Abrupt dietary changes between grass and hay alter faecal microbiota of ponies

Anna Garber; Peter Hastie; David McGuinness; Pauline Malarange; Jo-Anne Murray

doi:10.1371/journal.pone.0237869

. 2020 Aug 18;15(8):e0237869. doi: 10.1371/journal.pone.0237869

Abrupt dietary changes between grass and hay alter faecal microbiota of ponies

Anna Garber ^1,^*, Peter Hastie ², David McGuinness ³, Pauline Malarange ⁴, Jo-Anne Murray ²

Editor: Juan J Loor⁵

PMCID: PMC7446798 PMID: 32810164

Abstract

Abrupt dietary changes, as can be common when managing horses, may lead to compositional changes in gut microbiota, which may result in digestive or metabolic disturbances. The aim of this study was to describe and compare the faecal microbiota of ponies abruptly changed from pasture grazing ad libitum to a restricted hay-only diet and vice versa. The experiment consisted of two, 14-day periods. Faecal samples were collected on day 0 and days 1–3,7,14 after abrupt dietary change from grass to hay and from hay to grass. Microbial populations were characterised by sequencing the V3-V4 region of the 16S rRNA gene using the Illumina MiSeq platform, 4,777,315 sequences were obtained from 6 ponies. Further analyses were performed to characterise the microbiome as well as the relative abundance of microbiota present. The results of this study suggest that the faecal microbiota of mature ponies is highly diverse, and the relative abundances of individual taxa change in response to abrupt changes in diet. The faecal microbiota of ponies maintained on a restricted amount of hay-only was similar to that of the ponies fed solely grass ad libitum in terms of richness and phylogenetic diversity; however, it differed significantly in terms of the relative abundances at distinct taxonomic levels. Class Bacilli, order Lactobacillales, family Lactobacillaceae, and genus Lactobacillus were presented in increased relative abundance on day 2 after an abrupt dietary change from hay to grass compared to all other experimental days (P <0.05). Abrupt changes from grass to hay and vice versa affect the faecal microbial community structure; moreover, the order of dietary change appears to have a profound effect in the first few days following the transition. An abrupt dietary change from hay to grass may represent a higher risk for gut disturbances compared to abrupt change from grass to hay.

Introduction

The microbial population of the equine gastro-intestinal tract (GIT) is highly diverse [1–3], which may be related to the evolutionary survival/selection strategy of grazing and browsing herbivores. A diverse microbial population may provide the opportunity to respond quickly to various quantities and qualities of forage available to the horse depending on seasonal and geographical factors. However, seasonal changes in feral horses happen more gradually in comparison to their domesticated counterparts. A recent study suggested that domestication of the wild horse has dramatically changed its faecal microbial composition [4]. The faecal microbiota of Equus ferus przewalskii was more diverse compared to domestic horses, which was attributed to the more diverse plant material that wild living horses are likely to ingest. Forage-based diets were associated with a reduction in lactic-acid bacteria and Streptococcus bovis/equinus and more stable microbiota in the horse [5], whilst the more diverse the microbial population is, the more adaptable it is to new feed sources. Domesticated horses are sensitive to environmental stress and may develop drastic metabolic disturbances, such as colic and laminitis in response to sudden dietary changes [6, 7]. There is clearly an association between development of colic and changes in the relative abundance of microbial taxa [8, 9]. Equine nutritionists usually recommend a gradual transition (7–14 days) [10] during any dietary change. It is, however, relatively unknown how long it takes for the gastrointestinal microbiota to adapt to a new diet. Gaining a better understanding of what bacterial populations inhabit the equine gut when horses are fed certain feedstuffs and how those populations are altered following abrupt dietary changes is important to improve gastrointestinal health of the horse.

Nowadays, there is an increasing number of studies focused on the effects of abrupt changes in the diet on hindgut and faecal microbiota in horses under different dietary conditions [11–16]. Most of those studies focus on dietary transition when rapidly fermentable carbohydrates were incorporated into the diet. Forage-only based diets appear to mainly benefit the performance of the horse. Little is known about the effect of abrupt dietary transitions of forage-only diets. Regardless of the similarity in the nutrient composition of hays, differences in hindgut microbiota were seen after the abrupt hay change, suggesting a sensitive response of the hindgut microbiota [13]. To date, there is no information available on the effects of abrupt changes from grass (pasture) to a hay-only diet, and vice versa. These dietary challenges mimic real-world scenarios that horse owners are likely to encounter. To the best of our knowledge, a single published article that incorporated dietary change from pasture to good-quality and poor-quality hay was predominantly focused on reporting the effect of three diets (pasture, “good hay” and “bad hay”) on faecal microbiota of healthy horses and horses with asthma 3 weeks following adaptation to each diet [17]. Thus, the aim of this study was to investigate changes in equine faecal microbial population dynamics over time in response to an abrupt change from grass to hay and vice versa.

Materials and methods

Animals and experimental design

The experimental procedures in this study were approved by the University of Glasgow’s School of Veterinary Medicine Veterinary Ethics & Welfare Committee (Equine hindgut health, nutrition and microbiota–Ref. 05a/14).

The effects of an abrupt dietary transition from grass to hay and vice versa on the faecal microbiota of ponies were investigated using Illumina next generation sequencing (NGS) methodologies. The experimental trial consisted of two, 14-day periods. Prior to commencing the study, six adult Welsh Section A gelding ponies were grazing for a one-month period (May) on the same pasture as used during the trial. The experiment began during early summer (June). Day 0 (G-D0) was recorded as the last day on pasture before the first transition to stables. Diets comprised of 100% hay (H) diet fed at 17.5 g/kg body weight (BW) on a dry matter basis. Following day 14 of the first experimental period, the second abrupt change (from hay to grass) was implemented. During the second experimental period animals were grazed 100% on pasture (predominately ryegrass sward (Lolium perenne) (G) ad libitum. Nutritional analysis of the grass and hay were not performed. The detailed experimental design is provided in the Fig 1.

Faecal material was collected immediately (<2 minutes after being voided) between 1330 h and 1630 h on day 0 and days 1, 2, 3, 7, 14, of each experimental period using a spatula with minimum environmental contamination (spatula was sterilised between sampling from individual ponies to prevent cross contamination) and were frozen immediately (<2 minutes) at -20°C until downstream laboratory analysis.

DNA extraction, generation of 16S amplicon libraries and sequencing

DNA extraction from faecal material was performed using Qiagen QIAamp® Fast DNA Stool Mini Kit (Qiagen, Manchester, England, UK) with slight adaptations [18]. The 16S library preparation, sequencing and data analysis were performed at Glasgow Polyomics, University of Glasgow using an adapted protocol based on the protocol provided by Illumina (San Diego, CA, USA). Nextera XT DNA Library Prep Kit was used for library preparation workflow. Briefly, 16S metagenomics protocol comprised of the main steps listed below. The V3-V4 region of the 16S rRNA gene was amplified and Illumina index primers attached in a 2-step PCR process. In the first PCR, primers targeting the V3-V4 region of the 16S rRNA gene were used: forward 5’-CTTACGGGNGGCWGCAG-3’ and reverse 5’-GACTACHVGGGTATCTAATCC-3’. The primers were designed with overhanging adapters for annealing to Illumina primers with Nextera identifier indices and sequencing adaptors. The index primers were attached during the second PCR step. Prepared libraries were stored at 4°C until use. These amplicon libraries were run simultaneously and sequenced on an Illumina MiSeq platform using an Illumina Miseq 600 cycle cartridge v3. 5% PhiX adapter-ligated library was used as a control for Illumina sequencing runs for quality control.

Bioinformatics and statistical analysis

Each individual animal represented an experimental unit. The Illumina output was analysed with 16S metagenomics pipeline using Quantitative Insights Into Microbial Ecology (QIIME) v1.9.0 [19]. First, the Cutadapt Python package was used to search for the primers and adapters in all reads and remove them [20]. Cutadapt was also used to trim poor quality bases (using a Q-score 25 as a cut off) and to remove short reads (less than 250 bp long). Second, Pandaseq was used for merging paired-end reads into a single read. Third, add_qiime_labels.py QIIME script was used to create a single .fasta file with valid QIIME labels. Next, pick_open_reference_otus.py was used to perform open-reference operational taxonomic units (OTU) picking. After picking OTUs, a representative set of sequences was selected. Then, align_seqs.py script was used to align the sequences using PyNAST [21]. The NAST algorithm aligned each provided “candidate” sequence to the best-matching sequence in a pre-aligned database of “template” sequences. The template file was Greengenes core set (16S rRNA gene database)—available from http://greengenes.lbl.gov/. Further on, filter_fasta.py was used to remove chimeric sequences. Consequently, make_otu_table.py was used to tabulate the number of times an OTU was found in each sample and add the taxonomic predictions for each OTU. The script make_phylogeny.py was used to produce a phylogenetic tree relating the OTU. The script core_diversity_analyses.py used QIIME diversity analyses together to form a basic workflow beginning with a BIOM table, mapping file and phylogenetic tree. Koeken.py is a wrapper around the linear discriminant analysis effect size (LEfSe) script, it was used to create a convenient workflow from a QIIME OTU table to biomarker microbial taxa.

LEfSe is a tool [22] which was used to find biomarkers specific to each collection day using relative abundance (RA). Firstly, the non-parametric factorial Kruskal-Wallis sum-rank test was used to detect features with significant differential abundance with respect to the class of interest; subsequently, a set of pairwise tests among subclasses was carried out using the unpaired Wilcoxon rank-sum test. As a last step, LEfSe used linear discriminant analysis to estimate the effect size of each differentially abundant feature. The differences in microbial RA were considered significant when LDA score (log10) >2 [22]. LEfSe was performed using the OTU table with all the taxonomic levels and day of sampling as class.

α –diversity measures were used to show which microbial populations were the most compositionally diverse and which had most phylotypes or branch lengths [23, 24]. OTU or phylotype, is usually used instead of “species” for a cluster of related 16S rRNA sequences in 16S rRNA sequencing studies [25].

α –diversity of the microbial communities was evaluated using Chao1 richness non-parametric estimator (non-phylogenetic metric), PD_whole_tree (phylogenetic metric) and observed OTUs. The PD_whole_tree metric, also known as Faith's Phylogenetic Diversity, is based on the phylogenetic tree. It sums up all the branch lengths as a measure of diversity: if a new OTU is closely related to another OTU in the sample, there will be a small increase in diversity. Conversely, if a new OTU originates from a completely different lineage compared to all other OTUs in the sample, it will contribute significantly to increase in the diversity.

This 16S rRNA gene sequencing technique does not use reads long enough to identify bacterial and archaeal sequences down to species taxonomic level. Hence, any OTUs which were revealed to be significantly different in relation to diet were traced back to their corresponding genera.

To analyse β-diversity, weighted and unweighted UniFrac metrics in QIIME were used [26, 27]. β-diversity metric calculates the distance between a pair of samples and shows which factors correlate with differences in microbiota composition. In weighted UniFrac branch length were weighed by relative abundance, while unweighted UniFrac were based purely on sequence distance measure. UniFrac distance was calculated by constructing a phylogenetic tree from all the OTUs [27]. For each pair of samples, the distance was estimated as a (sum of unshared branch lengths)/ (sum of total branch lengths).

principal_coordinates.py and make_emperor.py were used to create principal coordinate analysis (PCoA) plots and to generate charts [28]. Clustering of samples was evaluated by plotting the resultant vector of the PCoA with 3 dimensions. The false discovery rate (FDR) P-value was used to adjust for multiple testing (Benjamini-Hochberg procedure). FDR P-value cutoff was preset at <0.1.

Results

Sequencing quality

Faecal samples (n = 66) from 6 ponies on 11 collection days were sequenced and generated 4,895,038 total reads. A total of 4,777,315 reads for the pooled microbial communities (bacterial and archaeal), which passed through quality filters, were approved for further analysis. The reads were assigned into OTU and 2,232 OTU observations in total were obtained with the counts of reads per sample (mean ± SD) 72,384 ± 14,322. A rarefication depth of 2,000 reads per sample was used in order to normalise the number of reads across all samples. The sub-sampling size was considered adequate, as evidenced by the plateau of rarefaction curves (S1–S3 Figs). OTUs with less than 100 reads were excluded from the analysis, leaving 2,221 OTUs in total.

α-diversity

There were no significant differences in phylogenetic diversity (PD_whole_tree), estimated total species richness (Chao_1), and richness expressed as the number of observed OTUs (Observed_OTUs) between different sampling days (S1–S3 Figs). However, it can be seen from Fig 2, that for H-D1 and G-D14 Chao1 diversity metrics were substantially higher and showed greater variance compared to H-D14 (H-D1 95% CI = 898, 1181; H-D14 95% CI = 737, 1033).

Fig 2 — Boxplot comparing estimated richness (n = 6). Red bars show abrupt changes: 1^st from grass (G) to hay (H), 2^nd from hay to grass. Blue line connects the means (orange lines). Circles represent outliers. D: days.

A PD_whole tree (S4 Fig) revealed that the microbial diversity was different between the following pairs of animals: pony 1 and pony 5, pony 2 and pony 4, pony 4 and pony 5, and pony 5 and pony 6 (P = 0.015).

β-diversity

Principal coordinate analysis on a phylogenetic weighted UniFrac dissimilarity matrix revealed no clustering of equine faecal samples by dietary group when all experimental days were compared (Fig 3A); however, there was clustering observed on H-D1 and G-D1 (Fig 3B) and H-D3 and G-D3 (Fig 3C) with more than half (52.19%) of the variation explained by principal coordinates (PC) 1, 2 and 3. Meanwhile, there was no clustering of equine faecal samples by dietary group observed on day 7 or 14 after the abrupt dietary change.

Fig 3 — (A) Days 1, 2, 3, 7, 14 on grass and hay (H-D1, H-D2, H-D3, H-D7, H-D14, G-D1, G-D2, G-D3, G-D7, G-D14) (B) Day 1 after abrupt change on grass and hay (H-D1, G-D1) (C) Day 3 after abrupt change on grass and hay (H-D3, G-D3). OTUs with <100 reads were excluded from the analysis. n = 6. H: hay; G: grass; D: day; PCA: Principal coordinate analysis.

Principal coordinate analysis on an unweighted-UniFrac dissimilarity matrix revealed no clustering of equine faecal samples by dietary group on any of the sampling days, with only 18.40% of the variation explained by principal coordinates (PC) 1, 2 and 3 (Fig 4).

Fig 4 — (A) Days 1, 2, 3, 7, 14 on grass and hay (H-D1, H-D2, H-D3, H-D7, H-D14, G-D1, G-D2, G-D3, G-D7, G-D14) (B) Day 1 after abrupt change on grass and hay (H-D1, G-D1) (C) Day 3 after abrupt change on grass and hay (H-D3, G-D3). OTUs with <100 reads were excluded from the analysis. n = 6. H: hay; G: grass; D: day; PCA: Principal coordinate analysis.

Relative abundance at various taxonomic levels

Within the domain Bacteria there were 18 phyla identified, while there was only 1 phylum detected within the domain Archaea. Bacterial phyla encompassed 32 different classes, 41 orders, 65 families and 109 genera. Archaeal phylum incorporated 3 different classes, 4 orders, 4 families and 5 genera. The detailed structure of the microbial populations is provided in S1 Table. Several microorganisms were detected, which have not been previously assigned in the Greengenes database.

LEfSe analysis revealed 46 biomarkers, associated with sample collection days (G-D0, H-D1, H-D2, H-D3, H-D7, H-D14, G-D1, G-D2, H-D3, H-D7, H-D14) (LDA score >2, P <0.05). Overall, there were 1 phylum, 4 classes, 6 orders, 11 families and 24 genera identified as biomarkers for one of the sample collection days and those are listed in Tables 1 and 2.

Table 1. Linear discriminate analysis effect size (LEfSe) showing significant taxa (biomarkers, in bold) in the faeces of the ponies for each sampling day (D) when ponies (n = 6) were maintained on hay (H).

Phylum>Class>Order>Family>Genus	LDA	P-value	Biomarker
Firmicutes>Clostridia>Clostridiales>Lachnospiraceae>Shuttleworthia	4.9	0.004	D1-H
Firmicutes>Clostridia>Clostridiales>Lachnospiraceae>Butyrivibrio	4.1	0.005	D2-H
Verrucomicrobia>Opitutae	4.4	<0.001	D2-H
>_Cerasicoccales_	4.5	<0.001	D2-H
>_Cerasicoccaceae_	4.6	<0.001	D2-H
>^*	4.3	<0.001	D2-H
Firmicutes>Clostridia>Clostridiales>Syntrophomonadaceae>Syntrophomonas	4.5	0.044	D3-H
>Veillonellaceae	5.9	0.003	D3-H
>^*	5.3	0.002	D3-H
>Phascolarctobacterium	5.8	0.003	D3-H
>Syntrophomonadaceae	4.5	0.044	D3-H
>Clostridiaceae>^*	4.9	0.049	D3-H
>Lachnospiraceae	6.6	0.029	D3-H
>Lachnospiraceae>^*	6.5	0.038	D3-H
>Lachnospiraceae>Dorea	4.5	0.011	D3-H
>Lachnospiraceae>Blautia	5.7	0.005	D3-H
>Lachnospiraceae>Anaerostipes	5.0	0.024	D3-H
Firmicutes>Erysipelotrichi>Erysipelotrichales>Erysipelotrichaceae>^*	5.0	<0.001	D14-H
> _Other	4.5	0.001	D14-H
Firmicutes>Clostridia>Clostridiales>Gracilibacteraceae	4.4	0.043	D14-H
>Lutispora	4.4	0.043	D14-H

Open in a new tab

* indicates unclassified taxonomic level. LDA: linear discriminate analysis score.

Table 2. Linear discriminate analysis effect size (LEfSe) showing significant taxa (biomarkers, in bold) in the faeces of the ponies for each sampling day (D) when ponies (n = 6) were maintained on grass (G).

Phylum>Class>Order>Family>Genus	LDA	P-value	Biomarker
Bacteroidetes>Bacteroidia>Bacteroidales>^*	7.1	0.010	D1-G
Bacteroidetes>Bacteroidia>Bacteroidales>^>^	7.1	0.010	D1-G
>_Paraprevotellaceae_>CF231	5.8	0.038	D1-G
Proteobacteria>Deltaproteobacteria>Desulfovibrionales	4.8	0.039	D1-G
> >Desulfovibrionaceae	4.8	0.039	D1-G
Firmicutes>Clostridia>Clostridiales>Lachnospiraceae>Roseburia	4.8	0.004	D1-G
Spirochaetes	6.7	0.002	D1-G
>Spirochaetes	6.7	0.001	D1-G
>Spirochaetales	6.7	0.001	D1-G
>Spirochaetaceae	6.7	0.001	D1-G
>Treponema	6.7	0.001	D1-G
Cyanobacteria>Chloroplast	4.8	0.014	D2-G
>Streptophyta	5.0	0.014	D2-G
>^*	5.0	0.014	D2-G
>^*	5.0	0.014	D2-G
Firmicutes>Clostridia>Clostridiales>Clostridiaceae>Sarcina	4.4	<0.001	D2-G
>Lachnospiraceae>Lachnobacterium	4.6	0.017	D2-G
Firmicutes>Bacilli	4.5	0.016	D2-G
>Lactobacillales	4.5	0.016	D2-G
>Lactobacillaceae	4.5	0.016	D2-G
>Lactobacillus	4.5	0.016	D2-G
Actinobacteria>Coriobacteriia>Coriobacteriales>Coriobacteriaceae>Slackia	4.5	0.002	D2-G
Proteobacteria>Alphaproteobacteria>Rickettsiales>mitochondria_Other	4.7	0.006	D3-G
Firmicutes>Clostridia>Clostridiales_Other_Other	5.6	0.017	D3-G
Actinobacteria>Coriobacteriia>Coriobacteriales>Coriobacteriaceae>Adlercreutzia	5.3	0.020	D3-G

Open in a new tab

* indicates unclassified taxonomic level. D: day; H: hay; G: grass; LDA: linear discriminate analysis score.

Though there was no clustering of equine faecal samples by dietary group on day 14 observed with PCoA, LEfSe analysis revealed 63 biomarkers that were different on day 14 between grass and hay diets (LDA score >2, P <0.05). Overall, there were 2 phyla, 3 classes, 9 orders, 18 families and 31 genera that were identified as different on day 14 when ponies were receiving hay or grass diets (H-D14 and G-D14).

Phylum level

There were 19 phyla identified among 66 samples. Within the domain Bacteria there were 7 phyla identified in the faeces of the ponies, with an average relative abundance across all samples of >1%. LEfSe identified significant differences in phylum Actinobacteria and phylum Spirochaetes on H-D14 and G-D14 (LDA score >2, P <0.05) (S5 Fig).

Two phyla, Bacteroidetes and Firmicutes dominated the bacterial community regardless of the diet or sampling time. Interestingly, on G-D0 (after ponies were grazing for a month before the first experimental period began), H-D14 and G-D14, Bacteroidetes was the most abundant phyla with Firmicutes being the second. However, on H-D1, H-D2, H-D3, G-D2 and G-D3 Firmicutes was identified as the most abundant phylum followed by Bacteroidetes (Fig 5).

Fig 5 — Bacterial phyla with relative abundance >5% are presented. Error bars represent standard error of the mean. Red vertical lines represent abrupt changes in the diet. D: day; H: hay; G: grass.

Genus level

There were 114 genera identified among 66 samples. There were 7 genera with an average RA across all samples higher than 5%; these 7 genera represented 73% of the samples’ genus RA on average. There were 26 genera identified with a RA >0.5% and, with the exception of unassigned genus from family Paraprevotellaceae, these 26 most abundant genera were present in the samples from all ponies regardless of diet or sampling day and represented 94% of the total RA at genus level. Among those genera were Fibrobacter, Ruminococcus, Treponema, Paludibacter, Prevotella, YRC22, CF231, BF311, Clostridium, RFN20, Blautia, and Phascolarctobacterium.

The most abundant genus belonged to the phylum Bacteroidetes, class Bacteroidia, order Bacteroidales unassigned at family and genus level and its RA across all samples was 22.8 ± 0.62% (mean ± SEM). The dynamics of relative abundance in the 1^st to 7^th most abundant genera (average RA across all sampling days >5%) after abrupt changes in the diets is depicted in Fig 6.

Fig 6 — Bacterial genera with average relative abundance (RA) >5% are presented. Error bars represent standard error of the mean. Red vertical lines represent abrupt changes in the diet. Unassigned genera were numbered in descending order based on their RA. D: day; H: hay; G: grass. All genera were traced back to their correspondent phylum > class > order > family > genus. 1: Bacteroidetes > Bacteroidia > Bacteroidales > unassigned > unassigned. 2: Verrucomicrobia > Verruco-5 > WCHB1-41 > RFP12 > unassigned. 3: Firmicutes > Clostridia > Clostridiales > Ruminococcaceae > unassigned. 4: Firmicutes > Clostridia > Clostridiales > Lachnospiraceae > unassigned. Treponema: Spirochaetes > Spirochaetes > Spirochaetales > Spirochaetaceae > Treponema. Fibrobacter: Fibrobacteres > Fibrobacteria > Fibrobacterales > Fibrobacteraceae > Fibrobacter. 7: Firmicutes > Clostridia > Clostridiales > unassigned > unassigned.

Discussion

This study was conducted to evaluate whether an abrupt change from grass to hay and vice versa would have an impact on faecal microbial communities in ponies. Such changes have been previously reported to be a risk factor for digestive upsets and colic manifestations in epidemiological studies [29, 30].

Alpha-diversity metrics were not different between diets and experimental days. This indicated that the differences observed in the PCoA plots were likely due to differences in the RA of some taxonomic levels, as the numbers of taxa and the evenness did not change significantly between faecal samples from ponies fed either diet. This does not, however, exclude the possibility of a radical shift in taxa without an overall change in taxa numbers or evenness, although the taxonomic distribution strongly suggests this is not the case. This is in agreement with the results reported by other researchers [12], and may be due to both types of forage (grass and hay) generally having a substantial fibre content, which is beneficial for the proliferation of fibrolytic bacteria: crude fibre—330 g/kg and 264 g/kg, NDF– 670 g/kg and 400–570 g/kg for grass hay and grass, respectively [31].

The clustering observed in the first few days after the dietary change suggests that a shorter more intense sampling period around the time of dietary transition should be considered. Future studies on the faecal microbiota of horses should consider more frequent sampling to gain better understanding of when any structural adaptation takes place. Furthermore, it is important to consider the differences in compositional adaptation and functional adaptation of gut microbiota. Compositional adaptation (alterations in structure) is known to be rapid, while functional adaptation (full adaptation to the new diet) may take place over an extended period of time. Previous research suggests that horse gut microbiota can adapt in response to new diets quickly within 4–6 days [12, 32]. This is in agreement with the results of the current study where according to PCoA and LEfSe, most of the differences in microbial community structure were observed during the first 3 days after dietary change. These results suggest a rapid microbial community response. Rapid adaptation to dietary changes have also been reported in human research. David, Maurice et al., reported that the human gut microbiome can revert to its original structure within two to four days post dietary transition from a plant-based diet to an animal-based diet or vice versa, which may reflect selection processes of human evolution [33]. However, there are an insufficient number of studies reporting functional adaptation of the equine gut microbiome to new dietary and environmental conditions. The change between two silages with different crude protein content resulted in altered colonic bacterial counts within the first 24 h after the abrupt change was introduced [15]. Moreover, days 7 to 22 after abrupt change were associated with a decrease in pH and an increase in volatile fatty acids concentration [15]. Hence, both short-term and long-term changes may occur in response to abrupt dietary changes in equine diets. Long-term functional changes require further investigation.

A recent study by Leclere and Costa [17] examined the effect of pasture, high-quality and poor-quality hay on healthy horses and horses with asthma. There were 2 genera overrepresented in faeces of the healthy horses maintained on pasture (Coprococcus and unclassified genus from family Enterobacteriaceae), 4 on good-quality hay (Denitrobacterium, Cellulosilyticum, unclassified genus belonging to Deltaproteobacteria class and unclassified genus belonging to order Sphingobacteriales), and none on dusty hay. However, the effect of abrupt dietary change from pasture to hay or vice versa was not discussed in the aforementioned study as samples were collected 3 weeks after diets were introduced to horses.

The current study found Bacteroidetes was the most abundant phylum in the faeces of ponies maintained on hay or grass on H-D7, H-D14, G-D7, and G-D14, closely followed by Firmicutes. However, short term after abrupt dietary change Firmicutes phylum generally dominated. This finding may suggest that Firmicutes are more adept at adjusting to abrupt dietary changes regardless of the order, compared to Bacteroidetes. Furthermore, it can be speculated that the Bacteroidetes:Firmicutes ratio was altered not by the diet itself (hay or grass), but by the abrupt change introduced to the diet. The proportions of Firmicutes and Bacteroidetes have been shown to be indicators of digestive health in humans [34], where obese people had fewer Bacteroidetes and more Firmicutes than their leaner counterparts. The same pattern was observed when obese and lean horses were studied [35]. Costa, Arroyo [36] reported that in healthy horses Firmicutes predominated (68%) followed by Bacteroidetes (14%), while in horses with colitis Bacteroidetes (40%) was the most abundant phylum followed by Firmicutes (30%). It appears that the Firmicutes to Bacteroidetes ratio might be an important predictor of metabolic changes in the GIT of the horse. The existing knowledge is, however, limited and does not allow us to interpret an increase or decrease in Firmicutes to Bacteroidetes ratio as a positive or negative change. The results of the current study are in disagreement with the findings of Grimm, Philippeau [14] who observed an increase in Bacteroidetes in the caecum of the horse on day 1 after abrupt dietary changes of two different hays. This may be due to different diets used in both studies, using faeces as a source of microbial exploration as opposed to caecum contents, as well as possible inter-breed variations, indicating that further research is required to validate current findings.

In the current study, the RA of faecal class Bacilli, order Lactobacillales, family Lactobacillaceae and genus Lactobacillus was increased on day 2 after the abrupt dietary change from hay to grass compared to all other sampling days. This finding suggests that an abrupt dietary change from hay to grass may represent a higher risk for hindgut pH to drop compared to abrupt change from grass to hay as Lactobacillus is known to be a major lactic acid producing bacteria. Conversely, in the study of Muhonen, Julliand [16] a change from hay to haylage or silage from the same swath harvested on the same date did not induce alterations in bacterial counts after 28 hours. However, volatile fatty acids and pH were associated with a modification of the colonic lactobacilli and streptococci bacterial counts on days 8 and 21 after the abrupt change took place, suggesting a long-term alteration in the hindgut environment, rather than the short-term changes observed in the current study.

There were differences observed between samples from each individual pony. Significant inter-horse variation in gut microbiota has been previously reported [37–39]. Regardless of the changes observed in the faecal microbiota of ponies as a result of the abrupt change in diet, none of the animals in this current study developed any clinical signs of digestive upset. This finding suggests that compositional differences in gut microbiota alone cannot explain the development of metabolic disorders, such as colic and laminitis, but may provide insights into potential mechanisms of development of these conditions. This hypothesis warrants further investigation using a larger animal sample size. It is likely that the adaptive capacity of equine gut microbiota in response to dietary changes depends on each individual animal. This might explain why some horses are more prone to the development of metabolic disorders in response to similar dietary changes compared to other individuals. More research on individual traits that may predispose horses to the development of metabolic disorders would be an essential next step for further work.

The ponies used in the current study were of the same sex and breed and were maintained on the same farm and fed the same diets during the experimental period. This is in contrast to other microbiome studies that involved horses maintained on various diets and managed differently [1, 39–41]. It is possible that by reducing individual animal variation under controlled conditions in this relatively small sample size, has enabled us to identify the differences truly related to the abrupt change from grass to hay and vice versa.

In the current experiment grass was fed ad libitum while hay intake was restricted to 17.5 g/kg BW on dry matter basis. Hence, this study investigated not only the change of the diets but also incorporated a change in the amount of feed offered. This choice was made with the purpose of replicating the real-world scenario by which horse owners are likely to manage their horses. Moreover, it is important to remember that the change was not only related to diet, but a complex interaction of social-behaviour change, change in exercise, climate effects attributed to differences in management of the stabled horse and horse maintained on pasture. Furthermore, some of the bacterial community structural changes observed in the grazing ponies may be a result of changes in the composition of grass and environmental conditions. Weather conditions, such as temperature and rainfall, influence horse faecal microbiota [42]. This effect can be attributed to the direct influence of weather on horses or an indirect influence of weather or soil microbiota and other environmental bacteria, which after being ingested, may affect equine gut microbiota. A limitation of this study design is the small sample size (6 animals) and lack of control group that remained on pasture. As mentioned previously, variables such as temperature, pasture composition, change in environment (housing) could have had an effect on their microbiota. If environmental factors were majorly responsible for the observed changes in microbial relative abundance, the microbiota of G-D14 would be most likely to be dissimilar to G-D0. In our study, according to LEfSe analysis there were no biomarkers identified for G-D0 or G-D14. According to weighted and unweighted UniFrac PCoA plots, there was no clustering observed on G-D0 and G-D14 (S6 Fig). Thus, it is likely that abrupt dietary changes were major factors leading to changes in faecal microbiota of the ponies abruptly transitioned from grass to hay and from hay to grass.

The OTU is a vague term that may be defined as an individual organism, a named taxonomic group such as a species or genus, or a group with undetermined evolutionary relationships that share a given set of observed characters [43]. In NGS studies, an OTU is typically defined as a cluster of reads with 97% similarity, expecting that these correspond approximately to species [44]. Correspondence of OTUs with species may fail because of various factors. Firstly, some species have genes that are >97% similar, resulting in merged OTU containing multiple species. Secondly, a single species may have paralogs that are <97% similar, causing the species to be split across two or more OTUs. Finally, some clusters may be false due to artefacts including read errors and chimeras. Due to the concerns discussed above, an exact correspondence between OTU and species cannot be determined. Therefore, when the Greengenes database attempted to place reads into a phylogenetic tree and then assign names to parts of that tree, sequences belonging to mitochondria and chloroplast rRNA were dispositioned. For the purposes of this study, the full taxonomy of the mitochondria and chloroplast reads were ignored, as they were from mitochondria and chloroplast and did not evolve from the Bacteria kingdom.

Abrupt changes from grass to hay and vice versa affected the faecal microbial community structure. Moreover, the order of dietary change had a profound effect in the first few days following the transition. There is a requirement to evaluate nutritional strategies with the potential to mitigate the effect of abrupt dietary changes on GIT microbiota. Clearly, further studies are required to identify a tool for supporting a more stable GIT environment in the horse during abrupt dietary transitions. Determining the role of microbial species in nutrient breakdown and their contribution to metabolism is fundamental to understanding digestive processes and maximising productivity, health and welfare of the horse; however, 16S rRNA gene sequencing only allows us to identify compositional changes in microbial populations. Results from this study can be used to further investigate the role of specific microbes that were affected by the abrupt dietary changes and their contribution to equine microbiota in terms of their immediate response to the change and longer-term adaptation to different diets.

Supporting information

S1 Fig. Rarefaction curves for bacterial communities in the faeces of the ponies (n = 6) showing α-diversity metric PD_whole_tree as a function of sequences per sample.

Each curve represents sampling day (D) on grass (G) or hay (H).

(TIF)

Click here for additional data file.^{(453.5KB, tif)}

S2 Fig. Rarefaction curves for bacterial communities in the faeces of the ponies (n = 6) showing α-diversity metric Chao1 as a function of sequences per sample.

Each curve represents sampling day (D) on grass (G) or hay (H).

(TIF)

Click here for additional data file.^{(489.6KB, tif)}

S3 Fig. Rarefaction curves for bacterial communities in the faeces of the ponies (n = 6) showing α-diversity metric observed OTUs as a function of sequences per sample.

Each curve represents sampling day (D) on grass (G) or hay (H).

(TIF)

Click here for additional data file.^{(511KB, tif)}

S4 Fig. Boxplots comparing faecal microbial α-diversity metric (PD_whole_tree) for each individual pony (P).

n: number of samples per pony.

(TIF)

Click here for additional data file.^{(199.4KB, tif)}

S5 Fig. Differential features in the faeces of the ponies (n = 6) fed grass and hay diets on day 14 (H-D14 and G-D14) at phylum level.

Solid and dashed horizontal lines indicate mean and median across all samples, respectively.

(TIF)

Click here for additional data file.^{(581.6KB, tif)}

S6 Fig

3-dimensional Principal Coordinate Analyses plot (on a phylogenetic weighted (A) unweighted (B) UniFrac dissimilarity matrix) for data on the microbial membership and structure in the faeces of ponies (n = 6) across H-D14 and G-D14. There was no clustering observed. OTUs with <100 reads were excluded from the analysis. H: hay; G: grass; D: day.

(TIF)

Click here for additional data file.^{(133.5KB, tif)}

S1 Table. Faecal bacterial and archaeal communities of mature Welsh Section A geldings (n = 6) during the study investigating abrupt change of the hay and grass diets.

The bacterial and archaeal taxa identified in the faeces of ponies in the present study are listed in the table according to the taxonomic ranks using Greengenes database.

(DOCX)

Click here for additional data file.^{(30KB, docx)}

Data Availability

All data files are available from the https://doi.org/10.5525/gla.researchdata.986

Funding Statement

The authors gratefully acknowledge the scholarship (for A.Garber) and funding support for this study from the Trustees of the Ronald Miller Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.O' Donnell MM, Harris HMB, Jeffery IB, Claesson MJ, Younge B, O' Toole PW, et al. The core faecal bacterial microbiome of Irish Thoroughbred racehorses. Lett Appl Microbiol. 2013;57(6):492–501. 10.1111/lam.12137 [DOI] [PubMed] [Google Scholar]
2.Shepherd ML, Swecker WS, Jensen RV, Ponder MA. Characterization of the fecal bacteria communities of forage-fed horses by pyrosequencing of 16S rRNA V4 gene amplicons. FEMS Microbiol Lett. 2012;326(1):62–68. 10.1111/j.1574-6968.2011.02434.x [DOI] [PubMed] [Google Scholar]
3.Steelman SM, Chowdhary BP, Dowd S, Suchodolski J, Janečka JE. Pyrosequencing of 16S rRNA genes in fecal samples reveals high diversity of hindgut microflora in horses and potential links to chronic laminitis. BMC Vet Res. 2012;8:231 10.1186/1746-6148-8-231 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Metcalf JL, Song SJ, Morton JT, Weiss S, Seguin-Orlando A, Joly F, et al. Evaluating the impact of domestication and captivity on the horse gut microbiome. Sci Rep. 2017;7(1):15497 10.1038/s41598-017-15375-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Willing B VÖRÖS A, Roos S, Jones C, Jansson A, Lindberg JE. Changes in faecal bacteria associated with concentrate and forage-only diets fed to horses in training. Equine Vet J. 2009;41(9):908–914. 10.2746/042516409x447806 [DOI] [PubMed] [Google Scholar]
6.Bailey SR, Marr CM, Elliott J. Current research and theories on the pathogenesis of acute laminitis in the horse. Equine Vet J. 2004;167:129–142. [DOI] [PubMed] [Google Scholar]
7.Hudson JM, Cohen ND, Gibbs PG, Thompson JA. Feeding practices associated with colic in horses. JAVMA. 2001;219(10):1419–1425. 10.2460/javma.2001.219.1419 [DOI] [PubMed] [Google Scholar]
8.Stewart HL, Southwood LL, Indugu N, Vecchiarelli B, Engiles JB, Pitta D. Differences in the equine faecal microbiota between horses presenting to a tertiary referral hospital for colic compared with an elective surgical procedure. Equine Vet J. 2019;51(3):336–342. 10.1111/evj.13010 [DOI] [PubMed] [Google Scholar]
9.Weese JS, Holcombe SJ, Embertson RM, Kurtz KA, Roessner HA, Jalali M, et al. Changes in the faecal microbiota of mares precede the development of post partum colic. Equine Vet J. 2015;47(6):641–649. 10.1111/evj.12361 [DOI] [PubMed] [Google Scholar]
10.Harris P, Dunnett C. Nutritional tips for veterinarians. Equine Vet Educ. 2018;30(9):486–496. [Google Scholar]
11.de Fombelle A, Julliand V, Drogoul C, Jacotot E. Feeding and microbial disorders in horses: 1-effects of an abrupt incorporation of two levels of barley in a hay diet on microbial profile and activities. J Equine Vet Sci. 2001;21(9):439–445. [Google Scholar]
12.Fernandes KA, Kittelmann S, Rogers CW, Gee EK, Bolwell CF, Bermingham EN, et al. Faecal Microbiota of Forage-Fed Horses in New Zealand and the Population Dynamics of Microbial Communities following Dietary Change. PLoS ONE. 2014;9(11):e112846 10.1371/journal.pone.0112846 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Grimm P, Julliand V, Philippeau C, Sadet-Bourgeteau S. Effect of yeast supplementation on hindgut microbiota and digestibility of horses subjected to an abrupt change of hays. Livest Sci. 2016;186:34–40. [Google Scholar]
14.Grimm P, Philippeau C, Julliand V. Faecal parameters as biomarkers of the equine hindgut microbial ecosystem under dietary change. animal. 2017;11(7):1136–1145. 10.1017/S1751731116002779 [DOI] [PubMed] [Google Scholar]
15.Muhonen S, Connysson M, Lindberg JE, Julliand V, Bertilsson J, Jansson A. Effects of crude protein intake from grass silage-only diets on the equine colon ecosystem after an abrupt feed change. J Anim Sci. 2008;86(12):3465–3472. 10.2527/jas.2007-0374 [DOI] [PubMed] [Google Scholar]
16.Muhonen S, Julliand V, Lindberg J, Bertilsson J, Jansson A. Effects on the equine colon ecosystem of grass silage and haylage diets after an abrupt change from hay. J Anim Sci. 2009;87:2291–2298. 10.2527/jas.2008-1461 [DOI] [PubMed] [Google Scholar]
17.Leclere M, Costa MC. Fecal microbiota in horses with asthma. J Vet Intern Med. 2020;34(2):996–1006. 10.1111/jvim.15748 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Bulmer LS, Murray J-A, Burns NM, Garber A, Wemelsfelder F, McEwan NR, et al. High-starch diets alter equine faecal microbiota and increase behavioural reactivity. Sci Rep. 2019;9(1):18621 10.1038/s41598-019-54039-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7: 335–336. 10.1038/nmeth.f.303 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnetjournal. 2011;17(1):10–12. [Google Scholar]
21.Caporaso JG, Bittinger K, Bushman FD, DeSantis TZ, Andersen GL, Knight R. PyNAST: a flexible tool for aligning sequences to a template alignment. Bioinformatics. 2010;26(2):266–267. 10.1093/bioinformatics/btp636 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, et al. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12(6):R60 10.1186/gb-2011-12-6-r60 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Costa MC, Weese JS. Understanding the Intestinal Microbiome in Health and Disease. Vet Clin N Am-Equine. 2018;34(1):1–12. [DOI] [PubMed] [Google Scholar]
24.Salem S. Epidemiological and Microbiome Studies of Equine Colic: University of Liverpool; 2018. Available from: https://livrepository.liverpool.ac.uk/3005140/ [Google Scholar]
25.Lozupone CA, Knight R. Species divergence and the measurement of microbial diversity. FEMS Microbiol Rev. 2008;32(4):557–578. 10.1111/j.1574-6976.2008.00111.x [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Lozupone C, Knight R. UniFrac: a New Phylogenetic Method for Comparing Microbial Communities. Appl and Environ Microb. 2005;71(12):8228–8235. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Lozupone C, Lladser ME, Knights D, Stombaugh J, Knight R. UniFrac: an effective distance metric for microbial community comparison. ISME J. 2011;5(2):169–172. 10.1038/ismej.2010.133 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Vazquez-Baeza Y, Pirrung M, Gonzalez A, Knight R. EMPeror: a tool for visualizing high-throughput microbial community data. Gigascience. 2013;2(1):16 10.1186/2047-217X-2-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Goncalves S, Julliand V, Leblond A. Risk factors associated with colic in horses. Vet Res. 2002;33(6):641–652. 10.1051/vetres:2002044 [DOI] [PubMed] [Google Scholar]
30.Al Jassim RAM, Andrews FM. The bacterial community of the horse gastrointestinal tract and its relation to fermentative acidosis, laminitis, colic, and stomach ulcers. Vet Clin N Am-Equine. 2009;25(2):199–215. [DOI] [PubMed] [Google Scholar]
31.Frape D. Equine Nutrition and Feeding. 4 ed UK: Wiley-Blackwell; 2013. 498 p. [Google Scholar]
32.van den Berg M, Hoskin SO, Rogers CW, Grinberg A. Fecal pH and Microbial Populations in Thoroughbred Horses During Transition from Pasture to Concentrate Feeding. J Equine Vet Sci. 2013;33(4):215–222. [Google Scholar]
33.David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014;505(7484):559–563. 10.1038/nature12820 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Larsen N, Vogensen FK, van den Berg FW, Nielsen DS, Andreasen AS, Pedersen BK, et al. Gut microbiota in human adults with type 2 diabetes differs from non-diabetic adults. PLoS One. 2010;5(2):e9085 10.1371/journal.pone.0009085 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Biddle AS, Tomb JF, Fan Z. Microbiome and Blood Analyte Differences Point to Community and Metabolic Signatures in Lean and Obese Horses. Front Vet Sci. 2018;5:225 10.3389/fvets.2018.00225 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Costa MC, Arroyo LG, Allen-Vercoe E, Stampfli HR, Kim PT, Sturgeon A, et al. Comparison of the fecal microbiota of healthy horses and horses with colitis by high throughput sequencing of the V3-V5 region of the 16S rRNA gene. PLoS ONE. 2012;7(7):e41484 10.1371/journal.pone.0041484 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Blackmore TM, Dugdale A, Argo CM, Curtis G, Pinloche E, Harris PA, et al. Strong stability and host specific bacterial community in faeces of ponies. PLoS ONE. 2013;8(9):e75079 10.1371/journal.pone.0075079 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Proudman C, Darby A, Escalona E. Faecal microbiome of the Thoroughbred racehorse and its response to dietary amylase supplementation. Equine Vet J. 2014;46(S46):35. [Google Scholar]
39.Costa MC, Weese JS. The equine intestinal microbiome. Anim Health Res Rev. 2012;13(01):121–128. [DOI] [PubMed] [Google Scholar]
40.Dougal K, de la Fuente G, Harris PA, Girdwood SE, Pinloche E, Newbold CJ. Identification of a core bacterial community within the large intestine of the horse. PLoS ONE. 2013;8(10):e77660 10.1371/journal.pone.0077660 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Dougal K, Harris PA, Edwards A, Pachebat JA, Blackmore TM, Worgan HJ, et al. A comparison of the microbiome and the metabolome of different regions of the equine hindgut. FEMS Microbiol Ecol. 2012;82(3):642–652. 10.1111/j.1574-6941.2012.01441.x [DOI] [PubMed] [Google Scholar]
42.Salem SE, Maddox TW, Berg A, Antczak P, Ketley JM, Williams NJ, et al. Variation in faecal microbiota in a group of horses managed at pasture over a 12-month period. Sci Rep. 2018;8(1):8510 10.1038/s41598-018-26930-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Sokal RR. The Principles and Practice of Numerical Taxonomy. Taxon. 1963;12(5):190–199. [Google Scholar]
44.Edgar RC. Accuracy of microbial community diversity estimated by closed- and open-reference OTUs. Peer-reviewed J. 2017;5:e3889. [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0237869.r001

Decision Letter 0

Juan J Loor

27 May 2020

PONE-D-20-08059

Abrupt dietary changes between grass and hay alter faecal microbiota of ponies

PLOS ONE

Dear Dr. Garber,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jul 11 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Juan J Loor

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere:

'An abstract based on the part of the current study was presented at EWEN (European Workshop in Equine Nutrition) 2018 in Sweden as oral presentation and published in the proceedings. This research study is a part of PhD thesis which is currently embargoed for 2 years. '

Please clarify whether this conference proceeding was peer-reviewed and formally published.

If this work was previously peer-reviewed and published, in the cover letter please provide the reason that this work does not constitute dual publication and should be included in the current manuscript.

3. We note you have included tables to which you do not refer in the text of your manuscript. Please ensure that you refer to Tables 3 and 4 in your text; if accepted, production will need this reference to link the reader to the Table.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dear Editor and authors.

This is a descriptive study evaluating changes in the fecal microbiota of 6 ponies undergoing diet change. The article is not completely novel, but brings interesting information to contribute to the current knowledge of factors affecting the intestinal microbiota of horses. The major limitation with the study design is the lack of control group that remained in pasture: variables such as temperature, pasture composition, change in environment (housing) could have affected their microbiota. This needs to be clearely addressed in the discussion. Also, authors give too much importance for the microbial variations at the phylum level, which is important, much less informative and meaningful. For example, which of the thousands species of Firmicutes increased after diet change? These genera should be presented in the main article (and not as supplementary data). As you will see in my comments below the article needs major rewriting to improve clarity.

Abstract:

The abstract needs major changes. You need to clearly state the study design including days of sampling, details on diet change, number of animals, etc. Many readers are not familiarized with concepts like LEfSe, LDA, PCoA, community structure, etc. You should use this space wisely to transmit the main findings and significance of your study in a way that most people will be able to understand it.

Line 53-56: I am not sure this statement should be used to justify your study. While fiber based diets likely benefit the horse microbiota, this is not feasible in performance horses in which high energy supplementation is required to fulfill their energy needs.

Line 59: please refer to Leclere and Costa, J Vet Intern Med. 2020 Mar;34(2):996-1006. doi: 10.1111/jvim.15748.

Material and methods:

What was the composition and quality of hay (any analysis done?)? What was the grass type/variety?

Table 1: it would be better to see that in a timeline rather than in a Table. You should not use the same nomenclature (e.g. D1) for both periods (e.g. line 170, 209, 294, etc.). In fact, every time you write “abrupt dietary change” you should specify to which of the changes you are referring to.

Line 130: what is a phylotype?

Line 132: what is a OTU? Please define those terms. Did you use a phylotype or OTU approach?

Line 147: adjusted using which test?

Line 157: why were low abundance sequences excluded? Was this done before or after alfa diversity calculation? This will likely impact Chao index.

Results:

Line 160: is phylogenetic diversity a measure of Alpha diversity? Please indicate which test was used for that in the material and methods.

Line 163: Please indicate the units after the values.

Please provide details how you performed LEfSe: using the OTU table with all the taxonomic levels? using day of sampling as class (or samples from G and H analysed separately, or performed pairwise comparisons?). Wouldn’t make more sense to use only the different genera?

Results:

Line 154: counts of reads per sample or of OTUs per sample?

Line 157: where are the rarefaction curves? This analysis can be quite subjective and I recommend that Good’s coverage is presented instead.

Figure 1: could indicate the diets and day of change, just like you did in Figure 3. Please indicate what circles represent (outliers?).

Line 180: OTUs

Figure 2: how can you distinguish the different days in figure 2A? Any clustering seen by individual?

Line 136: you mentioned that Illumina sequencing is not suitable for species level analysis (and you are completely right), so in my opinion it is not adequate to present this data, as it might mislead readers to erroneous information.

Line 188: were those really unassigned organisms or it could be products of sequencing error not detected during bioinformatics? This should be acknowledged especially considering the approach adopted here (sequencing of the V3-V4 regions), which considerably decreases the overlapping and therefore, increases sequencing errors:

https://mothur.org/blog/2014/Why-such-a-large-distance-matrix/

Line 190: by “day of dietary change” you mean D1H and D1G?

Line 192: please insert the number of the supplementary table.

Line 205: is this a sub-item of line 183?

Line 208: what does “p.” mean?

Figures S5 to S11 could be combined in one figure like you did in Fig 3. In fact, this new figure should be included as a main figure of the manuscript because its data (genus level) is much more meaningful than phylum level analysis.

Line 240: Table 4? I think this table along with OTUs data should be removed.

Line 242: different based on which test? LEfSe?

Line 256: please revise this statement. Alpha diversity indices do not take into account taxonomic information )membership), but just the number of different taxa.

Line 327: you are using only 6 animals in your study and this might be an overstatement.

Line 368: which diversity index was used to investigate diversity? It seems that only richness was included in your alpha diversity analysis.

Line 364–375: this paragraph is mainly repeating what was already said and might not be necessary.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2020 Aug 18;15(8):e0237869. doi: 10.1371/journal.pone.0237869.r002

Author response to Decision Letter 0

11 Jul 2020

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

A: I have added full affiliation including commune in France for one of the authors. I cannot find any other inconsistences with the template, could you please direct me to those if you notice anything else?

2. We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere:

Please clarify whether this conference proceeding was peer-reviewed and formally published.

The reason that this work does not constitute dual publication was included in the cover letter.

A: The following was included in the cover letter: “The abstract published in conference proceedings contains 350 words (0.5 page) compared to 7105 words (15 pages) in the current manuscript. To the best of our knowledge, it is a common practice to present preliminary results at the conferences before submitting complete manuscripts for publication. Based on our previous experience, this practice is usually well accepted by the journals. Please, also note, that the abstract published on conference proceedings does not repeat the abstract of this article.”

A: Reference to Table 3 and Table 4 was added (though table numbering was changed throughout the whole manuscript as reviewer has requested to replace one of the tables with figure).

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

3. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: No

A: Please, check Datacite DOI: 10.5525/gla.researchdata.986

If there anything else that needs to be included, could you please let me know what exactly that is?

4. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: No

A: The manuscript has been checked for typographical and grammatical errors and corrections have been made.

5. Review Comments to the Author

Reviewer #1: Dear Editor and authors.

This is a descriptive study evaluating changes in the fecal microbiota of 6 ponies undergoing diet change. The article is not completely novel, but brings interesting information to contribute to the current knowledge of factors affecting the intestinal microbiota of horses. The major limitation with the study design is the lack of control group that remained in pasture: variables such as temperature, pasture composition, change in environment (housing) could have affected their microbiota. This needs to be clearely addressed in the discussion. A: This was included and is now clearly addressed in the discussion. Also, authors give too much importance for the microbial variations at the phylum level, which is important, much less informative and meaningful. For example, which of the thousands species of Firmicutes increased after diet change? These genera should be presented in the main article (and not as supplementary data). As you will see in my comments below the article needs major rewriting to improve clarity.

Abstract:

Line 59: please refer to Leclere and Costa, J Vet Intern Med. 2020 Mar;34(2):996-1006. doi: 10.1111/jvim.15748. A: Thanks, this was published March 2020, same month as we have submitted our manuscript, this is why I was not aware of this article on the day of submission. I have included this article to improve our discussion section and changed introduction to make it up to date.

Material and methods:

What was the composition and quality of hay (any analysis done?)? What was the grass type/variety? A: Unfortunately, nutritional analysis of the grass and hay were not performed. I have included this statement in Material and Methods to make it clear for the reader same as was done in a study of Leclere and Costa, J Vet Intern Med. 2020 Mar;34(2):996-1006. In fact, it is clear now to me that it would be better to do nutritional analysis of the feeds in all diet-related studies and not only digestibility studies.

Line 130: what is a phylotype? A: Operational Taxonomic Units (OTU) or phylotype, is usually used instead of “species” for a cluster of related 16S rRNA sequences in 16S rRNA sequencing studies. This was clarified in the text and reference provided.

Line 132: what is a OTU? Please define those terms. A: OTU was defined above, further explanation is given in discussion section. Did you use a phylotype or OTU approach? A: OTU approach

Line 147: adjusted using which test? A: Benjamini-Hochberg procedure, this was included in the manuscript

Can you comment on the reliability of those primers on classifying Archaea based on a scientific publication? It is my impression that primers designed to target bacteria are not adequate for Archaea analysis. A: This was expected because the primers were targeted on the V3/V4 region of bacterial rRNA subunit. V3/V4 region primers provide good general amplification, they are designed to pick up bacterial species not archaeal; however, they are not designed to specifically exclude archaea.

Line 157: why were low abundance sequences excluded? Was this done before or after alpha diversity calculation? This will likely impact Chao index. A: Low abundance sequences were excluded to reduce the influence of rare genera/species on the overall results. Alpha rarefaction was performed before and after removal with very little difference.

Results:

Line 160: is phylogenetic diversity a measure of Alpha diversity? A: PD_whole_tree is a phylogenetic metric of alpha-diversity. This is also know as Faith's Phylogenetic Diversity (PD_whole tree means phylogenetic diversity using the whole “tree”) (more is included in M&M section). Please indicate which test was used for that in the material and methods. A: Observed OTUs, Chao_1 and PD_whole_tree tests were used to assess alpha-diversity. This is included in M&M and corresponding Figures are presented in supplementary material (S1-S3) and Figure 1.

Line 163: Please indicate the units after the values. A: Chao1 is based on the number of OTUs and is a measure of species richness. It has no units. The coherent unit for dimensionless quantities, also termed quantities of dimension one is symbol ”1” which is usually omitted. (Source: https://www.bipm.org/utils/common/pdf/si_brochure_8_en.pdf). These may also be referred to as arbitrary units.

Please provide details how you performed LEfSe: using the OTU table with all the taxonomic levels? (A: yes, this was including in the manuscript) using day of sampling as class (A: yes, day of sampling as class, this was including in the manuscript) (or samples from G and H analysed separately, or performed pairwise comparisons?). “LEfSe was performed using the OTU table with all the taxonomic levels and day of sampling as class.”

Wouldn’t make more sense to use only the different genera? A: This can be performed, but often misses higher level taxonomic features as biomarkers, for example where a family level marker is significantly differential but this is spread across several genera which are not independently differential.

Results:

Line 154: counts of reads per sample or of OTUs per sample? A: Counts of reads per sample, this was clarified in the text.

Line 157: where are the rarefaction curves? A: The rarefaction curves are presented in Supplementary figures S1-S3. This analysis can be quite subjective and I recommend that Good’s coverage is presented instead. A: Good’s coverage is 1 - the number of singleton OTUs (single count) divided by the total abundance for all OTUs. Since all singletons are discarded during the analysis this metric is not really relevant or in fact calculable on this data.

Figure 1: could indicate the diets and day of change, just like you did in Figure 3. Please indicate what circles represent (outliers?). A: This was changed as requested.

Line 177: is the weighted Unifrac really evaluating microbial membership and structure or only structure. In my understanding, membership is evaluated by unweighted Unifrac. A: Unweighted uses the number of different species/genera to determine differences between samples. Weighted factors in abundances for each species/genera. In other words, in weighted UniFrac branch length were weighed by relative abundance, while unweighted UniFrac were based purely on sequence distance measure. So in one sense unweighted is a better estimator of membership if you are using a binary definition – i.e. it is or it isn’t). Unweighted UniFrac plots were added (Fig 3).

Line 180: OTUs A: Done

Figure 2: how can you distinguish the different days in figure 2A? Any clustering seen by individual?

A: Figure 2A (now 3A after numbering has changed) is there to show that there was no clustering by diet (grass vs hay) when all sampling days (G-D0, H-D1, H-D2, H-D3, H-D7, H-D14, G-D1, G-D2, H-D3, H-D7, H-D14) are taken into account. You cannot distinguish between sampling days, just between diets. Yes, the following pairs of individual animals were different: 1 and 6, 2 and 6, 3 and 5, 3 and 6, 5 and 6 (weighted UniFrac). Unweighted UniFrac: 1 and 3, 1 and 5, 2 and 3, 2 and 4, 2 and 5, 3 and 4, 3 and 5, 3 and 6, 4 and 5, 4 and 6, 5 and 6. I don’t think that adding this information would be relevant and add to the manuscript, but I can do that if you suggest it would be better to do so.

A: I totally agree with what you say. Generally, we present genera level data. While it is not suitable for overall spp level analysis it can identify some spp accurately dependent on specific OTUs. Therefore, using the data to show species level differences is not particularly sound – but demonstrating the difference/presence of a one specific spp could be used. OTU data was previously used in other publications, but I removed it because I see your point. (Bulmer LS, Murray J-A, Burns NM, Garber A, Wemelsfelder F, McEwan NR, et al. High-starch diets alter equine faecal microbiota and increase behavioural reactivity. Scientific Reports. 2019;9(1):18621.

A: The level of sequencing error is not particularly high in Illumina data. Yes, some of these unassigned OTUs could be sequencing errors, they likewise could be novel bacteria, sequence variants or a number of other possibilities. The V3/V4 region is less than 450bp combined which means there is a 75bp overlap (even after trimming reads less than 250bp are discarded meaning the minimum overlap is 50bp) which is a very significant overlap therefore sequencing errors are unlikely.

https://mothur.org/blog/2014/Why-such-a-large-distance-matrix/

A: An interesting point and read, this is particularly important for their method of analysis (via Mothur) which clusters using the entire sequence. Qiime on the other hand clusters based on OTUs which minimises the influence of “off by one” and SNP variants eluded to in the blog post.

Line 190: by “day of dietary change” you mean D1H and D1G? A: This was clarified, thanks. “Day of dietary change” was replaced with “sample collection days” (G-D0, H-D1, H-D2, H-D3, H-D7, H-D14, G-D1, G-D2, H-D3, H-D7, H-D14). Same nomenclature is now used in the whole manuscript and all tables and figures.

Line 192: please insert the number of the supplementary table. A: This was done, thanks for noticing.

Line 205: is this a sub-item of line 183? A: Yes, thanks, this was corrected.

Line 208: what does “p.” mean? A: phylum, this was changed

Line 240: Table 4? I think this table along with OTUs data should be removed. A: This has been removed.

Line 242: different based on which test? LEfSe? A: PERMANOVA, but this was deleted as reviewer has suggested

Line 256: please revise this statement. Alpha diversity indices do not take into account taxonomic information )membership), but just the number of different taxa. A: this statement has been revised and the following was included instead: “This indicated that the differences observed in the PCoA plots were likely due to differences in the RA of some taxonomic levels, as the numbers of taxa and the evenness does not change significantly between faecal samples from ponies fed either diet. This does not, however, exclude the possibility of a radical shift in taxa without an overall change in taxa numbers or evenness, although the taxonomic distribution strongly suggests this is not the case”.

Line 327: you are using only 6 animals in your study and this might be an overstatement. A: this amended according to your recommendation. “This hypothesis warrants further investigation using a larger animal sample size.”

Line 368: which diversity index was used to investigate diversity? It seems that only richness was included in your alpha diversity analysis. A: PD_whole_tree was used as a measure of phylogenetic diversity. It was better explained in a M&M section.

Line 364–375: this paragraph is mainly repeating what was already said and might not be necessary. A part of this paragraph was deleted from the discussion section according to reviewer’s suggestion, another part was moved to the very end of the discussion section as it refers to future research that needs to be undertaken.

One of the first studies using NGS to evaluate microbiota (maybe from Turnbaugh’s group) showed similar results: marked changes during the first day after abrupt diet change with subsequent recovery of the microbiota. A: I think I found the article you were referring to and included it in our discussion section. Thanks a lot for your suggestion. (David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014;505(7484):559-63.

Ideally you should have a control group, bu if diet (or other environmental factors) was responsible for the observed microbiota changes, would you expect that the microbiota of D14G should be similar to D0? Was that the case? A: The comment regarding the absence of the control group remaining on pasture was added. If environmental factors were majorly responsible for the observed changes in microbial relative abundance, the microbiota of G-D14 most likely would not be similar to G-D0. In our study, according to LEfSe analysis there were no biomarkers identified for G-D0 or G-D14 (Table 2). According to weighted and unweighted UniFrac PCoA plots, there was no clustering observed on G-D0 and G-D14 (S6 Fig). Thus, it is likely that abrupt dietary changes were major factors leading to changes in faecal microbiota of the ponies abruptly transitioned from grass to hay and from hay to grass. (This was added to the manuscript).

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(33.3KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0237869.r003

Decision Letter 1

Juan J Loor

5 Aug 2020

Abrupt dietary changes between grass and hay alter faecal microbiota of ponies

PONE-D-20-08059R1

Dear Dr. Garber,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Juan J Loor

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

**********

6. Review Comments to the Author

Reviewer #1: Thank you very much for your clarifications and changes in the manuscript. It has largely improved the manuscript.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Marcio Costa

PLoS One. doi: 10.1371/journal.pone.0237869.r004

Acceptance letter

Juan J Loor

10 Aug 2020

PONE-D-20-08059R1

Abrupt dietary changes between grass and hay alter faecal microbiota of ponies

Dear Dr. Garber:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Juan J Loor

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. Rarefaction curves for bacterial communities in the faeces of the ponies (n = 6) showing α-diversity metric PD_whole_tree as a function of sequences per sample.

Each curve represents sampling day (D) on grass (G) or hay (H).

(TIF)

Click here for additional data file.^{(453.5KB, tif)}

S2 Fig. Rarefaction curves for bacterial communities in the faeces of the ponies (n = 6) showing α-diversity metric Chao1 as a function of sequences per sample.

Each curve represents sampling day (D) on grass (G) or hay (H).

(TIF)

Click here for additional data file.^{(489.6KB, tif)}

S3 Fig. Rarefaction curves for bacterial communities in the faeces of the ponies (n = 6) showing α-diversity metric observed OTUs as a function of sequences per sample.

Each curve represents sampling day (D) on grass (G) or hay (H).

(TIF)

Click here for additional data file.^{(511KB, tif)}

S4 Fig. Boxplots comparing faecal microbial α-diversity metric (PD_whole_tree) for each individual pony (P).

n: number of samples per pony.

(TIF)

Click here for additional data file.^{(199.4KB, tif)}

S5 Fig. Differential features in the faeces of the ponies (n = 6) fed grass and hay diets on day 14 (H-D14 and G-D14) at phylum level.

Solid and dashed horizontal lines indicate mean and median across all samples, respectively.

(TIF)

Click here for additional data file.^{(581.6KB, tif)}

S6 Fig

(TIF)

Click here for additional data file.^{(133.5KB, tif)}

S1 Table. Faecal bacterial and archaeal communities of mature Welsh Section A geldings (n = 6) during the study investigating abrupt change of the hay and grass diets.

The bacterial and archaeal taxa identified in the faeces of ponies in the present study are listed in the table according to the taxonomic ranks using Greengenes database.

(DOCX)

Click here for additional data file.^{(30KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(33.3KB, docx)}

Data Availability Statement

All data files are available from the https://doi.org/10.5525/gla.researchdata.986

[pone.0237869.ref001] 1.O' Donnell MM, Harris HMB, Jeffery IB, Claesson MJ, Younge B, O' Toole PW, et al. The core faecal bacterial microbiome of Irish Thoroughbred racehorses. Lett Appl Microbiol. 2013;57(6):492–501. 10.1111/lam.12137 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref002] 2.Shepherd ML, Swecker WS, Jensen RV, Ponder MA. Characterization of the fecal bacteria communities of forage-fed horses by pyrosequencing of 16S rRNA V4 gene amplicons. FEMS Microbiol Lett. 2012;326(1):62–68. 10.1111/j.1574-6968.2011.02434.x [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref003] 3.Steelman SM, Chowdhary BP, Dowd S, Suchodolski J, Janečka JE. Pyrosequencing of 16S rRNA genes in fecal samples reveals high diversity of hindgut microflora in horses and potential links to chronic laminitis. BMC Vet Res. 2012;8:231 10.1186/1746-6148-8-231 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref004] 4.Metcalf JL, Song SJ, Morton JT, Weiss S, Seguin-Orlando A, Joly F, et al. Evaluating the impact of domestication and captivity on the horse gut microbiome. Sci Rep. 2017;7(1):15497 10.1038/s41598-017-15375-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref005] 5.Willing B VÖRÖS A, Roos S, Jones C, Jansson A, Lindberg JE. Changes in faecal bacteria associated with concentrate and forage-only diets fed to horses in training. Equine Vet J. 2009;41(9):908–914. 10.2746/042516409x447806 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref006] 6.Bailey SR, Marr CM, Elliott J. Current research and theories on the pathogenesis of acute laminitis in the horse. Equine Vet J. 2004;167:129–142. [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref007] 7.Hudson JM, Cohen ND, Gibbs PG, Thompson JA. Feeding practices associated with colic in horses. JAVMA. 2001;219(10):1419–1425. 10.2460/javma.2001.219.1419 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref008] 8.Stewart HL, Southwood LL, Indugu N, Vecchiarelli B, Engiles JB, Pitta D. Differences in the equine faecal microbiota between horses presenting to a tertiary referral hospital for colic compared with an elective surgical procedure. Equine Vet J. 2019;51(3):336–342. 10.1111/evj.13010 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref009] 9.Weese JS, Holcombe SJ, Embertson RM, Kurtz KA, Roessner HA, Jalali M, et al. Changes in the faecal microbiota of mares precede the development of post partum colic. Equine Vet J. 2015;47(6):641–649. 10.1111/evj.12361 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref010] 10.Harris P, Dunnett C. Nutritional tips for veterinarians. Equine Vet Educ. 2018;30(9):486–496. [Google Scholar]

[pone.0237869.ref011] 11.de Fombelle A, Julliand V, Drogoul C, Jacotot E. Feeding and microbial disorders in horses: 1-effects of an abrupt incorporation of two levels of barley in a hay diet on microbial profile and activities. J Equine Vet Sci. 2001;21(9):439–445. [Google Scholar]

[pone.0237869.ref012] 12.Fernandes KA, Kittelmann S, Rogers CW, Gee EK, Bolwell CF, Bermingham EN, et al. Faecal Microbiota of Forage-Fed Horses in New Zealand and the Population Dynamics of Microbial Communities following Dietary Change. PLoS ONE. 2014;9(11):e112846 10.1371/journal.pone.0112846 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref013] 13.Grimm P, Julliand V, Philippeau C, Sadet-Bourgeteau S. Effect of yeast supplementation on hindgut microbiota and digestibility of horses subjected to an abrupt change of hays. Livest Sci. 2016;186:34–40. [Google Scholar]

[pone.0237869.ref014] 14.Grimm P, Philippeau C, Julliand V. Faecal parameters as biomarkers of the equine hindgut microbial ecosystem under dietary change. animal. 2017;11(7):1136–1145. 10.1017/S1751731116002779 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref015] 15.Muhonen S, Connysson M, Lindberg JE, Julliand V, Bertilsson J, Jansson A. Effects of crude protein intake from grass silage-only diets on the equine colon ecosystem after an abrupt feed change. J Anim Sci. 2008;86(12):3465–3472. 10.2527/jas.2007-0374 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref016] 16.Muhonen S, Julliand V, Lindberg J, Bertilsson J, Jansson A. Effects on the equine colon ecosystem of grass silage and haylage diets after an abrupt change from hay. J Anim Sci. 2009;87:2291–2298. 10.2527/jas.2008-1461 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref017] 17.Leclere M, Costa MC. Fecal microbiota in horses with asthma. J Vet Intern Med. 2020;34(2):996–1006. 10.1111/jvim.15748 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref018] 18.Bulmer LS, Murray J-A, Burns NM, Garber A, Wemelsfelder F, McEwan NR, et al. High-starch diets alter equine faecal microbiota and increase behavioural reactivity. Sci Rep. 2019;9(1):18621 10.1038/s41598-019-54039-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref019] 19.Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7: 335–336. 10.1038/nmeth.f.303 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref020] 20.Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnetjournal. 2011;17(1):10–12. [Google Scholar]

[pone.0237869.ref021] 21.Caporaso JG, Bittinger K, Bushman FD, DeSantis TZ, Andersen GL, Knight R. PyNAST: a flexible tool for aligning sequences to a template alignment. Bioinformatics. 2010;26(2):266–267. 10.1093/bioinformatics/btp636 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref022] 22.Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, et al. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12(6):R60 10.1186/gb-2011-12-6-r60 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref023] 23.Costa MC, Weese JS. Understanding the Intestinal Microbiome in Health and Disease. Vet Clin N Am-Equine. 2018;34(1):1–12. [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref024] 24.Salem S. Epidemiological and Microbiome Studies of Equine Colic: University of Liverpool; 2018. Available from: https://livrepository.liverpool.ac.uk/3005140/ [Google Scholar]

[pone.0237869.ref025] 25.Lozupone CA, Knight R. Species divergence and the measurement of microbial diversity. FEMS Microbiol Rev. 2008;32(4):557–578. 10.1111/j.1574-6976.2008.00111.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref026] 26.Lozupone C, Knight R. UniFrac: a New Phylogenetic Method for Comparing Microbial Communities. Appl and Environ Microb. 2005;71(12):8228–8235. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref027] 27.Lozupone C, Lladser ME, Knights D, Stombaugh J, Knight R. UniFrac: an effective distance metric for microbial community comparison. ISME J. 2011;5(2):169–172. 10.1038/ismej.2010.133 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref028] 28.Vazquez-Baeza Y, Pirrung M, Gonzalez A, Knight R. EMPeror: a tool for visualizing high-throughput microbial community data. Gigascience. 2013;2(1):16 10.1186/2047-217X-2-16 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref029] 29.Goncalves S, Julliand V, Leblond A. Risk factors associated with colic in horses. Vet Res. 2002;33(6):641–652. 10.1051/vetres:2002044 [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref030] 30.Al Jassim RAM, Andrews FM. The bacterial community of the horse gastrointestinal tract and its relation to fermentative acidosis, laminitis, colic, and stomach ulcers. Vet Clin N Am-Equine. 2009;25(2):199–215. [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref031] 31.Frape D. Equine Nutrition and Feeding. 4 ed UK: Wiley-Blackwell; 2013. 498 p. [Google Scholar]

[pone.0237869.ref032] 32.van den Berg M, Hoskin SO, Rogers CW, Grinberg A. Fecal pH and Microbial Populations in Thoroughbred Horses During Transition from Pasture to Concentrate Feeding. J Equine Vet Sci. 2013;33(4):215–222. [Google Scholar]

[pone.0237869.ref033] 33.David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature. 2014;505(7484):559–563. 10.1038/nature12820 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref034] 34.Larsen N, Vogensen FK, van den Berg FW, Nielsen DS, Andreasen AS, Pedersen BK, et al. Gut microbiota in human adults with type 2 diabetes differs from non-diabetic adults. PLoS One. 2010;5(2):e9085 10.1371/journal.pone.0009085 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref035] 35.Biddle AS, Tomb JF, Fan Z. Microbiome and Blood Analyte Differences Point to Community and Metabolic Signatures in Lean and Obese Horses. Front Vet Sci. 2018;5:225 10.3389/fvets.2018.00225 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref036] 36.Costa MC, Arroyo LG, Allen-Vercoe E, Stampfli HR, Kim PT, Sturgeon A, et al. Comparison of the fecal microbiota of healthy horses and horses with colitis by high throughput sequencing of the V3-V5 region of the 16S rRNA gene. PLoS ONE. 2012;7(7):e41484 10.1371/journal.pone.0041484 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref037] 37.Blackmore TM, Dugdale A, Argo CM, Curtis G, Pinloche E, Harris PA, et al. Strong stability and host specific bacterial community in faeces of ponies. PLoS ONE. 2013;8(9):e75079 10.1371/journal.pone.0075079 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref038] 38.Proudman C, Darby A, Escalona E. Faecal microbiome of the Thoroughbred racehorse and its response to dietary amylase supplementation. Equine Vet J. 2014;46(S46):35. [Google Scholar]

[pone.0237869.ref039] 39.Costa MC, Weese JS. The equine intestinal microbiome. Anim Health Res Rev. 2012;13(01):121–128. [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref040] 40.Dougal K, de la Fuente G, Harris PA, Girdwood SE, Pinloche E, Newbold CJ. Identification of a core bacterial community within the large intestine of the horse. PLoS ONE. 2013;8(10):e77660 10.1371/journal.pone.0077660 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref041] 41.Dougal K, Harris PA, Edwards A, Pachebat JA, Blackmore TM, Worgan HJ, et al. A comparison of the microbiome and the metabolome of different regions of the equine hindgut. FEMS Microbiol Ecol. 2012;82(3):642–652. 10.1111/j.1574-6941.2012.01441.x [DOI] [PubMed] [Google Scholar]

[pone.0237869.ref042] 42.Salem SE, Maddox TW, Berg A, Antczak P, Ketley JM, Williams NJ, et al. Variation in faecal microbiota in a group of horses managed at pasture over a 12-month period. Sci Rep. 2018;8(1):8510 10.1038/s41598-018-26930-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0237869.ref043] 43.Sokal RR. The Principles and Practice of Numerical Taxonomy. Taxon. 1963;12(5):190–199. [Google Scholar]

[pone.0237869.ref044] 44.Edgar RC. Accuracy of microbial community diversity estimated by closed- and open-reference OTUs. Peer-reviewed J. 2017;5:e3889. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Abrupt dietary changes between grass and hay alter faecal microbiota of ponies

Anna Garber

Peter Hastie

David McGuinness

Pauline Malarange

Jo-Anne Murray

Roles

Abstract

Introduction

Materials and methods

Animals and experimental design

Fig 1. Experimental design employed to study the effect of abrupt dietary changes on the faecal microbiota of ponies (n = 6).

DNA extraction, generation of 16S amplicon libraries and sequencing

Bioinformatics and statistical analysis

Results

Sequencing quality

α-diversity

Fig 2. Chao1 for bacterial communities in the faeces of the ponies for different days of experiment.

β-diversity

Fig 3. PCA on a weighted-UniFrac dissimilarity matrix for data on the microbial membership and structure in the faeces of ponies.

Fig 4. PCA on an unweighted-UniFrac dissimilarity matrix for data on the microbial membership and structure in the faeces of ponies.

Relative abundance at various taxonomic levels

Table 1. Linear discriminate analysis effect size (LEfSe) showing significant taxa (biomarkers, in bold) in the faeces of the ponies for each sampling day (D) when ponies (n = 6) were maintained on hay (H).

Table 2. Linear discriminate analysis effect size (LEfSe) showing significant taxa (biomarkers, in bold) in the faeces of the ponies for each sampling day (D) when ponies (n = 6) were maintained on grass (G).

Phylum level

Fig 5. Bacterial phyla dynamics in the faeces of the ponies during the abrupt dietary transitions.

Genus level

Fig 6. Bacterial genera dynamics in the faeces of the ponies (n = 6) during the abrupt transitions.

Discussion

Supporting information

Data Availability

Funding Statement

References

Decision Letter 0

Juan J Loor

Roles

Author response to Decision Letter 0

Decision Letter 1

Juan J Loor

Roles

Acceptance letter

Juan J Loor

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases