VECTOR-BORNE AND ZOONOTIC DISEASESVolume 9, Number 1, 2009© Mary Ann Liebert, Inc.DOI: 10.1089/vbz.2008.0010

Short Communication

Detection of a Novel Francisella in Dermacentor reticulatus:A Need for Careful Evaluation of PCR-Based Identification

of Francisella tularensis in Eurasian Ticks

Zsuzsa Sréter-Lancz,1,2 Zoltán Széll,3 Tamás Sréter,3 and Károly Márialigeti2

Abstract

Francisella tularensis, the causative agent of tularemia, has been detected in ixodid ticks in some regions of NorthAmerica, Europe, and Asia. In the present study, 245 Dermacentor reticulatus, 211 Ixodes ricinus, and 194 Haema-physalis concinna adults from Hungary were tested for the presence of F. tularensis by polymerase chain reac-tion (PCR) assays based on 16S ribosomal RNA (16S rDNA) and T-cell epitope of a Francisella membrane pro-tein (TUL4). No Francisella-specific amplification products were detected in I. ricinus and H. concinna ticks.Francisella DNA was identified using PCR assays based on 16S rDNA and TUL4 gene in D. reticulatus with sim-ilar prevalence (minimum 1.2%) as demonstrated in earlier European and Asian studies detecting F. tularensisin D. reticulatus. However, the 16S rDNA and TUL4 gene sequences of the Francisella-like agent occurring in D.reticulatus differed from the homologous sequences of Francisella spp. deposited in GenBank. Phylogenetic re-constructions showed that the new genotype detected in D. reticulatus was closely related to Francisella-like en-dosymbionts of North American Dermacentor ticks. Although further studies are needed on the relationship ofthis bacterium with ticks, the results highlight the need for careful evaluation of PCR-based identification inEuropean and Asian laboratories that screen ixodid ticks for F. tularensis.

Key words: Diagnostics—Tick(s)—Vector-borne—Zoonosis—Francisella.

123

Introduction

SEVERAL “NEW” BACTERIAL and protozoal tick-borne patho-gens have been described recently in Europe. The most

important zoonotic tick-borne infections include tick-borneencephalitis, Lyme borreliosis, anaplasmosis, spotted fevergroup rickettsioses, bartonelloses, tularemia, and babesioseson the continent. Tularemia caused by Francisella tularensisis a severe zoonotic disease transmitted to humans partly byvectors such as ticks, flies, and mosquitoes (Ellis et al. 2002).In the northern hemisphere, four subspecies of F. tularensishave been described: F. tularensis subsp. tularensis, F. tu-larensis subsp. holarctica, F. tularensis subsp. novicida, and F.tularensis subsp. mediasiatica (Petersen and Schriefer 2005). F. tularensis has been detected in ixodid ticks in some regionsof North America, Europe, and Asia (Ellis et al. 2002). Nev-ertheless, Francisella-like endosymbionts have also beenidentified in North American ixodid tick species (Niebyliskiet al. 1997, Sun et al. 2000, Scoles 2004, Kugeler et al. 2005).Herein we report a novel Francisella from a Eurasian ixodidtick species, Dermacentor reticulatus, which is closely related

to Francisella-like endosymbionts of North American Derma-centor ticks.

Materials and Methods

Overall, 245 D. reticulatus, 211 Ixodes ricinus, and 194Haemaphysalis concinna unengorged or partly engorgedadults were removed from carcasses of 263 red foxes (Vulpesvulpes). Foxes had been killed in 13 counties representing60% of the territory of Hungary. The origins, transportation,and storage of the fox carcasses, and the methods used tocollect and identify ticks found on them have been describedpreviously (Sréter et al. 2003). The ticks were stored as sep-arate pools (each of five or fewer ticks) from each fox at�20°C. The DNA was extracted from each pool (Sréter et al.2000), and fragments of genes coding for 16S ribosomal RNA(16S rDNA) and T-cell epitope of a Francisella membrane pro-tein (TUL4) were amplified using the primer pairs F11/F5,TUL4-435/TUL4-863, and FT393/FT642 (Long et al. 1993,Sjöstedt et al. 1997, Kugeler et al. 2005). To exclude the pos-sibility of contamination with specific DNA, a negative con-

1Laboratories for Microbiology, Food and Feed Safety Directorate and 3Laboratories for Parasitology, Fish, Bee and Wildlife Diseases,Veterinary Diagnostic Directorate, Central Agricultural Office, Budapest, Hungary.

2Department of Microbiology, Eötvös Loránd University, Budapest, Hungary.

trol was included and underwent the entire procedure start-ing with DNA extraction. A field isolate of F. tularensis wasused as positive control. PCRs were performed in an iCycler(BioRad, Hercules, CA) as described (Long et al. 1993, Sjöst-edt et al. 1997, Kugeler et al. 2005). All amplicons were fur-ther characterized by sequence analysis (Sréter et al. 2000)using the primers described earlier (Long et al. 1993, Kugeleret al. 2005). The chromatograms were visually inspected forreading errors and combined using Chromas 2.0 (Technely-sium, Tewantin, Australia) and MultAlin (Corpet 1988) pro-grams. The 16S rDNA and TUL4 gene sequences of Fran-cisella detected in D. reticulatus were identical in all poolsamples. The sequences were submitted to GenBank andhave the following accession numbers: EU234535 andEU126640. Sequences were identified by comparison withGenBank entries using the BLAST program (http://www.ncbi.nlm.nih.gov/blast). Multiple sequence alignment, esti-mation of evolutionary divergence, and phylogenetic analy-ses were conducted using the MEGA4 program (Tamura etal. 2007). Briefly, multiple sequence alignments were basedon the ClustalW algorithm, estimates of evolutionary diver-gence were calculated using the maximum composite likeli-hood method, and the evolutionary history was inferred us-ing the minimum evolution method. The following 16SrDNA and TUL4 gene sequences were included in the phy-logenetic analyses (accession numbers of sequences are inparentheses): F. tularensis subsp. tularensis (AY968225,

AM286280), F. tularensis subsp. holarctica (AY968229,AM233362), F. tularensis subsp. novicida (AY968237,CP000439), F. tularensis subsp. mediasiatica (AY968235,AM261162), F. philomiragia subsp. philomiragia (AY928394,EF153475), F. philomiragia subsp. noatunensis (DQ295795,DQ813267), Francisella-like endosymbionts of Dermacentor albipictus (AY375394, AY375408), Dermacentor andersoni(AY375424, AY375412), Dermacentor nitens (AY375400,AY375418), Dermacentor occidentalis (AY805304, AY375419),and Dermacentor variabilis (AY795980, AY375420). The 16SrDNA sequences of some Francisella strains detected in soilsamples (AY968285, AY968291, AY968296, AY968302,AY968303) (Barns et al. 2005) were also included in the phy-logenetic analyses.

Results and Discussion

F. tularensis was detected in previous studies using differ-ent biological, bacteriological, serological, and molecularmethods in 0.5%–3.5% of D. reticulatus, 0.1%–0.3% of I. rici-nus, and 0.0%–2.9% of H. concinna collected in European andAsian regions of Russia and some European countries(Hubálek et al. 1998, Gurycová et al. 2001, Astashina et al.2003). In the present study, no Francisella-specific amplifica-tion products were detected in I. ricinus and H. concinna pools.Knowing the low prevalence of F. tularensis in European ixo-did tick species, the negative result might be attributed to the

SRÉTER-LANCZ ET AL.124

FIG. 1. Phylogenetic relationships of Francisella subspecies, Francisella-like bacteria from environmental samples, Fran-cisella-like endosymbionts of North American Dermacentor ticks, and Francisella of D. reticulatus inferred from the minimumevolution analysis of the 1.1-kb fragment of the 16S ribosomal RNA gene. Numbers above and below nodes represent boot-strap values (%) based on 1000 replications. The scale bar indicates evolutionary distances as the number of substitutionsper nucleotide.

relatively low number of ticks included in the present study.Moreover, I. ricinus, H. concinna, and D. reticulatus ticks werecollected from foxes shot in different regions of the countryand did not come from enzootic foci as ticks included in otherstudies (Hubálek et al. 1998, Gurycová et al. 2001, Astashinaet al. 2003). Of the 102 pools of D. reticulatus investigated,Francisella DNA was detected in 3 pools from 3 different re-gions of Hungary by PCR based on 16S rDNA and TUL4gene. As there were 3 positive pools, there must have beenat least 3 infected ticks among the 245 investigated, giving aminimum prevalence of 1.2% for Francisella in D. reticulatus.However, the 16S rDNA and TUL4 gene sequences of theFrancisella-like agent differed from the homologous se-quences of all known Francisella sequences deposited in Gen-Bank and were most closely related to Francisella-like en-dosymbionts of North American ticks (identity: 98%–99% and94%–96%). The estimates of evolutionary divergence between16S rDNA sequences of Francisella detected in D. reticulatusand those of pathogenic F. tularensis subspecies and fran-cisellae occurring in environmental samples were 0.006–0.033, which were higher than those observed between 16SrDNA sequences of Francisella of D. reticulatus and Francisella-like endosymbionts of North American Dermacentor spp.(0.005–0.006). The 16S rDNA-based phylogenetic reconstruc-tion showed that the new Francisella genotype from D. re-ticulatus was relatively closely related to North AmericanDermacentor tick endosymbionts, which together form amonophyletic clade (Fig. 1). The estimates of evolutionary di-vergence between the TUL4 gene sequences of Franciselladetected in D. reticulatus and those of pathogenic F. tularen-sis subspecies were 4.667–10.837, which were considerablyhigher than those seen between the TUL4 gene sequences ofFrancisella of D. reticulatus and Francisella-like endosymbiontsof North American Dermacentor spp. (0.011–0.023). The TUL4gene-based phylogenetic reconstruction showed that Fran-cisella of D. reticulatus clustered together with North Ameri-can tick endosymbionts; nevertheless, as the amplicon wasshort (248 bp), bootstrap values were low (data not shown).As attempts to amplify a longer fragment of TUL4 gene us-ing the primer pair TUL4-435/TUL4-863 (Sjöstedt et al. 1997)failed, the phylogenetic position based on TUL4 gene se-quences could not be further analyzed.

Although francisellae are notorious for being encounteredin the environment (Barns et al. 2005), the identical sequencesof Francisella in D. reticulatus pools from different regions ofHungary, the estimates of evolutionary divergence, and theresults of the phylogenetic analyses indicate that Franciselladetected in D. reticulatus is not an external contamination ofthe outer surface of ticks with Francisella-like organisms oc-curring in soil, and the bacterium is most likely the en-dosymbiont of D. reticulatus. Nevertheless, as no attemptswere made to find the organisms in internal tissues of theticks (e.g., in the ovaries by microscopy or cultivation), fur-ther studies are needed in this field. The phylogenetic re-constructions seem to confirm that Francisella-like endosym-bionts adapted to symbiosis after transmissible ancestorsspread to a wide variety of tick species (Scoles 2004). As sali-vary glands of D. andersoni have not been found to be in-fected with Francisella-like endosymbionts, and guinea pigsfed on by Francisella-like endosymbiont infected D. andersonidid not develop clinical signs or immune response (Niebyl-ski et al. 1997, Sun et al. 2000), Francisella-like endosymbionts

are most likely nonpathogenic. Nevertheless, further studiesare needed in this field (Telford and Goethert 2004). Sinceno attempts were made to detect Francisella in foxes fromwhich the ticks were collected, it cannot be completely ex-cluded that the organism detected in D. reticulatus is patho-genic and originated in the blood meal acquired from thehost species. As Francisella-like bacteria in ticks and proba-bly some soil francisellae contaminating the outer surface ofticks give positive results in some PCR assays as seen in thepresent study, there is a need for careful evaluation of PCR-based identification of F. tularensis in European and Asianlaboratories that screen ixodid ticks for this pathogen.

Acknowledgments

The authors thank Lajos Tekes and Vilmos Pálfi for sup-porting the studies, András Zágon and József Szikraszer fortheir invaluable help in tick collection, and László Makrai forproviding the positive control strain. A Bolyai János Schol-arship from the Hungarian Academy of Science (BO/432/05,to T.S.) provided financial support.

References

Astashina, LI, Mironchuk, IV, Ivanov, DI. Comparative study ofIxodes ticks on their infection with tularemia pathogen usingbiologic and serological assays [in Russian]. Med Parazitol(Moscow) 2003; 3:27–30.

Barns, SM, Grow, CC, Okinaka, RT, Keim, P, et al. Detection ofdiverse new Francisella-like bacteria in environmental sam-ples. Appl Environ Microbiol 2005; 71:5494–5500.

Corpet, F. Multiple sequence alignment with hierarchical clus-tering. Nucleic Acids Res 1988; 16:10881–10890.

Ellis, J, Oyston, PC, Green, M, Titball, RW. Tularemia. Clin Mi-crobiol Rev 2002; 15:631–646.

Gurycová, D, Vyrosteková, V, Khanakah, G, Kocianová, E, et al.Importance of surveillance of tularemia natural foci in theknown endemic area of Central Europe, 1991–1997. Wien KlinWochenschr 2001; 113:433–438.

Hubálek, Z, Sixl, W, Halouzka, J. Francisella tularensis in Derma-centor reticulatus ticks from the Czech Republic and Austria.Wien Klin Wochenschr 1998; 110:909–910.

Kugeler, KJ, Gurfield, N, Creek, JG, Mahoney, KS, et al. Dis-crimination between Francisella tularensis and Francisella-likeendosymbionts when screening ticks by PCR. Appl EnvironMicrobiol 2005; 71:7594–7597.

Long, GW, Oprandy, JJ, Narayanan, RB, Fortier, AH, et al. De-tection of Francisella tularensis in blood by polymerase chainreaction. J Clin Microbiol 1993; 31:152–154.

Niebylski, ML, Peacock, MG, Fischer, ER, Porcella, SF, et al.Characterization of an endosymbiont infecting wood ticks,Dermacentor andersoni, as a member of the genus Francisella.Appl Environ Microbiol 1997; 63:3933–3940.

Petersen, JM, Schriefer, ME. Tularemia: emergence/re-emer-gence. Vet Res 2005; 36:455–467.

Scoles, GA. Phylogenetic analysis of the Francisella-like en-dosymbionts of Dermacentor ticks. J Med Entomol 2004;41:277–286.

Sjöstedt, A, Eriksson, U, Berglund, L, Tärnvik, A. Detection ofFrancisella tularensis in ulcers of patients with tularemia byPCR. J Clin Microbiol 1997; 35:1045–1048.

Sréter, T, Kovács, G, da Silva, AJ, Pieniazek, NJ, et al. Morpho-logic, host specificity, and molecular characterization of aHungarian Cryptosporidium meleagridis isolate. Appl EnvironMicrobiol 2000; 66:735–738.

A NOVEL FRANCISELLA IN TICKS 125

Sréter, T, Széll, Z, Varga, I. Ectoparasite infestations of red foxes(Vulpes vulpes) in Hungary. Vet Parasitol 2003; 115:349–354.

Sun, LV, Scoles, GA, Fish, D, O’Neill, SL. Francisella-like en-dosymbionts of ticks. J Invertebr Pathol 2000; 76:301–303.

Tamura, K, Dudley, J, Nei, M, Kumar S. MEGA4: Molecular Evo-lutionary Genetic Analysis (MEGA) software version 4.0. MolBiol Evol 2007; 24:1596–1599.

Telford, SR, III, Goethert, HK. Emerging tick-borne infections:rediscovered and better characterized, or truly ‘new’? Para-sitology 2004; 129:S301–S327.

Address reprint requests to:Tamás Sréter

Laboratories for Parasites, Fish, Bee and Wildlife DiseasesVeterinary Diagnostic Directorate

Central Agricultural OfficeTábornok u. 2

H-1149 BudapestHungary

E-mail: sretert@oai.hu

SRÉTER-LANCZ ET AL.126