An epidemiological investigation of coxiella burnetii and chlamydia spp. as infectious agents causing abortion in dairy cattle in Uruguay

Infectious diseases are the leading aetiological cause of bovine abortion globally as well as in Uruguayan dairy farms. Studies conducted in this country showed that a proportion of aborted fetuses without diagnosis present histological lesions compatible with the action of an infectious agent. These undiagnosed abortion cases could be due to infectious agents not currently included in the battery of diagnostic tests routinely applied. Infectious agents such as Coxiella burnetii and Chlamydia spp., typically considered to have a minor or relative marginal role in causing abortion in cattle, have never been systematically investigated in Uruguay. This thesis challenged the current diagnostic protocols, seeking evidence supporting the inclusion of these agents to dismiss the number of undiagnosed cases. This work investigated the association of these pathogens with cattle abortion in commercial dairy herds and the potential as zoonotic threats in the local context. The thesis will first introduce the main characteristics of dairy production in Uruguay (Chapter One) and thoroughly review Coxiella burnetii and Chlamydia abortus (Chapter Two). After this, two main sections will be presented. The first section including a systematic review and meta-analysis (Chapter Three), and the second section comprising the observational fieldwork studies conducted out on dairy farms in Uruguay, which includes a cross-sectional study of pooled milk (PM) samples from commercial herds (Chapter Four), a case-series sampling of placentas (Chapter Five), a case-control sampling of aborted and non-aborted dairy animals (Chapter Six), and finally a retrospective cohort evaluation of workers exposed to a local bovine Q fever outbreak (Chapter Seven). A final discussion (Chapter Eight) will consider the main findings' implications and integrate these results into the general body of knowledge. The systematic review and meta-analysis evidenced bulk tank milk (BTM) sampling as a broadly used epidemiological methodology for large-scale investigations of C. burnetii and provided a global framework of the prevalence of C. burnetii in collective milk samples from commercial farms. This study showed a widespread herd-level circulation of C. burnetii in bovine dairy farms by reporting a high overall global prevalence of 37.0% (CI95% 25.2-49.5%). A meta-regression showed the herd size as the most relevant moderator, with the odds of a BTM sample testing positive doubling with every unit increase. This C. burnetii meta-prevalence roles as a benchmark for comparison with the findings of the molecular investigation on PM samples obtained from local dairy farms. The mass-scale molecular evaluation on PM samples evidenced a low incidence (1.7%) of C. burnetii DNA. The findings showed that clinically healthy (asymptomatic) cows might shed the bacterium, raising awareness of Q fever as potential food safety and public health concern considering the C. burnetii survival as a milk-borne pathogen in unpasteurised milk and raw dairy products. Due to the exceptionally high infectivity, low infective doses, and aerosol transmission, the culture of C. burnetii and Chlamydiales should be done on BSL-3 laboratories not currently operating in Uruguay. Bacterial cultivation was not attempted in any of the studies conducted in this thesis. Hitherto the lack of diagnostic tools and the impossibility of doing culture have restricted local epidemiological investigation of these agents. Therefore, developing diagnostic tests to be used routinely in domestic laboratories is imperative to save costs and optimise currently available facilities and work with higher autonomy. For this purpose, a published protocol targeting well-evaluated genes was adapted to provide an available tool for local laboratories. Coxiella burnetii, C. abortus, and C. pecorum were investigated for the first time using an m-PCR in placentas from aborted dairy cows. Coxiella burnetii-DNA was detected and quantified in those samples, which supported this bacterium as an abortifacient agent in Uruguay. No co-infections of these pathogens were found. Evidence supporting Chlamydiales as a source of cattle abortion remain blurred. Coxiella burnetii was detected on the aborted placenta from a cow from an artisanal cheese-producing farm. Consumption of raw milk and dairy products represent a potential source for human infection. This finding underlined that the public health risk posed by C. burnetii should not be neglected and should be emphasised the need for on-farm milk pasteurisation by local artisanal cheesemakers. Molecular investigation of C. abortus in vulvo-vaginal swabs samples showed no evidence of this bacterium neither in aborted nor in control animals. Difficulties in identifying low-grade infection and evaluating a single sample per animal would have constrained the detection. The first attempted studies conducted so far support C. abortus as a no substantial abortifacient agent in cattle from Uruguay. Serological evidence confirmed the local bovine population as a potential reservoir for C. burnetii infection in humans. Anti-C. burnetii phase II IgM and IgG immunoglobulins were detected in a group of farmworkers and laboratory technicians exposed to aborted dairy cattle or aborted materials (fetuses and placenta) by indirect immunofluorescence. Molecular approaches were assessed, optimised and validated on veterinary clinical samples such as aborted placentas, vulvo-vaginal swabs or collective milk samples, providing valuable alternatives beyond the bacterial culture and isolation. The thesis presents original research studies that utilise different epidemiological strategies to search for evidence of an association between the infection by the pathogens and the occurrence of bovine abortion.

Agencia Nacional de Investigación e Innovación

Instituto Nacional de Investigación Agropecuaria