Immunotherapeutical approaches to Crohn's disease in view of a possible pathogenic role of Mycobacterial antigens.

Arnold Stronkhorst, Guido N.J. Tytgat, Sander J.H. van Deventer.

Dept. Gastroenterology, Academic Medical Center, Meibergdreef 9, 1105 AZ, Amsterdam, The Netherlands

INTRODUCTION

The immune system of the intestinal tract has been extensively studied. The lamina propria of adults is heavy populated with a variety of lymphoid cells: B- and T-lymphocytes, plasma cells, macrophages, mast cells, eosinophils and basophils. In normal subjects the majority of the B-cells are committed to IgA synthesis (90%), the residual 10% being represented (in order of decreasing frequency) by IgM, IgG, IgE and IgD producing plasma cells. These cells, which are located diffusely in the mucosa and more organized in Peyer's patches, together with an intact mucosal layer provide an effective barrier to the antigenic food load, that passes the bowel daily. Small amounts of antigen can transmigrate through the mucosa either by endo/pinocytic processes or via the tight junction complex. The production of antibodies to antigens penetrating the barrier is a physiological process that usually results in systemic tolerance. Changes in these immunological processes, caused by an altered mucosal barrier, an abnormal antigen-antibody or antigen-antigen presenting cell response, may play a role in the pathogenesis of Crohn's disease.

Although many immunological changes have been documented in Crohn's disease, it still is not clear whether these changes are a primary disorder of the immune system or secondary to the inflammatory process; in both cases the initiating process or agent that leads to inflammation remains unknown. In this chapter we will discuss some immunological changes that can be observed in Crohn's disease patients. We will speculate on the pathogenic role of mycobacterial antigens and suggest a novel immuno therapeutical intervention strategy.

Antigen presentation and T-cell activation in Crohn's disease

In Crohn's disease an increased permeability of the intestinal mucosa for larger molecules may exist^1,2, probably due to an abnormal tight junction structure³. Various (food derived-?) antigens can pass the mucosal layer by pino- or endocytosis, but oral and systemic tolerance for most of them is derived early in life by respectively mucosal IgA production and specific non-responsiveness of the systemic immune system to luminal antigens. It is not clear whether the permeability for very large molecular structures such as mycobacteria or mycobacterial antigens is increased in Crohn's disease. Besides a primary change in permeability it is also possible that penetration of antigens in the inflamed mucosa is a secondary process which perpetuates a destructive primary immune response. The supposed primary increased permeability is not obligate for immune activation, because enterocytes may exhibit antigen presenting capacities, shown by there expression of HLA-DR^4,5,6.

The importance of T-lymphocytes within the normal gut and gut-associated lymphoid tissue is generally accepted: their precise role, however, is uncertain. In the lamina propria most T-lymphocytes express the CD4 (helper) phenotype, whereas in the epithelial layer the CD8 (cytotoxic/suppressor) subtype dominates. The CD4 + cells are predominantly of the helper-inducer and cytolytic subtype, and less frequently of the suppressor-inducer phenotypes⁷. In Crohn's disease there is an influx of inflammatory cells into the diseased area and local production of soluble mediators of inflammation has been shown^8,9. The pivotal role of T-cells in Crohn's disease is well known¹⁰. There is an increased number of mucosal T-cells, but the CD4 to CD8 ratio is the same as in control subjects¹¹. Earlier immuno-histopathological reports could not show an increase in T-cell activation markers^12,13. This in contrast to the fact that mononuclear cells isolated from mucosa from Crohn's disease patients did show a slight increase in the expression of these markers¹⁴. Later studies showed that activated T-cells are abundantly present the submucosa of children with Crohn's disease, in contrast to mucosa of ulcerative colitis patients^15,16. T-cell activation is not only evident in histological specimens^17,18 but can also be shown in peripheral blood: even the serum level of soluble inter-leukin-2 receptor is increased ^19,20 The total numbers of CD4 and CD8 positive lymphocytes in peripheral blood of Crohn's disease to patients are normal, but the T-cells are activated as they express CD25 (interleukin-2 receptor) on their surface²¹.

In case of HLA class II restricted antigen presentation, as in type IV immune response, an antigen (fig. 1) is processed by the antigen presenting cell (enterocyte, mononuclear cell, dendritic cell) and presented on its surface by a MHC class II molecule. This complex is recognized by the T helper 1 (Th 1) cell receptor. Th 1 cells mediate delayed type hypersensitivity²². Additional bindings (integrins) and a costimulatory signal (interleukin- 1) are necessary for T-cell activation²³. The antigen(s) to which those T-cells react are still unknown. Their identification however would be a major step forward in our understanding of Crohn's disease. Mycobacteria elicit a cellular immune response in their hosts. This response usually leads to protective immunity, but may sometimes be accompanied by immunopathology due to delayed type hypersensitivity (like in tuberculoid leprosy). Mycobacterial antigens are capable of creating a granulomatous inflammation²⁴, in which mononuclear cells play an important role. Living mycobacteria are not necessary in this process. Mycobacterial antigens can be shown in mucosa of Crohn's disease patients²⁵ and Thayer showed that serum antibodies to mycobacterial antigens are more often present in Crohn's disease patients²⁶. Mycobacteria selectively induce human T-cells with a Th 1-like cytokine profile²⁷. Lymphocytes from Crohn's disease patients show marked suppressor activity when stimulated with M. paratuberculosis²⁸. These immunological findings may explain why Crohn's disease is a transmural inflammation: the mycobacterial antigen is passing through the mucosa one way or another and subsequently processed by the antigen presenting cell (APC). Then in the lamina propria the T-cell receptor of the T-helper (CD4 +) cell recognizes the antigen-MHC II complex. This results in T-cell activation and a granulomatous transmural inflammation moderated by tumor necrosis factor alpha, t interferon, and interleukin-2. Its however not clear what causes the perpetuating immune response and subsequent inflammation.

IMMUNOTHERAPY

The ideal immunosuppressive therapy would be either to neutralize the causative antigen, or to delete all reactive T-cell clones, without affecting the normal immune response. Somewhat less specifically, one could interfere with the antigen presenting process. As can be seen in figure 1. intervening in the antigen presenting pathway provides opportunities for immunotherapy. Because the antigen which causes the immune response in Crohn's disease is still unknown, no vaccination, antibodies or desensibilisation are available.

As T-cells play a central role in all immune responses, they are a primary target for any immuno-therapeutic intervention strategy. For many years polyclonal antibodies that suppress T-cell functions have been investigated as immunosuppressive agents in a variety of diseases. With the development of monoclonal antibodies, specific immunotherapy directed against T-cells became available and had promising results in both animal models and (auto-)immune diseases^29,30. Monoclonal antibody treatment has largely been applied clinically in cancer and modulation of the immune response to produce immunosuppression for treatment of autoimmune and graft versus host diseases. It is also very effective in the treatment of acute allograft rejection³¹. Human monoclonal antibodies have been used in both viral and bacterial infections³². Some monoclonal antibodies are now registered for clinical use, for example OKT3 in allograft rejection. Problems with monoclonal antibody therapy were recognized soon because anti-(murine-) antibody production limited therapeutic possibilities³³. Furthermore there are many side effects in murine- and non specific (for example OKT3-) monoclonal antibody therapy. New "humanized" chimeric and recombinant monoclonal antibodies have promising therapeutic prospects. Interference with the antigen presentation by blocking the T-cell receptor, the CD4 receptor, interleukin-1B pathway or the interleukin-2 receptor all are therapeutical options. Other , less specific, strategies consist of blockade of the inflammatory response by antibodies directed against tumor necrosis factor-alpha, interleukin-2, interferon-t or their receptors. Currently some of these monoclonal antibody therapies are tested in clinical trials.

The treatment of Crohn's disease with Corticosteroids, Azathioprine/6-mercaptopurine³⁴, and 5-Aminosalicylic acid is well accepted and needs no further discussion in this chapter.

Cyclosporine is a novel immunosuppressant, widely used in organ transplantation, that has a selective effect on T-cell media immune responses³⁵. Cyclosporine is able to reduce expression of Class II molecules on the epithelium in a dramatic way. It also reduces IL-2 and hence t-interferon production. Cyclosporine has proven to have a beneficial effect in active chronic Crohn's disease patients who were resistant to or intolerant of corticosteroids³⁶. Cyclosporine therapy during three month was well tolerated and no serious adverse events were observed, but is associated with a risk of nephropathy.

Anti CD4 antibody treatment

Anti-CD4 treatment in Crohn's disease seems rational in review of the antigen presenting pathway and T-cell activation in Crohn’s disease, as described above. Depletion of CD4 + cells is correlated with rapid improvement in a prototypic model of T-cell-mediated autoimmunity: experimental allergic encephalomyelitis³⁷. Reports of remissions of Crohn's disease in patients infected with the HIV virus provide additional arguments for anti-CD4 treatment in Crohn’s disease³⁸. In these cases remission of Crohn's disease occurred simultaneously with a decrease in CD4 cells. In phase I trials, rheumatoid arthritis, which is also characterized by a granulomatous inflammation and T-cell activation, responds to the treatment with murine monoclonal antibodies against the CD4+ (T helper) cells^39,40. The effect on remission of symptoms after one week of therapy lasted 3 weeks to more than 5 months. Only short-lasting low grade fever on the first day of treatment was seen. However in six out of eight patients an anti-mouse-Ig response developed. This can cause problems (ineffectiveness, allergic response) in longterm treatment of these chronic diseased patients. Currently we are evaluating a chimeric (mouse/human) anti-CD4 monoclonal antibody in Crohn's disease patients, with promising results and without serious side effects. These antibodies have a human IgG constant part and a mouse anti-CD4 variable part. We assume that humanizing the antibody causes less antigenetic response and a longer half live.

Anti TNF-alpha treatment

Both mycobacterial and Crohn's disease are characterized by a granulomatous, transmural inflammation of the bowel. Tumor necrosis factor alpha (TNF-alpha) plays a very important and perhaps pivotal role in granulomatous diseases^41,42,43. Blocking TNF-alpha prevents the formation of granuloma's⁴⁴. TNF-alpha is in increased amounts present in mucosa and stool of patients with active IBD^45,46. TNF-alpha can be controlled by reducing its production, TNF-neutralizing antibodies or by blocking the TNF-receptor. Several studies, using murine or murine/human chimeric antibodies, have been initiated in septicemia, and the results from a phase I study have been published⁴⁷. To our knowledge, anti-TNF therapy has not been investigated in Crohn's disease.

Pentoxifylline (a xanthine derivative) has been described to counteract endotoxin induced effects in different animal models^48,49. Pentoxifylline inhibits in macrophage cultures the synthesis of mRNA for TNF-alpha⁵⁰ and therefore is an alternative for anti-TNF antibody therapy. Pentoxifylline is also able to inhibit the inflammatory action of interleukin-1 and TNF-alpha on neutrophil function⁵¹. No data are available on the effect of pentoxifylline in granulomatous diseases. Thalidomide, which is presently used to treat some forms of leprosy, has also been shown to specifically inhibit transcription of the TNF gene in monocytes⁵².

Anti interleukin-2 receptor and interleukin-1B treatment

As shown in fig. 1 interleukin-l (IL-1) and interleukin-2 (IL-2) play a role in HLA class II restricted antigen presentation and (auto-)immune response. This is confirmed in IBD by studies showing enhanced IL-2 and IL-lB production⁵³ in mononuclear cells isolated from the mucosa and biopsy specimens¹⁷. In autoimmune diseases and in Crohn's disease serum soluble IL-2 receptor concentration is raised, indicating T-cell¹⁹. Intestinal mucosal mononuclear cells of Crohn's disease patients produce only one-third of the amount of interleukin-2 generated by control cells after stimulation, but were able to exhibit comparable cytotoxicity⁵⁴. This is probably due to either an increased number of interleukin-2 responsive cells or an exaggerated reactivity to interleukin-2. Lack of IL-2 production seems to play a role in the disability of a granulomatous response to Schistosoma mansoni in mice⁵⁵. Recombinant IL-2 therapy reversed this diminished response. Its not clear what the exact function of the granuloma formation in Crohn's disease is; Is it an exaggerated immune response to antigens or a decreased response to bacteria? Counteracting IL-2 production or neutralizing its effects by monoclonal antibodies directed against its receptor, may be an option for immunotherapy of Crohn's disease especially when we consider the antigen c.q. abnormal immune response. Monoclonal antibodies against IL-2 receptor are used in the prophylactic treatment of human kidney allograft recipients, with good result and only few side effects⁵⁶. No studies are as yet available monoclonal antibody treatment against the IL-1 receptor.

REFERENCES

Figure 1:The antigens on the luminal side of the mucosa transmigrate through the epithelial layer, either by endo-/pinocytosis or by the tight junction complex. In case of HLA-restricted antigen presentation as in Crohn's disease the antigen is processed by the antigen presenting cell (APC) and presented on its surface by the class II major histocompatibility complex (MHC II). The antigen presenting cell can either be an enterocyte, monocyte/macrophage or dendritic cell. The MHC II-antigen complex is recognized by the T-cell receptor. Additional binding by the CD4 molecule (integrin) and a costimulatory signal are necessary for T-cell activation. This in turn leads to the inflammatory response.