M cells--entryways of opportunity for enteropathogens

p eyer's patches have been thought of as areas of weakness in the intestinal wall since they were first noted in hog intestine used as casings for sausages (1). Butchers noted little bulging areas of decreased tensile strength in some regions of the casings, which were attributed to hogs swallowing too much air. These intestinal "glands" were mentioned by Hippocrates, and in 1677, Johann Conrad Peyer found that they corresponded to lymphoid nodules, interrupting the muscularis mucosae, which are now known as Peyer's patches. Clinically, Peyer's patches are also notable as areas of weakness in the mucosal barrier, especially as sites of intestinal penetration by Salmonella typhi, which is the etiologic agent of typhoid fever. In 1829, Louis described intestinal perforation as a primary complication of typhoid fever. He reported that perforation was usually single and located just above the cecum, in the center of an ulcer on the antimesenteric side of the ileum, which corresponds to the location of the major human Peyer's patch (2). In this issue of The Journal of Experimental Medicine, Jones et al. (3) provide new information regarding the pathogenic mechanism by which Salmonella spp. penetrate Peyer's patches and create micro-ulcerations, that allow intestinal flora to enter the epithelium and lamina propria of the ileum. The pathology of typhoid infection has long held clues pointing toward a major pathway by which intestinal infectious agents can produce systemic illness. After infection with S. typhi, Peyer's patches in the terminal ileum typically develop hyperplasia after the first week, which can either resolve or progress to necrosis in the second week. Ulcers can lead to perforation and hemorrhage, usually in the third week. Although septicemia and constitutional symptoms produce considerable morbidity, perforation has been and remains a major cause of mortality in typhoid. A review of mortality in typhoid patients with intestinal perforation, between 1937 and 1983, identified 2,112 patients treated surgically with 28% mortality and 409 patients treated medically with 70% mortality (4). Although typhoid has been largely eliminated in the industrialized world, it remains a major problem in third world countries. From 1975 to 1983 at the International Centre for Diarrhoeal Disease Research Hospital B in Dhaka, Bangladesh, 323 patients with typhoid were identified, of whom 15 (4.6%) developed intestinal perforation (4). A review of 57,864 cases of typhoid in the third world during the antibiotic era found 1,448 (2.5%) with perforation, which accounted for 25% of all typhoid deaths (5). A histopathological study of 184 cases of intestinal perforation due to typhoid in Bandung, Indonesia, from 1972 to 1976, found 142 cases (77.2%) in males and 42 (22.8%) in females, mostly in young adults and uncommonly in children. Pathological change began as edema and hyperplasia of Peyer's patch lymphoid elements, with stretching of the overlying mucosa, surface erosions, and later necrosis of lymphoid tissue (6). Interestingly, isolation of S. typhi from the peritoneal cavity is rare in cases of perforation. Instead there is usually polymicrobial infection with aerobic gram-positive cocci, aerobic gram-negative bacilli, and also anaerobic bacteria, presumably from "normal" luminal flora leaking across the perforation (4). Immunohistochemical examination of formalin-fixed ileal tissue from 13 patients with typhoid perforations, using rabbit polyclonal antisera to salmonella, failed to show any residual salmonella infection at the time of perforation (Mukawi, T. J., D. E Keren, and R. L. Owen, unpublished data). The pathophysiological basis for this remarkable association of perforation with Peyer's patches in the terminal ileum has long remained a source of puzzlement and speculation. Twenty years ago, Peyer's patches, tonsils, and the appendix were considered vestigial organs without known function, important only as possible sites of suppuration. New information regarding mechanisms of bacterial attachment and entry into host cells, and advances in understanding the structure and function of mucosal components of the intestinal immune system, have explained how salmonella attach to enterocytes and why uptake takes place in Peyer's patches (7). Kumagai (8) had noted in 1922 that Mycobacterium tuberculosis fed to rabbits were taken up by lymphoid follicles in Peyer's patches and in appendix, and the same process, was seen with powdered red blood ceils, suggesting that it was a general phenomenon and not a specific mechanism of bacterial invasion. Peyer's patches in human ileum, the avian bursa of Fahricius, and the rabbit appendix were subsequently found to contain epithelial cells over lymphoid follicles, which are specially adapted for uptake of microbes and other particles from the intestinal lumen (9, 10). These cells, termed M cells, initially because of "microfolds" over their luminal surfaces in the human terminal ileum, are now thought of as "membranous", because lymphoid cells invaginating their basolateral surfaces compress them into attenuated bands, which like membranes, separate the intestinal lumen from tissue compartments, yet let particles and molecules cross in a controlled manner. M cells rest on the basal lamina and adhere to adja-