A multi-kingdom metabarcoding study on cattle grazing Alpine pastures discloses intra-seasonal shifts in plant selection and faecal microbiota

Abstract

Diet selection by grazing livestock may affect animal performance as well as the biodiversity of grazed areas. Recent DNA barcoding techniques allow to assess dietary plant composition in faecal samples, which may be additionally integrated by the description of gut microbiota. In this high throughput metabarcoding study, we investigated the diversity of plant, fungal and bacterial taxa in faecal samples of lactating cows of two breeds grazing an Alpine semi-natural grassland during summer. The estimated plant composition of the diet comprised 67 genera and 39 species, which varied remarkably during summer, suggesting a decline of the diet forage value with the advancing of the vegetative season. The fungal community included Neocallimastigomycota gut symbionts, but also Ascomycota and Basidiomycota plant parasite and coprophilous taxa, likely ingested during grazing. The proportion of ingested fungi was remarkably higher than in other studies, and varied during summer, although less than that observed for plants. Some variation related to breed was also detected. The gut bacterial taxa remained stable through the summer but displayed a breed-specific composition. The study provided insights in the reciprocal organisms' interactions affecting, and being affected by, the foraging behaviour: plants showed a high temporal variation, fungi a smaller one, while bacteria had practically none; conversely, the same kingdoms showed the opposite gradient of variation as respect to the animal host breed, as bacteria revealed to be the group mostly characterized by host-specificity.

Figure 1

Clustered bar chart of the relative percent abundance showing the substitutional shifts of the plant taxa, grouped at family level, identified by DNA sequencing from bovine fecal material. The number of species featured by families in which more than one taxon was found is indicated in brackets. Samples belonging to the four time points (T1 trough T4) are ordered from left to right and coded by cow individual numbers and breed (SIM: Simmental; AG: Alpine Grey).

Figure 2

Variation of the relative abundance (% of total abundance) of the 16 plant genera/species identified in faecal samples with a relative abundance > 1% in relation with collection time (T1, T2, T3 and T4). Plant taxa are ordered from left to right and top to bottom according to decreasing average relative abundance. The significance of the effect of collection time in the linear mixed regression models analysing relative abundances (standardized with total sum scaling (TSS) normalisation followed by square route transformation) is indicated as ***P < 0.001; **P < 0.01; *P < 0.05. Boxplots in yellow indicate non-significant temporal trends, boxplots in green indicate a pattern of increase with collection time, boxplots in red indicate a pattern of decrease.

Figure 3

Principal coordinate analysis showing synoptically the ordination plots of standardized relative abundances of plant, fungal and bacterial taxa for the variables “collection time” and “breed”. Dotted arrows indicate gradients of time-consistent data partitioning. Dotted lines divide parts of the quadrants fitting a breed-consistent separation.

Figure 4

Canonical coordinate analysis of standardized relative abundances of plant, fungal and bacterial taxa in relation to sampling time and cow breed. The respective significance levels are indicated into the panels and highlighted in red when significant p values were obtained.

Figure 5

Depiction of the observed trends outlining the reciprocal gradients of community stability versus variation as a function of the two inspected variables of sampling time and cow breed. The pattern of plant diet composition (unstable in time, since ruled by their intra-seasonal availability succession but showing no preference in consumption by the two breeds) is opposite to that of bacteria (stable as gut residents irrespective of the seasonally changing botanical composition of the diet but displaying breed-specific composition). The fungal component, belonging partly to the grazed transient food and partly to the cow microbiota, showed an intermediate pattern. The fourth component is the Animal host kingdom, within which the above dynamics appear to unfold.