Scientific Papers

Investigation of milk microbiota of healthy and mastitic Sahiwal cattle | BMC Microbiology


Taxonomic profile at phylum level

The milk microbiome profile of Sahiwal cattle analysis revealed a total of 22 phyla, 49 classes, 132 orders, 207 families, 309 genera and 168 species with different levels of distribution among the groups. All the 15 samples comprised of 421,400 clean reads having a range of 11,022- 43,988 reads (median reads = 25,602 per sample) distributed among all groups. The OTU found in healthy group were 1,11,811, in clinical mastitis group 1,80,126 and 1,29, 463 in sub clinical mastitis group.

The Phylum level taxonomy assignment of the three groups revealed 22 phyla (Fig. 1). Only 4 phyla were shared among all the groups including Proteobacteria, Firmicutes, Actinobacteria and Cyanobacteria. The numbers of distinct phyla in subclinical mastitis group were eight, in the healthy group two and clinical mastitis group only one phylum was distinct (Fig. 2). Although the distribution of microbiota amongst different phyla varied but six of them comprised more than 90% of the total phyla with different abundance. The percentage abundance of Proteobacteria was highest in the healthy group (56%) followed by a percentage abundance of 48.06% in the subclinical mastitis group and significantly lower (p < 0.05) abundance was detected in clinical mastitis group (2.68%) (Table 1). Firmicutes dominance was observed in clinical mastitis group (62.24%) and subclinical mastitis group (39.98%), it was detected in lower abundance in healthy group (15.87%). Acidobacteriota abundance was observed to be highest in healthy animals (11.32%) followed by (2.98%) in subclinical group while it was not detected in clinical mastitis group. The percentage abundance of Bacteroidota was significantly higher (p < 0.05) in healthy group (6.61%), it was 1.90% in subclinical mastitis group while in clinical mastitis group it was not detected. The percentage abundance of Fusobacteriota was significantly higher (p < 0.05) in clinical mastitis group (31.79%) as compared to subclinical mastitis group (0.02%) while it was not detected in healthy group. The percentage abundance of Actinobacteriota was 8.94%, 3.86% and 0.61% in healthy, sub clinical mastitis and clinical mastitis groups respectively (Fig. 3) with no significant difference (p < 0.05).

Fig. 1
figure 1

Heat-map plot of the relative abundance of different phyla in Sahiwal cattle Milk Microbiota. Healthy (n = 5), Clinical Mastitis (n = 5) and Subclinical Mastitis (n = 5). Dendrograms show the clustering of different phyla in the same group

Fig. 2
figure 2

Venn diagram showing distinct and shared phyla among different groups in milk samples of Sahiwal cattle. Healthy (n = 5), Clinical Mastitis (n = 5) and Subclinical Mastitis (n = 5)

Table 1 Percent abundance (mean) comparison at phylum taxonomic level in milk samples from Sahiwal cattle grouped as healthy, clinical mastitis and with subclinical mastitis only signifcantly different taxa are denoted with letters
Fig. 3
figure 3

Donut graph of Percentage abundance base representation of taxonomic distribution at phylum level of Sahiwal cattle milk. Different colors represent different phyla amongst the three groups

Taxonomic profile at order level

The orders shared between all the groups were 14 with distinct orders being highest in subclinical mastitis group (Fig. 4). Amongst the orders Rhizobiales, Parvibaculales and Staphylococcales were the abundant orders identified in the healthy group. Fusobacteriales and Staphylococcales were most abundant in clinical mastitis group while the subclinical mastitis group was dominated by Staphylococcales and Pseudomonadales (Table 2 and Supplementary figure S2). Rhizobiales being the highest in terms of percentage abundance in the healthy group (19.82%) whereas a low percentage abundance was recorded in healthy group (0.17%) and 4.86% in subclinical mastitis group. The percentage abundance of Caulobacteriales showed a non significant difference (p < 0.05) between the groups, with most highest in the healthy group (6.65%) followed by clinical mastitis group (1.60%) and the least abundance in subclinical mastitis group (1.27%). Order Burkholderiales was dominant in subclinical mastitis group (6.64%) while clinical mastitis group was the lowest in terms of abundance (0.38%), Healthy group showed abundance of 2.11% of Burkholderiales. Although percentage abundance of order Staphylococcales was non significantly different amongst all the three groups. It was observed to be present in highest abundance in subclinical mastitis group (34.05%) followed by clinical mastitis group (20.38%) and healthy group (12.92%). Corynebacteriales showed a non significant difference in percentage abundance in all the groups being the most highest in healthy group (6.06%), in clinical mastitis group having a low abundance of (0.55%). Compared to the clinical mastitis group the percentage abundance of Corynebacteriales was high in subclinical mastitis group (1.41%) but it was low as compared to the healthy group. The abundance of order Enterobacteriales followed the same trend as found in Corynebacteriales but the values were 3.79% for healthy, 0.07% in clinical mastitis group and 3.59% in subclinical mastitis group. Fusobacteriales was significantly higher (p < 0.05) in clinical mastitis group (31.79%), lower in subclinical mastitis group (0.02%), while it was not detected in healthy group. Ignavibacteriales was only detected in healthy group while in other two groups it was not detected. Parvibaculales was not detected in clinical mastitis group while in healthy group (15.19%) it was the third most abundant order and the percentage abundance in subclinical mastitis group recorded was 8.30%. Subclinical mastitis group contained Pseudomonadales as 2nd most abundant (19.65%) whereas healthy group was found to have a percentage abundance of 6.11% and in clinical mastitis group (0.02%) (Fig. 5).

Fig. 4
figure 4

Venn diagram showing distinct and shared phyla among different groups in milk samples of Sahiwal cattle. healthy (n = 5), clinical mastitis (n = 5) and subclinical mastitis (n = 5)

Table 2 Percent abundance (mean) comparison of mean at Order taxonomic level in milk samples from Sahiwal cattle grouped as healthy, clinical mastitis and with subclinical mastitis. Only signifcantly different taxa are denoted with letters
Fig. 5
figure 5

Taxa bar plot representing taxonomic distribution at Order level of Sahiwal cattle milk. Different colors represent different orders amongst the three groups

Taxonomic profile at genus level

The genus level assignment of milk microbiome revealed a total of 309 genera with different abundance levels in all three groups. The abundant genera detected in healthy group included Parvibaculum, Staphylococcus, Ignavibacterium and Pseudomonas. Clinical mastitis group was dominated by Streptococcus, Caviibacter and Staphylococcus. Subclinical mastitis group included Staphylococcus, Pseudomonas and Parvibaculum as the top three abundant genera. Genus Novosphingobium was detected in all three groups with a higher percentage abundance detected in healthy group (0.22%) followed by clinical mastitis group (0.13%) and being the least abundant in subclinical mastitis group (0.06%). Caviibacter genus showed the significantly highest abundance ((p < 0.05) in clinical mastitis group (31.80%) as compared to other genera and it was also the highest as comparedto other groups because it was not detected in other groups. The pattern of percentage abundance of Corynebacterium genus showed a trend of being highest in subclinical mastitis group (0.79%) followed by healthy group (0.73%) whereas being the lowest in clinical mastitis group (0.55%). Escherichia-Shigella genus was observed as the least abundant in clinical mastitis group (0.49%) while subclinical group (0.07%) showed a high abundance as compared to clinical mastitis group while healthy group (1.16%) appeared as the most abundant group in terms of Escherichia-ShigellaIgnavibacterium genus was only detected in healthy group (5.7%) and was significantly higher as compared to other groupsThe percentage abundance of Lactobacillus was in decreasing order from clinical mastitis group, subclinical mastitis group and healthy group. Parvibaculum genus was identified as the most abundant genus in healthy group, in subclinical mastitis group it occurred as the 3rd abundant genus detected whereas it was found absent in clinical mastitis group. Pseudomonas genus appeared the 2nd most abundant genera in subclinical mastitis group (18.56%) while the percentage abundance of Pseudomonas in healthy and clinical mastitis group was 5.35% and 0.01% respectively. Staphylococcus ranked 1st, 2nd and 3rd amongst different genera in healthy (13.43%), Clinical mastitis (20.39%) and subclinical mastitis group (34.36%) respectively. Streptococcus was only detected in clinical mastitis group (42.68%) and was significantly higher (p < 0.05) as compared to other groups and ranked as the top genera of the group (Table 3 and Supplemntary figure S4).

Table 3 Percent abundance (mean) comparison of mean at Genus taxonomic level in milk samples from Sahiwal cattle grouped as healthy, clinical mastitis and with subclinical mastitis only signifcantly different taxa are denoted with letters

Taxonomic profile at specie level

A total of 168 bacterial species were found in all three groups after taxonomy assignment at the specie level. Uncultivated species of many phyla were given a cornucopia of classification. In some other cases, no taxonomy was assigned, and it was regarded as unknown. In our study, species were not resolved to the same degree as other taxonomic levels. While species percentage abundance varied between the groups, many species were not classified according to any particular taxonomic rank (Supplemntary figure S5). While some assigned species were assigned at higher levels of taxonomy, they lacked specie classification i.e. phylum, class, order and genus. We found that different groups had distinct representations of species from various phyla. The diversity among various groups was greater at the species level, such as Staphylococcus was identified at genus level but no specie of Staphylococcus was assigned. Ignavibacterium album (15.72%) was predominant in the healthy group followed by Novosphingobium capsulatum (10.42%), Akkermansia muciniphila (3.57%), uncultured Acidobacteriales (3.06%), uncultured Rubrobacteraceae (2.43%), uncultured Actinobacterium (0.64%), uncultured Phyllobacteriaceae (0.49%), Pseudomonas sp (0.16%), Lactobacillus fermentum (0.12%) and uncultured Rhizobiales (0.11%). Corynebacterium bovis and Streptococcus dysgalactiae were not detected in healthy group. The clinical mastitis group was dominated by Streptococcus dysgalactiae (49.15%) and Corynebacterium bovis (1.33%) along with Novosphingobium capsulatum (0.73%) while other species found in healthy and subclinical mastitis group were not detected in the group. Subclinical mastitis group was dominated by species that were not detected in clinical mastitis group while the dominant species that were detected in clinical mastitis group was not detected in clinical mastitis group. In subclinical mastitis group Lactobacillus fermentum (5.71%), uncultured Acidobacteriales (4.95%), Akkermansia muciniphila (1.63%), Pseudomonas sp (1.04%) and uncultured Actinobacterium (0.32%) were identified. Amongst the healthy group Ignavibacterium album was significantly higher (p < 0.05) as compared to the clinical mastitis group and subclinical mastitis group. In the clinical mastitis group the dominant specie i.e. Streptococcus dysgalactiae was significantly higher (p < 0.05) in the group as compared to the healthy group and subclinical mastitis group. The other species that was identified to be significantly higher (p < 0.05) in healthy group as compared to the clinical mastitis group was uncultured Rhizobiales.

The heat maps plots and group wise percentage tables abundance at Class and Family level is provided as Supplementary figure S 1 and S 4 and Supplementary Tables 10 and 11 respectively. While percentage abundance of individual samples at all levels are provided as Supplementary Table. 4– 9.

Alpha diversity analysis of Sahiwal cattle milk microbiota

Assessment of the three study groups using different alpha diversity metrics depicts a clear picture in terms of the microbial diversity in these groups. Shannon and Simpson diversity indices which deduce additional information about the community composition in addition to just species richness or evenness. This study shows that the microbial diversity was higher in healthy group as compared to clinical mastitis and subclinical mastitis group while lowest diversity was observed in clinical mastitis group. Chao 1 and observed features are diversity indices which are abundance base coverage estimators of specie richness also shows that alpha diversity was highest in healthy group as compared to other groups while being the lowest in clinical mastitis group (Fig. 6).Simpson and Shannon microbial diversity indices were significantly higher in healthy group (Table 4).

Fig. 6
figure 6

Alpha diversities of the milk microbiome collected from Sahiwal cattle (healthy = green, clinical Mastitis = blue and subclinical Mastitis group = red) calculated by a) Shannon index b) Simpson c) Chao 1 d) observed species

Table 4 Alpha diversity indices in healthy, clinical and subclinical mastitis in Sahiwal milk samples

Beta diversity analysis of sahiwal milk microbiota

The beta diversity indices (Jaccard Index and Bray Curtis dissimilarity index) shows a same pattern where in 80% (4 out of 5) healthy samples were clustered together separately from other samples of the two groups. Similarly, 60% (3 out of 5) of subclinical samples were clustered separately. while 60% (3 out of 5) samples of healthy group were clustered with 20% (1 out of 5) of subclinical samples seperatly. 40% (2 out of 5) of healthy, 20% (1 out of 5) clinical and 20% (1 out of 5) subclinical samples were clustered together separately from other samples. Weighted and unweighted unifrac Principal Component analysis plot shows that all clinical mastitis samples are clustered separately from the other two groups while there is clustering of 60% (3 out of 5) healthy samples in unweighted unifrac PCA plot and 40% are clustered together in weighted unifrac PCA plot. While remaining healthy and subclinical mastitis samples have been found to have mixed diversity patterns (Fig. 7).

Fig. 7
figure 7

Beta diversity indices for Sahiwal milk of different groups. Principal component analysis plots based on a) Jaccard index b) Bray Curtis dissimilarity index. c)  Unweighted d) Weighted unifrac metrics



Source link