- Email: [email protected]
Screening of various foodstuffs for occurrence of Coxiella burnetii in Switzerland
International Journal of Food Microbiology 116 (2007) 414 – 418 www.elsevier.com/locate/ijfoodmicro
Short communication
Screening of various foodstuffs for occurrence of Coxiella burnetii in Switzerland R. Fretz a,b , W. Schaeren c , M. Tanner b , A. Baumgartner a,⁎ a
c
Swiss Federal Office of Public Health, 3003 Berne, Switzerland b Swiss Tropical Institute, 4002 Basle, Switzerland Agroscope Liebefeld—Posieux Research Station, 3003 Berne, Switzerland Received 3 January 2007; accepted 7 March 2007
Abstract The epidemiology of Q-fever in Switzerland is largely unknown. For this reason, a screening programme for the presence of Coxiella burnetii in bulk milk samples from cows, sheep and goats and in shell eggs produced in and imported into Switzerland was conducted. In total, 17 of 359 (4.7%) of analysed bovine milk samples from two randomly selected cheese dairies were tested positive for C. burnetii by nested PCR. Furthermore, the findings with samples from one dairy showed that the agent seemed to persist over time in the herds of cattle of certain farms. Although no extensive prevalence study was undertaken, our results indicate that C. burnetii appears to be quite frequent in cattle. As for 81 ovine and 39 caprine bulk milk samples, they were all tested negative for C. burnetii. Finally, 504 shell eggs were also found to be negative for C. burnetii with PCR testing. The results of the study are discussed under inclusion of epidemiological data for human and animal coxiellosis and the current Swiss legal regulations for the control of C. burnetii in cattle. © 2007 Elsevier B.V. All rights reserved. Keywords: Coxiella burnetii; Screening; Milk; Eggs; Switzerland
1. Introduction Q-fever is an ubiquitous zoonosis caused by Coxiella burnetii, an obligate intracellular rickettsial organism and has been found throughout the world, except New Zealand (Maurin and Raoult, 1999). A wide variety of animals can be infected with C. burnetii, including cows, goats, sheep, dogs, cats, non human primates, wild rodents, small mammals, big game wildlife, non-mammalian animals, including reptiles, amphibians, birds (domesticated and wild), fish, and many ticks (Parker et al., 2006). Among farm animals, dairy cattles, sheep, and goats are the major reservoirs of C. burnetii (Kim et al., 2005). The uterus and mammary glands are primary sites of infection in the chronic phase of C. burnetii. Shedding into the environment occurs mainly during parturition by birth products, particularly the placenta. Shedding of C. burnetii in milk by ⁎ Corresponding author. Tel.: +41 31 322 95 82; fax: +41 31 322 95 74. E-mail address: [email protected] (A. Baumgartner). 0168-1605/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2007.03.001
infected dairy cattle is documented, too (Maurin and Raoult, 1999; Kim et al., 2005). Often, animals infected with C. burnetii do not show typical clinical symptoms such as abortion in sheep and goats or reproductive disorders in cattle (Kim et al., 2005). In humans, symptoms are highly variable and about 60% of infections are asymptomatic sero-conversions. Acute Q-fever presents mainly as flu-like disease, or atypical pneumonia or hepatitis (ArricauBouvery and Rodolakis, 2005). Q-fever is essentially an airborne disease. Infections occur after inhalation of aerosols generated from infected animal placentas, body fluids or contaminated dust resulting from contaminated manure and desiccation of infected placenta and body fluids (ArricauBouvery and Rodolakis, 2005). The significance of infection via the oral route is still a subject of discussion (Lorenz et al., 1998). Drinking contaminated milk has induced sero-conversion in human volunteers without clinical signs (ArricauBouvery and Rodolakis, 2005). Since birds are part of the host spectrum of C. burnetii, infected domestic poultry can transmit
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
the agent to humans by contaminated raw eggs or through fomites (Hirai and To, 1998). Previous studies on the prevalence of C. burnetii in dairy cattle were based mainly on serological tests (Kim et al., 2005). So far there are only few studies using PCR for screening. For example, a recent study reports a high prevalence of more than 94% of C. burnetii in samples of bulk tank milk from U.S. dairy herds tested by PCR (Kim et al., 2005). Furthermore, a screening of milk samples from 400 sheep in Turkey showed a PCR-positive rate for C. burnetii of 3.5% (Ongor et al., 2004). The findings of a large scale screening of shell eggs and egg products from Toyama prefecture in Japan revealed that 4.2% of the egg samples and 17.6% of the mayonnaise specimens were tested positive for C. burnetii by PCR (Tatsumi et al., 2006). The epidemiology of Q-fever in Switzerland is largely unknown. The last comprehensive sero-epidemiological study on Q-fever in Switzerland was undertaken in the 1980s (Metzler et al., 1983). For this reason, and because of the Japanese findings in the context of shell eggs and mayonnaise, the Swiss Federal Office of Public Health launched a screening programme about the occurrence of C. burnetii in foodstuffs where this infectious agent can be expected most likely. 2. Materials and methods 2.1. Analysed foodstuffs 2.1.1. Bovine milk From January to June 2006, bovine bulk milk samples obtained from two cheese dairies form the Canton of Berne in Switzerland were analysed for the presence of C. burnetii. The two cheese dairies were supplied by 12 and 15, respectively, milk cow farms which are gathered together in a 10 km radius around each of these cheese dairies. A monthly average of 9800 kg and 6900 kg, respectively, of cow milk were delivered to the two dairies which are situated around the city of Berne and are approximately 30 km apart from each other. Once every two weeks, bulk milk samples from each farm were taken directly at the cheese dairy during the early morning delivery. 2.1.2. Ovine milk From May to November 2005, according to the breeding season, bulk milk samples from sheep were obtained from 13 sheep breeding farms. The farms were settled within a radius of 20 km around a sheep cheese dairy in the Emmental region of the Canton of Berne, Switzerland. 2.1.3. Caprine milk During a two months period in 2006 (May–June), single bulk milk samples from goat-breeding farms were collected from a total of 39 farms originating in the adjoining cantons of Berne and Lucerne, Switzerland. The farms were randomly selected on the basis of a herd size of more than 30 animals. 2.1.4. Shell eggs From April to November 2005, 504 shell eggs were purchased quasi-randomly at supermarkets of 4 retailing companies in the
415
area of the city of Berne. A sampling unit consisted of 18 eggs of the same producer and the same date of laying. Shell eggs of Swiss origin were 5 sampling units from floor breeding farms and 5 units from outdoor breeding farms. Imported shell eggs were 5 sampling units from France, 4 from Poland, 4 from The Netherlands, 3 from Belgium, and 2 from Germany. 2.2. PCR-detection of C. burnetii 2.2.1. DNA extraction C. burnetii was isolated from milk samples by centrifuging and removing of cream and milk layers as described previously (Berri et al., 2003) and further processed by the protocol for purification of total DNA from tissue samples using the BioSprint 15 DNA Blood Kit (QIAGEN GmbH, Hilden, Germany) adapted to the KingFisher magnetic particle processor (Thermo Electron Corporation, Germany). From egg samples, the agent was extracted by placing a sterilized filter paper (7 cm × 1 cm) crosswise in the yolk of each egg sample. The moistened filter paper was put into a plastic tube (Sarstedt Ltd., Nümbrecht, Germany, tube 50 ml PP) containing 30 ml PBST washing solution (1× PBS + 0.5% Tween 20 (PBS, Tween 20: Sigma-Aldrich Chemie GmbH, Steinheim, Germany) and 50 times vigorously rotated by hand. After removing the filter paper, the tube was centrifuged at 400 ×g at 4 °C for 15 min. The supernatant was transferred to a new plastic tube (Labcon North America, Petaluma, CA, USA, tube 50 ml PP) and centrifuged with 15,000 ×g at 4 °C for another 15 min. The supernatant except a remaining of approximately 1.5 ml was discarded. The pellet with the remaining liquid was resuspended and transferred to a sterile 2 ml tube (Sarstedt Ltd., Nümbrecht, Germany). The 2 ml tube was centrifuged with 13,000 ×g at 4 °C for 40 min and the supernatant was discarded. In an additional washing step, the pellet was resuspended in 1 ml Tris Buffer (1% Np40 Substitute + 1% Tween 20+ 10 mM Tris–HCl [pH 8.0] (Np40 Substitute, Tween 20: Sigma-Aldrich Chemie GmbH, Steinheim, Germany; Tris–HCl: Merck, Darmstadt, Germany) and was centrifuged for 15 min at 13,000 ×g and 4 °C. Finally, the supernatant was discarded and the pellet was further processed as described with milk samples but following the QIAGEN protocol for purification of DNA from blood. 2.2.2. Nested PCR assay All oligonucleotide primers were obtained from a commercial source (Microsynth GmbH, Balgach, Switzerland). The nested PCR assay used to screen for C. burnetii was designed from the nucleotide sequence of the com1 gene encoding a 27kDa outer membrane protein as previously described (Zhang et al., 1998), but the amplification procedure was slightly modified. For the nested PCR with primers OMP1–OMP2 and OMP3–OMP4, the first amplification was performed in a total volume of 25 μl containing 5 μl of DNA sample, 0.5 mM MgCl2, 0.2 mM (each) dNTPs, 1 μM primer OMP1, 1 μM primer OMP2, and 0.5 U/reaction of Takara Taq™ DNA polymerase (Axon Lab GmbH, Baden-Dättwil, Switzerland). PCR was performed at 94 °C for 4 min and then for 36 cycles of
416
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
Fig. 1. Six-month screening (Jan–June 2006) by nested PCR for Coxiella burnetii of bovine bulk milk samples obtained from two cheese dairies in Switzerland.
94 °C for 1 min, 56 °C for 1 min, and 72 °C for 1 min in a DNA thermal cycler (Techne TC-312 FTC3105D; Witec GmbH, Littau, Switzerland). In the second amplification, the reaction was performed in a total volume of 25 μl containing 2 μl of DNA sample, 0.5 mM MgCl2, 0.2 mM (each) dNTPs, 0.8 μM primer OMP3, 0.8 μM primer OMP4, and 0.5 U/reaction of Takara Taq™ DNA polymerase. PCR was performed at 96 °C for 4 min and then for 30 cycles of 96 °C for 1 min, 57°C for 1 min, and 72 °C for 1 min. The PCR-amplified products (OMP1–OMP2: 501 bp; OMP3–OMP4: 438 bp) were examined by electrophoresis in a 1.5% agarose gel, stained with a 1% solution of ethidium bromide, and examined under UV illumination with the AlphaImager™ Imaging System (Witec GmbH, Littau, Switzerland). Positive results were confirmed by nested PCR, DNA sequencing (Microsynth GmbH, Balgach, Switzerland) and comparison of sequences with NCBI GenBank. 2.2.3. Positive control and sensitivity of nested PCR The positive control used in the study included heatinactivated isolates of C. burnetii, strain Nine Mile RSA493, obtained from the Institute for Hygiene and Infectious Diseases of Animals, Justus-Liebig University, Giessen, Germany. Milk and egg samples were spiked with 105 particles of C. burnetii, and used as positive controls in each test from DNA extraction to PCR. The sensitivity of the nested PCR on the various samples was assessed by spiking samples with serial 10-fold dilutions of particles of heat-inactivated C. burnetii.
milk producing farm was tested again positive in April and June 2006. Samples from a second and a third farm which all delivered milk to the same cheese dairy were also tested positive twice during the screening period. Noteworthy is the observation that there was a clear difference in the overall rate of positive results for C. burnetii between the two cheese dairies (dairy A and dairy B) and between the allocated milk supplying farms, respectively. In contrast, all ovine bulk milk samples from 13 sheep breeding farms (totally 81) and all caprine bulk milk sample from 39 farms (totally 39 samples) were shown to be negative for C. burnetii by the same detection method used for the analysis of cow milk samples. As for shell eggs, 504 were individually analysed with nested PCR and entirely tested negative for C. burnetii. The sensitivity of the applied nested PCR was assessed by spiking of samples with heat-inactivated particles of C. burnetii. The sensitivity limit was found to be 103 particles/1 ml cow or sheep milk, 102 particles/1 ml goat milk and 102 particles/15 mg egg yolk (see Fig. 2). The detection limit in the study of Lorenz et al. (1998) was reported to be 500 particles of C. burnetii/ml spiked milk sample. Berri et al. (2003) published detection
3. Results and discussion Fig. 1 shows the screening results for C. burnetii in bovine bulk milk samples. In total, 17 of 359 (4.7%) analysed milk samples were tested positive by nested PCR. The positive samples originated from 8 of 27 (29.6%) milk cow farms. Interestingly, in a number of farms, the agent seemed to persist over time. At least in one farm, C. burnetii could be detected in bi-weekly samples from January and February 2006. The same
Fig. 2. Sensitivity of the Coxiella burnetii-specific nested PCR with food samples after DNA preparation. Lane 1: positive control with 105 C. burnetii particles/ml; lane 2: egg yolk with 103 particles/15 mg; lane 3: egg yolk with 102 particles/15 mg; lane 4: bovine milk with 103 particles/ml; lane 5: bovine milk with 102 particles/ml; lane 6: ovine milk with 103 particles/ml; lane 7: ovine milk with 102 particles/ml; lane 8: caprine milk with 103 particles/ml; lane 9: caprine milk with 102 particles/ml; used marker: 50 bp DNA ladder.
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
limits of 10 particles/ml of bacterial suspension and 1 bacteria/ ml of milk sample. A recent Japanese screening on milk and eggs based on the same nested PCR methodology reported a detection limit of 3.2 × 101 particles of C. burnetii/1 g of egg yolk Hirai et al., 2005). A search in PubMed up to the year 2006 revealed that only in a few studies used PCR to conduct analyses of foodstuffs for C. burnetii. A recent publication from the U.S. indicates a prevalence of C. burnetii of over 94% in bulk tank milk samples from dairy herds in the period from 2002 to 2004. Furthermore, it was reported that the C. burnetii infection might be persistent in the dairy herds, with little temporal or regional variations. Shedding levels of positive cattle were estimated to be 101– 1014 particles of C. burnetii/ml. The bulk tank milk samples of the herd stayed at a level of 102 particles/ml over the whole study period (Kim et al., 2005). The fact that the sensitivity of our detection method was around 103 cells/ml of cows milk may explain the variation of positive and negative results with samples taken from the cheese dairy A (see Fig. 1). The reasons for the local differences in the overall rate samples positive for C. burnetii between the farms and the two cheese dairies remain unclear. Although the dairies were only situated about 30 km apart from each other, the according two cattle populations were strictly separated from each other because of geographical reasons. Our findings with ovine milk samples can only be compared to a recent Turkish study where single milk samples from 400 sheep of 22 flocks were analysed and a positive rate of 3.5% for C. burnetii was found. All positive results were obtained only from flocks with a history of abortion (Ongor et al., 2004). Unfortunately, no veterinary records from the Swiss sheep farms were available. The negative findings were unexpected, since sheep were identified to have caused human cases of coxiellosis in Switzerland. In 1983, a Q-fever outbreak occurred involving 415 serologically confirmed persons. The origin of infections were 12 flocks of sheep with apparently ill or C. burnetii shedding animals descending from the alpine pastures to the valley and passing through villages (Dupuis et al., 1987). In Japan, Tatsumi et al. (2006) tested 4.2% of shell eggs and 17.6% of mayonnaise samples positive for C. burnetii by PCR. Based on these findings, we conducted the first screening of shell eggs for C. burnetii outside of Japan by using the same detection methodology as the Japanese authors. But in contrast to their findings, no PCR-positive results were obtained. This is in line with the results of another Japanese screening of chicken eggs Hirai et al., 2005). Finally, a third Japanese study could not reproduce the detection of C. burnetii in mayonnaise (Sadamasu et al., 2006). The discrepancy between the three Japanese studies might be explained by the fact that the analysed eggs were produced in different regions of the country. Since we could not detect C. burnetii in shell eggs, it can be concluded that the epidemiological situation in Switzerland, and in countries from which eggs were imported, must be different to the situation in Japan. This conclusion is also supported by the fact that we used the same detection system as Tatsumi et al. (2006). Surprisingly, all sheep and goat milk samples were shown to be negative for C. burnetii. In future investigations,
417
these findings should be verified with most recently published PCR methods of higher sensitivity. The fact that the target micro-organism was demonstrated in the milk of both randomly selected cheese dairies, representing 27 milk farms, indicates a higher prevalence of C. burnetii in the Swiss cattle population than expected by epidemiological data. The morbidity for human coxiellosis is very low in Switzerland. In 1998, only 11 cases were registered within a population of 7,123,500 persons (Swiss Federal Office of Public Health, 1999). For this reason, mandatory reporting of human coxiellosis was stopped in the following year. Animal cases of coxiellosis are also quite rare. In the year 2005, only 30 cases with cattle, 5 with sheep and 5 with goats were reported (Swiss Federal Veterinary Office, 2006). The present study shows that C. burnetii occurs quite commonly in Swiss cattle. If this finding can be verified with more extensive prevalence studies, the legal regulations to control coxiellosis in cattle herds should be re-evaluated. Actually, the Ordinance on Epizootic Diseases still considers coxiellosis of animals as a disease to be controlled (Federal Department of Economic Affairs, 1995). Such a classification is discussible since C. burnetii seems to be widely spread in the cattle population and obviously has low clinical impact both in human and veterinary medicine. Acknowledgements The authors would like to express their gratefulness to Professor Dr. Noriyuki Tatsumi, Director, Dr. Ying Qiao and Dr. Ikkyu Yamamoto of the Institute for Preventive Medicine against Zoonosis (IPMZ), Komoridani, Oyabe City, Toyama Prefecture, Japan, for transferring the methodology to analyse chicken market eggs for C. burnetii. Our thanks also include Dr. Stefanie P. Templer, Swiss Federal Office of Public Health (SFOPH), for her support in PCR-analysis. References Arricau-Bouvery, N., Rodolakis, A., 2005. Is Q fever an emerging or reemerging zoonosis? Veterinary Research 36, 327–349. Berri, M., Arricau-Bouvery, N., Rodolakis, A., 2003. PCR-based detection of Coxiella burnetii from clinical samples. Methods in Molecular Biology 216, 153–161. Dupuis, G., Petite, J., Peter, O., Vouilloz, M., 1987. An important outbreak of human Q fever in a Swiss Alpine valley. International Journal of Epidemiology 16, 282–287. Federal Department of Economic Affairs, 1995. Tierseuchenverordnung (TSV) vom 27. Juni 1995, Stand am 2. Mai 2006 [in German]. Available at: http:// www.admin.ch/ch/d/sr/9/916.401.de.pdf. Hirai, A., Kaneko, S., Nakama, A., Ishizaki, N., Odagiri, M., Kai, A., Sadamasu, K., Shinkai, T., Yano, K., Morozumi, S., 2005. Investigation of Coxiella burnetii contamination in commercial milk and PCR method for the detection of C. burnetii in egg. Shokuhin Eiseigaku Zasshi 46, 86–92. Hirai, K., To, H., 1998. Advances in the understanding of Coxiella burnetii infection in Japan. Journal of Veterinary Medical Science 60, 781–790. Kim, S.G., Kim, E.H., Lafferty, C.J., Dubovi, E., 2005. Coxiella burnetii in bulk tank milk samples, United States. Emerging Infectious Diseases 11, 619–621. Lorenz, H., Jäger, C., Willems, H., Baljer, G., 1998. PCR detection of Coxiella burnetii from different clinical specimens, especially bovine milk, on the basis of DNA preparation with a silica matrix. Applied and Environmental Microbiology 64, 4234–4237.
418
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
Maurin, M., Raoult, D., 1999. Q fever. Clinical Microbiology Reviews 12, 518–553. Metzler, A.E., Nicolet, J., Bertschinger, H.U., Bruppacher, R., Gelzer, J., 1983. Distribution of Coxiella burnetii: a seroepidemiological study of domestic animals and veterinarians [in German]. Schweizer Archiv für Tierheilkunde 125, 507–517. Ongor, H., Cetinkaya, B., Karahan, M., Acik, M.N., Bulut, H., Muz, A., 2004. Detection of Coxiella burnetii by immunomagnetic separation-PCR in the milk of sheep in Turkey. Veterinary Record 154, 570–572. Parker, N.R., Barralet, J.H., Bell, A.M., 2006. Q-fever. Lancet 367, 679–688. Sadamasu, K., Tabei, Y., Shinkai, T., Hasegawa, M., Kaneko, S., Hirai, A., Nakama, A., Ishizaki, N., Odagiri, M., Kamata, S., Yano, K., Kai, A., Morozumi, S., 2006. Development of effective detection method for Coxiella burnetii in mayonnaise by real-time PCR and investigation of C.
burnetii contamination in commercial mayonnaise in Tokyo. Shokuhin Eiseigaku Zasshi 47, 1–8. Swiss Federal Office of Public Health, 1999. Weekly Bulletin no. 1, p. 6. Swiss Federal Veterinary Office, 2006. Weekly Bulletin no. 1, p. 12. Tatsumi, N., Baumgartner, A., Qiao, Y., Yamamoto, I., Yamaguchi, K., 2006. Detection of Coxiella burnetii in market chicken eggs and mayonnaise In: Hechemy, K.E., Oteo, J.O., Raoult, D.A., Silverman, D.J., Blanco, J.R. (Eds.), Century of Rickettsiology — Emerging, Reemerging Rickettsioses, Molecular Diagnostics, and Emerging Veterinary Rickettsioses. Annals of the New York Academy of Sciences, vol. 1078, pp. 502–505. Zhang, G.Q., Nguyen, S.V., To, H., Ogawa, M., Hotta, A., Yamaguchi, T., Kim, H.J., Fukushi, H., Hirai, K., 1998. Clinical evaluation of a new PCR assay for detection of Coxiella burnetii in human serum samples. Journal of Clinical Microbiology 36, 77–80.
Short communication
Screening of various foodstuffs for occurrence of Coxiella burnetii in Switzerland R. Fretz a,b , W. Schaeren c , M. Tanner b , A. Baumgartner a,⁎ a
c
Swiss Federal Office of Public Health, 3003 Berne, Switzerland b Swiss Tropical Institute, 4002 Basle, Switzerland Agroscope Liebefeld—Posieux Research Station, 3003 Berne, Switzerland Received 3 January 2007; accepted 7 March 2007
Abstract The epidemiology of Q-fever in Switzerland is largely unknown. For this reason, a screening programme for the presence of Coxiella burnetii in bulk milk samples from cows, sheep and goats and in shell eggs produced in and imported into Switzerland was conducted. In total, 17 of 359 (4.7%) of analysed bovine milk samples from two randomly selected cheese dairies were tested positive for C. burnetii by nested PCR. Furthermore, the findings with samples from one dairy showed that the agent seemed to persist over time in the herds of cattle of certain farms. Although no extensive prevalence study was undertaken, our results indicate that C. burnetii appears to be quite frequent in cattle. As for 81 ovine and 39 caprine bulk milk samples, they were all tested negative for C. burnetii. Finally, 504 shell eggs were also found to be negative for C. burnetii with PCR testing. The results of the study are discussed under inclusion of epidemiological data for human and animal coxiellosis and the current Swiss legal regulations for the control of C. burnetii in cattle. © 2007 Elsevier B.V. All rights reserved. Keywords: Coxiella burnetii; Screening; Milk; Eggs; Switzerland
1. Introduction Q-fever is an ubiquitous zoonosis caused by Coxiella burnetii, an obligate intracellular rickettsial organism and has been found throughout the world, except New Zealand (Maurin and Raoult, 1999). A wide variety of animals can be infected with C. burnetii, including cows, goats, sheep, dogs, cats, non human primates, wild rodents, small mammals, big game wildlife, non-mammalian animals, including reptiles, amphibians, birds (domesticated and wild), fish, and many ticks (Parker et al., 2006). Among farm animals, dairy cattles, sheep, and goats are the major reservoirs of C. burnetii (Kim et al., 2005). The uterus and mammary glands are primary sites of infection in the chronic phase of C. burnetii. Shedding into the environment occurs mainly during parturition by birth products, particularly the placenta. Shedding of C. burnetii in milk by ⁎ Corresponding author. Tel.: +41 31 322 95 82; fax: +41 31 322 95 74. E-mail address: [email protected] (A. Baumgartner). 0168-1605/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2007.03.001
infected dairy cattle is documented, too (Maurin and Raoult, 1999; Kim et al., 2005). Often, animals infected with C. burnetii do not show typical clinical symptoms such as abortion in sheep and goats or reproductive disorders in cattle (Kim et al., 2005). In humans, symptoms are highly variable and about 60% of infections are asymptomatic sero-conversions. Acute Q-fever presents mainly as flu-like disease, or atypical pneumonia or hepatitis (ArricauBouvery and Rodolakis, 2005). Q-fever is essentially an airborne disease. Infections occur after inhalation of aerosols generated from infected animal placentas, body fluids or contaminated dust resulting from contaminated manure and desiccation of infected placenta and body fluids (ArricauBouvery and Rodolakis, 2005). The significance of infection via the oral route is still a subject of discussion (Lorenz et al., 1998). Drinking contaminated milk has induced sero-conversion in human volunteers without clinical signs (ArricauBouvery and Rodolakis, 2005). Since birds are part of the host spectrum of C. burnetii, infected domestic poultry can transmit
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
the agent to humans by contaminated raw eggs or through fomites (Hirai and To, 1998). Previous studies on the prevalence of C. burnetii in dairy cattle were based mainly on serological tests (Kim et al., 2005). So far there are only few studies using PCR for screening. For example, a recent study reports a high prevalence of more than 94% of C. burnetii in samples of bulk tank milk from U.S. dairy herds tested by PCR (Kim et al., 2005). Furthermore, a screening of milk samples from 400 sheep in Turkey showed a PCR-positive rate for C. burnetii of 3.5% (Ongor et al., 2004). The findings of a large scale screening of shell eggs and egg products from Toyama prefecture in Japan revealed that 4.2% of the egg samples and 17.6% of the mayonnaise specimens were tested positive for C. burnetii by PCR (Tatsumi et al., 2006). The epidemiology of Q-fever in Switzerland is largely unknown. The last comprehensive sero-epidemiological study on Q-fever in Switzerland was undertaken in the 1980s (Metzler et al., 1983). For this reason, and because of the Japanese findings in the context of shell eggs and mayonnaise, the Swiss Federal Office of Public Health launched a screening programme about the occurrence of C. burnetii in foodstuffs where this infectious agent can be expected most likely. 2. Materials and methods 2.1. Analysed foodstuffs 2.1.1. Bovine milk From January to June 2006, bovine bulk milk samples obtained from two cheese dairies form the Canton of Berne in Switzerland were analysed for the presence of C. burnetii. The two cheese dairies were supplied by 12 and 15, respectively, milk cow farms which are gathered together in a 10 km radius around each of these cheese dairies. A monthly average of 9800 kg and 6900 kg, respectively, of cow milk were delivered to the two dairies which are situated around the city of Berne and are approximately 30 km apart from each other. Once every two weeks, bulk milk samples from each farm were taken directly at the cheese dairy during the early morning delivery. 2.1.2. Ovine milk From May to November 2005, according to the breeding season, bulk milk samples from sheep were obtained from 13 sheep breeding farms. The farms were settled within a radius of 20 km around a sheep cheese dairy in the Emmental region of the Canton of Berne, Switzerland. 2.1.3. Caprine milk During a two months period in 2006 (May–June), single bulk milk samples from goat-breeding farms were collected from a total of 39 farms originating in the adjoining cantons of Berne and Lucerne, Switzerland. The farms were randomly selected on the basis of a herd size of more than 30 animals. 2.1.4. Shell eggs From April to November 2005, 504 shell eggs were purchased quasi-randomly at supermarkets of 4 retailing companies in the
415
area of the city of Berne. A sampling unit consisted of 18 eggs of the same producer and the same date of laying. Shell eggs of Swiss origin were 5 sampling units from floor breeding farms and 5 units from outdoor breeding farms. Imported shell eggs were 5 sampling units from France, 4 from Poland, 4 from The Netherlands, 3 from Belgium, and 2 from Germany. 2.2. PCR-detection of C. burnetii 2.2.1. DNA extraction C. burnetii was isolated from milk samples by centrifuging and removing of cream and milk layers as described previously (Berri et al., 2003) and further processed by the protocol for purification of total DNA from tissue samples using the BioSprint 15 DNA Blood Kit (QIAGEN GmbH, Hilden, Germany) adapted to the KingFisher magnetic particle processor (Thermo Electron Corporation, Germany). From egg samples, the agent was extracted by placing a sterilized filter paper (7 cm × 1 cm) crosswise in the yolk of each egg sample. The moistened filter paper was put into a plastic tube (Sarstedt Ltd., Nümbrecht, Germany, tube 50 ml PP) containing 30 ml PBST washing solution (1× PBS + 0.5% Tween 20 (PBS, Tween 20: Sigma-Aldrich Chemie GmbH, Steinheim, Germany) and 50 times vigorously rotated by hand. After removing the filter paper, the tube was centrifuged at 400 ×g at 4 °C for 15 min. The supernatant was transferred to a new plastic tube (Labcon North America, Petaluma, CA, USA, tube 50 ml PP) and centrifuged with 15,000 ×g at 4 °C for another 15 min. The supernatant except a remaining of approximately 1.5 ml was discarded. The pellet with the remaining liquid was resuspended and transferred to a sterile 2 ml tube (Sarstedt Ltd., Nümbrecht, Germany). The 2 ml tube was centrifuged with 13,000 ×g at 4 °C for 40 min and the supernatant was discarded. In an additional washing step, the pellet was resuspended in 1 ml Tris Buffer (1% Np40 Substitute + 1% Tween 20+ 10 mM Tris–HCl [pH 8.0] (Np40 Substitute, Tween 20: Sigma-Aldrich Chemie GmbH, Steinheim, Germany; Tris–HCl: Merck, Darmstadt, Germany) and was centrifuged for 15 min at 13,000 ×g and 4 °C. Finally, the supernatant was discarded and the pellet was further processed as described with milk samples but following the QIAGEN protocol for purification of DNA from blood. 2.2.2. Nested PCR assay All oligonucleotide primers were obtained from a commercial source (Microsynth GmbH, Balgach, Switzerland). The nested PCR assay used to screen for C. burnetii was designed from the nucleotide sequence of the com1 gene encoding a 27kDa outer membrane protein as previously described (Zhang et al., 1998), but the amplification procedure was slightly modified. For the nested PCR with primers OMP1–OMP2 and OMP3–OMP4, the first amplification was performed in a total volume of 25 μl containing 5 μl of DNA sample, 0.5 mM MgCl2, 0.2 mM (each) dNTPs, 1 μM primer OMP1, 1 μM primer OMP2, and 0.5 U/reaction of Takara Taq™ DNA polymerase (Axon Lab GmbH, Baden-Dättwil, Switzerland). PCR was performed at 94 °C for 4 min and then for 36 cycles of
416
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
Fig. 1. Six-month screening (Jan–June 2006) by nested PCR for Coxiella burnetii of bovine bulk milk samples obtained from two cheese dairies in Switzerland.
94 °C for 1 min, 56 °C for 1 min, and 72 °C for 1 min in a DNA thermal cycler (Techne TC-312 FTC3105D; Witec GmbH, Littau, Switzerland). In the second amplification, the reaction was performed in a total volume of 25 μl containing 2 μl of DNA sample, 0.5 mM MgCl2, 0.2 mM (each) dNTPs, 0.8 μM primer OMP3, 0.8 μM primer OMP4, and 0.5 U/reaction of Takara Taq™ DNA polymerase. PCR was performed at 96 °C for 4 min and then for 30 cycles of 96 °C for 1 min, 57°C for 1 min, and 72 °C for 1 min. The PCR-amplified products (OMP1–OMP2: 501 bp; OMP3–OMP4: 438 bp) were examined by electrophoresis in a 1.5% agarose gel, stained with a 1% solution of ethidium bromide, and examined under UV illumination with the AlphaImager™ Imaging System (Witec GmbH, Littau, Switzerland). Positive results were confirmed by nested PCR, DNA sequencing (Microsynth GmbH, Balgach, Switzerland) and comparison of sequences with NCBI GenBank. 2.2.3. Positive control and sensitivity of nested PCR The positive control used in the study included heatinactivated isolates of C. burnetii, strain Nine Mile RSA493, obtained from the Institute for Hygiene and Infectious Diseases of Animals, Justus-Liebig University, Giessen, Germany. Milk and egg samples were spiked with 105 particles of C. burnetii, and used as positive controls in each test from DNA extraction to PCR. The sensitivity of the nested PCR on the various samples was assessed by spiking samples with serial 10-fold dilutions of particles of heat-inactivated C. burnetii.
milk producing farm was tested again positive in April and June 2006. Samples from a second and a third farm which all delivered milk to the same cheese dairy were also tested positive twice during the screening period. Noteworthy is the observation that there was a clear difference in the overall rate of positive results for C. burnetii between the two cheese dairies (dairy A and dairy B) and between the allocated milk supplying farms, respectively. In contrast, all ovine bulk milk samples from 13 sheep breeding farms (totally 81) and all caprine bulk milk sample from 39 farms (totally 39 samples) were shown to be negative for C. burnetii by the same detection method used for the analysis of cow milk samples. As for shell eggs, 504 were individually analysed with nested PCR and entirely tested negative for C. burnetii. The sensitivity of the applied nested PCR was assessed by spiking of samples with heat-inactivated particles of C. burnetii. The sensitivity limit was found to be 103 particles/1 ml cow or sheep milk, 102 particles/1 ml goat milk and 102 particles/15 mg egg yolk (see Fig. 2). The detection limit in the study of Lorenz et al. (1998) was reported to be 500 particles of C. burnetii/ml spiked milk sample. Berri et al. (2003) published detection
3. Results and discussion Fig. 1 shows the screening results for C. burnetii in bovine bulk milk samples. In total, 17 of 359 (4.7%) analysed milk samples were tested positive by nested PCR. The positive samples originated from 8 of 27 (29.6%) milk cow farms. Interestingly, in a number of farms, the agent seemed to persist over time. At least in one farm, C. burnetii could be detected in bi-weekly samples from January and February 2006. The same
Fig. 2. Sensitivity of the Coxiella burnetii-specific nested PCR with food samples after DNA preparation. Lane 1: positive control with 105 C. burnetii particles/ml; lane 2: egg yolk with 103 particles/15 mg; lane 3: egg yolk with 102 particles/15 mg; lane 4: bovine milk with 103 particles/ml; lane 5: bovine milk with 102 particles/ml; lane 6: ovine milk with 103 particles/ml; lane 7: ovine milk with 102 particles/ml; lane 8: caprine milk with 103 particles/ml; lane 9: caprine milk with 102 particles/ml; used marker: 50 bp DNA ladder.
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
limits of 10 particles/ml of bacterial suspension and 1 bacteria/ ml of milk sample. A recent Japanese screening on milk and eggs based on the same nested PCR methodology reported a detection limit of 3.2 × 101 particles of C. burnetii/1 g of egg yolk Hirai et al., 2005). A search in PubMed up to the year 2006 revealed that only in a few studies used PCR to conduct analyses of foodstuffs for C. burnetii. A recent publication from the U.S. indicates a prevalence of C. burnetii of over 94% in bulk tank milk samples from dairy herds in the period from 2002 to 2004. Furthermore, it was reported that the C. burnetii infection might be persistent in the dairy herds, with little temporal or regional variations. Shedding levels of positive cattle were estimated to be 101– 1014 particles of C. burnetii/ml. The bulk tank milk samples of the herd stayed at a level of 102 particles/ml over the whole study period (Kim et al., 2005). The fact that the sensitivity of our detection method was around 103 cells/ml of cows milk may explain the variation of positive and negative results with samples taken from the cheese dairy A (see Fig. 1). The reasons for the local differences in the overall rate samples positive for C. burnetii between the farms and the two cheese dairies remain unclear. Although the dairies were only situated about 30 km apart from each other, the according two cattle populations were strictly separated from each other because of geographical reasons. Our findings with ovine milk samples can only be compared to a recent Turkish study where single milk samples from 400 sheep of 22 flocks were analysed and a positive rate of 3.5% for C. burnetii was found. All positive results were obtained only from flocks with a history of abortion (Ongor et al., 2004). Unfortunately, no veterinary records from the Swiss sheep farms were available. The negative findings were unexpected, since sheep were identified to have caused human cases of coxiellosis in Switzerland. In 1983, a Q-fever outbreak occurred involving 415 serologically confirmed persons. The origin of infections were 12 flocks of sheep with apparently ill or C. burnetii shedding animals descending from the alpine pastures to the valley and passing through villages (Dupuis et al., 1987). In Japan, Tatsumi et al. (2006) tested 4.2% of shell eggs and 17.6% of mayonnaise samples positive for C. burnetii by PCR. Based on these findings, we conducted the first screening of shell eggs for C. burnetii outside of Japan by using the same detection methodology as the Japanese authors. But in contrast to their findings, no PCR-positive results were obtained. This is in line with the results of another Japanese screening of chicken eggs Hirai et al., 2005). Finally, a third Japanese study could not reproduce the detection of C. burnetii in mayonnaise (Sadamasu et al., 2006). The discrepancy between the three Japanese studies might be explained by the fact that the analysed eggs were produced in different regions of the country. Since we could not detect C. burnetii in shell eggs, it can be concluded that the epidemiological situation in Switzerland, and in countries from which eggs were imported, must be different to the situation in Japan. This conclusion is also supported by the fact that we used the same detection system as Tatsumi et al. (2006). Surprisingly, all sheep and goat milk samples were shown to be negative for C. burnetii. In future investigations,
417
these findings should be verified with most recently published PCR methods of higher sensitivity. The fact that the target micro-organism was demonstrated in the milk of both randomly selected cheese dairies, representing 27 milk farms, indicates a higher prevalence of C. burnetii in the Swiss cattle population than expected by epidemiological data. The morbidity for human coxiellosis is very low in Switzerland. In 1998, only 11 cases were registered within a population of 7,123,500 persons (Swiss Federal Office of Public Health, 1999). For this reason, mandatory reporting of human coxiellosis was stopped in the following year. Animal cases of coxiellosis are also quite rare. In the year 2005, only 30 cases with cattle, 5 with sheep and 5 with goats were reported (Swiss Federal Veterinary Office, 2006). The present study shows that C. burnetii occurs quite commonly in Swiss cattle. If this finding can be verified with more extensive prevalence studies, the legal regulations to control coxiellosis in cattle herds should be re-evaluated. Actually, the Ordinance on Epizootic Diseases still considers coxiellosis of animals as a disease to be controlled (Federal Department of Economic Affairs, 1995). Such a classification is discussible since C. burnetii seems to be widely spread in the cattle population and obviously has low clinical impact both in human and veterinary medicine. Acknowledgements The authors would like to express their gratefulness to Professor Dr. Noriyuki Tatsumi, Director, Dr. Ying Qiao and Dr. Ikkyu Yamamoto of the Institute for Preventive Medicine against Zoonosis (IPMZ), Komoridani, Oyabe City, Toyama Prefecture, Japan, for transferring the methodology to analyse chicken market eggs for C. burnetii. Our thanks also include Dr. Stefanie P. Templer, Swiss Federal Office of Public Health (SFOPH), for her support in PCR-analysis. References Arricau-Bouvery, N., Rodolakis, A., 2005. Is Q fever an emerging or reemerging zoonosis? Veterinary Research 36, 327–349. Berri, M., Arricau-Bouvery, N., Rodolakis, A., 2003. PCR-based detection of Coxiella burnetii from clinical samples. Methods in Molecular Biology 216, 153–161. Dupuis, G., Petite, J., Peter, O., Vouilloz, M., 1987. An important outbreak of human Q fever in a Swiss Alpine valley. International Journal of Epidemiology 16, 282–287. Federal Department of Economic Affairs, 1995. Tierseuchenverordnung (TSV) vom 27. Juni 1995, Stand am 2. Mai 2006 [in German]. Available at: http:// www.admin.ch/ch/d/sr/9/916.401.de.pdf. Hirai, A., Kaneko, S., Nakama, A., Ishizaki, N., Odagiri, M., Kai, A., Sadamasu, K., Shinkai, T., Yano, K., Morozumi, S., 2005. Investigation of Coxiella burnetii contamination in commercial milk and PCR method for the detection of C. burnetii in egg. Shokuhin Eiseigaku Zasshi 46, 86–92. Hirai, K., To, H., 1998. Advances in the understanding of Coxiella burnetii infection in Japan. Journal of Veterinary Medical Science 60, 781–790. Kim, S.G., Kim, E.H., Lafferty, C.J., Dubovi, E., 2005. Coxiella burnetii in bulk tank milk samples, United States. Emerging Infectious Diseases 11, 619–621. Lorenz, H., Jäger, C., Willems, H., Baljer, G., 1998. PCR detection of Coxiella burnetii from different clinical specimens, especially bovine milk, on the basis of DNA preparation with a silica matrix. Applied and Environmental Microbiology 64, 4234–4237.
418
R. Fretz et al. / International Journal of Food Microbiology 116 (2007) 414–418
Maurin, M., Raoult, D., 1999. Q fever. Clinical Microbiology Reviews 12, 518–553. Metzler, A.E., Nicolet, J., Bertschinger, H.U., Bruppacher, R., Gelzer, J., 1983. Distribution of Coxiella burnetii: a seroepidemiological study of domestic animals and veterinarians [in German]. Schweizer Archiv für Tierheilkunde 125, 507–517. Ongor, H., Cetinkaya, B., Karahan, M., Acik, M.N., Bulut, H., Muz, A., 2004. Detection of Coxiella burnetii by immunomagnetic separation-PCR in the milk of sheep in Turkey. Veterinary Record 154, 570–572. Parker, N.R., Barralet, J.H., Bell, A.M., 2006. Q-fever. Lancet 367, 679–688. Sadamasu, K., Tabei, Y., Shinkai, T., Hasegawa, M., Kaneko, S., Hirai, A., Nakama, A., Ishizaki, N., Odagiri, M., Kamata, S., Yano, K., Kai, A., Morozumi, S., 2006. Development of effective detection method for Coxiella burnetii in mayonnaise by real-time PCR and investigation of C.
burnetii contamination in commercial mayonnaise in Tokyo. Shokuhin Eiseigaku Zasshi 47, 1–8. Swiss Federal Office of Public Health, 1999. Weekly Bulletin no. 1, p. 6. Swiss Federal Veterinary Office, 2006. Weekly Bulletin no. 1, p. 12. Tatsumi, N., Baumgartner, A., Qiao, Y., Yamamoto, I., Yamaguchi, K., 2006. Detection of Coxiella burnetii in market chicken eggs and mayonnaise In: Hechemy, K.E., Oteo, J.O., Raoult, D.A., Silverman, D.J., Blanco, J.R. (Eds.), Century of Rickettsiology — Emerging, Reemerging Rickettsioses, Molecular Diagnostics, and Emerging Veterinary Rickettsioses. Annals of the New York Academy of Sciences, vol. 1078, pp. 502–505. Zhang, G.Q., Nguyen, S.V., To, H., Ogawa, M., Hotta, A., Yamaguchi, T., Kim, H.J., Fukushi, H., Hirai, K., 1998. Clinical evaluation of a new PCR assay for detection of Coxiella burnetii in human serum samples. Journal of Clinical Microbiology 36, 77–80.