Mycobacterium avium Subspecies paratuberculosis and Bovine Leukemia Virus Seroprevalence and Associated Risk Factors in Commercial Dairy and Beef Cattle in Northern and Northeastern China

Mycobacterium avium subspecies paratuberculosis (MAP) and bovine leukemia virus (BLV) are important pathogens, commonly responsible for economical loss to cattle farms all over the world, yet their epidemiology in commercial dairy and beef cattle in China is still unknown. Thus, from September 2013 to December 2014, a large-scale seroprevalence study was conducted to determine the seroprevalence and identify herd-level risk factors associated with MAP and BLV infection. The source sample was 3674 cattle from 113 herds in northern and northeastern China. Antibodies against MAP and BLV were detected using ELISA tests. At animal-level, the seroprevalence of antibodies against MAP and BLV was 11.79% (433/3674) and 18.29% (672/3674), respectively. At herd-level, the seroprevalence of antibodies against MAP and BLV was 20.35% and 21.24% (24/113), respectively. Herd size was identified to be associated with MAP infection while herd size and presence of cattle introduced from other farms were significantly associated with BLV infection. Further research is needed to confirm these findings and improve the knowledge of the epidemiology of these two pathogens in these regions and elsewhere in China.


Introduction
Paratuberculosis, also known as Johne's disease caused by Mycobacterium avium subsp. paratuberculosis (MAP), is a chronic infectious granulomatous enteritis of global importance in primarily domestic and wild ruminants [1,2]. This infection can induce significant economic losses in cattle due to reduced milk production and premature culling [3], decreased value at slaughter [4], and eventual death [5].
Other economic losses can also be caused by the presence of paratuberculosis in cattle herds including reduced feed efficiency, decreased milk's fat and protein content, decreased fertility, and increased incidence of mastitis [6].
Most of infected cattle do not develop clinical symptoms. However, nearly one-third of BLV carriers develop a form of the disease known as persistent lymphocytosis (PL), and malignant B-cell lymphosarcomas can only be found in 1-5% of BLV-infected animals [7]. BLV infection has a global distribution, and EBL is an important group of commonly major diseases that have huge implications in economic losses around the world and cause a sanitary barrier to international trade [8]. In some countries, EBL has been successfully uprooted via national control and elimination programs, such as those used in Europe [9,10].
To date, very limited information on the seroprevalence of MAP and BLV infections in cattle is available in China. In order to address this lack of publication issue, the objective of this study was to describe the seroprevalence of MAP and BLV infections in commercial dairy and beef cattle in northern and northeastern China and clarify risk factors associated with seroprevalence of MAP and BLV in the study regions.

Materials and Method
2.1. Ethics Statement. This study was approved by the Animal Ethics Committee of Jilin Agriculture University, China. Cattle of various categories (dairy cattle and beef cattle), from which serum samples were collected, were handled with good ethical animal practices required by the Animal Ethics Procedures and Guidelines of the People's Republic of China. The collection of serum samples was performed as part of routine process of disease monitoring and surveillance program for these cattle. The owners of cattle had given permission for collecting the serum samples.

Serum Samples.
A total of 3674 bovine blood samples were collected from four provinces (Heilongjiang, Jilin, Liaoning, Hebei, and Inner Mongolia Autonomous Region) ( Figure 1) between September 2013 and December 2014. Blood samples were collected from 12 administrative districts and 113 herds with 50 cattle and more across the five regions. Cattle sample sources from each farm were selected randomly using a table of random digits. Several large-scale farms (with more than 150 livestock animals) were not included because the owner was often not present to give permission. In the case of cattle, approximately 10% in each farm were sampled. All of the animals sampled were clinically healthy. Blood samples (10 mL) of each animal were collected by puncturing the tail vessel using sterile tubes sans/without anticoagulant. Samples were kept at 4 ∘ C and centrifuged at 1000 ×g for 15 min. The serum was separated and stored at −20 ∘ C until analysis.

Serological Examination.
The serological examination of MAP was performed by the commercial antibody detection kits (Pourquier-IDEXX ELISA Paratuberculosis Screening and Verification Ab Test) following the recommended protocol. The antibodies to BLV were detected using a commercially available verification ELISA kit (Pourquier, Montpellier, France) following the manufacturer's instructions.

Epidemiological Data Collection.
Data obtained with the epidemiological questionnaires were used in the analysis of risk factors associated with the animal-level and herd-level target prevalence. For animal-level variables, information about species, geographic origin (Heilongjiang/Jilin/Liaoning/Hebei/Inner Mongolia Autonomous Region), age, gender, and abortion history was acquired from farm administrators. For herd-level variables, the respective categories were as follows: region (Heilongjiang/Jilin/Liaoning/Hebei/Inner Mongolia Autonomous Region), herd size (small (50-100)/medium (100-150)/large (>150)), mixed farming (yes: dairy and beef cattle/no: dairy cattle only or beef cattle only), methods of cattle house cleaning (not practiced/sweeping/ water hosing), source of water (well/tap water), management system (intensive: very strict control for irrelevant personnel and animals/semi-intensive: slight control for irrelevant personnel and animals/extensive: no control for irrelevant personnel and animals), presence of cattle introduced from other farms (yes/no), presence of other animals (yes/no), type of production (beef/milk/mixed), and veterinary service (yes/no).

Statistical Analysis.
In this study, an infected farm was defined as a farm with one or more infected animals. Explanatory variables thought to influence the MAP and BLV seroprevalence were in two broad categories: animallevel variables and herd-level variables. Explanatory variables were tested for their relationship with the MAP and BLV seroprevalence by the Mantel-Haenszel Chi-square test and bivariate analyses. Variables were included in the multivariate analysis if they had a value of equal to or less than 0.20 in the bivariate analysis. Region was used as a concomitant variable in the multivariate analysis. Adjusted odds ratio (OR) and its 95% confidence interval were calculated by multivariate logistic regression analysis. A value less than 0.05 was considered statistically significant. Results were analyzed with SPSS 19.0 software package.   18.18% in Inner Mongolia Autonomous Region. The variables selected ( ≤ 0.2) for the multiple analysis were as follows: herd size, methods of cleaning, source of water, management system, presence of cattle introduced from other farms, type of production, and veterinary service. Herd size and presence of cattle introduced from other farms were significantly associated with BLV infection in the studied cattle in the multivariate analysis (Table 4).

Discussion
This is the first large-scale study of MAP and BLV antibody prevalence in commercial dairy cattle and beef cattle in China. High animal-level (11.79%; 95% CI: 7.3-15.4) and herdlevel (20.35%; 95% CI: 30.2-39.1) MAP prevalence was found in northern and northeastern China, whereas high animallevel (18.29%; 95% CI: 7.3-15.4) and herd-level (21.24%; 95% CI: 30.2-39.1) BLV prevalence was also found in the same regions, indicating that the infections are widely spread in the study regions.
In the present study, animal-level MAP seroprevalence was 11.79%, but animal-level prevalence might even be higher. It is important to highlight that paratuberculosis has a comparatively long latent period, and the antibody levels against MAP only can be detected at the end of the latent period by the ELISA test, results to show that this method was noneffective for detecting infected animals in the early stages of the infection [11]. Thus, the animals that are infected at a young stage (i.e., no more than one year) will still exist in the herd, shedding MAP via feces and contaminating water and food, and will only present clinical symptoms in the adult stage [12]. However, paratuberculosis is an unknown disease for most farmers in the study regions, and although the high infection rates at animal-level and herd-level prevalence were found in these regions, most farmers are not aware of the impact of the infection and the economic losses that it can cause.
The seroprevalence of MAP significantly increased with cattle age ( < 0.001) ( Table 3) and was also significantly higher in cattle herds with > 150 cattle compared with smaller herds ( = 0.024). An increase in seropositivity with herd size has also been reported in England [13] and Denmark [14]. This is not surprising because MAP is infective with an environmental component and density dependent effects are to be expected.
One of the critical issues is the probable zoonotic link between MAP and Crohn's disease, although this relation is still a vexed question and is the target of continual debates in science [15]. However, MAP has been detected in fresh and pasteurized milk [16][17][18], and the habit of consuming fresh and pasteurized milk is widespread in family farms in the study regions. Therefore, the validation of a zoonotic link between MAP and Crohn's disease could have great consequences for the cattle industry of the study regions and China.
BLV has a worldwide distribution and prevalence varies between countries [9,10]. In the present study, this is the first report of BLV infection with a 18.29% seroprevalence at animal-level sample in cattle in China, which is lower than the 70% seroprevalence in Argentina [19], 41% in USA [20], and 32.5% in Japan [8]; however, the prevalence of BLV is much higher than the values of 2.28% in dairy cattle in Turkey [21]. Several factors may contribute to such differences, such as geographical conditions, the types and size of the cattle tested, and the specificity and sensitivity of the detection methods.
Analyses of risk factors for the animal-level sample presence of BLV revealed that age was a common risk factor for dairy and beef cattle. The seroprevalence of BLV infection in older cattle was significantly higher than that in younger cattle (Table 3). This is probably due to the fact that older cattle had more chance to contact this pathogen comparing to younger cattle, thus increasing the risk of infection.
Analyses of risk factors for the herd-level presence of BLV revealed that the herd size and presence of cattle introduced from other farms were a common risk factor for dairy and beef farms. Prevalence has an increasing trend with increasing herd size. The reason for this phenomenon is because the population of susceptible animals is greater in large herds. In addition, the most important risk factor for the introduction of not only exotic but also endemic diseases is movement of animals [22]. In China, farm owners rarely test animals for BLV infection ahead of their introduction into their own farms. Resultant, control measures such as negative confirmation of BLV in cattle before their introduction onto a farm should be conducted to prevent BLV infection.
In conclusion, this study firstly reported seroprevalence of MAP and BLV infection in commercial dairy and beef cattle in northern and northeastern China and also identified several herd-level risk factors associated with the presence of seropositive cattle. Future research is required to compare the influence of each factor responsible for these pathogens infection at animal-and herd-level in these regions and elsewhere in China.