Subepithelial myofibroblasts express cyclooxygenase-2 in colorectal tubular adenomas.

PURPOSE: Recent data support the hypothesis that the inducible isoform of cyclooxygenase (COX-2) plays a role in the early stages of colonic carcinogenesis and that nonsteroidal anti-inflammatory drugs (NSAIDs) retard the development of colon cancer by modulating COX-2. However, the cell types responsible for producing COX-2 in colorectal adenomas remain a subject of controversy. EXPERIMENTAL DESIGN: COX-2 expression in normal colonic mucosa (n = 50), hyperplastic polyps (n = 43), sporadic adenomas (n = 67), and invasive colonic adenocarcinoma (n = 39) was studied in formalin-fixed and paraffin-embedded tissue sections from endoscopy biopsy and colonic resection specimens. Immunohistochemistry (avidin-biotin complex technique with double immunolabeling) was used to identify the phenotypes of COX-2-producing cells. RESULTS: In colorectal adenomas, increased expression of COX-2 was detected and localized to alpha smooth muscle actin ( proportional, variant SMA)-positive subepithelial stromal cells (myofibroblasts) in the periluminal region of the lamina propria in 63 (94%) of 67 cases. In contrast, in normal colonic mucosa and in hyperplastic polyps with intact epithelium, COX-2 expression was found only in macrophages and endothelial cells. In areas in which the surface epithelium was ulcerated in normal mucosa as well as hyperplastic or neoplastic polyps, COX-2 expression was increased in granulation tissue (and present in macrophages, endothelium, and myofibroblasts). In invasive carcinoma, COX-2 expression in myofibroblasts was limited to the adenomatous portion of the tumor and was detected in 62% of cases (n = 39). In addition, focal expression of COX-2 by malignant epithelial cells was observed in 23% of invasive adenocarcinoma. CONCLUSIONS: These results show that increased COX-2 expression in sporadic adenoma of the colon is common and is localized specifically to subepithelial intestinal myofibroblasts. These findings further support the hypothesis that myofibroblasts are important target cells for NSAID-mediated chemoprevention of colorectal cancer.

Immunohistochemical expression of cyclooxygenase-2 in normal kidneys.

Cyclooxygenase (COX)-2, the recently described inducible form ofcyclooxygenase, has been shown to be responsible for the inflammatory and tumorigenic effects of prostaglandins; hence the development and expanding clinical use of COX-2 selective inhibitors termed super aspirins. These pharmacologic agents block COX-2 without abrogating the desired physiologic roles of the constitutively expressed isoform COX-1. Therefore, they are now used in place of nonselective COX inhibitors in patients who require prolonged use of nonsteroidal anti-inflammatory agents. However, there are sporadic reports of COX-2-related nephrotoxicity, and the mechanism of this adverse reaction is not known. Also, the pattern of in situ expression of COX-2 in the human kidney is not known. We therefore studied the immunohistochemical expression of COX-2 in normal kidneys obtained from 53 consecutive total nephrectomy specimens. COX-2 immunohistochemistry was performed using affinity purified polyclonal murine antibody and avidin-biotin detection method with citrate antigen retrieval. Also, to localize COX-2 expression to specific cell types, double immunolabeling was performed using avidin-biotin (for COX-2 detection) and alkaline phosphatase (for detection of alpha-smooth muscle actin or factor VIII related antigen). In the cortex, COX-2 was found to be constitutively expressed in the endothelial cells of arteries, arterioles, and glomeruli in all 53 kidneys. COX-2 expression was also found in the cortical thick ascending limb of the loop of Henle (medullary rays and macula densa) in 50 of 53 cases. In the medulla, COX-2 expression was detected in the endothelial lining of the vasa recta in 52 cases and in the collecting ducts in 5 cases. These data show significant constitutive expression of COX-2 in normal kidney and underscore the need for caution in the use of COX-2 selective inhibitors, especially on a long-term basis.

Effect of tissue fixatives on the immunohistochemical expression of ABH blood group isoantigens.

Immunohistochemical analysis of ABH blood group isoantigens has been shown to be a useful ancillary technique for resolving problems associated with specimen mix-ups in the daily practice of surgical pathology. However, the effects of different fixatives on the expression of these antigens in paraffin-embedded tissues are not known. Therefore, the effects of seven different fixatives on the immunohistochemical expression of ABH blood group isoantigens were studied in tissues from several organs. The following fixatives were used: acetone, 70% ethanol, B5, Bouin, Carnoy, methanol, and 10% formalin. After fixation for 6, 12, and 72 hours, the tissue blocks were embedded in paraffin, and immunohistochemistry was performed on 4 microm-thick tissue sections using monoclonal antibodies to blood group isoantigens (A, B, and H) and the avidin-biotin detection method. Also, immunostaining was performed on step tissue sections with and without antigen retrieval using citrate buffer at pH 6.0. The expression of the blood group isoantigens was concordant with the blood group of the patient in all the cases studied, irrespective of the fixative and time of fixation. However, in the absence of antigen retrieval, the intensity of the staining reaction was diminished. These results showed that irrespective of the fixative used, immunohistochemical staining of paraffin-embedded tissue sections with ABH blood group antibodies is a rapid, reliable, and cost-effective method for sorting out interpretative problems of tissue contaminants (floaters) and specimen mix-ups in surgical pathology.