Growth performance and diarrhea incidence (Exp. 1)
As shown in Table 2, the initial BW, final BW, ADFI, and F: G of calves were similar among treatments (P > 0.05). However, the ADG of calves in 3% H + G and 5% H + G was higher (P < 0.05) than NC group. The decreased fecal scores and diarrhea incidence were significantly associated with the HNa and Gln supplementation. The diarrhea incidence of calves in the NC, 1% H + G, 3% H + G, and 5% H + G group was 25.81%, 17.08%, 21.42%, and 15.41%.
Table 2
Effects of HNa and Gln combined supplementation on the growth performance and diarrhea incidence of calves1
Item
|
Experimental Treatments
|
P-value
|
NC
|
1% H + G
|
3% H + G
|
5% H + G
|
BW, kg
|
|
|
|
|
|
Initial (51 d)
|
67.70 ± 2.01
|
66.43 ± 1.30
|
65.13 ± 1.82
|
68.05 ± 1.01
|
0.430
|
Final (72 d)
|
88.00 ± 2.29
|
90.43 ± 1.15
|
90.82 ± 1.81
|
93.90 ± 1.69
|
0.092
|
ADG, kg
|
0.96 ± 0.04b
|
1.14 ± 0.04ab
|
1.22 ± 0.06a
|
1.23 ± 0.08a
|
0.025
|
ADFI, kg
|
2.29 ± 0.04
|
2.49 ± 0.05
|
2.61 ± 0.06
|
2.58 ± 0.15
|
0.062
|
F:G
|
2.43 ± 0.14
|
2.26 ± 0.17
|
2.20 ± 0.15
|
2.19 ± 0.18
|
0.722
|
Fecal score
|
2.07 ± 0.12b
|
1.50 ± 0.14a
|
1.58 ± 0.11a
|
1.39 ± 0.07a
|
0.001
|
Diarrhea incidence, %
|
25.81
|
17.08
|
21.42
|
15.41
|
|
a−bMeans within a row with different letters differ significantly (P < 0.05). |
1Data are the mean of 10 replicates of 1 calf per treatment. |
Data were shown as means ± SEM (n = 10). |
Growth performance and diarrhea incidence (Exp. 2)
As shown in Table 3, compared with NC group, the calves in H + G group had greater final BW (P = 0.026) and ADG (P = 0.003). No significant differences in F: G and ADFI were observed among treatments. In addition, the calves in H + G group had lower fecal scores (P = 0.001) and diarrhea incidence than calves in NC group.
Table 3
Effects of HNa and Gln combined supplementation on the growth performance and diarrhea incidence of calves1
Item
|
Experimental Treatments
|
P-Value
|
NC
|
H + G
|
BW, kg
|
|
|
|
Initial (51 d)
|
68.50 ± 1.96
|
70.25 ± 2.07
|
0.548
|
Final (72 d)
|
91.17 ± 2.71b
|
101.38 ± 3.00a
|
0.026
|
ADG, kg
|
1.08 ± 0.06b
|
1.48 ± 0.06a
|
0.003
|
ADFI, kg
|
1.98 ± 0.15
|
2.37 ± 0.23
|
0.182
|
F:G
|
2.08 ± 0.27
|
1.63 ± 0.19
|
0.230
|
Fecal score
|
2.04 ± 0.15a
|
1.36 ± 0.17b
|
0.001
|
Diarrhea incidence, %
|
27.16
|
15.69
|
|
a−bMeans within a row with different letters differ significantly (P < 0.05). |
1Data are the mean of 10 replicates of 1 calf per treatment. |
Data were shown as means ± SEM (n = 10). |
Serum parameters (Exp. 2)
The concentration of serum DAO and D-lac are shown in Fig. 1A and 1B. Compared with NC group, supplemented with HNa and Gln significantly decreased the concentration of serum DAO and D-lac (P < 0.05). As shown in Table 4, compared with NC group, supplemented with HNa and Gln increased the IgG (P = 0.018) concentration, as well as activities of GSH-Px (P = 0.003) and T-AOC (P = 0.005) in the serum of calves. Furthermore, lower concentrations of serum TNF-α (P = 0.047) and MDA (P = 0.002) were observed in the H + G group compared with the NC group. There was no significant difference among treatments in serum IgA, IgM, IL-6, and T-SOD concentration (P > 0.05).
Table 4
Effects of HNa and Gln combined supplementation on the concentrations of cytokines and antioxidant capacity of calves in the serum1
Item
|
Experimental Treatments
|
P-Value
|
NC
|
H + G
|
IgA, µg/mL
|
89.51 ± 3.62
|
78.32 ± 4.61
|
0.073
|
IgG, µg/mL
|
1245.86 ± 41.61a
|
1083.57 ± 46.06b
|
0.018
|
IgM, µg/mL
|
74.24 ± 7.50
|
69.33 ± 6.79
|
0.295
|
IL-6, ng/L
|
8.64 ± 0.51
|
8.17 ± 0.62
|
0.569
|
TNF-α, ng/L
|
211.22 ± 8.34a
|
186.46 ± 8.10b
|
0.047
|
GSH-Px, U/L
|
128.65 ± 7.00b
|
167.48 ± 9.15a
|
0.003
|
T-SOD, pg/mL
|
59.64 ± 7.76
|
66.52 ± 4.70
|
0.458
|
T-AOC, U/mL
|
6.58 ± 0.27b
|
7.94 ± 0.33a
|
0.005
|
MDA, mmol/mL
|
3.08 ± 0.43a
|
2.29 ± 0.51b
|
0.002
|
a−bMeans within a row with different letters differ significantly (P < 0.05). |
1Data are the mean of 10 replicates of 1 calf per treatment. |
Data were shown as means ± SEM (n = 10). |
Analysis of intestinal microbiota in weaned calves (Exp. 2)
In the microbiome study, 5,116,836 effective tags were acquired after filtering the data quality, with an average number of 255,842 tags per sample. Based on the 97% identity level, these sequences were decomposed into 1,912 operational taxonomic units (OTUs), while 1,035 and 884 specific OTUs were observed in H + G and NC groups, respectively (Fig. 2B). The Chao1, Ace, Shannon, and Simpson indexes associated with bacterial richness and diversity were similar among groups (Fig. 2A). The principal coordinate analysis (PCoA) plots showed an overlap of partial samples between NC and H + G groups.
The relative abundances of different phyla were shown in Fig. 3. The microbial community was dominated by Firmicutes (76.28–30.67%), Bacteroidetes (45.46–15.45%), Proteobacteria (13.88 − 0.68%), Spirochaetes (11.91 − 0.08%), Fusobacteria (4.22 − 0.02%), and Actinobacteria (1.96 − 0.01%), which were more than 97% (Fig. 3A). Compared with NC group, HNa and Gln supplementation significantly increased (P < 0.05) the ratio of Firmicutes to Bacteroidetes (Fig. 3B) and the relative abundance (P < 0.001) of Firmicutes (Fig. 3C), but decreased (P < 0.05) the relative abundance of Bacteroidetes (Fig. 3D).
At the genus level (Fig. 4A), Ruminococcaceae.UCG-005 (40.38–3.03%), Others (55.06–23.37%), Succinivibrio (26.07–0.64%), Bacteroides (29.19–1.12%), Ruminococcaceae.UCG-010 (20.29–0.44%), Treponema.2 (17.99 − 0.03%), Rikenellaceae.RC9.gut.group (9.41 − 0.05%), Ruminococcaceae.UCG-014 (10.62 − 0.48%), Lachnospiraceae.AC2044.group (19.32–0.35%), Eubacterium.coprostanoligenes.group (6.45 − 0.07%), and uncultured bacterium (26.69 − 0.18%) were the most predominant genera in all the samples. Compared with NC group (Fig. 4B), H + G group had higher relative abundance of Bifidobacterium (P < 0.001), Lactobacillus (P < 0.001), Olsenella (P < 0.05), Ruminiclostridium 9 (P < 0.01), Howardella (P < 0.05), and uncultured organism (P < 0.05), but lower relative abundance of Helicobacter (P < 0.05) and Lachnoclostridium (P < 0.05).
Analysis of metabolic profiling in weaned calves (Exp. 2)
In the present study, the untargeted metabolomics analysis was generated based on fecal samples by ultra-high performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-QTOF/MS). There were overall 289 and 199 metabolites identified in the feces of weaned calves under the positive and negative mode, respectively. Fecal metabolic profiling was analyzed by PLS-DA, which showed clear segregation in the positive ion mode (R2 = 0.912, Q2 =–0.326) (Fig. 5A) and negative ion mode (R2 = 0.893, Q2 =–0.248) (Fig. 5B) between the NC and H + G groups, and suggesting the high reliability and predictive power. As a supervised method for pattern recognition, OPLS-DA analysis was performed on the data between the NC and H + G groups. As shown in Fig. 5C-D, the positive ion mode (R2 = 0.968, Q2 =–0.116) and negative ion mode (R2 = 0.873, Q2 =–0.222) for NC and H + G groups were distinctly separated in the OPLA-DS scatter plots, and illustrating the significant differences in the fecal metabolism between NC and H + G groups..
Metabolites with VIP values > 1.0 and P-value < 0.05 were considered significantly change. As shown in Table 5, a total of 18 (10 positive ion mode and 8 negative ion mode) significantly changed metabolites in fecal samples of weaned calves were detected among NC and H + G groups according tom ultivariate statistical analysis. Additionally, there were 16 (1-Palmitoylglycerol, Terbutaline, 3-Aminobutanoic acid, Leu-Ala, Thr-Arg, Nitrosobenzene, 1-Palmitoyl-sn-glycero-3-phosphocholine, Tyr-Met, N6-Methyladenine, Oleic acid, Sebacic acid, 13(S)-HODE, D-Mannose, Nname, cis-9,10-Epoxystearic acid, cis-9-Palmitoleic acid, and L-Malic acid) significantly upregulated metabolites and 2 (2-Methylbenzoic acid and N-Acetyl-D-lactosamine) significantly downregulated metabolites in the H + G group compared to NC group.
Table 5
Identified differentially expressed metabolites between H + G and NC group calves1. (Exp. 2)
Adduct
|
Metabolite
|
VIP
|
FC
|
p-value
|
m/z
|
Rt (s)
|
Trend
|
(M + H-H2O)+
|
1-Palmitoylglycerol
|
2.50
|
2.43
|
0.007
|
313.27
|
72.26
|
↑
|
(M + CH3CN + Na)+
|
Terbutaline
|
1.07
|
1.67
|
0.007
|
289.15
|
195.19
|
↑
|
(M + H)+
|
3-Aminobutanoic acid
|
2.38
|
3.07
|
0.01
|
104.07
|
45.25
|
↑
|
(M + CH3CN + H)+
|
Leu-Ala
|
1.01
|
3.12
|
0.012
|
244.16
|
254.23
|
↑
|
(M + H)+
|
Thr-Arg
|
2.01
|
2.01
|
0.019
|
276.16
|
398.92
|
↑
|
(M + NH4)+
|
Nitrosobenzene
|
1.83
|
1.3
|
0.02
|
125.07
|
67.04
|
↑
|
(M + H)+
|
1-Palmitoyl-sn-glycero-3-phosphocholine
|
4.94
|
1.67
|
0.037
|
496.34
|
190.05
|
↑
|
(M + H-H2O)+
|
Tyr-Met
|
1.2
|
1.41
|
0.044
|
295.11
|
129.32
|
↑
|
(M + H-H2O)+
|
N-Acetyl-D-lactosamine
|
1.16
|
0.54
|
0.045
|
366.14
|
364.75
|
↓
|
(M + H)+
|
N6-Methyladenine
|
1.28
|
1.55
|
0.05
|
150.08
|
126.98
|
↑
|
(M-H)-
|
Oleic acid
|
31.47
|
1.56
|
0.001
|
281.25
|
37.72
|
↑
|
(M-H)-
|
2-Methylbenzoic acid
|
1.04
|
0.74
|
0.005
|
135.045
|
155.13
|
↓
|
(M-H)-
|
Sebacic acid
|
1.23
|
2.03
|
0.008
|
201.11
|
321.5
|
↑
|
(M-H)-
|
13(S)-HODE
|
1.92
|
1.39
|
0.016
|
295.23
|
37.64
|
↑
|
(M + CH3COO)-
|
D-Mannose
|
3.22
|
1.58
|
0.018
|
239.08
|
369.75
|
↑
|
(M-H)-
|
Nname,cis-9,10-Epoxystearic acid
|
7.68
|
2.07
|
0.018
|
297.24
|
37.53
|
↑
|
(M-H)-
|
cis-9-Palmitoleic acid
|
5.98
|
1.43
|
0.032
|
253.22
|
37.69
|
↑
|
(M-H)-
|
L-Malic acid
|
1.55
|
1.62
|
0.045
|
133.01
|
405.99
|
↑
|
Difference metabolites identified by positive and negative ion mode. (multi-dimensional statistical analysis of VIP > 1 and univariate statistical analysis of P value < 0.05). VIP = variable importance in the projection, FC = Fold change, m/z = mass-to-charge ratio, Rt(s) = retention time,↑= the compound is up-regulated,↓= the compound is down-regulated. |
To organize and cluster the significantly different metabolites, two-way hierarchical cluster analysis was performed for comparisons between NC and H + G groups under positive ion mode (Fig. 6A) and negative ion mode (Fig. 6B), which indicated that the metabolites were highly differentiated among the groups.
To reveal the underlying mechanism, these changed metabolites were further performed by KEGG enrichment analysis. The data of pathway analysis reflected that there were 6 significant enriched pathways of metabolites in weaned calves supplemented with HNa and Gln, which included: 1) Fatty acid biosynthesis, 2) Proximal tubule bicarbonate reclamation, 3) C-type lectin receptor signaling pathway, 4) PPAR signaling pathway, 5) Lysosome, 6) Renal cell carcinoma (Fig. 6C).
Correlation analysis (Exp. 2)
To further investigate the correlation of the altered intestinal microbiota and altered metabolites in weaned calves supplemented with HNa and Gln, we performed a Spearman’s correlation analysis. In detail, as shown in Fig. 7A, Bifidobacterium was positively correlated (r > 0.52, P < 0.05) with Leu-Ala, Sebacic acid, and 1-Palmitoylglycerol. Lactobacillus was positively correlated (r = 0.56, P < 0.01) with Oleic acid, and negatively correlated (r =–0.50, P = 0.04) with N-Acetyl-D-lactosamine. Olsenella was positively correlated (r > 0.51, P < 0.01) with Oleic acid, Sebacic acid and negatively correlated (r =–0.50, P = 0.02) with 2-Methylbenzoic acid. Lachnoclostridium was positively correlated (r = 0.52, P = 0.01) with Nitrosobenzene. Ruminiclostridium 9 was positively correlated (r = 0.53, P = 0.01) with Sebacic acid.
The correlation of altered intestinal microbiota, metabolites and ADG, fecal score, and serum parameters in weaned calves are shown in Fig. 7B, Bifidobacterium, Lactobacillus, Leu-Ala, and Oleic acid was positively correlated (r > 0.51, P < 0.02) with ADG. Helicobacter was positively correlated (r = 0.53, P < 0.01) with fecal score, and Bifidobacterium, Lactobacillus, Olsenella, Oleic acid, and 1-Palmitoylglycerol was negatively correlated (r <–0.50, P < 0.04) with fecal score. Lachnoclostridium was positively correlated (r = 0.50, P = 0.01) with DAO and Bifidobacterium was negatively correlated (r =–0.51, P < 0.01) with DAO. Bifidobacterium was positively correlated (r = 0.51, P = 0.02) with IgG. Lachnoclostridium was positively correlated (r = 0.56, P = 0.01) with TNF-α. Olsenella and Oleic acid was positively correlated (r > 0.57, P < 0.03) with GSH-Px. Lachnoclostridium was positively correlated (r = 0.50, P = 0.02) with MDA and Bifidobacterium was negatively correlated (r =–0.55, P = 0.01) with MDA.