Expression Pattern of Fatty Acid Binding Proteins in Celiac Disease Enteropathy

Abstract

Celiac disease (CD) is an immune-mediated enteropathy that develops in genetically susceptible individuals following exposure to dietary gluten. Severe changes at the intestinal mucosa observed in untreated CD patients are linked to changes in the level and in the pattern of expression of different genes. Fully differentiated epithelial cells express two isoforms of fatty acid binding proteins (FABPs): intestinal and liver, IFABP and LFABP, respectively. These proteins bind and transport long chain fatty acids and also have other important biological roles in signaling pathways, particularly those related to PPARγ and inflammatory processes. Herein, we analyze the serum levels of IFABP and characterize the expression of both FABPs at protein and mRNA level in small intestinal mucosa in severe enteropathy and normal tissue. As a result, we observed higher levels of circulating IFABP in untreated CD patients compared with controls and patients on gluten-free diet. In duodenal mucosa a differential FABPs expression pattern was observed with a reduction in mRNA levels compared to controls explained by the epithelium loss in severe enteropathy. In conclusion, we report changes in FABPs' expression pattern in severe enteropathy. Consequently, there might be alterations in lipid metabolism and the inflammatory process in the small intestinal mucosa.

Figure 1

Increased levels of IFABP in serum samples of active CD. IFABP serum levels (pg/mL) were assessed by commercial quantitative ELISA. Serum samples from adults and pediatric patients were plotted together since no differences between both populations were observed. Serum samples from controls (n = 42), untreated CD patients (n = 40), CD patients on gluten-free diet (GFD) (n = 9), and intestinal bowel disease (IBD) patients (n = 7) were analyzed. Values are expressed as mean ± standard deviation (s.d.) of the mean. Representative experiments were analyzed statistically using the Mann-Whitney U test. (A) p < 0.0001; (B) p = 0.0002; (C) p = 0.0264.

Figure 2

Assessment of reactivity of the rabbit polyclonal antibodies. Western blot analysis was performed using recombinant purified LFABP (protein loaded: 0.25–2 μg) and IFABP (protein loaded: 0.25–2 μg). Rabbit anti-LFABP and anti-IFABP polyclonal antibodies were incubated at 1/6000 and 1/20000 dilutions, respectively.

Figure 3

LFABP expression in human small intestine. Indirect immunofluorescence analysis using anti-LFABP polyclonal antibodies (Alexa 488, green) and nuclei stained with DAPI (blue). Representative staining in duodenal sections of non-CD control ((a) and (b)) and CD patient at diagnosis (c) (magnification 20x). Confocal fluorescence microscopy using anti-LFABP polyclonal antibodies (Alexa 488, green) and nuclei stained with propidium iodide (red). Representative staining in duodenal sections of non-CD control ((d) and (e)) and CD patient at diagnosis (f) (magnification 63x + 1.7 zoom). Healthy tissue ((a), (b), (d), and (e)) shows LFABP expression in the villi enterocytes. Severe enteropathy ((c) and (f)) shows LFABP expression in the epithelium as well as the crypts closer to the epithelium.

Figure 4

IFABP expression in human small intestine. Indirect immunofluorescence analysis using anti-IFABP polyclonal antibodies (Alexa 488, green) and nuclei stained with DAPI (blue). Representative staining in duodenal sections of non-CD control ((a) and (b)) and CD patient at diagnosis (c) (magnification 20x). Confocal fluorescence microscopy using anti-IFABP polyclonal antibodies (Alexa 488, green) and nuclei stained with propidium iodide (red). Representative staining in duodenal sections of non-CD control ((d) and (e)) and CD patient at diagnosis (f) (magnification 63x + 1.7 zoom). Healthy tissue ((a), (b), (d), and (e)) shows IFABP expression in the villi enterocytes. Severe enteropathy ((c) and (f)) shows IFABP expression in the epithelium as well as the crypts.

Figure 5

mRNA levels of LFABP and IFABP assessed by quantitative PCR using β-actin as housekeeping gene. Quantitative PCR analysis was performed in whole duodenal biopsies from adult and pediatric populations of healthy non-CD controls and CD patients at diagnosis. Pediatric samples: controls (n = 13), CD patients (n = 14). Adult samples: controls (n = 9 for IFABP, n = 8 for LFABP), CD patients (n = 6). Results were plotted as relative unit (RU), using β-actin as housekeeping gene. Values are expressed as mean ± standard deviation (s.d.) of the mean. Representative experiments were analyzed statistically using the Mann-Whitney U test. (A) p = 0.0047; (B) p = 0.0423; (C, D) p < 0.0001.

Figure 6

mRNA levels assessed by quantitative PCR of (a) villin using β-actin as housekeeping gene and (b) LFABP and IFABP using villin as housekeeping gene. Quantitative PCR analysis was performed in whole duodenal biopsies from adult and pediatric populations of healthy non-CD controls and CD patients at diagnosis. mRNA relative levels from adults and pediatric patients were plotted together since no differences between both populations were observed. (a) mRNA levels of villin using β-actin as housekeeping gene. Pediatric samples: controls (n = 10), CD patients (n = 9). Adult samples: controls (n = 8), CD patients (n = 6). (b) mRNA levels of LFABP and IFABP using villin as housekeeping gene. Pediatric samples: controls (n = 12), CD patients (n = 13). Adult samples: controls (n = 8), CD patients (n = 6). Results were plotted as relative unit (RU). Values are expressed as mean ± standard deviation (s.d.) of the mean. Representative experiments were analyzed statistically using the Mann-Whitney U test. (A) p = 0.0374; (B) p = 0.0294.