Portals for Prions?

Prion disorders riddle the mammalian brain with plaque and holes, the precise pattern and resulting symptoms--dementia, extreme fatigue, or loss of balance--depending on whether one is human, bovine, or ovine. The agent of such a transmissible spongiform encephalopathy (TSE) is an infectious form of prion protein, called PrP scrapie (PrPSc), named after the long-known sheep illness. In bovine spongiform encephalopathy (mad cow disease) and its human version, variant Creutzfeldt-Jakob disease (vCJD), ingested PrPSc converts the normal "cellular" form of PrP (PrPC) to an abnormal form. Until now, most prion research explored brain tissue--the end point of infection. But a few investigators are seeking the site where prions pathogenesis begins. And their work is converging on a little-mentioned resident of the intestinal lining called an M cell. M cells link the digestive and immune systems.1 They provide a portal for ingested bacteria and viruses to lymphoid tissue beneath the intestinal lining. "M cells sample antigens so that the body can mount an effective protective immune response," explains Marian Neutra, a professor of pediatrics at Harvard Medical School. But the M cell defense isn't foolproof. "Pathogens have evolved faster than us to exploit this antigen sampling mechanism," she adds. Certain bacteria and viruses usurp the M cell connection, and prions may do so too. Their passage to humans from tainted food-beef in the case of vCJD, and human flesh decades ago in kuru among the Foré people in Papau New Guinea2 -place them, theoretically, at this crossroads of digestion and immunity. And now experimental evidence is beginning to support the idea. Discovery of M Cells Courtesy of Robert L. OwenThree-dimensional illustration of M cells. The exceedingly rare M cells peek out among the trillions of epithelial cells (enterocytes) that form the microvillus-fringed hills and valleys of the intestinal lining. They nestle among follicle-associated epithelium (FAE) that overlies scattered collections of lymphoid tissue called Peyer's patches. "M cells make up about 10 to 20 percent of the FAE, which in turn makes up 1 percent or less of the epithelium of the small intestine," says Lars Eckmann, assistant professor of medicine at the University of California, San Diego. Anatomists first described Peyer's patches in 1677. These cell aggregates become visible after death, when the epithelium above them recedes. In addition to their paucity, M cells are difficult to study because fresh material must be examined quickly. An unusual situation led to their initial description in 1973. Robert Owen and Albert Jones, then a postdoctoral researcher in gastroenterology at the Veterans Administration Hospital, San Francisco, and an associate professor of anatomy at the University of California at San Francisco School of Medicine, respectively, obtained sections of intestine from five obese individuals undergoing intestinal bypass surgery.3 "Only if there's an accident do you have a rare opportunity to look at normal human intestine. For these people, the small intestine worked too well," recalls Owen. Under the light microscope, the lining tissue appeared disorganized, but the scanning electron microscope told a different story. "I saw structures not previously recognized in the lowest part of the small intestine," Owen says. "Here, the Peyer's patches looked like little mounds of mashed potato among the fingerlike villi lining the intestine. I saw cells that were bare on the surface, rather than packed with a thicket of microvilli." The cell tops were thrown up into thicker microfolds; hence the "M" in M cell. Owen and Jones noted something even stranger at the M cells' other ends. Each cell dipped inward to form a large pocket that held immune system cells-T and B cells (lymphocytes) and macrophages, plus follicular dendritic cells, which present antigens to lymphocytes for further action. Protection from, or Pathway for, Pathogens? As is often true in biology, the M cell's unique structure makes possible its function. Normally, an M cell engulfs a microbe or particle invading its space, then passes it through the cytoplasm, in a process called transcytosis, the short distance to the pocket's inner edge. Here it is released and the congregating immune system cells gobble it up. This activity stimulates nearby B cell precursors to mature, and they eventually become plasma cells that secrete antibodies into the bloodstream. When the antibodies reach the intestinal lining, they destroy the pathogen. "M cells provide a window for the intestinal immune system to sample what's out there and begin responses to protect the rest of the intestine," explains Owen. The M cell circuit must work fast. In cholera, for example, bacteria bind to enterocytes and cause severe diarrhea before M cells can send in the antibodies. Some bacteria blast through the M cell blockade. "The factors that determine whether microbial transcytosis by M cells results in host invasion or, conversely, in a protective immune response, are unclear," says Ann Clark, a research fellow in the department of physiological sciences at the University of Newcastle upon Tyne in the United Kingdom. Her group showed that Salmonella typhimurium destroys M cells and adjacent FAE.4 "These factors likely contribute to massive bacterial invasion of the host and resulting disease," Clark explains. Yersinia pseudotuberculosis manufactures a protein that prevents M cells from engulfing them, and Shigella spreads via M cells to adjacent enterocytes instead of immune system cells. Certain viruses take advantage of M cells. Poliovirus replicates in Peyer's patches, traversing M cells to the lymphocytes that ferry them to the nervous system. Oral polio vaccine is a disabled virus that uses this route. Virus-M cell interactions occur farther along the gastrointestinal tract too. "M cells pick up HIV in rectal tissue and transfer the virus to lymphocytes," says Owen. A Possible Prion Connection Whether infectious prions are members of that select club of pathogens that uses M cells to invade a host is, at this point, an educated proposition. "Given that a wide variety of reagents are transcytosed across M cells, it seems a reasonable hypothesis that such transport of prions might occur," says Clark. Only the PrPSc forms should stimulate immunity, because PrPC is a normal cell surface protein. But catching the conversion as it happens has so far been elusive, although a few abnormal prions can massively convert PrPC in vitro.5 Paul Brown, senior investigator in the laboratory of central nervous system studies at the National Institute of Neurological Disorders and Stroke, says, "I'm sure nobody has experimental evidence for trafficking of prions. There is no information that deals with specific cells, showing that a prion goes to cell one, then two, then three, and then hits a nerve." But several groups are tantalizingly close to elucidating that pathway. Jean-Pierre Kraehenbuhl, a senior scientist at the Institute of Biochemistry at the University of Lausanne in Switzerland, and Eric Pringault, the head of the laboratory of lymphoepithelial interactions at the Pasteur Institute and their colleagues have developed a model of FAE that consists of lymphocytes from mouse-derived Peyer's patches seeded onto a monolayer of human enterocytes. The lymphocytes send signals that convert the enterocytes into FAE and M cells.6 Because the cells are highly differentiated, the in vitro culture lasts only a week, but can still illuminate M cell function. "We have performed experiments with Adrian Aguzzi at the Institute of Neuropathology at the University of Zurich that indicate that prions use M cells to infect underlying cells," says Kraehenbuhl. In vivo observations support Kraehenbuhl's claim that prions traverse M cells. Michael Beekes, a biochemist at the Robert Koch Institute in Berlin and Patricia McBride, a researcher in the neuropathology unit at the Institute for Animal Health in Edinburgh fed scrapie-tainted brain to hamsters, then probed the intestines with monoclonal antibody to PrPSc. By sacrificing animals at various times post-infection, the researchers traced parts of the prion trajectory.7 Telltale granules appeared first in the FAE, macrophages and follicular dendritic cells of Peyer's patches, and then in the enteric nervous system. PrPSc also apparently appeared in M cells. "Although the latter [FAE cells] were not specifically identified, their location and morphology was consistent with that of M cells," Beekes and McBride reported. A year earlier, Brown's group, along with D. Carleton Gajdusek's colleagues at the National Center for Scientific Research in Gif-sur-Yvette, France, reported that zoo lemurs fed brain tissue from BSE-afflicted cows showed PrPSc throughout the small intestinal epithelium, including the M cells and the lymphocytes that they embrace.8 Neil Mabbott's group in the neuropathology unit at the Institute for Animal Health in Edinburgh is also filling in some of the blanks for the post-M cell fate of PrPSc. They have shown that lack of complement protein C3 impairs the ability of follicular dendritic cells to take up prions.9 "How TSE infectivity reaches follicular dendritic cells from the gut lumen is not known, and it is an area we are actively trying to pursue," Mabbott says. "M cells are likely candidates, due to their location and function." But studying the elusive cells is challenging. Subrata Ghosh, a gastroenterologist at Western General Hospital in Edinburgh, has followed the fate of normal prion protein 10. "Recombinant bovine PrPC appears to be transcytosed by human intestinal cells," Ghosh relates. "We still have to study PrPSc. This is not easy - handling PrPSc requires special containment precautions, and it is quite difficult to work with human M cells. A lot more needs to be done." Despite these published experimental results, the idea that M cells could facilitate prion infection is not widely known. Perhaps this is simply because M cells are not as familiar as neurons or fibroblasts, or because the puzzle remains incomplete: prions seen fleetingly in M cells and their environs do not tell the full story. Researchers have yet to observe the conversion of normal to infectious prion protein and its replication in vivo. Until experiments capture the entire process, the "M" in "M cell" could stand for "mystery." Ricki Lewis (rickilewis@nasw.org) is a contributing editor for The Scientist. References 1. M.R. Neutra et al, "Epithelial M cells: gateways for mucosal infection and immunization," Cell, 86:345-8, 1996. 2. R. Lewis, Chapter 4, "In Pursuit of Prions," in Discovery: Windows on the Life Sciences, Blackwell Science, 2001." 3. R.L. Owen and A.L. Jones, "Epithelial cell specialization within human Peyer's patches: an ultrastructural study of intestinal lymphoid follicles," Gastroenterology, 66:189-203, 1974. 4. M.A. Clark, B.H. Hirst, and M.A. Jepson, "Lectin-mediated mucosal delivery of drugs and microparticles," Advances in drug delivery, 43:207-23, Sept. 30, 2000. 5. G.P. Saborio et al, "Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding," Nature, 411:810-3, June 14, 2001. 6. S. Kerneis et al, "Conversion by Peyer's patch lymphocytes of human enterocytes into M cells that transport bacteria," Science, 277:949-52, 1997. 7. M. Beekes and P.A. McBride, "Early accumulation of pathological PrP in the enteric nervous system and gut-associated lymphoid tissue of hamsters orally infected with scrapie," Neuroscience Letters, 278:181-4, 2000. 8. N. Bons et al., "Natural and experimental oral infection of nonhuman primates by bovine spongiform encephalopathy agents," Proceedings of the National Academy of Sciences, 96:4046-51, 1999. 9. N.A. Mabbott et al, "Temporary depletion of complement component C3 or genetic deficiency of C1q significantly delays onset of scrapie," Nature Medicine, 7:485-7, April 2001. 10. A.N. Shmakov and S. Ghosh, "Prion proteins and the gut: une liason dangereuse?" Gut, 48:443-7, April 2001.

Portals for Prions?

Interested in reading more?

Become a Member of