University of Groningen

Carbohydrate-active enzymes that modify the cell wall of Aspergillus nigervan Munster, Jolanda

DOI:10.1016/j.carres.2015.01.014

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2014

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):van Munster, J. (2014). Carbohydrate-active enzymes that modify the cell wall of Aspergillus niger:Biochemical properties and physiological functions during autolysis and differentiation. [S.n.].https://doi.org/10.1016/j.carres.2015.01.014

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 31-03-2020

165

Kinetic characterization of Aspergillus niger chi-tinase CfcI using a HPAEC-PAD method for native chitin oligosaccharidesJolanda M. van Munster 1, Peter Sanders 1, Geralt ten Kate 1, Lubbert Dijk-huizen 1, Marc J.E.C. van der Maarel 1,2

1 Microbial Physiology Research Group, Groningen Biomolecular Sciences and Biotechnology Institute (GBB), University of Groningen, Groningen, the Netherlands2 Aquatic Biotechnology and Bioproduct Engineering Department, Institute for Technology and Manage-ment (ITM), University of Groningen, Groningen, the Netherlands

Submitted for publication

Chapter 6

166

Chapter 6

Abstract The abundant polymer chitin can be degraded by chitinases (EC 3.2.1.14) and β-N-acetyl-hexosaminidases (EC 3.2.1.52) to oligosaccharides and N-acetyl-glu-cosamine (GlcNAc) monomers. Kinetic characterization of these enzymes requires product quantification by an assay method with a low detection limit, preferably com-patible with the use of native, non-labeled substrates. Here we report a quantitative HPAEC-PAD method that allows fast separation of chitin oligosaccharides (COS) ranging from (GlcNac)1-6 at detection limits of 1-3 pmol and a linear range of 5-250 pmol. Quantification under intra- and interday precision conditions was performed with 2.1-5.4 % relative standard deviation (RSD) and 1.2-10.3 % RSD, respectively. This method was successfully used for the determination of the kinetic parameters of the Aspergillus niger chitinase CfcI with native COS. CfcI was recently shown to release GlcNAc from the reducing end of COS, a new activity for fungal chitinases. A Carbohydrate Binding Module of family 18 (CBM18) is inserted in the CfcI cata-lytic domain. Site directed mutagenesis was used to assess the functionality of this CfcI-CBM18: four of its key amino acids were replaced by glycine residues, yielding CfcISYNF. Comparison of the kinetic parameters of CfcI and CfcISYNF confirmed that this CBM18 is functionally involved in catalysis.

167

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

IntroductionChitin, a β-(1,4) linked N-acetyl-glucosamine polymer, is abundantly present as a structural component of the arthropod exoskeleton and fungal cell walls. Chitinases (EC 3.2.1.14) and β-N-acetylhexosaminidases (EC 3.2.1.52) hydrolyze chitin to chi-tin oligosaccharides (COS) or N-acetyl-glucosamine GlcNAc monomers. Several methods are available to characterize chitinase activity, employing chromogenic substrates (McCreath & Gooday, 1992), HPLC (Fukamizo et al., 2001;Ghauharali-van der Vlugt et al., 2009;Krokeide et al., 2007), mass spectrometry (Dennhart & Letzel, 2006;Price & Naumann, 2010) or capillary electrophoresis (Beaudoin et al., 2005;Blanes et al., 2008;Hattori et al., 2010). The use of chromogenic substrates is convenient but kinetic parameters determined using these substrates may differ sig-nificantly from those found with native substrates (Krokeide et al., 2007). Although advances have been made in recent years, the number of methods for analysis of native COS in low concentrations is still limited. Most methods are semi-quantita-tive (Price & Naumann, 2010), require derivatization (Beaudoin et al., 2005;Ghau-harali-van der Vlugt et al., 2009), have a relatively high detection limit (Blanes et al., 2008;Hattori et al., 2010) or require specialized equipment (Dennhart & Letzel, 2006).

High performance anion exchange chromatography (HPAEC) combined with pulsed amperometic detection (PAD) is a powerful tool for the separation, sensitive detec-tion and quantitation of a range of underivatized carbohydrates (Lee, 1990). HPAEC-PAD is widely used in carbohydrate analysis and a wide range of protocols and methods is available (Cataldi et al., 2000). Separation of COS by HPAEC-PAD has previously been reported, but neither of these methods has found wide application (Horsch et al., 1996;Horsch et al., 1997;Lee, 1990;Lee, 1996;Lu et al., 2009). Sepa-ration of COS on a CarboPac PA1 column with 18 mM sodium hydroxide allowed detection of GlcNAc and (GlcNAc)2, but failed to separate (GlcNAc)3 to (GlcNAc)6 (Lu et al., 2009). Better separation of longer COS has been described with 25 mM sodium hydroxide using a MA1 column (Horsch et al., 1996), but the separation time of 3 h per sample is highly inefficient for analysis of multiple samples. Furthermore, the alkaline pH conditions usually employed during HPAEC are detrimental for the reducing end GlcNAc (as discussed in (Lee, 1990)). In this study, a HPAEC-PAD method for the separation and sensitive detection of non-labeled COS in the range of at least (GlcNAc)1-6 is described and validated. The separation of COS was per-formed in water, preventing high pH induced epimerization of COS on the column.

168

Chapter 6

Post column addition of sodium hydroxide allowed detection and quantification by PAD. Alternatively, the use of water as low salt, neutral pH liquid phase would allow direct coupling to on line mass spectrometry equipment.

To demonstrate the suitability of the validated HPAEC-PAD method in chitinase ac-tivity assays, it was applied to the determination of the kinetic parameters of Asper-gillus niger chitinase CfcI. A. niger is a common saphrophytic fungus of significant industrial importance. The genome of this fungus contains 13 putative chitinases belonging to glycoside hydrolase family 18 (GH18) (http://www.cazy.org). One of these chitinases, CfcI, is a representative of a phylogenetic cluster conserved within Aspergilli. We recently showed that CfcI differs from previously characterized fun-gal family GH18 members in displaying a novel reaction specificity: CfcI releases GlcNAc monomers from the reducing end of COS (van Munster et al., 2012).

As is the case for other exo-chitinases, the catalytic domain of CfcI consists of a TIM-barrel with an additional α+β domain inserted between β-sheet 7 and α-helix 7. However, only in CfcI and its homologs an additional carbohydrate binding module of family 18 (CBM18) is integrated in this α+β domain. As the position of the CBM18 insertion is in close proximity to the CfcI active site, this CBM18 may be functionally involved in catalysis. CBM18 domains are present in plant lectins, anti-microbial peptides and are found attached to fungal chitinases (Aboitiz et al., 2004;Harata & Muraki, 2000;Karlsson & Stenlid, 2009;Martins et al., 1996). Their characterization has shown that these modules can bind COS. Aromatic amino acids Trp21 and Tyr23 (numbering as in the lectin hevein) stack to the bound GlcNAc residues, resulting in three binding subsites (Asensio et al., 2000;Dam & Brewer, 2003). Mutation of these aromatic amino acids to alanine in the Amaranthus caudatus antimicrobial peptide resulted in a decrease in substrate affinity (Muraki et al., 2000). Mutational analysis indicated that the hydrogen bound between Ser19 and the acetamide group of the substrate significantly contributes to stability and specificity of COS binding. Fur-thermore, the acetamide group of the substrate interacts with the aromatic ring of Tyr30 (Chávez et al., 2010). Ser19 is conserved in CfcI-CBM18 (Ser322) and aromatic residues are present at positions 21 (Tyr324) and 30 (Phe333), but an Asn residue is found at position 23 (Asn326). Replacement of the aromatic residue at position 23 with Asn, Thr or Ser, is conserved in, but not limited to, Aspergillus chitinases of the phylogenetic cluster containing CfcI. To determine whether CfcI-CBM18 is functionally involved in catalysis, we constructed a mutant enzyme in which these

169

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

4 amino acids were replaced by Gly residues (CfcISYNF). Comparison of the kinetic parameters of the CfcI wild type and mutant enzyme CfcISYNF with native COS, us-ing the newly developed HPAEC-PAD method, confirmed that CBM18 is functionally involved in catalysis.

Materials and MethodsChemicals and instruments. GlcNAc and L-Fucose were obtained from Acros Or-ganics (Geel, Belgium), and (GlcNAc)2-(GlcNAc)6 from Megazyme (Bray, Ireland). The 4-nitrophenyl labeled GlcNAc-pNP, (GlcNAc)2-pNP and (GlcNAc)3-pNP were from Sigma Aldrich (Zwijndrecht, the Netherlands). Sodium hydroxide was obtained from J.T. Baker (Deventer, the Netherlands) as 50 % (w/v) solution and sodium ac-etate was from Sigma Aldrich. HPAEC-PAD was performed on an ICS-3000 system (Dionex, Sunnyvale, CA, USA) consisting of a detector-chromatography module, dual pump module, an autosampler and a thermal compartment. The system was equipped with a 4 x 250 mm CarboPac PA-1 analytical column (Dionex) and 4 x 50 mm CarboPac PA1 guard column (Dionex). A home-made device for mixing elu-ent with post column added solutions was used, consisting of 3 meter peek tubing with an internal diameter of 0.0254 cm (0.01 inches), and internal volume of 152 µl, wound around a plastic cylinder of 16.8 mm diameter. The mixer was installed in line between the mixing tee and the electrochemical detector. The detector was equipped with an Au electrode and a combined pH-Ag/AgCl reference electrode. Reagents were kept under constant helium pressure to ensure CO2 exclusion.

HPAEC-PAD conditions. The CarboPac PA-1 column was operated at 25 °C. Solu-tion A consisted of 0.1 M sodium hydroxide, solution B of 0.1 M sodium hydroxide with 0.6 M sodium acetate, and solution C consists of ultrapure water (0.22 µm filtered, deionized water, 18.2 MΩ cm-1, Milli-Q, Millipore, Amsterdam, the Nether-lands). The column was preconditioned in 5 steps by running at a flow speed of 1 ml min-1 for 10 min in 100 % solution A (step 1), then 10 min in 100 % solution B (step 2) followed by 10 min running with 3.7 % B, 96.3 % A, (step 3) and 20 min with 3.7 % B, 21.3 % A, 75 % C (step 4). The column was then equilibrated by running in solution C for 10 min before sample injection (step 5). Separations were performed while running isocratically in solution C. Generally, 25 µl of (GlcNAc)1-6 standards or samples were injected using the autosampler equipped with a 50 µl sample loop. Standards and samples contained an internal standard (IS) of 10 µM L-fucose. Fol-lowing separation, 225 mM sodium hydroxide was delivered at 2 ml min-1 and mixed

170

Chapter 6

in the mixing coil, with a 3 s residence time. Detection of sugars was performed dur-ing 16 min after injection using a standard quadruple potential waveform (Rocklin et al., 1998). Integration was performed using a Chromeleon 6.8 chromatography data system. The concentration of chitin oligosaccharides in each sample was cal-culated using a calibration curve giving the relation between the analyte amount and the ratios of the peak areas of standards prepared with known amounts of GlcNAc oligosaccharides divided by IS peak areas.

To investigate whether epimerization reactions were likely to cause the presence of extra peaks in the chromatogram, COS were incubated for 0, 2 and 7 h in 25 mM NaOH, neutralized and desalted using a Carbograph Ultra-Clean solid phase extraction column (Grace, Deerfield, USA). Subsequently, these samples were dis-solved in ultrapure water and injected on the HPAEC-PAD.

Method validation. Specificity and selectivity was assessed through determination of chromatography retention times of 20 µM of the carbohydrates N-acetyl-glucos-amine (with degree of polymerization (DP) 1 - 6), glucosamine (DP1 - 6), N-acetyl-galactosamine, galactosamine, glucose, galactose, mannose and 4-nitrophenyl la-beled GlcNAc-pNP, (GlcNAc)2-pNP and (GlcNAc)3-pNP.

The accuracy of the method was determined by measuring known amounts of analytes under conditions resembling those of actual samples. Sample-like back-grounds (10 mM sodium acetate buffer, 1 µg ml-1 BSA) were spiked with 2 µM, 10 µM or 25 µM COS and 25 µl was injected in duplicate. Five determinations were performed under intermediate precision conditions as detailed below.

Precision was determined by measuring known amounts of analyte as repeatability (intraday) and as intermediate (interday) precision. Sample-like backgrounds were spiked with 2 µM, 10 µM or 25 µM COS. Repeatability was determined from six replicate injections for each concentration performed in one HPAEC run. Interme-diate precision was estimated from five determinations with the same equipment, performed by two individuals on five separate days using five separately prepared batches of buffer.

To assess the linearity, calibration curves were produced in duplicate as described in the section HPAEC-PAD conditions, using analyte standards (n=12, injected at

171

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

least in four-fold) in the range of 0.25 pmol to 1 nmol (corresponding to 0.01-40 µM in 25 µl injections). The linear range of the curve was assessed by the deviation from the mean of the relative response of analytes, the value of the linear correlation coefficient and the residuals.

The limit of detection (LOD), defined as the “signal level which may be a priori ex-pected to lead to detection” (Currie, 1968) was estimated from the standard devia-tion of the calibration curve as LOD = I+2t(df)s (with t=student t-value for α=0.05, df=n-2, 1-sided test, I=intercept, s=standard deviation of intercept) and converted to the concentration domain (Hubaux & Vos, 1970). The quantification limit (LOQ), defined as the smallest amount of analyte that can be quantified with a limit of er-ror of 10 % RSD, was estimated from the calibration curve as LOQ = I+10s (with I=intercept, s=standard deviation). Values were calculated as mean ± s of 2 inde-pendent sets of measurements with n≥6 each. The LOD and LOQ were experimen-tally verified by determining the value for 3s and 10s from a six fold injection of 5 pmol COS, respectively.

Enzyme production. Vector pET-15b-25 (encoding A. niger CfcI, GenBank: XM_001400452 (van Munster et al., 2012)) served as template for the production of pET-15b-69, which encodes CfcISYNF harboring mutations S322G, Y324G, N326G and F333G. Plasmid pET-15b-69 was produced in two rounds of PCR, in a protocol based on the QuikChange kit (Stratagene). Mutations S322G, Y324G and N326G were introduced with primer F64QC (5’- GGAAGTTGCTGCGGTCAGGGCGGAG-GCTGCGGTACAACCG-3’, mutations underlined) and its identical, reverse comple-ment, primer R65QC. A second PCR with primers F69QC (5’-GGTACAACCGAC-GACGGCTGCCAAGCAG-3’) and its identical, reverse complement, primer R69QC, introduced mutation F333G. PCR was performed for 20 cycles with annealing at 62 °C and elongation for 14 min at 72 °C, using Pfu DNA polymerase. Introduction of the desired mutations was confirmed by nucleotide sequencing. CfcI and CfcISYNF were expressed in E. coli after induction with 0.1 mM IPTG at 18 °C and purified us-ing the N-terminal 6xHis-tag as described before (van Munster et al., 2012).

Enzyme assays. Enzyme activity with (GlcNAc)3-pNP was determined as described (van Munster et al., 2012) to quantify specific activity. One unit of enzyme activity (U) was defined as the amount of activity needed to release 1 µmol pNP min-1 from 1 mM (GlcNAc)3-pNP at 45 °C, pH 5.0.

172

Chapter 6

Kinetic constants were determined at 45 °C in 450 μl incubations containing 10 mM sodium acetate buffer pH 5.0 and 25-600 µM (GlcNAc)3 or 5-100 µM (GlcNAc)4-5. Reactions were started by the addition of 0.11 mU CfcI or CfcISYNF protein. At 1, 2, 3 and 4 min, 100 μl sample was inactivated in 100 μl ultrapure water containing 20 µM fucose, preheated at 95 °C. Product formation was quantified by HPAEC-PAD as described above. The initial rate of GlcNAc release was calculated from at least three data points. Product formation was linear in time under assay conditions used with at maximum 15% of the available substrate consumed. At least three measure-ments were performed at six substrate concentrations. No indications of transgly-cosylation or product inhibition were detected in any of the experiments and thus the rates obtained from two (CfcI) or three (CfcISYNF) independent experiments were each fitted to the Michaelis-Menten equation using OriginPro 8.0. Resulting values are given as mean ± SE. Statistical significance of differences between CfcI and CfcISYNF were tested using the Student’s t test.

Results

Chromatography conditions. To determine kinetic parameters of chitinases with native substrates, a sensitive and fast assay is needed to separate, identify and quantify reaction products. HPAEC-PAD has proven to be an effective, sensitive method which allows separation and quantification of a broad range of carbohy-drates without the need for labeling. We have investigated whether HPAEC-PAD also can be used as assay method for the separation and quantification of COS. After testing various conditions, we found that COS of DP2-6 could be separated reasonably well within 6 min at 15 °C and 20 °C using 25 mM sodium hydroxide sup-plemented with 22.2 mM sodium acetate, but complete baseline separation was not achieved. Separation was dependent on preconditioning of the column using steps 2, 3 and 4 as described in the Methods section. At 10 °C, also deacetylated chitin oligosaccharides (GlcN oligosaccharides) of DP2-6 could be separated. However, during COS separation additional peaks were found. Alkaline conditions above pH 9 are known to induce epimerization of GlcNAc residues to ManNAc (Blayer et al., 1999;Salo et al., 1976). Incubation of COS in sodium hydroxide before analysis in-creased the area of the additional peaks from 6 % to ~ 30 % of the total peak area, indicating epimerization of COS may indeed occur under the chromatography con-ditions used. Because the peaks of COS are adjacent in the 6 min chromatogram,

173

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

the additional peaks partly overlap with the COS peaks. The described chroma-tography conditions are therefore suitable for the fast separation and detection of COS present in roughly equal amounts, but less so for quantification of COS during kinetic studies where the substrate peak is high compared to the product peak.

To improve separation and avoid interference of epimerization peaks, we optimized the above protocol using separation in ultrapure water, followed by post-column ad-dition of base. Optimized preconditioning and temperature, as detailed in the meth-ods section, allowed baseline separation of COS in the range of at least (GlcNAc)1-6. Although small additional peaks (2 - 4% of total peak area) were still observed for some COS, these did not interfere with quantification. Injection of each COS in-dividually allowed determination of typical retention times. The order of elution of COS observed here is unusual compared to other oligosaccharides (Fig. 1). Oligo-saccharides with the highest DP eluted first, followed by COS with a lower DP, cor-responding to previously described HPAEC separations of COS (Lee, 1996). The elution order, however, was dependent on column temperature. At 25 °C, GlcNAc eluted between (GlcNAc)3 and (GlcNAc)2. At lower temperatures the retention time increased and at 20 °C or 15 °C the monomer peak overlapped with (GlcNAc)3,

either partially or completely, respectively (data not shown).

0 2 4 6 8 10 12 14 16-202468

101214161820

Valu

e (n

C)

Time (min)

IS (GlcNAc)6 (GlcNAc)5

(GlcNAc)4 (GlcNAc)3

(GlcNAc)1 (GlcNAc)2

Fig. 1. Chromatogram showing separation of 250 pmol COS varying in length from GlcNAc to (GlcNAc)6 in ultrapure water under conditions described in the HPAEC-PAD conditions section.

174

Chapter 6

Method validation. To investigate the selectivity of the method, a range of mono-mers was injected. Baseline separation of COS in the range of GlcNAc to at least (GlcNAc)6 or the 4-nitrophenol labeled GlcNAc-pNP, (GlcNAc)2-pNP and (GlcNAc)3-pNP was successfully performed. N-Acetyl-galactosamine eluted at a retention time similar to N-acetyl-glucosamine. Elution of glucosamine oligosaccharides GlcN – (GlcN)6 or galactosamine (GalN) was not detected upto 30 min after injection, sug-gesting that these carbohydrates are retained on the column. Taken together, within the group of amino sugars, the assay is specific for acetylated carbohydrates. In addition, various monomers without an amino group, such as glucose, mannose and galactose, were detected.

Precision of the optimized method was determined by measuring known amounts of analytes under sample conditions. Variations in peak areas (as ratio of I.S. peak area) and retention times were determined to establish repeatability (intraday) and intermediate (interday) precision (Table 1). Under repeatability conditions, the re-tention time of all tested analytes was stable with 0.1-0.4 % RSD over 6 h and 0.5-0.6 % RSD over 10 h. The detected analyte amounts varied between 1.2-5.4 % RSD with a mean of 2.8 % RSD. Measurements for intermediate precision were

Table 1. Method precision, given as mean ratio of analyte peak area (Aa) / IS peak area (AIS) and RSD (%). Results given under repeatability are representative values from one of two performed sets of measurements.

Repeatability (n=6) Intermediate precision (n=5)

Peak Area (Aa / AIS) Retention Peak Area (Aa / AIS) Retention

50 pmol 250 pmol time (min) 50 pmol 250 pmol time (min)Analyte Mean RSD Mean RSD Mean RSD Mean RSD Mean RSD Mean RSD

(GlcNAc)1 0.19 2.9 1.20 4.3 10.4 0.5 0.24 14.6 1.37 8.6 10.4 2.7

(GlcNac)2 0.30 3.2 1.56 4.0 13.5 0.5 0.35 14.8 1.78 7.3 13.5 3.5

(GlcNAc)3 0.32 2.1 1.77 3.9 9.4 0.6 0.37 10.3 1.97 6.6 9.4 3.8

(GlcNAc)4 0.37 2.9 1.84 4.8 7.2 0.6 0.42 5.7 2.10 7.8 7.2 4.0

(GlcNAc)5 0.40 5.3 1.96 4.2 5.8 0.6 0.44 7.8 2.27 8.6 5.8 4.2

(GlcNAc)6 0.38 2.9 1.87 4.0 4.8 0.6 0.43 8.6 2.18 9.4 4.8 4.3

175

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

performed by two individuals on five separate days using five separately prepared batches of buffer. The variation in detected analyte amounts was with a mean of 3.3 % and range of 1.2-10.3 % RSD higher than found under repeatability conditions. The accuracy of the method, determined under intermediate precision conditions, is satisfactory with recovery rates of 88.2-98.7 % (Table 2).

The linear range for (GlcNac)1-4 was 5-500 pmol (corresponding to 25 µl injections with 0.2-20 µM COS). For (GlcNAc)5 the range was 5-250 pmol and for (GlcNAc)6

2.5-250 pmol. Coefficients of determination (R2) were at least 0.9994. The limits of detection (LOD) (Table 3), defined as the “signal level which may be a priori ex-

Recovery (%)Analyte 50 pmol 250 pmol 625 pmol

(GlcNAc)1 88.2 ± 2.6 98.7 ± 0.9 96.1 ± 0.9

(GlcNac)2 94.0 ± 4.3 95.4 ± 1.3 93.0 ± 1.7

(GlcNAc)3 90.7 ± 1.9 95.5 ± 0.8 91.6 ± 1.2

(GlcNAc)4 97.7 ± 2.0 96.9 ± 0.7 92.2 ± 1.6

(GlcNAc)5 93.2 ± 0.6 97.2 ± 0.9 90.2 ± 1.6

(GlcNAc)6 94.0 ± 0.5 96.5 ± 1.1 88.4 ± 1.8

Table 2. Method accuracy given as mean recovery (n=6) ± SE

Table 3. Method detection limit (LOD) and limit of quantification (LOQ) estimated by calcula-tion (cal) or experimentally verified (exp).

Analyte LODcal (pmol) LODexp (pmol) LOQcal (pmol) LOQexp (pmol)

(GlcNAc)1 3.0 ± 0.3 0.8 7.5 ± 0.8 2.7

(GlcNac)2 3.1 ± 1.3 1.0 7.8 ± 3.0 3.3

(GlcNAc)3 0.9 ± 1.0 0.7 2.3 ± 2.3 2.2

(GlcNAc)4 2.4 ± 1.8 0.6 5.8 ± 4.2 2.2

(GlcNAc)5 1.5 ± 0.9 0.5 3.4 ± 2.1 1.7

(GlcNAc)6 1.1 ± 0.2 1.0 2.5 ± 0.6 3.3

176

Chapter 6

pected to lead to detection” (Currie, 1968) were in the range of 1-4 pmol (0.04-0.16 µM). The quantification limits (LOQ) (Table 3), defined as the smallest amount of analyte that can be quantified with a limit of error of 10 % RSD, were calculated to be in the range of 1-10 pmol (0.02-0.4 µM). Experimental verification gave values for LOD and LOQ of the same order of magnitude. Determination of kinetic parameters of CfcI and CfcISYNF. The suitability of the method was evaluated by determining kinetic parameters (Km and kcat) of A. ni-ger CfcI chitinase. CfcI was produced in E. coli with an N-terminal 6xHis-tag, and purified to near homogeneity. CfcI was obtained with a specific activity of 5.6 ± 0.6 U mg-1, determined with (GlcNAc)3-pNP, with a yield of 2.1 ± 0.3 mg protein per liter culture (mean ± SE, n=2). This is comparable to the 4-9 U mg-1 and 2.1 mg protein per liter culture reported previously for CfcI-MBP (van Munster et al., 2012). Kinetic parameters of CfcI were determined with three substrates using the validated HPAEC-PAD method. The results (Table 4) indicate that the catalytic ef-ficiency kcat/Km of CfcI increased with the chain length of the substrate. This increase was caused mainly by a higher affinity for the substrate, as the Km decreased with substrate chain length but kcat remained stable at 12-16 s-1.

To investigate whether the CBM18 in CfcI has a function in catalysis, we modified key amino acid residues by mutagenesis. From sequence alignments it appeared

Table 4. Kinetic parameters of the Aspergillus niger CfcI and CfcISYNF wild type and mutant enzymes, as determined by HPAEC-PAD. Values are given as mean ± SE of independent experiments.

CfcI CfcISYNF

Km

(µM)

kcat

(s-1)

kcat/Km

(µM-1 s-1)

Km

(µM)

kcat

(s-1)

kcat/Km

(µM-1 s-1)

(GlcNAc)3 49.5 ± 6.9 15.9 ± 0.9 0.32 ± 0.05 n.d.a n.d. n.d.

(GlcNAc)4 19.3 ± 2.4 12.4 ± 0.5 0.65 ± 0.09 34.5 ± 2.0 b 20.0 ± 2.7 0.52 ± 0.09

(GlcNAc)5 15.0 ± 2.0 15.1 ± 0.5 1.0 ± 0.14 23.3 ± 2.2 b 19.6 ± 2.2 0.84 ± 0.12

a n.d.; not determined b statistically significant differences between CfcI and CfcISYNF

177

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

that Ser322, Tyr324, Asn326 and Phe333 in CfcI CBM18 correspond to amino acids iden-tified as important in substrate interaction in characterized CBM18 members, such as Ser19, Trp21, Trp23 and Tyr30 in the lectin hevein. These four amino acids in CfcI CBM18 were replaced by glycine residues, resulting in CfcISYNF. CfcISYNF protein was produced as described in Methods with a yield of 1.1 ± 0.03 mg protein per liter cul-ture, showing a specific activity of 5.8 ± 0.2 U mg-1 with (GlcNAc)3-pNP, similar to the activity of wild type CfcI. Separation and identification of CfcISYNF reaction products showed a product profile similar to that of the wild type CfcI, with GlcNAc as the main reaction product. Comparison of the kinetic parameters of CfcISYNF and CfcI (Table 4) revealed that the affinity of CfcISYNF for substrates (GlcNAc)4 and (GlcNAc)5 is lower, since these Km values had significantly (p=0.01 and p=0.03, respectively) increased with approximately 150 %. Although not significant, the kcat values of CfcI-

SYNF for substrates (GlcNAc)4 and (GlcNAc)5 appeared to increase as well, resulting in a catalytic efficiency kcat/Km of CfcISYNF similar to that of CfcI. At relatively low sub-strate concentrations the activity of CfcISYNF thus is markedly lower than observed for the wild type enzyme. We conclude that CBM18 has a functional role in catalysis by the wild type CfcI protein, increasing affinity of binding at low substrate concen-trations.

DiscussionSpecific and sensitive separation, identification and quantification methods for COS are essential to assess their functions and potential applications in both glycobiol-ogy and enzymology. The HPAEC-PAD method described here provides such a sensitive, relatively fast, quantitative assay for these acetylated carbohydrates. The range of 2.5 pmol to at least 250 pmol in which quantitative detection is possible, and the complete baseline separation of a series of (GlcNAc)1-6 in 20 min per sam-ple, outperforms both previously described and commonly used HPLC methods for underivatized COS (Fukamizo et al., 2001;Krokeide et al., 2007) as well as HPAEC-PAD methods for COS separation (Horsch et al., 1996;Horsch et al., 1997;Lu et al., 2009). The described HPAEC-PAD method is suitable for the analysis of clean mixtures of GlcNAc oligosaccharides, which are for example used during enzyme assays. The method allows selective detection and quantification of acetylated COS in mixtures of N-acetyl-glucosamine and glucosamine monomers or oligosaccha-rides. The observed separation of COS when using water as mobile phase depends on preconditioning of the column. It must be noted that the assay is less suitable for the analysis of samples with a matrix containing large amounts of salt or buffers:

178

Chapter 6

injection of acetate or sodium hydroxide is detrimental to the repeatability of reten-tion times and peak areas. Furthermore, as the system is sensitive to carbonate

accumulation on the column, it is essential to prepare CO2 free eluent. This is con-sistent with the fact that after measuring an extended number of samples (70-140 in our system), removal of the carbonate accumulated on the column, by rinsing with solution B, improved method precision. We demonstrated the suitability of the method by identifying differences in the ki-netic properties of A. niger CfcI wild type and mutant enzymes acting on native substrates. Determination of kinetic parameters of CfcI revealed Km values in the range of 15-50 µM and kcat values in the range of 12-16 s-1. These kinetic param-eters of CfcI are in the same range as those of other fungal chitinases. Reports on kinetic constants of several fungal chitinases, obtained with labeled COS, show that Km values are in the lower µM to mM range (Boer et al., 2007;de La Cruz et al., 1992;Rao et al., 2005). For CfcI, the Km values found here on native substrates are significantly lower compared to the Km of 0.2 mM and 2.0 mM we reported for the substrates (GlcNAc)3-pNP and (GlcNAc)2-pNP, respectively (van Munster et al., 2012). Evidence has been reported previously that enzyme kinetic data obtained with labeled substrates may differ significantly from that obtained with native sub-strates (Krokeide et al., 2007). For example, for Serratia marcescens ChiB a kcat of 57 s-1 and Km of 7 µM were reported using (GlcNac)3-4MU but with the native (GlcNac)4 a kcat of 28 s-1 and Km of 4 µM were found (Krokeide et al., 2007;Synstad et al., 2008).

Our data shows that CfcI-CBM18 is important for substrate binding in the CfcI en-zyme: mutagenesis of key amino acids in this CBM18 caused a decrease in sub-strate affinity. Rather surprisingly, and not understood at present, CfcISYNF appears to have a higher turnover number kcat without a functional CBM18. Nevertheless, the overall catalytic efficiency kcat/Km of the two enzymes remained similar. In conclu-sion, comparison of the kinetic parameters of the CfcI wild type and mutant enzyme CfcISYNF with native COS, using the newly developed HPAEC-PAD method, showed that CBM18 is functionally involved in catalysis at low substrate concentrations.

AcknowledgementsWe acknowledge SenterNovem for funding of IOP Genomics project IGE07008 and we thank Prof. J.P. Kamerling for helpful discussions.

179

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

ReferencesAboitiz, N., Vila-Perelló, M., Groves, P., Asensio, J. L., Andreu, D., Cañada, F. J. & Jimé-

nez-Barbero, J. (2004).NMR and modeling studies of protein-carbohydrate interac-tions: synthesis, three-dimensional structure, and recognition properties of a mini-mum hevein domain with binding affinity for chitooligosaccharides. Chembiochem 5, 1245-1255.

Asensio, J. L., Cañada, F. J., Siebert, H. C., Laynez, J., Poveda, A., Nieto, P. M., Soedja-naamadja, U. M., Gabius, H. J. & Jiménez-Barbero, J. (2000).Structural basis for chitin recognition by defense proteins: GlcNAc residues are bound in a multivalent fashion by extended binding sites in hevein domains. Chem Biol 7, 529-543.

Beaudoin, M. E., Gauthier, J., Boucher, I. & Waldron, K. C. (2005).Capillary electrophore-sis separation of a mixture of chitin and chitosan oligosaccharides derivatized using a modified fluorophore conjugation procedure. J Sep Sci 28, 1390-1398.

Blanes, L., Saito, R. M., Genta, F. A., Donega, J., Terra, W. R., Ferreira, C. & do Lago, C. L. (2008).Direct detection of underivatized chitooligosaccharides produced through chitinase action using capillary zone electrophoresis. Anal Biochem 373, 99-103.

Blayer, S., Woodley, J. M., Dawson, M. J. & Lilly, M. D. (1999).Alkaline biocatalysis for the direct synthesis of N-acetyl-D-neuraminic acid (Neu5Ac) from N-acetyl-D-glucos-amine (GlcNAc). Biotechnol Bioeng 66, 131-136.

Boer, H., Simolin, H., Cottaz, S., Söderlund, H. & Koivula, A. (2007).Heterologous expres-sion and site-directed mutagenesis studies of two Trichoderma harzianum chitin-ases, Chit33 and Chit42, in Escherichia coli. Protein Expr Purif 51, 216-226.

Cataldi, T. R., Campa, C. & De Benedetto, G. E. (2000).Carbohydrate analysis by high-performance anion-exchange chromatography with pulsed amperometric detection: the potential is still growing. Fresenius J Anal Chem 368, 739-758.

Chávez, M. I., Vila-Perelló, M., Cañada, F. J., Andreu, D. & Jiménez-Barbero, J. (2010).Effect of a serine-to-aspartate replacement on the recognition of chitin oligosac-charides by truncated hevein. A 3D view by using NMR. Carbohydr Res 345, 1461-1468.

Currie, L. A. (1968).Limits for qualitative detection and quantitative determination. Applica-tion to radiochemistry. Anal Chem 40, 586-593.

Dam, T. K. & Brewer, C. F. (2003).Carbohydrate-lectin cross-linking interactions: structural, thermodynamic, and biological studies. Methods Enzymol 362:455-86., 455-486.

de La Cruz, J., Hidalgo-Gallego, A., Lora, J. M., Benitez, T., Pintor-Toro, J. A. & Llobell, A. (1992).Isolation and characterization of three chitinases from Trichoderma har-zianum. Eur J Biochem 206, 859-867.

Dennhart, N. & Letzel, T. (2006).Mass spectrometric real-time monitoring of enzymatic gly-cosidic hydrolysis, enzymatic inhibition and enzyme complexes. Anal Bioanal Chem 386, 689-698.

Fukamizo, T., Sasaki, C., Schelp, E., Bortone, K. & Robertus, J. D. (2001).Kinetic proper-ties of chitinase-1 from the fungal pathogen Coccidioides immitis. Biochemistry 40, 2448-2454.

Ghauharali-van der Vlugt, K., Bussink, A. P., Groener, J. E., Boot, R. G. & Aerts, J. M. (2009).Detection of chitinase activity by 2-aminobenzoic acid labeling of chito-oligosaccharides. Anal Biochem 384, 191-193.

180

Chapter 6

Harata, K. & Muraki, M. (2000).Crystal structures of Urtica dioica agglutinin and its complex with tri-N-acetylchitotriose. J Mol Biol 297, 673-681.

Hattori, T., Anraku, N. & Kato, R. (2010).Capillary electrophoresis of chitooligosaccharides in acidic solution: simple determination using a quaternary-ammonium-modified col-umn and indirect photometric detection with crystal violet. J Chromatogr B Analyt Technol Biomed Life Sci 878, 477-480.

Horsch, M., Mayer, C. & Rast, D. M. (1996).Stereochemical requirements of chitin synthase for ligand binding at the allosteric site for N-acetylglucosamine. Eur J Biochem 237, 476-482.

Horsch, M., Mayer, C., Sennhauser, U. & Rast, D. M. (1997).Beta-N-acetylhexosaminidase: a target for the design of antifungal agents. Pharmacol Ther 76, 187-218.

Hubaux, A. & Vos, G. (1970).Decision and detection limits for calibration curves. Anal Chem 42, 849-855.

Karlsson, M. & Stenlid, J. (2009).Evolution of family 18 glycoside hydrolases: diversity, domain structures and phylogenetic relationships. J Mol Microbiol Biotechnol 16, 208-232.

Krokeide, I. M., Synstad, B., Gåseidnes, S., Horn, S. J., Eijsink, V. G. & Sorlie, M. (2007).Natural substrate assay for chitinases using high-performance liquid chromatogra-phy: a comparison with existing assays. Anal Biochem 363, 128-134.

Lee, Y. C. (1996).Carbohydrate analyses with high-performance anion-exchange chromatog-raphy. J Chromatogr A 720, 137-149.

Lee, Y. C. (1990).High-performance anion-exchange chromatography for carbohydrate anal-ysis. Anal Biochem 189, 151-162.

Lu, Y., Yang, H., Hu, H., Wang, Y., Rao, Z. & Jin, C. (2009).Mutation of Trp137 to glutamate completely removes transglycosyl activity associated with the Aspergillus fumigatus AfChiB1. Glycoconj J 26, 525-534.

Martins, J. C., Maes, D., Loris, R., Pepermans, H. A., Wyns, L., Willem, R. & Verheyden, P. (1996).H NMR study of the solution structure of Ac-AMP2, a sugar binding antimi-crobial protein isolated from Amaranthus caudatus. J Mol Biol 258, 322-333.

McCreath, K. J. & Gooday, G. W. (1992).A rapid and sensitive microassay for determination of chitinolytic activity. J Microbiol Methods 14, 229-237.

Muraki, M., Morii, H. & Harata, K. (2000).Chemically prepared hevein domains: effect of C-terminal truncation and the mutagenesis of aromatic residues on the affinity for chitin. Protein Eng 13, 385-389.

Price, N. P. & Naumann, T. A. (2010).A high-throughput MALDI-TOF mass spectrometry-based assay of chitinase activity. Anal Biochem 411, 94-99.

Rao, F. V., Houston, D. R., Boot, R. G., Aerts, J. M., Hodkinson, M., Adams, D. J., Shiomi, K., Omura, S. & van Aalten, D. M. (2005).Specificity and affinity of natural product cyclopentapeptide inhibitors against A. fumigatus, human, and bacterial chitinases. Chem Biol 12, 65-76.

Rocklin, R. D., Clarke, A. P. & Weitzhandler, M. (1998).Improved long-term reproducibility for Pulsed Amperometric Detection of carbohydrates via a new quadruple-potential waveform. Anal Chem 70, 1496-1501.

Salo, W. L., Hamari, M. & Hallcher, L. (1976).Preparation of [2-2H]-2-acetamido-2-deoxy-D-glucose by epimerization of 2-acetamido-2-deoxy-D-mannose in basic deuteri-

181

HPAEC-PAD method for chitin oligosaccharides

Cha

pter

6

um oxide; and proposal of a unifying type of mechanism for the 2-epimerization of 2-acetamido-2-deoxyhexoses. Carbohydr Res 50, 287-291.

Synstad, B., Vaaje-Kolstad, G., Cederkvist, F. H., Saua, S. F., Horn, S. J., Eijsink, V. G. & Sorlie, M. (2008).Expression and characterization of endochitinase C from Serratia marcescens BJL200 and its purification by a one-step general chitinase purification method. Biosci Biotechnol Biochem 72, 715-723.

van Munster, J. M., van der Kaaij, R. M., Dijkhuizen, L. & van der Maarel, M. J. E. C. (2012).Biochemical characterization of Aspergillus niger CfcI, a glycoside hydrolase family 18 chitinase that releases monomers during substrate hydrolysis. Microbiol-ogy 158, 2168-2179.

182