INTRODUCTION
Chlamydiae are a frequently encountered intracellular eubacteria with a unique biphasic developmental cycle, whose cell wall resembles that of Gram-negative bacteria [Reference Barbour1–Reference Wyrick3]. Within the Chlamydiaceae family, nine distinct species have been found – Chlamydia trachomatis, C. muridarum, C. suis, C. abortus, C. caviae, C. felis, C. pecorum, C. pneumoniae, and C. psittaci [Reference Sachse4]. Of these, the species commonly detected in pigs include C. suis, C. abortus, C. psittaci, C. pecorum, and C. trachomatis [Reference Kauffold5, Reference Lis6]. In pigs, these bacteria in combination with Mycoplasma spp., porcine reproductive and respiratory syndrome virus (PRRSV), porcine circovirus type 2 (PCV-2), Pasteurella multocida, or Streptoccocus spp. contribute to the induction of a multi-etiological syndrome [Reference Rodolakis and Yousef Mohamad7–Reference Carrasco10]. One of the major characteristics of Chlamydia spp. is their ability to cause prolonged, often subclinical infections [Reference Hammerschlag11]. This syndrome can manifest with a wide range of symptoms, from a well-developed clinical infection to an asymptomatic infection, which occurs most often. Clinical symptoms manifest under the influence of certain conditions, such as stress, intensive pig production systems, or immunity deficiency. In pigs, Chlamydia contributes to disorders of the respiratory, digestive, and reproductive systems [Reference Bagdonas12–Reference Englund14]. Epidemiological studies conducted in European countries have shown the importance of Chlamydia infection; however, in case of reproduction problems in pig farms, veterinarians rarely suspect Chlamydia infection. In such epidemiological studies the basic research panel includes PRRSV, PCV-2, parvovirosis, leptospirosis and swine influenza (SI) [Reference Schautteet8, Reference Carrasco10, Reference Bagdonas12, Reference Truszczyński and Pejsak15].
The pig population in Poland amounts to almost 11 million animals, this includes almost 980 000 sows. The number of herds with sows/gilts on Polish territory is about 190 000 units. Regional distribution of pig herds is irregular but it can be concluded that the western part of the country is characterized by a greater number and density of pig production [Reference Dors16].
Pig production in Poland is a major branch of the economy. The large population of pigs, despite the irregular distribution of herds, promotes the spread of pathogens between farms and influences the pathogen prevalence in the general population. The presence of pathogens, even in the form of subclinical infection, has significant impact on the economic results of animal production.
Evaluation of the level of Chlamydia presence in pig herds will allow assessment of the scale of the problem and help in the implementation of preventive procedures.
Aim of the study
The aim of this study was to estimate the prevalence of Chlamydia spp. infection in Polish pig herds.
MATERIALS AND METHODS
Sample size
To determine the number of infected herds in our study population we used a list obtained from the Registration and Identification Animal System of the Agency for Restructuring and Modernization of Agriculture. This list includes approximately 380 000 pig farms [Reference Dmochowska17]. The number of farms necessary for us to evaluate in order to accurately represent Chlamydia infection was determined based on equation (1) (with a confidence level of 0·95) [Reference Thrusfield18]:
$$n = \displaystyle{{1\!\cdot\!96^2 \times P_{\exp} \times (1 - P_{\exp} )} \over {d^2}}, $$where n is the required sample size, P exp the expected morbidity, and d represents measurement accuracy.
Due to the lack of prevalence data for the pig population in Poland, the expected prevalence was arbitrarily set at 50% (P exp = 0·5) [Reference Thrusfield18]. Measurement accuracy was arbitrarily set to 10% (d = 0·1). Based on these parameters, the calculated sample size necessary to evaluate the epizootic situation of chlamydiosis in the Polish pig population was 97 herds. The number of sows/gilts examined in each herd was calculated according to equation (2):
$$\int n = [1 - (1 - p_1 )^{1/d} ] \times \left(N - \displaystyle{d \over 2}\right) + 1,$$where n is the number of examined sows/gilts in the herd; p 1 is the probability of detecting at least one infected sow/gilt; N represents the total number of sows/gilts in the herd; and d represents the number of infected animals in the herd. The probability of detecting at least one infected sow in the herd was 95% (p 1 = 0·95). Due to the lack of reference data regarding the prevalence of Chlamydia infection in pig herds in Poland, we arbitrarily set the herd prevalence at a low level of 30%. Based on these parameters we estimated the number of infected animals in each herd (d) and used that value to calculate the appropriate number of animals to examine (n). Samples were collected from randomly selected farms across the country. Polish pig production is mainly performed in closed indoor systems, therefore it was decided to analyse the results taking into account the farm size. The pig herds were considered separately based on the number of sows. A small herd was defined as a stock with ⩽20 sows, a medium herd had 21–120 sows, and a large herd had >120 sows.
After random selection of herds and obtaining information from the owners regarding basic herd size, the number of samples to be collected from animals in order to detect Chlamydia was calculated using using the WinEpi program (www.winepi.net). In small herds (⩽20 sows) seven samples were collected, in medium herds (21–120 sows) nine samples were collected and in large herds (>120 sows) nine samples were collected. The total number of samples taken from 97 herds was 797.
In the following study, data of biosecurity levels of pig farms were considered. Information regarding biosecurity levels was obtained from records of Veterinary Inspection, which carries out the periodic control of animal farms. During every year a control veterinary inspector complete a survey regarding the conditions of animal maintenance, welfare and biosecurity procedures. Collecting and comparing annual data allows us to see if biosecurity procedures in animal farms are effective or not.
Sample collection
Animals were sampled at random according to the following schedule: all animals were marked sequentially using PAINT-FARM SPRAY (Vetos-Farma, Poland). The numbers were written on separate animal record cards which were drawn at random to determine which animal was sampled. The record cards were returned to the container after selection. In the case of selecting the same random number, the record card was returned to the container, and the draw continued until the selection of an untested animal.
We collected swabs from the vagina (P) and conjunctival sac (O) of sows and gilts. Swabs from the conjunctival sac were collected to determine whether the pathogen was present outside the reproductive system. Vaginal swabs from clinically healthy animals were collected during oestrus or up to 5 days after delivery, when the volume of vaginal secretions is high. Vaginal swabs from animals with clinical reproductive disorders were collected regardless of the phase of their cycle or the physiological state of the sow or gilt. In addition, when possible, serum samples were also collected to investigate the presence of antibodies against Chlamydia spp. In total, 797 vaginal samples, 797 conjunctival samples, and 235 serum samples were collected and tested.
Immediately after samples were collected swabs were placed in transport containers (Equimed, Poland) and transported to a diagnostic laboratory (Faculty of Veterinary Medicine, Wroclaw) at a temperature of 4–8 °C. The laboratory implements a quality management system (ISO/IEC 17025:2005 + API:2007 + AC:2007).
Sampling began in autumn 2011 and continued until 2014. The number of herds tested each year was similar and amounted to 25 surveyed units in 2011, and 24 in each subsequent year.
DNA isolation
Bacterial genetic material was isolated directly from the swab using a DNA Syngen Tissue Mini kit (Syngen, Germany). Isolated DNA (20 µl) was either immediately used for real-time polymerase chain reaction (PCR) or stored at −20 °C.
Real-time PCR
To amplify the gene coding the 23S ribosomal mRNA subunit we used the primers TQF (5′-GAAAAGAACCCTTGTTAAGGGAG-3′) and TQR (5′- CTTAACTCCCTGGCTCATCATG-3′). The sequence of the probe was FAM-CAAAAGGCACGCCGTCAAC-TAMRA. The primers and probe used were specific for all nine species of the Chlamydiae family. The 15 µl reaction mixture contained 10 µl KAPA PROBE FAST Bio-Rad iCycler 2× qPCR master mix (Kapa Biosystems, USA), 0·2 µl of probe (10 µm), 0·3 µl of each primer (10 µm), and 4·5 µl water. To this reaction mixture we added 5 µl of the test matrix. Amplification was performed using a thermal cycler iQ5 Bio-Rad (Bio-Rad, USA) according to the following protocol: initial denaturation, 3 min at 94 °C, amplification over 40 cycles of 15 s at 95 °C and 1 min at 60·5 °C. To assess the analytical sensitivity of our assay we used DNA isolated from C. suis VR-1474 (ATCC, USA) starting with a titre of 5 × 106 bacterial cells/ml and performed sequential tenfold dilutions. For all reactions the standard C. suis strain VR-1474 (ATCC, USA) was used as a positive control [K(+)] at a dilution of 10−3.
Optimization and sensitivity evaluation of real-time PCR
In order to optimize our real-time PCR protocol, we used a standard C. suis strain VR-1474 (ATCC, USA) and a standard C. felis 905 (Merial, France) strain. Using these standard samples we performed real-time PCR with a gradient range of annealing temperatures from 55 °C to 66 °C, and determined that the optimal annealing temperature was 60·5 °C. Our assessment of sensitivity was performed by repeating the real-time PCR with tenfold dilutions of C. suis VR-1474 (ATCC, USA) with a titre of 5 × 106 cells. Using this method, our limit of detection was a 10−6 dilution, which equates to five copies of DNA.
DNA sequencing
The products obtained from samples positive in real-time PCR were sequenced (Genomed, Poland) and identified using BLAST (blast.ncbi.nlm.nih.gov).
Serological examination
To assess the prevalence of antibodies present in the serum we used the complement fixation test (CFT). The CFT is a quantitative test that detects the level of anti-C. psittaci, anti-C. abortus and anti-C.suis antibodies. Antibodies specific to Chlamydia spp. were detected using CFT according to EN ISO/IEC-17025:2005. The CFT was performed in the National Veterinary Research Institute (NVRI) in Puławy.
Statistical analysis
Statistical analysis was performed using Statistica v. 10 (StatSoft, USA) and Excel Microsoft (Microsoft Corp., USA) software. The nominal data collected (the number of farms according to certain criteria) are shown in Tables 1 and 2. The calculated relative risk (RR) of Chlamydia occurrence and the associated 95% confidence intervals (CI) are shown in Table 3, and were assessed using a χ 2 statistic with 1 d.f., at an arbitrary confidence level of P < 0·05.
Distribution of farms surveyed by the presence of Chlamydia suis in the conjunctival sac (O) and vagina (P)
CI, Confidence interval.
+, positive test result; –, negative test result.
Presence of Chlamydia in the conjunctival sac (O) and vagina (P) stratified by farm size (n = 97)
CI, Confidence interval; ChTR, Chlamydia in reproductive tract.
P < 0·05.
* Differences statistically significant.
Results of multiple comparisons of the presence of Chlamydia suis in different sized herds, and for different samples
RR, Relative risk; CI, confidence interval; ChTR, Chlamydia in reproductive tract.
P < 0·05.
* Differences statistically significant.
Ethics statement
The Ethical Committee for Animal Experiments, Wrocław, Poland approved this study, and all owners provided informed consent prior to initiation of the study.
RESULTS
Herds participating in the study
The study involved 97 pig herds in Poland. The herds were from the provinces of Wielkopolska (42 farms), Opole (12 farms), Podlaskie (10 farms), Małopolska (eight farms), the Lodz (five farms), Silesia (four farms), Lower Silesia (four farms), Mazowieckie (four farms), Lublin (four farms), and Kujawsko-Pomorskie (four farms). The studied farms comprised of 38·1% large herds, 22·7% medium herds, and 39·2% small herds.
Real-time PCR
For our analysis we divided the tested farms into groups depending on the presence of Chlamydia in the vagina and/or conjunctival sac (Table 1).
Without considering the origin of the swab, we found Chlamydia spp. in 71·2% of all tested farms. The percentage of animals testing positive on any given farm varied from 20% to 100%. Specifically, in the reproductive tract, we found Chlamydia spp. in 61·9% of the tested farms. In examining the prevalence of Chlamydia based on farm size, we found it in the vaginal swabs in 56·8% of large farms, 81·8% of medium farms, and 55·3% of small farms (Table 2).
While the proportion of vaginally infected pigs from medium-sized farms was higher than large and small farms, this difference was not statistically significant (P = 0·093).
Results of multiple comparison of proportion taking into account the farm group and farm size is shown in Table 3.
The proportion of O(–)P(+) farms in the large farm group was significantly higher than in small farms (18·9% vs. 0·0%, P = 0·016). The RR of Chlamydia occurrence in pigs kept in large farms is significantly higher than in pigs from small- and medium-sized farms (RR 5·68, 95% CI 1·24–25·9, P = 0·027). The probability of lack of Chlamydia in medium farms is less than in small farms (9·1% vs. 44·7%, P = 0·010). The risk of obtaining the O(–)P(+) results in large farms is more than five times higher than in medium and small farms.
We sequenced 26 samples that were positive for Chlamydia spp. by real-time PCR. All samples were identified as the species C. suis, except for one vaginal sample that was identified as C. pecorum. Notably, the sow whose vaginal swab was identified as C. pecorum also had antibodies specific for Chlamydia spp. in the blood.
Thirty-two herds were randomly selected in the first stage of sampling from the high percentage of herds in which Chlamydia was present. The number of sows/gilts from which serum samples were taken in each herd was calculated according to equation (2). Those were the same animals from which swabs were taken – individual numbers identified all animals. A total of 235 serum samples were collected and tested.
From all of these serum samples only two contained antibodies against Chlamydia spp. In these two sows, which were from the same farm, chlamydiae were present in both the vaginal and conjunctival sac swabs. Otherwise, the CFT did not detect anti-Chlamydia antibodies in the serum even when the pathogen was present in the vaginal and/or conjunctival sac swabs.
To fully assess of the prevalence of Chlamydia, we divided Poland along the eastern borders of Pomerania, Kuyavian Pomerania, and the Łódź and Silesia provinces. This divided the country into its western and eastern parts. Western Poland is characterized by the presence of medium and large pig farms with a high or very high biosecurity level. Eastern Poland, on the other hand, typically has small and very small (backyard) pig stocks with low biosecurity level [Reference Dors16].
The studies found that the presence of P(+)O(+) farms in Western Poland was 48·4% and in Eastern Poland 42·9%. Presence of P(+)O(–) farms in Western Poland was 24·2% compared to 5·7% in Eastern Poland. In Western Poland presence of P(–)O(+) farms was 9·7% while in Eastern Poland it was 20%. Farms free of the presence of Chlamydia in vaginal and conjunctival sac swabs [P(–)O(–)] were found in 17·8% of farms in Western Poland and 31·5% in Eastern Poland. The obtained results are shown in Figure 1 regarding the division of the country into western and eastern parts. The results are reported on a background of pig-herd density of >20 sows/1000 ha of agricultural land in each province.
DISCUSSION
Chlamydia spp. infection often occurs without symptoms; however, even in these cases the infection still influences production indicators. Additionally, Chlamydia spp. present a potential zoonotic hazard [Reference Longbottom and Coulter19, Reference Jiang20], therefore determining the prevalence of Chlamydia in pig populations is critical. The first study of Chlamydia prevalence in pigs was reported in 1966 in Great Britain. Capillary agglutination test was used to detect antibodies against Chlamydia spp. Antibodies were found in 23% of sera from tested pigs [Reference Wilson and Plummer21].
Some authors suggest that the type of pig farming may play a role in the prevalence of Chlamydia infection. Utilizing 16S RNA PCR, C. suis was detected in 1·5% of swabs from pigs with high rates of irregular return to oestrus and in 2·3% of swabs from pigs in control herds without reproductive problems. Moreover, C. abortus was detected in 33·3% of the examined fragments of genital tracts [Reference Camenisch22]. Another study on Chlamydia spp. infection was performed on 102 pigs from Germany (intensive farming) and 79 pigs from Switzerland (extensive farming). Both of these groups had a high prevalence of C. suis (90% in Germany, 79% in Switzerland). Additionally, Becker et al. suggest that intensively kept pigs are predisposed to ocular chlamydial infection [Reference Becker23]. Predisposition to ocular infection in intensively kept pigs might be explained among other things by the high density of animals in the farms, environmental conditions, reduction of costs of specific prevention and the pig's anatomy. A low-set head can result in increased emergence of ocular infections.
A similar observation was made in large pig production plants in Estonia where conjunctivitis and reproductive disorders associated with Chlamydia spp. presence were reported [Reference Schautteet8]. At present an assessment of the Chlamydia infection rate in pig herds has been reported for many countries throughout the world [Reference Kauffold5, Reference Schautteet and Vanrompay13, Reference Eggemann24, Reference Xu25].
The presence of Chlamydia spp. in pig herds was found in several European countries [Reference Vanrompay26–Reference Szeredi32]. Belgium, where 96·5% of pigs are seropositive, is considered a country with endemic Chlamydia. In the study, 249 fattening pigs were examined with a recombinant enzyme-linked immunosorbent assay (ELISA), and 240 pigs had Chlamydiaceae family-specific antibodies [Reference Vanrompay26]. Seroprevalence studies in Germany revealed the presence of Chlamydia in 33% of the tested herds. In these studies an ELISA was used to assess 1493 blood samples, and the rate of animals testing positive varied from 4·3% to 72·7% on individual farms. Therefore, the prevalence of chlamydial infections in breeding herds in Germany may be high and may play a role in reproductive disorders that indirectly affect pork production [Reference Eggemann24]. Furthermore, in 2004 a boar population from Thuringia province in Germany was tested, and Chlamydia spp. were detected in 57·1% of these animals. This high level of seroprevalence may suggest that wild fauna are an important reservoir of Chlamydia spp. [Reference Hotzel27]. A study conducted in Switzerland tested sows, older piglets (>4 weeks), and piglets (⩽4 weeks). The study used an LPS-based ELISA assay and reported a seroprevalence level of 6·9%, 48·1%, and 62% in piglets, older piglets, and sows respectively [Reference Camenisch22]. In pig populations in Italy the presence of Chlamydia was between 63·5% and 80·3% in tested herds [Reference Di Francesco28]. A study utilizing the ELISA test conducted in the Murcia region of Spain revealed that the prevalence of Chlamydia reached a level of 47·1% [Reference Buendía29], while in the Toledo region Chlamydia prevalence was reported at 36·4% [Reference Palomino30]. In Lithuania 2502 blood samples from pigs from 24 farms were tested and only 7·7% tested positive by CFT. However, 87·5% of the regions in Lithuania are affected by Chlamydia infection. Similar to observations in Switzerland [Reference Camenisch22], the Lithuanian study also reported a correlation between Chlamydia infection and animals' age, where older animals are more frequently seropositive than young piglets [Reference Bagdonas31], and there is an increase in Chlamydia infection in pigs aged >3–4 weeks [Reference Szeredi32].
Extensive studies regarding the prevalence of Chlamydia have also been carried out in Asia. In 2012 eleven administrative cities in Jiangxi province (southeastern China) were studied using an indirect haemagglutination assay (IHA) on 920 blood samples. The prevalence of Chlamydia ranged from 33·33% (Jingdezhen) to 90·91% (Pingxiang). The positive samples were distributed among all 11 administrative cities. In Guangdong province in southern China Chlamydia prevalence was reported as 63·38% in breeding boars, 41·10% in breeding sows and 36·25% in fattening piglets using an IHA. The differences in the prevalence in different age groups and gender is notable, and the authors suggest that breeding boars may be a source of the infection for other pigs on the farm [Reference Xu25]. In 2010, a study in Tibet examined the prevalence in different age groups and sexes using IHA. They reported an average seroprevalence level of 16·63%. However, sows had a higher seroprevalence (17·61%) than males (12·72%). Additionally, growing animals also had a higher incidence of Chlamydia [Reference Zhang33].
There is insufficient data of Chlamydia prevalence in the American continent. Research conducted in the United States to evaluate whether Chlamydia spp. may be the cause of diarrhoea in pigs reported 15% prevalence in the intestinal tract of tested animals. They concluded that Chlamydia spp. are common in the intestinal tract but not always the cause of clinical symptoms [Reference Nietfeld34].
Until our study, in Poland the presence of Chlamydia spp. had been described in cattle and companion animals [Reference Wieliczko and Płoneczka-Janeczko35, Reference Szymańska-Czerwińska, Niemczuk and Galińska36]. Research on Chlamydia infections using CFT in Polish pigs during 2008–2013 revealed a low level of seroprevalence, i.e. 0·32%. While research conducted by real-time PCR samples collected from pig herds with and without reproductive disorders revealed a high percentage of infected animals [Reference Rypuła37].
Herein, we report a total prevalence of Chlamydia of 71·2%, without taking into account the origin of swabs. The percentage of positive animals on the farm varied from 20% to 100%.
In considering the obtained results we took into account only the farm size because most pig production in Poland is carried out in closed indoor systems with implemented procedures such as artificial insemination.
However, we noted a higher prevalence of Chlamydia in herds located in Eastern Poland. Compared to Western Poland, Eastern Poland is characterized by a lower degree of biosafety in pig production and often suboptimal zoo hygienic conditions. Further, intensive pig farming is more common in the western provinces. For example, ~36% of national pork production is located in Wielkopolska province [Reference Gniot and Karpińska38]. In our study, DNA sequencing only identified C. suis, except for a single case of C. pecorum in a vaginal swab, which is supported by findings in a Swedish study [Reference Englund14].
Chlamydia infection affects the world's pork production [Reference De Puysseleyr39]. While the economic consequences of Chlamydia spp. infections in pig farms are not fully established, we know that reproductive disorders and other syndromes correlated with Chlamydia infection can lead to financial loss as a result of a reduction in pork production or by the need to purchase antibiotic therapy. Therefore, it is critical to conduct research on the prevalence of Chlamydia in pig populations and to continuously monitor the scale of the problem and develop a strategy for treatment and prevention. However, this requires the standardization of diagnostic tests to determine the situation of infection and spread not only in Poland but in other countries. Improvement of the existing situation can be facilitated by sharing experiences and epidemiological information between countries.
The obtained results suggest the ineffectiveness of biosecurity procedures against chronic Chlamydia infections. An attempt to reduce the prevalence of Chlamydia should focus on the ongoing infectious agent monitoring and optional treatment. Constant monitoring will allow carriers to be identified and the deployment of individual approaches to infected pigs.
Distribution of P(+)O(+), P(+)O(–), P(–)O(+) and P(–)O(–) herds in the Western and Eastern parts of Poland with a pig-herd density of >20 sows/1000 ha of agricultural land.
ACKNOWLEDGEMENTS
This study was supported by funds from grant NN 308 5782 40 from the National Committee for Scientific Research Poland.
DECLARATION OF INTEREST
None.
