Infection, Genetics and Evolution 31 (2015) 231–235
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Genetic diversity and variation over time of Coxiella burnetii genotypes in dairy cattle and the farm environment Alvaro Piñero, Jesús F. Barandika, Ana L. García-Pérez, Ana Hurtado ⇑ NEIKER – Instituto Vasco de Investigación y Desarrollo Agrario, Department of Animal Health, Berreaga 1, 48160 Derio, Bizkaia, Spain
a r t i c l e
i n f o
Article history: Received 15 December 2014 Received in revised form 2 February 2015 Accepted 5 February 2015 Available online 13 February 2015 Keywords: Coxiella burnetii Dairy cattle Multiple-Locus Variable number tandem repeats Analysis (MLVA) Genotyping Bulk-tank milk (BTM) Surface dust
a b s t r a c t The genetic diversity of Coxiella burnetii from 36 dairy cattle herds was determined by Multiple-Locus Variable number tandem repeats Analysis (MLVA), and genotypes from different sources (bulk-tank milk – BTM and surface dust) and sampling time (2009/10 and 2011/12) were compared. A total of 15 different genotypes were identified from 60 BTM and seven dust samples, including seven genotypes reported here for the first time (BN, BO, BP, BQ, BR, BS, BT). The two most prevalent genotypes (J and I), detected both in BTM and dust, accounted for 44.5% of the C. burnetii typed and have been reported infecting cattle worldwide. In 52% of herds more than one genotype was found, and mixed infection with two genotypes was observed in seven BTM samples. Comparison of C. burnetii genotypes at different samplings within each herd detected a change in genotype in 32% of herds, while a persistent genotype was identified in the remaining 68%. In addition, the genotype obtained from dust samples was always identical to that present in the BTM sample. Often persistent genotypes were among the most prevalent types. Clustering of the MLVA genotypes from this and other studies using the minimum spanning tree method separated our C. burnetii strains into two clusters, 10 genotypes clustered within genomic group (GG) III, and the remaining five types (AE, BQ, BR, BS and BT) grouped with GG II, which includes strains implicated in human outbreaks. Although presence in cattle of genotypes closely related to those identified in humans does not seem to be common event, it cannot be neglected and surveillance of genotype distribution is needed to fully understand the epidemiology of Q fever. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Q fever is a zoonotic disease whose etiological agent is Coxiella burnetii, a pathogen described in species throughout the animal kingdom (Babudieri, 1959; Kersh et al., 2012). Domestic ruminants are considered the main source of infection for humans (EFSA, 2010). Infected animals shed C. burnetii through birth or abortion products, as well as other routes, such as milk, vaginal fluids or feces. The inhalation of aerosols contaminated with the bacteria is the main route of infection for susceptible individuals. In recent years, an increased incidence of human infection in Europe has been reported (Georgiev et al., 2013), mainly due to the 4173 cases associated to the Q fever outbreak occurred in the Netherlands between 2007 and 2012 (Dijkstra et al., 2012) (http://www.rivm. nl/Onderwerpen/Q/Q_koorts).
⇑ Corresponding author. Tel.: +34 944034300; fax: +34 944034310. E-mail addresses: [email protected] (A. Piñero), [email protected] (J.F. Barandika), [email protected] (A.L. García-Pérez), [email protected] (A. Hurtado). http://dx.doi.org/10.1016/j.meegid.2015.02.006 1567-1348/Ó 2015 Elsevier B.V. All rights reserved.
Molecular characterization of C. burnetii is a useful tool for epidemiological investigation of Q fever outbreaks. Genotyping of C. burnetii strains obtained from domestic ruminants and the environment can be used to identify the source of human infection (Klaassen et al., 2009; Roest et al., 2011). The genetic diversity among C. burnetii isolates has been analyzed using different typing methods (Arricau-Bouvery et al., 2006; Glazunova et al., 2005; Hendrix et al., 1991; Hornstra et al., 2011), which sometimes hinders comparison of results. Among those typing methods that can be used directly on clinical samples without prior cultivation, Multiple-Locus Variable number tandem repeats Analysis (MLVA) provides the highest discriminatory power (Tilburg, 2013). Although interlaboratory comparison of MLVA results can be difficult, it is less laborious (Koh et al., 2012; Svraka et al., 2006) than other techniques like multispacer sequence typing (MST), and agreement between techniques has been demonstrated (Astobiza et al., 2012b; Santos et al., 2012; Sulyok et al., 2014). To facilitate comparison of MLVA results, a C. burnetii cooperative database is being developed by aggregating the genotyping data deduced by in silico analysis from published whole genome sequence data and the MLVA information provided by various
232
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
authors (http://mlva.u-psud.fr/MLVAnet/spip.php?rubrique50). Besides, a thorough analysis of results obtained by different typing techniques allowed the classification of C. burnetii strains into six genomic groups (GGs), each GG being defined by the presence of a specific plasmid, by a variant of the adaA gene and by the presence of certain MST and MLVA genotypes (Hendrix et al., 1991; Hornstra et al., 2011). Q fever is endemic in the Basque Country (northern Spain) and C. burnetii is widespread in cattle in the region (Astobiza et al., 2012a). The aim of this study was to identify the C. burnetii MLVA genotypes that infect dairy cattle herds in the Basque Country, and to evaluate the variation of genotypes over time (a two year period). Finally, C. burnetii genotypes found in the animals and the farm environment were compared and their relationship with genotypes described across the world was calculated. 2. Materials and methods 2.1. Sampling procedure C. burnetii included in this study were collected in the frame of a previous study carried out to evaluate the progression of C. burnetii infection in dairy cattle in an endemic region in northern Spain. In that study, 94 dairy cattle herds were sampled as fully detailed elsewhere (Piñero et al., 2014b). Briefly, presence of C. burnetii DNA was assessed by real-time PCR analysis of bulk-tank milk (BTM) samples collected twice (2009/10 and 2011/12) per herd, and dust samples collected in 2011/12 from five different surfaces within each farm premises. The PCR was performed using primers and probe targeting the IS1111 transposon-like repetitive region of C. burnetii as previously described (Schets et al., 2013). Herds that presented with C. burnetii real-time PCR positive BTM in the two sampling periods (N = 36) were selected. In addition, in nine of these herds, C. burnetii was also detected in 14 dust samples collected from the environment. Thus, a total of 86 C. burnetii positive samples from cattle (72 BTM) and the environment (14 dust) were included in this study.
that used the same MLVA typing scheme (Astobiza et al., 2012b; Santos et al., 2012; Tilburg, 2013). To determine the genetic similarity among the genotypes identified, the minimum spanning tree method was used. Genotypes were classified in predicted GGs based on Hendrix et al. (1991) and Hornstra et al. (2011) according to data described by other authors (Hornstra et al., 2011; Tilburg, 2013). The discriminatory ability of the MLVA method in general and of each marker individually was calculated using Simpson’s index of diversity (Hunter and Gaston, 1988), and the confidence intervals of Simpson’s index as proposed by Grundmann et al. (2001). 2.3. Other analyses Data regarding size of the herds and management of heifers (bred inside or outside the herd) were compiled in both samplings to investigate their association with changes in genotypes within the herds. Presence of small ruminants in the farm was also recorded to look for associations with genomic groups. Chi-square and Fisher tests were used to assess for any statistical significance in changes of frequencies. 3. Results A complete MLVA genotype was obtained for 83.3% (60/72) of the BTM samples and half of the dust samples (7/14) analyzed Table 1 C. burnetii genotypes according to type of sample and sampling date. ID herd
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
2.2. Multiple-Locus Variable number tandem repeats Analysis (MLVA) Two multicolor multiplex PCR assays were carried out targeting six microsatellite markers containing either six or seven base pairs (bp) repeat units: three hexanucleotide repeat markers (Ms27, Ms28 and Ms34) and three heptanucleotide repeat markers (Ms23, Ms24 and Ms33). Primer sequences were reported before (Klaassen et al., 2009; Tilburg et al., 2012c) and PCR conditions were detailed elsewhere (Astobiza et al., 2012b). PCR products (1 lL) were mixed with 11 lL of Hi-Di formamide (Applied Biosystems) and 0.5 lL of GeneScan 600 LIZ Size Standard (Applied Biosystems). After denaturation for 3 min at 96 °C, the samples were cooled on ice. The PCR products were separated on a 3130 Genetic Analyzer (Applied Biosystems) with a 36-cm array using POP7 polymer. The fragments were sized using GeneMapper version 4.0 software (Applied Biosystems). DNA from the Nine Mile strain (RSA 493) was used as reference. The number of repeats in each marker was determined by extrapolation using the sizes of the obtained fragments relative to those obtained for the Nine Mile strain. According to in silico analysis, the genotype of the Nine Mile strain is 9-27-4-6-9-5 for markers Ms23-Ms24-Ms27-Ms28-Ms33Ms34, respectively. This includes a correction to Ms33 of Nine Mile strain, until recently coded 4 units and now confirmed to be 9 repeat units (http://mlva.u-psud.fr/MLVAnet/spip.php?rubrique50). Data obtained from MLVA typing were imported into Bionumerics v7.1 (Applied Maths, Sint-Martens-Latem, Belgium) together with the C. burnetii genotypes reported in other studies
a
BTMa (2009/10)
J M BI AB nt nt M nt AB I BI/I BN J I/J BP BS nt nt BI J AB/BG nt nt M BS I AB BG BI N BT I J I BO I
BTM (2011/12)
J M J AB/BI J M I/J AE AB I nt J J I AB BS BQ nt BI J AB J M nt BR BI AB AB/BG nt I BR I J I nt AB/I
Environment (2011/12) PCRb
MLVA
Pos (1) Neg Neg Neg Neg Pos (1) Neg Neg Neg Neg Neg Pos (1) Neg Neg Neg Neg Neg Pos (2) Neg Neg Neg Pos (2) Neg Neg Neg Pos (2) Neg Neg Neg Neg Neg Neg Neg Pos (1) Pos (3) Pos (1)
ntc – – – – nt – – – – – nt – – – – – J/nt – – – J/J – – – BI/BI – – – – – – – I nt/nt/nt I
BTM, bulk-tank milk. Pos, positive, in brackets, number of dust samples positive to C. burnetii out the five samples tested; Neg, negative. c nt, non-typable. b
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
(Table 1). Non-typable samples were associated with a high realtime PCR Ct value (generally above 32) and therefore a lower concentration of C. burnetii. This was particularly common for dust samples. Most samples yielded a single genotype. However, in seven of the 60 BTM samples two different alleles were observed for marker Ms34, while a single allele was detected in the remaining five markers. This would indicate that two different C. burnetii strains were contaminating those seven BTM samples (Table 1). A total of 15 different genotypes were identified (Table 2), including seven new genotypes reported here for the first time (BN, BO, BP, BQ, BR, BS, BT). All 15 genotypes were found in BTM samples, with types J and I being the most prevalent. These two genotypes were also detected in dust samples (genotype J in three samples and genotype I in two), along with genotype BI, which was identified in a further two dust samples. Thus, genotypes J and I accounted for 44.5% of the C. burnetii typed. On the other end were the seven genotypes which were only identified once (Table 2). The Simpson’s index of diversity calculated for each marker ranged from 0.586 (95% CI: 0.557–0.615) for Ms27, the less discriminatory, to 0.860 (95% CI: 0.844–0.876) for Ms34, which provided the highest discrimination (Table 2). The complete typing scheme with the six markers had a final Simpson’s Index of Diversity of 0.933 (95% CI: 0.919–0.947). In 25 of the 36 herds included in the study, MLVA genotyping was successful in both BTM samples (2009/10 and 2011/12). A single genotype was found in 48% (12/25) of the herds, whereas in the remaining 52% (13/25) of herds more than one genotype was found (Table 1), the latter including six herds with BTM samples contaminated with two different genotypes. In one herd, a total of three genotypes were found (Herd 7, Table 1). Comparison of genotypes in both samples within each herd identified a persistent genotype in 68% (17/25) of them, whereas a change in genotype was observed in the remaining 32% (8/25) of herds (Table 1). In addition, when the dust surface samples were typable, the genotype obtained was identical to that present in the BTM sample collected that same year (2011/12) (Table 1). Often the persistent genotype was one of the three most prevalent genotypes (I, J or AB) which differed in only one of the six markers (Ms34). Clustering of MLVA genotypes using the minimum spanning tree method showed a high diversity between the strains (Fig. 1). The genotypes observed in this study (indicated with a golden star in Fig. 1) were located in two separate clusters, genotypes AE, BQ, BR, BS and BT grouped in GG II, and the remaining 10 types clustered within GG III. Genotypes from each cluster differed in at least five of the six markers, whereas genotypes within each cluster shared between three and five markers (Table 2; Fig. 1). Table 2 Description and frequency (%) of genotypes identified.
a
MLVA
Ms23
Ms24
Ms27
Ms28
Ms33
Ms34
Nb (%)
J I AB BI M BG BSa BRa AE N BQa BOa BTa BPa BNa
6 6 6 6 6 6 4 4 4 5 4 6 4 6 5
13 13 13 13 13 13 8 8 9 13 9 13 8 13 13
2 2 2 2 2 2 3 3 3 2 3 2 3 2 2
7 7 7 7 7 7 3 3 3 7 3 8 3 13 7
9 9 9 9 9 9 9 10 8 9 9 9 8 9 9
10 9 12 13 11 8 5 5 4 9 4 11 5 13 5
17 (22.9) 16 (21.6) 11 (14.8) 9 (12.2) 6 (8.1) 3 (4.1) 3 (4.1) 2 (2.7) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4)
Genotypes described in this study for the first time. A total of 74 genotype results were obtained, 67 from BTM (seven of 60 BTM samples included two different genotypes) and seven from dust samples. b
233
No significant associations were observed between changes in genotypes within the herd during two years and decrease/increase in the herd size, or type of heifers’ management (inside/outside the farm). Similarly no associations were observed between coexistence of small ruminants in the herd and C. burnetii genogroups (GG II/GG III).
4. Discussion In this study, C. burnetii positive samples collected from dairy cattle herds in northern Spain were genotyped by MLVA and types from different sources (BTM and dust) and sampling time (2009/10 and 2011/12) were compared. Fifteen different MLVA genotypes were identified in 36 dairy cattle herds, indicating a high genetic diversity of C. burnetii strains in this region. In a preliminary genotyping study, our group had already used MLVA to type a selection of C. burnetii positive samples from domestic ruminants, including 20 collected from 12 dairy cattle farms from different parts of Spain (Astobiza et al., 2012b). Seven genotypes were found in cattle herds in that study, including four of the genotypes identified here (J, I, AB and M). In fact, genotypes J and I had been reported in cattle milk products from other countries (Sulyok et al., 2014; Tilburg et al., 2012b; Tilburg, 2013), indicating the existence of a common pool of C. burnetii strains that infect cattle worldwide. On the other hand, seven of the 15 genotypes found here had not been described before, although certainly these novel genotypes were only sporadically found, each representing 1.4–4.1% of the detections (13.5% in total). Other studies that carried out MLVA typing for the first time in a particular region, also reported the description of novel genotypes (Frangoulidis et al., 2014; Santos et al., 2012; Sulyok et al., 2014). In any case, more MLVA data are needed to obtain a more comprehensive image of the complete population structure of C. burnetii. Comparison of C. burnetii genotypes at different samplings within each herd suggested that the same strain was able to persist for a long time in the herd. Often persistent genotypes were among the three most prevalent types (I, J or AB), which is in agreement with results from other studies carried out in dairy cattle herds (Piñero et al., 2014a) and sheep flocks (Astobiza et al., 2012b). In those cases when different genotypes were found between samplings (32% of the herds), types generally differed in only one marker. These closely related genotypes could represent microvariants within the C. burnetii population that infect cattle. These small variations were only detected by the marker that according to the Simpson’s index of diversity had the highest discriminatory power (Ms34). This same marker also allowed the detection of mixed infections in seven BTM samples. Presence of several genotypes in BTM samples was not unexpected and has been reported before (de Bruin et al., 2012; Tilburg et al., 2012b). In fact, PCR might have underestimated the real number of BTM samples contaminated with more than one genotype since preferential amplification of the most prevalent genotype in the mixture would be likely. The MLVA typing scheme using six markers showed a high discriminatory power (Simpson’s Index of Diversity of 0.933) and demonstrated its ability to estimate the genotypic diversity of C. burnetii infecting cattle and the farm environment. However, as previously reported (Roest et al., 2011; Tilburg et al., 2012c), typeability is compromised in samples with low C. burnetii burden (Ct > 31). Clustering of the MLVA genotypes of this and other studies using the minimum spanning tree method separated our C. burnetii strains into two clusters. Most of the strains included in this study clustered within GG III, which mainly includes C. burnetii from cattle. However, five genotypes identified in eight BTM samples from
234
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
Fig. 1. Minimum spanning tree of MLVA genotypes. Genotypes identified in this study (indicated with a golden star), along with others reported elsewhere (Astobiza et al., 2012b; Santos et al., 2012; Sulyok et al., 2014; Tilburg et al., 2012b, 2012c; Tilburg, 2013) and five sequenced reference strains of C. burnetii: CbuG_Q212 (GenBank accession number CP001019), CbuG_Q154 (CP001020), Dugway (CP000733), Nine Mile RSA493 (AE016828) and Henzerling RSA331 (CP000890), whose MLVA pattern were determined in silico using the published sequences (Tilburg et al., 2012c), are included. Each circle corresponds to a genotype and its size refers to the number of samples in which that type has been identified. Each color represents the species where the genotype was detected. Branch labels and connecting lines correspond to the number of different markers between the MLVA genotypes. Predicted genomic groups (GGs) based on (Hendrix et al., 1991; Hornstra et al., 2011) are highlighted with gray color and genotypes are classified in different GGs according to the information described by other authors (Hornstra et al., 2011; Tilburg, 2013). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
five herds grouped within GG II. Four of these five genotypes were described here for the first time. Strains within GG II have been associated with sheep and goats, and have been implicated in human outbreaks (Glazunova et al., 2005; Hornstra et al., 2011). In fact, the strain responsible for the human Q fever outbreak in the Netherlands (strain G, MST 33) belongs to GG II (see Fig. 1), and was linked to goats (Tilburg et al., 2012a). These results show that genotypes closely related to those identified in humans can also infect cattle, but still, this does not seem to be common event (Astobiza et al., 2012b; Santos et al., 2012; Tilburg et al., 2012b). In conclusion, a high genotypic diversity was observed among C. burnetii from epidemiologically unrelated dairy cattle herds in northern Spain. Yet, some predominant genotypes (J, I) with worldwide distribution were also identified. Genotype comparison demonstrated that only 11% of the overall detections corresponded
to genotypes closely related to those identified in humans, suggesting that dairy cattle play a limited role in human Q fever infection. Surveillance of genetic distribution of C. burnetii from different sources is needed to fully understand the epidemiology of Q fever. Acknowledgements This study was supported by Spanish National Institute for Agricultural and Food Research and Technology (INIA, RTA 201300051-C2-00) and the European Regional Development Fund (ERDF). We are grateful to Mónica Villoria (NEIKER, Spain) and Dr. Jeroen Tilburg (Canisius-Wilhelmina Hospital, The Netherlands) for technical support and helpful suggestions. AP is the recipient of a predoctoral fellowship from INIA. We acknowledge the interest and collaboration of the farmers.
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
References Arricau-Bouvery, N., Hauck, Y., Bejaoui, A., Frangoulidis, D., Bodier, C.C., Souriau, A., Meyer, H., Neubauer, H., Rodolakis, A., Vergnaud, G., 2006. Molecular characterization of Coxiella burnetii isolates by infrequent restriction site-PCR and MLVA typing. BMC Microbiol. 6, 38. Astobiza, I., Ruiz-Fons, F., Piñero, A., Barandika, J.F., Hurtado, A., García-Pérez, A.L., 2012a. Estimation of Coxiella burnetii prevalence in dairy cattle in intensive systems by serological and molecular analyses of bulk-tank milk samples. J. Dairy Sci. 95, 1632–1638. Astobiza, I., Tilburg, J.J., Piñero, A., Hurtado, A., García-Pérez, A.L., Nabuurs-Franssen, M.H., Klaassen, C.H., 2012b. Genotyping of Coxiella burnetii from domestic ruminants in northern Spain. BMC Vet. Res. 8, 241. Babudieri, B., 1959. Q fever: a zoonosis. Adv. Vet. Sci. 5, 81–182. de Bruin, A., van Alphen, P.T., van der Plaats, R.Q., de Heer, L., Reusken, C.B., van Rotterdam, B.J., Janse, I., 2012. Molecular typing of Coxiella burnetii from animal and environmental matrices during Q fever epidemics in the Netherlands. BMC Vet. Res. 8, 165. Dijkstra, F., van der Hoek, W., Wijers, N., Schimmer, B., Rietveld, A., Wijkmans, C.J., Vellema, P., Schneeberger, P.M., 2012. The 2007–2010 Q fever epidemic in The Netherlands: characteristics of notified acute Q fever patients and the association with dairy goat farming. FEMS Immunol. Med. Microbiol. 64, 3–12. EFSA, 2010. Scientific opinion on Q fever. EFSA J. 8, 1595. Frangoulidis, D., Walter, M.C., Antwerpen, M., Zimmermann, P., Janowetz, B., Alex, M., Bottcher, J., Henning, K., Hilbert, A., Ganter, M., Runge, M., Munsterkotter, M., Splettstoesser, W.D., Hanczaruk, M., 2014. Molecular analysis of Coxiella burnetii in Germany reveals evolution of unique clonal clusters. Int. J. Med. Microbiol. 304, 868–876. Georgiev, M., Afonso, A., Neubauer, H., Needham, H., Thiery, R., Rodolakis, A., Roest, H., Stark, K., Stegeman, J., Vellema, P., van der Hoek, W., More, S., 2013. Q fever in humans and farm animals in four European countries, 1982 to 2010. Euro Surveill. 18, pii=20407. Glazunova, O., Roux, V., Freylikman, O., Sekeyova, Z., Fournous, G., Tyczka, J., Tokarevich, N., Kovacava, E., Marrie, T.J., Raoult, D., 2005. Coxiella burnetii genotyping. Emerg. Infect. Dis. 11, 1211–1217. Grundmann, H., Hori, S., Tanner, G., 2001. Determining confidence intervals when measuring genetic diversity and the discriminatory abilities of typing methods for microorganisms. J. Clin. Microbiol. 39, 4190–4192. Hendrix, L.R., Samuel, J.E., Mallavia, L.P., 1991. Differentiation of Coxiella burnetii isolates by analysis of restriction-endonuclease-digested DNA separated by SDS-PAGE. J. Gen. Microbiol. 137, 269–276. Hornstra, H.M., Priestley, R.A., Georgia, S.M., Kachur, S., Birdsell, D.N., Hilsabeck, R., Gates, L.T., Samuel, J.E., Heinzen, R.A., Kersh, G.J., Keim, P., Massung, R.F., Pearson, T., 2011. Rapid typing of Coxiella burnetii. PLoS One 6, e26201. Hunter, P.R., Gaston, M.A., 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson’s index of diversity. J. Clin. Microbiol. 26, 2465–2466. Kersh, G.J., Lambourn, D.M., Raverty, S.A., Fitzpatrick, K.A., Self, J.S., Akmajian, A.M., Jeffries, S.J., Huggins, J., Drew, C.P., Zaki, S.R., Massung, R.F., 2012. Coxiella
235
burnetii infection of marine mammals in the Pacific Northwest, 1997–2010. J. Wildl. Dis. 48, 201–206. Klaassen, C.H., Nabuurs-Franssen, M.H., Tilburg, J.J., Hamans, M.A., Horrevorts, A.M., 2009. Multigenotype Q fever outbreak, The Netherlands. Emerg. Infect. Dis. 15, 613–614. Koh, X.P., Chiou, C.S., Ajam, N., Watanabe, H., Ahmad, N., Thong, K.L., 2012. Characterization of Shigella sonnei in Malaysia, an increasingly prevalent etiologic agent of local shigellosis cases. BMC Infect. Dis. 12, 122. Piñero, A., Barandika, J.F., Hurtado, A., García-Pérez, A.L., 2014a. Progression of Coxiella burnetii infection after implementing a two-year vaccination program in a naturally infected dairy cattle herd. Acta Vet. Scand. 56, 47. Piñero, A., Ruiz-Fons, F., Hurtado, A., Barandika, J.F., Atxaerandio, R., García-Pérez, A.L., 2014b. Changes in the dynamics of Coxiella burnetii infection in dairy cattle: an approach to match field data with the epidemiological cycle of C. burnetii in endemic herds. J. Dairy Sci. 97, 2718–2730. Roest, H.I., Ruuls, R.C., Tilburg, J.J., Nabuurs-Franssen, M.H., Klaassen, C.H., Vellema, P., van den Brom, R., Dercksen, D., Wouda, W., Spierenburg, M.A., van der Spek, A.N., Buijs, R., de Boer, A.G., Willemsen, P.T., van Zijderveld, F.G., 2011. Molecular epidemiology of Coxiella burnetii from ruminants in Q fever outbreak, the Netherlands. Emerg. Infect. Dis. 17, 668–675. Santos, A.S., Tilburg, J.J., Botelho, A., Barahona, M.J., Nuncio, M.S., Nabuurs-Franssen, M.H., Klaassen, C.H., 2012. Genotypic diversity of clinical Coxiella burnetii isolates from Portugal based on MST and MLVA typing. Int. J. Med. Microbiol. 302, 253–256. Schets, F.M., de Heer, L., de Roda Husman, A.M., 2013. Coxiella burnetii in sewage water at sewage water treatment plants in a Q fever epidemic area. Int. J. Hyg. Environ. Health 216, 698–702. Sulyok, K.M., Kreizinger, Z., Hornstra, H.M., Pearson, T., Szigeti, A., Dan, A., Balla, E., Keim, P.S., Gyuranecz, M., 2014. Genotyping of Coxiella burnetii from domestic ruminants and human in Hungary: indication of various genotypes. BMC Vet. Res. 10, 107. Svraka, S., Toman, R., Skultety, L., Slaba, K., Homan, W.L., 2006. Establishment of a genotyping scheme for Coxiella burnetii. FEMS Microbiol. Lett. 254, 268–274. Tilburg, J.J., 2013. Molecular Investigation of the Q Fever Epidemic in the Netherlands. The Largest Outbreak Caused by Coxiella burnetii Ever Reported. PhD Thesis, Radboud Universiteit Nijmegen, Boxpress, Nijmegen, The Netherlands. . Tilburg, J.J., Roest, H.J., Buffet, S., Nabuurs-Franssen, M.H., Horrevorts, A.M., Raoult, D., Klaassen, C.H., 2012a. Epidemic genotype of Coxiella burnetii among goats, sheep, and humans in the Netherlands. Emerg. Infect. Dis. 18, 887–889. Tilburg, J.J., Roest, H.J., Nabuurs-Franssen, M.H., Horrevorts, A.M., Klaassen, C.H., 2012b. Genotyping reveals the presence of a predominant genotype of Coxiella burnetii in consumer milk products. J. Clin. Microbiol. 50, 2156–2158. Tilburg, J.J., Rossen, J.W., van Hannen, E.J., Melchers, W.J., Hermans, M.H., van de Bovenkamp, J., Roest, H.J., de Bruin, A., Nabuurs-Franssen, M.H., Horrevorts, A.M., Klaassen, C.H., 2012c. Genotypic diversity of Coxiella burnetii in the 2007– 2010 Q fever outbreak episodes in The Netherlands. J. Clin. Microbiol. 50, 1076– 1078.
Contents lists available at ScienceDirect
Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid
Genetic diversity and variation over time of Coxiella burnetii genotypes in dairy cattle and the farm environment Alvaro Piñero, Jesús F. Barandika, Ana L. García-Pérez, Ana Hurtado ⇑ NEIKER – Instituto Vasco de Investigación y Desarrollo Agrario, Department of Animal Health, Berreaga 1, 48160 Derio, Bizkaia, Spain
a r t i c l e
i n f o
Article history: Received 15 December 2014 Received in revised form 2 February 2015 Accepted 5 February 2015 Available online 13 February 2015 Keywords: Coxiella burnetii Dairy cattle Multiple-Locus Variable number tandem repeats Analysis (MLVA) Genotyping Bulk-tank milk (BTM) Surface dust
a b s t r a c t The genetic diversity of Coxiella burnetii from 36 dairy cattle herds was determined by Multiple-Locus Variable number tandem repeats Analysis (MLVA), and genotypes from different sources (bulk-tank milk – BTM and surface dust) and sampling time (2009/10 and 2011/12) were compared. A total of 15 different genotypes were identified from 60 BTM and seven dust samples, including seven genotypes reported here for the first time (BN, BO, BP, BQ, BR, BS, BT). The two most prevalent genotypes (J and I), detected both in BTM and dust, accounted for 44.5% of the C. burnetii typed and have been reported infecting cattle worldwide. In 52% of herds more than one genotype was found, and mixed infection with two genotypes was observed in seven BTM samples. Comparison of C. burnetii genotypes at different samplings within each herd detected a change in genotype in 32% of herds, while a persistent genotype was identified in the remaining 68%. In addition, the genotype obtained from dust samples was always identical to that present in the BTM sample. Often persistent genotypes were among the most prevalent types. Clustering of the MLVA genotypes from this and other studies using the minimum spanning tree method separated our C. burnetii strains into two clusters, 10 genotypes clustered within genomic group (GG) III, and the remaining five types (AE, BQ, BR, BS and BT) grouped with GG II, which includes strains implicated in human outbreaks. Although presence in cattle of genotypes closely related to those identified in humans does not seem to be common event, it cannot be neglected and surveillance of genotype distribution is needed to fully understand the epidemiology of Q fever. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Q fever is a zoonotic disease whose etiological agent is Coxiella burnetii, a pathogen described in species throughout the animal kingdom (Babudieri, 1959; Kersh et al., 2012). Domestic ruminants are considered the main source of infection for humans (EFSA, 2010). Infected animals shed C. burnetii through birth or abortion products, as well as other routes, such as milk, vaginal fluids or feces. The inhalation of aerosols contaminated with the bacteria is the main route of infection for susceptible individuals. In recent years, an increased incidence of human infection in Europe has been reported (Georgiev et al., 2013), mainly due to the 4173 cases associated to the Q fever outbreak occurred in the Netherlands between 2007 and 2012 (Dijkstra et al., 2012) (http://www.rivm. nl/Onderwerpen/Q/Q_koorts).
⇑ Corresponding author. Tel.: +34 944034300; fax: +34 944034310. E-mail addresses: [email protected] (A. Piñero), [email protected] (J.F. Barandika), [email protected] (A.L. García-Pérez), [email protected] (A. Hurtado). http://dx.doi.org/10.1016/j.meegid.2015.02.006 1567-1348/Ó 2015 Elsevier B.V. All rights reserved.
Molecular characterization of C. burnetii is a useful tool for epidemiological investigation of Q fever outbreaks. Genotyping of C. burnetii strains obtained from domestic ruminants and the environment can be used to identify the source of human infection (Klaassen et al., 2009; Roest et al., 2011). The genetic diversity among C. burnetii isolates has been analyzed using different typing methods (Arricau-Bouvery et al., 2006; Glazunova et al., 2005; Hendrix et al., 1991; Hornstra et al., 2011), which sometimes hinders comparison of results. Among those typing methods that can be used directly on clinical samples without prior cultivation, Multiple-Locus Variable number tandem repeats Analysis (MLVA) provides the highest discriminatory power (Tilburg, 2013). Although interlaboratory comparison of MLVA results can be difficult, it is less laborious (Koh et al., 2012; Svraka et al., 2006) than other techniques like multispacer sequence typing (MST), and agreement between techniques has been demonstrated (Astobiza et al., 2012b; Santos et al., 2012; Sulyok et al., 2014). To facilitate comparison of MLVA results, a C. burnetii cooperative database is being developed by aggregating the genotyping data deduced by in silico analysis from published whole genome sequence data and the MLVA information provided by various
232
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
authors (http://mlva.u-psud.fr/MLVAnet/spip.php?rubrique50). Besides, a thorough analysis of results obtained by different typing techniques allowed the classification of C. burnetii strains into six genomic groups (GGs), each GG being defined by the presence of a specific plasmid, by a variant of the adaA gene and by the presence of certain MST and MLVA genotypes (Hendrix et al., 1991; Hornstra et al., 2011). Q fever is endemic in the Basque Country (northern Spain) and C. burnetii is widespread in cattle in the region (Astobiza et al., 2012a). The aim of this study was to identify the C. burnetii MLVA genotypes that infect dairy cattle herds in the Basque Country, and to evaluate the variation of genotypes over time (a two year period). Finally, C. burnetii genotypes found in the animals and the farm environment were compared and their relationship with genotypes described across the world was calculated. 2. Materials and methods 2.1. Sampling procedure C. burnetii included in this study were collected in the frame of a previous study carried out to evaluate the progression of C. burnetii infection in dairy cattle in an endemic region in northern Spain. In that study, 94 dairy cattle herds were sampled as fully detailed elsewhere (Piñero et al., 2014b). Briefly, presence of C. burnetii DNA was assessed by real-time PCR analysis of bulk-tank milk (BTM) samples collected twice (2009/10 and 2011/12) per herd, and dust samples collected in 2011/12 from five different surfaces within each farm premises. The PCR was performed using primers and probe targeting the IS1111 transposon-like repetitive region of C. burnetii as previously described (Schets et al., 2013). Herds that presented with C. burnetii real-time PCR positive BTM in the two sampling periods (N = 36) were selected. In addition, in nine of these herds, C. burnetii was also detected in 14 dust samples collected from the environment. Thus, a total of 86 C. burnetii positive samples from cattle (72 BTM) and the environment (14 dust) were included in this study.
that used the same MLVA typing scheme (Astobiza et al., 2012b; Santos et al., 2012; Tilburg, 2013). To determine the genetic similarity among the genotypes identified, the minimum spanning tree method was used. Genotypes were classified in predicted GGs based on Hendrix et al. (1991) and Hornstra et al. (2011) according to data described by other authors (Hornstra et al., 2011; Tilburg, 2013). The discriminatory ability of the MLVA method in general and of each marker individually was calculated using Simpson’s index of diversity (Hunter and Gaston, 1988), and the confidence intervals of Simpson’s index as proposed by Grundmann et al. (2001). 2.3. Other analyses Data regarding size of the herds and management of heifers (bred inside or outside the herd) were compiled in both samplings to investigate their association with changes in genotypes within the herds. Presence of small ruminants in the farm was also recorded to look for associations with genomic groups. Chi-square and Fisher tests were used to assess for any statistical significance in changes of frequencies. 3. Results A complete MLVA genotype was obtained for 83.3% (60/72) of the BTM samples and half of the dust samples (7/14) analyzed Table 1 C. burnetii genotypes according to type of sample and sampling date. ID herd
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
2.2. Multiple-Locus Variable number tandem repeats Analysis (MLVA) Two multicolor multiplex PCR assays were carried out targeting six microsatellite markers containing either six or seven base pairs (bp) repeat units: three hexanucleotide repeat markers (Ms27, Ms28 and Ms34) and three heptanucleotide repeat markers (Ms23, Ms24 and Ms33). Primer sequences were reported before (Klaassen et al., 2009; Tilburg et al., 2012c) and PCR conditions were detailed elsewhere (Astobiza et al., 2012b). PCR products (1 lL) were mixed with 11 lL of Hi-Di formamide (Applied Biosystems) and 0.5 lL of GeneScan 600 LIZ Size Standard (Applied Biosystems). After denaturation for 3 min at 96 °C, the samples were cooled on ice. The PCR products were separated on a 3130 Genetic Analyzer (Applied Biosystems) with a 36-cm array using POP7 polymer. The fragments were sized using GeneMapper version 4.0 software (Applied Biosystems). DNA from the Nine Mile strain (RSA 493) was used as reference. The number of repeats in each marker was determined by extrapolation using the sizes of the obtained fragments relative to those obtained for the Nine Mile strain. According to in silico analysis, the genotype of the Nine Mile strain is 9-27-4-6-9-5 for markers Ms23-Ms24-Ms27-Ms28-Ms33Ms34, respectively. This includes a correction to Ms33 of Nine Mile strain, until recently coded 4 units and now confirmed to be 9 repeat units (http://mlva.u-psud.fr/MLVAnet/spip.php?rubrique50). Data obtained from MLVA typing were imported into Bionumerics v7.1 (Applied Maths, Sint-Martens-Latem, Belgium) together with the C. burnetii genotypes reported in other studies
a
BTMa (2009/10)
J M BI AB nt nt M nt AB I BI/I BN J I/J BP BS nt nt BI J AB/BG nt nt M BS I AB BG BI N BT I J I BO I
BTM (2011/12)
J M J AB/BI J M I/J AE AB I nt J J I AB BS BQ nt BI J AB J M nt BR BI AB AB/BG nt I BR I J I nt AB/I
Environment (2011/12) PCRb
MLVA
Pos (1) Neg Neg Neg Neg Pos (1) Neg Neg Neg Neg Neg Pos (1) Neg Neg Neg Neg Neg Pos (2) Neg Neg Neg Pos (2) Neg Neg Neg Pos (2) Neg Neg Neg Neg Neg Neg Neg Pos (1) Pos (3) Pos (1)
ntc – – – – nt – – – – – nt – – – – – J/nt – – – J/J – – – BI/BI – – – – – – – I nt/nt/nt I
BTM, bulk-tank milk. Pos, positive, in brackets, number of dust samples positive to C. burnetii out the five samples tested; Neg, negative. c nt, non-typable. b
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
(Table 1). Non-typable samples were associated with a high realtime PCR Ct value (generally above 32) and therefore a lower concentration of C. burnetii. This was particularly common for dust samples. Most samples yielded a single genotype. However, in seven of the 60 BTM samples two different alleles were observed for marker Ms34, while a single allele was detected in the remaining five markers. This would indicate that two different C. burnetii strains were contaminating those seven BTM samples (Table 1). A total of 15 different genotypes were identified (Table 2), including seven new genotypes reported here for the first time (BN, BO, BP, BQ, BR, BS, BT). All 15 genotypes were found in BTM samples, with types J and I being the most prevalent. These two genotypes were also detected in dust samples (genotype J in three samples and genotype I in two), along with genotype BI, which was identified in a further two dust samples. Thus, genotypes J and I accounted for 44.5% of the C. burnetii typed. On the other end were the seven genotypes which were only identified once (Table 2). The Simpson’s index of diversity calculated for each marker ranged from 0.586 (95% CI: 0.557–0.615) for Ms27, the less discriminatory, to 0.860 (95% CI: 0.844–0.876) for Ms34, which provided the highest discrimination (Table 2). The complete typing scheme with the six markers had a final Simpson’s Index of Diversity of 0.933 (95% CI: 0.919–0.947). In 25 of the 36 herds included in the study, MLVA genotyping was successful in both BTM samples (2009/10 and 2011/12). A single genotype was found in 48% (12/25) of the herds, whereas in the remaining 52% (13/25) of herds more than one genotype was found (Table 1), the latter including six herds with BTM samples contaminated with two different genotypes. In one herd, a total of three genotypes were found (Herd 7, Table 1). Comparison of genotypes in both samples within each herd identified a persistent genotype in 68% (17/25) of them, whereas a change in genotype was observed in the remaining 32% (8/25) of herds (Table 1). In addition, when the dust surface samples were typable, the genotype obtained was identical to that present in the BTM sample collected that same year (2011/12) (Table 1). Often the persistent genotype was one of the three most prevalent genotypes (I, J or AB) which differed in only one of the six markers (Ms34). Clustering of MLVA genotypes using the minimum spanning tree method showed a high diversity between the strains (Fig. 1). The genotypes observed in this study (indicated with a golden star in Fig. 1) were located in two separate clusters, genotypes AE, BQ, BR, BS and BT grouped in GG II, and the remaining 10 types clustered within GG III. Genotypes from each cluster differed in at least five of the six markers, whereas genotypes within each cluster shared between three and five markers (Table 2; Fig. 1). Table 2 Description and frequency (%) of genotypes identified.
a
MLVA
Ms23
Ms24
Ms27
Ms28
Ms33
Ms34
Nb (%)
J I AB BI M BG BSa BRa AE N BQa BOa BTa BPa BNa
6 6 6 6 6 6 4 4 4 5 4 6 4 6 5
13 13 13 13 13 13 8 8 9 13 9 13 8 13 13
2 2 2 2 2 2 3 3 3 2 3 2 3 2 2
7 7 7 7 7 7 3 3 3 7 3 8 3 13 7
9 9 9 9 9 9 9 10 8 9 9 9 8 9 9
10 9 12 13 11 8 5 5 4 9 4 11 5 13 5
17 (22.9) 16 (21.6) 11 (14.8) 9 (12.2) 6 (8.1) 3 (4.1) 3 (4.1) 2 (2.7) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4) 1 (1.4)
Genotypes described in this study for the first time. A total of 74 genotype results were obtained, 67 from BTM (seven of 60 BTM samples included two different genotypes) and seven from dust samples. b
233
No significant associations were observed between changes in genotypes within the herd during two years and decrease/increase in the herd size, or type of heifers’ management (inside/outside the farm). Similarly no associations were observed between coexistence of small ruminants in the herd and C. burnetii genogroups (GG II/GG III).
4. Discussion In this study, C. burnetii positive samples collected from dairy cattle herds in northern Spain were genotyped by MLVA and types from different sources (BTM and dust) and sampling time (2009/10 and 2011/12) were compared. Fifteen different MLVA genotypes were identified in 36 dairy cattle herds, indicating a high genetic diversity of C. burnetii strains in this region. In a preliminary genotyping study, our group had already used MLVA to type a selection of C. burnetii positive samples from domestic ruminants, including 20 collected from 12 dairy cattle farms from different parts of Spain (Astobiza et al., 2012b). Seven genotypes were found in cattle herds in that study, including four of the genotypes identified here (J, I, AB and M). In fact, genotypes J and I had been reported in cattle milk products from other countries (Sulyok et al., 2014; Tilburg et al., 2012b; Tilburg, 2013), indicating the existence of a common pool of C. burnetii strains that infect cattle worldwide. On the other hand, seven of the 15 genotypes found here had not been described before, although certainly these novel genotypes were only sporadically found, each representing 1.4–4.1% of the detections (13.5% in total). Other studies that carried out MLVA typing for the first time in a particular region, also reported the description of novel genotypes (Frangoulidis et al., 2014; Santos et al., 2012; Sulyok et al., 2014). In any case, more MLVA data are needed to obtain a more comprehensive image of the complete population structure of C. burnetii. Comparison of C. burnetii genotypes at different samplings within each herd suggested that the same strain was able to persist for a long time in the herd. Often persistent genotypes were among the three most prevalent types (I, J or AB), which is in agreement with results from other studies carried out in dairy cattle herds (Piñero et al., 2014a) and sheep flocks (Astobiza et al., 2012b). In those cases when different genotypes were found between samplings (32% of the herds), types generally differed in only one marker. These closely related genotypes could represent microvariants within the C. burnetii population that infect cattle. These small variations were only detected by the marker that according to the Simpson’s index of diversity had the highest discriminatory power (Ms34). This same marker also allowed the detection of mixed infections in seven BTM samples. Presence of several genotypes in BTM samples was not unexpected and has been reported before (de Bruin et al., 2012; Tilburg et al., 2012b). In fact, PCR might have underestimated the real number of BTM samples contaminated with more than one genotype since preferential amplification of the most prevalent genotype in the mixture would be likely. The MLVA typing scheme using six markers showed a high discriminatory power (Simpson’s Index of Diversity of 0.933) and demonstrated its ability to estimate the genotypic diversity of C. burnetii infecting cattle and the farm environment. However, as previously reported (Roest et al., 2011; Tilburg et al., 2012c), typeability is compromised in samples with low C. burnetii burden (Ct > 31). Clustering of the MLVA genotypes of this and other studies using the minimum spanning tree method separated our C. burnetii strains into two clusters. Most of the strains included in this study clustered within GG III, which mainly includes C. burnetii from cattle. However, five genotypes identified in eight BTM samples from
234
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
Fig. 1. Minimum spanning tree of MLVA genotypes. Genotypes identified in this study (indicated with a golden star), along with others reported elsewhere (Astobiza et al., 2012b; Santos et al., 2012; Sulyok et al., 2014; Tilburg et al., 2012b, 2012c; Tilburg, 2013) and five sequenced reference strains of C. burnetii: CbuG_Q212 (GenBank accession number CP001019), CbuG_Q154 (CP001020), Dugway (CP000733), Nine Mile RSA493 (AE016828) and Henzerling RSA331 (CP000890), whose MLVA pattern were determined in silico using the published sequences (Tilburg et al., 2012c), are included. Each circle corresponds to a genotype and its size refers to the number of samples in which that type has been identified. Each color represents the species where the genotype was detected. Branch labels and connecting lines correspond to the number of different markers between the MLVA genotypes. Predicted genomic groups (GGs) based on (Hendrix et al., 1991; Hornstra et al., 2011) are highlighted with gray color and genotypes are classified in different GGs according to the information described by other authors (Hornstra et al., 2011; Tilburg, 2013). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
five herds grouped within GG II. Four of these five genotypes were described here for the first time. Strains within GG II have been associated with sheep and goats, and have been implicated in human outbreaks (Glazunova et al., 2005; Hornstra et al., 2011). In fact, the strain responsible for the human Q fever outbreak in the Netherlands (strain G, MST 33) belongs to GG II (see Fig. 1), and was linked to goats (Tilburg et al., 2012a). These results show that genotypes closely related to those identified in humans can also infect cattle, but still, this does not seem to be common event (Astobiza et al., 2012b; Santos et al., 2012; Tilburg et al., 2012b). In conclusion, a high genotypic diversity was observed among C. burnetii from epidemiologically unrelated dairy cattle herds in northern Spain. Yet, some predominant genotypes (J, I) with worldwide distribution were also identified. Genotype comparison demonstrated that only 11% of the overall detections corresponded
to genotypes closely related to those identified in humans, suggesting that dairy cattle play a limited role in human Q fever infection. Surveillance of genetic distribution of C. burnetii from different sources is needed to fully understand the epidemiology of Q fever. Acknowledgements This study was supported by Spanish National Institute for Agricultural and Food Research and Technology (INIA, RTA 201300051-C2-00) and the European Regional Development Fund (ERDF). We are grateful to Mónica Villoria (NEIKER, Spain) and Dr. Jeroen Tilburg (Canisius-Wilhelmina Hospital, The Netherlands) for technical support and helpful suggestions. AP is the recipient of a predoctoral fellowship from INIA. We acknowledge the interest and collaboration of the farmers.
A. Piñero et al. / Infection, Genetics and Evolution 31 (2015) 231–235
References Arricau-Bouvery, N., Hauck, Y., Bejaoui, A., Frangoulidis, D., Bodier, C.C., Souriau, A., Meyer, H., Neubauer, H., Rodolakis, A., Vergnaud, G., 2006. Molecular characterization of Coxiella burnetii isolates by infrequent restriction site-PCR and MLVA typing. BMC Microbiol. 6, 38. Astobiza, I., Ruiz-Fons, F., Piñero, A., Barandika, J.F., Hurtado, A., García-Pérez, A.L., 2012a. Estimation of Coxiella burnetii prevalence in dairy cattle in intensive systems by serological and molecular analyses of bulk-tank milk samples. J. Dairy Sci. 95, 1632–1638. Astobiza, I., Tilburg, J.J., Piñero, A., Hurtado, A., García-Pérez, A.L., Nabuurs-Franssen, M.H., Klaassen, C.H., 2012b. Genotyping of Coxiella burnetii from domestic ruminants in northern Spain. BMC Vet. Res. 8, 241. Babudieri, B., 1959. Q fever: a zoonosis. Adv. Vet. Sci. 5, 81–182. de Bruin, A., van Alphen, P.T., van der Plaats, R.Q., de Heer, L., Reusken, C.B., van Rotterdam, B.J., Janse, I., 2012. Molecular typing of Coxiella burnetii from animal and environmental matrices during Q fever epidemics in the Netherlands. BMC Vet. Res. 8, 165. Dijkstra, F., van der Hoek, W., Wijers, N., Schimmer, B., Rietveld, A., Wijkmans, C.J., Vellema, P., Schneeberger, P.M., 2012. The 2007–2010 Q fever epidemic in The Netherlands: characteristics of notified acute Q fever patients and the association with dairy goat farming. FEMS Immunol. Med. Microbiol. 64, 3–12. EFSA, 2010. Scientific opinion on Q fever. EFSA J. 8, 1595. Frangoulidis, D., Walter, M.C., Antwerpen, M., Zimmermann, P., Janowetz, B., Alex, M., Bottcher, J., Henning, K., Hilbert, A., Ganter, M., Runge, M., Munsterkotter, M., Splettstoesser, W.D., Hanczaruk, M., 2014. Molecular analysis of Coxiella burnetii in Germany reveals evolution of unique clonal clusters. Int. J. Med. Microbiol. 304, 868–876. Georgiev, M., Afonso, A., Neubauer, H., Needham, H., Thiery, R., Rodolakis, A., Roest, H., Stark, K., Stegeman, J., Vellema, P., van der Hoek, W., More, S., 2013. Q fever in humans and farm animals in four European countries, 1982 to 2010. Euro Surveill. 18, pii=20407. Glazunova, O., Roux, V., Freylikman, O., Sekeyova, Z., Fournous, G., Tyczka, J., Tokarevich, N., Kovacava, E., Marrie, T.J., Raoult, D., 2005. Coxiella burnetii genotyping. Emerg. Infect. Dis. 11, 1211–1217. Grundmann, H., Hori, S., Tanner, G., 2001. Determining confidence intervals when measuring genetic diversity and the discriminatory abilities of typing methods for microorganisms. J. Clin. Microbiol. 39, 4190–4192. Hendrix, L.R., Samuel, J.E., Mallavia, L.P., 1991. Differentiation of Coxiella burnetii isolates by analysis of restriction-endonuclease-digested DNA separated by SDS-PAGE. J. Gen. Microbiol. 137, 269–276. Hornstra, H.M., Priestley, R.A., Georgia, S.M., Kachur, S., Birdsell, D.N., Hilsabeck, R., Gates, L.T., Samuel, J.E., Heinzen, R.A., Kersh, G.J., Keim, P., Massung, R.F., Pearson, T., 2011. Rapid typing of Coxiella burnetii. PLoS One 6, e26201. Hunter, P.R., Gaston, M.A., 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson’s index of diversity. J. Clin. Microbiol. 26, 2465–2466. Kersh, G.J., Lambourn, D.M., Raverty, S.A., Fitzpatrick, K.A., Self, J.S., Akmajian, A.M., Jeffries, S.J., Huggins, J., Drew, C.P., Zaki, S.R., Massung, R.F., 2012. Coxiella
235
burnetii infection of marine mammals in the Pacific Northwest, 1997–2010. J. Wildl. Dis. 48, 201–206. Klaassen, C.H., Nabuurs-Franssen, M.H., Tilburg, J.J., Hamans, M.A., Horrevorts, A.M., 2009. Multigenotype Q fever outbreak, The Netherlands. Emerg. Infect. Dis. 15, 613–614. Koh, X.P., Chiou, C.S., Ajam, N., Watanabe, H., Ahmad, N., Thong, K.L., 2012. Characterization of Shigella sonnei in Malaysia, an increasingly prevalent etiologic agent of local shigellosis cases. BMC Infect. Dis. 12, 122. Piñero, A., Barandika, J.F., Hurtado, A., García-Pérez, A.L., 2014a. Progression of Coxiella burnetii infection after implementing a two-year vaccination program in a naturally infected dairy cattle herd. Acta Vet. Scand. 56, 47. Piñero, A., Ruiz-Fons, F., Hurtado, A., Barandika, J.F., Atxaerandio, R., García-Pérez, A.L., 2014b. Changes in the dynamics of Coxiella burnetii infection in dairy cattle: an approach to match field data with the epidemiological cycle of C. burnetii in endemic herds. J. Dairy Sci. 97, 2718–2730. Roest, H.I., Ruuls, R.C., Tilburg, J.J., Nabuurs-Franssen, M.H., Klaassen, C.H., Vellema, P., van den Brom, R., Dercksen, D., Wouda, W., Spierenburg, M.A., van der Spek, A.N., Buijs, R., de Boer, A.G., Willemsen, P.T., van Zijderveld, F.G., 2011. Molecular epidemiology of Coxiella burnetii from ruminants in Q fever outbreak, the Netherlands. Emerg. Infect. Dis. 17, 668–675. Santos, A.S., Tilburg, J.J., Botelho, A., Barahona, M.J., Nuncio, M.S., Nabuurs-Franssen, M.H., Klaassen, C.H., 2012. Genotypic diversity of clinical Coxiella burnetii isolates from Portugal based on MST and MLVA typing. Int. J. Med. Microbiol. 302, 253–256. Schets, F.M., de Heer, L., de Roda Husman, A.M., 2013. Coxiella burnetii in sewage water at sewage water treatment plants in a Q fever epidemic area. Int. J. Hyg. Environ. Health 216, 698–702. Sulyok, K.M., Kreizinger, Z., Hornstra, H.M., Pearson, T., Szigeti, A., Dan, A., Balla, E., Keim, P.S., Gyuranecz, M., 2014. Genotyping of Coxiella burnetii from domestic ruminants and human in Hungary: indication of various genotypes. BMC Vet. Res. 10, 107. Svraka, S., Toman, R., Skultety, L., Slaba, K., Homan, W.L., 2006. Establishment of a genotyping scheme for Coxiella burnetii. FEMS Microbiol. Lett. 254, 268–274. Tilburg, J.J., 2013. Molecular Investigation of the Q Fever Epidemic in the Netherlands. The Largest Outbreak Caused by Coxiella burnetii Ever Reported. PhD Thesis, Radboud Universiteit Nijmegen, Boxpress, Nijmegen, The Netherlands. . Tilburg, J.J., Roest, H.J., Buffet, S., Nabuurs-Franssen, M.H., Horrevorts, A.M., Raoult, D., Klaassen, C.H., 2012a. Epidemic genotype of Coxiella burnetii among goats, sheep, and humans in the Netherlands. Emerg. Infect. Dis. 18, 887–889. Tilburg, J.J., Roest, H.J., Nabuurs-Franssen, M.H., Horrevorts, A.M., Klaassen, C.H., 2012b. Genotyping reveals the presence of a predominant genotype of Coxiella burnetii in consumer milk products. J. Clin. Microbiol. 50, 2156–2158. Tilburg, J.J., Rossen, J.W., van Hannen, E.J., Melchers, W.J., Hermans, M.H., van de Bovenkamp, J., Roest, H.J., de Bruin, A., Nabuurs-Franssen, M.H., Horrevorts, A.M., Klaassen, C.H., 2012c. Genotypic diversity of Coxiella burnetii in the 2007– 2010 Q fever outbreak episodes in The Netherlands. J. Clin. Microbiol. 50, 1076– 1078.
