Effect of different dietary regimes on gut microbiota in water buffalo calves during the preweaning period

Abstract

Background

Early-life nutrition plays a crucial role in shaping gut microbiota and growth performance in ruminants. In water buffalo calves, the pre-weaning period is particularly critical, as the gastrointestinal tract undergoes rapid development and is highly susceptible to environmental and dietary influences. This study aimed to characterize fecal microbiota during early life and evaluate the effects of different milk-based diets on microbial development and growth performance.

Results

Calves exhibited low microbial diversity and homogeneous microbiota at birth, dominated by Proteobacteria, Firmicutes, and Bacteroidota, with no differences among groups, while microbial richness and evenness significantly increased in all groups during the early weeks of life, reflecting microbiota maturation (Observed features: p = 0.0042, p = 0.0266, and p = 0.0002 for WB, FM, and MD respectively; Shannon entropy: p ≤ 0.0031 for all groups). Beta diversity analysis revealed significant differences in microbial community structure between diet-based groups, particularly between water buffalo milk and formula-fed calves. The mixed diet group showed the highest beta diversity dispersion and temporal variation and exhibited the greatest weight gain, which was significantly higher than that observed in the formula-fed group. Genera traditionally considered beneficial, such as Lactobacillus increased across all groups, while Bifidobacterium was more abundant in formula and mixed diets. Functional analysis revealed enhanced metabolic activity associated with nutrient processing and immune-related pathways, particularly in calves fed natural milk.

Conclusions

Early dietary intervention significantly influences gut microbiota development and growth in water buffalo calves. A mixed milk feeding strategy appears to optimize both microbial maturation and growth performance, suggesting its potential as an effective management practice in the pre-weaning period.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12866-026-05356-2.

Introduction

The rearing of Italian Mediterranean buffalo is a major economic driver for Italy, thanks to the PDO supply chain for “Mozzarella di bufala Campana”. Italy accounts for 86% of Europe’s buffalo population (https://www.fao.org/faostat/en/#data/QCL), and of this, 70% of the herd is reared in the Campania region as reported in national animal registry database established by the Ministry of Health at the national service center for animal registration of the “G. Caporale” Experimental Zooprophylactic Institute in Teramo (year of consultation 2026), generating a consumer value of 891 million euros in 2024 [1]. In water buffalo farming, raising healthy calves is crucial, as it significantly impacts growth and milk production performance in adulthood. Optimum calf management represents the cornerstone for enhancing herd productivity and ensuring the long-term profitability of dairy enterprises. Indeed, growth trajectories during the pre- and post-weaning phases exert a direct influence on both age and live body weight (LBW) at puberty and first service; this biological mechanism is well-documented in both cattle [2] and water buffaloes [3]. Nevertheless, water buffalo calves exhibit a higher degree of physiological immaturity at birth compared to bovine calves, a trait characteristically reflected in lower initial milk intake and delayed dental eruption [4]. Furthermore, owing to their tropical ancestry, buffaloes possess a limited tolerance to cold climates, making them significantly more susceptible to neonatal disorders and adverse environmental stressors than other domestic ruminants [5]. In the Italian context, calving is strategically distributed throughout the entire year to ensure a continuous and steady supply of milk, satisfying the constant market demand driven by mozzarella cheese production. In water buffalo farming, raising healthy calves is crucial, as it significantly impacts growth and milk production performance in adulthood. Indeed, proper calf development represents a key factor for the success of the dairy industry. During the first weeks of life, the gastrointestinal tract (GIT) of calves undergoes profound morphological and metabolic physiological changes [6]. Following birth, calves are exposed to several stress factors, including the transition from the uterine environment to external conditions and the shift from maternal nutrition to independent digestion and nutrient absorption. These challenges make the newborn calves particularly vulnerable to gastrointestinal disorders, such as neonatal calf diarrhea, which frequently occurs during the first four weeks of life.

The health status and growth of calves are heavily influenced by the composition and activity of their gut microbiota [7]. This microbiota undergoes rapid development in the early weeks of life and plays an essential role in preventing gastrointestinal disorders. A well-balanced gut microbiota not only inhibits the colonization of pathogens but also supports proper physiological development, enhancing reproductive performance and milk production [8]. In ruminant species, several studies have been focused on the rumen microbiota, highlighting the central role of microorganisms on digestion, especially in adult animals [9]. However, calves are functional monogastric until weaning, therefore the digestive activity supported by microbial flora in the intestine plays a prominent role in digestion of nutrients. For these reasons, research studies in this area are essential to advance knowledge and promote quality and innovation in primary production.

Gut microbiota plays a critical role in early life development, influencing both immune system maturation and metabolic health [10]. Dietary intake is a primary factor shaping gut microbial diversity and composition [11]. High-throughput sequencing has been shown to be a suitable approach for studying the gut microbiota of newborn calves, allowing characterization of the intestinal microbiota during the pre-weaning phase. In recent years, high-throughput sequencing technologies have been increasingly applied to investigate the gastrointestinal system of water buffaloes, exploring both rumen microbial establishment and fecal microbiota dynamics under different management and dietary conditions [12, 13]. Monitoring changes in gut microbiota composition during this period, together with the characterization of microbial communities, may represent a useful strategy for identifying the most suitable dietary management practices capable of promoting calf health, optimal growth, and proper development. The aims of this study were (i) to characterize the fecal microbiota of water buffalo calves during the pre-weaning phase and (ii) to evaluate the effect of different diets (water buffalo milk, formula milk, or a mixed diet) on the development of gut microbiota.

Materials and methods

Animal care and procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals and Directive 2010/63/EU for animal experiments (National law: D.L. 26/2014). The ethics statement was approved by the Bioethics Committee of the Istituto Zooprofilattico Sperimentale del Mezzogiorno, ex art. 26, c. 2, DL n° 26 /2014.

Study population and sample collection

In the present study, a total of 60 newborn water buffalo calves were included. All calves were recruited from the same commercial dairy farm (Sud Rienergy) to avoid potential confounding factors that could affect the results, such as differences in farming practices, environmental conditions and climatic factors. In addition, all animals included in the study were born naturally during the winter period (from November to March).

Within six hours after birth, calves were fed with maternal colostrum for 3 to 5 days. After colostrum phase, animals were weighed and divided into three groups, each consisting of 20 animals kept in individual boxes (1 × 2.30 m2) and homogeneous with respect to body weight, birth month, sex and colostrum feeding duration. Each of these three groups of calves were fed with a different dietary regimen, as described below:

• calves fed twice a day with formula milk;

• calves fed twice a day with water buffalo milk;

• calves fed twice a day, once with formula milk and once with water buffalo milk.

Animals included in the study were both males and females equally distributed among groups. During the trial, calves were fed with 2 L of water buffalo milk or formula milk at 18% of dry matter (DM) per meal, twice daily, according to the assigned experimental group (see Supplementary Table 1). Formula milk (FM) was prepared according to the manufacturer’s recipe, and its raw materials, chemical composition, vitamin, and trace element supplementation are shown below (Table 1).

Table 1.

Milk replacer formula

Raw materials used in the formulation of the milk replacer
Skimmed milk powder (52%), whey, palm oil, coconut oil, wheat gluten, calcium carbonate, sodium bicarbonate, magnesium sulphate,
Chemical composition of milk Replacer
Nutrients	Value (%)
Dry Matter	96
Crude Protein	23.96a
Fat	25.83a
Ash	7.45a
Lysine	1.6a
Methionine	0.6a
Cysteine	0.2a
Calcium	1.0a
Sodium	0.5a
Phosphorus	0.7a
Milk Forage Unit	1.45a
Supplementation
Vitamins (per kg)
Vitamin A (UI-3a672a)	2000
Vitamin D3 (UI-3a671)	100
Vitamin E (mg- 3a700)	50
Trace element (mg/kg)
Iron (ferrous sulphate monohydrate – 3b103) (Iron (II) chelate of glycine hydrate − 3b108)	25
Zinc (zinc sulphate monohydrate – 3b605) (Zinc chelate of glycine, hydrate – 3b607)	40
Manganese (manganese sulphate monohydrate − 3b503) (manganese glycine chelate − 3b506)	70
Iodine (potassium iodide – 3b201)	1
Selenium (an organic form of selenium produced by S. cerevisiae NCYC R397- 3b811)	0.2
ANTIOXIDANTS (MG/KG)
bht (e321)	31

^avalue calculated as percentage of Dry Matter (DM)

Drinking water was made available to the calves ad libitum from 10 days of age using a dedicated bucket. The buffalo milk used to feed the calves was taken from the tank after each morning and evening milking and given to the calves. No waste of milk from sick or medicated animals was used to feed the calves. The reference levels for bulk milk were obtained from samples taken during monthly routine checks. The average chemical composition (mean ± SD) for bulk milk, measured monthly during the trial period, was as follows: Fat % 8.47 ± 0.001, Protein % 4.56 ± 0.001, Lactose 4.62 ± 0.000, SCC 78,000 ± 10,658, average daily/yield/head kg 10.18 ± 0.117.

To evaluate the impact of different dietary regimens on calf growth and development, body weight was recorded at the end of colostrum feeding (T0) and again after 21 days, the end of the experimental period (T1). Growth rate was calculated by comparing these two measurements.

Freshly voided fecal samples were collected from all calves at two different time points: at the end of colostrum feeding (T0), corresponding to the final day of colostrum feeding (either day 3, 4 or 5 of life) and prior to the introduction of the experimental diet, and 21 days (T1) after the beginning of the administration of the specific diet. All the samples were collected in a sterile container, transported on dry ice and immediately stored at -80 °C until DNA extraction. No euthanasia or anesthesia procedures were performed on subjects in this study.

Epidemiological and anamnestic data were collected for the enrolled animals and their mothers, including information on possible confounding factors such as health problems and/or antibiotics use during pregnancy. In addition, information related to weight (T0 and T1) and possible occurrence of gastroenteric, and respiratory disease were collected from calves. As set out in the Classyfarm protocol, the indicators for respiratory problems are designed to detect mild or severe clinical forms, drawing on the ABMs defined by the Welfare Quality^® protocol [14]. When assessing gastrointestinal problems, animals exhibiting diarrhea and bloating are observed and recorded. Stool consistency and soiling of the perianal area are indicators of diarrhea, whilst a prominent rumen and a distended abdomen are indicators of bloating [14].

Body weight gain was analyzed using a linear mixed model with diet and time (T0, T1) as fixed effects, including their interaction. Live bodyweight, sex, and birth month were tested as covariates; only birth weight was retained in the final model based on AIC comparison. Pairwise comparisons were adjusted using Tukey’s HSD test. Average daily gain (ADG) was calculated as (weight at T1 – weight at T0) / 21 days. All analyses were performed in R (v. 4.2).

DNA extraction and data analysis

All fecal samples were extracted using the standard operating protocol IHMS_SOP 07 V2 Version: 2 [15] with some modifications related to the use of the QIASimphony automatic DNA extractor and the QIAamp DSP DNA Mini Kit (Qiagen, Hilden Germany).

The quantification of the total bacterial DNA extracted was carried out using a high-sensitivity QubitTM fluorometer and final concentration was adjusted to 5 ng/µl for each sample.

All libraries were prepared according to the 16 S Metagenomic Sequencing Library Preparation protocol (Illumina, San Diego, CA, USA) amplifying the V3-V4 regions of the bacterial 16s rRNA gene. PCR amplification was performed using the KAPA HiFi HotStart Ready Mix DNA Polymerase (Roche Diagnostics, Mannheim, Germany) using 1 µM of each primer (341 F: CCTACGGGNGGCWGCAG; 805R: GACTACHVGGGTATCTAATCC) in a final reaction volume of 25 µl containing 12.5 ng of DNA template. The thermal cycling conditions included an initial denaturation at 95 °C for 3 min, followed by 25 cycles of denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s and an extension step at 72 °C for 30 s, followed by a final extension step at 72 °C for 5 min. Amplicons were then purified using the Agencourt AMPure XP-PCR Purification beads (Beckman Coulter, Brea, CA, USA).

At the end of the amplification, 1 µl of the PCR product was run on a Bioanalyzer DNA 1000 chip to verify that the amplicon size was ~ 550 bp. Libraries were then normalized, pooled and sequenced on a MiSeq platform (Illumina, San Diego, CA) in a 2 × 300 bp paired-end format. Negative controls for DNA extraction and PCR amplification were included in each working session as quality controls, even though they did not produce significant amplicon amounts and therefore could not be sequenced.

After sequencing, the raw sequences were checked for quality control using DADA2 software to remove low-quality ones (phred score < 20) which might cause bias in downstream analyses [16]. Sequences were then imported into Qiime2 software (v. 2020.8) [17] for microbiota data analysis as follows. Based on the quality control check, forward reads were filtered and trimmed at 300 bp, while reverse reads were trimmed at 180 bp and resolved to high-resolution Amplicon Sequence Variants (ASVs), which represent, as closely as possible, the original biological sequence of the sequenced amplicons [18]. ASVs were clustered at 99% similarity, and clustered sequences were aligned against the Silva 138 database [19] using feature-classifier classify-sklearn plug-in for taxonomy classification [20]. The sequences were aligned using gMAFFT and used to construct a rooted phylogenetic tree with the FastTree software [21], which was used for statistical diversity metrics. Chloroplasts and unclassified sequences were excluded from the analysis, as their total number counts were negligible. This filtering step was performed in accordance with standard QIIME 2 [18] and phyloseq [22] protocols to eliminate potential technical artifacts or non-bacterial noise that could artificially bias downstream microbial diversity metrics. Features with a minimum frequency ≤ 150were excluded from the analysis to avoid possible bias due to eventual contamination.

Gut microbiota composition was assessed by evaluating alpha and beta diversities using phyloseq R package [22]. To address sampling depth heterogeneity, alpha and beta diversity metrics were calculated after standardizing the sample size to 4,800 sequences per sample. Observed features index (a qualitative measure of species richness) and Shannon’s diversity index (a quantitative measure of community richness) were used to summarize the microbial diversity within a sample. Bray-Curtis (a quantitative measure of community dissimilarity), Unweighted Unifrac (a qualitative measure of community dissimilarity) and Weighted Unifrac dissimilarity matrix (a qualitative measure of community dissimilarity incorporating phylogenetic relationships between the features) instead, were used to compare differences in microbial community structure between samples. Furthermore, differential abundance analysis was performed in the R environment using the DESeq2 package [23] based on raw, not rarefied OTU/ASV counts.Taxa with an adjusted p-value (FDR) < 0.05 were considered significantly differentially abundant between sample groups. A sensitivity analysis excluding samples identified as outliers was performed, exhibiting that the overall beta-diversity patterns and the differential abundance of key taxa remained unchanged. Finally, functional annotation was performed using Tax4Fun2 [24] using the default database based on KEGG Orthology (KO), a tool for predicting functional profiles from marker gene sequences, then Kruskal-Wallis test was applied to evaluate statistical significance among sample groups and Spearman’s rank correlation analysis was performed to assess the correlation between the phenotypic measurements.

Results

Body weight gain

The mean birth body weight of calves was 42.29 ± 5.38 kg standard deviation for female and 45.97 ± 4.30 kg for male calves. At the end of the trial (T1), the mean body weight was 62.69 ± 6.02 kg for females and 66.41 ± 6.55 kg for males. Across feeding groups, calves receiving water buffalo milk showed an average weight gain of 20.80 kg (ADG = 0.99 kg/d), those fed on formula milk gained 18.61 kg (ADG = 0.89 kg/d) and those fed the mixed diet gained 23.23 kg (ADG = 1.11 kg/d) (Supplementary Table 1). Pairwise comparison using Tukey’s HSD test showed that the mixed diet group had significantly higher weight gain than the formula milk group (P < 0.005). No significant differences were observed between mixed diet and water buffalo milk (P = 0.312) or between water buffalo milk and formula milk (P = 0.178). Throughout the experimental period, all calves remained clinically healthy. Throughout the entire experimental period, all calves remained clinically healthy, with no evidence of enteric or respiratory disorders. No therapeutic interventions were required, and no antibiotics, multivitamin complexes, or oral rehydration solutions were administered. The observed differences in growth highlighted the potential impact of dietary regimens and underscored the importance of carefully evaluating how they may influence changes in the gut microbiota and its functions.

High-throughput sequencing

A total of 5,905,827 cleaned reads were obtained from all samples, with an average of 46,189 reads per sample ranging from 10,786 to 89,282 sequences per sample. The analysis identified 827 ASVs in T0 calves and 998 ASVs in T1 calves, which were used for taxonomic assignment and statistical analysis.

Fecal microbiota composition in newborn water buffalo calves at T0

The taxonomic composition of the 60 samples collected at T0 was analysed based on the relative abundance of amplicon sequence variants (ASVs). The analysis revealed the presence of four main phyla, 39 taxa at the Family level (with 7 families characterized by a mean relative frequency ˃2%), and 95 taxa at the genus level (with 7 genera with a mean relative frequency > 2%), see Fig. 1.These data were calculated on the totality of calves before partition into diet-based groups, based on the evidence that there were no significant differences at T0 among groups both in alpha (Observed features p-value = 0.884, Shannon index p-value = 0.913) and beta diversity (PERMANOVA, Bray-Curtis, Unweighted and Weighted Unifrac dissimilarity matrices, P > 0.2).

Fig. 1.

Lollipop plots show the mean relative abundance (dot) and SD (shaded band) for each taxon, color-coded by taxonomic level (blue = phylum, green = family, orange = genus) of fecal microbiota composition in newborn water buffalo calves at T0 and in calves fed on water buffalo milk at T1. Only taxa with mean relative frequency ˃2% are shown, and within each taxonomic level, taxa are ordered by decreasing mean abundance

Fecal microbiota composition in calves fed on water buffalo milk at T1

The taxonomic composition of the 20 fecal samples collected at the end of the experimental period (T1) from water buffalo milk fed calves was assessed at the phylum, family and genus levels.

Phylogenetic analysis exhibited the presence of five main phyla, 44 taxa at the Family level (with 10 families exhibiting a mean relative frequency˃ 2%), and 99 taxa at the Genus level (with 12 genera characterized by a mean relative frequency > 2%), see Fig. 1.

Fecal microbiota composition in calves fed on formula milk T1

The fecal microbiota of calves fed formula milk at T1 showed an overall taxonomic structure comparable to that observed in water buffalo milk-fed calves, with Firmicutes (49.2%) and Bacteroidota (21.1%) as the dominant phyla. However, a markedly higher relative abundance of Proteobacteria (12.8%) and Actinobacteriota (10.6%) was observed compared to buffalo milk-fed calves, while Verrucomicrobiota were slightly increased (5.0%). At the genus level, relevant differences emerged. Formula-fed calves exhibited higher abundances of Bifidobacterium (6.9%), Akkermansia (5.0%) and Clostridia_UCG-014 (6.6%), whereas beneficial butyrate-producing genera such as Faecalibacterium (4.0%) and Subdoligranulum (2.2%) were less represented compared to buffalo milk-fed animals. Lactobacillus (7.0%) and Bacteroides (12.9%) remained among the dominant genera, although at lower levels than those observed in buffalo milk-fed calves. Additionally, opportunistic taxa such as Pseudomonas (2.4%) were detected at higher relative abundance in formula-fed calves.

Fecal microbiota composition in calves fed on mixed diet T1

Calves fed a mixed diet displayed a microbiota composition largely overlapping with that of buffalo milk-fed animals. Firmicutes (62.0%) were the most abundant phylum, followed by Bacteroidota (15.8%) and Proteobacteria (8.3%), with the latter showing similar levels to buffalo milk-fed calves. Notably, Actinobacteriota (9.2%) were more represented compared to buffalo milk-fed animals, while Verrucomicrobiota (1.8%) were reduced.

At the genus level, the mixed diet group was characterized by relatively high abundances of Lactobacillus (8.6%), Faecalibacterium (7.5%) and Subdoligranulum (4.3%), indicating a microbial profile enriched in potentially beneficial taxa. Compared to buffalo milk-fed calves, Bacteroides (9.7%) were reduced, whereas Bifidobacterium (6.1%) and Collinsella (2.1%) were increased. Moreover, Clostridia_UCG-014 (7.6%) and [Eubacterium]_coprostanoligenes_group (3.8%) were more abundant than in buffalo milk-fed calves, suggesting diet-associated modulation of fermentative bacterial populations.

Overall, the comparison across dietary groups at T1 reveals distinct microbial profiles, particularly with variations in key genera traditionally considered beneficial, suggesting that potential variations in the intestinal ecosystem likely contribute to growth performance (specifically weight gain) and clinical health status observed in calves.

Comparison of microbial communities across time and diets

A statistical test on paired samples was used to assess changes in microbiota stability and diversity over time (T0 vs. T1) within subjects. Statistical significance was evaluated using the Wilcoxon signed-rank test (W), with P ≤ 0.05 as a measure of significance. Alpha diversity increased significantly from T0 to T1 in all experimental groups, both in terms of species richness (Observed Features) and diversity/evenness (Shannon entropy). Specifically, significant increases in Observed Features were detected in the water buffalo milk, formula milk, and mixed milk groups (P = 0.011, P = 0.025, and P = 0.003, respectively). Similarly, Shannon entropy increased significantly p-values (P ≤ 0.001 for all diet-based groups). These results suggest that variations in the gut microbiota of calves were evident as early as the first weeks of life and may be attributed to the natural maturation of the microbiota, characterized by an increase in richness and diversity. Notably, calves fed with mixed milk showed the strongest temporal shift in alpha diversity (P < 0.0005), although significant changes were observed in all dietary groups (Fig. 2).

Fig. 2.

Alpha diversity analysis on paired samples displaying Observed feature and Shannon entropy between the two time points among three experimental groups

Beta diversity analysis, used to assess variation in microbial communities between groups, confirmed these data. Bray-Curtis dissimilarity showed significant differences in microbial communities across time points (P = 0.001, PERMANOVA) and between different diets (P = 0.042, PERMANOVA, water buffalo milk vs. formula milk).

Consistently, both unweighted and weighted UniFrac distance matrices demonstrated significant differences in microbial communities over time (P < 0.005, PERMANOVA) (Fig. 3).

Fig. 3.

Principal coordinate analysis (PCoA) plot based on Bray-Curtis, Unweighted and Weighted Unifrac dissimilarity matrices show a separation over the time (T0 green and T1 yellow) among three experimental groups

The most abundant genera, as reported above, in the gut microbiota of newborn calves at T0 time were Escherichia-Shigella and Bacteroides followed by Butyricicoccus, Lactobacillus, Clostridium and Ruminococcus. The relative abundance of these genera at T1 time was significantly affected, exhibiting a decrease in the mean relative frequencies for these genera in all groups except for Lactobacillus (p-value < 0.005) which instead increased regardless of the type of diet (Fig. 4).

Fig. 4.

Boxplots showing the relative abundance of different bacterial species across times in the three dietary regimes. The plots highlight variations in microbial abundance of Escherichia-Shigella, Bacteroides, Butyricicoccus, Clostridium, Ruminococcus and Lactobacillus

Additional analyses were performed to evaluate the potential effect of sex on gut microbiota structure. No significant association between sex and microbial diversity or community composition was detected (P > 0.05), suggesting that dietary treatment represented the primary source of variation in the present cohort.

Moreover, differences among the dietary regimens were examined at T1 time. Alpha diversity analysis, used to assess the richness and the evenness of taxonomic units among diet-based groups, did not reveal significant differences in fecal microbiota for both indexes Observed features (p-value = 0.949) and Shannon (p-value = 0.774), see Fig. 5.

Fig. 5.

Alpha diversity analysis (Observed features and Shannon indexes) of microbiota at T1 in feces collected from animals destined to the water buffalo milk diet (blue), formula milk diet (green) and mixed milk diet (red)

Beta diversity analysis instead, showed both qualitative and quantitative differences in microbial community structure between groups, as assessed by Bray-Curtis (Pr(> F) = 0.007) and Unweighted UniFrac dissimilarity (Pr(> F) = 0.022). In contrast, PERMANOVA analysis on the Weighted UniFrac dissimilarity matrix did not show statistical significance (Pr(> F) = 0.08), excluding phylogenetic relationships between the features (Fig. 6). This discrepancy suggests that the dietary treatments primarily affected the presence/absence or the balance of specific sub-dominant taxa (captured by Bray-Curtis and Unweighted UniFrac), rather than causing massive shifts in the dominant phylogenetic lineages (Weighted UniFrac). To explore the specific taxonomic drivers responsible for these community structure differences, a differential abundance analysis was subsequently performed. Pairwise comparisons among dietary groups revealed significant differences between the formula milk and water buffalo milk groups based on both Bray–Curtis (Pr(> F) = 0.001) and unweighted UniFrac distances (Pr(> F) = 0.005), whereas no significant differences were detected between the other group pairs.

Fig. 6.

Beta diversity analysis (Bray-Curtis, Unweighted and Weighted Unifrac dissimilarity matrices) of microbiota at T1 in feces collected from animals fed with water buffalo milk diet (blue), formula milk diet (green) and mixed milk diet (red)

Differential abundance analysis identified Bifidobacterium as significantly enriched in the formula-fed and mixed-diet groups compared with the buffalo milk group (log2FC 2.71 padj 0.007 and log2FC 2.98 padj 0.0007, respectively), corresponding to an approximately eight-fold increase of this genus in both groups. (Fig. 7). These findings suggest that the composition of diet plays a crucial role in shaping gut microbiota, particularly with potentially beneficial bacterial taxa such as Bifidobacterium.

Fig. 7.

The violin plot shows the relative abundance of Bifidobacterium at T1 in feces collected from animals destined to the water buffalo milk diet (blue), formula milk diet (green) and mixed milk diet (red)

Finally, to gain deeper insights into nutrient assimilation, KEGG pathway enrichment analysis was performed to investigate the functional differences in gut microbiota of water buffalo calves fed on different milk-based diets. Results indicated a significant increase of taxa involved in energy metabolism (padj 0.004), amino acids (padj 0.02), cofactors and vitamins (padj 0.03), lipids (padj 0.04), carbohydrates (padj 0.04) and xenobiotics biodegradation (padj 0.04) as showed in Fig. 8A. Calves fed on water buffalo milk exhibited enrichment in the pathways involved in steroid biosynthesis, retinol metabolism, and neuroactive compound synthesis. Calves fed on formula milk showed increased activity in pathways such as carbon fixation in prokaryotes and porphyrin and chlorophyll metabolism. Furthermore, the mixed diet group showed intermediate features with notable enrichment in nicotinate and nicotinamide metabolism (vitamin B3 biosynthesis) and drug metabolism pathways (xenobiotics by cytochrome P450) (Fig. 8B). Unfortunately, none of the enriched pathways show a significant correlation with the phenotypic data.

Fig. 8.

(A) Bar plot showing the top KEGG metabolic categories enriched in the calf microbiota across all diet groups. The x-axis indicates the significance of enrichment as –log₁₀(p-value). (B) Heatmap of specific KEGG pathways showing differential abundance among calves fed water buffalo milk, formula milk, or a mixed diet at T1. Colors represent z-scores, with red indicating higher relative pathway activity and blue indicating lower activity

Discussion

The present study aimed to evaluate the impact of different diets on gut microbiota composition and growth performance in water buffalo calves during the first three weeks of life. Over a three-week experimental period following colostrum administration, calves were divided into three dietary groups: water buffalo milk, formula milk, and mixed diet (combining natural and formula milk). Our findings provide new insights into how early-life feeding practices influence both microbial diversity and body weight gain, two critical factors for the development and health of young ruminants.

Fecal samples collected at T0, when all calves were exclusively fed on colostrum, provided a baseline for gut microbiota development. In fact, alpha and multivariate beta diversity analyses revealed low microbial richness, consistent with the low bacterial richness reported after the meconium stage [25] and no significant differences in community structure, suggesting that all animals started from the same biological baseline at this stage of life, with a gutmicrobiota still developing and prone to evolve over time. During calfhood, the fecal microbiota demonstrated a high degree of homogeneity among newborn calves from the same farm. Microbiota was dominated mainly by the phyla Proteobacteria (particularly as regards the family Enterobacteriaceae and the genus Escherichia-Shigella), Firmicutes and Bacteroidota (including the families Bacteroides, Lachnospiraceae, and Ruminococcaceae) with progressive changes in microbial composition over time. These phyla are commonly reported as dominant during early gut colonization in young ruminants and play important roles in gastrointestinal development and metabolic activity [26, 27].

Indeed, as previously reported, the intestinal microbiota colonization begins during fetal life, and microbial exposure during the prenatal and immediate postnatal period contributes to the early establishment of the neonatal gut ecosystem, which subsequently undergoes rapid changes during the first weeks of life [15, 28].

Later, as the gut environment becomes more anaerobic and diet changes during early life, the relative abundance of Proteobacteria typically decreases while Firmicutes and Bacteroidetes increase, reflecting the gradual establishment of a more stable and functionally mature microbial community [27]. Although our two timepoints did not allow to evaluate any transient variations that may have occurred during the trial, our data clearly revealed a permanent, unidirectional gut maturation trajectory. In fact, we observed microbial transitions permanently away from the initial post-colostral state [7], and a significant increase in anaerobic species related to this natural maturation process, that make a potential return to the initial state (T0) unlikely [6, 27]. These findings were consistent with alpha diversity analysis at T1 characterized by an increase in microbial richness and evenness across all groups, indicating the natural and rapid maturation of gut microbiota during the first weeks of life. This microbial maturation coincided with substantial weight gain across all experimental groups, reflecting a fundamental developmental process essential for calf health and growth.

Consistently, beta diversity analysis revealed significant differences in microbial community structure, particularly between the water buffalo milk–fed and formula-fed groups. The mixed diet group showed the highest beta diversity dispersion and temporal variation in microbial community structure. This suggests that the combined administration of both milk sources shapes a more transitional and multi-faceted gut ecosystem, which is further supported by the enrichment of specific functional pathways related to carbohydrate and lipid metabolism observed in this group.

In particular, the observed greater variability and potentially enhanced stability in the mixed diet group microbiota correlated directly with the significantly higher weight gain recorded in these calves compared to the formula-fed group. This strong association suggested that the optimized microbial environment promoted by the mixed diet directly contributed to improved growth performance, potentially through more efficient nutrient acquisition and overall gut homeostasis.

The results of the present study also demonstrated that calves fed on the mixed diet had a higher average weight gain (AWG) than those fed exclusively on formula or water buffalo milk. This finding is consistent with previous data [29, 30], which indicates that dairy calves fed on whole milk achieve better growth performance than those fed on milk replacer, suggesting that milk replacer may be less effective in supporting growth when provided in equal volumes. Therefore, the combination of different milk sources may provide a more balanced and complete nutrient profile, potentially contributing to improved growth performance and metabolic development. Overall, the data obtained in this study suggest that the tested diets did not negatively affect the composition of the intestinal microbiota in calves. It is well established that diet plays a central role in the development and modulation of gut microbiota in ruminants. However, additional factors including age, environmental exposure, and host-related variables, also contribute significantly to shaping microbial colonization during early life [31]. Among these, sex has also been reported as a potential determinant of gut microbiota composition in mammals. However, in the present study, exploratory analyses did not reveal any significant association between sex and microbiota diversity or composition. This finding may be related to the very young age of the calves, in which sex-related physiological differences are still limited compared with the strong influence exerted by early-life diet and microbial maturation [32].

In this scenario, the microbial profile observed in the present study also included several bacterial taxa commonly considered biomarkers of optimal intestinal health and known to support various physiological processes, such as nutrient metabolism and immune modulation. These microbial groups included families such as Lachnospiraceae, Butyricicoccaceae, and Ruminococcaceae, and genera such as Bacteroides, Collinsella, Eggerthella, and Bifidobacterium. Many of these microorganisms are involved in carbohydrate fermentation and in the production of short-chain fatty acids (SCFAs), particularly acetate and butyrate, which are important for intestinal homeostasis and host energy metabolism in calves [33, 34]. In parallel, a reduction in the relative abundance of the Ruminococcus_gnavus_group was observed between T0 and T1. Although members of this taxonomic group have been associated with intestinal inflammation and dysbiosis in other host species [35], the lack of statistical significance prevents any biological interpretation of this trend in the present study. Therefore, whether this observation reflects a component of normal microbiota maturation or has potential implications for intestinal health remains to be clarified by future investigations.

Notably, the genus Bifidobacterium exhibited a higher relative abundance in formula-fed and mixed-fed calves compared to those fed exclusively on water buffalo milk, consistent with previous work [36] showing that milk replacer composition can significantly influence the establishment of Bifidobacterium and other beneficial taxa in neonatal calves. As early colonizers, these bacterial taxa, together with others, contribute to immune modulation [37].

In fact, Bifidobacterium offers numerous health benefits, including protection against intestinal infections by producing lactic acid and other organic acids, which lower gut pH and create an environment hostile to pathogenic microorganisms. For example, this genus has been shown to be more abundant in healthy calves compared to those with diarrhea and is positively associated with beneficial metabolic and immune markers, suggesting a protective role against enteric infections [38] such as those caused by enterotoxigenic Escherichia coli (ETEC). Moreover, Bifidobacterium influences cytokine production, supporting regulatory T-cell activity, and enhancing mucosal barrier integrity. Several species are also capable of synthesizing B-group vitamins, including B2, B6, B9 (folate), and B12, which act as essential cofactors for lymphocyte function, cytokine production, and other immune-related metabolic processes [39].

Moreover, the maturation of the gut microbiota was marked by increased levels of Lactobacillus across all diet groups. Lactobacillus is a probiotic genus that contributes to the maintenance of intestinal health enhancing the epithelial barrier, preserving mucosal integrity, preventing the colonization of pathogenic bacteria and mitigating mucosal damage [40]. In line with this, Varada et al. (2022) demonstrated that early-life supplementation with probiotic significantly improves gut health indices and enhances growth performance in Murrah buffalo calves [41]. Early colonization by Lactobacillus in newborn calves has been associated with a lower incidence of enteric disorders and improved gastrointestinal function [41].

Finally, the analysis of functional pathways in the gut microbiota of water buffalo calves fed on different milk-based diets indicates a microflora actively involved in host nutrient processing, vitamin synthesis, and detoxification functions. Calves fed on water buffalo milk exhibited a significant enrichment in several metabolic pathways, including steroid biosynthesis, which may affect the regulation of stress, growth and sexual development, retinol metabolism, which is implicated in immune development and neuroactive compound synthesis (e.g. dopamine and adrenaline). In contrast, calves fed on formula milk showed increased activity in pathways such as carbon fixation in prokaryotes and porphyrin and chlorophyll metabolism, typically associated with autotrophic microbial functions, therefore suggesting an immature microbial ecosystem less efficient in nutritional and immune support. Finally, the mixed diet group showed intermediate features with notable enrichment in nicotinate and nicotinamide metabolism (vitamin B3 biosynthesis) and drug metabolism pathways (xenobiotics by cytochrome P450), suggesting an enhanced capacity for cellular energy metabolism and compound detoxification. These data suggest a possible association between growth performance and specific taxonomic changes. Indeed, the significantly higher body weight gain observed in the mixed diet group was accompanied by a marked increase in taxa traditionally considered beneficial, including Lactobacillus and Bifidobacterium [42]. This microbial shift was further supported by predictive metabolic profiling, which showed an enrichment of pathways involved in carbohydrate and lipid metabolism in both formula and mixed diet groups. These functional changes may have improved energy utilization, contributing to the greater weight gain observed. Therefore, despite the absence of strong statistical correlations at the individual level, the dietary treatments appear to have shaped the gut microbiota toward a more metabolically efficient profile, providing a plausible biological explanation for the superior growth performance of calves fed the mixed diet. A limitation of this study is that individual feed intake and feed efficiency could not be calculated, as calves were fed a fixed volume (2 L/meal, twice daily) and consumed all milk offered. However, since the dry matter content was identical across diets (18% DM), the observed growth differences reflect diet composition rather than intake volume. Another limitation of this study is the limited sampling timeframe, that does not cover all the pre-weaning period. The choice of the study period was due to a compromise between the study design and complex herd management practices related to separation of animals according to different dietary regimens. However, a period of less than a month can still alow to observe significant variations in gut microbiota related to diet [43].

Conclusions

In conclusion, the study highlights that early dietary intervention plays a crucial role for both growth performance and gut microbiota development in newborn calves. Mixed diet emerges as the most beneficial strategy, as it provides a well-balanced nutritional profile, supports optimal weight gain, and fosters the establishment of a stable and appropriate microbial community. However, further research is needed to explore the long-term effects of various dietary regimens during the early weeks of life and the pre-weaning period on animal health, immunity, and productivity. Understanding the intricate interactions between diet, gut microbiota and host physiology could suggest the way for innovative strategies to optimize farming practices in the dairy calf industry. Such insights may contribute to improved growth performance, enhanced disease resistance, and overall sustainability in livestock management.

Abbreviations

GIT

Gastrointestinal Tract

DM

Dry Matter

FM

Formula Milk

DNA

Deoxyribonucleic Acid

PCR

Polymerase Chain Reaction

rRNA

Ribosomal RNA

ASVs

Amplicon Sequence Variants

Authors’ contributions

RP Study design, DNA extractions, Genome Sequencing, Analysis and Draft Preparation. AR Study design and Bioinformatic Analysis. MGR Study design and Draft preparation. DV Study design and Manuscript Review. EF Statistical analysis and Manuscript Review. GB Study design, Supervision and Manuscript Review. GG, GI, OP, EDC revised the main manuscript. All authors have read and approved the final manuscript.

Funding

This study was supported by funding of Health Ministry IZS ME 04/16 RC.

Data availability

The data that support the findings of this study have been deposited in the Sequenced Read Archive (SRA) under accession number PRJNA1213496.

Declarations

Ethics approval and consent to participate

Animal care and procedures were conducted in a commercial dairy buffalo farm (Sud Rienergy livestock farm), previous informed consent from the owner to use the animals in the study, in accordance with the Guide for the Care and Use of Laboratory Animals and Directive 2010/63/EU for animal experiments (National law: D.L. 26/2014). The ethics statement was approved by the Bioethics Committee of the Istituto Zooprofilattico Sperimentale del Mezzogiorno, ex art. 26, c. 2, DL n° 26 /2014. All the procedures used to measure and to collect samples in the study on the animals, are routinely applied by the vet farm during the normal process of calf weaning.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Conflict of interest

The authors declare that the research was conducted in the absence of any potential conflict of interest.