To investigate changes in advanced glycation end products (AGEs) and their receptor (RAGE) expression in the gastrointestinal (GI) tract in type 2 diabetic rats. Eight inherited type 2 diabetic rats Goto-Kakizak (GK) and ten age-matched normal rats were used in the study. From 18 wk of age, the body weight and blood glucose were measured every week and 2 wk respectively. When the rats reached 32 wk, two-centimeter segments of esophagus, duodenum, jejunum, ileum, and colon were excised and the wet weight was measured. The segments were fixed in 10% formalin, embedded in paraffin and five micron sections were cut. The layer thickness was measured in Hematoxylin and Eosin-stained slides. AGE [N epsilon-(carboxymethyl) lysine and N epsilon-(carboxyethyl)lysine] and RAGE were detected by immunohistochemistry staining and image analysis was done using Sigmascan Pro 4.0 image analysis software. The blood glucose concentration (mmol/L) at 18 wk age was highest in the GK group (8.88 ± 1.87 vs 6.90 ± 0.43, P < 0.001), a difference that continued to exist until the end of the experiment. The wet weight per unit length (mg/cm) increased in esophagus, jejunum and colon from the normal to the GK group (60.64 ± 9.96 vs 68.56 ± 11.69, P < 0.05 for esophagus; 87.01 ± 9.35 vs 105.29 ± 15.45, P < 0.01 for jejunum; 91.37 ± 7.25 vs 97.28 ± 10.90, P < 0.05 for colon). Histologically, the layer thickness of the GI tract was higher for esophagus, jejunum and colon in the GK group [full thickness (μm): 575.37 ± 69.22 vs 753.20 ± 150.41, P < 0.01 for esophagus; 813.51 ± 44.44 vs 884.81 ± 45.31, P < 0.05 for jejunum; 467.12 ± 65.92 vs 572.26 ± 93.60, P < 0.05 for colon]. In esophagus, the AGE and RAGE mainly distributed in striated muscle cells and squamous epithelial cells. The AGE distribution was much stronger in the GK group compared to the normal group both in the striated muscle layer and mucosa layer (immuno-positive area/ total measuring area %: 4.52 ± 0.89 vs 10.96 ± 1.34, P < 0.01 for muscle; 8.90 ± 2.62 vs 22.45 ± 1.26, P < 0.01 for mucosa). No visible difference was found for RAGE distribution between the two groups. In the intestine AGE and RAGE distributed in epithelial cells of villi and crypt. RAGE was also found in neurons in the myenteric and submucosal plexus. The intensity of AGE staining in mucosa of all segments and RAGE staining in neurons in all segments were strongest in the diabetes group. Significant difference for AGE was found in the epithelial cells of villi and crypt in duodenum (immuno-positive area/total measuring area %: 13.37 ± 3.51 vs 37.48 ± 8.43, P < 0.05 for villi; 0.38 ± 0.12 vs 1.87 ± 0.53, P < 0.05 for crypt) and for RAGE in neurons of all segments (e.g., for jejunum: no staining neurons% 0 vs 0, mild 36.0 ± 5.2 vs 28.7 ± 3.5, moderate 53.2 ± 4.8 vs 55.8 ± 5.4, strong 10.7 ± 1.1 vs 15.4 ± 2.0, P < 0.05). In the colon, RAGE was primarily found in neurons in the myenteric and submucosal plexus. It was stronger in the diabetes group than in the normal group (no staining neurons% 6.2 ± 0.2 vs 0.3 ± 0.04, mild 14.9 ± 2.1 vs 17.6 ± 1.5, moderate 53.1 ± 4.6 vs 44.7 ± 4.4, strong 25.6 ± 18 vs 43.6 ± 4.0, P < 0.05). In the rectum, RAGE was primarily found in the mucosa epithelial cells. The AGE and RAGE expression was up-regulated in the GI tract of GK diabetic rats and may contribute to GI dysfunction in type 2 diabetic patients.