Although NSAIDs are essential in the treatment of RA, patients with RA are more likely to experience gastrointestinal injuries via NSAIDs compared to patients without RA. These side effects affect the RA treatment. In this study, we investigated the effect of TEP on IMC-induced gastrointestinal injuries induced by IMC using AA rats and found that oral administration of TEP promoted the repair of IMC-induced gastrointestinal injuries in this RA model.

It is important to select an appropriate animal model to examine the effects of TEP on NSAID-induced gastrointestinal injury associated with RA. Collagen-induced arthritis, AA, and streptococcal cell wall-induced arthritis models are used as rodent models of RA [17, 18]. AA develops and rapidly progresses into polyarthritis. In the early stages of inflammation, interleukin 17 (IL-17), interferon, and tumor necrosis factor α, which are involved in macrophage activation, are expressed. In the later stages of inflammation, the levels of IL-4, IL-6, and transforming growth factor-β increase. Consequently, AA exhibits symptoms similar to human RA, such as joint swelling, cartilage deterioration, and lymphocyte damage [17, 18]. Therefore, it is often used as an animal model of RA. Moreover, in an RA model, gastrointestinal injuries were observed in normal rats after IMC administration, although they were more pronounced in AA rats [19]. Based on these reports, here we used AA rats instead of normal rats.

In this study, paw edema was confirmed 7 days after adjuvant administration in rats. Paw edema peaked on day 14 (Fig. 2). Paw edema was also observed after 42 days; however, the lesion areas in the stomach, jejunum, and ileum were significantly worse after 14 days (Fig. 3). When IMC was administered to normal rats, the lesion area (%) to the stomach, jejunum, and ileum was 0.56 ± 0.29, 0.56 ± 0.18, and 0.51 ± 0.15, respectively (n=6–15). Based on these results, we focused on the therapeutic effects of TEP on IMC-induced ulcers in AA rats. AA is a model of systemic inflammation, and the occurrence of paw edema in unadministered paws indicates the onset of systemic inflammation [7]. Gastrointestinal injury increases when systemic inflammation was caused [20]. In AA rats, systemic inflammation of AA rats at 14 days after adjuvant injection tended to increase more than at 21 and 42 days. In AA rats, local inflammation occurred 7 days after adjuvant injection, and systemic inflammation occurred after 14 days. Moreover, systemic inflammation peaked 14 days after adjuvant injection and decreased over time. Kato et al. reported that the onset of gastrointestinal injury caused by IMC occurs when systemic inflammation is enhanced [21]. IMC-induced gastrointestinal injuries were observed in AA rats on days 21 and 42 after adjuvant injection, and these injuries were similar to those observed in the control rats (IMC-induced normal rats) in this study. Therefore, in this study, the healing effect of TEP on IMC-induced gastric lesions was investigated in AA rats that experienced systemic inflammation 14 days after adjuvant administration. Paw edema of AA rats were similar in all groups.

OPN is an extracellular matrix protein [22] that is involved in biological processes such as bone remodeling, innate immunity, acute and chronic inflammation, and cancer [23, 24]. OPN expression is observed in immune cells such as intestinal epithelial cells, macrophages, dendritic cells, and T lymphocytes [25], suggesting an important role in acute and chronic inflammation [26]. OPN is expressed by various inflammatory cells and has been proposed as a biomarker for inflammatory diseases [24]. Therefore, in this study, we evaluated the therapeutic effects of TEP on IMC-induced gastrointestinal bleeding using digital imaging and immunohistochemical staining for OPN. Inflammation was defined as the presence of numerous OPN-positive cells. TEP administration resulted in fewer open arrows, indicating fewer OPN-positive cells, and reduced inflammation. Gastrointestinal injuries in normal rats were less severe than those in AA rats (Fig. 3). Our previous study showed that gastrointestinal injuries caused by IMC were similar in normal and sham-operated rats [9]. Analysis of the lesion area showed a significant decrease in the disability rate in the TEP-treated AA rats compared to that in the control group. Although, the HPβCD was used in combination with TEP to enhance the therapeutic effect of TEP, the efficacy did not increase in this study, since the lesion area (%) in the stomach, jejunum and ileum of IMC-induced AA rats treated with TEP alone (without HPβCD) was 3.8 ± 2.1, 1.6 ± 1.1, 2.5 ± 1.8%, respectively (n=3). Moreover, we previously reported that the lesion areas in IMC-induced AA rats treated with HPβCD were similar to those in IMC-induced AA rats [9]. These results suggest that HPβCD does not affect the therapeutic effects of TEP in rats with IMC-induced AA. However, it is unclear whether TEP is encapsulated by HPβCD. Therefore, further research on this topic is required. In addition, the oral administration of TEP significantly reduced NOS2 mRNA and NO levels in the stomachs of IMC-treated AA rats (Fig. 4). Similar results were obtained for the jejunum and ileum (Figs. 5 and 6). NO is abundant in the vascular endothelial cells and plays an important role in the regulation of blood flow. In addition, it is involved in the regulation of mucosal blood flow and the maintenance of mucus secretion in the small intestinal mucosa. However, excessive production of NOS2, an inducible isozyme, causes gastric and intestinal injuries [27]. Thus, the oral administration of TEP suppressed the factors involved in gastrointestinal injury. Thus, our study is the first to investigate the treatment of IMC-induced gastrointestinal injuries with TEP in AA rats, providing important information for the treatment of gastrointestinal injuries using NSAIDs in RA patients. On the other hand, PPIs are clinically used to treat gastrointestinal injuries in patients with RA and NSAIDs [8]. It remains to be determined whether PPIs or TEP are superior, and we believe that it is important to compare the therapeutic effects of PPIs and TEP, alone and in combination (Table 1).

In clinical practice, TEP is administered at a dose of 50–150 mg/day. In this study, we estimated that the average human body weight was approximately 50 kg, and the dose of TEP was determined to be 150 mg/50 kg (3 mg/kg). The RBM dose (2 mg/kg) was determined based on the same concept used for TEP. On the other hand, package inserts have shown that the recovery effect of TEP on acetic acid-induced ulcers increases with increasing concentration [28]. Moreover, Katoh et al. investigated the inhibitory activities of TEP (3.2–200 mg/kg, oral administration) on taurocholic acid/ HCl induced acute gastric mucosal lesions in rats, and showed that the lesion levels (%) were approximately 80% in the rat treated with 3.2 and 12.5 mg/kg of TEP [29]. In addition, the lesion levels (%) decreased to approximately 46% with 50 and 200 mg/kg TEP [29]. Previous studies have shown that the inhibitory activity of TEP is dose-dependent. In this study, we determined the dose of TEP based on the human dosage, and the dose (3 mg/kg) was lower than that reported in previous reports (100–200 mg/kg). Therefore, high doses of TEP may increase recovery from IMC-induced gastrointestinal injury in AA rats. However, the RBM dose was not within the upper limit. From these results, the RBM dosage can be increased, which may change the results when compared to TEP. That is a subject for future research.

We also discussed the mechanisms for the recovery effects of TEP on IMC-induced ulcers in the AA rat model. It has been reported that TEP induces and enhances the expression of heat shock protein (HSP) 70, suppresses the production of pro-inflammatory cytokines and NO, and induces HSP to inhibit NOS2 [30, 31]. Therefore, these factors are thought to improve NOS2 mRNA and NO levels and influence healing. On the other hand, it also suggests the involvement of increased eNOS expression [32, 33]. Although eNOS has also been reported to increase in an IMC-induced model of gastrointestinal injury [34], Kato et al. reported that NOS2 increased more than eNOS in an IMC-induced gastrointestinal injury model [34]. These findings suggest that NOS2 is more closely related to IMC-induced gastrointestinal bleeding than eNOS. RBM has also been reported to inhibit enhanced NO production via NOS2 activity induced by excess IMC [35]. This study confirmed that both drugs suppressed NOS2 mRNA and NO levels, which are thought to affect the healing of gastrointestinal injuries. Moreover, TEP protects the gastric mucosa by increasing the production of PGs [36], which has also been observed in RBM [37]. This suggests that a different factor underlies the effect of TEP in the treatment of gastrointestinal injuries via NSAIDs in AA rats. The secretion of bicarbonate ions (HCO3-) in the duodenum is reduced by IMC and inhibits PG production [38, 39]. HCO3- is secreted by secretin after TEP administration [40]. However, these effects were not observed with the RBM. Therefore, this factor may have a greater effects on the healing effects of TEP. On the other hand, many studies on the mechanism of NSAID-induced gastrointestinal injury in AA rats have been reported [41]. However, the exact mechanism remains unclear, and the difference in the inhibitory mechanism of TEP between AA rats and normal rats has not been fully elucidated. Further studies are needed to clarify the relationship between TEP and HCO3- in gastrointestinal injuries in AA rats.

Furthermore, we compared the effects of TEP on IMC-treated normal rats with those on IMC-treated AA rats. In this study, gastrointestinal injuries were not detected in IMC-treated normal rats when TEP was administered (N.D., n=3). Thus, TEP is effective in treating IMC-induced gastrointestinal injuries in both normal and AA rats and may be more effective in serious gastrointestinal injuries of RA induced by NSAIDs.

IMC, which is likely to cause gastrointestinal disturbances, was used in this study. IMC-induced gastrointestinal injury models have been used in other experiments [30]. However, it is also important to elucidate the therapeutic effect of TEP on gastrointestinal injuries induced in AA rats administered COX-2 selective inhibition such as celecoxib. Therefore, we investigated the efficacy of TEP treatment for gastrointestinal injuries induced by different NSAIDs in AA rats. In addition, we plan to consider inflammatory biomarkers in future studies.