Introduction

Amino acids (AAs) are not only needed for protein production but can be considered precursors of energy, signalling molecules and microbiota modulators. Therefore, AAs can contribute to restoring gut health and, in turn, improving general health1. L-Glutamate (Glu) and L-Glutamine (Gln) are abundant AAs in the body and have traditionally been considered dispensable AAs. However, they are currently regarded as a conditionally indispensable nutrient under stress conditions, including the weaning of piglets2. These two AAs can benefit the gut health of piglets by acting as sources of energy for the intestinal cells, as precursors for promoting cell proliferation in the gut and as precursors for the immune cells3,4,5,6. Glutamate is mainly utilised as a source of energy by enterocytes3,4. However, it can be used as a precursor for the synthesis of Gln and glutathione4; thus, it plays a key role in preserving the gut from oxidative damage.

The beneficial effects of Gln on the gut are mainly related to its use as a source of energy for enterocytes5, as metabolic fuel for the immune cells (including lymphocytes and macrophages) and by supplying substrates for the synthesis of glucosamines, nucleic acids, nucleotides and adenosine triphosphate (ATP)6,7,8. Furthermore, Gln can modulate the phosphorylation of the mammalian target of rapamycin (mTOR) which is involved in the regulation of protein synthesis in the intestine9.

Previous studies have shown that the supplementation of Glu alone and Gln alone could improve the growth performance, and gut health of weaning piglets in terms of gut integrity, acting as modulators of the mucosal gene expression and gut microbial community1,10,11,12,13,14.

The metabolism of Glu and Gln is closely connected; in fact, Glu is the immediate product of the Gln metabolism, produced by the action of glutaminase; Glu can be combined with ammonia (NH3) to produce Gln by the action of glutamine synthetase in some tissues and cells, such as the liver and skeletal muscles7. Therefore, it is plausible that the combined supplementation of Glu and Gln could have synergistic effects15. However, to date, only a few studies have investigated the effect of mixing Glu and Gln supplementation on piglets, suggesting a positive effect on growth performance and gut morphology parameters15,16. Up to now, no study has investigated the effect of mixing Glu and Gln at different doses regarding the growth, health and gut eubiosis of post-weaning piglets. Therefore, in the present study, the hypothesis that mixing different doses of Glu and Gln affected the growth, immune response and gut health of post-weaning piglets was tested. The aims of the present study were to (1) investigate the various beneficial effects of mixing different doses of Glu and Gln on the intestinal health and growth of post-weaning piglets and to evaluate whether mixing doses of Glu and Gln could be more promising than providing a single AA, (2) elucidate the mode of action of mixing different doses of Glu and Gln on the gut health of post-weaning piglets and (3) identify the best supplementation ratio of Glu and Gln for sustaining piglets during the post-weaning phase.

Results

To test the hypothesis that mixing different doses of Glu and Gln affected the growth, immune response and gut health of post-weaning piglets an in vivo trial was carried out; the piglets were assigned to 6 groups fed (1) a standard diet (CO), (2) CO plus 6 kg/Ton of Glu alone (100Glu), (3) CO plus 75% of Glu (4.5 kg/Ton) + 25% of Gln (1.5 kg/Ton) (75Glu + 25Gln), (4) CO plus 50% of Glu and Gln (3 kg/Ton) (50Glu + 50Gln), (5) CO plus 25% of Glu (1.5 kg/Ton Glu) and 75% of Gln (4.5 kg/Ton) (25Glu + 75Gln) and (6) CO plus 6 kg/Ton og Gln (100Gln) .

Performance and faecal score

During the trial, one pig, two pigs and one pig from the 75Glu + 25Gln, 25Glu + 75Gln and 100Gln groups respectively, were excluded due to health impairment and a substantial reduction in feed intake. The results of the growth performance are reported in Table 1. The live body weight (BW) at day (d) 0 (d0) (weaning), d7 and d14 did not differ among the dietary groups. At d21 post-weaning, the ever-increasing level of Glu tended to linearly reduce the BW (P = 0.065). The average daily gain (ADG) from d0 to d7 was not affected by the diet. From d7 to d14, the AAs increased the ADG of the piglets as compared with the CO (Control vs. AAs addition, P = 0.050). In the period d14–d21, the diet generally affected the ADG (P = 0.026), and the CO group had a higher ADG as compared with the groups supplemented with the AAs (Control vs. AAs addition, P = 0.015). Furthermore, from d14 to d21, the inclusion of Gln tended to linearly increase the ADG (P = 0.073). From d0 to d21 a linear effect of the Glu-Gln ratio was observed for the ADG with a favourable effect of the highest dose of Gln (P = 0.051). The feed intake (FI) was not affected by diet for the periods d0–d7 or d7–d14. In the periods d14–d21 and d7–d21, the FI tended to have a linear effect on the Glu:Gln ratio, which increased the value of the highest dose of Gln (P = 0.069 and P = 0.089, respectively). The gain to feed (G:F) ratio was not affected by diet for the periods d0–d7, d7–d14 or d14–d21. From d7 to d21, the G:F ratio tended to have a linear effect on the Glu:Gln ratio (P = 0.051), and the 100 Glu group tended to have a lower G:F ratio than the other groups supplemented with mixed doses of AAs (P = 0.078).

Table 1 Effect of the dietary supplementation (6 kg/T) with glutamate and glutamine in different ratio on growth performance of post-weaning piglets.
Full size table

Figure 1 shows the effect of diet on the faecal score and the faecal index of the piglets. The faecal score from d0 to d8 was reduced in the groups supplemented with the AAs as compared with the CO group (P = 0.002). The groups having the mixed doses of Glu and Gln had lower faecal scores as compared with the groups having Glu alone (P = 0.08) and Gln alone (P = 0.06). There was a quadratic effect of the Glu:Gln ratio (P = 0.005). From d9 to d21, the faecal score tended to be reduced in the groups supplemented with the AAs as compared with the CO group (P = 0.08). The diarrhoea index was reduced in the groups supplemented with the AAs as compared with the CO group in both periods (d0–d8, P = 0.006; d9–d21, P < 0.001). For both periods, there was a quadratic effect of the Glu:Gln ratio (d0–d8, P = 0.0007; d9–d21, P = 0.005). For the period d9–d21, the Glu:Gln ratio also exerted a linear effect (P < 0.0001). For both periods, the groups having the mixed doses of Glu and Gln had a lower diarrhoea index (d0–d8, Glu alone vs. mixed addition P = 0.0006, Gln alone vs. mixed addition P = 0.02; d9–d21, Glu alone vs. mixed addition P = 0.0004, Gln alone vs. mixed addition P = 0.05).

Figure 1
figure 1

The effect of dietary supplementation (6 g/T) with Glu and Gln in different ratios on the faecal score and the faecal index of weaned piglets from d0 to d8 (A: faecal score; C: faecal index) and from d9 to d21 (B: faecal score; D: faecal index). The data of the faecal index were transformed from log values. Diet: CO = standard diet; 100Glu = CO plus 6 kg/Ton Glu; 75Glu + 25Gln = CO plus 4.5 kg/Ton Glu and 1.5 kg/Ton Gln; 50Glu + 50Gln = CO plus 3 kg/Ton Glu plus 3 kg/Ton Gln; 25Glu + 75Gln = CO plus 1.5 kg/Ton Glu and 4.5 kg/Ton Gln; 100Gln = CO plus 6 kg/Ton Gln.

Full size image

Blood parameters

Table 2 shows the effect of the dietary supplementation on blood values at d8. No difference was observed for the values of haemoglobin (HGB), mean corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC), platelets and the percentage of red cell distribution width (RDW) in the number of leukocytes, neutrophils, lymphocytes, basophils and the percentage of basophils. The level of the red blood cell (RBC) count tended to be higher in the Glu alone group (100Glu) as compared with the other groups supplemented with mixed doses of AAs (P = 0.059). For hematocrit % (HCT %), a quadratic effect of the Glu:Gln ratio (P = 0.023) was observed, and the 100Glu group had a higher value as compared with the other groups supplemented with mixed doses of AAs (P = 0.025). The level of mean corpuscular volume (MCV) was higher in the Gln alone group (100Gln) as compared with the other groups supplemented with mixed doses of AAs (P = 0.049) (Table 2). Considering the white cell fraction, the diet influenced the number and the percentage of eosinophils (P = 0.011; P = 0.010, respectively). For the eosinophil count, a linear effect of the Glu:Gln ratio was observed (P = 0.026). For the percentage of eosinophils, a higher value was observed in the Gln alone group (100Gln) as compared with the other groups supplemented with mixed doses of AAs (P = 0.012), and a negative linear effect of Glu addition was observed (P = 0.012). A trend and a linear effect for the Glu:Gln ratio was observed regarding the number and the percentage of monocytes (P = 0.068; P = 0.028, respectively), and the Gln alone group tended to have a lower value of this white cell fraction as compared with the groups receiving the mixed doses of AAs (P = 0.063; P = 0.075, respectively). The diet tended to affect the percentage of neutrophils (P = 0.075); a quadratic effect of the Glu:Gln ratio was observed (P = 0.012) in which the 50Glu + 50Gln group had the highest value. The diet tended to affect the percentage of lymphocytes (P = 0.1); a quadratic effect of the Glu:Gln ratio was observed (P = 0.012) in which the 50Glu + 50Gln group had the lowest value. In addition, the Gln alone group (100Gln) tended to have a higher lymphocyte as percentage compared with the other groups supplemented with AAs (P = 0.092) (Table 2).

Table 2 Effect of the dietary supplementation (6 kg/T) with glutamate and glutamine in different ratio on haematological parameters of piglets at 8 days post-weaning.
Full size table

No difference due to diet was observed for any of the parameters, except for HGB for which the CO group tended to have a lower value as compared with the groups supplemented with AAs (P = 0.079) and for the percentage of eosinophils for which the diet showed a trend (P = 0.057); no significant contrasts were determined (Supplementary Table 1).

Caecal microbiota

A total of 4,098,584 quality checked reads on 103 samples were obtained resulting in 3523 different amplicon sequence variants (ASVs). The rarefaction curves relative to all the samples are shown in Supplementary Fig. 1. Considering the overall samples, eighteen different phyla were identified in the caecum; the most abundant phylum was Firmicutes (70%) followed by Bacteroidetes (17%). At the family level, eighty different genera were assigned; the most abundant were Peptostreptococcaceae (18%), and Eryspelotrichaceae (17%). At the genus level, the most abundant assigned genera were Turicibacter (15%) and Terrisporobacter (13%). Figure 2 shows the Alpha diversity indices of the dietary groups at the two time points. The dietary groups showed a fairly constant ASVs distribution in the caecum; the Alpha index values (Chao1, Shannon and InvSimpson) were not affected by diet. Time did not influence the Shannon and InvSimpson indices, but reduced the Chao1 index from d8 (265) to d21 (231) (P = 0.008). For the Beta diversity index (Bray–Curtis distance), a clear effect of time was observed (P = 0.01; R2 = 0.068). Two well-defined clusters regarding time were shown in the Non-Metric Multidimensional Scaling (NMDS) plot of the Bray–Curtis distance matrix (Supplementary Fig. 2). Diet did not affect the Beta diversity index. Results for the different taxa of the dietary groups at the family and genera levels at d8 and d21 are reported in Table 3. At d8, the 100Gln group had a lower abundance of family Fusobacteriaceae as compared with the CO group (adj.p < 0.0001); considering the comparison between the Glu alone group vs. the mixed addition groups, the 100Glu group had a lower relative abundance of Enterococcaceae (adj.p < 0.0001) and Lactobacillaceae (adj.p < 0.0001), and a higher abundance of Erysipelotrichaceae (adj.p = 0.02). Considering the comparison between the Gln alone group vs. the mixed addition groups, the 100Gln group had a lower abundance of Fusobacteriaceae (adj.p < 0.0001) and a higher level of Clostridiales_vadinBB60_group (adj.p = 0.001). Considering the comparisons at the genus level at d8, the CO group had a higher relative abundance of Selenomonas (adj.p = 0.007) and Mogibacterium (adj.p < 0.0001) than the 100Glu group, a higher relative abundance of Pelistega (adj.p < 0.0001) and Selenomonas (adj.p = 0.006) than the 50Glu + 50Gln group, a higher relative abundance of Selenomonas (adj.p = 0.004) and Blautia (adj.p = 0.023) than the 25Glu + 75Gln group and a higher relative abundance of Fusobacterium (adj.p < 0.0001) and UB1819 (Family Ruminococcaceae) (adj.p < 0.0001) as compared with the 100Gln group. The comparison between the Glu alone group or the Gln alone group vs. the AAs mixed diets showed a decrease in the relative abundance of Pediococcus (adj.p < 0.0001), Enterococcus (adj.p < 0.0001) and Lactobacillus (adj.p = 0.008), in the 100Glu group, and a decrease in Pediococcus (adj.p < 0.05) and Fusobacterium (adj.p < 0.0001) in the 100Gln group. At d21, the CO group had a lower abundance of Fibrobacteraceae vs. each of the groups supplemented with AAs (adj.p < 0.0001) and a lower relative abundance of Bacteroidales_RF16_group (adj.p = 0.028) and Peptococcaceae as compared with the 100Gln group (adj.p = 0.037). Considering the comparisons at the genus level at d21, the CO group had a lower relative abundance of Selenomonas_3 as compared with the 100Glu group (adj.p = 0.006) and the 25Glu + 75Gln groups (adj.p = 0.007), and a lower relative abundance of Succiniclasticum as compared with the groups supplemented with AAs (adj.p < 0.0001).

Figure 2
figure 2

Alpha diversity indices per Diet and time points (day 21 and day 8 post-weaning). Diet: CO = standard diet; 100Glu = CO plus 6 kg/Ton Glu; 75Glu + 25Gln = CO plus 4.5 kg/Ton Glu and 1.5 kg/Ton Gln; 50Glu + 50Gln = CO plus 3 kg/Ton Glu plus 3 kg/Ton Gln; 25Glu + 75Gln = CO plus 1.5 kg/Ton Glu and 4.5 kg/Ton Gln; 100Gln = CO plus 6 kg/Ton Gln.

Full size image
Table 3 Effect of the dietary supplementation (6 kg/T) with glutamate and glutamine in different ratio on microbial taxa composition of piglets at 8- and 21-days post-weaning.
Full size table

Intestinal morphology

Table 4 shows the effect of dietary supplementation on the intestinal morphology measurements at d8 and d21. At d8, no difference due to diet was observed for villus height, villus width, crypt depth, and the number of goblet cells and duodenal glands. However, at d8, in the duodenum, the number of duodenal glands was higher in the 100Gln group as compared with the other groups supplemented with mixed doses of AAs (P = 0.048); a quadratic effect of the Glu:Gln ratio was observed (P = 0.048). At d8, in the jejunum, the number of goblet cells was lower in the CO group as compared with the 50Glu + 50Gln group (P = 0.05); in the ileum, the villus height tended to be higher in the 100Glu group as compared with the other groups supplemented with mixed doses of AAs (P = 0.097). The villus width tended to be higher in the CO group as compared with the 100Glu group (P = 0.079) and the number of goblet cells tended to increase linearly with the Gln dose (P = 0.090). No difference due to diet was observed for the interstitial score in any of the intestinal segments at d8. Diet influenced the intraepithelial score in the jejunum at d8; the CO group had a higher probability of having a higher intraepithelial lymphocyte score as compared with the groups supplemented with Glu, Gln or both (P < 0.016), and the 25Glu + 75Gln group showed a lower probability of a high intraepithelial lymphocyte score as compared with the CO group (P = 0.05) (Fig. 3). At d21, no differences due to diet were observed for villus height, villus width, crypt depth, and the number of goblet cells and duodenal glands (Table 4). At d 21, in the duodenum, the crypt depth tended to be higher in the 100Gln group as compared with the other groups supplemented with mixed doses of AAs (P = 0.083). In the jejunum, the villus width tended to be higher in the CO group as compared with the 25Glu + 75Gln group (P = 0.081) at d21. No difference due to diet was observed for the interstitial and intraepithelial scores in any of the intestinal tracts at d21.

Table 4 Effect of the dietary supplementation (6 kg/T) with glutamate and glutamine in different ratio on the intestinal morphology of piglets at 8- and 21-days post-weaning.
Full size table
Figure 3
figure 3

The effect of the diet on the cumulative probabilities of having each score for the intraepithelial lymphocytes in the jejunum of piglets at 8 days post-weaning. (A) Comparison between the CO group and the groups supplemented with AAs; (B) Cumulative probabilities of having each score in each dietary group. Diet: CO = standard diet; 100Glu = CO plus 6 kg/Ton Glu; 75Glu + 25Gln = CO plus 4.5 kg/Ton Glu and 1.5 kg/Ton Gln; 50Glu + 50Gln = CO plus 3 kg/Ton Glu plus 3 kg/Ton Gln; 25Glu + 75Gln = CO plus 1.5 kg/Ton Glu and 4.5 kg/Ton Gln; 100Gln = CO plus 6 kg/Ton Gln.

Full size image

Intestinal gene expression

No difference due to diet was observed for the expression of the Innate Immune Signal Transduction Adaptor (MyD88), Nuclear Factor Kappa B Subunit (NFKB2), Tumor Necrosis Factor (TNF), C-X-C Motif Chemokine Ligand 8 (IL8;gene CXCL8), Occludin (OCLN), Tight Junction Protein 1(ZO-1), Mucin 13, Cell Surface Associated (MUC13), Glutathione Peroxidase 2 (GPX-2) and Glutamate-Ammonia Ligase (GLUL) and Regenerating Family Member 3 Gamma (REG3G) in the jejunal mucosa of piglets. Some effects or trends for GLUL, ZO-1 and GPX2 were found with the orthogonal contrasts as shown in Table 5. The CO group tended to have a lower expression of GLUL as compared with the groups supplemented with AAs (P = 0.088). The expression of ZO-1 was higher in the 100Gln group as compared with the groups supplemented with mixed doses of AAs (P = 0.025); a linear tendency of the diet was observed (P = 0.079). The expression of GPX-2 tended to be higher in the 100Glu group as compared with the other groups supplemented with AAs (P = 0.086).

Table 5 Effect of the dietary supplementation (6 g/T) with glutamate and glutamine in the different ratios on the jejunal gene expression of piglets at 8 days post-weaning.
Full size table

Discussion

Supporting the integrity, functionality and morphology of the gut is a key strategy for sustaining piglets during the weaning transition period1. The results obtained in the present study demonstrated that mixing doses of Glu and Gln can benefit gut health and the growth of post-weaning piglets during the first 2 weeks post-weaning, sustaining the thesis that these AAs play a crucial role in supporting and restoring the gut health of piglets.

The results suggested that both AAs were able to improve piglet growth during the more acute inflammatory period (until the second week post-weaning) as compared with the control diet, the piglets of which recovered in the third week post-weaning17,18. The recovery of the control group could have been due to compensatory or “catch up” growth which occur when animals are supplied with sufficient nutrition after restricting feed or following adverse conditions, such as weaning19. In fact, it is known that transient anorexia and gut inflammation can increase piglet morbidity in the first weeks post-weaning while, after 2 weeks, the gut can recover, and improve the digestion and absorption of nutrients, promoting piglet growth20. The faster recovery of piglets supplemented with Glu and Gln could have been related to their functional roles regarding gut health which will be discussed later in detail. The supplementation of Gln was associated with more promising effects than that of Glu; in fact, a linear effect with higher values regarding the ADG and the G:F ratio for the groups receiving more Gln than Glu was observed considering the overall period. to date, no direct comparison between supplementation using different quantities of Glu and Gln has been reported in the literature. However, previous studies have suggested a positive effect of mixed doses of Glu and Gln during the pre- and post-weaning phases (0.88% of 50% of Glu:Gln ratio21) and later in the piglets’ life16. In addition, Liu et al.Bioinformatic and statistical analysis

Statistical analysis was carried out using SAS version 9.3 (SAS Inst. Inc., Cary, NC, USA). The GLM procedure was used to fit the measurements carried on piglets with a linear model including batch, litter of origin, sex of piglets and diet. Orthogonal contrasts were used to assess the effect of Glu/Gln supplementation as follows: CO vs. AAs addition: (CO vs. 100Glu, 75Glu + 25Gln, 50Glu + 50Gln, 25Glu + 75Gln, 100Gln), Glu alone vs. mixed addition (100Glu vs. 75Glu + 25Gln, 50Glu + 50Gln, 25Glu + 75Gln), Gln alone vs. mixed addition (100Gln vs. 75Glu + 25Gln, 50Glu + 50Gln, 25Glu + 75Gln). In addition, the linear and quadratic effects of AAs supplementation were tested. Data regarding the faecal index were log-transformed before the statistical analysis.

Data regarding the inflammatory evaluation of the intestinal mucosa, such as categorical response variables in numerical order, were assessed using the Generalised Linear Mixed Model (GLIMMIX) procedure of SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), considering a multinomial distribution and calculating the cumulative probabilities of having each score for each experimental diet.

Microbiota analysis was carried out using the DADA2 pipeline64. Taxonomic categories were assigned using the Silva Database (release 138) as a reference65. Alpha (Shannon, Chao1 and InvSimpson indices) and Beta diversity (calculated as the Bray Curtis distance matrix), as well as the abundance of taxonomic categories, were analysed utilising R software 3.666 using the PhyloSeq67, Vegan68 and DESeq269 packages. The Alpha diversity indices were analysed using an ANOVA model (lm function) including batch, diet, time point (d8 and d21) and the interaction between diet and time-point as factors. Beta diversity was analysed using a PERMANOVA Adonis test model (adonis.test function) which included batch, diet, time point, and the interaction between diet and time point as factors. The effects on Beta diversity were visualised using a Non-Metric Multidimensional Scaling (NMDS) approach (plot_ordination function). The analysis was then carried out in the dataset composed of each single time point. Furthermore, orthogonal contrasts were carried out to assess the effect of Glu/Gln supplementation on Alpha and Beta diversity. The differences in taxonomic abundances among the diets were analysed using the DESeq2 package, based on negative binomial generalised linear models, including and applying the Benjamini–Hochberg method for multiple testing correction (estimateSizeFactors function)69. Statistical significance was set at P ≤ 0.05 and tendencies at P ≤ 0.10.

Ethics declarations

The procedures complied with Italian law pertaining to experimental animals and were approved by the Ethic-Scientific Committee for Experiments on Animals of the University of Bologna, Italy and by the Italian Ministry of Health (Approval No. 503/2018 of 2 July, 2018). The study was carried out in compliance with the ARRIVE guidelines.