Editorials

Do dietary lectins cause disease?

The evidence is suggestiveand raises interesting possibilities for treatment 

In 1988 a hospital launched a "healthy eating day" in its staff canteen at lunchtime. One dish contained red kidney beans, and 31 portions were served. At 3 pm one of the customers, a surgical registrar, vomited in theatre. Over the next four hours 10 more customers suffered profuse vomiting, some with diarrhoea. All had recovered by next day. No pathogens were isolated from the food, but the beans contained an abnormally high concentration of the lectin phytohaemagglutinin.¹ Lectins are carbohydrate binding proteins present in most plants, especially seeds and tubers like cereals, potatoes, and beans. Until recently their main use was as histology and blood transfusion reagents, but in the past two decades we have realised that many lectins are (a) toxic, inflammatory, or both; (b) resistant to cooking and digestive enzymes; and (c) present in much of our food.² It is thus no surprise that they sometimes cause "food poisoning." But the really disturbing finding came with the discovery in 1989 that some food lectins get past the gut wall and deposit themselves in distant organs. ^{3 4} So do they cause real life diseases?

This is no academic question because diet is one part of the environment that is manipulable and because lectins have excellent antidotes, at least in vitro. Because of their precise carbohydrate specificities, lectins can be blocked by simple sugars and oligosaccharides. Wheat lectin, for example, is blocked by the sugar N-acetyl glucosamine and its polymers.⁵ These natural compounds are potentially exploitable as drugs should lectin induced diseases be identified.

Wheat gliadin, which causes coeliac disease, contains a lectin like substance that binds to human intestinal mucosa,⁶ and this has been debated as the "coeliac disease toxin" for over 20 years.⁷ But coeliac disease is already managed by gluten avoidance, so nothing would change were the lectin hypothesis proved. On the other hand, wheat lectin also binds to glomerular capillary walls, mesangial cells, and tubules of human kidney and (in rodents) binds IgA and induces IgA mesangial deposits. This suggests that in humans IgA nephropathy might be caused or aggravated by wheat lectin; indeed a trial of gluten avoidance in children with this disease reported reduced proteinuria and immune complex levels.⁸

Of particular interest is the implication for autoimmune diseases. Lectins stimulate class II HLA antigens on cells that do not normally display them, such as pancreatic islet and thyroid cells.⁹ The islet cell determinant to which cytotoxic autoantibodies bind in insulin dependent diabetes mellitus is the disaccharide N-acetyl lactosamine,¹⁰ which must bind tomato lectin if present and probably also the lectins of wheat, potato, and peanuts. This would result in islet cells expressing both class II HLA antigens and foreign antigen togethera sitting duck for autoimmune attack. Certain foods (wheat, soya) are indeed diabetogenic in genetically susceptible mice.¹¹ Insulin dependent diabetes therefore is another potential lectin disease and could possibly be prevented by prophylactic oligosaccharides.

Another suspect lectin disease is rheumatoid arthritis. The normal human IgG molecule possesses carbohydrate side chains, which terminate with galactose. In rheumatoid arthritis much of the galactose is missing, so that the subterminal sugarN-acetyl glucosamineis exposed instead. These deficient IgG molecules feature strongly in the circulating immune complexes that cause fever and symptoms.¹² In diet responsive rheumatoid arthritis one of the commonest trigger foods is wheat, and wheat lectin is specific for N-acetyl glucosaminethe sugar that is normally hidden but exposed in rheumatoid arthritis. This suggests that N-acetyl glucosamine oligomers such as chitotetraose (derived from the chitin that forms crustacean shells) might be an effective treatment for diet associated rheumatoid arthritis. Interestingly, the health food trade has already siezed on N-acetyl glucosamine as an antiarthritic supplement.¹³

Among the effects observed in the small intestine of lectin fed rodents is stripping away of the mucous coat to expose naked mucosa and overgrowth of the mucosa by abnormal bacteria and protozoa.¹⁴ Lectins also cause discharge of histamine from gastric mast cells,¹⁵ which stimulates acid secretion. So the three main pathogenic factors for peptic ulceracid stimulation, failure of the mucous defence layer, and abnormal bacterial proliferation (Helicobacter pylori) are all theoretically linked to lectins. If true, blocking these effects by oligosaccharides would represent an attractive and more physiological treatment for peptic ulcer than suppressing stomach acid. The mucus stripping effect of lectins¹⁶ also offers an explanation for the anecdotal finding of many allergists that a "stone age diet," which eliminates most starchy foods and therefore most lectins, protects against common upper respiratory viral infections: without lectins in the throat the nasopharyngeal mucus lining would be more effective as a barrier to viruses.

But if we all eat lectins, why don't we all get insulin dependent diabetes, rheumatoid arthritis, IgA nephropathy, and peptic ulcers? Partly because of biological variation in the glycoconjugates that coat our cells and partly because these are protected behind a fine screen of sialic acid molecules, attached to the glycoprotein tips.¹⁰ We should be safe. But the sialic acid molecules can be stripped off by the enzyme neuraminidase, present in several micro-organisms such as influenzaviruses and streptococci. This may explain why diabetes and rheumatoid arthritis tend to occur as sequelae of infections. This facilitation of lectins by micro-organisms throws a new light on postinfectious diseases and makes the folklore cure of fasting during a fever seem sensible.

Alternative medicine popularisers are already publishing articles about dietary lectins,¹⁷ often with more enthusiasm than caution, so patients are starting to ask about them and doctors need to be armed with facts. The same comment applies to entrepreneurs at the opposite end of the commercial spectrum. Many lectins are powerful allergens, and prohevein, the principal allergen of rubber latex, is one. It has been engineered into transgenic tomatoes for its fungistatic properties,¹⁸ so we can expect an outbreak of tomato allergy in the near future among latex sensitive individuals. Dr Arpad Pusztai lost his job for publicising concerns of this type (20 February, p 483).

David L J Freed, Allergist. 

14 Marston Road, Salford M7 4ER

1. Gilbert RJ. Healthy eating day. Communicable Disease Report 1988; 33: 3-4.
2. Van Damme EJM, Peumans WJ, Pusztai A, Bardocz S. Handbook of plant lectins: properties and biomedical applications. London: Wiley, 1998:31-50.
3. Pusztai A, Greer F, Grant G. Specific uptake of dietary lectins into the systemic circulation of rats. Biochem Soc Trans 1989; 17: 481-482.
4. Wang Q, Yu L-G, Campbell BJ, Milton J, Rhodes JM. Identification of intact peanut lectinin peripheral venous blood. Lancet 1998; 352: 1831-1832[Medline].
5. Goldstein IJ, Poretz RD. Isolation and chemical properties of lectins. In: Liener IE, Sharon N, Goldstein IJ, eds. The lectins. Orlando: Academic Press , 1986.
6. Kolberg J, Sollid L. Lectin activity of gluten identified as wheat germ agglutinin. Biochem Biophys Res Comm 1985; 130: 867-872[Medline].
7. Weiser MM, Douglas AP. An alternative mechanism for gluten toxicity in coeliac disease. Lancet 1976; i: 567.
8. Coppo R, Amore A, Roccatello D. Dietary antigens and primary IgA nephropathy. J Am Soc Nephrol 1992; 2(10 suppl): S173-S180[Abstract].
9. What triggers auto-immunity? Lancet 1985; ii: 78-79.
10. Uchigata Y, Spitalnik SL, Tachiwaki O, Salata KF, Notkins AL. Pancreatic islet cell surface glycoproteins containing Gal (1-4)GNAc-R identified by cytotoxic monoclonal antibodies. J Exp Med 1987; 165: 124-139[Abstract/Free Full Text].
11. Scott FW, Kolb H. Cow's milk and insulin-dependent diabetes mellitus. Lancet 1996; 348: 613[Medline].
12. Bond A, Kerr MA, Hay FC. Distinct oligosaccharide content of rheumatoid arthritis derived immune complexes. Arthr Rheum 1995; 38: 744-749[Medline].
13. Toohey L. Natural substances combat arthritis with "immune power". Nutri Notes 1997; 2: 1-6.
14. Banwell JG, Howard R, Kabir I, Costerton JW. Bacterial overgrowth by indigenous microflora in the PHA-fed rat. Can J Microbiol 1988; 34: 1009-1013[Medline].
15. Greer F, Pusztai A. Toxicity of kidney bean (Phaseolus vulgaris) in rats: changes in intestinal permeability. Digestion 1985; 32: 42-46[Medline].
16. Freed DLJ, Buckley CH. Mucotractive effect of lectin. Lancet 1978; i: 585-586.
17. Anonymous, but attributed to D'Adamo P. Blood type: the link with diet and disease. What Doctors Don't Tell You 1998; 9: 1-4.
18. Lee HI, Raikhel NV. Prohevein is poorly processed but shows enhanced resistance to a chitin-binding fungus in transgenic tomato plants. Braz J Med Biol Res 1995; 28: 743-750[Medline].