Approaching the sterile womb theory in dogs and cats: A multi-technique investigation.

Abstract

The study investigated whether bacterial seeding occurs in utero in dogs and cats using a multi-technique approach, including 16S rRNA gene sequencing, culture, and fluorescence in situ hybridization (FISH). Healthy pregnant bitches and queens (n = 8) undergoing ovariohysterectomy between 30 and 45 days of gestation were included. Placenta and amniotic fluid samples from two foetuses per dam, along with multiple controls (uterine serosa, sampling table, and surgeon's gloves), were collected and analysed. Bacterial sequences were detected in all foetal samples, with no significant differences in intra- and inter-sample diversity (i.e., alpha and beta diversities, respectively) based on sample type or species. However, the surgeon's gloves showed species-specific differences in bacterial composition. After removing control sequences, significant differences based on foeto-maternal units emerged. Moraxella spp. was cultured from the canine placenta with the highest bacterial load, and Burkholderia cepacia was isolated from two feline placentae. FISH showed low bacterial presence in 50 % of placentae without histological signs of inflammation. No bacterial growth was observed in amniotic fluid or control samples. A multi-technique approach, including multiple controls, is essential in studies involving low-biomass samples, as the results of the present study indicated that contamination could mask real bacterial composition of foetal samples. After decontamination, differences were observed based on the foeto-maternal unit. This study confirmed the presence of bacterial DNA in placental and amniotic fluid samples during mid-term, healthy pregnancies in dogs and cats. Bacterial DNA is not equivalent to living bacteria, hence bacterial DNA fragments in the uterus probably originate from the dam but are unlikely to indicate the presence of a resident foetal microbiome.