3.1 Blood routine, physiological and biochemical indices
Compared with the control group, the counts of lymphocytes, monocytes, granulocytes, the hemoglobin concentration and lymphocyte percentage in the blood of Songliao Black pigs in the experimental groups increased (p < 0.05). There was no difference in any other index (p > 0.05). Compared with the number of leukocytes, lymphocytes, monocytes, granulocytes and their percentages, the above increased indices of group II were higher than those of groups I and III, i.e., 2% dietary Eucommia polysaccharide had the greatest effect, compared with 1, or 3% (Table 1).
Compared with the control group, the concentrations of albumin and total protein in the serum of the experimental groups increased (p < 0.01); the contents of urea nitrogen and triglyceride decreased (p < 0.05); the contents of globulin and glucose were not different (p > 0.05). There was no difference in these biochemical indices among the three groups (p > 0.05) (Table 2).
3.2 Immunoglobulin and cytokine assays
Compared with the control group, the serum levels of immunoglobulins IgA, IgE and its subclass IgG2a of pigs in the experimental groups increased (p < 0.05), and that of IgG2b increased slightly (p > 0.05; Figure 1). The serum level of IFN-γ in the experimental groups increased (p < 0.05), and those of IL-4 and TNF-α increased slightly (p > 0.05).
3.3 Growth performance
There was no difference between the final average weight of the experimental groups (p > 0.05; Table 3). There was also no difference between the daily weight gain of group I and the control group (p > 0.05), whereas group II was higher than control (p < 0.05). There was no difference in daily feed intake between the groups (p > 0.05). The material to weight ratios of groups I and II were lower than control (p < 0.01).
3.4 Carcass performance
There was no difference in the live weight before slaughter between the groups (p > 0.05; Table 4). The slaughter rate and lean meat rate group II were higher than control (p < 0.05), whereas the fat rate was lower (p < 0.01). Except for carcass weight, there was no difference between groups I and II and control (p > 0.05). There were no differences in carcass length, carcass oblique length, skin thickness, eye muscle area, bone rate and skin rate between the groups (p > 0.05).
3.5 Meat quality
Compared with control, the PH24 of group II decreased (p < 0.05; Table 5). The yellowness (CIELAB b* value) and centrifugal water loss rate of group II were lower than control (p < 0.01), whereas the cooked meat rate was higher (p < 0.05). The centrifugal water loss of group I was lower than control (p < 0.05). The brightness and redness (CIELAB L* and a*), meat color score, marbling score, drip loss, water loss rate, shear force and intramuscular fat varied slightly between the groups (p > 0.05).
3.6 RNA quality and RNA sequencing quality
The RNA quality test showed that the total amount of RNA in 5-1/T1, 6-1/T2, 7-3/T3 and 8-3/C1, 9-4/C2 and 10-1/C3 of the experimental groups, the RNA integrity index (≥7.7), and the rRNA ratio (≥1.0) met the requirements for database establishment. The numbers of reads from the three control high-throughput sequencing samples were 45,911,760, 47,517,310 and 48,766,628 respectively, and the effective rates were 97.87, 96.68 and 97.76% respectively. The numbers of reads from the three experimental group samples were 48,449,042, 46,867,530 and 45,715,882, and the effective rates were 97.73, 97.85 and 97.50%, respectively. The data comparison rate and effective rate were above 95% (Table 6) and the ratios of base quality (Q30) of the sequencing were all over 93%, indicating good sequencing quality that met the quality control requirements.
3.7 Differential gene analysis
Screening the DEGs with FC ≥ 1 and q < 0.05 between the experimental and control groups identified 32 DEGs (Figure 2). A hierarchical cluster analysis of the DEGs and treatment conditions was carried out (Figure 3), finding that 26 genes were up-regulated and six were down-regulated. A total of 19 genes was differently expressed among the four groups, of which 18 were up-regulated (Figure 4). There were eight DEGs with difference multiples of > 2, i.e., ADAMTS4, PER1, STAC, SERPINE1, FASN, THRSP, SP7 and KRT80, indicating that Eucommia polysaccharide had a significant influence on fat deposition genes.
3.8 Protein interaction network
Analysis of the protein interaction network (Figure 5) revealed 23 nodes, of which 20 were up-regulated and three down-regulated, and 14 were DEGs.
3.9 Significant enrichment analysis of GO functions and KEGG pathways of DEGs
GO function annotation (Figure 6) and enrichment analysis (Figures 7, 8) were performed on the above 32 DEGs, of which 34 were significantly enriched in transferase activity, actin binding, among which acetyl coenzyme A, acyl coenzyme A metabolism, adipose tissue development and acyl glycerol homeostasis were significantly different. KEGG analysis (Figure 9) showed that differential genes were significantly enriched in the AMPK (Figure 10) and PPAR (Figure 11) signaling pathways. These results indicate that Eucommia polysaccharide had a significant effect on intramuscular fat deposition.