Infertility in dairy cows – Possible bacterial and viral causes
A. Dobos*, I. Fodor, Z. Kreizinger, L. Makrai, B. Dénes, I. Kiss, D. Đuričić, M. Kovačić and L. Szeredi
Abstract
In this research uterine swab and biopsy samples were collected from 40 infertile dairy cows kept at five dairy cattle farms in Hungary.
Samples were tested for bacteria including Coxiella burnetii chlamydiae, Mycoplasma and Ureaplasma, and for the viruses Bovine herpesvirus 1 (BoHV-1) and Bovine viral diarrhoea virus (BVDV). Chlamydiaceae DNA was detected by real-time PCR in 22/40 (55%) samples. Coxiella burnetii DNA was detected in 3/40 (7.5%) cases by real-time PCR.
Mycoplasma and Ureaplasma DNA was found in 2/40 (5%) and 4/40 (10%) cows, respectively.
BVD and BoHV-1 DNA was not detected in any samples. Escherichia coli as a recognised uterine pathogen was found in two cases. The following potential uterine pathogens were found: Bacillus licheniformis (one case), non-haemolytic streptococci (five cases), Histophilus somni (two cases) and Candida krusei (two cases). Blood samples were collected at same time as swab samples from all 40 cows, and their examination for C. burnetii antibodies by ELISA revealed seropositivity in 26/40 cows (65%). Histological examination of the uterine biopsy samples showed the presence of mild lympho-histiocytic infiltration in the mucosa in 22 cases (59%). Moderate lympho-histiocytic infiltration of the endometrium was evident in 13 cases (35%), while in two cases (6%) severe inflammatory cell infiltration of the endometrium with lympho-histiocytes and neutrophil granulocytes was found.
Although no statistical correlation could be demonstrated between the severity of histological lesions of the endometrium and the uterine pathogenicity of the bacteria (P = 0.8555), endometritis of a certain severity grade and/or a recognised or potential uterine pathogen were found in all samples. The latter may play a role in the development of infertility either collectively or independently.Key words: Bovine infertility; Dairy cow; Histological examination; Uterine biopsy; Uterine swab
Introduction
Reproductive disorders are the second largest source of economic loss (after mastitis (Benić et al., 2018; Burović, 2020)) and the number one cause of involuntary culling among herd-level animal health problems in dairy herds (Ózsvári and Kerényi, 2004; Bonneville-Hébert et al., 2011; Folnožić et al., 2015; Kovács et al., 2020). Up to 40% of culled cows are removed from the herd because of infertility (Meadows et al., 2005). The most important economic consequences of suboptimal reproductive performance are the increased calving interval (leading to decreased milk production per year), higher culling risk, decreased calf sales, and increased treatment and semen costs (Cabrera, 2014; Folnožić et al., 2019). The role of several bacterial, viral and protozoal agents of infertility of dairy cattle has been reviewed (Yoo, 2010; Wathes et al., 2020), confirming their importance as sources of economic loss.
As regards viral infections, the average financial loss caused by a cow seropositive for bovine herpesvirus-1 (BoHV-1) was estimated at USD 379, compared to their seronegative counterparts (Can et al., 2016). This loss increased to USD 509 per case in the case of BoHV-1-induced abortion. De Vries (2006) estimated the average cost of any abortion at USD 555. The analysis of 3,660 calving records showed that the average economic loss due to uterine inflammations was EUR 122.8 per case, with an increased number of open days being the predominant source of loss, accounting for 57.6% of the total loss (Kern et al., 2018).
The aim of this study was to demonstrate the possible role of uterine infections caused by different bacteria and viruses in the infertility of dairy cattle.
Materials and methods
Animals
The samples were collected from 40 infertile dairy cows kept at five dairy farms in March 2021 (herd size: 650–1,800 cows; milk production: 9,500–11,600 kg/cow/year). At these farms, the culling rate for slaughter ranged between 32% and 40% in the previous year. On average, 40% of the total culling rate was associated with infertility (ranges: 32–48%). All cows were Holstein-Friesian, fed a total mixed ration (TMR) and bred by artificial insemination (AI). The cows included in this study exceeded 220 days in milk (DIM), had been inseminated at least three times, and had not become pregnant.
Uterine swab collection and bacteriology (n = 40)
After cleaning the vulva and the surrounding region with clean water, uterine swab samples were collected using swabs protected from vaginal contamination (Equivet uterine swab, Kruuse, Marslev, Denmark). All swab samples were transported to the laboratory in a refrigerator at 4°C within 2-4 hours of sample collection. Swabs of uterine contents were inoculated onto blood agar, MacConkey agar and Sabouraud Dextrose agar incubated aerobically, blood agar incubated anaerobically, Campylobacter-selective agar (Skirrow’s medium) incubated under microaerophilic conditions, and chocolate agar incubated in 10% CO2 at 37°C for up to 3 days. Bacterial and fungal isolates were identified to the genus level on the basis of cultural, morphological and biochemical features, and to the species level using the MALDI-TOF system.
Isolation of Ureaplasma and Mycoplasma species was also attempted in Ureaplasma medium (Mycoplasma Experience Ltd., UK) and in Mycoplasma broth medium (pH 7.8) [Thermo Fisher Scientific Inc., (Oxoid Inc.), Waltham, MA] supplemented with 0.5% (w/v) sodium pyruvate, 0.5% (w/v) glucose and 0.005% (w/v) phenol red. Additional swabs were taken for Chlamydiales, C. burnetii, U. diversum and Mycoplasma species-specific PCR tests. The DNA was extracted from the samples with the NucleoSpin Tissue Mini kit (Macherey-Nagel GmbH & Co. KG, Düren, Germany), according to the manufacturer’s instructions. A species-specific PCR assay targeting the 16S rRNA region was used to detect the presence of U. diversum in the samples (Tramuta et al., 2011). A genus-specific PCR assay targeting the 16S/23S rRNA intergenic spacer region was used to detect the presence of Mycoplasma species in the samples (Lauerman et al., 1995). Members of the Chlamydiales order were detected with the 16S rRNA gene-based Chlamydiales-specific qPCR assay as described previously (Lienard et al., 2011). Real-time PCR assay specific for the IS1111 element was used to detect the presence of C. burnetii in the uterine swabs. The sixth samples were tested by real-time PCR for BVDV and BoHV-1 DNA by the Bio-T kit® BVDV & BHV1-gE PCR kit (Biosellal, Dardilly, France) according to the manufacturer’s instructions. The detected bacteria were categorised (Table 1) according to their pathogenic potential reported in previous studies (Williams et al., 2007; Borel et al., 2018; De Biase et al., 2018; Appiah et al., 2020; Santos et al., 2021).
The category of uterine pathogens included bacterial species reported to be associated with uterine lesions (‘recognised uterine pathogens’), the category of potential bacterial pathogens included species frequently isolated from the uterus of cows presenting endometritis and commonly associated with uterine lesions, while the category of opportunistic pathogens included bacterial species occasionally isolated from the uterine lumen and not associated with endometritis (contaminants).
Blood samples (n = 40)
Blood samples were collected at same time as swab samples from all the 40 cows studied and were examined by ELISA.
Commercial ELISA kits (ID Screen® Q Fever Indirect Multispecies, IDVet Inc., Grabels, France) were used according to the manufacturer’s instructions. The serum samples were examined by two different complement fixation tests (CFT) utilising C. burnetii phase I and II antigens, according to the manufacturer’s instructions (Virion/Serion GmbH, Würzburg, Germany), and the Manual of Diagnostic Tests and Vaccines for Terrestrial Animals (World Organisation for Animal Health, 2018).
Histological examination of uterine biopsy samples (n = 40)
A previously disinfected biopsy apparatus (Kruuse, Marslev, Denmark) was introduced into the uterus, and an approx. 0.5 × 0.5 × 1 cm portion of the uterine mucosa was chipped off from the dorsal wall of the uterine body, at the junction of the uterine horns, under control by rectal palpation. All biopsies were performed by the same operator.
The biopsy samples were fixed in 10% buffered formaldehyde solution for 24 h. Subsequently the samples were embedded in paraffin, cut into 4 µm thick sections, which were then stained with haematoxylin and eosin. The samples were evaluated on the basis of the criteria described by Chapwanya et al. (2009).
Statistical method
Histological findings were classified as either mild (category 1) or severe (category 2 or 3). The bacteriological results of the samples were categorised according to uterine pathogenicity (recognised, potential, opportunistic, and no uterine pathogens). The samples were allocated to these four categories based on the bacterium with the highest expected uterine pathogenic potential (e.g., if both a recognised and a potential uterine pathogen was found in the sample, it was classified as recognised).
The relationship between severity class (mild or severe) and uterine pathogenicity was examined by mixed effects logistic regression with farm as the random effect. Model building was performed using the glmmTMB package in R (Brooks et al., 2017). Statistical analyses were performed in R version 4.0.5 (R Core Team, 2021). The level of significance was set to 0.05.
Results
Bacteriology and virology
The results of bacteriological and virological examinations are summarised in Table 1. Eight samples showed mixed infections and the remaining specimens yielded facultative anaerobic bacteria in pure culture. The mixed cultures contained mainly Bacillus spp. and Streptococcus spp. The most common facultative anaerobic pathogens were Streptococcus spp., Staphylococcus spp. and Corynebacterium spp. We found Escherichia coli as a recognised uterine pathogen in two cases. Potential uterine pathogens were Bacillus licheniformis (one case), non-haemolytic streptococci (five cases), Histophilus somni (two cases) and Candida krusei (two cases).
Opportunistic uterine contaminants included Clostridium perfringens (one case), Staphylococcus species, coagulase-negative (three cases), Bacillus spp. (ten cases), and Corynebacterium (four cases).
Chlamydiaceae DNA was detected by real-time PCR in 22/40 samples (55%). We detected C. burnetii DNA in 3/40 cases (7.5%) by real-time PCR. Mycoplasma and Ureaplasma DNA were found in 2/40 (5%) and 3/40 (7.5%) samples, respectively. BVDV and BoHV-1 DNA was not detected in any samples. ELISA testing showed 26/40 (65%) individual seropositivity among the cows examined, while 7/40 (17.5%) and 6/40 (15%) of the cows exhibited low titres (1:10–1:40) by Phase I and Phase II CFT, respectively.
One blood sample 1/40 (2.5%) showed high titres (1:640) by Phase II CFT.
Histological examination of the uterine biopsy samples
Three of the uterine biopsy samples were not suitable for evaluation due to a lack of endometrial mucosa. Twentytwo cases (59%) showed the presence of mild lympho-histiocytic infiltration in the mucosa, and were included in category 1 (Fig. 1a).
Prominent leukostasis was also evident in two of these cases (Fig. 1b).
Moderate lympho-histiocytic infiltration of the endometrium was evident in 13 cases (35%), which were included in category 2 (Fig. 2).
The remaining two cases were included in category 3, as they presented severe inflammatory cell infiltration in the endometrium, including lympho-histiocytes and neutrophil granulocytes (Fig. 3).
The number of samples in the recognised, potential, opportunistic, and no uterine pathogen categories based on the highest expected uterine pathogenic potential was 7 (17.5%), 22 (55.0%), 6 (15.0%), and 5 (12.5%), respectively.
Severe histological lesions occurred in 28.6% (2/7), 40.0% (8/20), 50.0% (3/6), and 50.0% (2/4) of the samples in the recognised, potential, opportunistic, and no uterine pathogen categories, respectively. No relationship was found between the severity of histological lesions and uterine pathogenicity (P = 0.8555).
Discussion
Unfortunately, there is little information on the bacterial causes of infertility in cows. Campylobacter fetus ssp. venerealis, Leptospira, Mycoplasma, Ureaplasma, Chlamydia, Histophilus somni and C. burnetii have all been associated with bovine infertility, but at present the relative prevalence and importance of these agents are unknown (Yaeger and Holler, 2007; Agerholm, 2013). The most frequently isolated bacteria were Bacillus spp. and Streptococcus spp., either in pure culture or in mixed cultures with other microorganisms, and they represented 56.9% of all isolates in this study.
Although Bacillus spp. are opportunistic microorganisms, they can occasionally contribute to the development of bovine abortion (Yaeger and Holler, 2007). In one case, we found Bacillus licheniformis, a potential abortifacient bacterium (Agerholm et al., 1999). The isolated Staphylococcus spp. and Streptococcus spp. are part of the normal microflora of mucosal surfaces. We detected Escherichia coli in two cases. This bacterium was cultured some days after parturition, but around breeding time, E. coli can cause endometritis and is a well-recognised uterine pathogen (Brodzki et al., 2014). Histophilus somni was found in two cases. This bacterium is known to cause diseases of the reproductive tract in cattle but the virulence of the strains may vary, the cause of which has not been defined (Pérez et al., 2010). The Corynebacterium spp. and Staphylococcus spp. found in this study generally constitute additional flora along with the major uterine pathogens (Watts et al., 2000; Ghanem et al., 2015).
The isolation of Mycoplasma spp. and Ureaplasma spp. has been linked with reproductive disorders in cattle. Several studies have indicated an association between the occurrence of Mollicutes infection and bovine infertility (Pfützner and Sachse, 1996; Macedo et al., 2018; Santos et al., 2021). We confirmed five cases of Mollicutes infection by mPCR and attempted to recover the pathogens by microbial culture but no growth was obtained. Chlamydiaceae DNA was detected by real-time PCR in 22/40 (55%) samples, thus Chlamydiaceae were the most frequently found pathogens in this study. Chlamydia abortus is incriminated as a cause of bovine endometritis with resulting infertility, while other Chlamydiales are associated with reproductive problems though further research is needed to reveal their exact aetiological role (Wittenbrink et al., 1993; Borel et al., 2018). Coxiella burnetii was demonstrated in the macrophages of the endometrium in cattle (De Biase et al., 2018). This study found mild to severe chronic endometritis in PCR-positive animals. We detected three C. burnetii PCR-positive cases by real-time PCR assay. However, only one of the three PCR-positive cases tested positive by ELISA of the blood as well, which is in agreement with the findings reported by Guatteo et al. (2006). Serological data and PCR detection of the pathogen in the uterus may not be correlated. We found diffuse endometritis in two cases, but no correlation between this agent and endometritis. Our previous research found 48.2% C. burnetii seropositivity by ELISA among pregnant cows, while the seropositivity of animals that had lost their pregnancy at an early stage was 80.5%. This study found a higher seropositivity rate (65%) in the infertile cows compared with the pregnant animals in the previous study. We detected only a single case with 1:80 titre by CFT (Phase II), which is indicate of an active phase of C. burnetii infection (Dobos et al., 2020). Clostridium spp. was detected in one case. Moderate endometritis was evident in 13 cases and we also detected two recognised uterine pathogens (E. coli and Bacillus licheniformis) in these cases. Chlamydiaceae DNA was detected in 22 cases. Three of these 22 cases were also positive for C. burnetii by PCR. Ten (45.4%) of the 22 cases showed diffuse and moderate endometritis (Category 2 or 3) while the rest of cases were mild (Category 1). Chlamydia spp. were identified in 22/40 cases (55%) in this study, which is higher than the 12.9% value obtained in a previous research (Fábián et al., 2007). We did not find a correlation between these agents and endometritis in this study. There was no correlation between Mycoplasma and Ureaplasma PCR positivity and the histological or bacteriological findings.
All the farms included in the study practised vaccination against infectious bovine rhinotracheitis (IBR), which was possibly the reason why BoHV-1 DNA was not detected in any samples.
However, as three of the five farms were infected with BVDV and none of the five farms were vaccinated against BVD, it was surprising that BVDV DNA was not detected by PCR.
In conclusion, the presence of pathogenic agents in the uterus evidently affects the chances of embryo survival (Sheldon et al., 2006). A healthy uterus and endometrium are key elements for embryo implantation. Interactions between the endometrium and the conceptus are influenced by many factors mostly related to the uterine environment. Pathogenic and potentially pathogenic bacteria often persist for long periods of time, causing uterine disease and changing the uterine environment, which are the key causes of infertility in cattle. This study highlights the need for a better understanding of the aetiology and pathogenesis of bovine infertility and the possible causative agents. As reproductive performance is a major factor influencing the profitability of dairy farms, further investigations into the possible causative agents and factors of infertility are needed.
References [… show]
Neplodnost u mliječnih krava – mogući bakterijski ili virusni uzroci
Attila DOBOS, István FODOR, CEVA-Phylaxia Co. Ltd., Szállás u. 5, H-1107 Budapest, Hungary; Zsuzsa KREIZINGER, Institute for Veterinary Medical Research, Budapest, Hungary; László MAKRAI, Department of Microbiology and Infectious Diseases, University of Veterinary Science, Budapest, Hungary; Béla DÉNES, National Food Chain Safety Office, Veterinary Diagnostic Directorate, Budapest, Hungary; István KISS, CEVA-Phylaxia Co. Ltd., Szállás u. 5, H-1107 Budapest, Hungary; Dražen ĐURIČIĆ, Mislav KOVAČIĆ, Mount Trade d.o.o., Garešnica, Croatia; Levente SZEREDI, National Food Chain Safety Office, Veterinary Diagnostic Directorate, Budapest, Hungary
U ovom su istraživanju prikupljeni brisevi maternice i biopsijski uzorci 40 neplodnih mliječnih krava s pet mliječnih farmi u Mađarskoj. Uzorci su testirani na bakterije, uključujući; Coxiella burnetii, klamidiju, mikoplazmu i ureaplazmu te na viruse uključujući goveđi herpesvirus 1 (BoHV-1) i virus virsnog proljeva goveda (BVDV).
DNK Chlamydiaceae otkriven je PCR testom u stvarnom vremenu u 22/40 (55 %) uzoraka.
DNK bakterije Coxiella burnetii otkriven je u 3/40 (7,5 %) slučajeva PCR testom u stvarnom vremenu. DNK mikoplazme i ureaplazme pronađen je u 2/40 (5 %), odnosno 4/40 (10 %) krava. DNK virusa BVD i BoHV-1 niti u jednom uzorku nije otkriven. Escherichia coli kao priznati maternični patogen pronađen je u dva slučaja. Pronađeni su sljedeći potencijalni maternični patogeni: Bacillus licheniformis (jedan slučaj), nehemolitički streptokoki (pet slučajeva), Histophilus somni (dva slučaja) i Candida krusei (dva slučaja). Uzorci krvi su istovremeno prikupljeni kad i brisevi od svih 40 pokusnih krava. Njihova pretraga na protututijela C. burnetii ELISA metodom otkrila je seropozitivnost u 26/40 krava (65 %). Histološka pretraga uzoraka biopsije maternice pokazala je prisutnost blage limfohistiocitotske infiltracije u sluznici u 22 slučaja (59 %). Umjerena limfohistiocitotska infiltracija endometrija bila je prisutna u 13 slučajeva (35 %), dok je u dva slučaja (6 %) otkrivena ozbiljna upalna stanična infiltracija endometrija s limfohistiocitima i neutrofilnim granulocitima. Premda nije bilo moguće dokazati statističku korelaciju između ozbiljnosti histoloških lezija endometrija i maternične patogenosti bakterija (P = 0,8555), endometrioza određenog stupnja ozbiljnost i/ili priznati ili potencijalni maternični patogeni pronađeni su u svim uzorcima. Ovi posljednji mogu prouzročiti razvoj neplodnosti bilo skupno ili pojedinačno.Ključne riječi: neplodnost goveda, mliječna krava, histološka pretraga, biopsija maternice, bris maternice