Disinfection of Maternal Environments Is Associated with Piglet Microbiome Composition from Birth to Weaning

Abstract

Maternal factors predetermine offspring development and health, including the establishment of offsprings' first microbiomes. Research in swine has shown that early microbial exposures impact microbiome colonization in piglets, but this phenomenon has never been tested in the context of delivery room disinfection. Thus, we exposed gestating sows to two delivery environments (n = 3/environment): stalls cleaned with a broad-spectrum disinfectant (disinfected environment [D]) or stalls cleaned only with hot-water power washing (nondisinfected environment [Nde]), 3 days prior to farrowing. Microbiomes of sows and farrowed piglets (n = 27/environment) were profiled at 4 different time points from birth to weaning via 16S rRNA sequencing. The results show that although vaginal, milk, skin, and gut microbiomes in mothers were minimally affected, sanitation of farrowing stalls impacted piglet microbiome colonization. These effects were mainly characterized by lower bacterial diversity in the gut and nasal cavity, specifically in D piglets at birth, and by distinct taxonomic compositions from birth to weaning depending on the farrowing environment. For instance, environmental bacteria greatly influenced microbiome colonization in Nde piglets, which also harbored significantly higher abundances of gut Lactobacillus and nasal Enhydrobacter at several time points through weaning. Different sanitation strategies at birth also resulted in distinct microbiome assembly patterns, with lower microbial exposures in D piglets being associated with limited interactions between bacterial taxa. However, increasing microbial exposures at birth through the lack of disinfection were also associated with lower piglet weight, highlighting the importance of understanding the trade-offs among optimal microbiome development, health, and growth performance in swine production systems. IMPORTANCE We show that levels of disinfection in farrowing facilities can impact early microbial exposures and colonization by pioneer microbes in piglets. Although previous research has shown a similar effect by raising pigs outdoors or by exposing them to soil, these practices are unattainable in most swine production systems in the United States due to biosecurity practices. Thus, our results underscore the importance of evaluating different disinfection practices in swine production to safely reduce pathogenic risks without limiting early microbial exposures. Allowing early exposure to both beneficial and pathogenic microbes may positively impact immune responses, reduce the stressors of weaning, and potentially reduce the need for dietary antimicrobials. However, the benefits of modified early microbial exposures need to be accomplished along with acceptable growth performance. Thus, our results also provide clues for understanding how disinfection practices in farrowing rooms may impact early microbiome development and assembly.

Keywords: disinfection; environment; gut; maternal; microbiome; piglet; programming; swine.

FIG 1

(a) Methodology and sample collection timeline for sows and piglets in disinfected (D) and nondisinfected (Nde) environments. (b) Diagram of the farrowing barn setup showing the separation of treatment groups into different rooms, with red stars denoting stalls randomly selected for the study.

FIG 2

Bacterial diversity (Shannon’s H) between piglets born in the two different farrowing environments, assessed for piglet fecal (a) and nasal (b) samples from birth to weaning (day 0 to day 21).

FIG 3

(a and b) Piglet gut (a) and nasal (b) microbiome compositions differed significantly from birth to weaning (Bray-Curtis, PCoA). Individual samples are represented by different shapes and colors according to farrowing stall sanitation. Dotted ellipses represent 95% confidence intervals in multivariate space. (c) Amount of variation of piglet microbiome composition explained by sanitation level in the farrowing environment (Bray-Curtis PERMANOVA, R²) for fecal and nasal samples.

FIG 4

(a) Top discriminant genera at birth in piglet gut microbiomes between farrowing environments were observed to be two common environmental bacteria (Aggregatibacter and Chryseobacterium). (b) Lactobacillus was identified as a discriminant genus in piglet gut microbiomes at days 14 and 21. (c) Discriminant genera in piglet gut microbiomes with an indicator species score of >0.6 are displayed for all four time points. (d) The top discriminant genera at birth in piglet nasal microbiomes mirrored those present in piglet gut microbiomes. (e) The genus Enhydrobacter was identified as a discriminant genus in piglet nasal microbiomes at all four time points. (f) Discriminant genera for piglet nasal microbiomes with an indicator species score of >0.6 are displayed for all four time points. Darker shading in panels c and f correspond to higher indicator values.

FIG 5

(a and b) Network analyses of piglet fecal samples from disinfected (a) and nondisinfected (b) farrowing environments. Each dot or node represents one taxon at the ASV level, with darker shading corresponding to higher degrees and larger node sizes corresponding to higher neighborhood connectivities. Edges represent the undirected interaction or correlation between two nodes or taxa. (c) Degree and neighborhood connectivity values were then quantified for each farrowing environment.

FIG 6

(a and b) Piglet growth performance from birth to weaning displayed through birth weights (BW) and weaning weights (WW) (a) as well as growth performance up to 6 weeks postweaning (b), with error bars representing standard errors. Period 1 (P1) represents the time period from birth to weaning (a), with P2 and P3 denoting weeks 1 and 2 postweaning, respectively, and P4 denoting the last 4 weeks of the nursery period. (c to f) PCoAs based upon discriminant ASVs between farrowing environments and Spearman correlations between piglet weights and PCoA scores along axis 1 at birth and weaning were created for piglet fecal (c and d) and nasal (e and f) samples, with shaded areas representing the 95% confidence intervals based on standard errors.