Ever since undergoing my training as a medical herbalist, I have maintained a particular interest in the cause and treatment of autoimmune disease. The term "autoimmune" is applied to a very wide spectrum of human disease, sometimes tentatively. However, many of these diseases do tend to share common factors in their pathophysiology. But we can confidently state that autoimmune disease does not arise from a single cause. In fact, it is quite possible that the same autoimmune disease will have different causes from person to person. The current approach in medical science is not very comfortable with such an idea. However traditionally based systems of medicine such as phytotherapy place great emphasis on the treatment of the individual.

In order to better treat autoimmune disease, it is necessary to have an understanding of its causes. This in turn would lead to a systematic approach to defining the key causative and sustaining factors operating in each case. For each individual, it is likely that the autoimmune process has been precipitated by a unique and complex interaction of causative events. What we therefore need is a multi-factorial model that allows us to individualize treatments, yet at the same time takes into account the most likely factors operating in each particular disease. This requires a blend of traditional understanding with the latest research findings. Such a synthesis is the goal of the modern phytotherapist.

Current Research – Is it Looking In The Wrong Direction?
Much of the current research into autoimmune disease is centered on two main areas:

a) the development of immune tolerance to self-tissue
b) the search for a pre-established defect in either immune cells or immune function that gives rise to autoimmunity.

Particularly with regard to the first area of research, I wonder if the wrong question is being asked. Rather than looking at how self-tolerance develops in a human being, perhaps the question should be, "How does loss of immune tolerance to self arise?" Scientists are very interested to learn how the embryo acquires self tolerance. But how relevant is this to a previously well 50-year-old who suddenly develops an autoimmune disease? In fact, self recognition by the immune system is part of its normal function. It is not appropriate to go into detail here but phenomena such as clonal anergy, idiotypic networks, and MHC antigens are all examples of the immune system recognizing self tissue. But it does this in a regulated non-destructive way. We need to understand the circumstances that cause the capacity for self recognition to become destructive.

Is there a pre-established defect in the immune system that causes autoimmune disease? Do some people have a biological time bomb pre-programmed to explode into tissue destruction at some point in their lives? We do not know for certain, but other scenarios are also possible. Consider the immune system as a computer. It has input and output. The former is antigenic material, and the latter is the immune response. It has hardware in the form of the white blood cells, and the software is the way these cells interact. The bias of current scientific investigations into autoimmunity is that there is some defect in the hardware or software that is awaiting discovery. But anyone who has worked with computers knows the old saying – garbage in equals garbage out. Using this analogy, we can see that it is possible that the inappropriate response of the immune system, which is autoimmunity, may be due to aberrant input.

A New Theory of Autoimmunity
In 1986, two American scientists, Fred Westall and Robert Root-Bernstein published a paper in the Lancet entitled "Cause and Prevention of Postinfectious and Postvaccinal Neuropathies in Light of a New Theory of Autoimmunity."¹ The authors were most interested in the incidence of postinfectious neuropathies, such as Guillain-Barré syndrome, and postvaccinal neuropathies, such as the potential reaction to measles vaccination. But they also linked this interest to a new theory about the development of autoimmunity in general. As far as I know, this paper did not cause much lasting interest in scientific circles, but their theory provides the basis for a practical multifactorial model for the treatment of autoimmune disease.

The basic rationale for their theory comes from experimental animal models of autoimmunity. Take, for example, experimental allergic encephalomyelitis (EAE), an autoimmune disease of the brain. If myelin basic protein (a brain protein) is injected into an animal, it does not cause EAE, no matter how often it is injected. The development of EAE only occurs when the myelin basic protein is injected together with Freund's complete adjuvant. Freund's complete adjuvant is a water in oil emulsion containing antigen in the aqueous phase, in this case myelin basic protein, and dead tuberculosis bacteria in the oil phase.

Extending this model to humans, the basic hypothesis is that three requirements must be met to induce autoimmune disease:

1) There must be an antigen present which is the same as or similar to some fragment of self tissue. We can call this the primary lesion.
2) There must be a second antigen which is chemically complementary to the first antigen (for EAE, this is the tuberculosis bacteria).
3) Both antigens must be present and immunologically active in the host at the same time.

I would like to extend this theory by examining the role of the second complementary antigen. The presence of this antigen probably confuses the immune system and causes an inappropriate response to the first antigen. Once this process is triggered, it becomes self-sustaining due to the inflammation and tissue damage that occurs. (This aspect is known as the "Hit and Run Hypothesis" of autoimmunity.) This tendency for the immune system to respond inappropriately can be called a state of immune dysregulation. The immune system is confused, but this need not always be due to a complementary antigen. It is possible that other inputs can cause a state of immune confusion.

The obvious implication of the above model is that we need to find and eradicate the causes of the primary lesion and the immune dysregulation. What may not be as obvious is that the autoimmune disease may still continue for a considerable time after these causes have been treated, since to some extent, the process can become self-sustaining. Therefore control of sustaining causes should also be sought, and this may involve symptomatic treatment and even the judicious use of natural agents that depress immunity.

The Primary Lesion
The primary lesion might often be created by the presence of a micro-organism (Table 1).² This need not necessarily be a clinical infection, but for the sake of brevity I will use this term. How can an infection create the primary lesion? Micro-organisms contain or produce proteins, which are antigenic; that is, the immune system is capable of recognizing parts of these proteins as foreign and mounting a reaction against them. One protein might provide many antigenic regions known as epitopes, and each micro-organism might have many antigenic proteins. It is now known that some of these foreign epitopes exactly resemble human epitopes. This phenomenon has been described as "molecular mimicry."² Originally, the term molecular mimicry was coined to describe the way that a micro-organism might escape detection from the immune system. It was felt that proteins from micro-organisms evolved to be antigenically similar to their hosts in order to escape immune detection and thereby enhance survival. Molecular mimicry certainly occurs more often than could be attributed to chance. (The probability that a random six amino acid sequence will be identical in two proteins is one in 20 million.)

Table 1: Factors That May Provide The Primary Lesion
Infection
Infestation
Chronic Tissue Destruction

It is now recognized that molecular mimicry may be an important factor in the etiology of autoimmune disease.² An immune reaction directed against the micro-organism cross-reacts with a similar self antigen. The evidence for molecular mimicry is summarized in Table 2. Many examples of the points listed in the table are documented in the literature.²

Table 2: Evidence For Molecular Mimicry
Amino acid sequence similarities between microbial and human proteins
Antibodies to viruses cross-react with uninfected host cells
Animal models of autoimmunity often involve viral infection
Post-infectious auto-antibody production is a common phenomenon

Some examples of the amino acid sequence similarities between microbial and human proteins are provided in Table 3.³ The similarity between a Klebsiella protein and HLA B27, and its implications for autoimmune disease, are supported by other research. The incidence of ankylosing spondylitis (AS) shows a high correlation with HLA B27. More than 90% of patients with AS are HLA B27-positive. Research has demonstrated that a substantial proportion of patients with AS have antibodies in their blood that react to the epitope shared by HLA B27 and Klebsiella pneumoniae nitrogenase.⁴ There was no reaction from antibodies in control subjects. Moreover antibodies to Klebsiella, but not to other bacteria, have been shown to be present in patients with active AS by seven different techniques.⁵ In this context, HLA B27 may be the self antigen that is the site of the cross-reactivity. The particular sites of inflammation in AS are also the tissues, which can express a relatively higher concentration of HLA B27 as a surface antigen.

Table 3: Sequence Similarities between Microbial Proteins and Human Host Proteins.³ Immunological cross-reactivity between the indicated pairs of proteins has been demonstrated in all but the last two cases.

Dr. Alan Ebringer of King's College Hospital, London, decided to test this association clinically. Patients with AS were placed on a low starch and sugar diet, because it was postulated that this would reduce the number of Klebsiella in the gut.⁶ Most patients on this program had their disease process halted, but the diet must be adhered to for at least six months.

Micro-organisms might provide the primary lesion in other ways. For example, haptenization may occur, where a fragment from the infecting agent may bind to a larger endogenous protein. This protein-hapten complex becomes antigenic, and an immune reaction may be directed against part of the protein and hence is a reaction to self.² Drugs can also produce autoimmune reactions in this way.

"Altered self" is another theory of autoimmunity involving micro-organisms.² An infecting agent, particularly a virus, may cause self tissue to change. The virus harnesses the host cell to make self-like proteins that are sufficiently close to self to create cross-reactivities, but different enough to enable strong immune recognition and attack.

Organisms that can provide the primary lesion include bacteria, viruses, protozoa, or fungi. The bacteria can be endogenous, that is they may be present in the gut flora or on the skin or mucous membranes. Although they may be endogenous, their presence could still be abnormal. Infestation could also provide the primary lesion. When I was studying in England, I met a herbalist who claimed good results in the treatment of juvenile arthritis, an autoimmune disease. His approach was to treat the child for worms. That was his particular insight into the problem, which is consistent with the proposed model.

Why is there not clear evidence for the role of micro-organisms in most autoimmune diseases? There are many aspects to this problem. Perhaps the best answer is that micro-organisms have been implicated in many autoimmune diseases, but the results have not been consistent. However, this meshes well with the multifactorial model. Autoimmune disease is not an infection per se, but it may be a response to an infection under particular circumstances. As such, it would be unreasonable to expect a single infectious agent to be implicated in any given autoimmune disease. Furthermore, if the micro-organisms involved are present at low levels or exist in unusual forms such as cell wall-deficient variants, they may be very difficult to detect. Some endogenous microflora that might be involved in autoimmune disease could also present in healthy subjects. Consequently, it would be difficult to define their significance in someone who was unwell.

Some additional examples of the association of infection with the autoimmune destruction of tissue are described below. This is not a comprehensive review of this topic. Guillain-Barré syndrome is viewed as a reactive, self-limited, autoimmune disease triggered by a preceding bacterial or viral infection. Campylobacter jejuni, a major cause of bacterial gastroenteritis, is the most frequent antecedent pathogen.⁷ The syndrome has also been linked to influenza vaccination in isolated instances.⁸

Patients with rheumatoid arthritis placed on a vegetarian diet for one year had a significant reduction in anti-Proteus mirabilis antibody levels, which was correlated with decreased disease activity.⁹ Patients with autoimmune thrombocytopenia (low platelet count) who also were positive for Helicobacter pylori experienced a significant increase in platelet count when the bacterium was eradicated from their stomachs.¹⁰

Eighty-nine percent of patients with untreated celiac disease were positive for the presence of human adenovirus serotype¹², an adenovirus isolated from the human intestinal tract, whereas the incidence in controls was 0 to 12.8%.¹¹ Amino acid sequence homology (and potential for molecular mimicry) was noted between a protein in gluten and this virus. Infection with Coxsackie B viruses has been linked to the development of insulin-dependent diabetes mellitus in children.¹² Other viruses have been linked to this disorder.¹²

A case was reported of a woman who developed ulcerative colitis during primary cytomegalovirus infection. She subsequently developed chronic recurrent disease.¹³ Retroviruses have been implicated in the development of various autoimmune diseases. For example, retroviral antibodies were detected in patients with primary biliary cirrhosis and other biliary disorders of unknown cause.¹⁴

Chronic tissue destruction can also potentially provide the primary lesion. This tissue damage may be due to an infection or may have other causes such as a tumor or toxins. The destruction creates facilitatory signals to the immune system. Local macrophages are induced to attack anything in the region thereby creating a snowball effect. This is known as the "bystander theory" since the macrophages are actually bystanders and are not involved in the initiation process.²

The multiplicity and the diversity of pathogenic auto-antibodies found in patients with systemic lupus erythematosus (SLE) has fascinated immunologists for decades. Surface blebs on apoptotic cells (cells partaking in programmed self destruction) now appear to provide antigenic material, together with the nuclear material released from apoptotic cells.¹⁵ It has thereby been suggested that the first event in lupus is an increased tendency towards apoptosis of certain cell populations, possibly lymphoid cells. What in turn causes this is an important question, but a high antioxidant status may hinder this process. Low antioxidant status has certainly been identified as a risk factor for lupus.¹⁶

Immune Dysregulation
As well as the primary lesion, the model also requires a state of immune system dysregulation (Table 4). Concurrent infection may cause immune dysregulation via several possible pathways. It may provide a complementary antigen as in animal models of autoimmunity. A state of endotoxinemia could result from some bacterial infection or from abnormal bowel flora combined with a leaky gut wall. Endotoxin is a potent general stimulator of the immune system. Many infecting agents, particularly viruses, can create a state of generalized activity of B lymphocytes known as polyclonal activation.²

Table 4: Factors That May Cause Immune System Dysregulation
Infection or Infestation
Endotoxinemia
Allergy or Chemical Sensitivity
Diet
Injury or Foreign Body
Stress
Cancer

Other mechanisms are also possible. Viral infection may create abnormal expression of immunoregulatory molecules, such as the MHC antigens, on the surface of somatic cells (that is, cells that are not part of the immune system). This then causes these cells to be attacked and destroyed by the immune system even though they may not actually be infected with the virus.²

Allergy and chemical sensitivity have the potential to cause immune dysregulation. There is a university in the UK where the medical treatment of multiple sclerosis includes the removal of amalgam fillings from patients. Allergy to the mercury in the fillings could create a state of immune imbalance. A direct toxic effect from the mercury could also be significant. Hair dyes have been implicated in autoimmune disease in women, and this could be another example of chemical sensitivity causing a state of immune dysregulation. This association has been recently questioned.¹⁷

Diet may be a factor. It is often observed that patients with Crohn's disease respond well to an elemental diet.¹⁸ This diet is free of antigens. Similarly there are anecdotal accounts of amelioration of multiple sclerosis after following a gluten-free diet. Antigens in the diet could therefore create a state of immune dysregulation. Consumption of cow’s milk by infants has been linked to insulin-dependent diabetes,¹⁹ and a dairy-free diet effected a substantial improvement in more than half of a group of patients with ankylosing spondylitis.²⁰

Injury or foreign bodies may be a factor. Cases of SLE due to breast implants were documented in the literature. When the implants were removed (they were not leaking), the SLE was cured. However, this association has been questioned in recent studies.²¹ In another example, a rose thorn was not effectively removed after injury, and the person developed rheumatoid arthritis.²² Miners with silicosis also suffer a higher incidence of systemic sclerosis compared to the general population.

Stress could create immune dysregulation. Many autoimmune diseases are precipitated during periods of extreme stress.² The best documented examples are for insulin-dependent diabetes.

Cancer increases the likelihood of developing autoimmune disease. It is possible that the mechanisms employed by the tumor to escape immune detection create a state of immune dysregulation.²

Auxiliary Factors in Autoimmunity
Some additional factors may contribute to the development of autoimmunity and need to be considered. Although these factors cannot be included under the above headings, they may facilitate the processes involved in the proposed model. As such they can be classified as auxiliary factors.

Table 5: Possible Auxiliary Factors in Autoimmunity
Intestinal Wall Hyperpermeability
Reduced Hepatic Integrity
Reduced Phagocytic Capacity
Reduced Proteolytic Activity

Intestinal wall hyperpermeability, or a leaky gut, increases the antigen and endotoxin load in the body. Blood from the digestive tract is screened by the phagocytic Kupffer cells in the liver. Kupffer cells have the special property of sequestering engulfed material. That is, they do not readily present such material to helper T cells in order to enlist an immune reaction. As such, they have a dampening effect on the reactivity of the immune system. If phagocytic activity is generally reduced due to poor health, or the liver is damaged and the activity of the Kupffer cells is thereby compromised, then the risk for antigen or endotoxin spillover into the general circulation is substantially increased.

Poor proteolytic capacity may also be an auxiliary factor in the development of autoimmunity. To understand the significance of this, we need to consider the theories of Pierre Grabar, who was a respected immunologist at the famous Pasteur Institute.²³ Grabar postulated that proteolytic enzymes in the body break down immunogenic material into smaller fragments. He proposed that these smaller fragments actually induce a state of tolerance in the immune system to their parent immunogenic proteins. Grabar called these fragments "tolerogens." He also felt that auto-antibodies were produced when the proteolytic processing of immunogenic material was overloaded. This in turn might lead to autoimmune disease in some cases.

Developing a Treatment from the Model
The proposed model supports many traditional naturopathic concepts that associate chronic disease with poor immunity, diet, intestinal dysbiosis, autotoxemia, poor liver function, and chronic infection. In particular, it stresses the importance of treating chronic disease by altering bowel flora and improving digestive function and integrity. This in turn emphasizes the relevance of a wholesome diet.

It is important when approaching autoimmune disease to develop a detailed case history. Specific causative factors must be identified for the individual as much as is reasonably possible. Particularly look for indications of problems in apparently unrelated systems or organs. Try to identify the process of the primary lesion. Childhood illnesses, especially serious or atypical ones, are relevant, as are previous surgery, accidents, or recurrent or chronic illnesses, especially in the respiratory, urinary, or digestive systems.

Diet is important in the treatment of autoimmune disease for three main reasons. An inappropriate diet can result in allergies and intolerances. Imbalanced diets can increase autoimmunity, e.g., high-protein or high-fat diets. Finally, diet can contribute to intestinal dysbiosis.

This association between diet, intestinal dysbiosis, and autoimmune disease has been highlighted in some recent studies. Excessive bacterial fermentation of sulfide in the colon could contribute to ulcerative colitis. In a pilot Australian study, four patients with ulcerative colitis were placed on a low-sulfur diet that included avoidance of eggs, dairy, nuts, sulfur-containing food additives, and cruciferous vegetables. All four patients showed substantial improvement.²⁴ Patients with rheumatoid arthritis assigned to a vegan diet displayed a significant change in fecal flora, which was correlated with clinical improvement.²⁵

The cause of any given autoimmune disease will vary from person to person, so individualization of the case is critical. The primary lesion and the source or sources of immune dysregulation should be identified as much as is possible and treated according to the problems uncovered. However, knowledge of the particular disease can assist to define these problems. The scientific literature now contains many publications dealing with the association of various factors such as micro-organisms, diet, and intestinal dysbiosis with specific autoimmune diseases.

Phytotherapy for Autoimmunity – General Strategy
Each heading in this section defines a treatment goal in the phytotherapy of autoimmune diseases and these treatment goals are based on the above model. Not all the treatment goals are relevant to every person. They must be tailored to each individual case at the time of presentation in the way described above. Moreover, the extensive list of herbs provided should be regarded as a menu from which items can be selected. It is not practical or appropriate to attempt to take a large number of herbs at any one time.

1. Reduce the Presence of Micro-organisms
Immune-enhancing, antiviral, and antimicrobial herbs are required. The use of immune-enhancing herbs in autoimmune disease is controversial. However, the above model clearly justifies their use to resolve the detrimental influence of micro-organisms. I have never found immune-enhancing herbs to aggravate an autoimmune disease. Echinacea root is particularly safe and useful, as it especially works on the input side of immune activation, mainly phagocytosis. Its influence will help the body to resolve the presence of micro-organisms that are having a deleterious influence on health, and yet it will not stimulate the autoimmune processes. Echinacea will also improve phagocytic screening of portal blood in the liver. In this case, it should be combined with Silybum (milk thistle) if there is evidence of hepatic impairment. Other useful immune-enhancing herbs include Picrorrhiza, Andrographis, and Astragalus. (Andrographis is contraindicated in pregnancy.)

Hypericum (St John's wort) and Thuja are two proven antiviral herbs that have the potential to work systemically (that is they can exert antiviral activity in the body after oral doses). Hypericum contains hypericin and pseudohypericin, which are active against enveloped viruses. Many of the viruses implicated in autoimmune disease have a viral envelope. To be effective, the daily dose of total hypericins should be in the range of 2 to 4 mg per day. Therefore, use of a high hypericin preparation is essential. Thuja is active against both enveloped and naked viruses.

Some antibacterial herbs are particularly useful against enteric infections. Herbs to be considered should include Hydrastis (goldenseal) and other berberine-containing plants, and preparations of Allium sativum (garlic), which release allicin after ingestion. Propolis, grapefruit seed extract, and Melaleuca (tea tree) oil may also be used. Herbs for bacterial contamination of the urinary tract include Juniperus, Barosma (buchu), and Arctostaphylos uva-ursi (bearberry). Antiseptics for the respiratory tract include Thymus (thyme) and Inula (elecampane). Melaleuca oil inhalation may also be useful.
Most of the above antibacterial herbs also have antifungal activity. Antiprotozoal herbs include Artemisia annua, Euphorbia, and berberine-containing plants.

2. Eliminate Infestation
Infestation may need to be treated. In this case, immune-enhancing treatment coupled with herbs such as garlic, Tanacetum vulgare (tansy), and Artemisia absinthium (wormwood) should be applied. Tansy and wormwood should be taken in controlled doses and for short periods only. They are contraindicated in pregnancy.

3. Control or Eliminate Dysbiosis
Immune-enhancing herbs again have a role here. A fiber-rich diet low in refined starch and sugar will encourage healthy bowel flora. Treat any constipation. Periodic treatment with gastrointestinal antiseptics (see above) alternating with mucilaginous herbs such as Ulmus (slippery elm) will also help to normalize gut flora. Chicory root is a source of inulin that can act as a probiotic.

4. Repair a Leaky Gut Wall
Allergens and alcoholic beverages should be eliminated from the diet. Demulcent herbs such as Althaea (marshmallow root), anti-inflammatory herbs such as Matricaria (chamomile), and healing treatments including propolis and Calendula are indicated. Improving phagocytosis with Echinacea and hepatic integrity with Silybum will reduce the systemic impact of a leaky gut wall. Similarly, the normalization of intestinal dysbiosis will eliminate one of the causes.

5. Reduce the Impact of Xenobiotic Toxins
Where possible, the source should be removed, e.g., dental amalgam, hair dyes, insecticide exposure. Otherwise depurative herbs such as Arctium (burdock) and stimulants of hepatic metabolism such as Schisandra may reduce the impact of introduced toxins.

6. Support the Organs Involved
If the organs involved in the processes of the primary lesion or immune dysregulation can be identified, then treatments to improve their health and function will assist in limiting the overall disease process. For example, if prostatitis is identified (as can be the case in ankylosing spondylitis), then treat with saw palmetto as well as urinary tract antiseptics. If there is a bladder or urethral "infection," add Glycyrrhiza, Crataeva, and urinary tract demulcents (e.g., Zea-corn silk) to the normal antiseptic treatment. Similarly, treat liver damage or "infection" with Silybum and lung damage or "infection" with anticatarrhal herbs (e.g., Verbascum – mullein), mucous membrane restoratives (e.g., Hydrastis) and Equisetum (horsetail) decoction for its silica content.

7. Improve Proteolytic Function
According to Grabar, improving the levels of proteolytic enzymes in the body will help to stabilize the immune system. Enhancing digestive proteolytic enzymes will decrease the immunogenic nature of food and possibly increase proteolytic enzymes in tissue fluid. Digestive stimulant herbs such as bitters and Coleus increase proteolytic enzyme release from the stomach and pancreas. Melilotus (sweet clover) contains coumarin, which enhances the breakdown by macrophages of potentially antigenic protein accumulated in extracellular spaces.

8. Reduce the Sustaining Causes
The use of anti-inflammatory and immune-depressing herbs is not just symptom control. Provided the causative factors are also being treated, judicious use of these herbs will help to break the vicious cycle sustaining the tissue destruction. Anti-inflammatory herbs include those containing saponins, which interact with endogenous corticosteroids. These include Glycyrrhiza (licorice), Bupleurum, and Aesculus (horsechestnut). Other anti-inflammatory herbs include Salix (willow bark) and Curcuma (turmeric). Tanacetum parthenium (feverfew) inhibits inflammatory degranulation of polymorphs. Ginkgo counters the effects of PAF, and Boswellia, the formation of leukotrienes. These will help to control inflammation in inflammatory bowel disease. Picrorrhiza contains apocynin, which is a selective and potent inhibitor of neutrophil oxidative burst. This selective anti-inflammatory activity may be important in controlling rheumatoid arthritis. Herbs with documented immune-depressing activity are Hemidesmus, Tylophora, and Stephania. Use of Stephania is not recommended due to the high risk of adulteration with the toxic herb Aristolochia. Neither is the use of Tripterygium wilfordii (the thunder god vine) due to the high risk of serious side effects.

Probably the most powerful but safest immune-depressing herb is Tylophora indica. In excessive (but not high) doses, it will act as an emetic, and it should not be taken continuously. The best way to have Tylophora is to take between 20 to 50 drops of a 1:5 tincture (the highest dose in that range which does not cause nausea) for the first 10 to 14 days of each month.

Desensitization to the original antigen or the cross-reacting antigen is beyond the scope of phytotherapy. However, it is worth discussing in the above context since it should also help to break the vicious cycle of autoimmunity. One way to achieve desensitization is the gastrointestinal administration of antigen. This technique has been successfully tested in clinical trials on multiple sclerosis and rheumatoid arthritis. Breakdown products from bacterial wall fragments are excreted in the urine, and their oral ingestion may cause desensitization to their effects. Tolerogens may also be excreted in the urine.

References
1. Westall FC, Root-Bernstein R. Cause and prevention of postinfectious and postvaccinal neuropathies in light of a new theory of autoimmunity. Lancet. 1986; 2: 251-252.
2. Shoenfeld Y, Isenberg D. The mosaic of autoimmunity (The factors associated with autoimmune disease). Amsterdam: Elsevier, 1989.
3. Oldestone MBA. Molecular mimicry and autoimmune disease. Cell. 1987; 50: 819-820.
4. Schwimmbeck PL, Oldestone MBA. Klebsiella pneumoniae and HLA B27-associated diseases of Reiter’s syndrome and ankylosing spondylitis. Current Topics in Microbiology and Immunology. 1989; 145: 45-55.
5. Ebringer A, Cox NL, Abuljadayel I, et al. Klebsiella antibodies in ankylosing spondylitis and proteus antibodies in rheumatoid arthritis. British Journal of Rheumatology. 1988; 27S: 72-85.
6. Ebringer A, Wilson C. The use of a low starch diet in the treatment of patients suffering from ankylosing spondylitis. Clinical Rheumatology. 1996; 15 (suppl 1): 62-66.
7. Hahn AF. Guillain-Barré syndrome. Lancet. 1998; 352: 635-641.
8. Ropper AH, Victor M. Influenza vaccination and the Guillain-Barré syndrome. New England Journal of Medicine. 1998; 339(25): 1845-1846.
9. Kjeldsen-Kragh J, Rashid T, Dybwad A, et al. Decrease in anti-Proteus mirabilis but not anti-Escherichia coli antibody levels in rheumatoid arthritis patients treated with fasting and a one year vegetarian diet. Annals of the Rheumatic Diseases. 1995; 54: 221-224.
10. Gasbarrini A, Franceschi F, Tartaglione R, et al. Regression of autoimmune thrombocytopenia after eradication of Helicobacter pylori. Lancet. 1998; 352: 878.
11. Kagnoff MF, Paterson YJ, Kumar PJ, et al. Evidence for the role of a human intestinal adenovirus in the pathogenesis of coeliac disease. Gut. 1987; 28(8): 995-1001.
12. Clements GB, Galbraith DN, Taylor KW. Coxsackie B virus infection and onset of childhood diabetes. Lancet. 1995; 346: 221-223.
13. Lortholary O, Perronne C, Leport J, et al. Primary cytomegalovirus infection associated with the onset of ulcerative colitis. European Journal of Clinical Microbiology and Infectious Diseases. 1993; 12: 570-571.
14. Mason AL, Xu L, Guo L, et al. Detection of retroviral antibodies in primary biliary cirrhosis and other idiopathic biliary disorders. Lancet. 1998; 351: 1620-1624.
15. Bach JF, Koutouzov S. New clues to systemic lupus. Lancet. 1997; 350 (suppl III): 11.
16. Comstock GW, Burke AE, Hoffman SC, et al. Serum concentrations of alpha tocopherol, beta carotene, and retinol preceding the diagnosis of rheumatoid arthritis and systemic lupus erythematosus. Annals of the Rheumatic Diseases. 1997; 56(5): 323-325.
17. Sanchez-Guerrero J, Karlson EW, Colditz GA, et al. Hair dye use and the risk of developing systemic lupus erythematosus. Arthritis and Rheumatism. 1996; 39(4): 657-662.
18. Papadopoulou A, Rawashdeh MO, Brown GA, et al. Remission following an elemental diet or prednisolone in Crohn’s disease. Acta Paediatrica. 1995; 84: 79-83.
19. Verge CF, Howard NJ, Irwig L, et al. Environmental factors in childhood IDDM. A population-based, case-control study. Diabetes Care. 1994; 17(12): 1381-1389.
20. Appelboom T, Durez P. Effect of milk product deprivation on spondyloarthropathy. Annals of the Rheumatic Diseases. 1994; 53: 481-482.
21. Nyren O, Yin L, Josefsson S, et al. Risk of connective tissue disease and related disorders among women with breast implants: A nation-wide retrospective cohort study in Sweden. British Medical Journal. 1998; 316: 417-422.
22. Hawkins SJ, Blake DR, Doherty M, et al. Rheumatoid arthritis developing after plant thorn synovitis. British Medical Journal. 1982; 285: 1620.
23. Grabar P. "Self" and "not-self" in immunology. Lancet. 1974; 1: 1320-1322.
24. Roediger WEW. Decreased sulphur aminoacid intake in ulcerative colitis. Lancet. 1998; 351: 1555.
25. Peltonen R, Nenonen M, Helve T, et al. Faecal microbial flora and disease activity in rheumatoid arthritis during a vegan diet. British Journal of Rheumatology. 1997; 36: 64-68.