Abstract
Objective:
The aim of the present work was to determine the presence of various rickettsiae and protozoan in raw milk and the assessment the potential, milk-borne route in the spread of selected zoonotic pathogens.
Materials:
A total of 119 raw milk samples collected randomly from 63 cows, 29 goats, and 27 sheep bred on 34 farms situated on eight communities in eastern Poland were examined by polymerase chain reaction (PCR) method for the presence of pathogenic rickettsiae (Coxiella burnetii, Anaplasma phagocytophilum, and Rickettsia spp.) and protozoan (Toxoplasma gondii).
Results:
The only prevalent pathogen was T. gondii, which was found in 10 samples of cow milk (15.9%), in one sample of goat milk (3.4%), and in one sample of sheep milk (3.7%). One sample of cow milk was positive for C. burnetii; however, the sequence analysis did not confirm any species of Coxiella or Coxiella-like organisms, but showed 100% homology to Psychrobacter alimentarius. None of the examined samples showed the presence of A. phagocytophilum or Rickettsia spp.
Conclusions:
The results of this study suggest a potential hazard of milk-borne Toxoplasma infection, mostly by consumption of raw cow milk. The milk-borne spread seems to be limited or nonsignificant in the case of C. burnetii, A. phagocytophilum, and Rickettsia spp. The false-positive sample for Coxiella spp. suggests that some care should be taken in the interpretation of the results obtained by using the PCR method.
Introduction
T
In contrast to TBEV and C. burnetii, the role of milk-borne infections in human by other zoonotic pathogens, including Anaplasma phagocytophilum, Rickettsia spp., and Toxoplasma gondii have been less studied.
A. phagocytophilum is a rickettsial species causing in humans granulocytic anaplasmosis, a tick-borne disease transmitted in Europe by Ixodes ricinus. In domestic ruminants it causes tick-borne fever (bovine ehrlichiosis), which may result in decreased milk yield, abortions, reduced fertility, and mastitis (Matsumoto et al., 2006; Torina and Caracappa, 2007). To date, there are only few studies on the potential role of the milk route in the spread of A. phagocytophilum (Zhang et al., 2016) and Rickettsia spp. classified as Spotted Fever Group Rickettsiae (SFGR), including Rickettsia rickettsii, Rickettsia conorii, Rickettsia raoultii, Rickettsia helvetica, Rickettsia slovaca, Rickettsia japonica, and other species also transmitted by ticks, mainly belonging to Dermacentor and Rhicephalus genera (McQuiston et al., 2012).
T. gondii is an apicomplexan protozoan widely distributed among animals and humans (Sroka et al., 2007; Dubey et al., 2014). It is assumed that it is transmitted mainly by the ingestion of raw or undercooked meat or by food contaminated with oocysts excreted by cats (Dubey, 1991). The consumption of unpasteurized milk from infected animals (especially from sheep and goats) is considered as a less common route for acquiring toxoplasmosis in humans (Bezerra et al., 2015).
Studies by Cusato et al. (2013, 2014) confirmed that the application of hazard analysis and critical control points and good manufacturing practices in a small dairy plant was adequately cost beneficial, could improve food safety, and a reduction in the yeast and mould count. The safety of food, especially of raw products, needs continuous monitoring of its handling and production, and training of personnel directly involved in food manufacturing (Gomes et al., 2014).
Considering the existing gaps in the knowledge of a potential significance of the milk-borne route in the spread of various rickettsial and protozoan zoonotic pathogens, this small-scale study was undertaken with the aim of evaluating, by the polymerase chain reaction (PCR) method, the presence of C. burnetii, A. phagocytophilum, Rickettsia spp. classified as SFGR, and T. gondii DNA in randomly selected samples of raw milk of cows, goats, and sheep.
Materials and Methods
Collection of milk samples
A total of 119 raw milk samples were collected directly after milking in 2009 and 2010 from clinically healthy 63 cows, 29 goats, and 27 sheep bred on farms situated on the territory of the Lublin Province in eastern Poland. The samples of cow milk were collected from 31 dairy farms, samples of goat milk from two farms, and samples of sheep have been raised from one farm. The samples were taken from the animals pastured on meadows located near deciduous and mixed forests. After milking, the samples in amount of 200 mL were centrifuged (1000 × g for 10 min at 0°C) according to Blackwell et al. (1982) and the fraction of skimmed milk stored at −80°C until DNA isolation and PCR investigation in 2016.
DNA isolation
Total DNA was isolated from 200 μL of each fraction of skimmed milk sample using the QIAmp DNA Mini Kit—protocol for blood and body fluids (Qiagen). The concentration of extracted DNA measured with NanoDrop ND1000 spectrophotometer (USA) was in the range of 5–15 ng/μL per sample. From each skimmed milk sample, at least four DNA isolates were prepared.
Detection of pathogens by PCR
The isolates were examined for the presence of C. burnetii, A. phagocytophilum, Rickettsia spp., and T. gondii by PCR.
Detection of C. burnetii DNA was based on amplification of insertion sequence with primers IS1111f and IS1111r described by Rolain and Raoult (2005) and Subramanian et al. (2012). Each PCR was carried out in a 25 μL reaction volume, which contained the following mix of reagents: 1 U Biotools DNA polymerase (Biotools B&M LABS, S.A., Spain), 1 × PCR buffer containing 20 mM MgCl2, 2.5 μL 2 mM dNTP (final concentration 0.2 mM) (Thermo Scientific), 1.25 μL 10 μM each of primer, 2.5 μL of matrix DNA, and nuclease-free water (Applied Biosystems). The amplification was carried out in C1000 Thermal Cycler (Bio-Rad) under the following conditions: preincubation at 95°C for 3 min, 35 cycles, each of 30 s at 95°C (denaturation), 30 s at 58°C (primers annealing), and 1 min at 72°C (elongation). The size of amplified DNA fragment was about 150 bp. Additionally, all samples were investigated with primer for 16S rDNA of Coxiella-like organisms according to Duron et al. (2014). As a positive control, DNA extracted from the antigen of C. burnetii on the substrate slide used for detection of anti-Coxiella antibodies (MRL Diagnostics) was used, whereas nuclease-free water was used as a negative control.
Detection of A. phagocytophilum was performed according to Massung et al. (1998) using amplification by PCR and confirmatory reamplification by nested-PCR. Each PCR was carried out in a 25 μL reaction volume, which contained the following mix of reagents: 1.25 U Taq DNA polymerase (Qiagen), 1 × PCR buffer containing 15 mM MgCl2, 2.5 μL 2 mM dNTP (final concentration 0.2 mM) (Thermo Scientific), 1.25 μL 10 μM each of primer for the first amplification (ge3a/ge10) or 0.5 μL 10 μM each of primer for nested-PCR (ge9/ge2), 2.5 μL of isolated DNA, and nuclease-free water (Applied Biosystems). For nested-PCR reaction, 1 μL from the primary PCR product was added. The size of amplified DNA fragment was 546 bp. DNA extracted from the A. phagocytophilum antigen, coating the substrate slide used for detection of anti-A. phagocytophilum antibodies (Focus Diagnostic), was used as a positive control, whereas nuclease-free water was used as a negative control.
In both cases, products of amplification were identified in agarose gel as described previously (Wójcik-Fatla et al., 2015).
Rickettsia spp. detection was performed as described previously by Wójcik-Fatla et al. (2013) using primers RpCS.887p and RpCS.1258n specific for a gene encoding the citrate synthase gene gltA. DNA isolated from antigen Rickettsia spp. on the substrate slide (Fuller Laboratories, CA) was used as a positive control.
Detection of T. gondii DNA was based on the amplification of B1 fragment gene according to Grigg and Boothroyd (2001) with modification described previously by Wójcik-Fatla et al. (2015). For preliminary identification of the genotype (I or II/III) of the isolated T. gondii strains, restriction fragment length polymorphism (RFLP)-PCR was performed, where the amplified B1 products were digested with restriction enzymes Eco 721 (PmII) and XhoI (Thermo Scientific). All PCRs were performed using positive and negative control as described in the abovementioned articles.
Psychrobacter spp. detection was performed using primers 27f and 1492r to target 16S rRNA gene (Moghadam et al., 2016). Species of Psychrobacter was determined by sequencing of 16S rRNA gene.
Sequence analysis of PCR products
DNA sequencing of all respective amplicons was performed with ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Inc., Foster City, CA) using an ABI PRISM Big Dye Terminator v. 3.1. Cycle Sequencing Kits and Big Dye XTerminator Purification Kit (Applied Biosystems). Sequence analysis of amplicons was accomplished with primers used for PCR described above. Only the sample positive with Coxiella primers was additionally analyzed by using M13 (−21) universal forward primer 5′-TGT-AAA-ACG-ACG-GCC-AGT-3′. The results were compared with sequences in the GenBank database using the BLAST server at the National Center for Biotechnology Information (Rockville Pike, Bethesda, MD).
Results
DNA fragments of A. phagocytophilum and Rickettsia spp. were not detected in the examined fraction of skimmed milk samples. Amplification with primers for C. burnetii gave positive result in one sample of cow milk, but the product was somewhat smaller (about 140 bp) than the positive control (154 bp). Sequence analysis of the insertions sequence IS11111 with the same primer set and additionally with universal primers did not confirm the presence of C. burnetii, but showed 100% homology to partial of 16S ribosomal RNA gene of Psychrobacter alimentarius strain PAMC 27889 (Accession No. CP014945.1).
The DNA of T. gondii was stated in 10 out of 63 cow milk samples (15.9%), in 1 out of 29 goat milk samples (3.4%), and in 1 out of 27 sheep milk samples (3.7%). On each of three farms, T. gondii was detected in two milk samples of cows, and on four farms, the presence of protozoan occurred individually. Totally, in 12 out of 119 examined milk samples (10.1%), the B1 gene fragment of T. gondii was identified (Accession No. LN714499.1). RFLP analysis of B1-positive samples showed that five Toxoplasma isolates from milk belonged to Type II or III (digested by Eco 721 and XhoI), whereas seven isolates belonged to Type I (not digested by enzymes).
Discussion
The results of this study suggest a potential hazard of milk-borne Toxoplasma infection, mostly by consumption of raw cow milk. The milk-borne spread seems to possess no significance in the case of C. burnetii, A. phagocytophilum, and Rickettsia spp.
At present, the role of the consumption of raw bovine milk in the spreading of toxoplasmosis is a matter of controversy. So far, T. gondii has been found mostly in caprine and ovine milk. The presence of the parasites was detected in 2.1–10.0% of the samples of goat milk in Brazil (Bezerra et al., 2015; da Silva et al., 2015), Italy (Mancianti et al., 2013), and Iran (Dehkordi et al., 2013) as well as in 3.4–5.0% of the samples of sheep milk in Italy (Fusco et al., 2007), Brazil (Camossi et al., 2011), and Iran (Dehkordi et al., 2013). Surprisingly, there are only few studies on the occurrence of T. gondii in the samples of raw bovine milk, which is often consumed worldwide. As early as 1953, Sanger et al. (1953) reported the presence of toxoplasmas in the milk of cows with clinical toxoplasmosis. More recently, Dehkordi et al. (2013) detected T. gondii in 3.5% of raw cow milk samples collected in Iran.
To date, milk-borne cases of clinical toxoplasmosis in humans have been related only with the consumption of raw caprine milk (Sacks et al., 1982; Skinner et al., 1990). The role of milk consumption as a general risk factor increasing the probability of acquiring toxoplasmosis in humans is equivocal in the light of epidemiological studies performed in various countries: Poland (Paul, 1998), USA (Jones et al., 2009), Iran (Fouladvand et al., 2010), and Brazil (da Silva et al., 2014), which indicated a statistically significant association of raw milk consumption with Toxoplasma infection, whereas other studies performed in Egypt (Elsheikha et al., 2009), Brazil (Santos et al., 2009), Kyrgyzstan (Minbaeva et al., 2013), and Mexico (Alvarado-Esquivel et al., 2013) have not revealed such a relationship.
Some authors, such as Dubey (1991), tend to underestimate the role of drinking cow's milk in the epidemiology of toxoplasmosis, whereas others (Tenter, 2009; Boughattas, 2015a) postulate that any type of milk cannot be excluded as a potential source of infection, particularly if consumed raw. The latter view is much more reliable, all the more so because there are serious gaps in studies on the occurrence of Toxoplasma in milk samples of various animals, mostly cows (Boughattas, 2015b). Against the underestimation of bovine milk as a potential way of Toxoplasma spread there are the results of the study by Hiramoto et al. (2001), who demonstrated the long persistence of the T. gondii cysts in artificially infected bovine milk and homemade fresh cheese at the refrigerator temperature.
The results obtained in the current work, which to the best of the knowledge of the authors, present the highest prevalence of T. gondii in milk samples from healthy cows ever reported, and provide further support for the potential role of bovine milk in the transmission of toxoplasmosis. Due to the small amount of farms with goats and sheep, the present study could not ascertain if the presence of T. gondii in cow milk is higher than in goat or sheep milk in eastern Poland. To solve this important problem unequivocally, further studies involving a much larger number of milk samples are needed. For now, avoiding drinking raw milk is recommended to prevent Toxoplasma infection.
Unpasteurized milk or cheeses have been generally considered to be microbiologically unsafe because it is possible for pathogenic bacteria such as Escherichia coli, Listeria monocytogenes, Salmonella spp., or Staphylococcus aureus to contaminate dairy products on the farms or post cheese making (Yoon et al., 2016). In earlier studies in Estonia, Kalmus et al. (2015) showed that raw milk intended for direct consumption cannot be considered microbiologically safe without heat treatment, and according to obtained results (36% of L. monocytogenes in-line milk filters), the official criteria for raw milk should be improved. There is no study confirming whether or not contamination of raw milk and raw milk products by T. gondii is possible.
Out of four pathogens examined in the presented study, a milk-borne route of transmission was studied to a greatest extent in the case of C. burnetii, a rickettsial species causing Q fever in humans and coxiellosis in animals (Porter et al., 2011), which could be evoked by the ingestion of raw milk (Signs et al., 2012). The prevalence of C. burnetii DNA in individual samples of cow milk collected in The Netherlands, Hungary, and France was 8.7%, 8.7%, and 18.5%, respectively (Guatteo et al., 2007; Muskens et al., 2011; Gyuranecz et al., 2012). These results are similar to those obtained by Kampen et al. (2012) in Norway who did not find the presence of this bacterium in samples of bovine, caprine, and ovine milk.
So far, a milk-borne route of infection with the tick-borne pathogen A. phagocytophilum has not been considered to any great extent. However, such a route is possible in the light of the study by Pusterla et al. (1997) who identified A. phagocytophilum in 1–5% of neutrophils in the milk of experimentally infected cows. Also, the occurrence of mastitis in ruminants with clinical granulocytic anaplasmosis increases the probability of such a route. Recently, Zhang et al. (2016) detected A. phagocytophilum by PCR in 2.5% of caprine milk samples in China, whereas examination of the ovine milk samples gave negative results. A previous study on the presence of A. phagocytophilum in milk samples suggests a low risk of infection by the milk-borne route. Nevertheless, this suggestion requires confirmation by the examination of a bigger number of samples.
To the best of the authors' knowledge, the presented study is the first in which milk samples collected from healthy ruminants were examined for the presence of Rickettsia spp. from the SFGR. To date, the only report indicating the possibility of milk-borne transmission of the SFGR species is that by Ducroiset (1967) who described two cases of R. conorii fever in cheese producers, presumably due to the ingestion of raw milk. The negative results obtained in this study do not confirm the possibility of the transmission of SFGR by milk, but this presumption needs confirmation on a larger amount of milk samples.
The genus of Psychrobacter involves Gram-negative, environmental, psychrophilic, and halotolerant bacteria, found mostly in low-temperature marine environments (Borsodi et al., 2010). They have been described as rare opportunistic pathogens in humans causing meningitis (Le Guern et al., 2014).
Furthermore, Psychrobacter spp. was found in dairy products, such as cheeses and milk (Schmidt et al., 2012; Delcenserie et al., 2014). The presence of P. alimentarius in one sample of raw milk in this study could be explained by a possible fecal contamination, considering the detection of Psychrobacter genus in fecal samples from dairy cows (Kaevska et al., 2016). The PCR method for detection of Coxiella used in this study needs confirmation. Considering the results of the study, the positive samples obtained by PCR method for detection of Coxiella need to be confirmed by sequencing. The potential pathogenicity and circulation of Psychrobacter spp. in the environment remains unknown and needs further investigations.
Conclusion
In conclusion, the results of the present study suggest a severe hazard of T. gondii infection by the consumption of raw cow milk. However, the obtained results do not suggest the possibility of a milk-borne route of infection in the case of examined tick-borne pathogens C. burnetii, A. phagocytophilum, and Rickettsia spp., classified as SFGR, but this presumption needs confirmation on a greater number of milk samples. The influence and significance to human health of Psychrobacter spp. detected in a milk sample is still unknown. The consumption of raw milk is currently increasing due to several beneficial aspects, however, potential risk could occur due to the presence of some zoonotic pathogens in raw milk, although this hypothesis needs further investigation.
Footnotes
Disclosure Statement
No competing financial interests exist.
