Diabetes mellitus increases susceptibility to acute gastric injury and impairs
ulcer healing.
Pioglitazone as an agonist of
peroxisome proliferator-activated receptor gamma (
PPARgamma) is used as anti-diabetic drug and has additionally gastroprotective activities. However, the effect of
pioglitazone on the protection and healing of gastric mucosa under diabetic conditions is poorly understood. The aim of the present study was: 1) to compare the effects of treatment with PPARg
ligand (
pioglitazone) on healing of
acetic acid-induced
gastric ulcers and prevention of acute water immersion and restraint stress (WRS)-induced gastric lesions in normal rats and those with
streptozotocin (STZ)-induced
diabetes mellitus; 2) to assess the effects of
pioglitazone on the
mRNA expression of
cyclooxygenase-2 (COX-2), c-NOS,
interleukin-1beta and
hypoxia inducible factor-1 alpha (HIF-1alpha) in the gastric mucosa of rats with or without STZ-induced
diabetes mellitus; 3) to investigate the involvement of endogenous NO and proinflammatory
cytokines (IL-1beta,
TNF-alpha) in healing of chronic
gastric ulcers and in prevention of acute stress lesions by
pioglitazone in rats with or without STZ-induced
diabetes mellitus. Diabetes was induced in rats by single injection of STZ (70 mg/kg i.p.) four weeks prior to production of
gastric ulcers by
acetic acid method or induction of stress lesions by 3.5 hours of WRS. Non-diabetic rats were used as controls. Two major animal groups (A and B) were tested; A) diabetic and non-diabetic rats with chronic
gastric ulcers treated with 1)
pioglitazone (40 mg/kg-d i.g.), 2)
pioglitazone in combination of blocker of
NO synthase (L-NNA 20 mg/kg-d i.p.), and 3) saline (vehicle-control); and B) diabetic and non-diabetic rats exposed to 3.5 hours of WRS and pretreated with 1)
pioglitazone (40 mg/kg i.g.), 2)
pioglitazone in combination of blocker of
NO synthase (L-NNA 20 mg/kg i.p.), and 3) saline (vehicle-control). The gastric mucosal blood flow was assessed by H(2)-gas clearance method. The area of chronic
acetic acid ulcers and number of acute WRS-induced gastric lesions were assessed by planimetry or by counting of number of lesions, respectively. In rats with chronic
ulcers, the
mRNA expression of HIF-1alpha, IL-1beta and COX-2 was assessed by RT-PCR and
protein expression of
platelet endothelial cell adhesion molecule-1 (PECAM-1), COX-2 and cNOS was examined by Western blot. In rats with stress lesions, the
protein expression of COX-2, cNOS,
catalase,
PPAR and
heat shock protein 70 (HSP70) was examined by Western blot. In diabetic rats, a marked delay in
ulcer healing and increased susceptibility to WRS lesions were observed and these effects were accompanied by a significant decrease in GBF.
Pioglitazone significantly increased healing of chronic
gastric ulcers and exerted a strong protective effect against WRS-induced lesions, but these effects were attenuated by NO-inhibition with L-NNA. Interestingly, the
ulcer healing and gastroprotective effects of
pioglitazone were weak under diabetic conditions, and this effect on
ulcer healing was accompanied by impaired angiogenesis due to decreased PECAM-1 expression, attenuated expression of COX-2 and the increased expression of proinflammatory
cytokines compared to those in diabetic rats treated with vehicle. We conclude that: 1) experimental diabetes in rats impairs healing of chronic
ulcers and enhances acute stress lesions due to an increase in the expression and release of proinflammatory
cytokines such as
TNF-alpha and IL-1beta; 2) the
ulcer healing effect of
pioglitazone, which is, at least in part, mediated by endogenous NO, is significantly attenuated by L-NNA in diabetic rats despite increased COX-2 expression at the
ulcer edge; 3) the formation of acute gastric lesions induced by WRS is also attenuated by pretreatment with
pioglitazone due to increased GBF probably mediated by NO, as the administration of L-NNA reversed, in part, the preventive action induced by this
PPARgamma ligand, and 4)
pioglitazone is effective both in healing of chronic
ulcers and protection against WRS lesions though its action under diabetic conditions seems to be attenuated, possibly due to reduction in NOS-NO system, angiogenesis and increased expression and release of proinflammatory
cytokines.