Altered Structural Expression and Enzymatic Activity Parameters in Quiescent Ulcerative Colitis: Are These Potential Normalization Criteria?

Abstract

Mucosal healing determined by endoscopy is currently the remission standard for ulcerative colitis (UC). However, new criteria for remission are emerging, such as histologic normalization, which appears to correlate better to the risk of relapse. Here, we study mucosal healing on a molecular and functional level in quiescent UC. We obtained endoscopic biopsies from 33 quiescent UC patients and from 17 controls. Histology was assessed using Geboes score. Protein and mRNA levels were evaluated for the tight junction proteins claudin-2, claudin-4, occludin, and tricellulin, as well as Cl^-/HCO₃^- exchanger DRA, and cyclo-oxygenase enzymes (COX-1, COX-2). The mucosal activity of COX-1 and COX-2 enzymes was assessed in modified Ussing chambers, measuring electrogenic ion transport (short-circuit current, SCC). Chronic inflammation was present in most UC patients. The protein level of claudin-4 was reduced, while mRNA-levels of claudin-2 and claudin-4 were upregulated in UC patients. Surprisingly, the mRNA level of COX-1 was downregulated, but was unaltered for COX-2. Basal ion transport was not affected, while COX-2 inhibition induced a two-fold larger decrease in SCC in UC patients. Despite being in clinical and endoscopic remission, quiescent UC patients demonstrated abnormal mucosal barrier properties at the molecular and functional level. Further exploration of mucosal molecular signature for revision of current remission standards should be considered.

Keywords: COX-1; COX-2; Inflammatory bowel disease; PGE2; mucosal barrier integrity; mucosal healing; short-circuit current; tight junction; ulcerative colitis.

Figure 1

Patient recruitment flowchart.

Figure 2

Histological features in quiescent ulcerative colitis (UC). H&E staining of colonic mucosal biopsies (×200 magnification). Two UC patients, both in clinical and endoscopic remission but with different histologies: (A) No inflammation, (B) chronic inflammation. Blue arrow: increased chronic inflammatory infiltrate with lymphocytes. Green arrow: Paneth cell metaplasia. Red arrows: crypt irregularity.

Figure 3

Western blot and RT-qPCR. (A) Expression levels of barrier related proteins by densitometric analysis of Western blot (WB). Values are expressed as % of a mean of all controls (ctrls). Bars indicate mean ± standard error of the mean. Asterisk indicates statistically significant difference between groups (* p < 0.05). (B) mRNA levels by RT-qPCR. Values are expressed relative to a mean of all controls. Three controls (18%, 3/17) presented a different baseline of housekeeping genes, while two controls (12%, 2/17) presented outliers, which were removed, leaving the final number of controls for mRNA analysis at 12 (70% 12/17). One outlier (3%, 1/33) was removed from the UC group. n = number of UC observations: COX-1 (n = 16), COX-2 (n = 11), claudin-2 (n = 23), claudin-4 (n = 33), occludin (n = 31), tricellulin (n = 29), and DRA (n = 33). Bars indicate mean ± standard error of the mean. Asterisks indicate statistically significant difference between two groups (* p < 0.05, ** p < 0.01).

Figure 4

The localization of downregulated-in-adenoma (DRA), occludin and claudin-4. Representative confocal images of human colonic biopsies from control, CTRL, (n = 11) and UC patients (n = 11) stained for (A) DRA, (B) occludin and (C) claudin-4. Occludin and claudin-4 display intracellular accumulation in the surface epithelium of a subset of UC patients (white arrows). Stains for Na⁺/K⁺-ATPase or beta-catenin were included to mark the lateral membranes and the nuclei were visualized with DAPI. The localization depicted for UC patients was observed in the indicated subset of biopsies (2/11, 1/11, respectively). The upper panels in A–C show low magnification images and below high magnification images, while the lower panels show high magnification images of the surface epithelium. Scale bars: upper panel in A–C: 50 µm, lower panels in A–C: 20 µm.

Figure 5

Prostaglandin E2 (PGE₂)-dependent chloride (Cl⁻) secretion in a colonocyte. Cyclooxygenase enzyme 1 (COX-1) and -2 (COX-2) produce PGE₂ from activated arachidonic acid (AAA). PGE₂ leaves the cell through the basolateral membrane and exerts its function either by auto- or paracrine stimulation. Binding to its receptor (EP-receptor) stimulates the synthesis of second messenger cyclic adenosine monophosphate (cAMP) from adenosine triphosphate (ATP) by adenylate cyclase (AC). cAMP in turn stimulates luminal Cl⁻ secretion and is degraded to adenosine monophosphate (AMP) by phosphodiesterase-enzymes (PDE). SC-560 and rofecoxib specifically inhibits COX-1 and COX-2, while indomethacin is a non-specific COX-inhibitor. Theophylline is a non-specific phosphodiesterase inhibitor. Amiloride inhibits Epithelial Sodium Channel (ENaC)-mediated luminal sodium absorption. Chloride influx through basolateral Na⁺/K⁺/2Cl⁻-cotransporter and efflux through apical chloride channel cystic fibrosis transmembrane conductance regulator (CFTR). PGE₂ synthesis also occurs in the subepithelium. + and – indicate stimulatory and inhibitory effects, respectively.

Figure 6

Examples of short-circuit current (SCC) data. Measurements in biopsies from an UC patient (A), and control (B). Biopsies mounted in the mini-Ussing-air-suction (MUAS) chambers were exposed to: amiloride (sodium absorption inhibitor, 20 µM), theophylline (non-specific phosphodiesterase inhibitor, 400 µM), a specific COX inhibitor (either COX-1, SC-560, or COX-2, rofecoxib, 500 nM) followed by indomethacin (non-specific COX inhibitor, 13 µM), prostaglandin E2 (PGE₂) in increasing concentrations (five-step, factor five from 5 to 3125 nM), and ultimately either bumetanide (Na⁺/K⁺/2Cl⁻-cotransporter inhibitor, 25 µM) or ouabain (Na⁺/K⁺-ATPase inhibitor, 200 µM).

Figure 7

Mucosal function, expressed as a decrease in short-circuit current (SCC) after COX-2 enzyme inhibition with rofecoxib, correlated with endoscopy and histology for patient groups. Endoscopy (left): comparing functional data for controls (n = 11) to UC patients in clinical and endoscopic remission (n = 28). Histology (right): comparing functional data for UC patients with (n = 19) and without (n = 9) chronic inflammation. A larger decrease in SCC indicates elevated levels of PGE₂. Values are expressed as mean ± SEM. Asterisk indicates statistically significant difference between two groups (* p = 0.01).