Expression of corticotropin releasing factor receptor type 1 (CRF1) in the human gastrointestinal tract and upregulation in the colonic mucosa in patients with ulcerative colitis.

Activation of corticotropin releasing factor (CRF) signaling in the brain plays a key role in orchestrating the endocrine, behavioral and gastrointestinal (GI) responses to stress [ 3 , 39 ]. Convergent evidence indicates that CRF is also expressed in peripheral tissues where the peptide acts locally as a proinflammatory mediator [ 2 , 7 , 16 , 27 , 44 ]. In rodents, the increased expression of CRF has been found in inflamed tissues during carrageenin-induced granulomas [ 16 ], inflammatory arthritis [ 7 ], experimental autoimmune uveoretinitis [ 27 ], chronic granulomatous enterocolitis [ 44 ] and Clostridium difficile toxin A-mediated ileitis [ 2 , 20 ]. Our previous study also showed that CRF is upregulated at both mRNA and protein levels by peripheral injection of low dose of endotoxin in the rat colon [ 47 ]. Conversely, blockade of CRF signaling pathways by CRF immunoneutralization using anti-CRF antibody, pharmacological blockade of CRF receptors with antagonists or genetic approaches using CRF deficient mice or CRF RNA interference, dampened the local inflammatory response triggered in these tissues including in the intestine [ 2 , 10 , 16 , 20 , 27 , 44 ]. In humans, likewise, peripheral CRF is up-regulated at sites of rheumatoid arthritis [ 6 ], Hashimoto thyroiditis [ 36 ], endometriosis [ 18 ], psoriasis [ 19 ] as well as in the colon of patients with ulcerative colitis (UC) [ 17 ]. The biological actions of CRF and related peptides, urocortin (Ucn) 1, Ucn 2 and Ucn 3 are mediated via binding to two distinct receptor types, CRF 1 and/or CRF 2 , both belonging to the class B of G-protein coupled receptor superfamily [ 12 , 22 ]. CRF 1 has high affinity for CRF or Ucn 1 and no appreciable binding affinity for Ucn 2 and Ucn 3. In contrast, CRF 2 displays high affinity for Ucn 1, Ucn 2 and Ucn 3 and low affinity for CRF [ 12 ]. Our functional studies in rodents indicate that the colonic secretory motor alterations and visceral hyperalgesia induced by peripheral administration of CRF and Ucn 1 are mediated by peripheral CRF 1 receptor [ 21 , 26 , 35 , 41 ]. This is further supported by anatomical evidence that CRF 1 receptors are expressed in colonic myenteric neurons, immune and endocrine cells in experimental animals [ 4 , 23 , 47 ]. By contrast, gene expression of CRF 1 receptor in the human whole GI tract has been little investigated [ 31 , 43 ] particularly in the human intestinal mucosa which contains the majority of cells involved in immune reactions including macrophages, mast cells, granulocytes, T and B lymphocytes, and dentritic cells [ 14 ]. Among the immune cells, macrophages play a key role in orchestrating mucosal inflammatory responses [ 37 ]. Gaining insight to CRF 1 receptor expression in the human GI tract, and in particular the cellular location in the colon and possibly altered expression under the conditions associated with inflammatory bowel disease (IBD) would provide anatomical support for potential sites of CRF/Ucn 1 action in clinical setting. This will be particularly relevant in the context of UC based on clinical reports that CRF, a preferential CRF 1 endogenous agonist [ 12 ], and Ucn 1 are upregulated in the colonic mucosal of UC patients [ 17 , 34 ]. Therefore in the present study, we first mapped the CRF 1 mRNA distribution in the different segments of GI tract in healthy human subjects using reverse transcription polymerase chain reaction (RT-PCR). Then we assessed the expression and localization of CRF 1 at the protein level by immunohistochemistry on whole thickness sigmoid colonic sections and in sigmoid colonic mucosal biopsies of healthy subjects. Lastly, in sigmoid colonic biopsies of healthy control and UC patients, we compared the numbers of cells immunostained with CRF 1 antibody and double labeled with CRF 1 and CD163, a marker of resident macrophages [ 8 ].

Total RNA samples extracted from whole thickness GI segments including the esophagus, gastric fundus, pylorus, duodenum, jejunum, ileum, cecum, ascending colon, descending colon, transverse colon, sigmoid colon and rectum and paraffin tissue slides of sigmoid colon were purchased from BioChain Institute, Inc. (Hayward, CA, USA). These tissues were collected from 4 male healthy subject donors (24-29 years) who died from sudden death due to accidents. Tissue collection was done at the time of death with a post mortem delay between 4-6 h (BioChain Institute, Inc). These specimens were used to perform the mapping of CRF 1 mRNA distribution along the GI tract and cellular localization of CRF 1 receptor in the sigmoid colon. Approval for the use of human tissue samples and all procedures to process them was obtained from the Committee for Medical Research Ethics (VA Greater Los Angeles Health System, VA project#: 0016). All sigmoid colonic mucosal biopsy specimens were obtained from the Mucosal Immunology Core (UCLA AIDS Institute Center for AIDS Research). Approval to conduct the study was obtained from the UCLA Human Subjects Protection Committee. All participants provided written consent. Sigmoid colonic biopsies were collected by flexible sigmoidoscopy, 10-20 cm from the anal verge, from 6 healthy controls (33-65 yrs; 1 male, 5 females) and 10 UC patients (32-83 yrs; 6 males and 4 females). Patients had UC for >10 years, based on history, endoscopic and pathology reports over the years. Four UC patients had active clinical disease presentation and met the criteria for mild-moderate active disease (graded as 2-3) and the 6 UC patients met criteria for minimal to no active inflammation with markers of quiescent disease (graded as 1-2). Grades were based on Matts’ UC classification [ 28 ] as used in other studies [ 34 ]. No participants were taking steroids and one was taking a low dose immunosuppressive (6-mercaptopurine at 25mg/day). All were treated with 5-aminosalicylate except one patient who was taking only omega-3 fatty acid gel caps. cDNA was synthesized using total RNA extracted from whole thickness segments of healthy subject GI tract (BioChain Institute, Inc.) in a total volume of 20 μl reaction by ThermoScript™ RT-PCR system (Invitrogen, CA) as reported previously [ 46 ]. Briefly, total RNA (5 μg) was denatured at 65°C for 5 min and then followed by the procedures of the reverse transcription which was stopped by incubation at 85°C for 5 min. Then the reaction mixture was incubated with RNase H at 37 °C for 30 min to remove total RNA template. The sequences of oligonucleotide primers specific for human CRF 1 used in RT-PCR are as follows. Forward primer: 5′-ACAAACAATGGCTACCGGGA-3′, reverse primer: 5’-TCGCAGGCACCGGATGCTC-3′, which cover human CRF 1 Exon 4 and Exon 5, thus amplify the functional wild type form of CRF 1a. PCR reactions were performed in a final volume of 30 μl using RedTaq System (Sigma-Aldrich, Saint Louis, MO). The reaction was pre-denatured at 95°C for 2 min, and then amplified 34 times (94°C, 40 sec; 59°C, 40 sec; 72°C, 2 min), followed by a 5 min extension at 72°C in Thermal Cycler (MyCycler™ Thermal Cycler, BIO-RAD Laboratories, Hercules, CA). The house keeping gene, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control to assure cDNA quality and equal loading. The sequences of the primers for human GAPDH are as follows. Forward primer: 5′-GGTCGGAGTCAACGGATTTG -3′, reverse primer: 5′-ATGAGGTCCACCACCCTGTT -3′. Negative control contained all reagents, except that 1 μl H 2 O was substituted for reverse transcriptase in RT reaction to exclude the possibility of genomic or other DNA contamination. PCR products were separated by 1% agarose gel electrophoresis and visualized with ethidium bromide. Gel images were acquired by Kodak EDAS 290 system and the band density from tissues was quantified by NIH Image system (Scion Corporation, Frederick, MD) and standardized by taking the ratio to that of GAPDH in each sample respectively. Paraffin sections (5 μ m) of whole thickness sigmoid colon from healthy male subjects (BioChain Institute Inc., n=3) and sigmoid colonic biopsies mostly containing the mucosa from healthy controls (n=6) and patients with UC (n=10) were processed for immunohistochemistry. Tissue sections were deparaffinized in xylene and hydrated in descending grades of ethanol. After washing 2 times (5 min each) in phosphate buffered saline (PBS), slides were placed in a plastic Coplin jar filled with 10 mM citrate buffer (pH 6.0), boiled for 8 min, and followed by cooling to room temperature. Endogenous peroxidase activity was blocked by incubation for 30 min with 0.3% hydrogen peroxide in PBS and nonspecific reaction was blocked by incubation in 3% normal donkey serum for 30 min at room temperature. Slides were incubated overnight at 4°C with goat anti-CRF 1 antibody (1:500; C-20, Santa Cruz Biotechnology, CA, USA) diluted in PBS containing 0.3% Triton X-100, followed by incubation with biotinylated donkey anti-goat IgG (1:1000; Jackson ImmunoResearch, West Grove, PA) for 2 h at room temperature. Sections were subsequently processed for avidin-biotin-peroxidase procedure using diaminobenzidine as a chromogen, and then counterstained with hematoxylin. Sections of human cerebral cortex (BioChain Institute, Inc.), well known to express CRF 1 [ 13 ] as in rodents [ 42 ], were taken as a positive control. Immunohistochemical negative control was routinely performed following the same procedures, except that the primary antibody was replaced by PBS. Paraffin sections (5 μm) of sigmoid colonic biopsies from healthy controls (n=6) and patients with UC (n=10) were deparaffinized in xylene and hydrated in descending grades of ethanol, followed by an antigen retrieval procedure as described above, tissue slides from biopsy specimens were washed three times at 10-min intervals with PBS, then were incubated in 10% normal donkey serum (Jackson ImmunoResearch Laboratories) in 0.3% Triton-X 100/PBS for 30 min at room temperature followed by 48-h incubation with the mixture of goat anti-CRF 1 antibody (1:50, C-20, Santa Cruz Biotechnology) and mouse anti-rat/human CD163 (1:200, Serotec, Inc. Raleigh, NC, USA), a marker of macrophages [ 8 ]. Sections were washed with PBS three times at 10-min intervals and incubated for 2 h at room temperature with the mixture of Rhodamine Red™-X-conjugated donkey anti-goat IgG and FITC-conjugated donkey anti-mouse IgG (1:100. Jackson ImmunoResearch Laboratories). After washing three times in PBS, sections were subsequently mounted on slides with anti-fade mounting media (Vector Laboratory Inc., Burlingame, CA) and visualized by standard fluorescence microscopy. The number of CRF 1 immunoreactive (IR) and CRF 1 /CD163 double labelled cells in the lamina propria of colonic biopsies were counted and quantified in an average number of 5 fields (340 μm×260 μm/field) from each specimen in a blinded fashion such that the information on the clinical, endoscopic and pathological findings were unknown until all counting was completed. All data are expressed as mean ± SE. Comparison of CRF 1 mRNA levels among GI segments was performed with one-way ANOVA followed by Duncan's test. The numbers of CRF 1, and CRF 1 /CD163 double-labeled cells in the lamina propria of the sigmoid biopsies of healthy subjects and UC patients were compared and analyzed by Student t -test. A P value <0.05 was considered statistically significant.

CRF 1 mRNA expression was detected in all GI segments examined from the esophagus to the rectum and the distribution pattern was consistent among 4 healthy subjects ( Fig.1A ). CRF 1 mRNA was expressed predominantly in the small intestine and the rectum with the highest levels found in the ileum and rectum followed by the esophagus, gastric fundus, duodenum, jejunum, cecum, with lower levels in the pylorus, and colon (ascending, transverse, descending, sigmoid). In small intestinal segments CRF 1 mRNA expression is linearly increased from the duodenum to the ileum. In colonic segments, however, CRF 1 mRNA levels in the ascending, transverse, descending and sigmoid colon were 31.4%, 41.8%, 31.2%, and 31.7% respectively significantly lower than in the ileum ( Fig. 1B ). In the whole thickness tissue sections of the sigmoid colon from healthy subjects, CRF 1 immunoreactivity was located in cells scattered within the lamina propria ( Figs. 2A , 3A ), submucosal ( Fig. 3C ) and myenteric plexus ( Fig. 2A , 3E ). A weak staining was also found at the base of the absorptive surface epithelial cells ( Fig. 2A ). In biopsy sections of sigmoid colon from healthy subjects, CRF 1 immunoreactivity was mainly located in cells of the lamina propria underneath the epithelia ( Fig. 4A ). In sections of human cerebral cortex taken as a positive control [ 13 ], CRF 1 immunoreactivity was located in neurons ( Fig. 3G ). The immunostaining was not observed when the CRF 1 antibody was replaced by PBS ( Figs. 2B , 3B, 3D, 3F, 3H , and 4B ). To characterize CRF 1 immunoreactive (IR) positive cells in the lamina propria, double staining of CRF 1 and CD163 using a marker antibody for macrophages, was conducted in tissue sections of sigmoid colonic biopsies from healthy and UC subjects. All cells labeled by monoclonal antibody against CD163 also showed CRF 1 immunoreactivity. CRF 1 /CD163 double-labeled cells represented 79% of total CRF 1 IR positive cells in the lamina propria while only 21% of CRF 1 IR positive cells was not labeled by CD163 indicating that CRF 1 receptor is mainly located in the colonic lamina propria macrophages ( Fig. 5A, B, C, G ). The number of CRF 1 IR positive cells in the mucosal lamina propria was significantly increased in sigmoid biopsies of UC patients (30±8 vs. 7±3/field in healthy subjects, n=10 and 6 respectively p<0.01). The increase occurred irrespectively of these chronic UC patients being in remission with minimal inflammation (31±10/field n=6) or active with mild to moderate inflammation (29±5 /field, n= 4). Similarly, CRF 1 /CD163 double-labeled cells (macrophages) in the mucosa were also significantly elevated in the UC patients (24±6 vs 6±2/field in healthy subjects, n=10 and 6 respectively, p<0.01) ( Fig. 5D, E, F, G ) with similar increase in UC in remission and active (25±6, n=6 and 23±5/field, n=4, respectively).

In the present study we provided the first comprehensive description of CRF 1 gene expression in the different segments throughout the GI tract of healthy human subjects based on levels of CRF 1 mRNA. There was a very consistent pattern in CRF 1 mRNA expression in the GI tract among the 4 subjects investigated with the highest levels found in the ileum and rectum, followed by the esophagus, gastric fundus, duodenum, jejunum, cecum, and lesser levels in the pylorus, and colon (ascending, transverse, descending and sigmoid). A previous study detected only CRF 2 mRNA in the whole stomach of healthy human subjects but not CRF 1 [ 5 ] which may be related to the difference in the primer efficiency between the two studies. In small intestinal segments, CRF 1 mRNA expression is linearly increased from the duodenum to the ileum while in the large bowel segments, it is decreased from the cecum to the colon and then abruptly elevated in the rectum. Such a CRF 1 transcriptional expression pattern in small intestine in humans differs from that reported in the rat which has a greater expression in the duodenum than in the ileum [ 33 ]. Previous functional studies in healthy human subjects showed that systemic injection of the preferential CRF 1 agonist, CRF [ 12 ] stimulates pyloric and duodenal pressure waves switching postprandial duodenal motor activity to non-propagated high frequency contractions [ 29 , 40 ]. The present demonstration of CRF 1 expression provides anatomic support for a local CRF 1 mediated action within the human gut wall to influence GI motor function. Immunohistochemical cellular detection of CRF 1 in whole thickness tissue sections of the sigmoid colon from healthy subjects showed CRF 1 immunoreactivity with strong intensity in submucosal and myenteric neurons. The specificity of CRF 1 antibody to detect human CRF 1 receptor was characterized in our previous study by Western blot and immunofluorescence in HEK-293 cells transected with human CRF 1a and in human BON-1B cells [ 45 ]. These data extend to humans reports so far only in experimental animals demonstrating CRF 1 expression at gene and protein levels within colonic myenteric neurons [ 4 , 11 , 30 , 32 , 46 ]. Functional studies in healthy subjects showing that systemic injection of CRF induces segmental contractions in the descending and sigmoid colon [ 9 ] support this contention. In experimental animals, CRF and selective CRF 1 agonist exert a direct CRF 1 mediated excitatory action on colonic myenteric neurons to stimulate propulsive colonic motility [ 24 , 25 , 46 ]. In addition to neurons in the enteric plexi, CRF 1 immunoreactivity in humans was detected in cells scattered in the lamina propria with a similar staining pattern regardless of whether the samples were whole thickness colonic tissues (post mortem) or freshly-acquired colonic biopsies. These data are consistent with previous studies showing CRF 1 mRNA expression in lamina propria cells isolated from the mucosa-submucosal layer of normal human colorectal tissues [ 31 ] and CRF 1 immunoreactivity in the lamina propria of human sigmoid biopsies [ 43 ]. In addition, we found that all of CD163 labeled cells were CRF 1 IR and 79% of total CRF 1 IR cells in the lamina propria were double-labeled by CRF 1 /CD163, providing the first evidence that CRF 1 receptor is prominently located in human colonic lamina propria macrophages. The GI mucosa contains the largest reservoir of resident macrophages in the body and intestinal macrophages are the first phagocytic cells of the innate immune system to interact with microorganisms and microbial products that have breached the epithelium [ 37 ]. Located in the subepithelial lamina propria, mucosal macrophages have a dual role of protecting the host against foreign pathogens and regulating mucosal responses to commensal bacteria by the production of proinflammatory cytokines during inflammation [ 37 , 38 ]. The expression of CRF 1 on macrophages in the lamina propria provides anatomical evidence that macrophages may represent an immune target for CRF 1 cognate agonists, CRF and Ucn 1 also located in macrophage of the human colon [ 31 ]. Besides macrophages, there is a remaining 21% of CRF 1 IR positive cells which were not labeled by CD163 in the lamina propria. A previous report showed that CRF 1 IR cells were double labeled with mast cell tryptase in sigmoid biopsy of healthy subjects [ 43 ]. This may account for the other cell components in the lamina propria also expressing CRF 1 . Related to the immunoinflammatory influences of CRF 1 and the tissue macrophage immunocyte as well, the modulation of mucosal macrophages (and associated CRF 1 expression) might be significantly altered in certain pathological conditions [ 15 ]. In the present study, the quantification of CRF 1 IR positive cells and CRF 1 /CD163 double labeled macrophages in the distal colonic (sigmoid) mucosa lamina propia showed that CRF 1 IR positive cells were significantly increased by 4.2 fold in non glucocorticoid treated UC patients compared to healthy subjects. Likewise, a previous report indicates the accumulation of CRF 1 mRNA hybridization signal in colonic lamina propria cells of non glucocorticoid-treated UC patients although no quantification was performed comparative to control subjects [ 34 ]. Given the small number of patients investigated in this pilot study, we did not find difference between the UC samples from those patients with a moderate active inflammation or in a clinical state of remission. It is well known that in UC, a variety of inflammatory cells infiltrate the colon but an increase in the population of activated macrophages has been demonstrated to be a central characteristic of the inflammatory process involved in the disease [ 14 ]. Thus, the increased CRF 1 IR positive cells most likely originate from the increased resident population of macrophages in UC patients (based on the near equivalency in active and remission disease). This was ascertained by our double immunostaining showing that all CD163 labeled cells are CRF 1 IR positive and CRF 1 /CD163 double-labeled cells (macrophages) represent 79% of total CRF 1 IR positive cells. CRF 1 /CD163 double-labeled macrophages were also significantly elevated in the UC patients by 4-fold compared to tissues of healthy controls. In addition other reports showed that CRF and Ucn 1 in the colonic lamina propria inflammatory (plasma) cells are upregulated in UC patients [ 17 , 34 ]. Taken together, these data support the hypothesis of autocrine/paracrine mechanisms whereby CRF/Ucn 1 act on CRF 1 expressing macrophages of the lamina propria in UC patients which may have a bearing with the pro-inflammatory state. Indeed, there is in vitro evidence that CRF exerts a CRF 1 mediated enhancing effect on endotoxin-induced cytokine production on primary culture of mouse macrophages or macrophage cell line [ 1 , 16 ]. In summary, the present study delineated a widespread and regionally different pattern of CRF 1 mRNA expression along the GI tract of healthy subjects with a prominent expression in the ileum and rectum. As established in experimental animals [ 4 , 23 , 46 ], CRF 1 IR was present in submucosal and myenteric neurons of the human colon, supporting that CRF/Ucn 1 can exert a direct action on colonic enteric nervous system. In addition, all macrophages in sigmoid colonic biopsies express CRF 1 representing 79% of total CRF 1 IR positive cells in the colonic lamina propria. Significant to the UC pathophysiology, total CRF 1 IR cells and CRF 1 IR present on macrophages in the mucosal lamina propria were significantly increased by 4.2 and 4.0 folds in UC patients respectively. This up regulation was maintained independent of UC being active or in remission. The enhanced CRF 1 expression observed in the healthy human ileum raises the testable and consistent hypothesis that ileal samples from Crohn's disease patients may also show a similar alteration in total CRF 1 IR cells and CRF 1 IR located on macrophages. The present results along with reports that CRF and Ucn 1 are upregulated in the lamina propria cells of colonic mucosa in UC patients [ 17 , 34 ], are indicative that enhanced CRF/Ucn1/CRF 1 signaling might be an important process in the mediation and/or exacerbation of inflammatory process in UC and may provide a new therapeutic target for UC.