The intestinal barrier maintains intestinal immune response by affecting the integrity of the intestine [29]. The intestine is mainly composed of the mucous layer in the intestinal mucosa, intestinal epithelial cells and their close connections and the underlying layers of the mucosa. Tight junctions are the most important intercellular connections. Various proteins constitute the tight junction, such as occludin, claudin, junction adhesion molecule and peripheral cytoplasmic proteins such as ZO-1 and ZO-2. Occludin is a transmembrane protein that maintains and regulates tight connections [30]. Intestinal injury in piglets is one of the well-established animal models for investigating infant nutrition and gastrointestinal physiology. In this study, spermine intake enhanced the occludin mRNA levels in the jejunum of piglets. Claudin is the most important component of tight connections and mainly regulates the permeability of the intestinal barrier [31]. Our results showed that spermine ingestion increased the claudin-1, claudin-2, claudin-3, claudin-12, claudin14, and claudin-15 gene expression levels in the jejunum of piglets. Additionally, the occludin and actin skeleton can form a stable system by the C terminal of the peripheral membrane protein ZO-1 by combining with actin and stress fibre. ZO-2 interacts with ZO-1 to co-exist in the tight junction of epithelial cells and the adhesion site of non-epithelial cells [32]. Our study showed that ZO-1 and ZO-2 gene expression levels were improved after spermine intake in the jejunum of piglets. ZO-1 distribution was affected by MLCK [33], which is a kind of calmodulin-dependent kinase that causes the contraction of the actomyosin through phosphorylating myosin light chain, causing the cytoskeleton to contract and damage the barrier function [34]. This finding is in agreement with our result that spermine supplementation decreased the MLCK mRNA levels in the 7-hour and 3-day supplementation in the jejunum of piglets. D-lactate is the metabolite of bacterial fermentation. A large amount of D-lactate produced by bacteria in the intestinal tract enters the blood through the damaged mucosa after the loss of intestinal villus tip or the increase of intestinal permeability after intestinal injury. Mammals do not have an enzymatic metabolism system for D-lactate; thus, the levels of serum D-lactate can reflect the degree of intestinal mucosal damage and permeability [35]. The results of our experiment showed that feeding spermine for 7 hours, 3, 6 and 9 days significantly reduced the content of serum D-lactate. At present, no report is available regarding the effect of spermine on the content of serum D-lactate in piglets. Diamine oxidase is a highly active intestinal enzyme in the intestinal villus tip of mammals that plays a role in histamine and polyamine metabolism, and its activity is closely related to the synthesis of nucleic acid and protein in mucosal cells [36]. Diamine oxidase reflects the degree of integrity and damage of the intestinal physical barrier. Fang et al found that extended spermine treatment time decreased the activity of diamine oxidase in the jejunum of piglets [27]. In this study, the activity of diamine oxidase was negatively correlated with claudin-1, occludin mRNA levels in the jejunum of piglets, respectively. Therefore, spermine supplementation may protect the intestinal integrity in vivo. This result was consistent with in the vitro study, that is, spermine significantly increased TER and the expression of tight junction proteins, and decreased paracellular permeability in IPEC-J2 cells. Occludin and claudin-1 proteins were significantly increased after 0.1 µM spermine treatment. The results of this experiment are consistent with those of previous studies. For example, Yu et al showed that the expression of occludin protein in IEC-Cdx2L1 cells was significantly decreased after depletion of polyamine treatment with DFMO [37]. Moreover, 0.2 and 0.7 mM arginine improved the expression of occludin and claudin-1 proteins in IPEC-J2 cells [38]. To further investigate the protective effect of spermine from intestinal damage, we also explored the injury model in IPEC-J2 cells induced by TNF-α that can induce inflammation. The results showed that TNF-α significantly reduced the expression of occludin protein in porcine intestinal epithelial cells, indicating that the injury model of porcine intestinal epithelial cells induced by TNF-α was established successfully, which was consistent with the results of Xiao et al [39]. In the present study, 0.1 µM spermine significantly increased TER and the protein expression of ZO-1, occludin and claudin-1, and decreased paracellular permeability after TNF-α challenge, indicating that 0.1 µM spermine could restore the injury of barrier of porcine intestinal epithelial cells. This finding was consistent with the results of previous studies, that is, arginine alleviated the damage of ZO-1 protein in IPEC-J2 cells induced by hypoxia [40] and asparagine significantly increased the expression of occludin and claudin-1 protein in the jejunum and ileum of weaned piglets after LPS challenge [41]. Glial-derived S-nitroso-glutathione can induce the increase in ZO-1, occludin and claudin-1 protein expression levels in transgenic mice whose intestinal barrier function was impaired because of removed intestinal glial cells [42]. This result was also basically consistent with the results of previous studies, that is, the dietary addition of tributyrin reduced intestinal damage caused by acetic acid in porcine colitis model [43], and arginine had a significant protective effect on intestinal injury in weaned piglets induced by LPS [44]. Taken together, spermine protects intestinal integrity by improving the TER and expression of tight junction protein, and decreasing paracellular permeability.
The RhoA, Rac1, PLC-γ1 and CDC42 gene expression levels are closely related to the protein expression of tight junction [45–48]. In our experiment, the Rac1 gene expression was positively related to the mRNA levels of occludin. PLC-γ1 gene expression was positively related to the claudin-3, claudin-14, claudin-15 and ZO-2 mRNA levels in the jejunum. In addition, a positive correlation was observed between RhoA and claudin-1, claudin-12 mRNA levels, respectively. Moreover, the mRNA levels of occludin were positively correlated with the CDC42 gene expression. The gene expression of MRLC was also positively correlated with the mRNA levels of occludin (r = + 0.775, P = 0.024) and claudin-1 (r = + 0.815, P = 0.014). These results suggested that the improvement of occludin, claudin-1, claudin-3, claudin-12, claudin-14, claudin-15 and ZO-2 mRNA levels may be partly ascribed to the upregulation of the mRNA levels of Rac1/PLC-γ1 signalling pathway in the jejunum of piglets. However, whether the Rac1/PLC-γ1 signalling pathway is involved or not in the restoration process is unclear. Therefore, we explored the relationship between spermine and the Rac1/PLC-γ1 signalling pathway.
Rac1, a small GTPase 1 protein belonging to the Rac family, controls glucose uptake, cell growth, cytoskeleton recombination and protein kinase activation [49]. PLC is involved in cell signal transduction. PLC-γ1 is a member of the PLC family and plays an important role in cell signal transduction induced by growth factors, such as cell proliferation, cell migration, cell differentiation and membrane fold formation [50]. Our results also showed that spermine treatment significantly increased the expression of Rac1, PLC-γ1, RhoA, CDC42 and MRLC genes in jejunum, indicating that spermine may restore the intestinal epithelium of piglets possibly by regulating the gene expression related to Rac1/PLC-γ1 signalling pathway. Moreover, CDC42 is the downstream of the activated Rac1 in primary keratinocytes from mice [51]. In this study, a significant positive correlation was found between the Rac1 and CDC42 gene expression levels in the jejunum, suggesting that the increased CDC42 mRNA levels were partly related to the improved Rac1 gene expression. In vitro study, spermine significantly increased expression of Rac1, PLC-γ1, RhoA, CDC42, Rock1 and MRLC genes in cells. Spermine can also significantly increase the expression of Rac1, PLC-γ1, RhoA, CDC42 and Rock1 genes in the intestinal epithelial barrier injury model of IPEC-J2 cells induced by TNF-α. These findings were consistent with previous studies indicating that spermidine increases the gene expression levels of Rac1, PLC-γ1, RhoA and CDC42 in IEC-6 cells [52, 53], and asymmetric dimethylarginine increases the expression level of Rock1 gene in rat pulmonary artery smooth muscle cells [54]. Moreover, results of our experiment were also in accordance with the effects of ornithine decarboxylase on polyaminated RhoA protein expression in ODC-deficient cell line, Rat-1 tsRSVLA29 cells, NIH 3T3, and Jurkat cells [55] and the RhoA protein content after DFMO treatment in IEC-6 cells [56, 57]. In the present study, spermine significantly increased GTP-rac1 protein content and GTP-rac1/total rac1 ratio in cells. This results was consistent with previous study indicating that Rac1 protein expression is reduced by putrescine depletion in IEC-Cdx2L1 cells [22] and DFMO treatment in differentiated IEC-Cdx2L1 cells [22]. Moreover, putrescine depletion decreased the expression level of GTP-rac1 protein in IEC-6 cells [24, 26, 58]. Further experiment showed that 40 ng/mL TNF-α + 0.1 µM spermine + 160 µM NSC-23766 significantly decreased TER and protein expression of ZO-1, occludin, claudin-1, and the content of GTP-rac1 and PLC-γ1 phosphorylated proteins, as well as the gene expression of Rac1, MRLC, CDC42 and Rock1, and increased paracellular permeability in the presence of 40 ng/mL TNF-α and 0.1 µM spermine. The 40 ng/mL TNF-α + 0.1 µM spermine + 3 µM U73122 significantly decreased TER and the expression of ZO-1, occludin, claudin-1, GTP-rac1, PLC-γ1 phosphorylated protein and Rac1, PLC-γ1, MRLC genes and increased paracellular permeability, suggesting that spermine protects intestinal barrier integrity through the Rac1/PLC-γ1 signalling pathway in pigs. Results of our experiment are consistent with those of the previous study: with the prolongation of the treatment time of Rho kinase inhibitor Y27632, the expression of GTP-rac1, GTP-RhoA protein decreased significantly in scratch wound IEC-6 cells [25].