Exploring the association between microbiota and behaviour in suckling piglets

Abstract

It is increasingly recognised that the microbes residing in the gastrointestinal tract can influence brain physiology and behaviour, via the microbiota-gut-brain axis. Here, we made a first explorative evaluation at the association between the gut microbiota and behaviour in suckling piglets. 16S microbiota profiling information was obtained from two independent replicate experiments at 2 and 4 weeks of age. Piglets underwent a backtest to assess their personality or coping style at 2 weeks of age, and were subjected to a combined open field and novel object test at 3.5 weeks of age, recording anxiety-related and exploratory behaviour. The number of squeals vocalised during the open field test was associated with microbial groups such as Coprococcus 3 and CAG-873, whereas in the novel object test, explorative behaviour was significantly associated with microbial genera like Atopobium and Prevotella. Overall, this study explores the microbiota-behavioural relation by employing multivariate analysis and exemplifies the importance of individualised analyses when evaluating such relationships.

Figure 1

Schematic study design to assess association between the microbiota and behaviour. Microbiota and behavioural data were obtained from two independent experiments. Rectal swabs were collected at day 15 (2 weeks of age) and day 28 (4 weeks of age) pre-weaning for microbiota analysis. Personality (or back) test and Anxiety test (open field test, novel object test) were performed at day 15 (2 weeks of age) and day 25 (3.5 weeks of age) respectively, in suckling piglets.

Figure 2

Assessing microbial associations of individual behavioural variables in open field test (OFT). (A) Partial redundancy analysis (pRDA) of the frequency of squeals (F_squeal_OFT), corrected for experiment and treatment. In the pRDA triplot, individual animals are indicated in coloured balls belonging to either one of the experiments and treatment group. Microbial groups having (1) RDA response score ≥ 0.35 as well as (2) minimum 0.1% relative abundance in at least 10% of the samples, are visualised in the triplot. The perpendicular distance between microbes and environmental variable axes in the plot reflects their correlations. The smaller the distance, the stronger the correlation. (B) Spearman correlation of individual pRDA identified microbes with the frequency of squeals.

Figure 3

Assessing microbial associations of individual behavioural variables in novel object test (NOT). (A) Partial redundancy analysis (pRDA) of the nosing behaviour (% of time spent nosing) during the novel object test, corrected for experiment and treatment. (B) pRDA of the exploring behaviour (nosing + rooting: both walls and floor; % of time spent exploring), corrected for experiment and treatment. In the pRDA triplot, individual animals are indicated in coloured balls belonging to either one of the experiments and treatment group. Microbial groups having (1) RDA response score ≥ 0.35 as well as (2) minimum 0.1% relative abundance in at least 10% of the samples, are visualised in the triplot. The perpendicular distance between microbes and environmental variable axes in the plot reflects their correlations. The smaller the distance, the stronger the correlation. (C) Spearman correlation of pRDA identified microbes with nosing behaviour. (D) Spearman correlation of pRDA identified microbes with exploring behaviour.