polish journal of microbiology

P O L S K I E T O WA R Z Y S T W O M I K R O B I O L O G Ó WP O L I S H S O C I E T Y O F M I C R O B I O L O G I S T S

Polish Journal of Microbiology

I am pleased to inform you that Polish Journal of Microbiology has been selectedfor coverage in Thomson Scientific products and customers information services.Beginning with No 1, Vol. 57, 2008 information on the contents of the PJM isincluded in: Science Citation Index Expanded (ISI) and Journal Citation Reports(JCR)/Science Edition.

Stanisława Tylewska-WierzbanowskaEditor in Chief

2011

INSTRUCTIONS TO AUTHORS ANd FULL TExT ARTICLES (IN PdF FORM) AVAILABLE AT:www.microbiology.pl/pjm

Polish Journal of Microbiologyformely Acta Microbiologica Polonica

2011, Vol. 60, No 2

CONTENTS

MINIREWIEV

Staphylococcal cassette chromosome mec (SCCmec) classification and typing methods: an overviewTURLEJ A., HRYNIEWICZ W., EMPEL J. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

ORIGINAL PAPERS

Immobilized cells of recombinant Escherichia coli strain for continuous production of L-aspartic acidSZYMAńSKA G., SOBIERAJSKI B., CHMIEL A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Reaction conditions for maximal cyclodextrin production by cyclodextrin glucanotransferase from Bacillus megateriumZHEKOVA B.Y., STANCHEV V.S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Efficacy of UV treatment in the management of bacterial adhesion on hard surfacesKOLAPPAN A., SATHEESH S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Comparition of the nucleotide sequences of wheat dwarf virus (WdV) isolates from Hungary and UkraineTOBIAS I., SHEVCHENKO O., KISS B., BYSOV A., SNIHUR H., POLISCHUK V., SALÁNKI K.,PALKOVICS L. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

β-glucanase productivity improvement via cell immobilization of recombinant Escherichia coli cells in different matricesBESHAY U., EL-ENSHASY H., ISMAIL I.M.K., MOAWAd H., ABd-EL-GHANY S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Utilization of UF-permeate for production of β-galactosidase by lactic acid bacteriaMURAd H.A., REFAEA R.I., ALY E.M. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Strains differentiation of Microsporum canis by RAPd analysis using (GACA)4 and (ACA)5 primersdOBROWOLSKA A., dęBSKA J., KOZłOWSKA M., STąCZEK P. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

SdS-PAGE heat-shock protein profiles of environmental Aeromonas strainsOSMAN K.M., AMIN Z.M.S., ALY M.A.K., HASSAN H., SOLIMAN W.S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Species-specific sensitivity of coagulase-negative staphylococci to single antibiotics and their combinationsSZYMAńSKA G., SZEMRAJ M., SZEWCZYK E.M. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Association between existence of integrons and multi-drug resistance in Acinetobacter isolated from patients in Southern IranJAPONI S., JAPONI A., FARSHAd S., ALI A.A., JAMALIdOUST M. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

dissemination of class 1, 2 and 3 integrons among different multidrug resistant isolates of Acinetobacter baumannii in Tehran hospitals, IranTAHERIKALANI M., MALEKI A., SAdEGHIFARd N., MOHAMMAdZAdEH d., SOROUSH S., ASAdOLLAHI P.,ASAdOLLAHI K., EMANEINI M.SHORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

COMMUNICATIONS

Host response to the presence of Helicobacter spp. dNA in the liver of patients with chronic liver diseasesRYBICKA M., NAKONIECZNA J., STALKE P., BIELAWSKI K.P. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

Polish Journal of Microbiology2011, Vol. 60, No 2, 95–103

MINIREVIEW

Introduction

Staphylococcus aureus is one of the leading causes of bacterial infections in developed countries and is responsible for a wide spectrum of diseases, ranging from minor skin infections to fatal necrotizing pneu-monia. Since the introduction of penicillin into medical treatment in early 1940s, the resistance for beta-lactams has started to develop. It was a result of the acquisi-tion of a plasmid, coding for penicillinase, a penicil-lin-hydrolyzing enzyme, which is able to cleave the beta-lactam ring and thus inactivate antibiotic mol-ecule. Penicillin resistant strains soon spread not only in healthcare facilities, but also in the community. To overcome infections caused by beta-lactamase-produc-ing S. aureus, a narrow spectrum semi-synthetic peni-cillin (meticillin) was introduced. However, soon after that, in 1961, first meticillin-resistant Staphylococcus aureus (MRSA) strain was identified. Initially, MRSA strains were encountered only in the hospitals, but in the late 1990s first virulent community-acquired MRSA (CA-MRSA) clones, characterized by the presence of the toxin Panton-Valentine leukocidin (PVL), appeared rapidly and unexpectedly. They quickly spread world-wide, initially only in the community, but later on also in the healthcare facilities, displacing in some caun-

* Corresponding author: Agata Turlej, National Medicine Institute, Chełmska 30/34, 00-725 Warsaw, Poland; phone: (48-22) 841 57 69; fax: (48-22) 8412949; e-mail: [email protected]

tries typical HA-MRSA. For this reason, nowadays, distinction between CA-MRSA and mostly multire-sistant HA-MRSA become challenging (Chambers and deleo, 2009; deurenberg and Stobberingh, 2008). The resistance of S. aureus to meticillin is caused by the presence of the mecA gene, encoding for an addi-tional 78-kda penicillin binding protein 2a, (PBP2a or PBP2’). Compared to other PBP, PBP2a has a low affin-ity for all beta-lactam antibiotics. As a result of that, even in the presence of a beta-lactam antibiotic, the peptidoglycan layer biosynthesis is not disrupted and the bacterium can survive (Berger-Bachi and Rohrer, 2002; deurenberg and Stobberingh, 2008). The mecA gene is located within a mec operon together with its regulatory genes: mecI and mecR1. (Berger-Bachi and Rohrer, 2002). The mec operon is carried by staphylo-coccal cassette chromosome mec (SCCmec). The ori-gin of SCCmec is still unknown, but it is proposed that it was acquired by S. aureus from S. sciuri and that the mecA-positive coagulase-negative staphylococci (CoNS), especially S. epidermidis, may be a potential reservoir for the SCCmec element (Mongkolrattanothai et al., 2004; Wu et al., 2001). On the other hand, it is sug-gested that the main source of SCCmec could be MRSA itself (Aires de Sousa and de Lencastre, 2004). There are also suggestions of possible acquisition of the mecA

Staphylococcal Cassette Chromosome mec (SCCmec) Classificationand Typing Methods: an Overview

AGATA TURLEJ1, WALERIA HRYNIEWICZ1 and JOANNA EMPEL2

1 department of Epidemiology and Clinical Microbiology, 2 department of Molecular MicrobiologyNational Medicines Institute, Warsaw, Poland

Received 6 April 2011, accepted 6 May 2011

A b s t r a c t

Meticillin-resistant Staphylococcus aureus (MRSA) is one of the main causes of hospital-acquired infections, but since late 1990s also the community-acquired. For better understanding of the S. aureus epidemiology there is an urgent need for creation of new typing method for SCCmec element. The molecular typing of MRSA for epidemiological purposes is investigated by pulsed-field gel electrophoresis (PFGE), multilocus sequence typing (MLST), spa typing and the SCCmec type assignment. In last few years not only new type of SCCmec (VI to xI) have been identified, but also additional subtypes (i.e. IVg-j) and different variants of already existed one (i.e. 5C2&5 and 2B2&5) were discovered. The aim of this review is to briefly summarize current knowledge about SCCmec classification and to discuss advantages and disadvantages of selected SCCmec typing methods.

K e y w o r d s: Staphylococcus aureus (MRSA), SCCmec classification and typing

Turlej A. et al. 296

region from S. fleurettii, which is a commensal bacte-rium of animals (Tsubakishita et al., 2010). SCCmec typing, which classifies SCCmec elements on the basis of their structural differences, is applied in epidemio-logical studies to distinguish MRSA strains or to define an MRSA clone in combination with the genotype of meticillin-susceptible S. aureus (MSSA) strain in which a SCCmec element has integrated.

SCCmec element composition

SCCmec elements, detected in almost all MRSA strains, belong to particular type of the staphylococcal mobile genetic elements coding for meticillin-resist-ance and designated as staphylococcal cassette chromo-some mec (Katayama et al., 2000). In S. aureus strains, SCCmec elements always integrate sequence specifically at the unique site of the chromosome, attBscc (bacterial chromosomal attachment site). The attBscc is located near the origin of replication, at the 3’ end of orfX, cod-ing for an open reading frame x of unknown function, well conserved among both MRSA and MSSA strains (Hiramatsu et al., 2001; Ito et al., 1999; Ito et al., 2001). The attachment site contains a core 15-bp sequence, called the integration site sequence (ISS), that is neces-sary for ccr-mediated recombination (IWG-SCC, 2009; Katayama et al., 2000). After integration of SCCmec into the chromosome, ISS is found in direct repeat sequences at left and right SCCmec/chromosomal junctions of the integrated SCCmec element. different SCCmec ele-ments share similar backbone structure, that consists of (i) mec complex, composed of mecA operon, (ii) ccr gene complex, composed of cassette chromosome recombinase (ccr) gene(s) and (iii) three regions bor-dering the ccr and mec complexes, designated as join-ing (J) regions. The composition of almost all SCCmec elements identified so far in S. aureus can be presented as follow: (orfX)J3-mec-J2-ccr-J1 (Chongtrakool et al., 2006; Hiramatsu et al., 2002). The exception constitute SCCmecVII and a newly described SCCmecIx, with the ccr gene complex positioned between J3 and J2 regions and the mec gene complex between J2 and J1 regions (Berglund et al., 2008; Li et al., 2011). It is noteworthy, that different authors, and sometimes even the same, present the structure of the same SCCmec elements in reverse orientations, what creates difficulties or even confusions, especially for these not experienced in the field. The orientation with SCCmec located at the right site of orfX gene seems to be more correct as it is con-sistent with direction on genomic maps of S. aureus and will be used for needs of this article. Since the struc-tural components mentioned above play a crucial role in classification of SCCmec elements they will be pre-sented below in more details.

The ccr gene complex. The ccr gene complex is com-posed of the ccr gene(s) surrounded by orfs. The ccr genes encode for dNA recombinases of the invertase-resolvase family, enzymes that can catalyze precise excision of the SCCmec as well as its integration, both site- and orientation-specific, into staphylococcal chromo some, being thus responsible for mobilization of the cassette (Katayama et al., 2000). Based on the com-position of ccr genes, two distinct ccr gene complexes have been reported to date, one carrying two adjacent ccr genes, ccrA and ccrB, and the second carrying ccrC. The ccrA and ccrB genes identified to date in S. aureus strains have been classified into four and five allo- types respectively, resulting in six ccr gene complex types, designated as type 1 (ccrA1B1), type 2 (ccrA2B2), type 3 (ccrA3B3), type 4 (ccrA4B4), type 7 (ccrA1B6) and type 8 (ccrA1B3). All identified so far ccrC variants have shown high nucleotide similarity and are assigned to only one allotype, ccrC1, constituting type 5 of ccr gene complex (Chongtrakool et al., 2006; IWG-SCC, 2009) (http://www.sccmec.org/).

The mec gene complex. Two evolutionary different lineages of mec gene complexes have been described in S. aureus. The first one, which encompasses the vast majority of known and well characterized mec gene complexes, have been observed in MRSA isolates of human origin since the nineties. The prototype of this lineage is the mec gene complex designated as class A, composed of an intact mec operon, the hyper-variable region (HVR) and the insertion sequence IS431 (Ito et al., 2001; Katayama et al., 2001). The mec operon includes mecA gene and located upstream of mecA its regulatory genes: mecR1 and mecI, coding for the signal transducer and the repressor, respectively. dif-ferences between class A mec gene complex and other mec gene complexes of this lineage, described to date, result mainly from insertions of IS elements, IS1272 or IS431, into the region of mecA regulatory genes, causing complete deletion of mecI and, different in size, partial deletions of mecR1. depending on the structural diver-sity of mecI-mecR1 region, five major classes of mec gene complexes, of the said lineage, have been defined by IGW-SCC (IWG-SCC, 2009):

•Class A, which contains intact mec gene complex: IS431-mecA-mecR1-mecI;

•Class B, where mecR1 is truncated by insertion sequence IS1272: IS431-mecA-ΔmecR1-IS1272;

•Class C1, where mecR1 is truncated by insertion sequence IS431 having the same direction as the IS431 downstream of mecA: IS431-mecA-ΔmecR1- -IS431;

•Class C2, where mecR1 is truncated by insertion sequence IS431 having the reverse direction to the IS431 downstream of mecA: IS431-mecA-ΔmecR1-IS431; and

SCCmec classification and typing methods2 97

•Class d, where mecR1 is partly deleted but there is no IS element downstream of ΔmecR1: IS431-mecA-ΔmecR1. This class has been observed in S. caprae only (Katayama et al., 2001).

Recently, data from genome sequencing project of the bovine S. aureus isolate LGA251, have revealed a mec gene complex of the second evolutionary lineage (McCarthy and Lindsay, 2010)) (http://www.sanger.ac.uk/pathogens). This new complex, depicted as: bla Z-mecALGA251-mecR1LGA251-mecILGA251, constitutes the sixth defined major class, assigned as class E (http://www.sccmec.org/).

Besides the major classes of the mec gene complex several variants within the classes have also been distin-guished, for example: class A3, where mecI is truncated by insertion sequence IS1182: IS431-mecA-mecR1-ΔmecI-IS1182 and class A4, where mecI is disrupted by insertion sequence IS1182: IS431-mecA-mecR1-ΔmecI-IS1182-ΔmecI (Shore et al., 2005) or class B2, which has an insertion of the transposone Tn4001 upstream of mecA in ΔmecR1: IS431-mecA-ΔmecR1- Tn4001-IS1272 (Heusser et al., 2007).

The joining (J) regions. Apart from the ccr and mecA gene complexes, essential for the SCCmec biolo-gical functions, the cassette comprises also three joining regions (J1-J3), previously called “junkyard” regions (Hiramatsu et al., 2002). J regions from different SCCmec elements are arranged in the same order. The J1 region is located at right site of the cassette, the J2 region between the ccr and the mec complexes and the J3 region at the left chromosomal junction adja-cent to orfX (IWG-SCC, 2009). Although considered as less important in terms of SCCmec functions, these regions are epidemiologically significant since they may serve as targets for plasmids or transposons, carrying additional antibiotic and heavy metal resistance deter-minants. Acquisition and accumulation of resistance genes by mobile elements like SCCmec enables their dissemination and in consequence leads to emerge of multidrug resistance strains. Examples of antibiotic resistance determinants, that may be carried within

J regions are summarized in Table I (Ito et al., 2003; Malachowa and deLeo, 2010). Sequence analysis of J regions from different SCCmec elements revealed that they are unique to particular types of ccr-mec gene com-plex combinations and that variations of these regions within the same ccr-mec gene complex combination are specific for SCCmec subtypes (Hisata et al., 2005; Ito et al., 2003; IWG-SCC, 2009; Kwon et al., 2005; Ma et al., 2002; Ma et al., 2006; Milheirico et al., 2007b; Oliveira et al., 2001; Oliveira and de Lencastre, 2002; Shore et al., 2005).

The SCCmec element classification

The first SCCmec element was identified in Japa-nese S. aureus strain, N315 in 1999 and shortly after two additional SCCmec from different MRSA strains were determined (Ito et al., 1999; Ito et al., 2001). Based on detailed structural analysis these three SCCmec ele-ments were classified as types I to III (Ito et al., 2001). In time, both new types of SCCmec, such as SCCmecIV (Ma et al., 2002), SCCmecV (Ito et al., 2004), SCCmecVI (Oliveira et al., 2006a), SCCmecVII (Berglund et al., 2008), SCCmecVIII (Zhang et al., 2009), SCCmecIx, SCCmecx (McCarthy and Lindsay, 2010), SCCmecxI (http://www.sccmec.org/) and many new variants of already known SCCmec types have been reported (Boyle- Vavra et al., 2005; Cha et al., 2005; Chlebowicz et al., 2010; Heusser et al., 2007; Higuchi et al., 2008; Hisata et al., 2005; Kwon et al., 2005; O’Brien et al., 2005; Oliveira and de Lencastre, 2002; Shore et al., 2005; Shukla et al., 2004; Stephens et al., 2007; Takano et al., 2008). With growing number of SCCmec types, sub-types or variants published in the literature it became obvious that without approved international rules of nomenclature system it would be difficult in the near-est future to keep in order suitable naming of new emerged SCCmec elements. To meet the urgent need the International Working Group on Classification of Staphylococcal Cassette Chromosome (SCC) Elements

pUB110 I, II, IVA ble bleomycin ant4’ tobramycinTn554 II, SCCHg,VIII ermA erythromycin aad9/spc streptomycin / spectinomycinSCCHg – mer mercurypT181 III tet tetracyclineΨTn554 III cad cadmiumTn4001 IV (IVc and 2B&5) aacA-aphd aminoglycosides

Table IAdditional resistance genes located on mobile elements within the SCCmec elements

Genetic element SCCmec type/subtype Gene Resistance


(IWG-SCC) was established in 2009. The main objec-tives of the group was to define consensus rules of a uniform nomenclature system for SCCmec ele-ments, determine minimum requirements for the description of the new SCCmec elements and estab-lish guidelines for the identification of SCCmec ele-ments for epidemiological studies (IWG-SCC, 2009). In published guidelines IWG-SCC decided to retain the previous nomenclature of SCCmec with addi-tional information about combination of ccr complex type and class of mec complex present in the element. Thus, classification of SCCmec element into the types (SCCmec typing), should be based on the combination of the type of ccr gene complex and the class of the mec gene complex present in the cassette, while variants within SCCmec types (SCCmec subtyping) should be defined by differences in their J regions, as it was pro-posed earlier by Hiramatsu group (Chongtrakool et al., 2006; IWG-SCC, 2009). Accordingly, SCCmec type I, was described additionally as 1B, what indicates the SCCmec element harboring the type 1 ccr and a class B mec gene complexes. To date, eleven SCCmec types have been defined. The other known SCCmec types are designated type II (2A), type III (3A), type IV (2B), type V (5C2), type VI (4B), type VII (5C1), type VIII (4A), type Ix (1C2), type x (7C1) and type xI (8E). They all are summarized in Table II.

The improved way of SCCmec classification allowed also to assign the mosaic variants of SCCmec. For

example, SCCmec element from ZH47 strain, harbor-ing type 2 ccr gene complex and additional ccrC1 in combination with mec class B2 was designated type IV (2B&5), while SCCmec element from TSGH17 and PM1 strains, carrying two different ccrC1 allels, 2 and 8, that was previously reported as SCCmec type VII or Taiwanese SCCmec type V (SCCmecVT), was designa ted type V (5C2&5) (Boyle-Vavra et al., 2005; Heusser et al., 2007; Higuchi et al., 2008; IWG-SCC, 2009; Takano et al., 2008).

due to the increasing diversity among SCCmec sub-types, IWG-SCC proposed the preparation of a com-puterized system, which will be able to characterize and assign certain SCCmec subtype based on the occur-rence of specific elements within the J regions.

Together with the discovery of new SCCmec types, also a need for new more complex SCCmec typing methods has emerged.

Available SCCmec typing methods

It had already been observed that the worldwide spread of MRSA is driven by the dissemination of a number of clones with a specific genetic background. Epidemiological studies revealed that for proper clone assignment not only the multilocus sequence typing (MLST) and spa typing is required, but also SCCmec typing is needed (deurenberg et al., 2007). Since that

Table IIReference strains for SCCmec types, which have been described up to date

SCCmectype/subtype Strain GenBank

accession no descriptionIsolationdateOrigin

I (1B) NCTC 10442 AB033763 UK 1961 (Ito et al., 2001)II (2A) N315 d86934 Japan 1982 (Ito et al., 1999)III (3A) 85/2082 AB037671 New Zealand 1985 (Ito et al., 2001)IVa (2B) CA05 AB063172 USA 1999 (Ma et al., 2002)IVb (2B) 8/6-3P (JCSC1978) AB063173 USA 1996 (Ma et al., 2002)IVc (2B) 81/108 (MR108) AB096217 Japan NA* (Ito et al., 2003)IVd (2B) JCSC4469 AB097677 Japan 1982 (Ma et al., 2006)IVg (2B) M03-68 dQ106887 Korea 2003 (Kwon et al., 2005)IVh (2B) HAR22 NA Finland 2002 (Milheirico et al., 2007b)IVi (2B) JCSC6668 AB425823 Sweden 1999 (Berglund et al., 2009)IVj (2B) JCSC6670 AB425824 Sweden 1990 (Berglund et al., 2009)V (5C2) WIS (JCSC3624) AB121219 Australia 1999 (Ito et al., 2004)VI (4B) HdE 288 AF411935 Portugal 1996 (Oliveira et al., 2006a)VII (5C1) JCSC6082 AB373032 Sweden 2002 (Berglund et al., 2008)VIII (4A) C10682 FJ670542 Canada 2003 (Zhang et al., 2009)Ix 1(C2) JCSC6943 NA Thailand 2006 (Li et al., 2011)x (7C1) JCSC6945 NA Canada 2006 (Li et al., 2011)xI (8E) LGA251 Na NA NA http://www.sccmec.org/

* NA – information not available


time, it has become necessary to find an easy and robust method for SCCmec element identification and typ-ing. The more complex our knowledge about SCCmec elements is, the more challenging invention of the typ-ing method becomes and the more relevant becomes the question, how accurate should we be in assigning SCCmec element type for epidemiological purposes. The SCCmec typing methods has been developed along with the new SCCmec types descriptions and appear-ance of the novel techniques or approaches for their analysis. Three different schemes of SCCmec typing can be distinguished: methods based on the restriction enzymes digestion, methods based on PCR or multiplex PCR (M-PCR) and methods based on real-time PCR.

First SCCmec typing methods. The first methods for detecting polymorphisms in the mecA vicinity were based on hybridizations of the mecA probe and Tn554 probe with ClaI-digested genomic dNA from the ana-lyzed isolates (Leski et al., 1998). This method was very useful for epidemiological studies before the SCCmec element’s structure was described. Nowadays there are some concepts on using restriction enzymes digestion in combination with PCR, like in multienzyme mul-tiplex PCR-amplified fragment length polymorphism (ME-AFLP) or in SCCmec typing method by PCR amplification of the ccrB gene in combination with restriction fragment length polymorphism (RFLP) employing endonucleases HinfI and BsmI (van der Zee et al., 2005; Yang et al., 2006).

In the first one, the obtained typing patterns were found to cluster together according to the SCCmec type of the strain, with the discriminatory power com-parable to PFGE. However, it is just a pattern based typing, which might be an interesting method for pre-screening of a large strain collection, but is by far not sensitive enough to proper assign SCCmec type for epi-demiological purposes, since it does not recognize any characteristic features for already described SCCmec elements. The second method is simple and time-effec-tive, but far not elaborate enough, because it focuses only on the ccrB typing. Since the ccrC genes were also described, it is not complex enough to be useful, concerning the current knowledge of SCCmec types. Moreover, for proper SCCmec assignment the mec class description in combination with ccr type is necessary. It seems that the scheme of SCCmec typing based on restriction enzymes digestion is no longer superior. The most common methods used nowadays for SCCmec typing based on PCR are summarized below.

PCR based SCCmec typing methods. during the past several years, a number of SCCmec typing meth-ods based on multiplex PCR (M-PCR) have been devel-oped (Boye et al., 2007; Hisata et al., 2005; Kondo et al., 2007; Milheirico et al., 2007a; Oliveira and de Lencastre, 2002; Zhang et al., 2005). Two different approaches were

applied in this methods; one was focused on analysis of J regions, whereas the other determine mainly mec class and ccr type. The first M-PCR method was described by Oliveira et al. (Oliveira and de Lencastre, 2002). At that time it was innovative technique that enabled to increase analysis scale and exchange the information about SCCmec types all over the world. It was based on identification of specific genes or motifs located mostly in the J regions of particular cassettes. Potentially, this method should detected SCCmec type I–IV, but in practice detection of SCCmec type III was problematic as in fact the primers were designed not to SCCmec type III but to so-called SCCmercury, which at that time was believed to be the integral part of the element. For more details see Chongtrakool et al. (Chongtrakool et al., 2006). This SCCmec typing strategy also did not discrim-inate SCCmec type IV and VI. (Oliveira et al., 2006a). In 2007, the same group published an update for Oliveira’s method, which improved the detection of SCCmec type I to IV and includes the structure determination of the SCCmec type V and VI. How-ever, the SCCmecVI was suggested to be confirmed by ccrB sequencing, which is costly and time-consuming (Milheirico et al., 2007a; Oliveira et al., 2006b) (http://www.ccrbtyping.net). The most significant advantage of this method is that it is a quick, single-tube M-PCR reaction for all detectable by this method SCCmec types. Unfortunately, the method is still based on mark-ers located within the J regions. This can cause some problems, as for example in our practice we sometimes see the pattern for SCCmec type I similar to type VI, which is confusing. Almost at the same time Zhang et al. has described a complex method for SCCmec I to V typing and SCCmec type IV subtyping in a mul-tiplex PCR reactions (Zhang et al., 2005). This method include three M-PCR reactions and one single target PCR reaction with sets of primers specific to mec, ccr and J1 region. First M-PCR reaction uses a set of primers specific for SCCmec type I to V, with SCCmec subtypes IVa to IVd, the second M-PCR reaction uses primers for assigning mec class A and B and third M-PCR reaction uses primers for type 1–3 ccr described previously by Ito et al. For detection of type 5 ccr the authors propose PCR reaction with single pair of prim-ers (Ito et al., 2001; Zhang et al., 2005). The proposed approach seems to be not useful for large-scale analysis, since four separate reactions should be done. More-over, it does not allow to detect neither mec class C1 and C2 nor type 4 ccr. On the other hand, this method allows to detect Taiwanese SCCmec type V (5C2&5). This mosaic variant of SCCmec type V gives one extra band of 1599 bp, compared to the reference type V (5C2). Another SCCmec typing method was devel-oped in 2005 by Hisata et al. who proposed another set of primers (Hisata et al., 2005). The method allowed


to detect SCCmec type I, IIa, IIb, III and IVa to IVd and was based mainly on mec class and ccr type assign-ment. Unfortunately, it was not possible to perform the analysis in a single M-PCR reaction and the method did not become popular widely. Another recently devel-oped technique was presented by Boye et al. (Boye et al., 2007). It is a quick and easy to interpret method based on single-tube M-PCR reaction, using primers for spe-cific detection of both mec class and ccr type of SCCmec type I to V. It seems to be very useful for first screen-ing of large amount of strains, but for the detection of SCCmec type I to III it is not complex enough. For example, to detect SCCmec type III only ccrC specific pair of primers is used, which confirms just the pres-ence of SCCmercury (Boye et al., 2007). Moreover, this method also misclassifies SCCmec type I with type VI. However, this method can also be used for confirma-tion of doubtful SCCmec types. The most complex and promising system for SCCmec assignment, especially in the light of the new guidelines for SCCmec elements classification was developed by Kondo et al. (Kondo et al., 2007). This PCR scheme combines six M-PCR reactions: M-PCR 1 for amplification of ccr type (1–4) along with mecA gene; M-PCR 2 for amplification of mec class (A, B and C2); M-PCR 3 for amplification of ORFs from J1 region of SCCmec type I and IV; M-PCR 4 for amplification of ORFs from J1 region of SCCmec type II, III and V; M-PCR 5 and 6 for amplification of gene alleles located in J2 and J3 region of SCCmec elements, respectively. M-PCRs 5 and 6 are used for the identification of integrated copies of transposons (Tn554 or ΨTn554) and plasmids (pUB110 or pT181). The most significant advantage of this method is its flexibility, since it does not detect any particular SCC-mec type, but only crucial loci, which in combination gives at the end the SCCmec type. This approach poten-tially allows detection of SCCmec types I to Ix except SCCmecVII and x since primers specific for mec class C1 have not been included to this system yet. Using M-PCRs 3 to 6 it is possible also to identify variety of SCCmec subtypes. However, since this method requires a relatively large number of PCR reactions to determine the structure of SCCmec it is quite complicated and time consuming. That is why it is suggested to perform just M-PCR 1 and 2 to assign type of SCCmec elements and in most cases, it may be enough for epidemiologi-cal purposes. Recently new SCCmec type V variants, which are getting epidemiologically important, were also described (Chlebowicz et al., 2010; Higuchi et al., 2008). It turned out that despite the high similarity among ccrC1 genes, its’ specific alleles 1, 2, 8, 9 and 10 typing could be also important for SCCmec type V precise characterization. For detection of mosaic SCCmecV (5C2&5) variant, Higuchi et al. provides a set of primers specific for two types 5 ccr (ccrC1 allel 2 and

ccrC1 allel 8), two characteristic ORFs (orf33 and orf35) and the mec class C2 variant, in which nucleotide sub-stitution in IS431 results in truncated transposase in ΨIS431 in mec complex C2 (Higuchi et al., 2008). The last method we would like to mention here concerns detection of recently discovered SCCmec type VIII (McClure et al., 2010; Zhang et al., 2005). It is based on single-tube M-PCR reaction using a set of primers specific for detection of characteristic features of this particular of SCCmec element, namely type 4 ccr, mec class A, and a unique junction within the J region as well as internal controls. Regarding the growing num-ber of MRSA clones harbouring different SCCmec IV subtypes one of the most important issue become also invention of a robust method for SCCmec type IV subtyping. In our opinion dedicated to solve this prob-lem is method published by Milheirico et al., which is to our knowledge the most complex among available methods and allow to detect SCCmec type IV subtypes from a to h (Milheirico et al., 2007b). We also use sub-typig descriebed by Zhang et al., but we find that quite often it is difficult to detect SCCmec type IVc using this method, while we never had such problems when using Milheirico et al. method.The PCR based methods for SCCmec typing are the most common and the easiest to implement in laboratories, since they do not require additional expensive equipment, but on the other hand, they are labor and time-consuming, compared to real-time PCR based analysis.

Real-time PCR based SCCmec typing methods. In parallel to PCR based methods for SCCmec typing, the methods based on real-time PCR have also been developed. A multiplex scheme based on a real-time PCR targeting the ccrB regions of SCCmec types I to IV was published in 2004 by Francois et al., but with cur-rent knowledge about SCCmec types this method is not elaborate enough (Francois et al., 2004). Recently, however, a new approach using ccr-specific padlock probes and tag microarray analysis for simultaneous probing of core genome diversity and identification of SCCmec was developed. However, the set of padlock probes includes only oligonucleotides targeting diag-nostic regions in the mecA, ccrB and ccrC genes with-out mec class recognition (Kurt et al., 2009). A signifi-cant disadvantage of these methods is that they detect only ccr locus and ignore the mec complex, which may result in misclassification and do not allow the detec-tion of novel combinations of the mec and ccr complex in SCCmec elements. On the other hand, in the same year, a very interesting method for SCCmec element typing was published, based on a rapid molecular beacon real-time PCR (MB-PCR) assay (Chen et al., 2009). The design of the system is based on the defini-tion of SCCmec types as a combination of the ccr allo-type along with the mec class complex. The assay con-


sists of two multiplex panels, the combination of which results in two targets (mec class, ccr) for each SCCmec type. MB-PCR panel I targets mecA, ccrB2, mecI, and the ΔmecR1-IS1272 junction (mec class B) and thus can definitively identify SCCmec types II and IV. MB-PCR panel II detects ccrC, ccrB1, ccrB3, ccrB4, and the ΔmecR1-IS431 junction (mec class C2) and is there-fore capable of identifying SCCmec types I, III, V, and VI in combination with panel I. This method can also detect the recently described novel SCCmec type VIII (ccrAB4 with mec class A). The authors of this method ascertain that it is possible to easily classify isolates within 3 to 4 h, including dNA isolation, PCR cycling and analysis, which is extremely quick. However, in this analysis it is impossible to detect SCCmec subtypes, since there are no primers designed for J-regions and this method does not detect the mec class C1, which is characteristic for SCCmec type VII and x.

The most significant advantage of real-time PCR based methods is the small amount of time and labor required for the analysis. They do not combine many preparatory steps and are easy to interpret. On the other hand, they require special equipment and reagents, which are very expensive.

Conclusion

A variety of different methods for SCCmec typing are now available. To cope with the increasing diversity of SCCmec elements being reported, methods for their detection should be elaborated and flexible. Conven-tional PCR assays using commonly up to ten primer pairs in a single-tube assay can give various sensitivi-ties, depending on the template quality and may be easily contaminated. On the other hand, the real-time PCR assay requires expensive reagents and instru-ments, which can limit its use in many microbiological laboratories. Unfortunately, there is still no method available for SCCmec type VII and x–xI typing. High heterogenicity and variability of SCCmec elements make them complicated as a markers for epidemiological clone assignment. In our opinion, up to date, the best way of assigning the SCCmec type is to prepare the M-PCR 1 and 2 according to Kondo et al. or real-time PCR according to Chen et al., for ccr type and mec class detection. For SCCmec type IV subtyping we recom-mend method by Milheirico et al. (Milheirico et al., 2007b). This proceeding should be enough in most cases. For more accurate subtyping we suggest using the other M-PCRs described by Kondo et al. and if neces sary also methods established by Higuchi et al. and McClure et al. Recent work done by IWG-SCC clarified the classification of the major SCCmec elements type, but there is still a lot of ambiguity regarding naming of the

SCCmec variants, which requires further investigations and agreements. This includes answering the question, how accurate should we be in assigning SCCmec ele-ment type/variant for epidemiological purposes.

AcknowledgementsThis work was a part of the activities of the CONCORd Col-

laborative project supported by the European Commission grant HEALTH-F3-2008-222718. WH and JE was also partly supported by finding from the European Community MOSAR network contract LSHP-CT-2007-037941.

Literature

Aires de Sousa M. and H. de Lencastre. 2004. Bridges from hos-pitals to the laboratory: genetic portraits of methicillin-resistant Staphylococcus aureus clones. FEMS Immunol. Med. Microbiol. 40: 101–111.Berger-Bachi B. and S. Rohrer. 2002. Factors influencing methicil-lin resistance in staphylococci. Arch. Microbiol. 178: 165–171.Berglund C., T. Ito, M. Ikeda, X.X. Ma, B. Soderquist and K. Hira-matsu. 2008. Novel type of staphylococcal cassette chromosome mec in a methicillin-resistant Staphylococcus aureus strain isolated in Sweden. Antimicrob. Agents. Chemother. 52: 3512–3516.Berglund C., T. Ito, X.X. Ma, M. Ikeda, S. Watanabe, B. Soderquist and K. Hiramatsu. 2009. Genetic diversity of methicillin-resistant Staphylococcus aureus carrying type IV SCCmec in Orebro County and the western region of Sweden. J. Antimicrob. Chemother. 63: 32–41.Boye K., M.D. Bartels, I.S. Andersen, J.A. Moller and H. Westh. 2007. A new multiplex PCR for easy screening of methicillin-resis-tant Staphylococcus aureus SCCmec types I–V. Clin. Microbiol. Infect. 13: 725–727.Boyle-Vavra S., B. Ereshefsky, C.C. Wang and R.S. Daum. 2005. Successful multiresistant community-associated methicillin-resis-tant Staphylococcus aureus lineage from Taipei, Taiwan, that carries either the novel staphylococcal chromosome cassette mec (SCCmec) type VT or SCCmec type IV. J. Clin. Microbiol. 43: 4719–4730.Cha H.Y., D.C. Moon, C.H. Choi et al. 2005. Prevalence of the ST239 clone of methicillin-resistant Staphylococcus aureus and dif-ferences in antimicrobial susceptibilities of ST239 and ST5 clones identified in a Korean hospital. J. Clin. Microbiol. 43: 3610–3614.Chambers H.F. and F.R. Deleo. 2009. Waves of resistance: Staphy-lococcus aureus in the antibiotic era. Nat. Rev. Microbiol. 7: 629–641.Chen L., J.R. Mediavilla, D.C. Oliveira, B.M. Willey, H. de Lencastre and B.N. Kreiswirth. 2009. Multiplex real-time PCR for rapid staphylococcal cassette chromosome mec typing. J. Clin. Microbiol. 47: 3692–3706.Chlebowicz M.A., K. Nganou, S. Kozytska, J.P. Arends, S. Engel-mann, H. Grundmann, K. Ohlsen, J.M. van Dijl and G. Buist. 2010. Recombination between ccrC genes in a type V (5C2&5) staphylococcal cassette chromosome mec (SCCmec) of Staphylococ-cus aureus ST398 leads to conversion from methicillin resistance to methicillin susceptibility in vivo. Antimicrob. Agents. Chemother. 54: 783–791.Chongtrakool P., T. Ito, X.X. Ma et al. 2006. Staphylococcal cassette chromosome mec (SCCmec) typing of methicillin-resistant Staphy-lococcus aureus strains isolated in 11 Asian countries: a proposal for a new nomenclature for SCCmec elements. Antimicrob. Agents.Chemother. 50: 1001–1012.Deurenberg R.H., C. Vink, S. Kalenic, A.W. Friedrich, C.A. Brug-geman and E.E. Stobberingh. 2007. The molecular evolution of


methicillin-resistant Staphylococcus aureus. Clin. Microbiol. Infect. 13: 222–235.Deurenberg R.H. and E.E. Stobberingh. 2008. The evolution of Staphylococcus aureus. Infect. Genet. Evol. 8: 747–763.Francois P., G. Renzi, D. Pittet, M. Bento, D. Lew, S. Harbarth, P. Vaudaux and J. Schrenzel. 2004. A novel multiplex real-time PCR assay for rapid typing of major staphylococcal cassette chromosome mec elements. J. Clin. Microbiol. 42: 3309–3312.Heusser R., M. Ender, B. Berger-Bachi and N. McCallum. 2007. Mosaic staphylococcal cassette chromosome mec containing two recombinase loci and a new mec complex, B2. Antimicrob. Agents. Chemother. 51: 390–393.Higuchi W., T. Takano, L.J. Teng. and T. Yamamoto. 2008. Struc-ture and specific detection of staphylococcal cassette chromosome mec type VII. Biochem. Biophys. Res. Commun. 377: 752–756.Hiramatsu K., L. Cui, M. Kuroda and T. Ito. 2001. The emergence and evolution of methicillin-resistant Staphylococcus aureus. Trends. in microbiology 9: 486–493.Hiramatsu K., Y. Katayama, H. Yuzawa and T. Ito. 2002. Molecular genetics of methicillin-resistant Staphylococcus aureus. Int. J. Med. Microbiol. 292: 67–74.Hisata K., K. Kuwahara-Arai, M. Yamanoto et al. 2005. dissemi-nation of methicillin-resistant staphylococci among healthy Japa-nese children. J. Clin. Microbiol. 43: 3364–3372.Ito T., Y. Katayama and K. Hiramatsu. 1999. Cloning and nucleo-tide sequence determination of the entire mec dNA of pre-methi-cillin-resistant Staphylococcus aureus N315. Antimicrob. Agents. Chemother. 43: 1449–1458.Ito T., Y. Katayama, K. Asada, N. Mori, K. Tsutsumimoto, C. Tiensasitorn and K. Hiramatsu. 2001. Structural comparison of three types of staphylococcal cassette chromosome mec integrated in the chromosome in methicillin-resistant Staphylococcus aureus. Antimicrob. Agents. Chemother. 45: 1323–1336.Ito T., K. Okuma, X.X. Ma, H. Yuzawa and K. Hiramatsu. 2003. Insights on antibiotic resistance of Staphylococcus aureus from its whole genome: genomic island SCC. Drug. Resist. Updat. 6: 41–52.Ito T., X.X. Ma, F. Takeuchi, K. Okuma, H. Yuzawa and K. Hiramatsu. 2004. Novel type V staphylococcal cassette chromo-some mec driven by a novel cassette chromosome recombinase, ccrC. Antimicrob. Agents. Chemother. 48: 2637–2651.IWG-SCC. 2009. Classification of staphylococcal cassette chromo-some mec (SCCmec): guidelines for reporting novel SCCmec ele-ments. Antimicrob. Agents. Chemother. 53: 4961–4967.Katayama Y., T. Ito and K. Hiramatsu. 2000. A new class of genetic element, staphylococcus cassette chromosome mec, encodes meth-icillin resistance in Staphylococcus aureus. Antimicrob. Agents. Chemother. 44: 1549–1555.Katayama Y., T. Ito and K. Hiramatsu. 2001. Genetic organization of the chromosome region surrounding mecA in clinical staphy-lococcal strains: role of IS431-mediated mecI deletion in expres-sion of resistance in mecA-carrying, low-level methicillin-resistant Staphylococcus haemolyticus. Antimicrob. Agents. Chemother. 45: 1955–1963.Kondo Y., T. Ito, X.X. Ma, S. Watanabe, B.N. Kreiswirth, J. Etienne and K. Hiramatsu. 2007. Combination of multiplex PCRs for staphylococcal cassette chromosome mec type assignment: rapid identification system for mec, ccr, and major differences in junkyard regions. Antimicrob. Agents. Chemother. 51: 264–274.Kurt K., A. Alderborn, M. Nilsson, B. Strommenger, W. Witte and U. Nubel. 2009. Multiplexed genotyping of methicillin-resis-tant Staphylococcus aureus isolates by use of padlock probes and tag microarrays. J. Clin. Microbiol. 47: 577–585.Kwon N.H., K.T. Park, J.S. Moon et al. 2005. Staphylococcal cas-sette chromosome mec (SCCmec) characterization and molecular

analysis for methicillin-resistant Staphylococcus aureus and novel SCCmec subtype IVg isolated from bovine milk in Korea. J. Anti-microb. Chemother. 56: 624–632.Leski T., D. Oliveira, K. Trzcinski, I.S. Sanches, M. Aires de Sousa, W. Hryniewicz and H. de Lencastre. 1998. Clonal distribution of methicillin-resistant Staphylococcus aureus in Poland. J. Clin. Microbiol. 36: 3532–3539.Li S., R.L. Skov, X. Han et al. 2011. Novel types of staphylococcal cassette chromosome mec elements identified in CC398 methi-cillin resistant Staphylococcus aureus strains. Antimicrob. Agents. Chemother.Ma X.X., T. Ito, C. Tiensasitorn, M. Jamklang, P. Chongtrakool, S. Boyle-Vavra, R.S. Daum and K. Hiramatsu. 2002. Novel type of staphylococcal cassette chromosome mec identified in commu-nity-acquired methicillin-resistant Staphylococcus aureus strains. Antimicrob. Agents. Chemother. 46: 1147–1152.Ma X.X., T. Ito, P. Chongtrakool and K. Hiramatsu. 2006. Pre- dom inance of clones carrying Panton-Valentine leukocidin genes among methicillin-resistant Staphylococcus aureus strains isolated in Japanese hospitals from 1979 to 1985. J. Clin. Microbiol. 44: 4515–4527.Malachowa N. and F.R. DeLeo. 2010. Mobile genetic elements of Staphylococcus aureus. Cell. Mol. Life Sci. 67: 3057–3071.McCarthy A.J. and J.A. Lindsay. 2010. Genetic variation in Staphy-lococcus aureus surface and immune evasion genes is lineage asso-ciated: implications for vaccine design and host-pathogen inter-actions. BMC Microbiol. 10: 173.McClure J.A., J.M. Conly, S. Elsayed and K. Zhang. 2010. Multi-plex PCR assay to facilitate identification of the recently described staphylococcal cassette chromosome mec type VIII. Molecular and cellular probes 24: 229–232.Milheirico C., D.C. Oliveira and H. de Lencastre. 2007a. Update to the multiplex PCR strategy for assignment of mec element types in Staphylococcus aureus. Antimicrob. Agents. Chemother. 51: 3374–3377.Milheirico C., D.C. Oliveira and H. de Lencastre. 2007b. Multiplex PCR strategy for subtyping the staphylococcal cassette chromosome mec type IV in methicillin-resistant Staphylococcus aureus: ‘SCCmec IV multiplex’. J. Antimicrob. Chemother. 60: 42–48.Mongkolrattanothai K., S. Boyle, T.V. Murphy and R.S. Daum. 2004. Novel non-mecA-containing staphylococcal chromosomal cassette composite island containing pbp4 and tagF genes in a com-mensal staphylococcal species: a possible reservoir for antibiotic resistance islands in Staphylococcus aureus. Antimicrob. Agents. Chemother. 48: 1823–1836.O’Brien F.G., G.W. Coombs, J.C. Pearson, K.J. Christiansen and W.B. Grubb. 2005. Type V staphylococcal cassette chromosome mec in community staphylococci from Australia. Antimicrob. Agents. Chemother. 49: 5129–5132.Oliveira D.C., A. Tomasz and H. de Lencastre. 2001. The evolution of pandemic clones of methicillin-resistant Staphylococcus aureus: identification of two ancestral genetic backgrounds and the associ-ated mec elements. Microb. Drug. Resist. 7: 349–361.Oliveira D.C. and H. de Lencastre. 2002. Multiplex PCR strategy for rapid identification of structural types and variants of the mec element in methicillin-resistant Staphylococcus aureus. Antimicrob. Agents. Chemother. 46: 2155–2161.Oliveira D.C., C. Milheirico and H. de Lencastre. 2006a. Redefin-ing a structural variant of staphylococcal cassette chromosome mec, SCCmec type VI. Antimicrob. Agents. Chemother. 50: 3457–3459.Oliveira D.C., C. Milheirico, S. Vinga and H. de Lencastre. 2006b. Assessment of allelic variation in the ccrAB locus in methicillin-resistant Staphylococcus aureus clones. J. Antimicrob. Chemother. 58: 23–30.


Shore A., A.S. Rossney, C.T. Keane, M.C. Enright and D.C. Cole-man. 2005. Seven novel variants of the staphylococcal chromosomal cassette mec in methicillin-resistant Staphylococcus aureus isolates from Ireland. Antimicrob. Agents. Chemother. 49: 2070–2083.Shukla S.K., S.V. Ramaswamy, J. Conradt, M.E. Stemper, R. Reich, K.D. Reed and E.A. Graviss. 2004. Novel polymorphisms in mec genes and a new mec complex type in methicillin-resistant Staphy-lococcus aureus isolates obtained in rural Wisconsin. Antimicrob. Agents. Chemother. 48: 3080–3085.Stephens A.J., F. Huygens and P.M. Giffard. 2007. Systematic deri-vation of marker sets for staphylococcal cassette chromosome mec typing. Antimicrob. Agents. Chemother. 51: 2954–2964.Takano T., W. Higuchi, T. Otsuka et al. 2008. Novel characteristics of community-acquired methicillin-resistant Staphylococcus aureus strains belonging to multilocus sequence type 59 in Taiwan. Anti-microb. Agents. Chemother. 52: 837–845.Tsubakishita S., K. Kuwahara-Arai, T. Sasaki and K. Hiramatsu. 2010. Origin and molecular evolution of the determinant of methi-cillin resistance in staphylococci. Antimicrob. Agents. Chemother. 54: 4352–4359.van der Zee A., M. Heck, M. Sterks, A. Harpal, E. Spalburg, L. Kazobagora and W. Wannet. 2005. Recognition of SCCmec

types according to typing pattern determined by multienzyme multiplex PCR-amplified fragment length polymorphism analysis of methicillin-resistant Staphylococcus aureus. J. Clin. Microbiol. 43: 6042–6047.Wu S.W., H. de Lencastre and A. Tomasz. 2001. Recruitment of the mecA gene homologue of staphylococcus sciuri into a resistance determinant and expression of the resistant phenotype in Staphylo-coccus aureus. J Bacteriol 183, 2417–2424.Yang J.A., D.W. Park, J.W. Sohn and M.J. Kim. 2006. Novel PCR-restriction fragment length polymorphism analysis for rapid typ-ing of staphylococcal cassette chromosome mec elements. J. Clin. Microbiol. 44, 236–238.Zhang K., J.A. McClure, S. Elsayed, T Louie and J.M. Conly. 2005. Novel multiplex PCR assay for characterization and concomitant subtyping of staphylococcal cassette chromosome mec types I to V in methicillin-resistant Staphylococcus aureus. J. Clin. Microbiol. 43: 5026–5033.Zhang K., J.A. McClure, S. Elsayed and J.M. Conly. 2009. Novel staphylococcal cassette chromosome mec type, tentatively desig-nated type VIII, harboring class A mec and type 4 ccr gene com-plexes in a Canadian epidemic strain of methicillin-resistant Staphy-lococcus aureus. Antimicrob. Agents. Chemother. 53: 531–540.


ORGINAL PAPER

Introduction

When extracted from cells, intracellular enzymes can be used in solution for only one batch process if they are not immobilized. This shortcoming can be eliminated by enzyme immobilization; however, such preparations are frequently not sufficiently stable and their productivity is usually unsatisfactory for indus-trial purposes. These disadvantages can be overcome by immobilization of whole cells, enabling both cheaper and less laborious biotechnology to be developed. In their first technological description (Chibata et al., 1974) examined various methods for the immobili-zation of E. coli cells with high aspartase activity. As a result, L-aspartic acid production on an industrial scale using Escherichia coli cells immobilized in poly-acrylamide was developed in 1973 (Chibata et al., 1974). Subsequently, various other polymers for this process have been proposed and applied (Chibata et al., 1985, Fusee et al.,1981); however, it is obvious that continuous improvement of both the biological agents and process technology is necessary. We have previously described our investigations on strain improvement (Gadomska et al., 2007; Papierz et al., 2007) and in this paper, we present our own immobilization method of whole E. coli cells in chitosan for L-aspartic acid biosynthesis in continuous process in column bioreactors.

For industrial production of L-aspartic acid, column bioreactors, single or in combination of two or more columns, are proposed (Lee and Hong, 1988; Tosa et al., 1973; Kawabata et al., 1990; Sato et al., 1975). Contin-uous L-aspartic acid biosynthesis is carried out using immobilized cells of bacteria with high aspartase activity by passing an ammonium fumarate solution through the biocatalyst bed. Bioreactor productivity is closely related to aspartase activity, ammonium fuma-rate solution concentration and substrate flow rate. L-aspartic acid biosynthesis is profitable if the con-

version rate of ammonium fumarate to the product is over 90% (Mukouyama et al., 2000; Mukouyama and Komatsuzaki, 2001). In this paper, the continuous pro-cess of L-aspartic acid production in column bioreac-tors was optimized using immobilized cells of recom-binant strain Escherichia coli P1.

Experimental

Materials and Methods

Bacteria. For preliminary elaboration of cell immo-bilization procedure E. coli mutant B-715 (Papierz et al., 2007) was applied. For continuous process investiga-tions in bioreactors a recombinant strain of E. coli P1

Immobilized Cells of Recombinant Escherichia coli Strainfor Continuous Production of L-aspartic Acid

GRAŻYNA SZYMAńSKA*, BOGUSłAW SOBIERAJSKI and ALEKSANdER CHMIEL

department of Biosynthesis of drugs, Chair of Biology and Pharmaceutical Biotechnology,Medical University of łódź, Poland

Received 6 September 2010, revised 15 February 2011, accepted 20 February 2011

A b s t r a c t

For L-aspartic acid biosynthesis, high production cells of Escherichia coli mutant B-715 and P1 were immobilized in chitosan gel using a technique developed in our laboratory. The immobilization process reduced initial activity of the intact cells, however, the biocatalyst produced was very stabile for long-term use in multi-repeated batch or continuous processes. Temperature influence on the conver-sion of ammonium fumarate to L-aspartic acid was investigated. In long-term experiments, over 603 hours, the temperature 40°C was found to be the best for both biocatalyst stability and high conversion rate. The optimum substrate concentration was 1.0 M. Continuous production of L-aspartic acid was investigated in three types of column bioreactors characterized by different volumes as well as different high to biocatalyst bed volume rations (Hz/Vz). The highest conversion rate, 99.8%, and the productivity 6 g/g/h (mass of L-aspartic acid per dry mass of cells in biocatalyst per time unit) was achieved in the bioreactor with the highest value Hz/Vz = 3.1, and liquid hour space velocity value of 5.2, defined as the volume of feeding substrate passed per volume of catalyst in bioreactor per one hour.

K e y w o r d s: E. coli, immobilized cells, L-aspartic acid

* Corresponding author: G. Szymańska, department of Biosynthesis of drugs, Chair of Biology and Pharmaceutical Biotechnology, Medical University of łódź; ul. Muszyńskiego 1, 90-151 łódź, Poland; e-mail: [email protected]

Szymańska G. et al. 2106

(Gadomska et al., 2007) with high aspartase activity was used. The bacterial suspension in LB-medium was mixed with 50% glycerol (1:1), frozen and stored at –70°C as stocks for further use.

Media (as described earlier, Gadomska et al., 2007). (1) FF medium for biomass cultivation: Yeast Extract (difco) 20 g/l, ammonium fumarate 5.0 g/l, KH2PO4 11.4 g/l, MgSO4 × 7 H2O 0.5 g/l, pH 7.2. (2) Medium for cell activation (activation medium): ammonium fumarate 50.0 g/l, MgSO4 × 7 H2O 0.25 g/l, 1% Triton 0.5 ml/l, pH 8.5. (3) Medium for L-aspartic acid pro-duction (productive medium): ammonium fumarate 150.0 g/l, MgSO4 × 7 H2O 0.25 g/l, pH 8.5. Chemicals (at analytical grade), if not indicated otherwise, were purchased from POCh S.A.

Cell multiplication and activation. Bacteria were cultured in shaking flasks as described earlier (Gadom-ska et al., 2007), then the cells were centrifuged at 4000 rpm for 20 min and introduced into the activation medium (1 g of wet mass/20 ml). The cell suspension was activated by shaking for 24 hours in shaking flasks at 37°C. Activated cells were centrifuged at 4000 rpm for 20 min and washed twice with distilled water (1 g wet mass/20 ml water).

Immobilization of E. coli cells with chitosan. Acti-vated cells of E. coli were suspended in weight propor-tion 1:1 in a solution of chitosan consisting of 5.0 g of chitosan (Marine Institute, Gdynia) dissolved in 100 ml of 2% acetic acid (Chemical Company of Lub-lin) and stored for approximately 20 hours at ambient temperature. The mixture was instilled into cross- linking reagent via syringe. Sodium hexametaphos-phate (NaPO3)12–13Na2O (Fluka), sodium ortophosphate (POCh) and penta-sodium triphosphate Na5P3O10 (Fluka) were used in different concentrations as cross-linking reagents. The immobilisation process was opti-mized in this study. After 15–45 min of hardening, the gel pellets obtained were washed with distilled water and placed into the activation medium.

L-aspartic acid biosynthesisProcess with cell suspension. Activated cells were

mixed with the production medium (1 g wet mass/20 ml) in a 100 ml flask and shaken at a temperature of 37°C. After 15, 30, 60 and 120 min of incubation, samples of 0.1 ml were withdrawn for analysis.

Process with immobilized cells in shake flask. Immobilized cells (2 g biocatalysts containing 1g of wet biomass) were washed with distilled water and intro-duced into the shaken 100 ml flasks with 20 ml produc-tion media. The biosynthesis process was conducted as described above.

Continuous production of L-aspartic acid in bio-reactors by immobilized cells. Water jacket column bioreactors with different working volumes, i.e.: 2 ml,

20 ml and 40 ml were used. The substrate solution was passed through the column using a peristaltic pump. Both the medium and biocatalyst bed were kept at the same selected temperature.

Aspartic acid analysis. To estimate the amount of L-aspartic acid, HPLC was applied using a column 250–4 LichrospherTM 100RP-18 (Merck) and Waters fluorescence detector, type 474. The details of the ana-lytical procedure were described in a previous study (Papierz et al., 2007).

Results

Immobilization procedure optimization. In pre-liminary trials of immobilization of E. coli cells using mutant B-715 the bacterial cells were immobilized as chitosan pellets of 1.5–2.0 mm diameter using three cross-linking reagents: sodium hexametaphosphate, sodium ortophosphate or penta-sodium triphosphate. In the optimum immobilization procedure, 5% chitosan sol containing E. coli cells was reacted with 4% hexametaphosphate solution for a duration of approximately 30 min. The effect of the kind of phosphate ions used, their concentration and the time of pellet cross-link-ing on the activity and mechanical stability of the biocatalysts was investigated, and for further study, the recombinant cells of E. coli P1 were immobilized with 5% chitosan sol cross-linking 4% sodium hexameta-phosphate solution for 15–45 min as this is the best means of cell immobilization.

during the recombinant E. coli P1 cell immobiliza-tion process, some difficulties with obtaining a homo-geneous suspension of the cells in chitosan sol were observed. The biomass of the recombinant was sticky and stringy and the obtained pellets were irregular, with varying diameters. An irregular shape of the immobilized biocatalysts (Fig. 1) was unfavourable for packing them into a bioreactor. Those difficulties can influence the quality of the study and the reliability of the results obtained. With the aim of eliminating those difficulties, we examined the effect of compo-nents within the medium used for bacterial multiplica-tion, and added a surface-active substance (surfactant) for the immobilization of recombinant cells. Two variants of the FF medium (for multiplication) were applied: a) with the previously) determined amount of Yeast Extract (rich medium – FF20), b) with half the previously determined amount of Yeast Extract (poor medium – FF10). The activation of biomass was carried out for 24 hours at 37°C in two activation media: a) without Tween 80, b) with added Tween 80. The application of the poor medium for the multi-plication of recombinant E. coli P1, followed by the addition of surfactant for cell activation, caused easier

L-aspartic acid production by E. coli immobilized cells2 107

immobilization and more effective L-aspartic acid bio-synthesis (Fig. 2).

Activity of immobilized biocatalyst. Bacteria were cultivated and activated as described in Materials and Methods. The active biomass was divided into two parts. One part was used directly for the L-aspartic acid biosynthesis process in cell suspension; the second part was immobilized as described above. 1 g of intact cells or 2 g of immobilized cells were introduced into 20 ml of the productive medium in 100 ml flasks and shaken at a temperature of 37°C. After 15 min of incubation, the samples were taken for analysis in order to estimate an initial (i.e. maximum) process rate for both prepara-tions. The immobilized cells were almost 60% less active than the intact cells.

Effect of temperature. It is common knowledge that the rate of any biochemical process depends on its temperature; the rate increasing steadily as the tem-perature increases, up to the level at which enzymes are inactivated. Immobilization of the cells or enzymes can affect the thermal stability of biocatalysts. Exothermic

reactions, such as L-aspartic acid biosynthesis, com-plicate temperature optimization in long-term, large-scale bioreactor processes. There are different optimum temperature values for highest reaction rate in a batch short-term process and in a long-term continuous pro-cess. Biosynthesis in shaking flasks at the growth-opti-mum temperature 37°C and much higher temperatures: 48, 50, 52, 54 and 56°C was studied. In a short (up to 2 hours) test, biocatalyst activity increased together with an increase in temperature, obtaining a maximum at 54 and 56°C (Fig. 3). In the next experiment the activity of newly-immobilized cells at 56°C for a 72-hour bioreac-tor run was investigated with a substrate flow rate of 44 ml/h. In this process the conversion of ammonium fumarate to L-aspartic acid decreases during the first day by nearly one-third (Fig. 4). On the basis of this result

Fig. 1. E. coli P1 cells immobilized in chitosan gel: a) in rich medium without surfactant, b) in poor medium with surfactant.

Fig. 2. Effect of multiplication and activation of bacteriaon L-aspartic acid biosynthesis by immobilized cells E. coli P1.

Fig. 3. Effect of temperature on biosynthesis of L-aspartic acid byimmobilized cells E. coli P1.


a new process was conducted at temperatures of 37°C and 40°C. during a 603-hour operation of the bioreactor, biocatalyst activities at both temperatures were at simi-lar levels (about 55% conversion). For further experi-ments, temperatures in the range 37–40°C was applied.

The effect of ammonium fumarate concentration. The aim of evolving technology is to obtain the highest possible concentration of a given product. In one-step enzymatic reaction, such as L-aspartic acid biosynthe-sis, it is possible to obtain this result by increasing sub-strate concentration. The effect of production medium ammonium fumarate concentration on L-aspartic acid biosynthesis was investigated in three bioreac-tors running in parallel. Every bioreactor was supplied (100 ml/h) with different concentration of ammonium fumarate, i.e.: 1.0 mol/l (150 g/l), 1.2 mol/l (180 g/l) and 1.5 mol/l (225 g/l) in substrate solution. The initial rate of ammonium fumarate to L-aspartic acid conversion in every bioreactor was over 50%. However, over the next 40 days, a decrease in conversion yield, dependent on substrate concentration, was observed. In the days that followed, productivity was stable in every bioreac-tor; however, the best results were achieved with a sub-strate concentration of 1mol/l (Fig. 5).

After the continuous process had run for 603 hours in the bioreactors, the biocatalyst preparations were removed, and their activities in shaking flasks in fresh ammonium fumarate solution of 1 mol/l were investi-

gated as a short-term experiment for residual activity of biocatalysts estimation. An experiment with newly-immobilized cells was conducted as the control. The highest activity was observed for biocatalyst previously working in substrate solution of 1 mol/l (Fig. 5). The use of higher concentrations of ammonium fumarate during a long-term continuous process resulted in bio-catalyst inactivation.

Continuous production of L-aspartic acid. Appro-priate quantities of immobilized cells of E. coli P1 were placed into the three column bioreactors: A, B and C, as described in Table I. The conversion of ammonium fumarate to L-aspartic acid in continuous process at a substrate solution flow rate of below 100 ml/h in these reactors is shown in figure 6. In the first experiment carried out in bioreactor A, a fresh biocatalyst was used after cell immobilization. In bioreactor A with 43 ml of biocatalyst (3.8 g dry mass of E. coli P1 cells) production medium was passed through at flow rates rz from 5 to 858 ml/h at 37°C (Table II, Fig. 6). The highest conversion rate, over 95%, was obtained at a substrate flow rate of rz = 53–136 ml/h. Increasing the substrate flow rate to over 136 ml/h, a permanent decrease in substrate to product conversion ratio was observed. At a maximum flow rate of 858 ml/h, the conversion ratio decreased below 50%.

It is possible to assume that the freshly prepared bio-catalyst used in this experiment, despite earlier activa-

A 43.0 3.80 15.0 0.35B 10.0 0.83 3.5 0.35C 1.3 0.15 4.0 3.10

Table ICharacteristics of bioreactors A, B and C

Bioreactor Hz/Vz ratioHeight

of biocatalystbed (Hz) [cm]

dry mass of cells in

biocatalyst [g]

Volume of a biocatalystbed (Vz) [ml]

Fig. 4. decrease of conversion of ammonium fumarateto L-aspartic acid by immobilized cells E. coli P1 during 3 days

test in 56°C.

Fig. 5. Effect of ammonium fumarate concentrationon continuous production of L-aspartic acid in a column

bioreactor with immobilized cells E. coli P1.


tion, did not achieve its maximum activity. It could be that further cell activation occurred in the biocatalyst bed through the initial period of the biosynthesis pro-cess in the bioreactor. In the next experiment, freshly prepared pellets of biocatalyst were placed into pro-duction medium for 3 days before they were used for continuous process in bioreactor B. This biocatalyst preparation stage may well have caused further reduc-tion of the diffusion barrier for substrate and product through the external envelopes of the cells. As the result, the maximum conversion ratio of nearly 100% was achieved in bioreactor B during the initial period of the process for a flow rate rz = 2.8–10.0 ml/h. The productivity of the biocatalyst in this experiment was 0.4–1.6 g/g/h (Table III).

An important parameter introduced in this work, according to Mukouyama and Komatsuzaki (2001), was the ratio of the bioreactor height to the bioreactor volume (Hz/Vz). In both experiments for different bio-catalyst volumes in bioreactor A and B, the same ratio Hz/Vz = 0.35 was maintained. In the next experiment bioreactor C was used with a biocatalyst bed height of 4 cm and working volume of 1.3 ml; Hz/Vz = 3.1. The maximum conversion ratio of over 99% for the low flow rate of 6.8 ml/h, was achieved. The productivity of the biocatalyst in this experiment was 6 g/g/h. The increased substrate solution flow rate through bioreactor C resul-ted in the same decrease of conversion rate as in bio-reactors A and B, however with a lower ratio (Table IV).

Discussion

For L-aspartic acid biosynthesis, the high pro-duction cells of Escherichia coli were immobilized in chitosan gel using a technique developed in our labo-ratory. In the process of cell immobilization it is cru-cial to obtain a homogenous suspension of bacterial cells. In the case of the immobilization of recombinant E. coli P1, the addition of the surfactant Tween 80 to the medium for biomass cultivation was necessary. This

5 0.1 102.0 0.1 76.7 21 0.5 113.3 0.6 85.9 44 1.0 118.0 1.5 89.4 53 1.2 124.0 1.7 95.7 82 1.9 129.3 2.8 97.2 136 3.1 126.7 4.5 95.1 162 3.8 119.3 5.0 89.8 192 4.5 119.3 6.0 89.8 258 6.0 116.7 8.1 87.9 312 7.3 110.7 9.1 83.1 360 8.4 108.7 10.3 81.6 675 15.7 71.0 12.6 53.3 858 20.0 62.7 14.1 47.1

Table IIEffect of substrate medium flow rate through bioreactor A during

L-aspartic acid biosynthesis

* productivity as grams of L-aspartic acid calculated as 1 g dry weight of biomass per 1 hour [g/g/h]

Flowrate

[ml/h]

Liquid hourspace velocity

LHSV

Conversion[%]

Productivity*[g/g/h]

L-asparticacid[g/l]

2.8 0.3 132.6 0.4 99.7 10.0 1.0 132.3 1.6 99.5 24.8 2.5 113.8 3.4 85.6 40.2 4.0 109.5 5.3 82.3 115.5 11.6 76.9 10.6 57.8 123.0 12.3 75.1 11.1 56.5

Table IIIEffect of substrate medium flow rate through bioreactor B during



Flowrate

[ml/h]


LHSV

Conversion[%]


L-asparticacid[g/l]

Fig. 6. Effect of substrate flow rate on conversion of ammonium fumarate to L-aspartic acid in bioreactors A, B and C.

6.8 5.2 132.8 6.0 99.8 12.8 9.8 117.3 10.0 88.2 19.2 14.8 107.5 14.0 80.8 30.4 23.4 93.1 18.7 70.0 45.6 35.1 74.1 22.7 55.7

Table IVEffect of substrate medium flow rate through bioreactor C during



Flowrate

[ml/h]


LHSV

Conversion[%]


L-asparticacid[g/l]


surfactant facilitated the immobilization of the recom-binant cells; as mentioned above, their biomass was sticky and difficult to homogenize without it. Among three reagents, sodium orthophosphate, penta-sodium triphosphate and sodium hexametaphosphate, the final one was selected as the best cross-linking agent. Chi-bata et al. (1974) and Sato et al. (1975) have suggested the use of polyacrylamide gel; however, it is mechani-cally unstable and, following polymerization, some toxic monomer (acrylamide) usually remains inside the gel. Sato et al. (1979) and Umemura et al. (1984) have immobilized the cells and enzymes for L-aspartic acid production in κ-carrageenan gel. The main disadvan-tage of this method is the high temperature, 45–55°C, which is required for κ-carrageenan sol preparation. Very popular for biocatalyst preparation for various biochemical reactions is an alginate gel, which is cheap and extremely easy to prepare. We tested the alginate gel containing active E. coli cells for conversion of ammonium fumarate to L-aspartic acid (Chmiel et al., data not published). However, alginate beds proved very unstable in the process conditions.

It is commonly known that any immobilization technique causes reduction in the initial activity of intact free cells or enzymes. In the case of our immobi-lized biocatalyst, the activity reduction was about 60%, however the main aim of cell or enzyme immobiliza-tion is in stabilizing the biocatalyst for long-term use in multi-repeated batch or continuous processes.

It is necessary to optimize the basic reaction param-eters for the newly immobilized biocatalyst. The control of temperature during L-aspartic acid biosynthesis is extremely important because a considerable amount of heat is produced during the process. Conducting the process at a higher temperature may reduce the neces-sary cooling of the biocatalyst bed, thereby reducing the cost of production. In our work, the increase of tem-erature from 37 to 56°C causes a significant increase in the rate of L-aspartic acid biosynthesis in short-term experiments. However, after 24, 49 and 72 hours, the cell activity has decreased to 75%, 72% and 69% respectively. The thermal instability of the immobilized L-aspartic acid producing E. coli cells was described earlier (Chibata et al., 1974; Tosa et al., 1974). The aspartase instability in E. coli cells has been described in patents (Mukouyama et al., 2000; Mukouyama et al., 2001). The thermal stability of biocatalysts was tested at 37 and 40°C in long-term experiments over 603 hours. The activity of the biocatalyst was roughly equal at both temperatures and after 25 days, had decreased by only about 3–5%. In comparison with published data (Chibata et al., 1974; Tosa et al., 1974) immobilized recombinant E. coli P1 cells show a higher thermal stability, and their maximum activity was attained at a temperature approximately 4°C higher in our studies.

Substrate concentration is another important pro-cess parameter influencing both the reaction rate and product concentration. In our study, the optimum ammonium fumarate concentration was 1.0 M what was agreement with other publications. Increases of substrate concentration to 1.2 and 1.5 M decreased the efficiency of the continuous process and caused additional biocatalyst inactivation of about 8–13% and 13–28% respectively.

In accordance with literature (Mukouyama and Komatsuzaki, 2000; Mukouyama and Komatsuzaki, 2001) the process of L-aspartic acid biosynthesis is profitable for a conversion rate over 90%, which was obtained in our experiment at a flow rate of rz<162 ml/h for bioreactor A. In those conditions, the productivity of L-aspartic acid, defined as mass of product per cell dry mass in biocatalyst per time unit, was 4.5 g/g/h. For a description of the efficiency of the continuous L-aspartic acid production process, the parameter liq-uid hour space velocity (LHSV, i.e. ml/ml/h) or sim-ply space velocity (SV) was proposed by Mukouyama and Komatsuzaki (2001). It is defined as the volume of feeding substrate passed per volume of catalyst bed in bioreactor per one hour. Analyzing the LHSV, it is easy to compare continuous processes conducted in different bioreactors. The maximum conversion yield obtained for the process in bioreactor A was 97% for a substrate solution flow rate of 82 ml/h through the bioreactor. However LHSV for these conditions was 1.9 only. A dramatic decrease in the conversion ratio was observed in bioreactor A at LHSV over 3.1 and in bioreactor B at LHSV over 1.0 (Fig. 7).

As in both experiments for different biocatalyst vol-umes in bioreactor A and B, the same ratio Hz/Vz = 0.35 was maintained, in the next experiment bioreactor C was used with Hz/Vz = 3.1. For the liquid hour space velocity = 5.2 a maximum conversion ratio of over 99% and productivity of 6 g/g/h in this bioreactor was achieved. The increased substrate solution flow rate through bioreactor C resulted in the same decrease of conversion rate as in bioreactors A and B, however with a lower ratio. A conversion rate of about 90% was obtained even for the high LHSV value of about 10 (Fig. 7).

Generally, the best result of the L-aspartic acid bio-synthesis process was achieved in bioreactor C, which has the lowest volume of biocatalyst (1.3 ml). For the whole range of the tested flow rates, the substrate to product conversion ratio did not decrease below 50%. By extrapolating the experimental results, it is estimated that a conversion level equal to 50% can be achieved at LHSV values up to about 40 (Fig. 7). Analysis of the relationships between conversion ratio and LHSV and between bioreactor productivity (calculated by biomass unit) and LHSV shows also the importance of the rela-


tionship Hz/Vz in bioreactors (Fig. 7 and 8). The higher this relationship, the more effective L-aspartic acid biosynthesis is in continuous process. The substrate to product conversion rate of over 90% in bioreactors B and A with similar Hz/Vz = 0.35, was achieved with the L-aspartic acid productivity below 3.4 g/g/h and below 5.0 g/g/h respectively. In bioreactor C with Hz/Vz = 3.1, i.e. 9 times higher than in bioreactors A and B, this yield limit (90%) was achieved with a significant higher pro-ductivity of nearly 10.0 g/g/h. These results correspond to the literature data of Tosa et al. (1973) and Mukouy-ama and Komatsuzaki (2001) who reported a positive correlation between bioreactor column height (length) and L-aspartic acid formation efficiency.

Conclusions

For the development of L-aspartic acid biosynthesis technology, the new high yielding recombinant E. coli P1 was constructed (Gadomska et al., 2007). In this paper the new immobilization method of the E. coli P1 cells in chitosan gel was standardized. It was the starting point for laboratory process elaboration using column bioreactors characterized by different volumes and different height to volume ratios. Optimization of the biosynthesis process was based on both these parameters and the substrate flow rate through the bio-reactors. Two factors were critical for the bioreactor achieving optimum production capacity: Firstly, suf-ficient preliminary activation of the biocatalyst through its incubation in substrate solution before its use for biosynthesis process, and secondly, optimal substrate flow rate through biocatalyst bed during the continuous process of L-aspartic acid. Two parameters, i.e. the bio-catalyst bed height to volume ratio (Hz/Vz) and liquid hour space velocity (ml/ml/h) were used as essential criteria for the process optimization, i.e. maximum of

yield (%) and maximum productivity, defined as mass of product per mass of biocatalyst per time unit (g/g/h). In bioreactors A and B with an Hz/Vz of 0.35, the maxi-mum substrate to product conversion rate was 95.7% and productivity 1.7 g/g/h with an LHSV of 1.2 for bio-reactor A, and conversion over 99% and productivity 0.4–1.6 g/g/h with an LHSV of 0.3–1 for bioreactor B. However in bioreactor C with a Hz/Vz of 3.1, i.e. 9 times higher than in bioreactors A and B, the maximum yield was 99.8%, with significant higher productivity, 6 g/g/h, for a much higher LHSV of 5.2 have been achieved. In conclusion, it is clear that the height to volume ratio of a biocatalyst bed (and column bioreactor) is a key factor in the process development of L-aspartic acid biosynthesis, and the liquid hour space velocity (LHSV) is an important index of the optimization procedure.

AcknowledgementsThe authors are grateful to George Gasyna and Edward Lowc-

zynski for helpful writing this paper.

Literature

Chibata I., T. Tosa and T. Sato. 1974. Immobilized aspartase-con-taining microbial cells: preparation and enzymatic properties. Appl. Microbiol. 27: 878–885.Chibata I., T. Tosa and T. Sato. 1985. Aspartic acid. Comprehen-sive Biotechnology, vol. 3: 633–640, Moo-Young M. (ed.) Pergamon Press, Oxfort-New York.Fusee M.C., W.E. Swann and G.J. Calton. 1981. Immobilization of Escherichia coli cells containing aspartase activity with polyurethane and its application for L-aspartic acid production. Appl. Environ. Microbiol. 42: 672–676.Gadomska G., A. Płucienniczak and A. Chmiel. 2007. Recombi-nant strains of Escherichia coli for L-aspartic acid. Pol. J. Microbiol. 56: 77–82Kawabata N., S. Nishimura and T. Yoshimura. 1990. New method of immobilization of microbial cells by capture on the surface of insoluble pyridinium-type resin. Biotechnol. Bioeng. 35: 1000–1005.

Fig. 7. Effect of liquid hour space velocity of production through bioreactors A, B and C on ammonium fumarate conversion to

L-aspartic acid.

Fig. 8. Effect of liquid hour space velocity of production through bioreactors A, B and C on biocatalyst bed productivity.


Lee C. K. and J. Hong. 1988. Membrane reactor coupled with electro phoresis for enzymatic production of aspartic acid. Biotech-nol. Bioeng. 32: 647–654.Mukouyama M., Masaharu, Komatsuzaki S. and Satomi. 2000. Pro-cess for production of L-aspartic acid. United States Patent. 6: 150,142.Mukouyama M. and S. Komatsuzaki. 2001. Method for producing L-aspartic acid. United States Patent: 6: 214,589.Papierz M., G. Gadomska, B. Sobierajski and A. Chmiel. 2007. Selection and activation of Escherichia coli strains for L-aspartic acid biosynthesis. Polish Journal of Microbiology 56: 71–76.Sato T., T. Mori, T. Tosa and I. Chibata. 1975. Engineering analysis of continuous production of L-aspartic acid by immobilized Escheri-chia coli cells in fixed beds. Biotech. Bioeng. 17: 1797–1804.

Sato T., Y. Nishida, T. Tosa, and I. Chibata. 1979. Immobiliza-tion of Escherichia coli cells containing aspartase activity with κ-carrageenan: enzymic properties and application for L-aspartic acid production. Biochim. Biophys. Acta 570: 179–186.Tosa T., T. Sato, T. Mori, Y. Matuo and I. Chibata. 1973. Continious production of L-aspartic acid by immobilized aspartase. Biotechnol. Bioeng. 15: 69–84.Tosa T., T. Sato, T. Mori and I. Chibata. 1974. Basic studies for continous production of L-aspartic acid by immobilized Escherichia coli cells. Appl. Microbiol. 27: 886–889.Umemura I., S. Takamatsu, S. Sato , T. Tosa and I. Chibata. 1984. Improvement of production of L-aspartic acid using immobilized microbial cells. Appl. Microbiol. Biotechnol. 20, 291–295.


ORGINAL PAPER

Introduction

Cyclodextrins (Cd) are cyclic nonreducing oligosac-charides composed of α-1,4-linked glucose units, which are designated α, β and γ, according to the number of glucose units (6, 7 or 8, respectively). Cd molecules possess a hydrophobic inner cavity, in which hydro-phobic compounds can be incapsulated. As a result, the properties of the guest molecules are altered. This abil-ity of Cd determines their broad application in differ-ent areas of industry. They are applied in food industry for removal of unwanted flavour and aroma, for pro-tection of guest molecules from degradation under the action of light and heat, for reduction of side effects of drug formulations, for improvement of water solubil-ity of insoluble compounds, for stabilization of volatile substances, etc. (del Valle, 2004).

Cd are produced by enzyme conversion of starch with cyclodextrin glucanotransferase (CGTase, 2.4.1.19). CGTase is a unique enzyme produced only by micro-organisms, usually Bacillus species (Tonkova, 1998). All known CGTases form the three types of Cd, but in different ratio (Leemhuis et al., 2010; Qi and Zimmermann, 2005). According to the predominant type of Cd formed they are classified as α-, β- and

γ-Cd. The product specificity of CGTases determines their application for production of certain type of Cd.

The yield and ratio of Cd depend on the proper-ties of CGTase, kind of substrate used (Alves-Prado et al., 2008), its preliminary treatment (Goh et al., 2007; Pishtiyski and Zhekova, 2006; Sakinah et al., 2009) and reaction conditions (Martins and Hatti-Kaul, 2003; Matioli et al., 2001). A great number of reports, descri-bing the effect of the reaction conditions on the produc-tion of the predominant type of Cd and determination of the optimal parameters of the process, are available (Charoenlap et al., 2004; Gawande and Patkar, 2001; Rauf et al., 2008; Szerman et al., 2007). However, there is no data for determination of the conditions for maxi-mal production of certain type of Cd at the terms of minimal amounts of concomitant types of Cd. This is a subject of a certain interest, as the presence of several types of Cd in the reaction mixture requires separation and purification of the desired product.

On the other hand the enzyme reaction can be directed to formation of a certain type of Cd by selec-tion of proper conditions. Additionally this fact allows the enzyme to be used not only for production of the predominant type of Cd, but also for production of the concomitant types of Cd. The enzyme from Bacillus

Reaction Conditions for Maximal Cyclodextrin Production by CyclodextrinGlucanotransferase from Bacillus megaterium

BORIANA Y. ZHEKOVA1* and VESELIN S. STANCHEV2

1 department of Biochemistry and Molecular Biology, University of Food Technologies, Plovdiv, Bulgaria2 department of Automatics, Information and Control Systems, University of Food Technologies

Plovdiv, Bulgaria

Received 14 September 2010, revised 7 April 2011, accepted 15 April 2011

A b s t r a c t

The effect of the reaction conditions (substrate concentration, enzyme dosage, and pH) on cyclodextrin production by cyclodextrin glucanotransferase from Bacillus megaterium was investigated by applying mathematical modeling methods. Adequate models were devel-oped and they were used for determination of the optimal conditions for maximal formation of β-cyclodextrins at minimal concentrations of α- and γ-cyclodextrins. The main factor affecting the ratio of the products was pH of the reaction mixture. At pH 9 the enzyme formed mainly β- and γ-cyclodextrins and the ratio α:β:γ was 2.6:83.5:13.9; at pH 5 the ratio changed to 8.6:84.6:6.8. Mathematical models were used for determination of the conditions for maximal conversion of the substrate into cyclodextrins. 45.88% conversion of starch was achieved at 5% substrate concentration, 3.5 U/g enzyme dosage, and pH 7.4.

K e y w o r d s: cyclodextrins, cyclodextrin glucanotransferase, mathematical modeling, Bacillus megaterium

* Corresponding author: B.Y. Zhekova, 26 Maritza Boulevard, 4002 Plovdiv, Bulgaria; phone: (+359) 32 603 605; fax: (+359) 32 644 102; e-mail: [email protected]

Zhekova B.Y. and Stanchev V.S. 2114

megaterium used in this study formed β-Cd as the pre-dominant product of the cyclization reaction (Zhekova et al., 2008; Zhekova et al., 2009).

The aim of the current research was establishment of an optimal working point for the process of Cd pro-duction, which determines maximal amount of β-Cd at the conditions of minimal presence of α- and γ-Cd, and determination of the conditions for main formation of α- and γ-Cd and maximal conversion of starch into Cd.

Experimental

Material and Methods

Substrate and enzyme preparation. The substrate for Cd production was corn starch obtained from Amilum. The enzyme preparation used was a crude CGTase from B. megaterium (from the collection of the department of Biochemistry and Molecular Biology, University of Food Technologies, Plovdiv) with activity of 2.54 U/ml. The cultivation of the strain and biosyn-thesis of the enzyme were performed as described in previous research (Pishtiyski et al., 2008).

Enzyme reaction. Substrate solutions were prepared in citrate-phosphate buffer (pH 5.0–9.0), in a manner allowing the desired concentration to be reached after the addition of the enzyme preparation. The substrate was gelatinized in a steam water bath for 10 min, cooled to 30°C and the necessary amount of CGTase was added. The enzyme reaction was conducted in 100-ml Erlen-meyer flasks containing 50 ml of reaction medium, at 30°C on a reciprocal shaker for 10 h. CGTase was inac-tivated by boiling for 10 min in a water bath and the content of α-, β- and γ-Cd formed was determined.

Experimental design. The effect of starch concen-tration, CGTase dosage and pH on cyclodextrin pro-duction was studied by using optimal composite design with distance from the centre of the design space to a factorial point ±1 (Mason et al., 2003).

The static of the process was described by using nonlinear mathematical models of the type:

med using ANOVA. For determination of the function maxima MATLAB 6.0 was used.

Assays. CGTase activity was determined by Kestner’s method (Kestner et al., 1989) as described previously (Zhekova et al., 2008).

The concentration of α-Cd was determined by the method with methyl orange (Lejeune et al., 1989), of β-Cd – with phenolphthalein (Kestner et al., 1989), of γ-Cd – with bromcresol green (Kato and Horikoshi, 1984).

Cd concentrations were confirmed by HPLC sys-tem Shimadzu 20 AHT with refractive index detector. For estimation of α-, and γ-Cd YMC-Pack OdS-AQ column (YMC Europe) was used. The mobile phase was a mixture of methanol and water in a ratio of 3:97 with a flow rate of 1.3 ml/min, and the column tem-perature was 30°C. β-Cd were analyzed on Ultrahydro-gel column (Waters), at 30°C, and bidestiled water as a mobile phase with a flow rate of 0.8 ml/min.

Results and Discussion

The effect of the investigated factors in the varia-tion intervals, presented in Table I on α-, β- and γ-Cd production was analyzed by using nonlinear regression equations (1).

The optimal composite design, the experimental data for α-, β- and γ-Cd (designated Yα, Yβ and Yγ respectively) and the corresponding predicted values (Ŷα, Ŷβ, Ŷγ) are presented in Table II.

The following regression equations were obtained after removal of the insignificant terms:

Ŷα = 1.703 + 0.206.x1 + 0.19.x2 – 0.376.x3 – 0.16.x13 – – 0.13.x23 – 0.42.x1

2 – 0.3.x22 – 0.12.x3

2 (2)

Ŷβ = 12.571 + 5.234.x1 + 3.138.x2 + 1.669.х12 – – 1.693.x1

2–1.973.x22 (3)

Ŷγ = 1.832 + 0.521.x1 + 0.332.x2 + 0.716.x3 + + 0.156.x12 + 0.099.x13 – 0.249.x2

2 – 0.249.x32 (4)

The analysis of variance (Table III) showed that the regression equations are statistically significant at 95% confidence level.

The extremums of equations (2), (3) and (4) are pre-sented in Table IV.

It was noticed that they were achieved at different values of the independent variables. This confirmed the

(1)

where the variable Ŷ is the predicted response, xi and xj are the independent variables, β0 is the offset term, βi is the linear effect, βij is the interaction effect, βii is the squared term and k is the number of the independ-ent variables.

Real and coded values of the independent variables and their variation intervals are presented in Table I.

The test for statistical significance of the regression coefficients and the models developed was perfor-

Independent variable –1 0 +1x1 – starch (%) 1 3 5x2 – CGTase (U/g ) 0.5 2 3.5x3 – pH 5 7 9

Table IVariation intervals of the independent variables

Cyclodextrin production by cyclodextrin glucanotransferase from B. megaterium2 115

hypothesis that the ratio of α-, β- and γ-Cd can be con-trolled by a change in the reaction conditions.

The mathematical models obtained can be inter-preted in several aspects. It is of great interest to deter-mine the working conditions at which equations (2), (3) and (4) achieved their maximal value. This allows the models to be used for selection of conditions at which maximal amount of a certain Cd type is produced at beforehand known concentration of other types of Cd.

The enzyme used in this study formed mainly β-Cd. For these reasons the models developed were inter-preted in the respect of production of maximal amount of β-Cd at minimal concentration of α- and γ-Cd.

Minimal values of Ŷα and Ŷγ were achieved at equal levels of the first and second independent variable x1 = –1 and x2 = –1 (Table IV). However, under these conditions Ŷβ also reached its minimum. Consequently, these two factors could not be used for control of the

1 –1 –1 –1 0.38 0.55 2.98 2.20 0.15 0.02 2 –1 –1 1 0.32 0.38 2.46 2.20 1.20 1.25 3 –1 1 –1 1.24 1.19 5.30 5.14 0.32 0.37 4 –1 1 1 0.40 0.50 5.52 5.14 1.61 1.61 5 1 –1 –1 1.48 1.28 10.48 9.33 0.66 0.55 6 1 –1 1 0.52 0.47 8.86 9.33 2.07 2.18 7 1 1 –1 1.83 1.92 17.29 18.95 1.42 1.53 8 1 1 1 0.61 0.59 20.78 18.95 3.14 3.16 9 –1 0 0 1.36 1.08 4.07 5.64 1.27 1.31 10 1 0 0 1.32 1.49 15.26 16.11 2.47 2.38 11 0 –1 0 1.20 1.21 5.75 7.46 1.18 1.25 12 0 1 0 1.72 1.59 13.02 13.74 2.09 1.92 13 0 0 –1 1.98 1.96 11.56 12.57 0.79 0.87 14 0 0 1 1.30 1.21 11.25 12.57 2.48 2.33 15 0 0 0 1.70 1.70 14.88 12.57 1.61 1.83 16 0 0 0 1.78 1.70 15.38 12.57 1.51 1.83 17 0 0 0 1.52 1.70 12.81 12.57 2.06 1.83 18 0 0 0 1.58 1.70 11.97 12.57 1.97 1.83

Table IIOptimal composite design for three factors and three levels of their variation

α-Cd (mg/ml) γ-Cd (mg/ml)β-Cd (mg/ml)№

ŶγYγŶβYβŶαYα

x3x2x1

Parameter df SS MS Fsign df SS MS Fsign df SS MS Fsign

Regression 8 4.83 0.60 8.7E-5 5 441 88.2 1.2E-6 7 10.1 1.44 1.9E-6Residual 9 0.28 0.03 12 31.5 2.6 10 0.37 0.04Total 17 5.11 17 473 17 10.4

Table IIIStatistical analysis results according to Anova

Equation (2) Equation (3) Equation (4)

df – degree of freedom; SS – Sum square; MS – Mean square; Fsign – F significance

Value (mg/ml) 0.381 2.12 2.20 18.95 0.02 3.16Conditions: x1 –1 0.436 –1 1 –1 1 x2 –1 0.533 –1 1 –1 0.98 x3 1 -1 –1÷1 –1÷1 –1 1

Table IVExtremum of equations (2), (3) and (4)

ExtremumEquation (4)Equation (3)Equation (2)

Ŷαmin Ŷγ

maxŶγminŶβ

maxŶβminŶα

max


ratio of α-, β- and γ-Cd. This result is normal taking into consideration the fact that increase in substrate and enzyme concentrations lead to enhancement of the product yield in enzyme reactions.

With regard to the third independent variable (pH of the reaction mixture) there were significant differ-ences at the values at which Ŷα and Ŷγ reached their minimums. Ŷα

min = 0.38 mg/ml was achieved at x3 = 1, and Ŷγ

min =0.02 mg/cm3 – at x3 = –1 (Table IV). On the other hand formation of β-Cd did not depend on pH of reaction mixture – the concentration of β-Cd was max-imal at the whole investigated interval of pH (5.0–9.0). These results allowed the process for Cd production to be performed at reaction conditions which ensured for-mation of only two types of Cd (α and β) or (γ and β).

The minimal values of α- and γ-Cd concentrations were formed at the lowest levels of x1 and x2, which also had a negative effect on β-Cd. For these reasons the values of the functions were calculated at the highest levels of x1 and x2, and variation of x3.

When the process was performed at x1 = 1, x2 = 1 and x3 = 1, the predicted concentrations of Cd accord-ing to equations (2), (3) and (4) were as followed: Ŷα = 0.59 mg/ml, Ŷβ = 18.95 mg/ml and Ŷγ = 3.16 mg/ml. Since at these conditions the concentration of α-Cd was only 3.1% in regard to β-Cd concentration, they were considered as optimal for production of β-Cd in the absence of α. These results were confirmed by perfor-ming 4 parallel experiments and the following mean val-ues of α-, β- and γ-Cd were registered: Yα

exp=0.56 mg/ml, Yβ

exp = 18.79 mg/ml, Yγexp = 3.13 mg/ml. A test for equal-

ity of the mathematical expectation of the experimen-tal results and the predicted data was conducted. As the values of tcalc (tα, calc= 1.058, tβcalc = 0.392, tγ, calc = 0.821 respectively) were lower than tcrit = 3.182 at 95% signifi-cant level and degree of freedom 3, it was established

that there was no statistically significant difference between the experimental and predicted results.

It can be concluded that at starch concentration of 5.0%, enzyme dosage 3.5 U/g and pH 9 CGTase from Bacillus megaterium formed mainly β- and γ-Cd and the ratio α:β:γ was 2.6:83.5:13.9. At these conditions α-Cd were only 2.6% of the total Cd amount.

For determination of the conditions at which the enzyme formed maximal amount β-Cd at minimal γ-Cd concentration, the predicted values of equations (2), (3) and (4) at x1 = 1, x2 = 1, x3 = –1 were calculated. These were Ŷα = 1.92 mg/ml, Ŷβ = 18.95 mg/ml and Ŷγ = 1.53 mg/ml, respectively. The experimental results (mean value of 4 experiments) under these conditions were as follows: Yα

exp=1.87 mg/ml, Yβexp=18.21 mg/ml,

Yγexp = 1.48 mg/ml. No statistical difference was obser-

ved between the experimental and predicted data (tα, calc = 1.475, tβcalc = 2.039 and tγ, calc = 1.65 were lower than tcrit = 3.182). The ratio of the three types of Cd was 8.6:84.6:6.8. There was no significant decrease in γ-Cd concentration and its percent ratio was close to the value of α-Cd. The analysis of (2), (3) and (4) showed that the amount of γ-Cd can be decreased only by a change in x1 and x2. However, these two fac-tors influenced the concentration of β-Cd in a similar way. For example at the lowest levels of the factors the enzyme did not formed γ-Cd, but the concentration of the main product was only 2.20 mg/ml (Table IV). This was probably due to the characteristic feature of CGTase from B. megaterium to form β- and γ-Cd in a certain ratio independently of the reaction conditions.

The established considerations are presented graphi-cally in Fig. 1. It can be seen that formation of β-Cd does not depend on pH in the interval from 5.0 to 9.0. The ratio of the other two types of Cd was significantly influenced by this factor. At low values of pH CGTase formed predominantly α-Cd, and at high values of pH it produced γ-Cd. A change in the ratio of Cd depend-ing on pH of the reaction mixture was reported for other CGTases as well (Atanasova et al., 2009; Martins and Hatti-Kaul, 2003).

The mathematical models developed can be used also for determination of the conditions in which CGTase formed maximal amount of α- and γ-Cd, which are not the predominent product of the reaction. This applica-tion of the models is not only important in the theoreti-cal aspect. It has a practical impact if the process for Cd production is performed with the aim of maximal conversion of the substrate in Cd, independently of the ratio of the products.

The effect of starch and CGTase concentrations on α- and γ-Cd production at optimal pH is presented in Fig. 2. data about β-Cd was not included, since their production was investigated in a previous research (Zhekova et al., 2008).

Fig. 1. Effect of CGTase concentration and pH on Cd production (5.0% starch).

Cyclodextrin production by cyclodextrin glucanotransferase from B. megaterium2 117

The increase in starch concentration led to an enhan - cement of the two types of Cd. With regard to α-Cd (Fig. 2a) a presence of a saturation substrate concen-tration (3.44%) was observed. This may be due to sub-strate or product inhibition of α-Cd forming activity of the enzyme. In the case of γ-Cd production no satu-ration of the substrate was reached in the interval of the experiment.

CGTase concentration had an effect on α-Cd for-mation up to 2.8 U/g, and it influenced γ-Cd yield up to 3.47 U/g. In a previous research a similar depend-ence was established for β-Cd. The reason for this fact was found to be product inhibition of CGTase by β-Cd (Zhekova et al., 2008). Probably α- and γ-Cd forming activities of the enzyme were also inhibited by the cor-responding type of product. Similar results were estab-lished for other CGTases (Gawande and Patkar, 2001; Tomita et al., 1990).

Maximal sum of the concentrations of the three types of Cd 22.94 mg/ml (Ŷ1 = 1.24 mg/ml, Ŷ2 = 18.95 mg/ml и Ŷ3 = 2.75 mg/ml) was achieved under the following conditions x1 = 1, x2 = 1 and x3 = 0.202. This concentra-

tion corresponded to 45.88% conversion degree of starch into Cd, which was a good yield taking into account that it was obtained without the use of organic solvents.

The mathematical models developed were used for working out a regression equation for the effect of the factors on the conversion degree of substrate in Cd (5):

%conversion = 56.43–10.97.x1+14.163.x2+2.06.x3– –3.68.x1.x2+2.99.x1

2–11.8.x22–4.66.x3

2, (5)

Maximal conversion of starch 77.19% was achieved at x1 = –1, x2 = 0.753 и x3 = 0.221. However, under these conditions the sum of the concentrations of the three types of Cd was only 8.03 mg/ml. This was probably

due to the low level of substrate concentration. In a pre-vious work it was established that increase in starch concentration led to enhancement of concentration of Cd, but decreased the conversion degree. The reason for this was the product inhibition of CGTase by β-Cd (Zhekova et al., 2008).

Conclusions

Adequate mathematical models for the effect of starch concentration, CGTase dosage and pH on α-, β-, and γ-Cd production were developed. They were suc-cessfully used for determination of the conditions, in which CGTase formed maximal amount of β-Cd, at possibly minimal concentration of α- and γ-Cd. The ratio of the product can be controlled by change in the pH of the reaction mixture. Another application of the models was determination of the conditions for maximal conversion of starch into Cd. This approach led to 45.88% degree of substrate conversion, which was a good yield for a process without organic solvents.

Literature

Alves-Prado H.F., A.A.J. Carneiro, F.C. Pavezzi, E. Gomes, M. Boscolo, C.M.L. Franco and R. da Silva. 2008. Production of cyclodextrins by CGTase from Bacillus clausii using different starches as substrates. Appl. Biochem. Biotechnol. 146: 3–13.Atanasova N. , T. Kitayska, D. Yankov, M. Safarikova and A. Tonkova. 2009. Cyclodextrin glucanotransferase production by cell biocatalysts of alkaliphilic bacilli. Biochem. Eng. J. 46: 278–285.Charoenlap N., S. Dharmsthiti, S. Sirisansaneeyakul and S. Lertsiri. 2004. Optimization of cyclodextrin production from sago starch. Bioresource Technol. 92: 49–54.Del Valle E. 2004. Cyclodextrins and their uses: a review. Process Biochem. 39:1033–1046.

Fig. 2. Effect of starch (x1) and CGTase (x2) concentration on production of (а) α-Cd and (b) γ-Cd at optimal pH values(pH 5.0 and pH 9.0 respectively).


Gawande B. and A. Patkar. 2001. α-Cyclodextrin production using cyclodextrin glycosyltransferase from Klebsiella pneumoniae AS-22. Starch/Stärke 53: 75–83.Goh K.M, N.M. Mahadi, O. Hassan, R.N.Z.R.A. Rahman and R. Md. Illias. 2007. The effects of reaction conditions on the pro-duction of γ-cyclodextrin from tapioca starch by using a novel recombinant engineered CGTase. J. Mollecul. Catal. B: Enzymatic 49: 118–126.Kato T. and K. Horikoshi. 1984. Colorimetric determination of γ-cyclodextrin. Analyt. Chem. 54: 1738–1740.Kestner A., R. Vokk, E. Papel and A. Papeman. 1989. determina-tion of cyclodextrin glucanotransferase activity. Prikladnaja Bio- chimia i Microbiologia, 25: 425–430 (in Russian).Leemhuis H., R.M. Kelly and L. Dijkhuizen. 2010. Engineering of cyclodextrin glucanotransferases and the impact for biotechnologi-cal applications. Appl. Microbiol. Biotechnol. 85:823–835.Lejeune A., K. Sakaguchi and T. Imanaka. 1989. A spectrophoto- metric assay for the cyclization activity of cyclomaltohexaose (α-cyclodextrin) glucanotransferase. Analyt. Biochem. 181: 6–11.Martins R. and R. Hatti-Kaul. 2003. Bacillus agaradhaerens LS-3C cyclodextrin glycosyltransferase: activity and stability features. Enzyme Microbial Technol. 33: 819–827.Mason R., R. Gunst and J. Hess. 2003. Statistical design and analy-sis of experiments with applications to engineering and science, John Wiley & Sons.Matioli G., G. Zanin and F. De Moraes. 2001. Characterization of cyclodextrin glycosyltransferase from Bacillus firmus strain No 37. Appl. Biochem. Biotechnol. 91–93: 643–654.Pishtiyski I. and B. Zhekova. 2006. Effect of different substrates and their preliminary treatment on cyclodextrin production. World J. Microbiol. Biotechnol. 22: 109–114.

Pishtiyski I., V. Popova and B. Zhekova. 2008. Characterization of cyclodextrin glucanotransferase produced by Bacillus megaterium. Appl. Biochem. Biotechnol. 144 (3): 263–272.Qi Q. and W. Zimmermann. 2005. Cyclodextrin glucanotrans-ferase: from gene to applications. Appl. Microbiol. Biotechnol. 66: 475–485.Rauf Z.A., R.Md. Illias, N.M. Mahadi and O. Hassan. 2008. Experimental design to optimization of beta cyclodextrin produc-tion from ungelatinized sago starch. Eur. Food Res. Technol. 226: 1421–1427.Sakinah A.M.M., A.F. Ismail, O. Hassan, A.W. Zularisam and R.Md. Illias. 2009. Influence of starch pretreatment on yield of cyclodextrins and performance of ultrafiltration membranes. Desali-nation 239: 317–333.=Szerman N., I. Schroch, A.L. Rossi, A.M. Rosso, N. Krymkiewicz and S.A. Ferrarotti. 2007. Cyclodextrin production by cyclodex-trin glycosyltransferase from Bacillus circulans dF 9R. Bioresource Technol. 98: 2886–2891.Tomita K., T. Tanaka, Y. Fujita and K. Nakanishi. 1990. Some factors affecting the formation of γ-cyclodextrin using cyclodextrin glycosyltransferase from Bacillus sp. AL 6. J. Ferment. Bioeng. 70: 190–192.Tonkova A. 1998. Bacterial cyclodextrin glucanotransferase. Enzyme Microbial Technol. 22: 678–686.Zhekova B., I. Pishtiyski and V. Stanchev. 2008. Investigation on cyclodextrin production with cyclodextrin glucanotransferase from Bacillus megaterium. Food Technol. Biotechnol. 46: 328–334.Zhekova B., G. Dobrev, V. Stanchev and I. Pishtiyski. 2009. Approaches for yield increase of β-cyclodextrin formed by cyclodex-trin glucanotransferase from Bacillus megaterium. World J. Micro-biol. Biotechnol. 25: 1043–1049.


ORGINAL PAPER

Introduction

Biofouling is a major problem for maritime opera-tions such as shipping, off shore oil mining, coastal power generation, marine electronics, mariculture, marine construction or naval operations (Armstrong et al., 2000). Generally, industries like coastal power plants and desalination plants face biofouling problem due to the microbial growth and other higher organ-isms. The most common biofouling control method in industrial cooling water systems is chlorination (Yebra et al., 2004). However there is an increasing pressure to reduce or eliminate the use of chlorine due to the production of by-products. The use of chemo-biocides to control biofouling forms the major contaminant of marine environment. The chemical antifouling agents applied on the industrial objects, kill not only the foulers, but also have negative effect on other benthic community (Fingerman, 1988). The hazards of heavy metals in marine environment include their high toxi-city, circulation in food chain and bioaccumulation.

The biofouling process involves various steps, from the initial conditioning of the surface by organic and inorganic molecules to the colonization by microorga-nisms and leading to the establishment of biofilm. There has been a growing interest in biofilms due to their significance in environmental, industrial and medi cal areas.

Since, biofilm formation on technical objects submerged in aquatic environments is a major problem with huge economic loss, there is a need to formulate adequate eco- friendly control measures. Flemming (1991) suggested that UV and ultrasound waves would be a potential source to minimize biofouling in industrial units. There are also reports on the inhibitory effect of UV on macro- fouling organisms and is also effective in cleaning previ-ous fouled surfaces (Zelver et al., 1981). UV radiation does not lead to any large-scale accumulation of toxic by-products in the ecosystem. Ultraviolet light is an estab-

lished and increasingly popular alternative to chemicals for the disinfection of drinking water, wastewater, and industrial waters of various qualities. Ultraviolet radia-tion (10–400 nm wavelength) is of special interest because it is used in certain environments (e.g. hospital operat-ing rooms) to kill microorganisms. Ultraviolet light is that portion of the electromagnetic spectrum that lies between x-rays and visible light. Four regions of the UV spectrum have been defined – vacuum UV between 100 and 200 nm, UV-C between 200 and 280 nm, UV-B between 280 and 315 nm, and UV-A between 315 and 400 nm. Practical application of UV disinfection relies on the germicidal ability of UV-C and UV-B.

The effect of UV light on the microbial community is widely studied by various investigators (Chang et al., 1985; Harris et al., 1987). The efficiency of UV treatment

Efficacy of UV Treatment in the Managementof Bacterial Adhesion on Hard Surfaces

A. KOLAPPAN and S. SATHEESH*

Centre for Marine Science and Technology, Manonmaniam Sundaranar University,Rajakamangalam Tamil Nadu, India

Received 8 May 2010, revised 22 december 2010, accepted 8 January 2011

A b s t r a c t

The efficacy of UV treatment to control bacterial adhesion onto hard surfaces was investigated in laboratory conditions. The major characteristics necessary for biofilm formation like extracellular polymeric substance (EPS) production, carbohydrate and protein concentration in EPS, and adhesion ability onto hard surface were studied using two bacterial strains isolated from marine biofilms. The results showed that there was a considerable difference between the control and UV treated bacterial cultures in their viability, production of EPS, and adhesion ability. The protein and carbohydrate concentration of the EPS and the adhesion of bacterial cells to surface were also considerably reduced due to UV treatment. This study indicates that treatment of water with UV light may be used to control biofilm development on hard surfaces.

K e y w o r d s: biofilm, adhesion, antifouling, extracellular polymeric substance (EPS)

* Corresponding author: S. Satheesh, Centre for Marine Science and Technology, Manonmaniam Sundaranar University, Marina campus, Pannaiyoor, Rajakamangalam-629502, Kanyakumari district, Tamil Nadu, India; e-mail: [email protected]

Kolappan A. and Satheesh S. 2120

to control the biofilm formation and biofouling com-munity development was studied by Munshi et al. (1999; 2001; 2005), Sharrer et al. (2007) and Wenjun and Wenjun (2009). However, there is a lack of informa-tion on the response of bacterial community to UV treatment, particularly the biofilm forming character-istics like EPS production and adhesion ability. There is also a scarcity of information on the effectiveness of the UV irradiation of incoming water for preventing microorganisms depositing on the surface. Hence, in the present study, an attempt has been made to evaluate the efficacy of UV treatment to control biofilm forma-tion. The main objective was to investigate the effect of UV light on the EPS production and adhesion ability of marine bacteria involved in biofilm formation.

Experimental


Biofilm development assay. Two bacterial cultures, Alteromonas sp. (SS03) and Pseudomonas sp. (SS04) maintained in our laboratory were used for the pre-sent study. These bacteria were originally isolated from the marine biofilm developed on hard surfaces and tentatively identified based on the biochemical char-acteristics. A loop full of pure culture from the slant was inoculated into 100 ml Zobell marine broth taken in 250 ml conical flasks. The flasks were incubated for 24 h at room temperature. An aliquot of the broth was taken on microscopic slides to enumerate the number of bacterial cells present in per millilitre of the broth.

500 ml glass beakers were filled with 300 ml sterile seawater (Millipore filtered and autoclaved) and 10 ml of bacterial culture (approx. 106 cells ml–1) was added. This bacterial culture introduced into seawater medium was exposed to UV light for 10 minutes (UV-C, TUV 30W/G30) in a laminar airflow chamber. The distance between the beakers with seawater medium and UV lamb was maintained at 15 cm. After UV light treatment, the beakers were covered with parafilm and transferred into a sterile chamber. Experimental set-ups prepared as above without UV treatment were considered as con-trols. Five glass slides (7.5×2.5 cm) were placed inside the beakers in slanting position as a substratum for biofilm development. The slides were removed from the beaker after 1, 2, 3, 4 and 5 hours of immersion. The slides were then air-dried, heat fixed and stained with methylene blue. The number of bacteria adhered to the slides were counted under a binocular microscope. The experiment was replicated (N = 6) and the mean values were taken. One-way ANOVA (analysis of variance) was used to evaluate the effect of ultraviolet treatment on the adhesion of bacteria on hard surface.

Alternatively, separate experiments were carried out to assess the viability of bacterial cells in the biofilms. For this, the glass coupons were incubated in the bacte-rial culture introduced medium (UV treated and control) for 24 hours at room temperature. After incubation, the coupons were removed from the beaker and rinsed with sterile seawater to remove the unattached organisms. The biofilm developed on the coupon was scrapped off using a sterile nylon brush (each coupon was analysed separately) and dispersed in to 1 ml sterile seawater (Millipore filtered and autoclaved). This biofilm sam-ple was used for the isolation of extracellular polymeric substance and to assess the viability of bacterial cells.

Viable cell counts. In order to enumerate viable counts of bacterial population in both UV treated and control cultures, an aliquot of the biofilm sample isola-ted from the coupons was serially diluted using sterile seawater. The appropriate dilutions were spread on Zobell marine agar plates. The plates (n = 3) were incubated at room temperature for 24 hours, and colonies were counted manually.

Estimation of extracellular polymeric substance (EPS). The amount of EPS produced by the bacteria isolated from the both UV treated and control experi-mental set-up was analysed by estimating the total car-bohydrate and protein concentration. For this, 5 ml of the biofilm sample obtained as above was centrifuged at 10,000 g for 10 min at 4°C. The cell pellets was discarded and the supernatant was mixed with equal amount of cold absolute ethanol. The precipitated EPS was diluted to known volume with distilled water and stored at 4°C. Carbohydrate was estimated by Phenol Sulphuric acid method using glucose as standard (dubois, 1956). The total protein content of EPS produced by the cultures was estimated by the Lowry et al. (1951) method, using Bovine serum albumin as the standard.

Characterization of EPS by thin-layer chromatography.The EPS isolated from the UV treated and con-trol experiments was characterized by thin-layer chro-matography (TLC). The EPS was loaded on a silica gel plate. n-butanol, acetic acid and distilled water (2:1:1) were used as the solvent system for TLC. Iodine crystals were used for the visualization of spots in the TLC plate.

Results

The number of Alteromonas sp. cells adhered on the coupons submerged in control medium was 3140 cm–2 (after 1 hour). The number of cells attached on the cou-pons submerged in UV treated medium was 1364 cm–2 after 1 hour. After five hours, the number of bacteria adhered on the coupons submerged in control medium was 8291 cells cm–2. The coupons submerged in UV treated medium showed a density of 1451 cells cm–2

Influence of UV on biofilm formation by marine strains2 121

after 5 hours of exposure (Fig. 1). The cell attachment assay revealed that the number of Alteromonas sp. adhered to glass surface was reduced significantly after treated with UV (one-way ANOVA, F = 16.16; d.f = 1, 9; P<0.05).

Similarly, the biofilm adhesion assay with Pseudo-monas sp. showed considerable variations between control and ultraviolet treated cultures (Fig. 2). After one hour of glass surface exposure 2033 cells cm–2 were observed in the control medium. The number of cells adhered on the coupons submerged in the UV treated medium was 1160 cells cm–2. At the end of the exposure period, 9233 cells cm–2 were observed in the control and 2395 cells cm–2 on the coupons immersed in UV treated medium. One-way ANOVA showed that the adhesion of Pseudomonas sp. to glass surface did not differ signif-icantly between the coupons submerged in UV treated and control medium (F = 3.49; d.f = 1,9; P>0.05).

The carbohydrate concentration of EPS of the Altero-monas sp. biofilm isolated from the coupons submerged in UV treated medium was 5.44 mg ml–1. The EPS of the Alteromonas sp. biofilm isolated from the cou-pons submerged in control medium was 7.34 mg ml–1

(Table I). The carbohydrate concentration of the EPS of Pseudomonas sp. isolated from the control medium coupons was 3.37 mg ml–1 and the coupons submerged in UV treated medium showed a value of 1.26 mg ml–1.

The protein concentration of the EPS produced by Alteromonas sp. from the UV treated medium was 4.01 mg ml–1 and in the control, the protein concentra-

tion was 5.23 mg ml–1 (Table I). The protein concentra-tion of EPS synthesized by the Pseudomonas sp. isolated from the coupons submerged in UV treated medium was 8.62 mg ml–1 and in the control, the concentration was 9.48mg ml–1. In general, results showed that the total protein concentration of the EPS was not reduced much due to the UV treatment.

The EPS synthesized by the bacteria adhered to coupons submerged in both UV treated and untreated systems were subjected to thin-layer chromatography to understand the changes in the carbohydrate composi-tion of the EPS. Results revealed that the carbohydrate composition of EPS was considerably changed due to UV treatment. The EPS synthesized by Alteromonas sp. adhered to coupons submerged in control medium showed three distinct spots on the thin layer chroma-togram. The EPS isolated from the Alteromonas sp. adhered on the coupons submerged in the UV treated medium showed two spots. Further, the EPS of the Alte-romonas sp. isolated from the coupons submerged in control and UV treated systems did not show any simi-larity in the RF values. Rf values of the EPS produced by the Alteromonas sp. control culture were 0.196, 0.312 and 0.651 cm. In UV treated Alteromonas culture, the Rf values were 0.16 and 0.607 cm. The EPS synthesized by the Pseudomonas sp. isolated from the coupons submerged in control medium also showed three spots and the bacteria isolated from the coupons submerged in UV treated medium showed only one spot (Fig. 3). The Rf values of the spots exhibited by the EPS of

Fig. 1. Adhesion of UV treated and control Alteromonas sp. cells on glass slides.

Fig. 2. Adhesion of UV treated and control Pseudomonas sp. cells on glass slides.

Alteromonas sp. 7.34±0.84 5.44±0.48 5.23±0.799 4.01±0.97Pseudomonas sp. 3.379±0.14 1.26±0.113 9.48±1.72 8.62±2.14

Table IEffect of UV treatment on the carbohydrate and protein concentration (mean ± stan-

dard deviation) of extracellular polymeric substance (EPS) synthesized by bacteria

Bacterial species UV treated Control UV treatedControlProtein (mg ml-1) Carbohydrate (mg ml-1)

Kolappan A. and Satheesh S. 2122

Pseudomonas sp. isolated from the control medium coupons were 0.165, 0.247 and 0.661 cm. The Rf value of Pseudomonas sp. isolated from the UV treated medium coupons was 0.454 cm.

The viable counts of biofilm sample isolated from the UV treated and the control medium were also var-ied. The viable count of Alteromonas sp. isolated from the coupons submerged in UV treated medium was 4.5±1.8×105 CFU ml–1. The viability of Alteromonas sp. isolated from the control medium coupons was 38.5± 9.3×105 CFU ml–1. The viable count of Pseudomonas sp. biofilm sample isolated from the coupons submerged in the control medium was 34.25±11.4×105 CFU ml–1. The biofilm sample of coupons submerged in UV treated medium showed a viability of 11.25±3.71×105 CFU ml–1. In general, the viability of bacterial cultures was reduced after UV treatment.

Discussion

The development of biofilm and fouling commu-nities is a multiple event with numerous interactions taking place between fouling organisms colonizing the surface. The first colonizers on any newly exposed

surface in marine waters are bacteria and they have been found to affect the subsequent recruitment of both microorganisms and macrofoulers. Hence, con-trol of biofilms on surfaces is an important strategy in any biofouling management programme. The results of the present study indicate that the viability of the bacterial cultures treated with UV light was reduced considerably. Previous studies by Munshi et al. (2005) also reported a reduction in bacterial load after UV treatment from a desalination plant. Generally, sensi-tivity of micro organisms to UV radiation may vary with species (Gaudy and Gaudy 1980). In the present study, Alteromonas sp. showed lower viability after UV treat-ment than that of Pseudomonas sp.

The adhesion of bacteria to glass surface was reduced considerably on the coupons submerged in UV treated medium. This may be due to the change in cell surface properties after UV treatment. Previous studies by Li and Logan (2005) reported a 40% reduction in the adhesion of Bacillus subtilis cells to the hard surface. They also reported that UV treatment oxidizes the surface polymers of the bacteria and decreases their adhesion to surface. The active attachment of bacterial cells is facilitated by the cell surface by the cell surface proper-ties such as adhesion proteins, capsules, surface charge, flagella and pili (Kumar and Anand 1998). Any change in the cell surface properties influences adhesion of the cells to the solid surfaces. Primary effect of UV light on bacterial adhesion was to reduce the hydrophobicity of the bacterial cell surface. The hydrophobicity of the cell surface is important in adhesion because hydrophobic interactions tent to increase with an increasing non-polar nature of one or both surfaces involved (i.e., the microbial cell surface and the substratum).

Microorganisms are inactivated by UV light as a result of photochemical damage to their nucleic acids (Sonntag and Schuchmann, 1992). Absorbed UV promotes the formation of bonds between adjacent nucleotides, creating double molecules or dimmers (Jagger, 1967). While the formation of thymine – thymine dimmers are the most common, cytosine – cytosine, cytosine-thy-mine, and uracil dimerization will also occur. Forma-tion of a sufficient number of dimers within a microbe prevents it from replicating the dNA and RNA.

The amount of carbohydrate and protein in the EPS synthesized by the bacteria isolated from the coupons submerged in the UV treated medium was lesser than that of the control. This indicated that the extracellular polymeric substance production was affected due to the UV treatment. Most of the fouling organisms includ-ing bacteria use adhesive materials with permanent or temporary adhesive capabilities to attach to surfaces (Callow and Callow, 2002). Extracellular polymeric substances are considered as adhesive material involved in the process of biofilm formation (Flemming et al.,

Fig. 3. Thin-layer chromatogram of the extracellular polymeric sub-stance isolated from both UV treated and control bacterial cells.

A – Alteromonas sp. (A1 – Control, A2 – UV treated) B – Pseudomonas sp.(B1 – Control, B2 – UV treated).

Influence of UV on biofilm formation by marine strains2 123

2000). The EPS consists of polysaccharides, polyuronic acids, proteins, nucleic acids and lipids (decho, 1990; Schmidt and Ahring, 1994). EPS may account for 50–90% of the total organic carbon of biofilms (Flemming et al., 2000) and can be considered as the primary matrix material of the biofilm. The EPS also bridge the micro-bial cells with the substratum and permit negatively charged bacteria to adhere both negatively and posi-tively charged surfaces. Hence, the reduction in EPS production may be one of the possible reasons for the low abundance of bacteria on the coupons submerged in UV treated medium.

The efficiency of UV on the removal of well-estab-lished biofilm matrix has some practical constraints. Flemming (1991) indicated that the effectiveness of UV treatment for removing an established biofilm might be low due to entrapped particles within the biofilm. Hence, in the present study, the effect of UV irradiation of the incoming water to control biofilm was analysed. The results of the present study suggest that UV light is a promising source for the control of bacterial fouling. There are large amounts of experience and data avail-able on its use for medical sterilization. The UV may be effective in control biofilm especially in cooling sys-tems if the incoming water is irradiated using specific UV devices. More studies on this aspect will certainly improve our understanding on the role of UV light in biofouling control.

Literature

Armstrong E., K.G. Boyd, A. Pisacane, C.J. Peppiatt, and J.G. Bur-gess. 2000. Marine microbial natural products in antifouling coat-ings. Biofouling 16: 215–224. Callow M.E. and J.A. Callow. 2002. Marine biofouling: a sticky problem. Biologist 49: 1–5.Chang J.C.H., S.F. Ossoff, D.C. Lobe, M.H. Dorfman, C.M. Dumais, R.G. Dualls and J.D. Johnson. 1985. UV inactivation of patho-genic and indicator microorganisms. Appl. Environ Microbiol. 49: 1361–1365Decho A.W. 1990. Microbial expolymer secretions in ocean envi-ronments: their role in food webs and marine processes. Oceanogr. Mar. Biol. Ann. Rev. 28: 73–153.Dubois M. 1956. Colorimetric method for determination of sugars and related substances. Anal.Chem.28: 50–356.Fingerman S.W. 1988. Environmental pollution by heavy metals from antifouling compounds. pp. 781–789. In: Thompson M.F, R. Sarojini and R. Nagabhushanam (eds). Marine Biodeterioration: Advanced techniques applicable to the Indian Ocean. Oxford & IBH, New delhi. Flemming H.C. 1991. Biofouling: effects and control. pp. 47–81 In: Flemming H.C and G. Gregory (eds). Biofouling in Industrial Water Systems. Springer Publisher, New York.

Flemming H.C., J. Wingender, C. Griegbe and C. Mayer. 2000. Physico-chemical properties of biofilms. Pp. 19–34. In: Evans L.V. (ed). Biofilms: recent advances in their study and control. Harwood Academic Publishers, Amsterdam. Gaudy A.F and E.T. Gaudy. 1980. Microbiology for Environmental scientists and Engineers. Mc Garw Hill book co. New York.Harris G.D., V.D. Adams, D.L. Sorensen and M.S. Curtis. 1987. Ultraviolet inactivation of selected bacteria and virus with photo-reactivation of the bacteria. Water Res. 21: 687–692Jagger J. 1967. Introduction to research in ultraviolet photobiology. Englewood Cliffs, New Jersey: Prentice-Hall.Jain A., K.K. Nishad, K.K. Narayan, B. Bhosle. 2007. Effects of dNP on the cell surface properties of marine bacteria and its impli-cation for adhesion to surfaces. Biofouling 23: 171–177.Kumar C.G and S.K. Anand. 1998. Significance of microbial bio-films in food industry. Int. J. Food Microbiol. 42: 9–27.Li B. and B.E. Logan. 2005. The impact of ultraviolet light on bacte-rial adhesion to glass and metal oxide-coated surface. Colloid sur-face B. 41: 153–161.Lowry O., H. Roseburg, A. Farr and R. Randwall. 1951. Protein Measurement with the Folin-Phenol reagent. J. Biol.Chem. 193: 265–275.Munshi H.A., N. Sasikumar, A.T. Jamaluddin and K. Mohammed. 1999. Evaluation of ultra-violet radiation disinfection on the bacterial growth in the SWRO pilot plant Al-Jubail, seawater. pp. 603–618. In: Proceedings of the fourth Gulf water conference, 13–18 Feb-ruary 1999, Bahrain. Water science and Technology Association, Manama, Bahrain.Munshi H.A., O.M. Saeed, T.N. Green, A.A. Al-Hamza, M. Farooque and A.R.A. Ismail. 2001. Application of ultraviolet radiation to control bacterial growth in the RO feed water from nanofiltration membranes. Technical Report: TR: APP 3805/9001, Saline water conversion corporation, Saudi Arabia. Munshi H.A., O.M. Saeed, T.N. Green, A.A. Al-Hamza, M. Farooque and A.R.A. Ismail. 2005. Impact of UV radiation on controlling biofouling problems in NF SWRO desalination process. In: International desalination Association (IdA) world congress, Singapore. Schmidt J.E. and B.K. Ahring, 1994. Extracellular polmers in granular sludge from difference up flow anaerobic sludge blanket (UASB) reactors. Appl. Microbiol. Biotechnol. 42: 457–462Sharrer M.J. and S.T. Summerfelt. 2007. Ozonation followed by ultraviolet irradiation provides attractive bacteria inactivation in a freshwater recirculating system. Aquat. Eng. 37: 180–191. Sonntag C.V. and H.P. Schuchmann. 1992. UV disinfection of drinking water and by-product formation-some basic considera-tions. J. Water SRT Aqua. 41: 67–74. Sutherland I.W. 2001. Biofilm exopolysaccharides: A story and sticky frame work. Microbiology, 147: 3–9.Wenjun S. and L. Wenjun. 2009. Impact of the ultraviolet disinfec-tion process on biofilm control in a model drinking water distribu-tion system. Environ. Engi. Sci. 26:809Yebra D.M., S. Kiil and K. Dam-Johansen. 2004. Antifouling technology-past, present and future steps towards efficient and environmentally friendly antifouling coatings. Prog. Org. Coat. 50: 75–104.Zelver N., R. Legan and W.G. Charaklis. 1981. Biofouling control with UV/peroxide. A laboratory study. Pp. 1164–1183. In: Proceed-ings of the water reuse symposium II Washington d.C.


ORGINAL PAPER

Introduction

during the last decade Wheat dwarf virus (WdV) has been the most frequently isolated and most ubi-quitous cereal-infecting virus in Hungary and it has now become a serious problem also in the Ukraine (Mesterházy et al., 2002, Szunics et al., 2003, Snihur et al., 2007). WdV is a frequent causal agent of dwarf-ing, mottling, yellowing or reddening in cereals and suppressed heading and root growth in infected plants can drastically reduce yield. WdV was first described by Vacke (1961) in the former Czechoslovakia and subsequently found in Sweden (Lindsten et al., 1970), Bulgaria (Stephanov and dimov, 1981), Hungary (Bisztray et al., 1989), France (Lindsten and Lindsten, 1993), Germany (Huth, 2000), Poland (Jezewska, 2001), Finland (Lemmetty and Huusela-Veistola, 2005), Romania (Jilaveanu and Vacke, 1995), Spain (Achon et al., 2006), Tunisia (Najar et al., 2000), Turkey (Köklü et al., 2007), Zambia (Kapooria and Ndunguru, 2004), Ukraine (for the first time approximately in 1975 (Razvyazkina 1975), then in 2007 (Snihur et al., 2007) and China (xie et al., 2007). WdV is transmitted by the European grass-feeding leafhopper Psammotettix alienus (Vacke, 1961) in a circulative, non-propagative

manner (Lindsten and Vacke, 1991), therefore the occurrence of diseased plants in the field depends on the presence of the vector. during the crop screening for WdV conducted in 2009–2010, the unique virus vector, Psammotettix alienus, has been found in abun-dance in Ukrainian agroecosystems, providing indirect proof of widespread presence of WdV in the Ukraine.

WdV belongs to the genus Mastrevirus (family Geminiviridae) infecting monocotyledonous plants. Mastreviruses have a monopartite single-stranded genome of circular dNA and the genome encodes four different proteins: movement protein (MP) and coat protein (CP) on the viral sense strand, and two replica-tion-associated proteins (Rep and RepA) on the com-plementary strand (Gutierrez, 1999). The presence of an intron in the Rep gene makes it possible for WdV to produce two different forms of the replication protein. The non-coding long intergenic region (LIR) and short intergenic region (SIR) contain sequence elements nec-essary for viral replication and transcription. The LIR comprises the origin of rolling circle replication of the virus (Heyraud et al., 1993). The SIR contains polyade-nylation signals and a region to which a short comple-mentary primer for the second strand synthesis binds (Kammann et al., 1991).

Comparison of the Nucleotide Sequences of Wheat Dwarf Virus (WDV)isolates from Hungary and Ukraine

ISTVÁN TÓBIÁS1*, OLEKSIY SHEVCHENKO2, BALÁZS KISS1, ANdRIY BYSOV2, HALINA SNIHUR2,VALERY POLISCHUK2, KATALIN SALÁNKI3 and LÁSZLÓ PALKOVICS4

1 Plant Protection Institute, Hungarian Academy of Sciences, Budapest, Hungary2 Taras Shevchenko’ Kyiv National University, Kyiv, Ukraine

3 Agricultural Biotechnology Centre, Gödölló, Hungary4 Corvinus University of Budapest, department of Plant Pathology, Budapest, Hungary

Received 10 October 2010, revised) 21 January 2011, accepted 5 February 2011

A b s t r a c t

Wheat dwarf virus (WdV) is the most ubiquitous virus in cereals causing huge losses in both Hungary and Ukraine. The presence of barley- and wheat-adapted strains has been confirmed, suggesting that the barley strain is restricted to barley, while the wheat strain is present in both wheat and barley plants. Five WdV isolates from wheat plants sampled in Hungary and Ukraine were sequenced and compared with known WdV isolates from GenBank. Four WdV isolates belonged to the wheat strain. Our results indicate that WdV-Odessa is an isolate of special interest since it has originated from wheat, but belongs to the barley-adapted strain, providing novel data on WdV biology and raising issues of pathogen epidemiology.

K e y w o r d s: Wheat dwarf virus (WdV), nucleotide sequence of WdV

* Corresponding author: I. Tóbiás, Plant Protection Institute, Hungarian Academy of Sciences; H-1525 Budapest, P.O.Box 102, Hungary; fax: +36 1 3918655; e-mail: [email protected]

Tóbiás I. et al. 2126

Two different forms of WdV exist: a wheat-adapted form (WdV wheat strain) and a barley-adapted form (WdV barley strain) (Lindsten and Vacke, 1991; Bendahmane et al., 1995, Kvarnheden et al., 2002). Both strains infect a number of plant species in the family Poaceae (Lindsten and Vacke, 1991). There are, however, contradictory reports on whether the wheat strain can infect barley, and whether the barley strain can infect wheat. Lindsten and Vacke (1991) and Tóbiás et al. (2009) observed no transmission of the barley strain to wheat, whereas the wheat strain was trans-missible to barley. Similar conclusion was made by Kundu et al. (2009), who found that the barley strain is restricted to the barley host, while the wheat strain is present in both wheat and barley plants. Mechner et al. (2003) detected both WdV strains in barley plants in the fields, using strain specific primers. Commandeur and Huth (1999) and Schubert et al. (2007) found that the barley strain can infect wheat only under laboratory conditions.

The genomes of the barley and wheat strains of WdV share an average of 85% identity. The isolates within the wheat strain show a high degree of homo-logy (>98% identity), whereas the isolates of the barley strain are more variable (>94% identity). As the demar-cation criterion for mastrevirus species has been set to 75% nucleotide sequence identity by the International Committee for Taxonomy of Viruses (Fauquet et al., 2008), both strains are currently considered to belong to the same species. Schubert et al. (2007) recently pro-posed two new mastrevirus species Barley dwarf virus

(BdV) and Oat dwarf virus (OdV) based on dNA sequence differences. OdV was accepted as a new ten-tative mastrevirus species sharing 70% genome-wide nucleotide sequence identity with the wheat and barley strains of WdV (Fauquet et al., 2008).

The aim of the present study was the molecular characterisation of WdV isolates from Hungary and Ukraine and their comparison with the available sequences of WdV.

Experimental


Virus isolates. Symptoms of viral infection were found during spring observations carried out in wheat crops in Martonvasar (Middle Hungary), Pula (South-ern Hungary) and Mironivka (Middle Ukraine). Plants displaying yellowing of leaves or dwarfing were placed in an insect-proof greenhouse and were tested for WdV by ELISA using a WdV kit (Bio-Rad). The col-lected WdV-infected plants were replanted into clay pots and placed in an insect-proof isolation net. For virus transmission, thirty individuals of virus-free Psammotettix alienus dahlb. were placed underneath each net. One week later the leafhoppers were trans-ferred to young seedlings of wheat being in two leaves stage. Six weeks later the plants were tested again for WdV by ELISA. Three isolates, WdV-HU-2Marton (collected in 2008 from Martonvasar), WdV-HU-Pula

WdV-Barley forw 468–488 (B) ATCCCGGGTCCTCCGACTACWdV-Barley rev 478–458 (B) GACCCGGGATCGTAAGGGGCWdV-Barley 540 555–531 (B) TAAGCCAAACAAACAACTCCTACGGWdV-P1 611–631 (B) GACCGAGGAAATTGGTTACGGWdV-5’ 1045–1067 (B) CCACTGACATCTTTACGATGCCWdV-Barley 1200 1200–1225 (B) AACTACGTAGTGGGGAAGAATATCGWdV-Barley 1900 1895–1917 (B) CATAGGTCGTGAAATTCAACTAGWdV-Barley 2110 2094–2122 (B) TTCGAGGCTTACGGAGTAGAGATGTTCATWdV-Wheat P1 475–494 (W) GACCGAGGAAATTGGTTACGGWdV-Wheat 483 506–485 (W) GCTTATACACAGCCCCCTTCCWdV-Wheat 5’ 809–831 (W) CCACTGACATCTTTACGATGCCWdV-Wheat 1076 1069–1087 (W) TAAGAAAGGAGCACTGTATCWdV-Wheat 1410 1428–1406 (W) GCGAGTCATTCATCAACTACTCGWdV-Wheat 1850 1850–1482 (W) CCACTCCTGCGGATCAAGCWdV-Wheat forw 2305–2326 (W) ACGAAGCTTGTTCTGCACGAGAWdV-Wheat rev 2316–2295 (W) AACAAGCTTCGTGCTTCCATCWdV-Wheat 2521 2521–2542 (W) CAGAAGTCCGGCAGGTCCTTA

Table IPrimers used for sequencing

Name Sequence (5’-3’)Genome position1

1 With reference to WdV-Heves (FM999833) – B and WdV-2 Marton (FN806785) – W.

WdV isolates from Hungary and Ukraine2 127

(collected in 2007 from Pula) and WdV-Uk-Miron (collected in 2009 from Mironivka and maintained in our greenhouse by subsequent transmission) were selected for further studies. Ten wheat samples from

the Odessa region (South Ukraine) and one from Glevakha (Central North Ukraine) were initially tested by PCR with WdV specific primers. Two samples (WdV-Uk-Odessa and WdV-Uk-g collected from Odessa and Glevakha, respectively) were selected for molecular characterization.

Isolation of virus DNA, cloning and sequence analysis of the WDV isolates. dNA extraction was done according to Shepherd et al. (2008) with a slight modification (fresh leaf material was used instead of dry leaves). The samples were then stored at –20°C or used directly as a template for rolling circle amplifica-tion (RCA) of the WdV genome (Haible et al., 2006). One microliter of the final Extract-n-Amp dNA solution was mixed with 4 μl of Templi PhiTM sample buffer (TempliPhiTM, Amersham Biosciences), heated for 2 min at 94°C, and then brought to room tempera-ture. Five μl of reaction buffer and 0.2 μl of enzyme mix were added to the cooled mixture and the Templi PhiTM extension reaction was run at 30°C for 18–20 h. WdV genome concatemers (multiple copies of unit-length virus genomes covalently linked end-to-end) gener-ated during Phi29 dNA polymerase amplification were digested with HindIII (wheat strain) or SmaI (barley strain) to release unit-length genomes. After diges-tion genomic dNA was separated in 1% agarose gel and extracted with a dNA purification kit (Fermentas dNA Extraction Kit). The WdV genome was inserted into a HindIII or SmaI digested pBSK+ plasmid (Strata-gene). The recombinant plasmids were transformed into Escherichia coli dH5α (Sambrook et al., 1989).

Clones containing inserts with the expected size of 2.7 kb were sequenced with the dyedeoxyTermina-tor Kit (Applied Biosystems) using reverse, universal (–20) and internal primers (Table I). Sequence analysis was performed using University of Wisconsin Genetics Computer Groups (GCG) sequence analysis software package version 9.1.

In order to determine the phylogenetic relationships between different WdV isolates complete genomes were analysed (Table II). Sequence alignment, tree for-mation, and bootstrap analysis were done with the help of the software Clustal x 1.83.


This work has been focused on the screening of Hungarian and Ukrainian cereal ecosystems for the presence of Wheat dwarf virus and its unique vec-tor, P. alienus. The outcomes of the 2-year monitoring clearly demonstrated significant spread of the virus in Ukraine and confirmed the positive tendency in its spread compared to previous years of observations. In addition, for the first time a virus vector has been

Table IIAbbreviation, accession number and origin of Wheat dwarf virus

isolates used in this study

Abbreviation Countryof collection

Accessionnumber

WdV isolates sequenced in this study are indicated in bold type.

WdV-HU-B AM040732 HungaryWdV-HU-F AM040733 HungaryWdV-HU-H07 FM210034 HungaryWdV-HU-Heves FM999833 HungaryWdV-HU-dunakiliti FM999832 HungaryWdV-HU-Martonbar AM747816 HungaryWDV-HU-2Marton FN806785 HungaryWDV-HU-Pula FN806786 HungaryWDV-Uk-g FN806783 UkraineWDV-Uk-Miron FN806784 UkraineWDV-Uk-Odessa FN806787 UkraineWdV-BU-Bg17 AM989927 BulgariaWdV-Swe-Enk1 AJ311031 SwedenWdV-Swe-Enk2 AM491490 SwedenWdV-Swe-SE x02869 SwedenWdV-Chi-hbsjz061 EF536870 ChinaWdV-Chi-ynkm062 EF536881 ChinaWdV-Chi-sxyl052 EF536878 ChinaWdV-Chi-gsgg050 EF5368591 ChinaWdV-Chi-sxyl051 EF536877 ChinaWdV-Ge-SxA22 AM296022 GermanyWdV-Ge-SxA23 AM296023 GermanyWdV-Ge-SxA24 AM296024 GermanyWdV-Ge-SxA25 AM296025 GermanyWdV-Ge-SCBB21 AM296021 GermanyWdV-Ge-BaW1 AM411651 GermanyWdV-Ge-BaW2 AM411652 GermanyWdV-Ge-McP20 AM296020 GermanyWdV-Ge-Sx18 AM296018 GermanyWdV-Cz-6217 FJ546189 Czech RepublicWdV-Cz-6239 FJ546190 Czech RepublicWdV-Cz-W FJ546188 Czech RepublicWdV-Cz-1841 FJ546191 Czech RepublicWdV-Cz-19 AM296019 Czech RepublicWdV-Cz-11105 FJ546180 Czech RepublicWdV-Cz-8100 FJ546179 Czech RepublicWdV-Cz-11229 FJ546181 Czech RepublicWdV-Cz-6482 FJ546178 Czech RepublicWdV-Cz-B FJ546193 Czech RepublicWdV-Tr-bar AJ783960 Turkey


shown to be common to Ukrainian fields since it has been detected in virtually every region of the country where the virus was identified.

depending on the place of origin, the degree of WdV infection of collected wheat plants was 20–70% as confirmed by ELISA tests and PCR. In laboratory, the virus was transmitted by P. alienus in insect-proof isolation net and maintained on oat or wheat plants. Nucleic acids were isolated from plants infected with WdV-HU-2Marton, WdV-HU-Pula, WdV-Uk-g and WdV-Uk-Odessa, WdV-Uk-g and WdV-Uk-Miron and used for subsequent molecular and phylogenetic characterization.

The Extraction-n-Amp dNA extraction method was very rapid and simple in order to isolate intact viral dNA. WdV genome amplification via the RCA method generates large dNA concameters, from which unit-length genome were subsequently cleaved with HindIII or SmaI enzymes. The size of the full length genome obtained for WdV-HU-2Marton and WdV-HU-Pula constituted 2750 nucleotides, while the genomes of WdV-Uk-g and WdV-Uk-Miron were 2749-nucleotide-long, and the WdV-Uk-Odessa genome was exactly 2734 nucleotides in length. The very special property of this latter isolate is hat it has originated from the winter wheat variety Selyanka. To our knowledge, this is the first report of the barley strain of WdV isolated from naturally infected wheat plants. The genomes of these character-ized isolates contained all four expressed mastrevirus ORFs (MP, CP, Rep, RepA), and the intergenic regions LIR and SIR.

The nucleotide sequences were deposited in Gen-Bank as WdV-Uk-g: FN806783, WdV-Uk-Miron: FN806784, WdV-HU-2Marton: FN806785, WdV-HU- Pula: FN806786 and WdV-Uk-Odessa: FN806787, and

were further compared to previously characterized WdV isolates (Tobias et al., 2006, 2009, 2010) (Table III).

The analysis of the full genome sequences revealed high levels of identity among wheat strains and higher level of diversity among barley strains. The sequences’ identities between isolates of the wheat strain of dif-ferent geographical origins were very similar (>98.7% identity). For the movement protein (MP) and coat protein (CP), we observed high sequence identity (>98.8%) at the predicted amino acid level, in some cases MPs (Hungarian and Swedish isolates) and CPs (Hungarian isolates originating from different parts of the country) were identical. For the short (SIR) and large intergenic region (LIR), we observed a higher var-iability (97% and 96.6% identity, respectively) (data not shown). Regarding the diversity of the WdV isolates of the barley strain, we observed a relatively high vari-ability (96.3–99.4%). Interestingly, however, the MP and CP also revealed a high level of amino acid sequence identity among barley strain isolates originating from different geographical regions (98.5–100%). Similar to wheat strain isolates, barley strain isolates showed greater variability also in the LIR and SIR.

Molecular characterization of Ukrainian and Hun-garian WdV isolates was followed by phylogenetic analysis in order to compare their relationships with previously characterized wheat and barley isolates avail-able from the GenBank database (Fig. 1). The phyloge-netic analysis of WdV isolates showed that they were clearly distinguishable, both barley and wheat strains formed two clades. Isolates from Hungary, Germany, Czech Republic, Ukraine and Sweden clustered in clade 1. Interestingly, both Ukrainian isolates WdV-Uk-g and WdV-Uk-Miron showed closer relation-ship to WdV-HU-Pula and WdV-Swe-Enk2 isolates, respectively, than to each other. This is surprising as

Pula 99.5 99.5 99.2 85.3 85.5 85.3 85.4 98.7 99.5 85.5B 99.6 99.4 85.1 85.3 85.1 85.2 98.7 99.6 85.3F 99.3 85.2 85.4 85.2 85.3 98.7 99.4 85.32Marton 85.3 85.2 85 85.1 98.4 99.3 85.2H07 99.3 99 96.3 84.9 84.9 96.5Heves 99.4 96.6 85.1 85.1 96.8dunakiliti 96.6 84.9 84.9 96.9Bg17 85.1 85 99.3Mironivka 98.7 85.2.g 85.1

Table IIISequence identity of complete genomes of the WdV isolates characterized in our laboratory

WdV B Mironivka Odessa.gBg17dunakilitiHevesH072MartonF

Abbrevations and accession numbers: WdV-HU-B: AM040732, WdV-HU-F: AM040733, WdV-HU-H07: FM210034,WdV-HU-Heves: FM999833, WdV-HU-dunakiliti: FM999832, WdV-BU-Bg17: AM989927, WdV-Uk-g: FN806783,WdV-Uk-Miron: FN806784, WdV-HU-2Marton: FN806785, WdV-HU-Pula: FN806786 and WdV-Uk-Odessa: FN806787


Fig. 1. Phylogenetic tree constructed by the UPGMA method for complete genome sequences of Wheat dwarf virus isolates. (Bootstrap values are indicated) The isolate OdV-Ge-SxA25 was used as the outgroup with a ca. 70% genome-wide nucleotide sequence identity

with barley and wheat strain isolates of WdV.


according to the available literature data phylogenetic relationships of WdV isolates normally show high degree of dependence on the geographical origin of the virus (Köklü et al., 2007). However, both WdV-Uk-g and WdV-Uk-Miron isolates came from the same geographical region of the Ukraine (Kiev region, central north of the Ukraine) and the sites of sampling were situated just approximately 100 km apart from each other. Apparently, other issues (such as vector occur-rence and behaviour, plant cultivars cultivated at a given territory, agricultural techniques and, primarily, the initial virus source in the country) should be consid-ered as well when evaluating spread and evolutionary divergence of WdV.

Clade 2 could be divided into two subgroups, one with wheat isolates from China and the other contain-ing only the WdV-Swe-SE isolate from Sweden.

As for the isolates of barley strain of WdV, clade 3 could be divided into two subgroups, one with a diver-gent pool of isolates from Hungary, Germany and Czech Republic. The other subgroup comprised Ukrainian and Bulgarian isolates of WdV-BU-Bg17 and WdV-Uk-Odessa. In Clade 4, WdV-TR-bar isolate formed one subgroup and WdV-Cz-11105 and WdV-Cz-8100 formed the other one. These observations are in a good agreement with previous results (Schubert et al., 2007, and Kundu et al., 2009).

In conclusion, the results presented in this work have shown that Hungarian and Ukrainian isolates of WdV were divided into two distinct groups of wheat and barley strains. WdV-Uk-Odessa is the first barley strain isolate originating from wheat infected under natural conditions. At this point it should be men-tioned that we have managed to identify the barley strain of WdV in a single wheat plant only once dur-ing the intensive two-year strain-specific PCR-based screening of WdV isolates in naturally grown cereal crops in Hungary and Ukraine. Hence the proven fact of WdV barley strain transmission to wheat plants by P. alienus is obviously an uncommon and rare event. Seemingly it may happen under natural conditions but only occasionally and possibly when virus concentra-tions in host plants are high enough to allow host range extension by overcoming typical limitations on virus-plant relationships.

In our opinion, the issue of WdV transmission by its vector needs further characterization especially by employing molecular approaches to identify virus genes and/or gene products (and preferably their vector coun-terparts) responsible for the transmission of the virus and its efficiency.

AcknowledgementsThis research was supported by a Hungarian Scientific Research

Found (OTKA 61644 and 68589). TI, KB, PL, SO, SH and BA were supported by TéT (UA-14/8).

Literature

Achon M.A., L. Serrano, L. Ratti and C. Rubies-Autonell. 2006. First detection of Wheat dwarf virus in barley in Spain associated with an outbreak of barley yellow dwarf. Plant Dis. 90: 970.Bendahmane M., H-J. Schalk and B. Gronenborn. 1995. Identifica-tion and characterization of wheat dwarf virus from France using a rapid method for geminivirus dNA preparation. Phytopathology 85: 1449–1455.Bisztray Gy., R. Gáborjányi and J. Vacke. 1989. Isolation and char-acterisation of wheat dwarf virus found for the first time in Hungary. J. Plant. Dis. Protect. 96: 449–454.Commandeur U. and W. Huth. 1999. differentiation of strains of wheat dwarf virus in infected wheat and barley plants by means of polymerase chain reaction. J. Plant. Dis. Protect. 106: 550–552.Fauquet C.M., R.W. Briddon, J.K. Brown, E. Moriones, J. Stanley, M. Zerbini and X. Zhou. 2008. Geminivirus strain demarcation and nomenclature. Arch. Virol. 153: 783–821. Gutierrez C. 1999. Geminivirus dNA replication. Cell. Mol. Life Sci. 56: 313–329.Haible D., S. Kober and H. Jeske. 2006. Rolling circle amplification revolutionizes diagnosis and genomics of geminiviruses. J. Virol. Meth. 135: 9–16.Heyraud F., V. Matzeit, M. Kammann, S. Schaefer, J. Schell and B. Gronenborn. 1993. Identification of the initiation sequence for viral-strand dNA synthesis of wheat dwarf virus. EMBO J. 12: 4445–4452.Huth W. 2000. Viruses of Gramineae in Germany – a short over-view. J. Plant. Dis. Protect. 107: 406–414.Jezewska J. 2001. First report of Wheat dwarf virus occurring in Poland. Phytopath Polonica 21: 93–100.Jilaveanu A. and J. Vacke. 1995. Isolation and identification of wheat dwarf virus (WdV) in Romania. Probleme de Protectia Plan-telor 23: 51–62.Kammann M., H.J. Schalk, V. Matzeit, S. Schaefer, J. Schell and B. Gronenborn. 1991. dNA replication of wheat dwarf virus, a gemini virus, requires two cis-acting signals. Virology 184: 786–790.Kapooria R.G. and J. Ndunguru. 2004. Occurrence of viruses in irrigated wheat in Zambia. EPPO/OEPP Bulletin 34: 413–419.Köklü G., J.N.E. Ramsell and A. Kvarnheden. 2007. The complete genome sequence for a Turkish isolate of Wheat dwarf virus (WdV) from barley confirms the presence of two distinct WdV strains. Virus Genes 34: 359–366.Kundu J.K., S. Gadiou and G. Cervená. 2009. discrimination and genetic diversity of Wheat dwarf virus in Czech Republic. Virus Genes 38: 468–474.Kvarnheden A., M. Lindblad, K. Lindsten and J.P.T. Valkonen. 2002. Genetic diversity of Wheat dwarf virus. Arch. Virol. 147: 205–216.Lemmetty A. and E. Huusela-Veistola. 2005. First report of Wheat dwarf virus in winter wheat in Finland. Plant Dis. 89: 912.Lindsten K. and B. Lindsten. 1993. Occurrence and transmission of Wheat dwarf virus (WdV) in France. In: Proceedings of Third International Conference on Pest in Agriculture 7–9 december 1993, Montpellier, France, pp. 41–48.Lindsten K. and J. Vacke. 1991. A possible barley adapted strain of Wheat dwarf virus (WdV). Acta Phytopath. Entomol. Hung. 26: 175–180.Lindsten K, J. Vacke and B. Gerhardson. 1970. A preliminary report on three cereal virus diseases new to Sweden spread by Macrosteles and Psammotettix leafhoppers. Nat. Swedish Inst. Plant. Protec. Cent. 14: 285–297.Mechner S., B. Manurung, M. Grüntzig, A. Habekuss, W. Witsack and E. Fuchs. 2003. Investigations into the ecology of the Wheat dwarf virus (WdV) in Saxony-Anhalt, Germany. J. Plant Dis. Pro-tect. 110: 313–323.


Mesterházy Á., R. Gáborjányi, M. Papp and P. Fónad. 2002. Mul-tiple virus infection of wheat in South Hungary. Cer. Res. Com. 30: 329–334.Najar A., K.M. Makkouk, H. Boudhir, S.G. Kumari, R. Zarouk, R. Bessai and F.B. Othman. 2000. Viral diseases of cultivated legume and cereal crops in Tunisia. Phyt. Medit. 39: 423–432.Razvyazkina G.M. 1975. Virus diseases of Cereals – Novosibirsk, 1975 – 292 p. (in Russian)Sambrook J., E.F. Fritsch and T. Maniatis. 1989. Molecular Clon-ing. A Laboratory Manual. 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.Schubert J., A. Habekuss, K. Kazmaier and H. Jeske. 2007. Survey-ing cereal-infecting geminiviruses in Germany – diagnostics and direct sequencing using rolling circle amplification. Virus Research 127: 61–70.Shepherd D., PD Martin, P. Lefeuvre, A. Monjane, B. Owor, E.P. Rybicki and A. Varsani. 2008. A protocol for the rapid isola-tion of full geminivirus genomes from dried plant tissue. J. Virol. Meth. 149: 97–102.Snihur H., V. Polischuk and U. Kastirr. 2007. dissemination of viruses of cereal crops in agrocoenosises of Ukraine. // 10 th Interna-tional Plant Virus Epidemiology Symposium. 15–20 October 2007, ICRISAT, Hyderabad, India. P. 107.

Stephanov J. and A. Dimov. 1981. Bolestta vdjudjavanje po spe- nittsata Bolgaria. Rasteniev Nauki 18: 124–128 (in Bulgarian with English summary). Szunics L., E. Pocsai, G. Vida, O. Veisz, L. Láng and Z. Bedő. 2003. Kalászos gabonák vírusok okozta betegségei 2002-ben. Növényter-melés 52: 33–39 (in Hungarian with English summary).Tóbiás I., B. Kiss and L. Palkovics. 2006. The nucleotide sequence of two Hungarian isolates of Wheat dwarf virus. Acta Phytopath. et Entomol. Hung. 41: 47–52. Tóbiás I., B. Kiss, É. Pájtli, G. Tholt and K. Salánki. 2008. Molecu-lar characterization and transmission experiments with barley strain of Wheat dwarf virus. Növényvédelem 44(11), 545–552. (in Hungar-ian with English summary)Tóbiás I., B. Kiss, N. Bakardjieva and L. Palkovics. 2009. The nucleotide sequence of barley strain of Wheat dwarf virus isolated in Bulgaria. Cer. Res. Com. 37: 237–242.Tóbiás I., B. Kiss, K. Salánki and L. Palkovics. 2010. The nucleotide sequence of barley strain of Wheat dwarf virus isolated in Hungary. Cereal Research Communication 38: 67–74.Vacke J. 1961. Wheat dwarf virus. Biol. Plant. 3: 228–233.Xie J., X.Wang, Y. Liu, Y. Peng and G. Zyhou. 2007. First report of the occurrence of Wheat dwarf virus in wheat in China. Plant Dis. 91: 111.


ORGINAL PAPER

Introduction

β-1,3-glucanase, synthesized by many bacteria, hydro lyzes glucan polymers containing β-1,3-linkages (Kourkoutas et al., 2004). This enzyme plays a key role in both the pulp drainability and beatability changes. The use of pure endoglucanase was responsible for most of the success in deinking (Gusek et al., 1991). Enzymes displaying β-glucanase activity have been found to have a variety of uses. For example they are used as a biologi-cal control of soil-borne plant pathogens and as a food supplementation (Weuster-Botz, 1993). β-1,3-glucanase is an important enzyme in the field of industrial and agricultural processing. The resistance of this enzyme to denaturation by high temperature and pH extremes makes it particularly important.

Cell immobilization can be defined as the confine-ment or localization of viable microbial cells to a certain defined region (Hamdy et al., 1990). This is achieved by significantly increasing the effective size or density of the cells by their aggregation or by attachment of the

cells to some support surface. Thus, flocculated cells in the form of large aggregates are considered to be immobilized. The advantages of inorganic supports compared to organic supports were studied by several groups (Kourkoutas et al., 2004). These include their abundance and low price, higher mechanical strength, high thermal stability, higher resistance to organic solvents and microbial attack, easy handling and easy regenerability (Gusek et al., 1991; Weuster-Botz, 1993). Many inorganic supports were studied for immobili-zation such as polygorskite, montmorilanite, hydro-mica, porous porcelain, pumice stone and glass beads (Colagrande et al., 1994; Beshay, 1998).

The aim of the present work wasto optimize the pro-duction of β-glucanases using a high bacterial cell den-sity cultivation strategy. Cell immobilization is applied to improve the productivity of the recombinant strain. different inorganic supports such as porous sintered glass SIRAN® SIKUG 041, Ceramic supporting matri-ces and Broken Pumice stone as well as SIRAN Raschig-rings were tested.

β-glucanase Productivity Improvement via Cell Immobilizationof Recombinant Escherichia coli Cells in Different Matrices

USAMA BESHAY1, *, HESHAM EL-ENSHASY1, I.M.K. ISMAIL2, HASSAN MOAWAd3

and SAWSAN ABd-EL-GHANY1

1 Bioprocess development department, Genetic Engineering and Biotechnology Research Institute (GEBRI)Mubarak City for Scientific Research and Technology Applications, New Bourg El-Arab City

Universities and Research district, 21934 Alexandria, Egypt2 Botany department, Faculty of Science, Cairo University, Cairo, Egypt

3 Agricultural Microbiology department, National Research Centre, dokki, Cairo, Egypt

Received 10 September 2010, revised 3 March 2011, accepted 10 March 2011

A b s t r a c t

The studies have been performed to analyze the production of β-glucanase by a recombinant strain of Escherichia coli immobilized in different matrices. Porous sintered glass SIRAN®, Ceramic supporting matrices and Broken Pumice stone as well as SIRAN Raschig-rings were examined for the immobilization of whole bacterial cells. The β-glucanase activity of bacteria immobilized in CeramTec PST 5 (4–5 mm) was very low. CeramTec PST 5 (1.5–2.5 mm) was found to be the best carrier compared to all other matrices regarding glucanase production (630 U/ml) and compared to enzyme activity produced by free cells (500 U/ml). different doses of matrices were applied (2, 5, 7, 10 g/lask) in the form of “matrix weight”. Using 2 g/flask of CeramTec PST 5 (1.5–2.5 mm) yielded enzyme activity of 630 U/ml). CeramTec gives highest operational stability of β-glucanase by repeated batch fermentation to 5 cycles, and activity reached 660 U/ml. Scanning electron microscopy observations showed a high number of vegetative cells that continued growth inside the matrices, indicating that β-glucanase activity improvement was due to the immobilization of the cells.

K e y w o r d s: β-glucanase, organic and/or inorganic matrices, cell immobilization, scanning electron microscope

* Corresponding author: U. Beshay, Mubarak City for Scientific Research and Technology Applications, Bioprocess develop-ment depart ment, New Bourg El-Arab City, Universities and Research district; 21934 Alexandria, Egypt; phone: +2 03 4593 422; fax: +2 03 4593 407; e-mail: [email protected]; u.beshay@mucsat,sci.eg

Beshay U. et al. 2134

Experimental


Microorganism. The E. coli strain BL21(dE3) pET-bgl- hisactophilin-Sec used in this work was kindly provided by prof. dr Erwin Flaschel, the head of the Fermen-tation Engineering department, Bielefeld University, Germany. This strain contained a plasmid coexpress-ing protein of a chimeric Bacillus amyloliquefaciens β-glucanase C-terminally linked with hisactophilin from Dictyostelium discoideum as metal chelating affinity tag and kill protein from the plasmid CoIE1 as a permea-bilizing agent for the outer membrane of Escherichia coli. The fusion protein was expressed under control of the genuine promoter of β-glucanase from Bacillus amyloliquefaciens, whereas the kill protein was under the control of stationary phase promoter of the fic gene (Miksch et al., 1997; Flaschel et al., 1998).

Media for vegetative cell growth and fermenta-tion. The optimized medium Terrific broth glycerol TBG (Beshay et al., 2003) used in this work contained the following gm per liter: 12.0 peptone, 24.0 yeast extract, 5.0 NaCl, 7.0 lactose. These components were dissolved in 800 ml distilled water and pH was adjusted at 7.0. The medium was distributed into flasks contain-ing 40 ml medium each and then autoclaved at 121°C for 20 minutes. Kanamycin was added after cooling (50 µl/50ml). The second part of the medium, i.e. “salts”, consisted of g per liter: 2.30 potassium dihydrogen phosphate, 12.50 di-potassium hydrogen phosphate. Salts were dissolved separately in 200 ml distilled water and distributed into 20 test tubes each containing 10 ml, then autoclaved. Sterilized salts were added to the basal medium before inoculation.

Porous supports for immobilization of E. coli. Three inorganic porous supports, porous sintered glass SIRAN, broken pumice stone, and the ceramic cata-lyst carrier CeramTec F1/porous PST 5 were used to immobilize the cells. The first supporting matrix used to immobilize E. coli was SIRAN® in the form of porous sintered glass beads (Schott Engineering GmbH, Mainz, Germany). Two different shapes of porous sintered glass SIRAN® beads and Raschig-rings SIRAS 09 were used in this study. The characteristics of porous sintered glass SIRAN are gathered in Table I.

The broken pumice stone was supplied by Joseph Raab GmbH & Cie. KG (Neuwied, Germany). It had a diameter of 1.5–2.5 mm, a pore volume of 0.93 ml/g, a particle density of 670 g/l and a specific surface area of 27 cm2/cm3. The chemical composition of broken pumice stone is as follows: silicic acid SiO2 55%, alu-minum oxide Al2O3 22%, alkalis K2O+Na2O 12%, iron oxide Fe2O3 3%, calcium oxide CaO 2%, magnesium oxide MgO 1%, titanium dioxide TiO2 0.5%. Ignition loss was 4%.

The CeramTec® carrier F1/porous PST 5 was obtained from CeramTec® AG (Innovative Ceramic Engineering, Wunsiedel, Germany). Two different particle sizes were used, 1.5–2.5 mm and 4–5 mm. It has a particle density of 1430 g/l, a pore volume of 0.25 ml/g and a specific surface area of 20 cm2/cm3.

Before use, the supports were cleaned in 10% aqueous hydrogen peroxide solution by gentle agitation at 80°C for 30 min. After the liquid was decanted, the beads were washed with water and dried overnight at 110°C.

Cultivation of E. coli in immobilized formEffect of using different inorganic matrices on

β-glucanase production. Batch cultivation was car-ried out in Erlenmeyer flasks (250 ml), each containing 50 ml of TBG-medium. The flasks were inoculated with 2.5 ml (Od600 = 4) fresh bacterial cells grown for 24 h. To each flask, 2 g of pre-sterilized supporting materials such as, porous sintered glass beads SIRAN SIKUG 041, CeramTec PST 5 (1.5–2.5 mm and 4–5 mm), Broken Pumice stone (1.5–2.5 mm) and Porous sintered glass Raschig-rings SIRAS 09 were added in order to select the most suitable matrix for maximal production of β-glucanase. The flasks were incubated in a shaker incu-bator (200 rpm) at 37°C for 72 h. As a control, a free-cell suspension was cultivated under the same culture conditions. Samples of the liquid culture were taken at indicated intervals for the determination of β-glucanase activity and free cell concentration. In addition, car-rier samples were taken for the determination of immo-bilized cells and in some cases for scanning electron microscope (SEM) observations.

Effect of using different doses of inorganic matri-ces on β-glucanase production. different quantities of selected matrices were tested to find out the most suitable one. The quantities were (2, 5, 7 and 10) g per flask containing 50 ml culture medium. Samples were taken during the experimental runs to determine cell growth and enzyme activity.

Repeated batch cultivations of E. coli on inorganic matrices. For establishing the long-term stability of β-glucanase production by immobilized cells, repeated batch cultivations were carried out. Every 48 h the bio-catalysts were washed several times with sterile water and transferred into fresh production medium under

SIKUG 041 0.4–1 55–60 <120 0.15 SIRAS 09 8.8 × 9 70 1.6–400 0.4

Table ICharacteristics of porous sintered glass (SIRAN®)

Carriertype

Characteristicdimensions

(mm)

Surfaceaream2/g

Porediameter

(µm)

Porevolume

(%)

β-glucanase production by immobilized E. coli2 135

repeated batch cultivation conditions. This process was carried out by decanting the spent medium every 48 h and replacing it with fresh medium after washing the matrix with sterile saline. On the other hand, a similar experiment was carried out with free cells to compare the efficiency of free and immobilized cells for the pro-duction of glucanase enzyme under these conditions. Samples were taken after each cycle (48 h) to determine cell density and β-glucanase activity.

Sample analysis. All determinations reported here were performed in triplicate and the experiments at least in duplicate. The results given are mean values.

The cell density in the medium was monitored by measuring the absorbance at 600 nm using a spectro-photometer (Pharmacia Biotech). Optical densities values were converted to cell dry weight according to a standard calibration curve previously obtained for E. coli cells. Optical density at 600 nm = 1 is equal to 0.4 g/l cell dry weight.

The cell density of immobilized cells was deter-mined indirectly by measuring the total amount of pro-tein of the immobilized cells by a method described by (Henry and Clim, 1964). For the determination of the cell density of the immobilized cells, carrier samples were washed with 1 ml phosphate buffer, followed by adding 1 ml lysis buffer and strong vortexing for 5 min-utes at room temperature. The assay test was performed with the supernatant of the lysis solution.

β-Glucanase activity was determined according to Borriss et al., (1989) using lichenan (Sigma), a polysac-charide from Cetraria islandica as substrate. The activity test described here expresses the activity in units (U) corresponding to the liberation of 1 µmol glucose per min at pH 5.6 at 50°C from lichenan. Briefly, lichenan solution (200 µl) is thermostated at 50°C for 5 min in 1.5 ml reaction vials equipped with plastic covers. A series of reactions was started by adding the sam-ple solution (20 µl) at 20 s intervals. Each reaction was stopped exactly after 20 min by adding 2-hydroxy-3,5-dinitrobenzoic acid (HdNBA) solution (100 µl) with intensive mixing. The reaction vials were placed on ice. After cooling, the reaction vials were centrifuged for a few seconds in order to collect the liquid phase. Each cover of the vials was perforated with a needle prior to placing the vials for 10 min in a boiling water bath followed by rapid cooling in ice water. After cool-ing, the residual solution in the vials was complemented with distilled water (1 ml). The absorption of this solu-tion was measured at 530 nm.

For morphological studies of E. coli cells in carrier beads, scanning electron microscopy was used. Car-rier samples were taken from the flasks, washed twice with phosphate buffer and treated in a laboratory microwave oven according to the method of Giberson et al., 1997. Samples were fixed for 60 seconds in 2.5%

glutaral dehyde (in 50 mM phosphate buffer, pH 7.2) at 37°C. The second fixation was carried out with a solu-tion of 2% osmium tetraoxide in deionized water for 2.5 minutes at 37°C. The fixed carriers were dehydrated by treatment with 30, 45, 60, 75 and 90% acetone at 40°C-each step for 5 minutes. At the end of the dehy-dration process the samples were treated overnight with absolute acetone and dried by the critical point method. The specimens were coated with a thin layer of gold to make the surface more efficient in electron scattering. Finally, the specimens were observed with a JxA-JEOL scanning electron microscope.


Immobilization of E. coli cells by adsorption method using inorganic matrices

Effect of using different matrices on cell growth and β-glucanase production. As shown in Fig. 1A, freely suspended cells grew and reached maximum CdW of 3.5 g/l after 72 h incubation while biomass concentrations, on Pumice stone and CeramTec (1.5–2.5 mm) after 72 h, were 3.8 and 3.1 g/l, respectively.

Fig. 1. Effect of different inorganic matrices on cell density (A) and β-glucanase activity (B) of a recombinant E. coli BL21(dE3)

pET-bgl-hisactophilin-Sec strain in immobilized state.


β-glucanase production started to increase markedly after 28 h and reached a maximum activity of 517 U/ml at 48 h for free cells. On the other hand, the results for β-glucanase production by immobilized cells showed that almost all of the biocatalysts used were more effec-tive enzyme producers than free cells. It was noticed that immobilization bacterial cells on CeramTec PST 5 (1.5–2.5 mm) and Broken Pumice stone (1.5–2.5 mm) lead to an increase in β-glucanase production (630 U/ml and 590 U/ml at 72 h, respectively) in comparison to freely suspended cells (500 U/ml).

This was probably due to the high amount of cells immobilized inside the matrix during the cultivation process. The difference in enzyme activity among the various matrices could be attributed to the difference in porosity and consequently, oxygen transfers which facilitated better growth of the immobilized cells and higher enzyme production (Fiedurek and Lobarzewski 1990; Beshay and Moreira 2003). Thus, E. coli cells may be immobilized on commercially available porous

carriers. Immobilization has been shown to be an ade-quate means for obtaining high cell densities.

Effect of matrix dose used for cell immobilization on both cell growth and enzyme production by E. coli. By comparing different quantities of the matrix for cell immobilization it was found that increasing the amount of carrier weight from 2.0 to 10.0 g/50 ml medium was accompanied by decreasing β-glucanase production as shown in Fig. 2A, B for CeramTec (1.5–2.5 mm) and broken Pumice stone (1.5–2.5 mm), respectively.

This could be attributed to the fact that a rise in carrier amount leads to increase in the possible shear stress force and abrasion effect between microbial cells and carrier particles. It was also found that 2.0 g of CeramTec (1.5–2.5 mm) is an optimal carrier dose for the highest glucanase production (630 U/ml) after 72 h cultivation. However the glucanase activity obtained by immobilized cells in Broken Pumice stone was com-parable with that obtained by cell immobilization in CeramTec. This is because of the fact that broken pum-ice stone shows a tendency to be broken by shaking which leads to a high turbidity in culture medium.

Operational stability of β-glucanase produced by immobilized cells in CeramTec matrix (1.5–2.5 mm) by the adsorption method. The possibility of the multiple use of E. coli cells immobilized in CeramTec (1.5–2.5 mm) for β-glucanase production was studied by repeated batch cultivation for 10 days (5 cycles). Each fermentation cycle continued for 48 h as previously described in materials and methods. The substitution of exhausted (spent) medium with fresh production medium was carried out at the time of maximum enzyme production (48 h for immobilized cells). Figure 3 illustrates also that β-glucanase production by immobi-lized cells increased with number of reaction cycles and reached a maximum at the third cycle (660 U/ml) where β-glucanase production (650 U/ml) remained almost stable for the following two batches. On the other hand,

Fig. 2. Effect of CeramTec PST5 (1.5–2.5 mm) (A) and Broken Pumice stone (1.5–2.5mmm) (B) doses on β-glucanase production by immobilized E. coli BL21(dE3) pET-bgl-hisactophilin-Sec cells.

Fig. 3. Repeated batch cultivation of E. coli BL21(dE3)pET-bgl-hisactophilin-Sec strain immobilized

in CeramTec PST 5 (1.5–2.5 mm).

β-glucanase production by immobilized E. coli2 137

the activity of free cells decreased after the first batch while that of the immobilized cells increased. High and stable enzyme production by immobilized cells in com-parison to free cell culture was obtained in cycles from 3 to 5 days (144 ) and activity was almost 1.3 times higher than that of that of free cells (500 U/ml). This result was compatible with those reported by other researchers (Manolov, 1993; Petruccioli et al., 1994; Petruccioli et al., 1996), who concluded that the activ-ity of free cells decreased after the first batch while that of the immobilized cells increased.

Scanning electron microscopy observations of immobilized E. coli cells. Fig. 4 shows an electron micrograph taken immediately after 24 h cultivation of immobilized

E. coli cells in CeramTec 1.5–2.5 mm (A) and CeramTec 4–5 mm (B). It is clear that a high number of cells also penetrated into the open pores within the beads and formed dense aggregates. Comparison of the electron micrographs indicates that there are a much higher number of E. coli cells in CeramTec (1.5–2.5 mm) than in CeramTec (4–5 mm). Images showed the immobi-lized cells on the surface and in the pores of a bead par-ticle after 24 h of cultivation. This shows the successful immobilization of E. coli on the studied matrices.

Conclusions. Of the three carriers assayed, the best for the immobilization for enzyme production and with adhered biomass on a laboratory scale are broken

Fig. 4. Scanning electron microscope images of immobilized E. coli BL21(dE3) pET-bgl-hisactophilin-Sec cellsin CeramTec PST 5: 1.5–2.5 mm (a, b, c) and 4–5 mm (d, e, f) after 24 h cultivation time.


pumice stone and CeramTec. Broken pumice stone allows the immobilization of a large number of cells (4 g cells/l of carrier) in a short time (24 h). Moreover, it leads to a high enzyme activity (550 U/ml), which makes the overall yields higher. Furthermore, it is an inert material and is very cheap, thus making the process potentially suitable for industrial scale-up. However, because of the abrasion problem of the broken pumice stone, which could lead to a serious error in the obtained results, CeramTec will be the second choice for bacterial immobilization. Glucanase activity reached by immobi-lized cells under batch cultivation is 650 U/ml after 60 h.

AcknowledgementsProf. Flaschel (University of Bielefeld, Fermentation Engineer-

ing dept.) is acknowledged for providing the Escherichia coli strain.

Literature

Beshay U. and A. Moreira. 2003. Repeated batch production of alkaline protease by using porous sintered glass as carriers. Process Biochemistry. 38: 1463–1469.Beshay U., H. El-Enshasy, I.M.K. Ismail, H. Moawad, E. Wojcie- chowska and S. Abd-El-Ghany. 2003. β-Glucanase production from genetically modified recombinant Escherichia coli: Effect of growth substrates and development of a culture medium in shake flasks and stirred tank bioreactor. Process Biochemistry 39: 307–313.Beshay U. Ph.D. Thesis 1998. Continuous cultivation with immo-bilized Dictyostelium discoideum, Faculty of Technology, University of Bielefeld, Germany. Borriss R., O. Olsen, K.K. Thomsen and D. Von Wettstein. 1989. Hybrid Bacillus endo-(1-3,1-4)-β-glucanases: Construction of recom-binant genes and molecular properties of the gene product. Carls-berg Res. Commun. 54: 41–54.

Colagrande O., A. Silva and M.D. Fumi. 1994. Recent applications of biotechnology in wine production. Rev. Biotechnol. Progr. 10: 2–18.Fiedurek J. and J. Lobarzewski. 1990. Glucoamylase biosynthesis by cells of Aspergillus niger C58-III immobilized in sintered glass and pumice stones. Starch. 42:3 58–362.Flaschel E., L. Poppenborg, R. Neitzel, G. Miksch and K. Friehs. 1998. Affinitaetstrennverfahahren zur Gewinnung Rekombinanter Protein. Biotech. 9:26–29.Giberson R.T., R.S. Jr Demaree and R.W. Nordhausen. 1997. Four-hour processing of clinical/diagnostic specimens for electron microscopy using microwave technique. J. Vet. Diagn. Invest. 9: 61–7.Gusek T.A., M.T. Tyn and J.E. Kinsella. 1991. Immobilization of the serine protease from Thermomonospora fusca Yx on porous glass. Biotechnol. Bioeng. 36: 411–416. Hamdy M.K., K. Kim and C.A. Rudtke. 1990. Continuous ethanol production by yeast immobilized on to channeled alumina beads. Biomass. 21: 189–206.Henry R. and C. Clim. 1964. Principles and Technics. Harper-Row. N. York, pp. 182.Kourkoutas Y., A. Bekatorou, I.M. Banat, R. Marchant and A.A. Koutinas. 2004. Immobilization technologies and support materials suitable in alcohol beverages production: a review. Food Microbiol. 21: 377–397.Manolov R.J. 1993. Ribonuclease production by free and immobilized Aspergillus clavatus cells. World J. Microbiol. Biotechnol. 9: 29–33.Miksch G., R. Neitzel, E. Fiedler, K. Friehs and E. Flaschel. 1997. Extracellular production of a hybrid β-glucanase from Bacillus by Escherichia coli under different cultivation conditions in shaking cultures and bioreactors. Appl. Microbiol. Biotechnol. 47: 120–126.Petruccioli M., P. Piccioni, M. Fenice and F. Federici. 1994. Glu-cose oxidase, catalase and gluconic acid production by immobilized mycelium of Penicillium variable P16. Biotechnol. Lett. 16: 939–942. Petruccioli M., E. Angiani and F. Federici. 1996. Semi-continu-ous Fumaric acid production by Rhizopus arrhizus immobilized in polyurethane sponge. Proc. Biochem. 31: 463–469.Weuster-Botz D. 1993. Continuous ethanol production by Xouto-monas mobilis in a fluidized bed reactor. Part I. Kinetic studies of immobilization in macroporus glass beads. Appl. Microbial. Bio-technol. 39:679–684.


ORGINAL PAPER

Introduction

The disposal of whey remains a significant problem for the dairy industry. As whey contains 5 to 6% dis-solved solids, including 3 to 5% lactose, the biological oxygen demand (BOd) is high. Generally, whey must be treated prior to discharge into the environment (Marwaha and Kennedy, 1988).

A number of applications for whey-permeate have been developed in an effort to overcome the problem of its disposal. One alternative is the use of whey as the basic medium for various fermentation processes including the production of ethanol, methane, yeast protein, xanthan gum (Fu and Tseng, 1990) or organic acids such as lactate, propionate or acetate (Mawson, 1994; Huang and Yang, 1998). Another application with a high tech-nological and dietetic interest is the enzymatic hydro-lysis of lactose, whose economic importance has been increasing ever since the 1960s (Novalin et al., 2005).

The enzyme β-d-galactoside galactohydrolase (β-galactosidase, E.C. 3.2.1.23, trivially lactase) hydrolyzes lactose, the milk sugar, into two moieties glucose and galactose (Rings et al., 1994). This technically and eco-nomically feasible process would also open new possi-bilities for the utilization of whey and whey-permeate (Zadow, 1993).

While β-galactosidase has been found in numerous biological systems, microorganisms such as yeasts, molds and bacteria still remain the only commercially exploited sources (Agrawal and dutta, 1989). More recently, thermophilic bacteria have become an object of inter-est for the commercial production of β-galactosidase (Petzelbauer et al., 1999). Among these, special atten-tion has been paid to lactic acid bacteria (LAB) because of their GRAS status (Stiles and Holzapfel, 1997). Lactic acid bacteria have a long tradition of use in the food industry. Their potential uses as a source of enzymes, especially β-galactosidase, has been shown to be prom-ising (Murad, 1998).

The aim of the current work was to optimize the growth conditions to maximize the production of β-galac-

tosidase by some LAB strains grown on the newly mod-ified permeate-based medium.

Experimental


Organisms. Seven LAB strains were obtained from the department of dairy Microbiology, National Research Centre. The obtained strains were Lactobacil-lus reuteri, Lactobacillus acidophilus, Lactococcus lactis

Utilization of UF-Permeate for Production of β-galactosidaseby Lactic Acid Bacteria

H.A. MURAd1*, R.I. REFAEA2 and E.M. ALY3

1 dairy Science department, National Research Centre, Cairo, Egypt2 Microbiology department, Faculty of Agriculture, Cairo University

3 Environmental Compliance Office, Federation of Egyptian Industries, Cairo, Egypt

Received 7 April 2010, revised 20 december 2010, accepted 15 January 2011

A b s t r a c t

Four lactobacilli strains (Lactobacillus bulgaricus, Lactobacillus acidophilus, Lactobacilus casei and Lactobacillus reuteri) were grown in MRS broth and three lactococci strains (Streptococcus thermophilus, Lactococcus lactis subsp. Lactis and Lactococcus lactis subsp. lactis biovar. diacetilactis) were grown in M17 broth. L. reuteri and S. thermophilus were chosen on the basis of the best mean β-galactosidase activity of 10.44 and 10.01 U/ml respectively, for further studies on permeate-based medium. The maximum production of β-galactosidase by L. reuteri was achieved at lactose concentration of 6%, initial pH 5.0–7.5, ammonium phosphate as nitrogen source at a concentration of 0.66 g N/L and incubation temperature at 30°C/24 hrs to give 6.31 U/ml. While in case of S. thermophilus, maximum β-galactosidase production was achieved at 10% lactose concentration of permeate medium, supplemented with phosphate buffer ratio of 0.5:0.5 (KH2PO4:K2HPO4, g/L), at initial pH 6.0–6.5, ammonium phosphate (0.66g N/L ) as nitrogen source and incubation temperature 35°C for 24 hrs to give 7.85 U/ml.

K e y w o r d s: L. reuteri, S. thermophilus, β-galactosidase, lactic acid bacteria (LAB), permeate

* Corresponding author: H.A. Murad; P.O.B 12622, dokki, Cairo, Egypt.; phone: 202 33068626; fax: 202 33370931; e-mail: [email protected]

Murad H.A. et al. 2140

subsp., lactis biovar diacetilactis, Lactobacillus bulgari-cus (241), Streptococcus thermophilus, Lactococcus lac-tis subsp. lactis and Lactobacillus casei. The obtained strains were maintained and activated in sterile litmus milk, stored in a refrigerator at 4°C and tested for their ability to produce β-galactosidase.

β-Galactosidase production mediaand growth conditions

Synthetic media. Production of β-galactosidase by lactococci and lactobacilli cultures was examined in M17 synthetic broth medium (difco) and MRS syn-thetic broth medium, respectively. The media were inoculated with freshly activated 1% inoculum and incubated at 37°C for 24 hrs according to Vasiljevic and Jelen (2001) and Herreros et al. (2003).

Permeate-based medium. The milk permeate, which is considered as a waste during the production of cheese by ultrafiltration, was obtained from a cheese production factory located in Mansoura City (Nile delta) and used for the preparation of the permeate-based medium. In the following tests, permeate-based medium components and environmental conditions were investigated for optimum conditions leading to maximize β-galactosidase production as mentioned by Murad (1998).The pH of permeate-based media in the experiments was adjusted to 6.5 by 1N NaOH or 1N HCl and then sterilized for 30 minutes at 110°C. The media were inoculated with 1% inoculum of the freshly activated strains (24 hrs) and then incubated at for 24 hrs 37°C.

Effect of media composition and environmentalconditions on the production of β-galactosidase

The previously mentioned permeate-based medium was employed for improvement of β-galactosidase pro-duction by varying its components qualitatively and quantitatively under variable environmental conditions as follows:

Effect of lactose concentration. different concen-trations of lactose in permeate medium (2, 4, 6, 8, 10 and 12%) were prepared either by diluting or adding lactose (difco) to the medium.

Effect of potassium phosphate. different concen-trations of KH2PO4 : K2HPO4 mixtures were added to permeate medium at a ratio concentration of 0.5 :0.5, 0.5 :1.0 and 0.5 :1.5 g/l, respectively, compared with the control.

Effect of initial pH. The influence of different initial pH values was examined. The pH values were adjusted at 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0 and 8.5 using either NaOH or 1 N HCl.

Effect of nitrogen source. Various nitrogen sources were used separately at an equivalent concentration of 0.33 g N/l media (Murad, 1998). The nitrogen sources

included 4 inorganic forms [(NH4)2PO4, (NH4)2SO4, NH4Cl and NaNO3] and 4 organic sources (yeast extract, beef extract, peptone and tryptone).

Effect of nitrogen concentration. The optimum nitrogen source chosen was tested for its most suitable concentration ranging from 0.165 to 0.990 g N/l.

Effect of incubation temperature. The effect of incubation temperature was studied. The incubation temperatures ranged from 25 to 45°C (with 5°C incre-ments), except for 37°C (the control), using electric chamber incubators.

Enzyme activity assay. β-galactosidase was assayed according to the method of Lederberg (1950) as described by Sanchez and Hardisson (1979). The method was strictly applied, except for the centrifu-gation speed which was used for separation of the bacterial cells and the sonicated cell debris at 4°C. The centrifugal speeds used were modified to 8,000 rpm/10 min and 15,000 rpm/20 min, respectively using Sigma 2K15 centrifuge.

Statistical analysis. The significances of the results were determined by the analysis of variance (ANOVA) evaluated by duncan’s multiple range tests (at 0.05), using COSTAT software, product of Cohort software Inc., Berkley, California, (duncan, 1955).


Production of β-galactosidase by selected LAB strains. As shown in Fig. 1, β-galactosidase production showed no significant differences between the 7 tested strains, ranging from 9.3 to 10.4 U/ml under the given experimental conditions. According to the obtained results, S. thermophilus and L. reuteri (representing lactococci and lactobacilli, respectively) were chosen on the basis of the best mean β-galactosidase activity of 10.01 and 10.44 U/ml, respectively for further studies on permeate-based medium.

Fig 1. β-galactosidase production by LAB strains M17 and MRS media expressed as enzyme activity (U/ml)

Permeate for β-galactosidase production by LAB2 141

Effect of lactose concentration. The data shown in Fig. 2 indicate that the production of β-galactosidase by L. reuteri and S. thermophilus strains increased with the increasing of lactose concentration (up to 6%), reach-ing 2.54 and 2.59 U/ml, respectively. Further increasing lactose concentration up to 10% decreased the L. reuteri enzyme production dramatically, while it increased the S. thermophilus enzyme production (3.01 U/ml). Hasan and durr (1974) stated that lactose induced the synthe-sis of β-galactosidase.

As a fact, the amounts of carbon source in the medium may affect the expression of β-galactosidase by microorganisms (Fiedurek and Szczodrak, 1994; Inchaurrondo et al., 1998). The β-galactosidase activity increased as the concentration of lactose in the medium was increased up to 4.0%. Further increasing the lactose content resulted in the reduction of β-galactosidase activity. A similar phenomenon was observed by Fiedurek and Szczodrak (1994) who investigated the biosynthesis of β-galactosidase by Kluyveromyces fragilis.

Inchaurrondo et al. (1998) clarified that the decreased β-galactosidase activity in the medium containing 5% or more lactose might be attributed to the increased concentration of internally released glucose which represses the biosynthesis of β-galactosidase by the test organism. Furthermore, it was demonstrated that 4% lactose was sufficient to induce the highest expression of β-galactosidase under the tested conditions.

This gives an indication about the efficiency of S. thermophilus to consume higher concentrations of lac-tose (10%) than L. reuteri (6%), indicated by the high-est production of β-galactosidase. Besides, the upper limit of lactose concentration that the two strains can produce β-galactosidase efficiently couldn’t be exceeded due to the accumulation of glucose by-products intra-cellularly, as discussed before.

Effect of potassium phosphate. The data shown in Fig. 3 illustrate the effect of potassium phosphate

(acidic and alkaline form) ratio on the β-galactosidase production by both strains. In case of L. reuteri there was no significant difference in β-galactosidase produc-tion magnitude for the control and the phosphate buffer treatments (KH2PO4:K2HPO4, g/L) at ratios 0.5:0.5 and 0.5:1.0 (2.38 U/ml, in average). Increasing the alkaline phosphate form of potassium decreased the produc-tion level significantly at the ratio of 0.5:1.5. Apparently, there is no need to add potassium phosphate buffer because the control gave similar level of β-galactosidase production, but there is always a great need for the buffer to capture any possible excess in organic acid by-product (lactic acid), specially when the permeate sources are varied.

The production of β-galactosidase by S. thermophi-lus showed a significant difference between the con-trol test and the whole added phosphate buffer forms ratios. At 0.5:1.0 and 0.5:1.5 ratios (KH2PO4:K2HPO4, g/L) the β-galactosidase production mean increased by 12% compared with the control, but at 0.5:0.5 ratio the increase of β-galactosidase production was 17.5% (2.68 U/ml), proving its dominancy over the control and other treatments.

These data agree with what Ramana Rao and dutta (1977) discussed in their work on S. thermophilus pro-duc tion for β-galactosidase that was obviously stimu-lated by the monobasic phosphate buffer form more than di- and tri-basic forms, while in the case of K. fragi-lis maximum β-galactosidase production was achieved by addition of KH2PO4:K2HPO4 buffer at a ratio of 1:3, as stated by Fiedurek and Szczodrak later (1994). Murad (1998) tested the β-galactosidase production level from L. bulgaricus using different KH2PO4:K2HPO4 buffer ratios at 1:2 and 1:3 and found that at the latter ratio the organism produced maximum β-galactosidase than at the former. Hsu et al. (2005) found that the best buffer ratio at which the Bifidobacterium produced maximum β-galactosidase was 1:3 (KH2PO4:K2HPO4).

Fig 2. Effect of lactose concentration in permeate-based medium on the production of β-galactosidase.

Fig 3. Effect of KH2PO4:K2HPO4 ratio (g/l) in permeate-based medium on the production of β-galactosidase.


Effect of initial pH. The obtained data showed no effect of various initial pH values on L. reuteri produc-tion of β-galactosidase, except that there was an obvious decline in the enzyme production at the alkaline border beginning at pH 8 and above. This proved the capability of L. reuteri to produce β-galactosidase at maximum level (3.21 U/ml, in average) through a wide range of initial pH (5.0 to 7.5). Maximum β-galactosidase pro-duction by S. thermophilus was obtained at pH 6.5 (2.77 U/ml). The experimental results revealed that the increase in the enzyme production was gradually increased at pH 5 up to 6.5, followed by significant decrease in the production at pH 7 up to 8.5.

The effect of the initial pH of the medium on enzyme production by S. thermophilus was studied over a pH range of 4.0 to 9.0 by Ramana Rao and dutta (1977) and they reported that maximum enzyme pro-duction was observed between pH 6.5 and 7.5. Sridhar and dutta (1991) worked on the production of β-galactosidase from Streptococcus cremoris on whey. They reported that the optimum pH was between 6.5 and 7. The results presented by Murad (1998) showed that the highest enzyme production by L. bulgaricus was obtained at pH 5, while, Hsu et al. (2005) reported the optimum initial pH for β-galactosidase produc-tion by Bifidobacterium sp. to be 6.5. It is obvious that there is a great similarity in the effect of optimum ini-tial pH (6.5) on the production of β-galactosidase by both strains (L. reuteri and S. thermophilus) and those reported previously by Ramana Rao and dutta (1977), Sridhar and dutta (1991) and Hsu et al. (2005).

The effect of nitrogen source. The observed results revealed that among the 8 nitrogen sources tested, the ammonium phosphate serving as a source for both nitrogen and phosphate, was found to be the best for maximum production of β-galactosidase by L. reuteri (5.74 U/ml) and S. thermophilus (6.83 U/ml). L. reuteri managed to use both organic and inorganic sources efficiently compared with the control, while, S. ther-mophilus preferred the simple nitrogen form (i.e. inor-ganic) more than the complex ones (i.e. organic).

This agreed with what was reported by Murad (1998), namely that the highest production of β-galactosidase by L. bulgaricus was obtained using (NH4)2HPO4. Also, he mentioned that no activity was detected in the presence of (NH4)2SO4. This fact was confirmed previ-ously by Selim and EL-diwany (1985) and Fiedurek and Szczodrak (1994) who reported that some salts such as (NH4)2SO4 showed an inhibitory effect on β-galactosidase production by K. fragilis.

On the contrary, Ramana Rao and dutta (1977) found that the best nitrogen source for the production of β-galactosidase by S. thermophilus was protease pep-tone, followed by ammonium sulphate, and they did not use any phosphate form of nitrogen.

These findings about the importance of nitrogen source were mentioned by Ramana Rao and dutta (1977) and Shaikh et al. (1997), who found that nitro-gen sources may affect the microbial biosynthesis of β-galactosidase.

Effect of nitrogen concentration. The preferred nitrogen source type (ammonium phosphate) chosen for both strains from the previous experiment was tested for its best concentration as shown in figure 4. It is clear that S. thermophilus production of β-galactosidase was much higher than L. reuteri over all the ammonium phosphate concentrations tested.

Both strains showed the same trend in response to nitrogen concentration. The β-galactosidase production significantly increased as the nitrogen concentration increased; until it reached its best level when nitrogen concentration was twice (0.66 g N/L) that of the con-trol (0.33 g N/L), as L. reuteri yielded 5.56 U/ml and S. thermophilus gave 7.11 U/ml, then declined when it reached triple that of the control.

This trend agreed with the findings of Hsu et al. (2005) in their work on Bifidobacteria. The activity of β-galactosidase increased upon increasing the nitrogen source concentration but further increasing resulted in a sharp reduction in the activity of β-galactosidase and a reduced final population of the test organism.

Murad (1998) found that the concentration of (NH4)2HPO4 in the growth medium of L. bulgaricus exhibited a profound effect on the production of β-galactosidase, as the highest activity was obtained using 0.4%. While in case of L. reuteri and S. thermophilus, the the highest β-galactosidase activity was obtained using 0.3% only.

It is worth mentioning that Jokar and Karbassi (2009) maximized β-galactosidase production by Lactobacil-lus delbruekii when grown in permeate based medium enriched with a combination of yeast extract, whey

Fig 4. Effect of supplemented nitrogen concentration in permeate-based medium on the production of β-galactosidase.

Permeate for β-galactosidase production by LAB2 143

powder and wheat steep liquor as organic nitrogen sources to reach 4.924 U/ml, less than that obtained in the current study (5.56 and 7.11 U/ml) using only ammonium phosphate as an inorganic source.

Effect of incubation temperature. L. reuteri strain produced maximum β-galactosidase (6.31 U/ml) at 30°C followed by a significant decline in enzyme production with further increase in the incubation temperature (Fig. 5). The same trend in the production of β-galactosidase by S. thermophilus strain was observed with its maximum enzyme production (7.85 U/ml) at 35°C.

Hsu et al., (2005) stated that the activity of β-galacto-sidase produced by Bifidobacteria increased as the cultivation temperature increased from 22°C to 37°C. Further increases in the cultivation temperature led to a reduction of enzyme production accompanied by a reduction in the final viable population. These obser-vations agree with those of Fiedurek and Szczodrak (1994) as well as Smith et al. (1985), which demon-strated that the highest β-galactosidase production by B. longum was obtained at 37 ºC.

It could be concluded that the permeate (4.5% lac-tose) as an industrial waste can be used efficiently in the production of β-galactosidase by either L. reuteri or S. thermophilus strains (6.31 and 7.85 U/ml, respec-tively) when grown at initial pH 6.5, if optimized by adding KH2PO4:K2HPO4, at ratio of 0.5:0.5 g/L and ammonium phosphate at 0.66 g N/L as nitrogen source, with an incubation temperature of 30°C and 35°C, respectively.

Literature

Agrawal S. and S.M. Dutta. 1989. Microbial β-galactosidase: pro-duction, properties and industrial applications. Indian Journal of Dairy Science 42: 251–262.

Duncan D. B. 1955. Multiple ranges and multiple F-Tests. Biomet-rics 11: 1–24.Fiedurek J. and J. Szczodrak. 1994. Selection of strain, culture conditions and extraction procedures for optimum production of β-galactosidase from Kluyveromyces fragilis. Acta Microbiol. Pol. 43: 57–65. Fu J. and Y. Tseng. 1990. Construction of lactose-utilizing Xan-thomonas campestris and production of xanthan gum from whey, Appl. Environ. Microbiol. 56: 919–923.Hasan N. and I.F. Durr. 1974. Induction of β-Galactosidase in Lac-tobacillus plantarum. Journal of Bacteriol. 120: 66–73.Herreros M.A., J.M. Fresno, M.J. Gonzalez Prieto and M.E. Tor-nadijo. 2003. Technological characterization of lactic acid bacteria isolated from Armada cheese (a Spanish goats’ milk cheese). Inter-national Dairy Journal 13: 469–479.Hsu C.A., R.C. Yu and C.C. Chou. 2005. Production of β-galacto-sidase by Bifidobacteria as influenced by various culture conditions. International Journal of Food Microbiology 104: 197–206.Huang Y. and S. Yang. 1998. Acetate production from whey lactose using co-immobilized cells of homolactic and homoacetic bacteria in a fibrous-bed bioreactor. Biotechnol. Bioeng. 60: 498–507. Inchaurrondo V.A., M.V. Flores and C.E. Voget. 1998. Growth and β-galactosidase synthesis in aerobic chemostat cultures of Kluyvero-myces lactis. J. Ind. Microbiol. Biotech. 20: 291– 298. Jokar A. and A. Karbassi. 2009. determination of proper conditions for the production of crude beta-galactosidase using Lactobacillus delbrueckii ssp. bulgaricus. J. Agric. Sci. Technol. 11: 301–308Lederberg J. 1950. The beta-d-galactosidase of Escherichia coli, strain K-12. J. Bacteriol. 60: 381–392.Marwaha S.S. and J.F. Kennedy. 1988. Review: Whey pollution problem and potential utilization. Int. J. Food Sci. Technol. 23: 323–336. Mawson A.J. 1994. Bioconversions for whey utilization and waste abatement. Bioresource Technol. 47: 195–203. Murad H.A. 1998. Utilization of ultrafiltration permeate for pro-duction of β-galactosidase from Lactobacillus bulgaricus. Milchwis-senchaft 53: 273–276.Novalin S., W. Neuhaus and K.D. Kulbe. 2005. A new innovative process to produce lactose-reduced skim milk. J. Biotechnol. 119: 212–218.Petzelbauer I., B. Nidetzky, D. Haltrich and K.D. Kulbe. 1999. development of an ultra high temperature process for the enzy-matic hydrolysis of lactose. I. The properties of two thermostable β-glycosidases. Biotechnology and Bioengineering 64: 322–332. Ramana Rao M.V. and S.M. Dutta. 1977. Production of Beta-Galactosidase from Streptococcus thermophilus Grown in Whey. Appl. and Environ. Microbiol. 34: 185–188.Rings E.H.H.M., E.H. Van Beers, S.D. Krasinski, M. Verliave, R.K. Montgomery, R.J. Grand, J. Dekker and H.A. Büller. 1994. Lactase: origin, gene expression, localization and function. Nutrition Research 14: 775–797. Sanchez J. and C. Hardisson. 1979. Induction of β-galactosidase in Streptomyces violaceus. Can. J. Microbiol. 25: 833–840.Selim M.H. and A.J. El-Diwany. 1985. Chem. Mikrobiol. Technol. Lebensm. 9: 81–86. Cited in: Murad, H.A. 1998. Utilization of ultra-filtration permeate for production of β-galactosidase from Lactoba-cillus bulgaricus. Milchwissenchaft 53: 273–276.Shaikh S.A., J.M. Khire and M.I. Khan. 1997. Production of β-galactosidase from thermophilic fungus Rhizomucor sp. J. Ind. Microbiol. Biotech. 19: 239–245.Smith P.K., R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson and D.C. Klenk. 1985. Measurement of protein using bicinchoninic acid. Anal. Biochem. 150: 76–85. )

Fig 5. Effect of incubation temperature on productionof β-galactosidase.


Sridhar N. and S.M. Dutta. 1991. Ind. J. Dairy Sci. 44: 283. Cited in: Murad, H.A. 1998. Utilization of ultrafiltration-permeate for pro-duction of β-galactosidase from Lactobacillus bulgaricus. Milchwis-senchaft 53: 273–276.Stiles M. and W.H. Holzapfel. 1997. Lactic acid bacteria of foods and their current taxonomy. International Journal of Food Micro-biology 36: 1–29. Vasiljevic T. and P. Jelen. 2001. Production of β-galactosidase for lactose hydrolysis in milk and dairy products using thermophilic

lactic acid bacteria. Innovative Food Science & Emerging Techno-logies 2: 75–85.Zadow J.G. 1993. Economic considerations related to the production of lactose and lactose by-products. Lactose hydrolysis, IdF Bulletin 289. IdF, Brussels, pp. 10–15. Cited in: T. Vasiljevic and P. Jelen. 2001. Production of β-galactosidase for lactose hydrolysis in milk and dairy products using thermophilic lactic acid bacteria. Innovative Food Science & Emerging Technologies 2: 75–85.


ORGINAL PAPER

Introduction

dermatophytes are a group of keratinophilic fungi causing infections called dermatophytoses. These microorganisms belong to the 3 genera of fungi, Trichophyton, Microsporum and Epidermophyton. Microsporum canis, zoophilic dermatophyte, is responsible for most cases of tinea capitis in children and tinea corporis in adults. The geographic spread of this infection is worldwide, because it is transmitted by cats (the principal reservoir of this fungus) or dogs. Human infection caused by M. canis may occur by direct contact with infected animals or with their hair. M. canis infections were also reported in humans who did not have a history of exposure to animals, which suggests that the infection may be spread indirectly from other humans colonized with the fungus or by contact with arthrospores that have con-taminated object such as grooming equipment, furni- ture or environment (Weitzman and Summerbell, 1995).

Molecular PCR-based techniques provide a means of comparing isolates from patients and domestic ani-mals to determine a source of contamination. Faggi et al. (2001) described the application of PCR for molecu-lar identification and differentiation of dermatophyte species and strains using a simple repetitive oligonu-

cleotide (GACA)4 as a primer, which was previously successfully used by Meyer et al. (1993, 1997) to distin-guish strains of Cryptococcus neoformans and Candida species. In case of clinical strains of M. canis isolated in Italy from humans, cats and dogs, the authors obtained only species-specific profiles, there was no intraspecies variation among them. The same results were obtained by Shehata et al. (2008) who also employed (GACA)4 primer for molecular analysis of dermatophytes iso-lated from Egyptian patients. Our results using PCR MP method (Leibner-Ciszak et al., 2010) revealed also only one genotype among M. canis strains isolated in Central Poland. It seems that the lack of intraspecies polymorphism could be due to the genetic stability of M. canis genome. These data are in conflict with the results obtained by Cano et al. (2005), who found a total of 21 genotypes among 24 human isolates of M. canis using ISSR-PCR (Inter-Simple-Sequence Repeat PCR) what suggested that distribution of this zoophilic der-matophyte was restricted even to a single patient.

In the present study, we have used (GACA)4 and (ACA)5 primer used previously by Shehata et al. (2008) and Cano et al. (2005), respectively to analyze M. canis strains isolated from patients and from animals, mainly from Central Poland.

Strains Differentiation of Microsporum canis by RAPD Analysis Using (GACA)4and (ACA)5 Primers

ANITA dOBROWOLSKA1*, JOANNA dęBSKA1, MAGdALENA KOZłOWSKA2 and PAWEł STąCZEK1

1 department of Genetics of Microorganisms, University of łódź2 department of dermatology and Venerology, Medical University of łódź

Received 15 december 2010, revised 21 January 2011, accepted 2 February 2011

A b s t r a c t

Molecular analysis of dermatophytes (based on PCR fingerprinting) revealed high clonal differentiation between the genus and species. Microsporum canis (zoophilic dermatophyte, belonging to genus Microsporum), responsible for most cases of tinea capitis in children, tinea corporis in adults and dermatophytoses in cats, is very unique in comparison with other dermatophytes. Results of most molecular studies show that there is no clonal differentiation within M. canis as distinct from other species. The aim of this study was application of (GACA)4 repetitive primer and (ACA)5 primer for typing of M. canis strains isolated from human and animals in Central Poland. Fungal strains: 32 clinical isolates of M. canis, originated from patients from Central Poland; 11 strains isolated from infected cats (6) and dogs (7), reference strains of M. canis (CBS 113480), T. rubrum (CBS 120358), T. mentagrophytes (CBS 120357) and E. floccosum (CBS 970.95). The genomic dNAs of the strains were used as a template in RAPd reaction. No differentiation was observed for the analyzed M. canis strains using (GACA)4 and (ACA)5 typing.

K e y w o r d s: M. canis, dermatophytes, strains differentiation, RAPd method

* Corresponding author: A. dobrowolska, University of łódź, department of Genetics of Microorganisms; 90-237 łódź, Poland, ul. Banacha 12/16; phone: +48 42 635-47-72; e-mail: [email protected]

dobrowolska A. et al. 2146

Experimental


Fungal strains. Thirty-two clinical isolates of M. ca - nis, canis, seven strains isolated from dogs and six strains isolated from cats were used in this study (Table I). Among them, 12 were isolated from females between 2 and 54 years old, 20 were isolated from males between 3 and 64 years old. Traditional identification based on the fungus morphology observed under microscope, was performed in the department of der-matology and Venerology, Medical University of łódź.

DNA extraction. Genomic dNA was extracted from a small amount of mycelium by a self-modified mini preparation method (Liu et al., 2000). Mycelium was suspended in 700 µl of lysis buffer (400 mM Tris-HCl, 60 mM EdTA, 150 mM NaCl, 1% sodium dodecyl sulfate), and incubated at 60°C for 1 hour. After addi-tion of 210 µl of 3 M sodium acetate, the homogen-ate was centrifuged at 12 000 rpm for 10 minutes. The supernatant was successively extracted with phenol-chloroform-isoamyl alcohol (25:24:1). dNA was treated with RNAase at a final concentration of 50 µg/ml for 20 minutes at 57°C. Than, samples were precipitated using 3 volumes of cold ethanol in the presence of 300 mM sodium acetate and dNA was centrifuged for 10 minutes. The pellet was washed with 70% ethanol and air dried. dNA was dissolved in 50 μl 1x TE buffer, 1 µl of the resulting solution was used as a template in the following PCR reaction.

Molecular identification by PCR-RFLP method. Morphological identification was confirmed by PCR-RFLP identification targeting ITS1-5.8S-ITS2 region and using HinfI restriction enzyme (dobrowolska et al., 2006).

(GACA)4 and (ACA)5 typing. PCR amplification was performed using (GACA)4 and (ACA)5 primers. Each PCR mixture (30 µl) contained 1 µl of genomic dNA, 1 µl of dMSO, 4 µl of 25 mM MgCl2, 0.5 µl of 50 pmol (GACA)4/(ACA)5 primer, 8.5 µl of distilled water and 15 µl of MasterMixTaq Polymerase (Qiagen). PCR was performed as described previously by She-hata et al. (2008) and Cano et al. (2005). detection of PCR products was performed by electrophoresis in 1% agarose gel stained with ethidium bromide and visu-alized by UV light.

Results

Molecular identification of M. canis isolates. The genomic dNAs of the reference strain CBS 113480 and forty-five examined M. canis isolates from patients and animals in Central Poland were amplified using univer-

sal primers ITS1 and ITS4 (Jackson et al., 1999). The size of the obtained PCR products was approximately 700 bp for all strains. We performed RFLP analysis for all 45 PCR products and the reference strain, using HinfI restriction enzyme (dobrowolska et al., 2006). Additionally using computer software we generated hypothetical RFLP profiles (213, 194, 142, 108, 8 bp) based on the dNA sequence (AF 168127) of exam-ined region. HinfI digestion of ITS1-5.8S-ITS2 region revealed a pattern unique for M. canis (Fig. 1).

RAPD analysis. All of the examined M. canis isolates obtained from the patients and animals in Poland gave band patterns after RAPd amplification with (GACA)4 and (ACA)5 primer. Among them, we distinguished only one type, designated A, which was characteristic for all isolates of M. canis originating from Poland and for the reference strain. The (GACA)4 fingerprints of 45 analyzed isolates, yielded up to 11 bands, ranging from approximately 300 to 2000 bp in length (Fig. 2) while the results obtained by Shehata et al. (2008) revealed profiles, with up to 11 bands but ranging from 600 bp to 2500 bp. In case of the (ACA)5 fingerprints we observed two bands, approximately 500 bp, 850 bp (Fig. 3).

Discussion

dermatomycosis induced by M. canis has became a serious problem in Poland in recent years. Epidemio-logical studies conducted in Poland between January 1,

Fig. 1. Exemplary polyacrylamide-gel electrophoresisof PCR products digested with HinfI restriction enzyme.

The ITS1-ITS4 set of primers was used to amplify ITS1-5,8S rdNA-ITS2 region. Obtained profiles correspond to Microsporum canis.

Abbrevations above the lanes correspond to the species names assigned during traditional identification (see Table I); bp – base pair.

Strains differentiation of Microsporumcanis by RAPd2 147

1989 and december 31, 2007 (Macura et al., 2008) revealed an increase in frequency of tinea corporis and tinea capitis. For example, in this period of time infections caused by M. canis raised from 2.64% up to 5.17%. The authors explained that the increase in M. canis infection resulted from buying domestic ani-mals like cats, dogs, hamsters which can be a source of infection of this dermatophyte. The infection agents are arthrospores, which are asexual spores (present in the parasitic state of M. canis) formed by segmentation of the fungal hyphae. Sexual spores are absent in the parasitic phase (Sparkers et al., 2000).

Molecular analysis of dermatophytes revealed diver-sity between genus and species. M. canis is very unique in comparison with other dermatophytes. Typing me - thods based on molecular markers used in Randomly Amplified Polymorphic dNA (RAPd) (Mochizuki et al., 1997; Kaszubiak et al., 2004; Gräser et al., 2000), the specific amplification of internal transcribed spacer and non-transcribed spacer regions of rdNA genes

(Mochizuki et al., 2003; Yazdanparast et al., 2003) were performed, but the level of intraspecies polymorphisms within M. canis isolates was low.

In our study, we have utilized (GACA)4 repetitive primer previously used by Faggi et al. (2001) and Shehata et al. (2008) and (ACA)5 primer previously used by Cano et al. (2005) for differentiation of M. canis strains iso-lated from patients and animals in Central Poland. The results obtained by Faggi et al. (2001) showed that there was no intraspecies variability among 49 analyzed M. canis strains isolated from human and animals (cats and dogs) in Spain. Shehata et al. (2008) analyzed only 4 strains of M. canis isolated from Egyptian patients and among them also identified only one genotype. Our (GACA)4 typing results are in agreement with those mentioned above, as we distinguished only one geno-type (A) among analyzed 45 M. canis strains isolated from patients and animals. This can be due to the low frequency of changes in dNA among populations of analyzed strains. Results confirming this thesis have been shown by Sharma et al. (2007), who observed over- representation of one genotype (cluster I) containing 74% of the human isolates and 23% of the animals strains, which had a higher degree of virulence and had a pandemic distribution. The authors showed that some M. canis strains had a higher infective potential to humans but there was no association between genotypes and type of tinea caused by this zoophilic dermatophyte.

In case of different geographical origins of the M. canis used in the present study and those used by Shehata et al. (2008), the obtained profiles consisted of 11 bands but the size range was slightly different. These results may suggest that (GACA)4 primer has utility for differentiation of M. canis strains isolated from patients and animals geographically isolated areas. However, RAPd techniques causes some problems regarding reproducibility especially when the results are compared between the laboratories. Variation in the obtained patterns can occur as a results of subtle differences in the primer or template concentration, the temperature variations during amplification or differ-ences in magnesium concentration in the reaction mix-ture (Ellsworth et al., 1993).

Interestingly, T. rubrum (120358) and T. mentagro-phytes (120357) originated from CBS (Centraalbureau voor Schimmelcultutes, Utrecht, The Netherlands) gave the same pattern using (GACA)4 and (ACA)5 primer (Fig. 2, Fig. 3) what may suggest that these oligonucleo-tides do not have very high differentiation power. On the other hand (GACA)4 – based PCR revealed the presence of four profiles among Trichophyton ajelloi strains (geophilic dermatophyte) isolated from soil in Poland (dobrowolska et al., unpublished data).

The (ACA)5 – based typing of analyzed M. canis col-lection gave only one pattern (A; Fig. 3). These results

Fig. 2. Exemplary amplification profiles of dNA fragments by RAPd using repetitive primer (GACA)4.

M – molecular marker 1Kb Plus, K– negative control, K+ – positive control (1 – T. rubrum, 2 – T. mentagrophytes, 3 – E. floccosum), profile A corresponds to Microsporum canis strains isolated from parients and animals in Central Poland. Abbrevations above the lanes correspond to the species names

assigned during molecular identification (see Table I); bp – base pair.

Fig. 3. Exemplary amplification profiles of dNA fragments by RAPd using primer (ACA)5.

M – molecular marker 1 Kb Plus, K- negative control, K+ – positive control (1 – T. rubrum, 2 – T. mentagrophytes, 3 – E. floccosum), profile A corresponds to Microsporum canis strains isolated from parients and animals in Central Poland. Abbrevations above the lanes correspond to the species names

assigned during molecular identification (see Table I); bp – base pair.

dobrowolska A. et al. 2148

are in conflict with data obtained by Cano et al. (2005) who proposed ISSR-PCR (Inter Single Sequence Repeat PCR) for typing of M. canis isolates. differentiation of 21 genotypes among 24 examined isolates of M. canis using (ACA)5 and (CCA)5 was very promising but the authors did not confirm traditional mycological identification with a molecular method such as e.g. PCR-RFLP (which was done in this study) and no reference strain was analysed. Sharma et al. (2007) disputed the high discriminatory power of ISSR-PCR, explaining it its low reproducibility. Other studies in which typing of M. canis strains was conducted, e.g. using RAPd, reported low variability among epidemio-logically unrelated strains from cats, dogs and humans, despite their morphological diversity (Faggi et al., 2001; Brilhante et al., 2005).

In our study clinical isolates of M. canis were obtained from patients and animals from łódź, thus it cannot be excluded that this area is dominated by the sin-gle clone of M. canis. However, the reference strain, which was isolated in Germany, has given the identical pattern. This problem would be probably solved after examina tion of the greater number of M. canis isolates from different regions of Poland and other countries, as well as after application of other methods such as typing method based on microsatellite markers using specific primers, McGT17 and McGT13, proposed by Sharma et al. (2007). Such analyses are actually per-formed by our group.

AcknowledgementsThis work was supported by grant (N 303 568938) from The

Ministry of Science and Higher Education. This research was also supported by the European Union within the European Social Fund and National Budget in the framework of the Operation 2.6 Inte-grated Operational Program of Regional development, in connec-tion with realization of project “Scholarships supporting innovative scientific researches of Phd students”

Literature

Brilhante R.S., R.A. Cordeiro and M.F. Rocha. 2005. Antifungal susceptibility and genotypical pattern of Microsporum canis strains. Can. J. Microbiol. 51: 507–510.Cano J., A. Rezusta and J. Guarro. 2005. Inter-single-sequence-repeat-PCR typing as a new tool for identification of Microsporum canis strains. J. Dermatol. Sci. 39: 17–21.

Dobrowolska A., P. Stączek and M. Kozłowska. 2006. PCR-RFLP analysis of the dermatophytes isolated from patients in Central Poland. J. Dermatol. Sci. 42: 71–74.Ellsworth D.L., K.D. Rittenhouse and R.L. Honeycutl. 1993. Arti-factual variation in tandomly amplified polymorphic dNA banding patterns. Biotechniques 14: 214–217.Faggi E., G. Pini and F. Mancianti. 2001. Application of PCR to distinguish common species of dermatophytes. J. Clin. Microbiol. 39: 3382–3385.Gräser Y., A.F.A. Kuijpers and G.S. de Hoog. 2000. Molecular taxonomy of the Trichophyton rubrum complex. J. Clin. Microbiol. 38: 3329–3336.Jackson C.J., R.C. Barton and E.G. Evans. 1999. Species identifi-cation and strain differentiation of dermatophyte fungi by analysis of ribosomal dNA intergenic spacer regions. J. Clin. Microbiol. 37: 931–936.Kaszubiak A., S. Klein and Y. Gräser. 2004. Population structure and evolutionary origins of Microsporum canis, M. ferrugineum and M. audouinii. Infect. Genet. Evol. 4: 179–186.Leibner-Ciszak J., A. Dobrowolska and P. Staczek. 2010. Evalua-tion of a PCR melting profiles of Trichophyton rubrum and Tricho-phyton interdigitale. J. Med. Microbiol. 59: 185–192.Liu D., S. Coloe and J. Pederson. 2000. Rapid mini-preparation of fungal dNA for PCR. J. Clin. Microbiol. 38: 471.Macura A.B., P. Krzysciak and M. Bochenek. 2008. Trends in the spectrum of dermatophytes causing superficial mycoses in the past decade. Mikol. Lek. 15: 76–79. Meyer W., T.G. Mitchell and R. Vilgalys. 1993. Hybrydization probes for conventional dNA fingerprinting used as single primers in the polymerase chain reaction to distinguish strains of Cryptococ-cus neoformans. J. Clin. Microbiol. 31: 2274–2280.Meyer W., G.N. Latouche and T. Sorrel. 1997. Identification of pathogenic yeasts of the imperfect genus Candida by polymerase chain reaction fingerprinting. Electrophoresis 18: 1548–1559.Mochizuki T., N. Sugie and M. Uehara. 1997. Random amplifica-tion of polymorphic dNA is useful for differentiation of several antrhopophilic dermatophytes. Mycoses 40: 405–409.Mochizuki T., H. Ishizaki and E.G.V. Evans. 2003. Restriction frag-ment length polymorphism analysis of ribosomal dNA intergenic regions is useful for differentiating strains of Trichophyton menta-grophytes. J. Clin. Microbiol. 41: 4583–4588.Sharma R., S. de Hoog and Y. Gräser. 2007. A virulent genotype of Micropsorum canis is responsible for the majority of human infec-tion. J. Med. Microbiol. 56: 1377–1385.Shehata A.S., P.K. Mukherjee and M.A. Ghannoum. 2008. Single-step PCR using (GACA)4 primer: utility for rapid identification of dermatophyte species and strains. J. Clin. Microbiol. 46: 2641–2645.Sparkers A.H., A. Robinson and S.E. Shan. 2000. A study of the efficacy of topical and systemic therapy for treatment of M. canis infections. J. Feline Med. and Surg. 2: 135–142.Weitzman I. and R.C. Summerbell. 1995. The dermatophytes. Clin. Microbiol. Rev. 8: 240–259.Yazdanparast A., C.J. Jackson and E.G.V. Evans. 2003. Molecular typing of Trichophyton rubrum indicated multiple strain involve-ment in onychomycosis. Br. J. Dermatol. 148: 51–54.


ORGINAL PAPER

Introduction

Species of Aeromonas are widely distributed in nature, and some of them are of medical and veterinary interest (Albert et al., 2000; Beutin et al., 2004). Aeromonas spe-cies are now recognized as important enteropathogens. Some strains are able to infect the human gastrointes-tinal tract and possess virulence properties such as the ability to produce enterotoxins, cytotoxins, haemolysins and/or the ability to invade epithelial cells (Figueras, 2005; Coburn et al., 2007; Janda and Abbott, 2010).

Avian species such as chicken, normally have high body temperatures (40 to 42°C). Therefore, it was note-worthy to hypothesize that microorganisms, such as Aeromonas spp., responsible for a variety of diseases in poultry respond to tissue invasion with the production of heat shock proteins. Moreover, in several bacterial species heat shock proteins have been shown to play an important role in pathogenesis (Lathigra et al., 1991; Kaufmann, 1992; Mauchline et al., 1994; Macario, 1995; Schurr and deretic, 1997). It was therefore proposed that heat shock proteins produced by bacteria affecting

avian hosts, may be virulence determinants, since with-out the ability to produce this unique class of protein, bacteria would be incapable of producing disease (Love and Hirsh, 1994).

Bacterial heat-shock response is a global regulatory system required for effective adaptation to changes (stress) in the environment. The heat-shock response involves the induction of many proteins – called heat-shock proteins, or HSPs in response to rise of tempera-ture (Neidhardt and Van Bogelen, 1987). Induction of this response improves thermotolerance (Kusukawa and Tura, 1988; LaRossa et al., 1991; Volker et al., 1992).

Consequently, as HSP produced by bacteria may be virulence determinants, it was essential to deter-mine the influence of temperature and time of incu-bation on their production in vitro. This initiated the present endeavor to determine and evaluate whether Aeromonas species exhibits a heat shock response to different temperatures and time factors and to indicate any diversity between the HSPs produced by Aero-monas hydrophila, A. caviae and A. veronii biovar sobria through dendrogram analysis.

SDS-PAGE Heat-Shock Protein Profiles of EnvironmentalAeromonas Strains

KAMELIA M. OSMAN1*, ZEINAB M. S. AMIN2, MAGdY A.K. ALY3,HANY HASSAN4 and WALEEd S. SOLIMAN5

1 department of Microbiology, Faculty of Veterinary Medicine, Cairo University, Egypt2 department of Poultry diseases, National Research Center, dokki, Egypt

3 department of Microbiology, National Research Centre, dokki, Egypt4 department of Immunology, Animal Reproduction Research Institute, El-Haram, Egypt

5 department of Fish diseases, National Research Center, dokki, Egypt

Received 23 August 2010, revised 15 February 2011, accepted 20 February 2011

A b s t r a c t

Aeromonas microorganisms normally grow at temperatures between 5°C and 45°C and therefore should have high thermotolerance. Thus it was of interest to find out whether A. hydrophila, A. caviae and A. veronii biovar sobria serovars respond to abrupt temperature changes with a heat shock-like response. To this end the present study was undertaken to determine whether Aeromonas species exhibits a heat shock response to different temperatures and time factors. The response of Aeromonas serovars to 24 h and 48 h of thermal stress at 25°C, 42°C and 50°C involved the synthesis of 12–18 heat shock proteins (HSPs) bands with molecular weights ranging between 83.5–103.9 kda in the high HSP molecular mass and 14.5–12.0 as low molecular mass HSP. Electrophoretic analysis of the HSPs showed that the serovars do not cluster very tightly and also that they are distinct from each other.

K e y w o r d s: environmental Aeromonas spp., heat-shock proteins (HSP), SdS-PAGE

* Corresponding author. K.M. Osman, department of Microbiology, Faculty of Veterinary Medicine, Cairo University, Egypt; phone: +20233854762; e-mail: [email protected]

Osman K.M. et al. 2150

Experimental


Aeromonas strains and growth conditions. Eighty-two strains of presumptive Aeromonas spp. isolated from the sucrose positive strains, which were mainly associ-ated with epizootic outbreaks that occurred in chicken and fish farms from 2006 to 2007, were stored long-term as freeze-dried cultures in 7.5% horse glucose serum as a cryoprotector. They were genetically re-identified although previously biochemically identified as Aero-monas spp. isolated from diseased chicken and fish.

Phenotypic identification. All environmental strains were cultured aerobically on trypticase soy agar (TSA) at 25°C for 24 h. The Aeromonas isolates were tested for their physiological and biochemical properties as previously described (Abbott et al., 2003; Murray et al., 2003). Before each test, all the cultures were grown on TSA (Oxoid) at 37°C for 18 h. Strains were first iden-tified as Aeromonas spp. and growth in 6% sodium chloride was used to discriminate Aeromonas from Vibrio fluvialis, V. splendidus, V. harvevyi and V. algine-

lyticus (Castro-Escarpulli et al., 2003). All Aeromonas spp. were re-identified biochemically by using 14 tests which were: indole, gas from glucose, cytochrome oxi-dase, nitrate reduction, arginine dihydrolase, lysine decarboxylase and ornithine decarboxylase by the Moeller’s method, esculin hydrolysis, Voges-Proskauer test, acid production from L-arabinose, lactose, sucrose, salicin, m-inositol, d-manitol and the h-hemolysine. To identify the biovars of Aeromonas veronii, the fol-lowing tests were used: arginine dihydrolase, ornithine decarboxylase, acid from salicin and esculin hydrolysis.

Serological testing. All motile Aeromonas strains had been previously tested by the O-serogrouping system of the Hydrobiology department (National Research Centre) according to Eurell et al. (1978). Strains were grown on TSA slants overnight at 30°C, har-

vested with phosphate-buffered saline (>109 cells/ml), and heated for 1 h at 100°C. After being heated, 20 µl of the boiled cell suspensions (thermostable O antigen of the strains) was mixed with 20 µl of each specific rab-bit antiserum (O:1 to O:30) in ceramic rings on aggluti-nation glass sides. The mixtures were rotated for 2 min, and the degree of agglutination (0 to 2+) was recorded. Two negative controls were used, boiled cell suspen-sions mixed with phosphate-buffered saline and boiled cell suspensions mixed with rabbit serum obtained from nonimmunized animals.

Molecular identification. All strains were re-identified on the basis of the restriction fragment length polymor-phism patterns (RFLP) obtained from the 16S rdNA (Ausubel et al., 1994). Aeromonas strains ATCC 7966, ATCC 43979, and ATCC 15468, and Escherichia coli (ATCC 25922) and Staphylococcus aureus (ATCC 25923), were included as quality controls.

Bacterial DNA extraction for PCR. Approximately 100 µl of a TSA culture grown for 16 h at 28°C was used for dNA extraction using the InstaGene matrix (Bio-Rad Laboratories AG, Glattbrugg, Switzerland) in accordance with the manufacturer’s instructions. Sub-sequently, 5 µl of the dNA solution was used as a tem-plate for PCR amplification. The sequence, specifici-ties, the primer combination and the size and length of the amplified products are summarized in Table I. The general reaction conditions in a Gene Cycler were 94°C for 1 min, 60°C for 1 min and 72°C for 3 min. This was repeated for 35 cycles. The program also included a preincubation at 94°C for 2 min before the first cycle and an incubation at 72°C for 3 min followed by cooling at 4oC after the last cycle.

Stress induction. Environmental strains of A. hydrophila, A. caviae and A. veronii biovar sobria each was separately grown in 250 ml of Luria-Bertani (LB) broth containing 0.3 M NaCl. The HSP bands were visible if the strains were grown in a broth medium but not after cultivation on an agar medium. Heat shock stress

Aeromonas species Aero-F GGAAACTTCTTGGCGAAAAC 20 IGS 550 Aero-R GGTTCTTTTCGCCTTTCCCT 20 23SA. hydrophila A-hydro-F CCAAACGAGAGAAGCCCTT 19 iGS 700 A-hydro-R CCATTCCACTAACTTCCAAGAA 23SA. caviae Acav-F CGCGCCGTTGCAAACATG 18 IGS 320 Acav-R GCGATACCTAGCTTATGCTAA 21 23SA. veronii biovar sobria Aver-F TGGTAGCTAATAACTGCCAG 20 16S 1170 Aver-R GGCTTCTCTCGTTTGGCGT 19 IGS

Table IOligonucleotide primer sequences and size of the PCR-targeted products of selected species of Aeromonas

SpecificitySize

of amplifiedproduct (bp)

LocationLength(bp)Primer sequence (5’–3’)

Forwardand Reverse

sequence

Heat shock proteins of environmental Aeromonas strains2 151

conditions were imposed according to the following procedures: 1) each strain was cultured at a tempera-ture of 42°C and 50°C, 2) at each temperature, samples were taken at intervals after heat shock, and crude cell extracts were prepared. Culture samples were taken after 3 h, 6 h, 24 h, 48 h and 72 h.

Analysis of Aeromonas spp. culture supernatant proteins by SDS-PAGE. Following the procedure of Love and Hirsh (1994), bacterial cells were centrifuged (15,600×g, 5 min, room temperature). Bacteria were lysed in gel sample buffer consisting of 9.5 M urea, 2% Nonidet P-40, 2% ampholytes (Bio-Lyte 5/7 [Bio-Rad Laboratories, Richmond, Calif.] and 2-d Pharmalyte [Pharmacia LKB, Uppsala, Sweden]), and 5% 2-mer-captoethanol. An equal volume of glass beads (212- to 300-[Lm diameter; Sigma Chemical Co., St. Louis, Mo.) was then added to each tube. After the tubes were vortexed for 3 min, they were centrifuged (15,600×g, 5 min, room temperature). The proteins in the super-natants were separated by electrophoresis through a 3% stacking gel and a 10% separating gel. Polyacrylamide gel electrophoresis, Coomassie blue staining analysis of proteins was carried out by standard protocols.

Computer-aided analysis of the gels. Images of the gels were captured using a Sharp Jx-330 flat-bed scan-ner, and image analysis of the protein profiles was perfor-

med using Amersham Pharmacia Biotech Image Master 2-d Elite software. The relative amount of each pro-tein spot was calculated and expressed by the software as the percentage of the spot volume and represented

the intensity of each individual spot compared to the intensity of the whole gel. The genetic similarity coef-ficient between two genotypes was estimated according to dice. The similarity-derived dissimilarity matrix was used in the cluster analysis by using the unweighted pair-group method with arithmetic averages (UPGMA).

Results

Effect of the incubation temperature on for 24 and 48 h on the SDS-PAGE protein pattern. The Aeromonas species that were subjected to heat stress in the present investigation were A. hydrophila, A. caviae and A. veronii biovar sobria. The SdS-PAGE electrophoresis HSP pat-terns for cells subjected to temperature downshifts from 42 to 25°C and upshifts to 50°C were examined after 24 and 48 h. The protein profile for isolates of Aeromonas species was established out by running eight per cent SdS- PAGE with an objective to find variation in the protein banding pattern of all the isolates. It was found that there was variation in the protein banding pattern. Most pro-teins were similarly expressed at the three temperatures.

Aeromonas intraspecies generation of heat-shock proteins (HSP). The SdS-PAGE protein profiles of the

A. hydrophila, A. caviae and A. veronii biovar sobria grown in LB- broth revealed that, Aeromonas pheno-species produced protein patterns containing several discrete bands in the most important area, with molec-ular masses in the range of 103.9–83.5 kda (Table II). differences between strains were evident in this range of molecular weights in the number of bands that were expressed under different temperatures (25°C, 42°C and 50°C) when incubated for 24 and 48 h. The SdS-PAGE protein pattern revealed that the 3 Aeromonas species bands ranged from 12–18 bands (Table II). The A. vero-nii biovar sobria serovar which was isolated had a strong band of molecular mass in the range of 103.1–101.0 kda (Table II) which lacked from all the other serovars. The band with molecular masses of >9 kda and <100 kda, in particular, was lacking in all investigated Aeromonas serovars (Table II).

The molecular weight of the band of a strain was consistent in repeated SdS-PAGE.

Aeromonas cluster analysis of heat-shock proteins (HSP). Pattern storage and comparison were performed with GelCompar version 4.2 (Applied Maths). Of the 400 digitized points in each densitometric trace, only points 11±317 were used for calculation of the simi-larities between individual pairs of patterns. The den-drogram is derived from the (unweighted pair group arithmetic average-linkage algorithm) clustering of cor-relation coefficients of the SdS-PAGE protein patterns.

The incubation of A. hydrophila (CR +ve) at 25°C revealed that the HSPs that were produced after 24 h and 48 h of their incubation when analysed dendro-gramically varied from each other by a percentage of 26.1%. But, when A. hydrophila (CR +ve) was incu-bated at 42°C, the HSPs that were produced after 24 h and 48 h were found to exhibit a 42.8% variation in their dendrogram analyses. dramatically, the dendrogramic variation between the HSPs produced by A. hydrophila (CR +ve) when incubated at 50°C for 24 h and 48 h dropped to an extremely low level of variation (7%) (Fig. 1).

The dendrogramic analysis of A. caviae when sub-jected to incubation at 25°C for 24 h induced a vari-ation in its protein electrophoresis by about 25.9% when compared to the HSPs that were produced during incubation at 25°C for 48 h. The comparative changes in the electrophoretic profile between the two incuba-tion periods increased to 35.5% when the incubation temperature was increased to 42°C. Heat stress became more evident on the A. caviae strain when the incuba-tion was elevated to 50°C and the dendrogramic anal-ysis recorded a difference of 39% between the HSPs produced after 24 h and 48 h incubation (Fig. 2).

The incubation of A. veronii biovar sobria at 25°C for 24 h disclosed HSPs changes by 38.1% from the HSPs that were produced by the strains that were incubated


for 48 h. These changes were also close to the changes that were exhibited by the strains that were incubated at 50°C for 24 h and 48 h (35.8%). The HSPs differ-

ences dramatically dropped to 29% when the strains of A. veronii biovar sobria were incubated at 42°C for 24 h and 48 h (Fig. 3).

Table IIHSPs molecular mass analysis of A. hydrophila, A. caviae and A. veronii biotype sobria incubated

at 25°C, 42°C and 50°C for 24 h and 48 h

Time of incubation (hours) 24 h 48 h 24 h 48 h 24 h 48 hNumber of bands 15 13 12 14 12 14Range of MW (kda) 94.0–13.0 95.0–14.5 96.0–13.5 95.5–13.2 95.3-13.2 94.5–13.0

Aeromonas species A. hydrophila (CR+ve)Incubation temperature 25°C 42°C 50°C

Time of incubation (hours) 24 h 48 h 24 h 48 h 24 h 48 hNumber of bands 16 15 15 16 14 15 Range of MW (kda) 96.5–12.5 98.7–14.5 98.4–12.1 99.8–12.4 98.4–12.0 97.8–12.4

Aeromonas species A. hydrophila (CR+ve)Incubation temperature 25°C 42°C 50°C

Time of incubation (hours) 24 h 48 h 24 h 48 h 24 h 48 hNumber of bands 15 14 15 13 13 14Range of MW (kda) 88.2–14.4 90.7–13.6 95.5–14.1 90.4–13.8 86.2–13.5 83.5–14.5

Aeromonas species A. caviaeIncubation temperature 25°C 42°C 50°C

Time of incubation (hours) 24 h 48 h 24 h 48 h 24 h 48 hNumber of bands 17 14 15 18 18 17Range of MW (kda) 103.1–13.4 103.5–13.5 103.9–13.8 99.0–13.3 103.1–12.4 101.0–12.2

Aeromonas species A. veronii biotype sobriaIncubation temperature 25°C 42°C 50°C

Fig. 2. dendrogram derived from SdS-PAGE HSP pattern analysis of Aeromonas caviae when subjectedto different temperature conditions and time.

Fig. 1. dendrogram derived from SdS-PAGE HSP pattern analysis of Aeromonas hydrophila (CR +ve) when subjectedto different temperature conditions and time.

Heat shock proteins of environmental Aeromonas strains2 153

Discussion

Foods are exposed to a wide range of processing treatments intended to kill or control the growth of microorganisms. Thermal treatment of foods is exten-sively used for the purpose of killing or reducing popu-lations of pathogenic and spoilage microorganisms. The exposure of microorganisms to sublethal heat treatment can, however, occur in food processing plants (Marriott, 1994). Physical stresses can be induced by exposure of microorganisms to elevated or reduced temperatures. The ability of foodborne spoilage and infectious micro-organisms in food processing environments to survive under adverse conditions and become cross protected against subsequent stresses depends on intrinsic fac-tors as well as mechanisms induced in the cell upon exposure to sublethal stresses (Abee and Wauters, 1999; Bower and daeschel, 1999).

Although the identification of Aeromonas strains to species level concerns the current number of recog-nized taxa (n = 24) (demarta et al., 2008; Beaz-Hidalgo et al., 2009; Alperi et al., 2010), yet certain strains of Aeromonas species, in particular A. hydrophila and A. sobria, are of potential public health significance (Korbel and Kosters, 1989; demarta et al., 2008; Beaz-Hidalgo et al., 2009; Alperi et al., 2010). While approxi-mately half of the clinical isolates can grow well at 45°C (Palumbo et al., 1985; Knochel, 1990), most clinical strains grow well at 42°C (Palumbo et al., 1985) and only a few isolates from vegetables stored at 5°C are capable of growing at this temperature (Majeed et al., 1990). Some isolates from cold water may not even grow or grow very slowly at 37°C (Knochel, 1990). depending on the isolate the maximum growth tem-perature seems to be from 37°C to 43–44°C (Merino et al., 1995). An observation of growth of several iso-lates at 55°C (Abou-Shanab, 2007) has not been sub-stantiated by other researchers.

A short exposure of cells to elevated temperatures reduces the synthesis of normal cellular proteins and at the same time induces a transient overproduction of a specific group of proteins, the so-called heat shock

proteins (HSPs) (Freeman et al., 1989; Abou-Shanab, 2007). The optimum temperature for the production of HSPs varies from organism to organism. The heat shock temperature range for E. coli is 43– 47°C, for the yeast 36°C and for the sickle fungus Fusarium oxy-

sporum 40°C or 43°C. In general, a rise of 5°C above the normal physiological temperature will induce the synthesis of HSPs (Freeman et al., 1989; Abou-Shanab, 2007). They are classified according to their respective molecular weights and are divided into six families: HSP100 family, the HSP90 family, the HSP70 family, the HSP60 family, the HSP40 family, and the small HSPs (sHSPs). The small heat shock family members vary in their respective molecular weights; they range in size from 15 kda to 30 kda. The HSP70 family rep-resents one of the most widely examined heat shock families. Remarkably, the general features of the HSP70 and HSP60 molecule functional roles are similar. Both types of chaperones are abundant proteins whose rate of synthesis can be enhanced by stress conditions such as heat shock.

The study showed that there are differences in the SdS-PAGE protein profiles between the three Aero-monas species grown in LB. The molecular weight of the bands differed between strains. We have shown that at an increased temperature Aeromonas species produce heat shock proteins in vitro that are typical of the heat shock responses described for other bac-teria (Love and Hirsh, 1994). In addition, the heat shock proteins produced in A. hydrophila, A. caviae or A. veronii biovar sobria are of the appropriate size for the previously described families of heat shock pro-teins (HSP90, HSP70, HSP60, and small heat shock proteins) (Love and Hirsh, 1994). Statner et al. (1988) observed temperature-dependent changes in the pro-tein profiles of aeromonads. Kuijper et al. (1989) found that A. hydrophila strains had unique patterns in the region of 3 and 45 kda, and they lacked a protein band of 22–26 kda, which does not agree with the present study. This could be hypothetically attributed to the fact that Kuijper et al. (1989) did not expose their cultures to different temperatures and time of incubation.

Fig. 3. dendrogram derived from SdS-PAGE HSP pattern analysis of Aeromonas veronii biovar sobria when subjectedto different temperature conditions and time.


Concluding remarks. Foods and food processing environments can impose extreme temperature stress on spoilage and pathogenic bacteria. The behavior of bacterial cells exposed to such a condition is dictated by the duration of exposure, degree, constitutive and induced ability to respond, and ability to synthesize shock proteins. These factors also affect the extent to which cells become cross protected to a secondary stress encountered after exposure to extreme tempera-ture. Investigations of the behavior of bacteria at tem-perature extremes are made difficult by the fact that cells may act differently upon exposure to simultaneous or sequential stresses

Literature

Abbott S.L., W.K.W. Cheung and J.M. Janda. 2003. The genus Aeromonas: biochemical characteristics, atypical reactions, and phenotypic identification schemes. J. Clin. Microbiol. 41: 2348–2357.Abee T. and J.A. Wauters. 1999. Microbial stress response in mini-mal processing. Int. J. Food Microbiol. 50: 69–91.Abou-Shanab R.A.I. 2007. Characterization and 16S rdNA identifi-cation of thermo-tolerant bacteria isolated from hot springs. J. Appl. Sci. Res. 3: 994–1000.Albert M.J., M. Ansaruzzaman, K.A. Talukder, A.K. Chopra, I. Kuhn, R.M.A.S.G. Faruque, I.M. Sirajul, R.O. Bradleysack and R. Mollby. 2000. Prevalence of enterotoxin genes in Aeromonas spp. isolated from children with diarrhea, healthy controls and the envi-ronment. J. Clin. Microbiol. 38: 3785–3700.Alperi A., A.J. Martínez-Murcia, A. Monera, M.J. Saavedra and M.J. Figueras. 2010. Aeromonas fluvialis sp. nov., isolated from a Spanish river. Int. J. Syst. Evol. Microbiol. 60: 72–77Ausubel F.M., R. Brent and R.E. Kingston. 1994. Current proto-cols in molecular biology. 2nd Edition, John Wiley and Sons Inc, New York.Beaz-Hidalgo R., A. Alperi, M.J. Figueras and J.L. Romalde. 2009. Aeromonas piscicola sp. nov., isolated from diseased fish. Syst. Appl. Microbiol. 32: 471–479.Beutin L., D. Ullmann, D. Knabner, S. Zimmermann and H. Weber. 2004. Isolation and characterization of toxigenic Aeromonas from seafood in Germany 5th World Congress Foodborne Infections and Intoxications 7–11 June Berlin, Germany Federal Institute for Risk Assessment.Bower C.K. and M.A. Daeschel. 1999. Resistance responses of microorganisms in food environments. Int. J. Food Microbiol. 50: 33–44.Castro-Escarpulli G., M.J. Figueras, G. Aguilera-Arreola, L. Soler, E. Ferna’ndez-Rendo’n, G.O. Aparicio, J. Guarro and M.R. Chaco’n. 2003. Characterisation of Aeromonas spp. isolated from frozen fish intended for human consumption in Mexico. Int. J. Food Microbiol. 84: 41– 49Coburn B., I. Sekirov and B.B. Finlay. 2007. Type III secretion systems and disease. Clin. Microbiol. Rev. 20: 535–549.Demarta A., M. Küpfer, P. Riegel, C. Harf-Monteil, M. Tonolla, R. Peduzzi, A. Monera, M.J. Saavedra and A. Martínez-Murcia. 2008. Aeromonas tecta sp. nov., isolated from clinical and environ-mental sources. Syst. Appl. Microbiol. 31: 278–286.Eurell T.E., D.H. Lewis and L.C. Grumbles. 1978. Comparison of selected diagnostic tests for detection of motile Aeromonas septicae-mia. Am. J. Vet. Res. 39: 1384–1386.

Figueras M.J. 2005. Clinical relevance of Aeromonas. Rev. Med. Microbiol. 16: 145–153.Freeman S., C. Ginzburg and J. Katan. 1989. Heat shock protein synthesis in propagules of Fusarium oxysporum f. sp. Niveum. Phytopathology 79: 1054–1058.Janda J.M. and S.L. Abbott. 2010. The genus Aeromonas: taxonomy, pathogenicity, and infection. Clin. Microbiol. Rev. 23: 35–73.Kaufmann S.H. 1992. Heat-shock proteins in health and disease. Int. J. Clin. Lab. Res. 21: 221–226.Knochel S. 1990. Growth characteristics of motile Aeromonas spp. isolated from different environments. Int. J. Food Microbiol. 10: 235–244.Korbel R. and J. Kösters. 1989. Epidemic deaths of wild birds after Aeromonas hydrophila infection. Tierarztl. Prax. 17: 297–298.Kuijper E.J., L. van Alphen, E. Leenders and H.C. Zanen. 1989. Typing of Aeromonas strains by dNA restriction endonuclease analysis and polyacrylamide gel electrophoresis of cell envelopes. J. Clin. Microbiol. 27: 1280–1285.Kusukawa N. and T. Yura. 1988. Heat-shock protein GroE of Esche- richia coli: key protective roles against thermal stress. Genes Dev. 2: 874–882.La Rossa R.A. and T.K. Van Dyk. 1991. Physiological roles of the dnaK and GroE stress proteins: catalysts of protein folding or macro-

molecular sponges? Mol. Microbiol. 5: 529–534.Lathigra R.B., P.D. Butcher, T.R. Garbe and D.B. Young. 1991. Heat-shock proteins as virulence factors of pathogens. Curr. Top. Microbiol. Immunol. 167: 125–143.Love B.C and D. Hirsh. 1994. Pasteurella multocida produces heat shock proteins in turkeys. Infect. Immun. 62: 1128–1130.Macario A.J. 1995. Heat-shock proteins and molecular chaperones: implications for pathogenesis, diagnostics, and therapeutics. Int. J. Clin. Lab. Res. 25: 59–70.Majeed K.N., A.F. Egan and I.C. MacRae. 1990. Production of exotoxins by Aeromonas spp. at 5 degrees C. J. Appl. Bacteriol. 69: 332–337. Marriott N.G. 1994. Principles of Food Sanitation, 4th ed. Chapman and Hall, New York.Mauchline W.S., B.W. James, R.B. Fitzgeorge, P.J. Dennis and C.W. Keevil. 1994. Growth temperature reversibly modulates the virulence of Legionella pneumophila. Infect. Immun. 62: 2995–2997.Merino S., X. Rubires, S. Kngchel and J.M. Tomas. 1995. Emerg-ing pathogens: Aeromonas spp. Int. J. Food Microbiol. 28: 157–168.Murray P.R., E.J.O. Baron, M.A. Pfaller, J.H. Jorgensen and R.H. Yolken. 2003. Manual of clinical microbiology 8th Ed. Vol. 1, ASM Press, Washington d.C.Neidhardt F. and R.A. Van Bogelen. 1987; Heat shock response. In: Escherichia coli and Salmonella Typhimurium, vol. 2 (Neidhardt F.C., Ingraham J.L., Low K.B., Magasanik B., Schaechter M., Umbarger E., eds.), pp. 1334–1345, American Society of Microbiology, Washing-ton, d.C.Palumbo S.A., D.R. Morgan and R.L. Buchanan. 1985. Influence of temperature, NaCI, and pH on the growth of Aeromonas hydrophila. J. Food Protect. 50:1417–1421.Schurr M.J. and V. Deretic. 1997. Microbial pathogenesis in cystic fibrosis: co-ordinate regulation of heat-shock response and conver-sion to mucoidy in Pseudomonas aeruginosa. Mol. Microbiol. 24: 411–420.Statner B., M.J. Jones and W.L. George. 1988. Effect of incuba-tion temperature on growth and soluble protein profiles of motile Aeromonas strains. J. Clin. Microbiol. 26: 392–393.Volker U., H. Mach, R. Schmid and M. Hecker. 1992. Stress pro-teins and cross-protecion by heat-shock and salt stress in Bacillus subtilis. J. Gen. Microbiol. 38: 2125–2135.


ORGINAL PAPER

Introduction

Coagulase-negative staphylococci (CNS) have been regarded as a threat since the 80s of the last century due to advances in medicine and implementation of new medical methods. This type of infection is difficult to diagnose as it is often not easy to distinguish an etiolo-gical disease factor from a natural flora contamination. Recently, much more attention has been paid to the need of CNS species identification and searching for a connection between species affiliation and the clinical importance of particular staphylococci (Tan et al., 2006; Hamels et al., 2001).

CNS are the leading causes of nosocomial blood-stream infections (Chandran and Rennie, 2005; Frigatto et al., 2005). According to Finkelstein et al., (2002) mor-bidity among patients with CNS bacteriemia amounts to 16%, hence investigations concerning differences in resistance profiles of particular species of this group, especially multi-resistant strains are of particular sig-nificance. The increase of bacterial resistance forces rational antibiotic therapy in order to maintain the effi-cacy of these drugs for as long as possible (Wang and Lipstich, 2006). Methods of alternative therapy against these infections are being sought. One of them can be

combination therapy (dawis et al., 2003; Guerrero and Gorgolas, 2006; Miranda-Nowales et al., 2006).

In the presented investigations, the activity of β-lac-

tam antibiotics (oxacillin, cloxacillin, cephalotin), van-comycin, gentamicin and rifampicin applied in vitro as single or in combination against nosocomial methicil-lin-resistant CNS belonging to S. epidermidis, S. haemo-lyticus, S. cohnii and S. hominis was investigated.

Experimental


37 methicillin-resistant CNS from our own col-lection were tested. They belonged to S. epidermidis (n = 12), S. haemolyticus (n = 9), S. cohnii (n = 10), S. hominis (n = 6). The strains originated from hospi-tal environment and from the skin of medical staff of the intensive care unit at a paediatric ward of a uni-versity hospital. Identification to the species level was performed with API-Staph System (BioMérieux). Only a single isolate per patient was tested. Methicillin resist-ance was detected phenotypically with a cefoxitin test (FOx30 – Becton-dickinson) and confirmed by mecA

Species-Specific Sensitivity of Coagulase-Negative Staphylococcito Single Antibiotics and Their Combinations

GRAŻYNA SZYMAńSKA1*, MAGdALENA SZEMRAJ2 and ELIGIA M. SZEWCZYK2

1department of Biosynthesis of drugs , 2department of Pharmaceutical MicrobiologyMedical University of łódź, Poland

Received 25 September 2010, revised 2 April 2011, accepted 4 April 2011

A b s t r a c t

The activity of β-lactam antibiotics (oxacillin, cloxacillin, cephalotin), vancomycin, gentamicin and rifampicin applied in vitro individu-ally and in combination against 37 nosocomial methicillin-resistant strains of coagulase-negative staphylococci (CNS) was assessed to demonstrate the heterogeneity of this group of bacteria and estimate the chance of the efficacy of such therapy. The strains belonged to four species: Staphylococcus epidermidis, Staphylococcus haemolyticus, Staphylococcus cohnii, Staphylococcus hominis. They originated from a hospital environment and from the skin of medical staff of the intensive care unit of a paediatric ward at a university hospital. All strains were methicillin-resistant, according to CLSI standards, but individual strains differed in MICOx values. Susceptibility to other tested antibiotics was also characteristic for the species. The increased susceptibility to antibiotics in combinations, tested by calculating the fractional inhibitory concentration (FIC) index, concerned 26 out of 37 investigated strains and it was a feature of a particular species. Combinations of vancomycin and cephalotin against S. epidermidis and oxacillin with vancomycin were significant, as well as cephalotin and rifampicin in growth inhibition of multiresistant S. haemolyticus strains.

K e y w o r d s: Staphylococcus sp., antibiotic synergism, methicillin-resistant CNS, staphylococci infections treatment

* Corresponding author: G. Szymańska, phone: +48 42 677 93 01; fax: +48 42 677 93 00; e-mail: [email protected]


gene detection in PCR method (PK14 kit – dNA-GdańskII) (Van Griethuysen et al., 1999). β-lactamase was detected in a cefinase test (Becton-dickinson). Strains were stored frozen in glycerol-broth at the tem-perature of –70°C.

Strain susceptibility tests were performed according to the standards of the CLSI (2008) by the MIC micro-dilution tests using oxacillin (Ox), cloxacillin (Cx), cephalotin (CF), gentamicin (GM), vancomycin (VA) and rifampicin (RA) (Fluka and Sigma) in Mueller-Hinton broth (BioMérieux) without 2% NaCl. MIC is the lowest concentration of antibiotic that yielded no visible growth after incubation at 35°C for 24 h.

Activity of antibiotics in combination was checked, at first using disc sensitivity tests, where discs (all from Becton-dickinson) were placed on Mueller-Hinton agar plates at a distance of 19–20 mm in pairs: Ox1+VA30, CF30+VA30, CF30+RA5, CF30+GM10 and CF30+Cx1. Subsequently, strains forming characteristic elongated shape of inhibit growth zones around discs with antibi-otics were tested in a microdilution synergy test. Com-binations of antibiotics at appropriate concentrations in Mueller-Hinton broth were tested in 96-well microtiter plates (Kartell). Plates were incubated at 35°C for 24 h and Fractional Inhibitory Concentration (FIC) index was calculated. FIC is MIC of drug in combination divided by the MIC of drug acting alone (MIC of drug A in combination with drug B/MIC of drug A alone). The sum of the FICs of both antibiotics gave the FIC index. Synergism was identified when the FIC index was ≤0.5 (Climo et al., 1999; dawis et al., 2003). All tests were performed at least twice. Reference strains were S. aureus ATCC 29213 and ATCC 25923. Quantitative variables were calculated using the Kruskal-Wallis test.

Results

All investigated strains were methicillin-resistant accor - ding to CLSI standards but individual strains presented different levels of oxacillin susceptibility, expressed in MICOx values (Table I). The greatest number of strains with low MIC (≤4 mg/l) was found among S. hominis,

with an average MIC (8–128 mg/l) among S. epider-minis and with high MIC (≥256 mg/l) among S. cohnii. There were significant differences in MIC values (from 0.5 to 2048 mg/l) within species, particularly in S. haemolyticus. Nevertheless, the differences in antibiotic-resist-ance between species were clearly visible and statisti-cally important in the Kruskal-Wallis test.

Ranges of MIC values for the tested antibiotics: cloxacillin, cephalotin, gentamicin, vancomycin and rifampicin were also different for individual species (Fig. 1). Resistant strains with high MIC values were the most numerous among S. haemolyticus. In spite of great standard deviations, it can be stated that statis-tically significant differences among species in their sensitivity to particular antibiotics not only concerned differences in absolute MIC values, but also sensitivity or resistance in the clinical sense, thus placing above or below the breakpoint values for individual antibiotics.

Using UPGMA (unweighted pair group mathemati-cal averages), a dendrogram was created. The dendro-gram grouped all investigated strains in relation to their antibiotic-susceptibility and the level of their resistance was expressed in MIC values. Strains formed several clusters and particular species were located in separated areas on the axis. S. haemolyticus strains were located mostly on the left side, S. epidermidis on the right, whereas S. cohnii and S. hominis in the central part of the dendrogram (Fig. 2).

In further experiments, the activity of antibiotics in combination was tested. disc-diffusion method allowed to observe the occurrence of synergism between cephalotin and vancomycin (CF+VA), cephalotin and rifampicin (CF+RA), oxacillin and vancomycin (Ox+VA), cephalotin and cloxacillin (CF+Cx). In none of the investigated combinations either antagonism or addi-tion occurred. For the strains in which, on the basis of growth inhibition zone shape around discs, the beneficial activity of antibiotics used in combination was observed, FIC index was calculated confirming or excluding synergism. Synergistic action was confirmed in most cases. It concerned 26 out of 37 investigated strains (Fig. 3). Most susceptible strains were among S. epidermidis – 10 out of 12 of the tested ones. The

S. epidermidis (12) 1–256 5 (42%) 6 (50%) 1 (8%)S. haemolyticus (9) 0.5–2048 4 (45%) 2 (22%) 3 (33%)S. cohnii (10) 4–512 1 (10%) 3 (30%) 6 (60%)S. hominis (6) 0.5–64 5 (83%) 1 (17%) 0

Table IMIC values of investigated species to oxacillin

Species (numberof strains examined)

MICox [mg/l]range in species

Number of strains of low, medium and high MICox values

≤4 mg/l 8–128 mg/l ≥ 256 mg/l

Sensitivity of CNS to antibiotics2 157

majority of them reacted this way to two or three pairs of antibiotics. Mostly, it was the combination of CF+VA. The same effect was achieved for most of the strains by CF+RA binding. Only for S. epidermidis CF+Cx combination synergism occurred. However, for none of the S. epidermidis strains, synergism of β-lactams and gentamicin was determined. Synergistic activity of antibiotics against S. haemolyticus was observed for six out of nine tested strains. Mostly, combination of Ox+VA (5 strains) presented synergism, less frequently

combination of CF+RA (3 strains). Synergistic activity in eight out of ten of the tested S. cohnii strains pre-sented CF+RA and in half of the tested strains a com-bination of CF+GM occurred. In S. hominis synergism was noticed only against two strains and only for one pair of antibiotics (CF+ GM). Sensitivity to particular pairs of antibiotics was a feature connected with spe-cies. The combination presenting a synergistic effect in relation to the greatest number of tested strains was CF+RA. This combination was effective when MIC of

Fig. 1. differences in MIC values of the tested antibiotics for strains of the investigated species.In the case of vancomycin and rifampicin, breakpoints were higher than the maximum points on a chart scale.


cephalotin as a single antibiotic was ≤8 µg/ml (sensi-tive strain) but also when it was ≥32, µg/ml (resistant strain). Production of β-lactamases was a feature char-acteristic for almost all strains of the three tested spe-cies. The exception was S. cohnii strains, the majority

of which did not produce β-lactamases. There was no connection between synergism occurrence and produc-tion of those enzymes.

Strains presenting a synergistic reaction to the high-est number of antibiotics combinations are shown in

Fig. 2. dendrogram presenting similarity in susceptibility to 6 antibiotics of 37 strains belonging to four tested species.

Fig. 3. Synergistic effect of antibiotics in combination against strains of the investigated CNS.


Table II, in which MIC values for single antibiotics, strain sensitivity in clinical sense and antibiotic com-binations giving synergistic effect and MIC values achieved for antibiotics in pairs are presented.

The use of combinations of antibiotics gave an effective growth inhibition of the tested bacteria with simulta-neously lowered antibiotic concentration. For S. epidermidis strains vancomycin concentration could be lowered 8–64 fold and for S. haemolyticus and S. cohnii 4-fold. Cephalotin could be applied mostly in the dose of 4–16 times lower than when it was applied individually.

Discussion

CNS include several dozens of species and are a he- terogenic group of microorganisms, but only some of them are frequently isolated from humans. There is no doubt nowadays that these bacteria, especially methicil-lin-resistant strains, are important nosocomial patho-gens. In this research strains belonging to four of these species were investigated. Three of those are species considered the most important in hospital infections

(S. epidermidis, S. haemolyticus, S. hominis), relatively often isolated from clinical samples (Kloos and Banner-man, 1994; Weinstein et al., 1998) and numerously rep-resented in the hospital environment. S. cohnii strains are less frequent in clinical materials, but often occur in the hospital environment and on the skin of medical personnel. S. hominis may be considered a resistance reservoir in the environment (Kloos, 1997; Szewczyk et al., 2004). All strains were multiresistant and it could be assumed that this resistance was a result of selective pressure. Potentially, each of them could become the cause of a hospital infection.

The need to identify CNS species, wrongly treated as the same, is often perceived. It is stressed by many authors that such approach is a significant obstacle in research and diagnostics (Tacconelli et al., 2003). Nevertheless, there are still papers, even quite recent ones, in which all isolates from this group are treated equally (Kuti et al., 2008). There were also some radical opinions questioning antibiotic resistance evaluation of microorganisms from this group (Chandran and Rennie, 2005). This work was to defy such approach by showing the results of our research.

S. epidermidis 32d 256 8 128 2 0.008 1024 Ox+VA (16+0.25) R S R S S R + CF+VA (2+0.125) CF+RA (1+0.0019)S. epidermidis 1008 8 8 32 4 0.008 512 CF+VA (2+0.25) R S R S S R _ CF+Cx (1+1) CF+RA (2+0.0005)S. epidermidis 1061 16 8 32 2 0.008 512 Ox+VA (4+0.25) R S R S S R + CF+VA (2+0.0625) CF+RA (2+0.0019)S. epidermidis 1135 64 2 32 2 0.016 1024 CF+VA (0.25+0.25) R S R S S R + CF+Cx (0.25+8 ) CF+RA (0.5+0.00312)S. haemolyticus 18 512 1 8 2 0.016 0.25 Ox+VA (128+0.5) R S R S S S + CF+VA (0.125+0.0625) CF+Cx (0.125+0.125) CF+RA (0.25+0.0019)S. haemolyticus 1148 128 32 1024 2 0.016 256 Ox+VA (4+0.5) R R R S S R + CF+VA (2+0.5) CF+RA (8+0.00375)S. cohnii 105 64 32 64 2 0.125 1 Ox+VA (16+0.5) R R R S S S – CF+GM (4+0.125) CF+RA (4+0.0312)

Table IISensitivity of strains susceptible to a number of antibiotic combinations

MIC [mg/l]Strain Antibiotics in syner-

gistic combinationsMIC [mg/l]

β-lactamaseproduction

CLSI breakpointsOx GMRACxCF VA0.25 4140.258

R – resistant, S – sensitive


Methicillin resistance estimation of CNS still rises doubts and discussion. This stems from the differences among particular species. The differences between novobiocin-sensitive and novobiocin-resistant species are clearly visible, however, the similarity of S. haemo-lyticus to the latter group is noticeable (Nowak et al., 2006; Frigatto et al., 2005). This is presumably con-nected with the content of different cassettes includ-ing mecA gene in staphylococci (Martins and Maria de Lourdes Cunha, 2007) and perhaps also with other features of these bacteria. There is an urgent need for research in this matter.

The multiresistant strains tested herein clearly show the differences in antibiotic susceptibility profiles of the tested species. It may be suspected that they also differ in genetic equipment. The results showed the notice-able resistance of strains of S. haemolyticus and S. coh-nii. John et al., 2002, who conducted a research on the spectrum of 658 clinical staphylococcal isolates belong-ing to 13 species, pointed out that the highest levels of participation of isolates resistant to oxacyllin, erythro-mycin, klindamycin and telithromycin were presented by S. haemolyticus and S. epidermidis while for S. cohnii to quinupristin/dalfopristin. The presence of multiple plasmids in the cells was shown in the latter species (Szewczyk et al., 2004).

Sensitivity to cephalosporins maintained in methi-cillin-resistant staphylococci strains (S. epidermidis, S. hominis) was observed previously. In the research of Krediet et al., (1999) this feature was characteristic for 25 (almost 70%) of the tested strains including all from S. epidermidis and S. hominis, five strains of S. haemo-lyticus and two of S. cohnii. The use of this antibiotic in CNS infection therapy should be analysed thoroughly, particularly because of earlier suggestions to apply it. All environmental strains of staphylococci tested by Szewczyk et al., (2000) were susceptible to cefuroxime and cefotaxime. 87% and 71% isolates from infants were respectively susceptible to two of these cephalospor-ins. All investigated strains of S. epidermidis from both groups were susceptible to cephalotin.

All of the strains from the collection tested in this research were sensitive to vancomycin and rifampicin, though clear differences in MIC values among species were visible (Fig. 1). Vancomycin is still frequently used in treatment of serious infections caused by gram-posi-tive microorganisms, but stepping away from vancomy-cin therapy and replacing it with, e.g. cloxacillin both in cases of infections caused by S. aureus and CNS is recommended (Lawrence et al., 2005). As can be seen in the results presented in this work, therapy with cloxacil-lin would not meet the expectations in case of all spe-cies investigated CNS except S. haemolyticus, especially in combination with cephalotin. Lowered sensitivity to glycopeptides of S. aureus and S. epidermidis strains was

described by authors who analysed CNS clinical iso-lates’ sensitivity to vancomycin and claimed appearance of tolerance to that antibiotic and low sensitivity level (Walsh et al., 2001; Bourgeois et al., 2007).

Several studies have previously reported the syner-gistic effect of antibiotics on methicillin-resistant strains of staphylococci. Such studies have been undertaken mostly towards infections caused by multiresistant clinical strains i.e. MRSA (dawis et al., 2003). Also research on treatment of infections with those bacteria in experimental models has been carried out. Fox et al. (2006), showed that using nafcillin and vancomycin in combination cleared bloodstream infections in experi-mental endocarditis in rabbit models caused by van-comycin-resistant S. aureus. Efficacy of treatment with vancomycin in combination with aminoglysides against MRSA has also been tested (Lee et al., 2003). In the study of Miranda-Nowales (2006), synergy was evident for dicloxacillin or cephalotin in combination with ami-kacin against methicillin-resistant Staphylococcus spp.

In this research it has been shown that there is a chance to obtain an efficient antibiotic therapy in combination also in infections caused by CNS. Combi-nations of vancomycin and cephalotin against S. epider-midis, staphylococcus most frequently isolated from gram-positive hospital infections, seem to be particu-larly promising. A synergistic effect on growth inhibi - tion of multiresistant S. haemolyticus strains was obtained using oxacillin and vancomycin as well as cephalotin and rifampicin. However, clinical research including species identification of CNS causing infections is necessary, because as can be seen, individual species of this group differ in antibiotic resistance.

AcknowledgementsThis study was supported by a grant 503-3012-3 from Medical

University of łódź.

Literature

Bourgeois I., M. Pestel-Caron, J. Lemeland, J.L. Pons and F. Caron. 2007. Tolerance to the glycopeptides vancomycin and teicoplanin in coagulase-negative staphylococci. Antimicrob. Agents Chemiother. 51: 740–743.Chandran A.U. and R. Rennie. 2005. Routine antimicrobial sus-ceptibility testing of coagulase-negative staphylococci isolated from blood cultures: is it necessary? Clin. Microbiol. Infect. 11: 1037–1040.Climo M.W, R.L. Patron and G.L. Archer. 1999. Combinations of vancomycin and β-lactams are synergistic against staphylococci with reduced susceptibilities to vancomycin. Antimicrob. Agents Chemiother. 43: 1747–1753.Clinical and Laboratory Standards Institute. 2008. Performance standards for antimicrobial Susceptibility testing. Eighteenth Informa-tional Supplement. M100-S18. CLSI, Wayne, PA, USA. Dawis M.A., H.D.Isenberg, K.A. France and S.G. Jenkins. 2003. In vitro activity of gatifloxacin alone and in combination with cefepime, meropenem, piperacillin and gentamicin against multi-drug-resistant organism. J. Antimicrob. Chemother. 51: 1203–1211.


Finkelstein R., R. Fusman, I. Oren, I. Kassis and N. Hashman. 2002. Clinical and epidemiologic significance of coagulase-negative staphylococci bacteremia in a tertiary care university Israeli hospital. Am. J. Infect. Control. 30: 21–5Fox P.M., R.J. Lampen, K.S. Stumpf, G.L. Archer and M.W. Climo. 2006. Succesful therapy of experimental endocarditis caused by vancomycin-resistant Staphylococcus aureus with a combination of vancomycin and β-lactam antibiotics. Antimicrob. Agents Chemio-ther. 50: 2951–2956.Frigatto E.A., A.M. Machado, A.C. Pignatari and A.C. Gales. 2005. Is the cefoxitin disk test reliable enough to detect oxacillin resistance in coagulase-negative staphylococci? J. Clin. Microbiol. 43: 2028–2029.Guerrero M.L.F. and M. Górgolas. 2006. Comparative activity of cloxacillin and vancomycin against methicillin-susceptible Staphy-lococcus aureus experimental endocarditis. J. Antimicrob. Chemother. 58: 1066–1069.Hamels S., J.L. Gala, S. Dufour, P. Vannuffel, N. Zammatteo and J. Remacle. 2001. Consensus PCR and microarray for diagnosis of the genus Staphylococcus species, and methicillin resistance. Bio-techniq. 31: 1364–6John M.A., C. Pletch and Z. Hussain. 2002. In vitro activity of quinupristin/dalfopristin, linszolid, telithromycin and comparator antimicrobial agents against 13 species of coagulase-negative staphy-lococci. J. Antimicrob. Chemother. 50: 933–938.Kloos W.E. 1997. Taxonomy and systematic of staphylococci indig-enous to humans. pp. 123–125. In: Crossley K.B. and G.L. Archer (eds). The Staphylococci in Human disease. Churchill Livingstone, London.Kloos W.E. and T.L. Bannerman. 1994. Update on clinical signifi-cance of coagulase-negative staphylococci. Clin. Microbiol. Rev. 7: 117–140.Krediet T.G., M.E. Jones, L.J. Gerards and A. Fleer. 1999. Clini-cal outcome of cephalotin versus vancomycin therapy in the treat-ment of coagulase-negative staphylococcal septisemia in neonates: Relation to methicillin resistance and mec A gene carriage of blood isolates. Pediatrics 103: 3 Kuti J.L., C.R.V. Kiffer, C.M.F. Mendes and D.P. Nicolau. 2008. Pharmacodynamics comparsion of linezolid, teicoplanin and van-comycin against clinical isolates of Staphylococcus aureus and coa-gulase-negative staphylococci collected from hospitals in Brasil. Clin. Microbiol. Infect. 14: 116–123.Lawrence S.L., V. Roth, R. Slinger, B. Toye, I. Gaboury and B. Lemyre. 2005. Cloxacillin versus vancomycin for presumed late-onset sepsis in the Neonatal Intensive care Unit and the impact upon outcome of coagulase negative staphylococcal bacteremia: a retro-spective cohort study. BMC Pediatrics 5: 49.

Lee D., H. Chun, D. Yim, S.M. Choi, J.H. Choi, J.H. Yoo, W.S. Shin and M.W. Kang. 2003. Efficacies of vancomycin, arbekacin, and gentamicin alone or in combination against methicillin-resistant Staphylococcus aureus in an in vitro infective endocarditis model. Antimicrob. Agents Chemiother. 47: 3768–3773.Martins A. and R.S. Cunha Maria de Lourdes. 2007. Methicillin resistance in Staphylococcus aureus and coagulase-negative staphy-lococci: Epidemiological and molecular aspects. Microbiol. Immunol. 51: 787–795.Miranda-Nowales G., B.E. Leanos-Miranda, M. Vilchis-Perez and F. Solórzano-Santos. 2006. In vitro activity effects of combinations of cephalotin, dicloxacillin, imipenem, vancomycin and amikacin against methicillin-resistant Staphylococcus spp. strains. Ann. Clin. Microb. Antimicrob. 5: 25.Nowak T, E. Balcerczak, M. Mirowski and E.M. Szewczyk. 2006. detection of methicillin resistance in hospital environmental strains of coagulase-negative staphylococci. Pol. J. Microbiol. 55: 339–343.Szewczyk E.M., A. Piotrowski and M. Różalska. 2000. Predomi-nant staphylococci in the intensive care unit of a pediatric hospital. J. Hosp. Infec. 45: 145–154.Szewczyk E.M., M. Różalska, T. Cieślikowski and T. Nowak. 2004. Plasmids of Staphylococcus cohnii isolated from the intensive-care unit. Folia Microbiol. 49: 123–131.Tacconelli E., E.M.C. D’Agata and A.W. Karchmer. 2003. Epide-miological comparison of true methicillin-resistant and methicillin-susceptible coagulase-negative staphylococcal bacteremia at hospital admission. Clin. Infect. Dis. 37: 644–649.Tan T.Y., S.Y. Ng, W.X. Ng. 2006. Clinical significance of coagu-lase-negative Staphylococci recovered from nonsterile sites. J. Clin. Microbiol. 44: 3413–3414.Van Griethuysen A., M. Pouw, van Leewen M. Heck, P. Willemse, A. Buiting, and J. Kluytmans. 1999. Rapid slide latex agglutination test for detection of methicillin resistance in Staphylococcus aureus. J. Clin. Microbiol. 37: 2789–2792.Walsh T.R., A. Bolmström, A. Qwärnström, P. Ho, M. Wootton, R.A. Howe, A.P. MacGowan, and D. Diekema. 2001. Evaluation of current methods for detection of Staphylococci with reduced susceptibility to glycopeptides. J. Clin. Microbiol. 39: 2439–2444.Wang Y.C. and M. Lipstich. 2006. Upgrading antibiotic use within a class: Tradeoff between resistance and treatment success. PNAS. 103: 9655–9660Weinstein M.P., S. Mirrett, L. van Pelt, M. Mc-Kinnon, B.L. Zim-mer, W. Kloos and L.B. Reller. 1998. Clinical importance of identi-fying coagulase-negative staphylococci isolated from blood cultures: Evaluation of microscan rapid and dried overnight gram-positive panels versus a conventional reference method. J. Clin. Microbiol. 36: 2089–2092


ORGINAL PAPER

Introduction

Acinetobacter is a gram-negative bacterium, which is a very effective human colonizer found in many health care environments (Bergogne-Berezin and Towner, 1996; Perez et al., 2007). The combination of its envi-ronmental resilience and its wide range of resistance determinants renders it a successful nosocomial patho-gen (Nordmann, 2004). Nowadays, A. baumannii is emerging as a cause of numerous global outbreaks, displaying ever-increasing rates of resistance (Villegas and Hartstein, 2003).

Although six classes of integrons exist (Neild et al., 2001), three main classes have been described (Rowe-Magnus and Mazel, 1999). All integrons have a 5’ con-served segment, including an intI gene encoding an integrase and an attI recombination site, but have dis-tinct 3’ conserved segments. As for the class 1 integrons, the 3’ conserved segment includes three open reading

frames (ORFs): qacEΔ1, a deletion derivative of the anti-septic resistance gene qacE; the sul1 sulfonamide resistance gene; and ORF5, of unknown function (Radström et al., 1994). The second class of integrons was found in transposon Tn7 and its derivatives, and its 3’ conserved segment contains five tns genes involved in the move-ment of the transposon. A single class 3 integron has been reported to date, but its 3’ conserved segment has not been characterized (Arakawa et al., 1995). Class 1 integron is more frequent in Acinetobacter species (Koeleman et al., 2000; Galleco and Towner, 2001; Gaur et al., 2006; Ploy et al., 2000; xu et al., 2008).

The objectives of this study were to determine the prevalence of different classes of integrons in isolated Acinetobacter and to find out the association between the presence of integrons and antibiotics resistance. Furthermore, the relationship between the productions of specific bands in PCR assay with the extension of multi-drug resistance was studied.

Association between Existence of Integrons and Multi-Drug Resistancein Acinetobacter Isolated from Patients in Southern Iran

SARA JAPONI1, AZIZ JAPONI2*, SHOREH FARSHAd2, AHYA ABdI ALI1

and MARZIEH JAMALIdOUST2

1 department of Biology, Alzahra University, Tehran, Iran2 Alborzi Clinical Microbiology Research Center, Nemazee hospital, Shiraz, Iran

Received 2 September 2010, revised 5 January 2011, accepted 20 January 2011

A b s t r a c t

Nosocomial infections caused by multi-drug resistant Acinetobacter pose a serious problem in many countries. This study aimed at determining the antibiotic susceptibility patterns and prevalence of different classes of integrons in isolated Acinetobacter. In addition, the association between production of specific bands in PCR assay and magnitude of multi-drug resistance was investigated. In total, 88 Acinetobacter strains were isolated from patients from October 2008 through September 2009. The Minimal inhibitory concentration (MIC) of 12 antibiotics conventionally used in clinics against the isolates, was determined by E-test method. The existence of integron classes was investigated by PCR assay through the amplification of integrase genes. The most effective antibiotic against Acinetobacter was colistin with 97.7% activity, followed by imipenem (77.3%) and meropenem (72.7%). The presence of integron classes 1 and 2 in 47 (53.4%) isolates was confirme, However, no class 3 was detected. The proportion of class 1, compared with class 2, was high (47.7% vs. 3.4%). The association between multi-drug resistance to norfloxacin, ceftazidime, gentamicin, ciprofloxacin, cefepime and amikacin and the presence of integrons was statistically significant. However, the association was not remarkable in many of the isolates which exhibited resistance to the rest of antibiotics. This may imply that in addition to integrons, other resistance determinants such as transposon and plasmid may also contribute to resistance. To reduce the pressure on sensitive isolates, comprehensive control measures should be implemented. Furthermore, wise application of effective antibiotics could help alleviate the situation. Colistin is the most effective antibiotic in vitro against Acinetobacter.

K e y w o r d s: Acinetobacter, integrons, multi-drug resistance, PCR assay

* Corresponding author: A. Japoni, Alborzi Clinical Microbiology Research Center, Nemazee hospital, Shiraz University of Medical Sciences, post code: 71037-11351, Shiraz, Iran; phone: +98-711-6474264; fax: +98-711- 6474303; e-mail: [email protected]

Japoni S. et al. 2164

Experimental


Isolation of Acinetobacter. Eighty eight Acinetobac-ter strains were isolated from patients hospitalized in Nemazee hospital, affiliated with Shiraz University of Medical Sciences, Iran during the period from October 2008 to September 2009. Identification of the isolates was carried out using the API 20E system (bioMérieux, Marcy l’Etoile, France).

MIC determination. Minimal inhibitory concentrations (MICs) of 12 antibiotics including, ciprofloxacin, colistin, ceftazidime, ampicillin/sulbactam, imipenem, meropenem, gentamicin, norfloxacin, amikacin, cefe- peme, tobramycin and cefoperazon/sulbactom against the isolates were determined by the E-test method and the results were interpreted as recommended by the manufacturer’s instructions.

DNA extraction and PCR assay. Bacteria dNA were harvested by conventional phenol-chloroform extrac-tion method. The quantity of dNA was determined by Nanodrop (Nanodrop Technologies, Wilmington, delaware USA) and adjusted 50 ng µl–1. determina-tion of integron classes was performed by multiplex PCR using the primers described in Table I. The prim-ers were obtained from TIB MOLBIOL Syntheselabor GmbH (Berlin, Germany). PCR assay was performed in 20 µl volume, containing 0.4 mM deoxynucleoside triphosphate (dNTP), 2 ml of 10x PCR buffer, 1 U of Taq polymerase (Fermentas, Lithuania), 0.6 mM MgCl2, 0.25 mM of each primer and 250 ng dNA in 5 µl vol-ume were added to the reaction mixtures. PCR was performed under pre-denaturation at 94°C for 5 min-utes, followed by 35 cycles of 94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds ending with a final extension step at 72°C for 5 min and held at 8°C. Products were electrophoresed in 2% agarose, stained by ethidium bromide and video images were obtained by gel documentation (Uvtec, Sigma, Germany) system.

Primers Int1F and Int1R were used to amplify a 160 bp fragment of the intI1 gene for the class 1 integrase and

the primers Int2F and Int2R amplified a fragment of 288 bp, specific for the intI2 gene. Primers Int3F and Int3R were used to amplify a specific intI3 gene.

detection of the complete gene composition of class 1 integrons was performed with primers for the 5’ and 3’ conserved segments. This PCR also permitted the determination of the size of any inserted gene cassette. The size range of the detected inserted gene cassette varied from 500 to >3000 bp.

PCR assay to detect complete gene makeup of class 1 integron was carried out in 20 µl volume with the same concentration mixture mentioned above. The only modification was increased to 3U amount of Smart Taq polymerase (Fermentas, Lithuania). PCR amplification program was as follows: 5 minutes of initial denatura-tion at 94°C , 1 min of denaturation at 94°C , 1 min of annealing at 55°C, and 30 seconds of extension at 72°C for a total of 35 cycles. Five seconds were added to the extension time at each cycle.

Statistical analysis. Correlation between antibiotic resistance patterns and presence of different classes of integron was determined by Chi-square and Fisher’s Exact test by SPSS version 15 software. The significant level was defined as P<0.05.

Results

Acinetobacter spp was mostly isolated from the blood, 35 (39.8%) and predominantly in men (70%), than in women (30%). A. baumannii was dominant in the isola - ted bacteria 79 (89.8%), followed by Acinetobacter lwoffii 8 (9.1 %). 97.7% of Acinetobacter isolates were suscep tible to colistin, 77.3% to imipenem, 72.7% to meropenem, 67% to cefoperazon/sulbactam, 63.6% to tobramycin, 61.4% to ampicillin/sulbactam, 26.1% to ciprofloxacin, 25% to amikacin, 23.8% to norfloxacin, 20.4% to gen-tamicin, 19.3% to cefepime and 18.2 % to ceftazidime.

diagnosis of integrons was carried out by multi-plex PCR assay. The presence of intI1 and intI2 was confirmed while intI3 was not detected. Figure 1 dem-onstrates the presence of integrase 1 and 2. In total,

Int1-F 5’ CAG TGG ACA TAA GCC TGT TC 3’Int1-R 5’ CCC GAG GCA TAG ACT GTA 3’

Koeleman et al., 2000Int2-F 5’ TTG CGA GTA TCC ATA ACC TG 3’Int2-R 5’ TTA CCT GCA CTG GAT TAA GC 3’Int3-F 5’ ACG GAT CTG CCA AAC CTG ACT 3’

Ploy et al., 2000Int3-R 5’ GCC TCC GGC AGC GAC TTT CAG 3’ CS-F 5’ GGC ATC CAA GCA GCA AG 3’

Lévesque et al., 1995CS-R 5’ AAG CAG ACT TGA CCT GA 3’

Table IPrimers used in this study

Primer Nucleotide sequence References

Multidrug resistant Acinetobacter and integrons 2 165

47 (53.4%) Acinetobacter strains exhibited either a class 1 integrase or class 2 integrase or both of them. Table II shows the prevalence of different classes of integron in Acinetobacter isolates.

Forty five (57%) of the 79 A. baumannii strains, and 2 (25%) of the 8 A. lwoffii strains carry an inte-gron. The frequencies of the different antibiotic resist-ance patterns and their association with integrons were assessed. Based on this evaluation, strains exhibiting resistance to the panel of antibiotics including norfloxa-cin, ceftazidime, gentamicin, ciprofloxacin, amikacin, cefepime and norfloxacin, ceftazidime, gentamicin, ciprofloxacin, ampicillin/sulbactam, amikacin, cefope-razon/sulbactom, cefepime showed high prevalence of class 1 integrons.

data also indicate an association between the panel of antibiotics including norfloxacin, ceftazidime, gen-tamicin, ciprofloxacin, amikacin, tobramycin, cefepime and class 2 integron (Table III). Amplification of inte-gron class 1 produced bands ranging between 500 to

47.7 42 Class І 3.4 3 Class ІІ 2.3 2 Class І & Class ІІ 0 0 Class ІII 46.6 41 Without integron 100 88 Total

Table IIPrevalence of different classes of integrons

in Acinetobacter isolates

Percentage Number Integron

Table IIIAssociation between Antibiotic resistance and integron classes

Abbreviations: CI – ciprofloxacin; CO – colistin; TZ – ceftazidime; AB –ampicillin/sulbactam; IP – imipenem; MP – meropenem; GM – gentamicin; Nx–norfloxacin; AK – amikacin; PM – cefepime; TM – tobramycin;CPS – cefoperazon/sulbactom.

IntegronAntibiotic resistance pattern

Class ІClass ІІClass І & IITotal

TZ 1 – – 1TZ–AK – – 1 1AB-PM – – – 1AK-TM – – – –GM-TM – – – 1AB-IP-MP – – – 1Nx-TZ-GM-TM-PM – – – 1TZ-GM-AK-TM-PM 1 – – 1Nx-TZ-GM-CI-AK-PM 11 – – 13Nx-TZ-GM-CI-AB-PM 2 – – 2Nx-TZ-GM-AK-TM-PM – – – 1Nx-TZ-CI-AK-MP-PM – – – 1Nx-TZ-GM-CI-AK-CPS-PM 1 – – 2Nx-TZ-GM-CI-AK-TM-PM 4 3 – 14Nx-TZ-GM-CI-AK-PM-CO – – – 1Nx-TZ-GM-AK-IP-TM-MP-PM 1 – – 1Nx-TZ-GM-CI-AB-AK-CPS-PM 8 – – 11Nx-TZ-GM-CI-AK-TM-MP-PM 1 – – 2Nx-TZ-GM-CI-AB-AK-TM-PM 1 – – 1TZ-GM-AB-IP-TM-MP-PM-CO – – – 1TZ-GM-CI-AK-IP-TM-MP-PM 1 – – 1Nx-TZ-GM-CI-AB-AK-TM-MP 1 – – 1Nx-TZ-GM-CI-AB-IP-CPS-MP-PM 2 – – 2Nx-TZ-GM-CI-AB-AK-IP-CPS-MP-PM 5 – 1 6Nx-TZ-GM-CI-AB-AK-IP-TM-CPS-MP-PM 1 – – 8Sensitive 1 – – 12Total 42 3 2 88


>3000 bp and the strains containing bands with 500, 600, 800 and 1200 bp were more frequent (Table IV).

The association between drug resistance to norfloxacin, ceftazidime, gentamicin, ciprofloxacin, cefepime, amikacin and the presence of integrons was statistically signifi cant, while no association was observed between colistin, imipenem, meropenem, cefoperazon/sulbactom, tobramycin, ampicillin/sulbactam and integron (Table V).

Discussion

Acinetobacter infections are complicated in hospital-ized patients due to the acquisition of multi-drug resist-ance. Most samples in the present study were isolated from the blood (39.8%). Similar data were obtained previously in the same region (Feizabadi et al., 2008). dissemination of Acinetobacter through blood may

Fig. 1. detection of integrons by amplification of integrase.Lane 8, 100 bp dNA ladder (MBI Fermentas, Hanover, Md); lanes 1, 3, 5, integrase 1 amplicons (160 bp); lane 2, both integrase 1(160 bp) and inte-

grase 2 amplicons (288 bp). Lane 4, 6 and 7 were integron negative.

Colistin 0 (0) 2.3 (2) 2.3 (2) 0.126Imipenem 13.6 (12) 9.1 (8) 22.7 (20) 0.501Meropenem 15.9 (14) 11.4 (10) 27.3 (24) 0.571Cefoperazone/sulbactam 21.6 (19) 11.4 (10) 33 (29) 0.110 Tobramycin 14.8 (13) 21.6 (19) 36.4 (32) 0.069Ampicillin/sulbactam 25 (22) 13.6 (12) 38.6 (34) 0.092Ciprofloxacin 48.9 (43) 25 (22) 73.6 (65) P< 0.05Amikacin 47.7 (42) 27.3 (24) 75 (66) 0.001Norfloxacin 48.9 (43) 27.3 (24) 76.2 (67) P< 0.05Gentamicin 51.1 (45) 28.4 (25) 79.5 (70) P< 0.05Cefepime 51.1 (45) 29.5 (26) 80.6 (71) P< 0.05Ceftazidime 52.3 (46) 29.5 (26) 81.8 (72) P< 0.05

Table VAssociation between the existence of integron and antibiotic resistance in 88 Acinetobacter isolates

Associationwith integron3

% Resistanceof total (total no)

% Resistanceint-negative2 (no)

% Resistanceint-positive1 (no) Antibiotic

1 – int-positive: integron positive in multiplex PCR assay; 2 – int-negative: integron negative in multiplex PCR assay.3 – Significant values are in bold.

500, 600, 1300 1500, 800, 1200 1500, 600, 800, 1200 13500, 600, 800, 1200, 2500 5500, 600, 800, 1200, >3000 1500, 800, 900, 1200, 2500 1500, 800, 1000, 1200, 2500 2620, 900, 1300, 1700, >3000 1500, 600, 800, 900,1200, 2500 2500, 600, 800, 1000, 2300, 2500 1500, 600, 800, 1200, 1500, 2500 2500, 600, 800, 1200, 2500, 3000 2500, 800, 1000, 1200, 1500, 2500 1500, 600, 800, 900, 1200, 1500, 2500 4500, 600, 800, 1000, 1200, 1500, 2500 4500, 600, 800,1200, 1500, 2400, 2500 1500, 600, 800, 900, 1200, 1500, 1700, 2500, >3000 1500, 600, 750, 800, 900, 1200, 1300, 1500, 2500, >3000 1

Table IVSize of amplicons when primers used to amplify integron class 1

Pattern of integron I bands (pb) No. of isolates

Multidrug resistant Acinetobacter and integrons 2 167

indicate the role of the bloodstream in spreading the infection (Gisneous and Rodriguez-Bano, 2002). Con-sistent with previous studies, A. baumannii is the pre-dominant species in clinical isolates (Seifert et al., 1993; Towner, 2009).

The three most effective antibiotics against Acine-tobacter were found to be colistin, imipenem and meropenem. despite being the most effective antibio-tic against Acinetobacter in vitro, colistin use is limited only to life threatening conditions due to its serious side effects (Reed et al., 2001; Lewis and Lewis, 2004). Nevertheless, observation of high resistance rate of Acinetobacter to the majority of the tested antibiotics has limited the use of alternative effective antibiotics. More likely, resistance genes are acquired via genetic elements such as integrons, plasmids and transposons (Perez et al., 2007). In this regard, the role of integron is remarkable due to possessing a strong capturing sys-tem (Gonzalez et al., 1998; Seward, 1999; Turton et al., 2005). Continuous capturing of antibiotic resistance genes in Acinetobacter will extend quickly, so with more uncontrolled administration of antibiotics in hospitals and clinics, the possibility of acquiring resistance will be increased. To overcome progressive antibiotic resist-ance, rational and timely administration of effective antibiotics should be implemented. The present study on the existence of integron revealed that 53.4% of the isolates contained integron classes 1 or 2. These results are in agreement with published reports that Acine-tobacter harbors high prevalence of integron class 1, lower class 2 and no class 3 (Koeleman et al., 2000; Ploy et al., 2000; Galleco and Towner, 2001; Gaur et al., 2006; xu et al., 2008). The lack of integron class 3 may indi-cate its null role in antibiotic resistance. As mentioned above, the prevalence of class 1 integron, as compared to class 2 may imply that class 1 integron is more impor-tant in capturing resistant determinants. Alternatively, both systems acquire the same resistance genes but class 1 integrons may express these genes more efficiently. To determine this possibility, sequencing and cloning of resistance genes of the isolates containing class 1 or 2 integrons might be helpful. Comparison of antibiotic resistance patterns and their association with class 1 and 2 integrons confirms that both classes of integrons exhibit similar resistance patterns to the tested antibio-tics (Table III). However, class 1 integron is more likely involved in emerging resistance to antibiotics.

data in the present study show a statistical associa-tion between the presence of integrons and resistance to 6 antibiotics. Because we did not detect any association between resistance to other antibiotics and the presence of integrons, this can implicate the role of other resist-ance determinants (Gaur et al., 2006; Chen et al., 2010).

In conclusion, Acinetobacter expressed high resist-ance to most of the prescribed antibiotics. To reduce

the resistance rate, comprehensive control measures along with determination of periodical antibiotic sen-sitivity pattern may alleviate the situation to an accept-able level. Colistin, imipenem and meropenem are the most effective agents against Acinetobacter. However, the clinical application of colistin is limited due to its inappropriate side effects.

Acknowledgementsdeep thanks are due to prof. A. Alborzi for his invaluable help

with provision of the laboratory facilities in Prof. Alborzi Clinical Microbiology Research Center. We are thankful to Hassan Khajehei, Phd for critical reading of the manuscript.

Literature

Arakawa Y., M. Murakami, K. Suzuki, H. Ito, R. Wacharotayankun, S. Ohsuka, N. Kato and M. Ohta, M. 1995. A novel integron-like element carrying the metallo-b-lactamase gene blaIMP. Antimicrob. Agents. Chemother. 39: 1612–1615.Bergogne-Berezin E. and K.J. Towner. 1996. Acinetobacter spp. as nosocomial pathogens: Microbiological, clinical & epidemiological features. Clin. Microbiol. 8: 148–165.Chen T.L., W.C. Chang, S.C. Kuo, Y.T. Lee, C.P. Chen, L.K. Siu, W.L. Cho and C.P. Fung. 2010. Contribution of a plasmid borne blaOxA-58 with its hybrid promoter provided by IS1006 and ISAba3-like to {beta}-lactam resistance in Acinetobacter genomic species 13TU. Antimicrob. Agents. Chemother. 54: 3107–3112. Feizabadi M.M., B. fatollahzadeh, M. Taherikalani, M. Rasoolinejad, N. Sadeghiferd, M. Aligholi, S. Soroush and S. Mohammadi-Yegane. 2008. Antimicrobial susceptibility patterns and distribution of bla OxA genes among Acinetobacter spp. isolated from patients at Tehran hospitals. Jpn. J. Infect. Dis. 61: 274–278.Galleco L. and K.J Towner. 2001. Carriage of class 1 integrons and antibiotic resistance in clinical isolates of Acinetobacter baumannii from Northern Spain. J. Med. Microbiol. 50: 71–77.Gaur A., P. Prakash, S. Anupurba and T.M. Mahapatra. 2006. Possible role of integrase gene polymerase chain reaction as an epi-demiological marker: study of multi-drug resistant Acinetobacter baumannii isolated from nosocomial infections. Int. J. Antimicrobial. Agents. 29: 446–450.Gisneous J.M. and J. Rodriguez-Bano. 2002. Nosocomial bactere-mia due to Acinetobacter baumannii: epidemiology, clinical features and treatment. Clin. Microbiol. Infects. 8: 687–693.Gonzalez G., K. Sossa, H. Bello, M. Dominguez, S. Mella and R. Zemelman. 1998. Presence of integrons in isolates of different biotypes of Acinetobacter baumannii from Chilean hospitals. FEMs. Microbiol. lett. 161: 125–128.Koeleman J.G., J. Stoof, M.W. Van Der bijl, C. M.Vandenbroucke-Grauls and P.H. Savelkoul. 2000. Identification of epidemic strains of Acinetobacter baumannii by integrase gene PCR. J. Clin. Microbiol. 39: 8–13.Lévesque C., L. Piché, C. Larose and P.H. Roy. 1995. PCR mapping of integrons reveals several novel combinations of resistance genes. Antimicrob. Agents. Chemother. 39: 185–91.Lewis J.R. and S.A. Lewis. 2004. Colistin interactions with mam-malian urothelium. Am. J. physiol. Cell. physiol. 286: C913– C922.Neild B.S., A.J. Holmes, M.R. Gillings, G.D. Recchia, , B.C. Mabbutt, K.M. Nevalainen and H.W. Stokes. 2001. Recovery of new integron classes from environmental dNA. FEMS. Microbiol. Lett. 195: 59–65.Nordmann P. 2004. Acinetobacter baumannii, the nosocomial pathogen par excellence. Pathol. Biol. 52: 301–303.


Perez F., A.M. Hujer, K.M. Hujer, B.K. Decker, P.N. Rather and R.A. Bonomo. 2007. Global challenge of multidrug-resistant Acine-tobacter baumannii. Antimicrob. Agents. Chemother. 51: 3471–3484.Ploy M.C., F. Denis, P. Courvalin and T. Lambert. 2000. Molecular characterization of integrons in Acinetobacter baumannii: descrip-tion of a hybrid class 2 integron. Antimicrob. Agents. Chemother. 44: 2684–2688.Radström P., O. Sköld, G. Swedberg, J. Flensburg, P.H. Roy and L. Sundström. 1994. Transposon Tn5090 of plasmid R751, which carries an integron, is related to Tn7, Mu, and the retroelements. J. Bacteriol. 176: 3257–3268.Reed M.D., R.C. Stern, M.A. O’Riordan and J.L. Blumer. 2001. The pharmacokinetics of colistin in patients with cystic fibrosis. J. Clin. Pharmacol. 41: 645–54.Rowe-Magnus D. A. and D. Mazel. 1999. Resistance gene capture. Curr. Opin. Microbiol. 2: 481–486.

Seifert H., R. Baginsky, A. Schulze and G. Polverer. 1993. The distribution of Acinetobacter species in clinical culture materials. Zentralbl. Bakteriol. 279: 544–552.Seward R.J. 1999. detection of integrons in worldwide nosocomial isolates of Acinetobacter spp. Clin. Microbiol. Infect. 5: 308–318.Towner K.J. 2009. Acinetobacter: an old friend, but a new enemy. J. Hosp. Infect. 73: 355–63. Turton J.F., M.E. Kaufmann, J. Glover, J.M. Coelho, M. Warner, R. Pike and T.L. Pitt. 2005. detection and typing of integrons in epidemic strains of Acinetobacter baumannii found in the United Kingdom. J. Clin. Microbiol. 43: 3074–3082.Villegas M.V. and A.I. Hartstein. 2003. Acinetobacter outbreaks, 1977–2000. Infect. Control. Hosp. Epidemiol. 24: 284–95.Xu X., F. Kong, X. Cheng, B. Yan X. Du, J. Gai, H.Ai, L. Shi and J. Iredell. 2008. Integron gene Cassettes in Acinetobacter spp. Strains from south China”. Int. J. Antimicrobial. Agents. 32: 441–445.


ORGINAL PAPER

Introduction

Acinetobacter baumannii is an important opportun-istic pathogen responsible for a variety of nosocomial infections, including ventilator-associated pneumonia, bacteremia, surgical-site infections, secondary menin-gitis, and urinary tract infections (von dolinger et al., 2005; Fontana et al., 2008; Peleg et al., 2008).

Most A. baumannii infections are caused by the outbreak strains, which can spread widely and rapidly between patients. Since these strains also exhibit multi-ple-antibiotic resistance, it has been suggested that epi-demic potential among isolates of A. baumannii may be linked to the presence of integrons.

Integrons are dNA elements capable of capturing genes by a site-specific recombination mechanism that often carry gene cassettes, containing antibiotic resist-ance genes (Turton, et al., 2005). Various studies have reported the existence of antibiotic resistance genes located on integrons among Acinetobacter spp. (Gallego

and Towner 2001; Navia et al., 2002; Nemec et al., 2004; Zarrilli et al., 2004).

Several classes of integrons have been described, with class I integrons being the most common and widely distributed among Gram-negative bacteria. Integrons have been found in isolates of Acinetobacter spp. from differ-ent locations of the world and it has been suggested that multi-resistant isolates of Acinetobacter spp. may act as a reservoir of integron-associated antibiotic resistance gene, which could then spread to other pathogens in the hospital environment (Gallego and Towner 2001).

Few studies have hitherto focused on the distribu-tion of antibiotic resistance genes among Acinetobacter spp. in Iran (Feizabadi et al., 2008; Taherikalani et al., 2008; Taherikalani et al., 2009; Akbari et al., 2010); how-ever, there is limited information on the detection of different classes of integrons in Iran.

This study aimed to determine the distribution of class 1, 2 and 3 integrons among A. baumannii isolates, collected from different clinical specimens in selected

Dissemination of Class 1, 2 and 3 Integrons among Different Multidrug ResistantIsolates of Acinetobacter baumannii in Tehran Hospitals, Iran

MOROVAT TAHERIKALANI1,2*, ABBAS MALEKI2, NOURKHOdA SAdEGHIFARd1,2,dELBAR MOHAMMAdZAdEH3, SETAREH SOROUSH2, PARISA ASAdOLLAHI2,

KHAIROLLAH ASAdOLLAHI4 and MOHAMMAd EMANEINI5

1 department of Medical Microbiology, School of Medicine, Ilam University of Medical Sciences, Ilam, Iran2 Clinical Biology Research Center, Ilam University of Medical Sciences, Ilam, Iran

3 department of Microbiology, Science and Research Branch, Islamic Azad University of Tehran, Iran4 department of Epidemiology, School of Medicine, Ilam University of Medical Sciences, Ilam, Iran

5 department of Microbiology, School of Medicine, Tehran University of Medical Sciences, Tehran, Iran

Received 3 January 2011, revised 10 April 2011, accepted 12 April 2011

A b s t r a c t

A total of 100 non-duplicate Acinetobacter baumannii isolates were collected from different hospitals in Tehran and were confirmed as A. baumannii by conventional biochemical and API testing. Antimicrobial susceptibility of these isolates was checked by a disk diffusion method in accordance with CLSI guidelines. The isolates were then detected as carrying class 1 and 2 integron gene cassettes by PCR evaluation and then genotyped by REP-PCR. More than 50% (n = 50) of the isolates were multidrug resistant. The results showed that more than 80% of all multidrug resistant A. baumannii strains carry a class 1 integron. distribution of IntI 1 and IntI2 among A. baumannii isolates was 58% and 14%, respectively. Analysis of a conserved segment of class 1 integron showed a range from 100 bp to 2.5 kb. REP-PCR fingerprinting showed more than 20 genotypes among A. baumannii strains. There was no relationship between REP genotypes and the distribution of different classes of integrons. This is a comprehensive study on the distribution of different classes of integrons among A. baumannii in Iran. Considering the exact role of integrons in coding drug resistance in bacteria, the findings of this study could help us find antimicrobial resistant mechanisms among A. baumannii isolates in Iran.

K e y w o r d s: A. baumannii, hospital isolates in Iran, integron classes

* Corresponding author: M. Taherikalani, department of Microbiology, School of Medicine, and Clinical Microbiology Research Center, Ilam University of Medical Sciences, Banganjab, Ilam, IR of Iran, Postal Zip: 69391-77143; phone: +98-841-223-5747; fax. +98-841-2227136; e-mail: [email protected]

Taherikalani M. et al. 2170

hospitals, in Tehran and to evaluate any correlations between antibiotic resistance and the carriage of differ-ent classes of integrons among A. baumannii isolates.

Experimental


Bacterial isolates. A total of 100 non-duplicate isolates of Acinetobacter spp. were collected from differ-ent clinical specimens during 2007–2009. These isolates were confirmed as A. baumannii by conventional bio-chemical testing and 20NE API galleries (BioMerieux, Inc) used in the previous studies (Feizabadi et al., 2008; Taherikalani et al., 2009; Akbari et al., 2010). These studies were carried out in the laboratory of microbio-logy in Ilam University, Iran. Fifty six percent (n = 56) of the isolates were recovered from wound and tra-chea. The strains isolated were then stored at –80°С in nutrient broth containing 50% glycerol v/v for fur-ther investigation.

Antimicrobial susceptibility testing. Antimicro-bial susceptibility testing was performed by disk agar diffusion, according to CLSI guidelines. The applied antimicrobials were as follows: ampicillin-sulbactam (10/10 µg), piperacillin (100 µg), cefotaxime (30 µg), ceftazidime (30 µg), cefteriaxone (30 µg), cefepime (30 µg), imipenem (10 µg), ciprofloxacin (5 µg), amikacin (30 µg), gentamicin (10 µg) and tetracycline (10 µg). Inoculums of the A. baumannii isolates (106 CFU) were swabbed on several Muller-Hinton agar plates, and different disks, impregnated each with different antibiotics, were then placed on these plates. Incubation at 37°C for 24 h was then carried out after which the inhibition zones were read. Escherichia coli ATCC 25922, Staphylococ-cus aureus ATCC 29213 and Pseudomonas aeruginosa ATCC 27853 were used as control strains.

PCR amplification of integron-associated genes for different integron classes. dNA extraction was car-ried out using commercial standard kit (Bioner, Repub-lic of Korea) and 4 μl of the suspension were used as the template dNA for PCR.

PCR annealing temperature, primer sequences and the amplicon sizes are listed in Table I (Srinivasan, Rajamohan et al., 2009).

The PCR conditions were as follows: initial dena-turation at 95°C for 5 min; 30 cycles with denaturation at 95°C for 30s, annealing at 50°C, 51°C, 52°C and 53°C for 30s for integrons class 1, 2 and 3 and conserved sequence of integron class 1 respectively, and extension at 72°C for 45s followed by final extension at 72°C for 7 min. PCR products were separated by electrophoresis on a 1% agarose gel and were detected by comparison against a 100 bp dNA ladder as a size marker under the visualization of UV light on Geldoc apparatus.

REP-PCR Finger-printing. dNA extraction was carried out by dNA extraction kit (Bioner, Republic of Korea) and 4 μl of the extract was used as the tem-plate dNA. The primer pair REP1, 5’-IIIGCGCCGI-CATCAGGC-3’ and REP2, 5’-ACGTCTTATCAG-GCCTAC-3’ were used to amplify putative REP-like elements in the genomic bacterial chromosomes (Bou et al., 2000). Amplification reaction was performed in a final volume of 25 μl. Each reaction mixture contained 2.5 μl of 10x PCR buffer, 1.25 U Taq dNA polymerase (Fermentas, UK), and 0.8 μl of mixed dNTPs (Fermen-tas, UK), 1.5 μl of 25 Mm MgCl2, 1 μl of 10 pmol prim-ers and 50 ng of bacterial dNA. Amplification reaction was carried out by thermal cycler (Ependorff, Germany) with an initial denaturation at 94°C for 10min, followed by 30 cycle of denaturation at 94°C for 1 min, anneal-ing at 45°C for 1 min, and extension at 72°C for 1 min, followed by final extension at 72°C for 16 min. Ali-quots of each sample were subjected to electrophoresis in 1.2% agarose gels. Amplified products were detected by Geldoc apparatus after staining with ethidium bro-mide (50 mg/L) and the created photographs were then analysed visually and with the TotalLab TL120 software.

Results

The most effective antimicrobial agents against A. baumannii isolates were as follows: gentamicin 55% (n=55), imipenem 47% (n=47), ampicillin-sulbactam, ami-

intI 1 5’-ACATGTGATGGCGACGCACGA-3’ 5’-ATTTCTGTCCTGGCTGGCGA-3’ 50 300intI 2 5’-CACGGATATGCGACAAAAAGGT-3’ 5’-GTAGCAAACGAGTGACGAAATG-3’ 51 962intI 3 5’-GCCTCCGGCAGCGACTTTCAG-3’ 5’-ACGGATCTGCCAAACCTGACT-3’ 52 1041ConservedSegment of IntI 1 (5’-CS) 5’-GGCATCCAAGCAGCAAG-3’ (3’-CS) 5’-AAAGCAGACTTGACCTGA-3’ 53 Variable

Table IPrimers used in PCR amplification of integron classes 1 to 3

Target geneSize of

amplicon(bp)

AnnealingTemperature

(°С)ReverseForward

Classes of integrons among hospital isolates of A. baumanniiin Iran2 171

kacin 38% (n = 38) and tetracycline 31% (n = 31). Most isolates showed high resistance to piperacillin (100%) and cephalosporin drugs (more than 95%).

The REP- fingerprinting of some A. baumannii iso-lates are shown in Fig. 1. All the isolates not previously compared with REP or other typing methods were revealed to have 20 REP patterns. No reliable REP pat-tern was observed among 15 isolates.

PCR detecting integrase gene showed that 58% (n=58) of all the isolates had intI 1; however, intI 2 was only identified in 14% (n = 14) of the isolates and intI 3 was not revealed in any of the clinical isolates. The coexist-ence rate of intI 1/intI 2 was 9% (n = 9). The relation-ship between antibiotic resistance and the existence of different integrons is shown in Table II. More than 50% of penicillin and cephalosporin resistant isolates har-

Piperacillin 100 58 (58) 14 (14) 9 (9)Ampicillin- sulbactam 62 24 (37.5) 3 (4.83) 1 (1.61)Ciprofloxacin 85 51 (60) 14 (16.4) 9 (10.58)Amikacin 62 27 (43.54) 10 (16.12) 7 (11.29)Imipenem 53 24 (45.28) 7 (13.20) 4 (7.54)Cefotaxime 97 57 (58.76) 13 (13.4) 9 (9.27)Cefepime 99 58 (58.58) 14 (14.14) 9 (9.09)Ceftazidime 97 57 (58.76) 13 (13.4) 9 (9.27)Ceftriaxone 97 56 (57.73) 14 (14.43) 9 (9.27)Tetracycline 31 16 (51.6) 7 (22.58) 5 (16.12)Gentamicin 45 18 (40) 8 (17.7) 4 (8.88)

Table IIdistribution of intI 1 and intI 2 among A. baumannii isolates resistant to different antibiotic agents

Antibiotic agent Class 2 integronn (%)t

Class 1&2integrons n (%)

Class 1 integronn (%)

Number of resistantisolates t

Fig. 1. Variable amplicon size of conserved segmentsof integron class 1.

Lane 1 (negative control: ddW); Lanes 3 and 8 (clinical positive sample); Lanes 2 and 4–7 and 9–11 and 13–21 (clinical negative samples).

Fig. 4. REP PCR. Pattern of genomic dNA from 19 clinicalA. baumannii isolates.

Lanes 1–10, 12–20 (clinical isolates of A. baumannii), Lane 11(dNA Ladder 100 bp to 3000 bp).

Fig. 2. PCR of class 1 integron among clinical A. baumannii isolates.Lanes 1–2, 4–5, 7–8, 10 (clinical positive isolates), Lane 11

(100 bp dNA size marker), Lanes 12–20 (clinical positive isolates).

Fig. 3. PCR of class 2 integron among clinical A. baumannii isolates. Lanes 1–2, 8, 9 13–14 (clinical positive isolates), Lane 11 (100 bp dNA size

marker); Lanes 3–7, 9, 12, 15–20 (clinical negative isolates).


bored different integrons. Ceftriaxone and cefotaxime resistant isolates were the most common intI 1 harboring isolates (58.76%), followed by ceftriaxone and cefepime with 58.73% and 58.58%, respectively. However, tetra-cycline resistant isolates were the most common intI 2 harboring isolates (22.58%), followed by gentamicin and amikacin with 17.7% and 16.12%, respectively (Figs 3, 4). Amplification of the integron gene cassettes of the inte-grase positive isolates gave PCR products of various sizes ranging from 100 bp to 2.5 kb (Fig. 4). distribution of the integron gene cassettes among integrase positive isolates was accounted for 10% (n = 10). Various amplicons of integron gene cassettes were not seen among all integrase positive isolates. The correlations between REP-genotype, different hospitals, wards, sample origins and the existence of integrons are shown in dendogram (Fig. 5).

Discussion

Acinetobacter baumannii is typically resistant to various antimicrobial agents such as penicillins, cepha-losporines, macrolides, aminoglycosides, tetracyclines and fluoroquinolones (Wang et al., 2007).

Because of the multidrug resistance and tendency to spread in hospital population, A. baumannii has a spe-cial clinical significance, requiring epidemiologic moni-toring as a measure for control of nosocomial infection.

On the basis of the sequence of integrase gene, inte-grons are divided into at least six classes, with class 1 integron being the most common among the clinical isolates of Gram-negative bacteria, including acineto-

bacters (Koeleman et al., 2001; Turton et al., 2005). It seems that class 2 integrons were rarely detected in Aci-netobacter spp., but class 3 integrons were not detected in those bacteria at all (Ploy et al., 2000; Koeleman et al., 2001; Turton et al., 2005).

PCR detecting integrase gene used has advantages over the integron cassette PCR in screening for inte-grons, in that it is designed to give a small product which is easily amplified. Integron cassette PCR can give a negative result even when the integrons are pre-sent, if the cassette array is difficult to amplify or if there are no cassettes present. The PCR detecting inte-grase gene was simple, reliable, and easy to perform (Turton et al., 2005).

In the current study a high prevalence of intI 1 and intI 2 was found among multidrug resistant A. bau-mannii strains, isolated from different specimens. The results related to class1 integron were in concordance with other studies (Gonzalez et al., 1998; Seward 1999). Although some studies clarified the presence of intI 2 among A. baumannii strains, only 14% of the isolates in this study seemed to harbor this class of integrons. These findings are relatively significant since most stud-ies report that class 2 integrons are not found or are found in low rates among A. baumannii strains (Ploy et al., 2000; Turton et al., 2005). Class 3 integron was not found among A. baumannii strains, which was in accordance with other reports (Ploy et al., 2000). Although amplicons with variable sizes were found in integron gene cassette, in agreement with other studies, these variable amplicon could not be detected among all integrase positive strains (Turton et al., 2005).

Fig. 5. UPGMA dendogram illustrating the relationships between sample origins, hospitals, wards and the existence of class 1 and 2 integron genes among the genotypes of A. baumannii isolates.

Classes of integrons among hospital isolates of A. baumanniiin Iran2 173

Antibiotic resistance is an important factor in the spread of nosocomial infection. It is generally consid-ered that the existence of integrons confers the advan-tage of antibiotic resistance upon the strains. Among the multiresistant strains described in this study, there were still some strains which did not contain inte-grons (of classes 1 and 2 at least); however, most of them were susceptible to gentamicin, imipenem, and ampicillin-sulbactam.

Integrons containing the same organization of cas-settes were found in various REP genotypes, suggesting horizontal transfer of integrons, also reported in other studies (Sallen et al., 1995; Seward 1999). In addition, the promoter sequences were mostly conserved, even in isolates from different countries with distinct selec-tive pressure, suggesting that acquisition of resistance is likely due to transfer of entire integrons via plasmids and/or transposons rather than of individual cassettes (Ploy et al., 2000). In concordance with other studies, it was proven that the isolates of the same genotype possess different integrons and in the same way, the unrelated isolates with different genotypes could con-tain the same integrons.

Similar 2.5-kb integrons, with integron cassettes, found by PCR in the present study, have been widely found in isolates of European clones I and II from many countries (Nemec et al., 2004; Turton et al. 2005). The 2.5-kb integron has also been found in a number of outbreak strains of A. baumannii of different genotypes in Italy, Russia and Ireland (Gombac et al., 2002; Zarrilli et al., 2004; Turton et al., 2005). In conclusion, inte-grons could be a feature of epidemic strains or clones of A. baumannii currently found in Iran. Information on both the genotype and integron type is useful in epidemiological studies. The association of integrons with epidemic behaviors merits further studies.

AcknowledgmentThis work was supported by Ilam University of Medical Scien ces.

We acknowledge the microbiology lab workers of Ilam University of Medical Sciences who cooperated in this work.

Literature

Akbari M., M. Niakan, M. Teherikalani, M.M. Feizabadi, N.A. Azadi, S. Soroush, M. Emaneini, A. Abdolkarimi, A. Maleki and A. Hematian. 2010. Rapid identification of Iranian Acinetobac-ter baumannii strains by single PCR assay using bla OxA-51-like carbenemase and evaluation of the antimicrobial resistance profiles of the isolates. Acta Microbiol. Immunol. Hung. 57: 87–94.Bou, G., G. Cervero, M.A. Dominguez, C. Quereda and J. Martinez- eltran. 2006. PCR-based dNA fingerprinting (REP-PCR, AP-PCR) and pulsed-field gel electrophoresis characterization of a nosocomial outbreak caused by imipenem- and meropenem-resistant Acinetobacter baumannii. Clin. Microbiol. Infect. 6: 635–643

Clinical and Laboratory Standard Institute. 2006. Performance Standards for Antimicrobial disk Susceptibility tests; 9th for infor-mational Supplement, M2-A9.Clinical and Laboratory Standards Institute, Wayne, Pa.von Dolinger de Brito D., E.J. Oliveira, V.O. Abdallah, A.L. da Costa Darini and P.P. Filho. 2005. An outbreak of Acinetobacter baumannii septicemia in a neonatal intensive care unit of a univer-sity hospital in Brazil. Braz. J. Infect. Dis. 9: 301–309.Fontana C, M. Favaro, S. Minelli, M.C. Bossa, G.P. Testore, F. Leonardis, S. Natoli and C. Favalli. 2008. Acinetobacter bauman-nii in intensive care unit: a novel system to study clonal relationship among the isolates. BMC Infect. Dis. 8: 79.Feizabadi M.M., B. Fathollahzadeh, M. Taherikalani, M. Rasooli- nejad, N. Sadeghifard, M. Aligholi, S. Soroush and S. Mohammadi- egane. 2008. Antimicrobial susceptibility patterns and distribution of blaOxA genes among Acinetobacter spp. Isolated from patients at Tehran hospitals. Jpn. J. Infect. Dis. 61: 274–278.Gallego L. and K.J. Towner. 2001. Carriage of class 1 integrons and antibiotic resistance in clinical isolates of Acinetobacter baumannii from northern Spain. J. Med. Microbiol. 50: 71–77.Gonzalez G., K. Sossa, H. Bello, M. Dominguez, S. Mella and R. Zemelman. 1998. Presence of integrons in isolates of different biotypes of Acinetobacter baumannii from Chilean hospitals. FEMS. Microbiol. Lett. 161: 125–128.Gombac F., M.L. Riccio, G.M. Rossolini, C. Lagatolla, E. Tonin, C. Monti-Bragadin, A. Lavenia and L. Dolzani. 2002. Molecular characterization of integrons in epidemiologically unrelated clinical isolates of Acinetobacter baumannii from Italian hospitals reveals a limited diversity of gene cassette arrays. Antimicrob. Agents. Chem-other. 46: 3665–3668.Koeleman J.G, M.W. van der Bijl, J. Stoof, C.M. Vandenbroucke-Grauls and P.H. Savelkoul. 2001. Antibiotic resistance is a major risk factor for epidemic behavior of Acinetobacter baumannii. Infect. Control. Hosp. Epidemiol. 22: 284–288.Navia M.M., J. Ruiz and J. Vila. 2002. Characterization of an inte-gron carrying a new class d beta-lactamase (OxA-37) in Acineto-bacter baumannii. Microb. Drug. Resist. 8: 261–265.Nemec A., L. Dolzani, S. Brisse, P. van den Broek and L. Dijk-shoorn. 2004. diversity of aminoglycoside-resistance genes and their association with class 1 integrons among strains of pan-European Acinetobacter baumannii clones. J. Med. Microbiol. 53: 1233–1240.Peleg A.Y, H. Seifert and D.L. Paterson. 2008. Acinetobacter bau-mannii: emergence of a successful pathogen. Clin. Microbiol. Rev. 21: 538–582.Ploy M.C, F. Denis, P. Courvalin and T. Lambert. 2000. Molecular characterization of integrons in Acinetobacter baumannii: descrip-tion of a hybrid class 2 integron. Antimicrob. Agents. Chemother. 44: 2684–2688.Sallen B., A. Rajoharison, S. Desvarenne ans C. Mabilat. 1995. Molecular epidemiology of integron-associated antibiotic resistance genes in clinical isolates of enterobacteriaceae. Microb. Drug. Resist. 1: 195–202.Seward R.J. 1999. detection of integrons in worldwide nosocomial isolates of Acinetobacter spp. Clin. Microbiol. Infect. 5: 308–318.Srinivasan V.B, G. Rajamohan, P. Pancholi, K. Stevenson, D. Tadesse, P. Patchanee, M. Marcon and W.A. Gebreyes. 2009. Genetic relatedness and molecular characterization of multidrug resistant Acinetobacter baumannii isolated in central Ohio, USA. Ann. Clin. Microbiol. Antimicrob. 8: 21.Turton J.F, M.E. Kaufmann, J. Glover, J.M. Coelho, M. Warner, R. Pike and T.L. Pitt. 2005. detection and typing of integrons in epidemic strains of Acinetobacter baumannii found in the United Kingdom. J. Clin. Microbiol. 43: 3074–3082.


Taherikalani M., B. Fatolahzadeh, M. Emaneini, S. Soroush and M.M. Feizabadi. 2009. distribution of different carbapenem resistant clones of Acinetobacter baumannii in Tehran hospitals. New Microbiol. 32: 265–271.Taherikalani M., G. Etemadi, K.N. Geliani, B. Fatollahzadeh, S. Soroush and M.M. Feizabadi. 2008. Emergence of multi and pan-drug resistance Acinetobacter baumannii carrying blaOxA-type -carbapenemase genes among burn patients in Tehran, Iran. Saudi. Med. J. 29: 623–624.

Wang H., P. Guo, H. Sun, H. Wang, Q. Yang, M. Chen, Y. Xu and Y. Zhu. 2007. Molecular epidemiology of clinical isolates of carbapenem-resistant Acinetobacter spp. from Chinese hospitals. Antimicrob. Agents. Chemother. 51: 4022–4028.Zarrilli R., M. Crispino, M. Bagattini, E. Barretta, A. Di Popolo, M. Triassi and P. Villari. 2004. Molecular epidemiology of sequen-tial outbreaks of Acinetobacter baumannii in an intensive care unit shows the emergence of carbapenem resistance. J. Clin. Microbiol. 42: 946–953.


SHORT COMMUNICATION

On the basis of various epidemiological studies H. pylori has been classified as a type I carcinogen by the Working Group of the World Health Organiza-tion International Agency for Research on Cancer in 1994 (IARC 1994). H. pylori is specialized in coloniz-ing human gastric mucosa of more than 50% of world population. It can be a cause of chronic gastritis, peptic ulcers, and gastric adenocarcinoma (Wedi et al., 2002). The bacterium has also been implicated in extra-gastric conditions such as ischemic heart disease, vascular and immunological disorders, halitosis, migraine, and poor growth in children (Pellicano et al., 2008). Recently, Helicobacter spp. dNA has been found in the liver of patients with various chronic liver diseases, such as pri-mary sclerosing cholangitis, hepatocellular carcinoma (HCC), hepatitis C virus-related chronic infection, and cirrhosis. Inflammatory disease is characterized by increased levels of pro-inflammatory cytokines, such as interleukins 1 and 8 (IL-1 and IL-8). The higher prevalence of Helicobacter spp. associated with more advanced stages of the liver disease supports the possi-bility of their role in the progression of chronic hepatitis towards cirrhosis and HCC (Pellicano et al., 2008).

In this study we aimed at detecting Helicobacter spp. genetic material in patients with chronic liver diseases in the population of Northern Poland. Further, the host

response to the presence of Helicobacter spp. in the liver was investigated.

The study included 27 patients suffering from differ-ent chronic liver diseases (CLd): Hepatitis B (HBV) and C (HCV) virus infections, HBV/HCV double infection, non-alcoholic steatohepatitis, non-alcoholic fatty liver disease with HCV infection, non-alcoholic fatty liver disease with HBV infection, hereditary hemochromato-sis, alcoholic steatohepatitis and autoimmunohepatitis. The biopsy from each patient was halved, and used for dNA and RNA extraction, respectively.

dNA was extracted using a High Pure PCR Template Preparation Kit (Roche). Twenty-seven dNA sam ples with high quality and quantity were amplified. Helico-bacter spp. dNA was detected by nested polymerase chain reaction with genus specific primers targeting Helico- bacter spp. 16S rRNA gene. The reaction mixture for the first step contained (25 µl):100 ng of genomic dNA, 1x chelating buffer, 2.5 mM Mg(OAc)2, 0.2 mM dNTP, 0.4 U of Taq dNA polymerase (Fermentas), 0.1 mg/ml of casein, 0.01% (v/v) formamide and 0.125 μM of primers: 1F and 1R (Al-Soud et al., 2003). Amplifica-tion conditions for the first PCR were: 94°C for 2 min; then 30 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 30 s; and finally 72°C for 5 min. The reaction mix-ture for the second step (25 µl) contained: 1x chelating

Host Response to the Presence of Helicobacter spp. DNA in the Liver of Patientswith Chronic Liver Diseases

MAGdA RYBICKA1, JOANNA NAKONIECZNA1, *, PIOTR STALKE2

and KRZYSZTOF PIOTR BIELAWSKI1

1 Intercollegiate Faculty of Biotechnology, University of Gdańsk and Medical University of Gdańsk, Poland2 department of Infectious diseases, Medical University of Gdańsk, Gdańsk, Poland

Received 1 december 2010, revised 18 January 2011, accepted 25 January 2011

A b s t r a c t

Literature data indicate an association between the presence of Helicobacter spp. in the liver and the development of hepatocellular carcinoma (HCC). However, the role of H. pylori infections in chronic liver diseases (CLd) remains controversial. The aim of this study was to detect Helicobacter spp. dNA in patients with CLd, and to investigate the host response to the presence of the bacterium in the liver. Helicobacter spp. dNA was detected in 59% samples. H.pylori was the most prevalent species (94%). We estimated the expression level of IL-1 and IL-8 genes. The presence of Helicobacter spp. did not have a significant effect on the gene expression of IL-8 and IL-1.

K e y w o r d s: Helicobacter, interleukin 1, interleukin 8, nested-PCR

* Corresponding author: J. Nakonieczna, Intercollegiate Faculty of Biotechnology, University of Gdańsk and Medical University of Gdańsk, Poland; Kladki 24, 80-822 Gdańsk, Poland; phone: 48 58 5236332; e-mail: [email protected]

Rybicka M. et al. 2176

buffer, 2.5 mM MgCl2, 0.2 mM dNTP, 2.5 U of Ampli-Taq Gold polymerase, 0.25% (v/v) glycerol, 0.4% (v/v) BSA, 0.125 μM of primers: 2F (5’-AGGGAATATT-GCTCAATGGG-3’, designed by the authors) and 2R (Al-Soud et al., 2003), and 1 μl of the first amplification step product. Amplification conditions were: 95°C for 10 min; than 35 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 30 s; and finally 72°C for 5 min.

Helicobacter genus-specific PCR products were sequenced, aligned and compared with the sequences from GenBank database using BLASTN 2.2.1 (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Total RNA was extracted by a RNeasy® kit (Qiagen), 100 ng of RNA was used for reverse strand cdNA synthesis according to the manufacturer’s protocol (Quanti Tect® Reverse Transcription Kit, Qiagen).

Expression of IL-1 and IL-8 genes was quantified using real-time RT-PCR (LightCycler®; Roche diag-nostic). β-glucuronidase gene (GUS) was used as refer-ence (Romanowski et al., 2008).

The reaction mixture contained (20 μl): 2 µl of cdNA, 5 μM concentration of each primer, 3 mM of MgCl2 and 2 µl of ready-to-use Light Cycler® dNA Master SYBR Green I (Roche diagnostic). The polymerase was acti-vated at 95°C for 10 min. The following cycling condi-tions were used in the reaction: 1 s (IL-1, IL-8) and 5 s (GUS) 95°C denaturation step, 15 s annealing at 64°C (IL-1, IL-8) and 60°C (GUS), and 20 s (IL-1, IL-8) and 10 s (GUS) extension at 72°C. Melting-curve analysis followed 45 and 50 cycles of IL-1, IL-8 and GUS genes amplification, respectively. Results were normalized with respect to the reference gene.

To determine expression levels of genes of interest in the studied samples, standard curves for IL-1 and IL-8 were generated (serial dilutions of a standard sample used at this step were as follows: 1x, 0.5x, 0.2x, 0.1x). As a standard we used a mixture of all cdNAs. Each

sample was studied in three repetitions and normalized with respect to the reference gene.

The role of H. pylori in various gastric and duodenal diseases is well documented. In recent years H. pylori dNA has been identified in the bile, gallbladder and liver tissue of patients with different hepato-biliary dis-eases. Recently published detection rates of Helicobacter spp., in tissue samples collected from patients with liver cancer, varied from 0% to 100% (Avenaud et al., 2000; Nilsson et al., 2001; Verhoef et al., 2003; Huang et al., 2004; Ito et al., 2004; Rocha et al., 2005; Vivekanandan and Torbenson 2008). In our study we selected 16S rdNA as the target because it contains both conserved and highly variable regions and gene sequences for almost all Helicobacter species available in public data bases. We identified Helicobacter genus-specific PCR products in 16/27 (59%) of the liver specimens, which constitutes a more than 2-fold increase of the detection frequency in comparison to the data previously pub-lished by our group (Stalke et al., 2005). The differences in detection level might be related to improvements in PCR reaction conditions. Previously only BSA was used to avoid the inhibitory effect of bile, whereas at present we used several PCR facilitators which improved the frequency of Helicobacter spp. dNA detection. Moreo-ver, the amount of dNA per reaction, which we used as a template was 20 ng, whereas previously 5 μl of each dNA solution was added, regardless of the concentra-tion. Additionally, in the previously published study, diluted product from the first PCR reaction was used as a template in the second step, while here, we added undiluted product. All this differences could influence detection sensitivity.

The dNA sequence of 15 out of 16 (94%) positive samples showed the highest similarity to H. pylori 16S rRNA gene, whereas in 1 sample (6%), H. cetorum-like dNA was detected. detailed chromatogram analysis

ALT (alanine aminotransferase) [IU/L] 80.73±107.5 83±78.8 10–409 0.45AST (aspartate aminotransferase) [IU/L] 59.26±77.31 51.4±39.92 12–316 0.62ALP (alkaline phosphatase) [IU/L] 97±45.72 87.6±61.5 29–212 0.42GGTP (γ-glutamyl transpeptidase) [IU/L] 108.53±164.82 178.5±208.86 9–707 0.2HGB (hemoglobin) [mg/dL] 13.58±1.88 0.77±0.34 0.24–2.31 0.49Bilirubin [mg/dL] 0.94±0.61 0.86±0.44 0.24–2.31 1Liver biopsy grading 1±0.83 0±0.52 0–3 0.13Liver biopsy staging 0.47±0.74 0.5±1.08 0–3 0.66IL-8 relative expression (in arbitrary units) 1.1±0.9 1.3±2.5 0–6 0.35IL-1 relative expression (in arbitrary units) 27±66.1 16±31.4 0.12–259 0.62

Table IResults of liver function test, blood morphology and expression of interleukin 1 and 8 in liver in patients with CLd

Helicobacter positive Helicobacter negative Range P*

P* Statistical significance was assessed by U Mann-Whitney Test

Short communication2 177

revealed that 3 samples contained mixed dNA mate-rial. In those three samples H. rodentium-like dNA was identified based on double fluorescent signals in positions where the 16S rRNA genes differ for H. pylori and H. rodentium. This may indicate that in humans, different isolates of particular Helicobacter species may exist, like it was found in other species (Nilsson et al., 2004). In our recently published results we also found H. rodentium-like dNA by means of denaturing gradi-ent gel electrophoresis (dGGE) and subsequent dNA sequencing (Nakonieczna et al., 2010). This H. roden-tium-like dNA was actually the most prevalent one in the previously studied group of patients, which in com-parison to the presented data was less heterogeneous.

Literature data indicate that expression levels of IL-1 and IL-8 are higher in gastric epithelial cells of H. pylori-infected than in uninfected patients (Backhed et al., 2003). As the phenomenon of Helicobacter species pres-ence in the human liver is widely discussed with respect to its participation in disease state and/or progression, we analyzed the level of IL-8 and IL-1 in the group of patients, in which Helicobacter dNA was detected. As a control group, patients with negative result of nested-PCR were used. There were no differences in IL-8 gene expression between Helicobacter-positive and Helico-bacter-negative patients. The biopsy samples obtained from H. pylori-positive patients expressed about two times higher levels of IL-1, however this data was not statistically significant (Table I). Moreover, comparison of Helicobacter spp. positive and negative group did not show any statistically relevant differences in the func-tional liver test, including the levels of alanine ami-notransferase, aspartate aminotransferase, γ-glutamyl transpeptidase, alkaline phosphatase, hemoglobin and bilirubin (Table I). However, we did not study the sta-tus of cytotoxin associated gene A (cagA) in the Heli-cobacter-positive group, which might have influenced the results. It is known that cagA+ strains are associated with enhanced secretion of interleukins, especially IL-8 (Backhed et al., 2003; Rieder et al., 2005). The relatively low expression level of the studied interleukins can be explained by defective immune system of investigated individuals or their general low immunity. Another interpretation might be low bacterial load and their adaptation to a special environment. Besides, the expression level of cytokines is associated with disease progression and is highly increased at the initial stage of disease. Because the investigated material originated from patients with chronic liver diseases we can sup-pose that levels of IL-1 and IL-8 were low.

Lack of significant differences in expression levels of studied genes may be associated with the fact that H. pylori possesses lipopolysaccharide (LPS) with a lower virulence compared to the typical bacterial endotox-ins, such as Escherichia coli LPS. Furthermore, H. pylori

adheres and is internalized into hepatocytes more effi-ciently than into gastric epithelial cells. It appears that H. pylori may survive inside hepatocytes and effectively avoid host response (Ito et al., 2008).

Sequence analysis showed that H. pylori is the most prevalent species (94%) in the studied population. In contrast to gastric epithelial cells, the presence of H. pylori in the liver of patient with CLd had no influ-ence on IL-8 and IL-1 mRNAs status as well as bio-chemical parameters describing liver functioning.

AcknowledgmentsThis study was funded by University of Gdańsk, grant no B051-5-

-0315-8 and Medical University grant no BW 155 and ST no 79.

Literature

Al-Soud W.A., M. Bennedsen, S.L. On, I.S. Ouis, P. Vandamme, H.O. Nilsson, A. Ljungh and T. Wadström. 2003. Assessment of PCR-dGGE for the identification of diverse Helicobacter species, and application to faecal samples from zoo animals to determine Helicobacter prevalence. J. Med. Microbiol. 52: 765–771.Avenaud P., A. Marais, L. Monteiro, B. Lebail, P. Bioulac Sage and C. Balabaud. 2000. detection of Helicobacter species in liver of patients with and without primary liver carcinoma. Cancer 89: 1431–1439.Backhed F., E. Torstensson, D. Seguin, A. Richter-Dahlfors and B. Rokbi. 2003. Helicobacter pylori Infection Induces Inter-leukin-8 Receptor. Expression in the Human Gastric Epithelium. Infection and Immunity 71: 3357–3360. Huang Y., H.G. Fan, Z.M. Wang, J.H. Zhou, X.F. Tian and N. Li. 2004. Identification of Helicobacter species in human liver samples from patients with primary hepatocellular carcinoma. J. Clin. Pathol. 57: 1273–1277.IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Shistosomes, liver flukes and Helicobacter pylori. Lyon: IARC monographs, 1994: 177–243.Ito K., M. Nakamura, G. Toda, M. Negishi, A. Torii and T. Ohno. 2004. Potential role of Helicobacter pylori in hepatocarcinogenesis. Int. J. Mol. Med. 13: 221–227.Ito K., Y. Yamaoka, H. Ota. 2008. Adherence, Internalization, and Persistence of Helicobacter pylori in Hepatocytes. Dig. Dis. Sci. 53: 2541–2549.Nakonieczna J., P. Stalke, W.A. Al-Soud, T. Wadström and K.P. Bielawski. 2010. detection of Helicobacter rodentium-like dNA in the liver tissue of patients with chronic liver diseases by polymerase chain reaction-denaturing gradient gel electrophoresis and dNA sequence analysis. Diagn Microbiol Infect Dis. 68: 201–7.Nilsson H.O., R. Mulchandani, K.G. Tranberg. 2001. Helicobacter species identified in liver from patients with cholangiocarcinoma and hepatocellular carcinoma. Gastroenterology 120: 323–324.Pellicano R., A. Ménard. 2008. Helicobacter species and liver dis-eases: association or causation? Lancet Infect Dis. 8: 254–60. Rieder G., W. Fischer. 2005. Interaction of Helicobacter pylori with host cells: Function of secreted and translocated molecules. Curr. Opin. Microbiol. 8: 67–73.Rocha M., P. Avenaut, A. Menard, B. Le Bail, C. Balabaud and P. Bioulac-Sage. 2005. Association of Helicobacter species with hepatitis C cirrhosis with or without hepatocellular carcinoma. Gut 54: 396–401.

Rybicka M. et al. 2178

Romanowski T., K. Sikorska and K.P. Bielawski. 2008. GUS and PMM1 as suitable reference genes for gene expression analysis in the liver tissue of patients with chronic hepatitis. Med. Sci. Monit. 14: 147–152.Stalke P., W.A. Al-Soud, K.P. Bielawski, A. Bakowska, H. Trocha and J. Stępiński. 2005. detection of Helicobacter species in liver and stomach tissue of patients with chronic liver diseases using polimerase chain reaction-denaturin gradient gel electro-phoresis and immunohistochemistry. Scand. J. Gastroenterol 40: 1032–1041.

Verhoef C., R.G. Pot, R.A. de Man, P.E. Zondervan, E.J. Kuipers, J.N. Ijzermans and J.G. Kusters. 2003. detection of identical Helicobacter dNA in the stomach and in the non-cirrhotic liver of patients with hepatocellular carcinoma. Eur. J. Gastroenterol. Hepatol. 15: 1171–1174.Vivekanandan P. and M. Torbenson. 2008. Low frequency of Heli-cobacter dNA in benign and malignant liver tissues from Baltimore, United States. Human Pathology 39: 213–216.Wedi B. and A. Kapp. 2002. Helicobacter pylori infection in skin diseases: a critical appraisal. Am. J. Clin. Dermatol. 3: 273–282

polish journal of microbiology

Documents