Glutamic Acid Decarboxylase and its role in the onset of Insulin-Dependent Diabetes Mellitus.

By Cara Neese

Insulin-dependent diabetes mellitus (IDDM), also known as Type I diabetes, is a severe metabolic disease of the pancreas. The major clinical manifestation of IDDM is the inability of the body to manufacture insulin, therefore patients must be provided with synthetic insulin. The disease can lead to long-term vascular and neural problems that drastically reduce the patient's quality of life (Schrof, 1999).

IDDM (which is also called juvenile-onset diabetes), often affects individuals before the age of 15. These individuals go through a long asymptomatic period where the cells of the pancreas (insulin-producing cells) are destroyed. It has been discovered in IDDM that the human immune system targets one of its own proteins, the pancreatic enzyme glutamic acid decarboxylase (GAD). GAD, therefore, may play an important role in the onset of Type I diabetes and the destruction of the pancreas (Yoon, et al., 1999).

In order to understand the impact that GAD has on IDDM, it is beneficial to first explore the main characteristics of IDDM. As mentioned before, IDDM is caused by a progressive loss of cells in the pancreas. Studies have shown that the loss of the cells caused by an autoimmune reaction of the patient's immune system against the pancreatic tissues. De Blasio, et al. (1999) described the onset of IDDM as a "dynamical instability" in a local compartment of the pancreatic islet cells ( cells) which spreads like a virus throughout the whole pancreas. Beta cell death is caused by a massive infiltration of macrophages into the islets of Langerhans (the part of the pancreas that houses the cells) followed by an invasion of lymphocytes (DeBlasio, et al., 1999). This period of cell death is asymptomatic; therefore, the patient has no idea anything is wrong until the clinical onset of the disease (O'Brien, et al., 1997).

Another main characteristic of IDDM is the presence in the bloodstream of circulating autoantibodies. These include islet cell autoantibodies (ICAs), insulin antibodies (IAAs) and the focus of this paper, glutamic acid decarboxylase (GAD) antibodies (Schmidli, et al., 1995). All of these types of antibodies have been detected before and after clinical onset of IDDM (Hawa, et al., 2000) and are now the focus of study as disease indicators.

GAD is an enzyme found in high concentration in the pancreas and brain where it catalyzes the conversion of glutamic acid to gamma-aminobutyric acid (GABA). GABA is an inhibitory neurotransmitter that is important in the pancreas as a messenger between neurons and pancreatic cells (Marieb, 1998). Two major isoforms of GAD exist in both the pancreas and the brain. They are products of two separate genes and are homologous (Ujihara, et al., 1994). GAD65 is a 65,000 molecular weight protein made of 585 amino acids and is the principal target of antibodies in IDDM. The second isoform, GAD67, is a 67,000 molecular weight protein with 593 amino acids which has antibodies in a minority of patients with IDDM (Bonifacio, et al., 2000, Schmidli, et al., 1995). Both isoforms of GAD have carboxy terminal and amino terminal ends, as well as a middle portion which contain regions that are unique to each isoform (Bonifacio, et al., 2000). (An important characteristic of the middle portion of the GAD sequence, which will be discussed later, is that it contains a conformation dependent region (Bonifacio, et al., 2000).) GAD65 is dominant in the islet cells of the pancreas (Ujihara, et al., 1994) where it is found in membranes and nerve endings (Soghomonian & Martin, 1998). GAD67 is widely distributed in cells (Soghomonian & Martin, 1998). Both types of GAD are found in hydrophilic forms localized in the cytoplasm and in post-translationally modified hydrophobic forms associated with the membranes of small synaptic vesicles (Syren, et al., 1996).

Many questions still remain regarding the mechanism of action of GAD in the onset of IDDM. As presented in numerous studies, however, GAD does play an important (if somewhat unknown) role in IDDM. Specific experiments have shown that GAD expression is necessary for the autoimmune destruction of cells. One study in particular performed on non-obese diabetic (NOD) mice (which are close to humans in their autoimmune manifestation of IDDM), incorporated an antisense GAD transgene (which suppresses GAD expression in the cells) into a subset of mice and found that IDDM did not develop (Yoon, et al., 1999). In other words, the immune system cannot provoke the destruction of cells in NOD mice in the absence of GAD (Yoon, et al., 1999). Another specific study (undertaken by Myers, et al. (1998)) using fetal mouse tissue, showed hyperexpression of GAD in pancreatic islet cells. This hyperexpression was caused by an impairment of mitochondria complex I as a result of IDDM. Impairment of the mitochondria resulted in an accumulation of glutamate that directly induced the hyperexpression of GAD in the cells. This hyperexpression can dramatically increase the rate of cell loss as GAD is targeted by the immune system (Myers, et al., 1998). If these same qualities hold true with humans, then these experiments have a great significance in the treatment of IDDM.

In the study of IDDM, the specific immune system is the major player in the destruction of cells, therefore it is important to understand its main characteristics. The specific immune system has two main branches: the cellular (or cell-mediated immune system (CMI)) and the humoral immune system. Both parts of the immune system are stimulated when foreign substances (called antigens) enter the body. The response to these antigens, however, varies according to the branch of the immune system that responds. When an antigen enters the body, antigen-presenting cells (APCs) (macrophages, B cells and dendritic cells) process the antigen and present it on their surfaces. (The part of the antigen that is presented here is called the epitope.) After the antigen is presented, one branch of the immune system will be activated--via its specific T helper (Th) lymphocytes. Th1 lymphocytes activate the CMI, which develops natural killer cells and cytotoxic T cells that attack and destroy infected cells through an inflammatory response involving cell lysis. Th2 lymphocytes, on the other hand, stimulate the humoral immune system which produces B cells (Schmidt & Roberts, 2000). These B cells produce plasma cells, which manufacture antibodies specific to the antigens presented by the APCs. The antibodies mark the antigens for destruction by other cells of the immune system (Marieb, 1998). The problem with IDDM, which is exacerbated by GAD, is that the immune system has problems recognizing self-proteins, so it sees proteins like GAD as foreign, attacks them, and breaks down the cells in which they are found (namely cells) (Marieb, 1998).

The three main characteristics of autoimmune diseases may explain the increased occurrence of IDDM with the expression of GAD (Marieb, 1998). Inefficient or ineffective lymphocyte programming may cause autoimmune diseases, which occur when the immune system turns against itself. In the case of GAD, this would mean that T lymphocytes would attack GAD instead of recognizing it as a self-protein. Another mechanism of autoimmune action occurs when self-proteins appear in the circulation that were not previously exposed to the immune system. The immune system in this scenario is not able to recognize these proteins as self because the lymphocytes were never exposed to them before. Von Boehmer and Sarukhan (1999) propose that GAD in cells "may be proteolytically cleaved to generate a completely different set of peptides not found in other cells." Since these specialized peptides are not found elsewhere in the body, the immune system considers them non-self and mounts an attack against them. The result when combined with the other pathological factors of IDDM, then, is the systematic destruction of GAD and the cells. The third characteristic of autoimmune diseases relates to a problem with the antibodies produced by the immune system. Antibodies produced against foreign proteins may be close in form to self-proteins, therefore the immune system attacks the self-antigens as though they were foreign. All of these characteristics do not truly explain the mechanism by which GAD causes IDDM, but they do promote a better understanding of how the immune system may malfunction to attack GAD (Marieb, 1998).

The onset of IDDM is associated with a genetic defect in chromosome six where certain mutant alleles that code for the major histocompatibility complex class II (MHCII) proteins are found (DeBlasio, et al., 1999). MHCII proteins provide the molecular basis for non-self recognition and are found embedded in the surface of immune system cells (Schmidt & Roberts, 2000). The mutation in the MHCII allele in NOD mice is found at only one locus; where a serine residue replaces an aspartic acid residue. Mice with this miscoded allele show a markedly higher incidence of IDDM. This higher incidence may be caused by the fact that the diabetogenic MHCII alleles select (and present to T cells) different epitopes of GAD65 than normal MHCII alleles (Chao, et al., 1999).

Through experimentation and investigation specific to the onset of IDDM, a mechanism for the pathogenesis of the disease has been proposed. The islet cells of the pancreas are first infiltrated by antigen presenting cells (APCs) which internalize self-antigens. Once the self-antigens are processed by the APCs, they are presented in association with MHCII molecules on the surface of the APCs. The APCs, then, are presented to T cell receptors. The nature of the MHCII molecule-antigen-T cell receptor complex and the cytokines (chemical compounds involved in the immune system (Marieb, 1998)) that are in close proximity to the complex determine which T lymphocyte helper cell will be activated (thus determining which arm of the specific immune system will be activated). The self-proteins, then, are processed by the APCs, activate the immune system, and cause a cascade mechanism that leads to the destruction of cells and exposure of other antigens (Menard, et al., 1999). GAD is important in this process because the immune response to GAD precedes that of other cell antigens tested (Zekzer, et al., 1998). GAD is the primary autoantigen involved in the cascade effect of IDDM (Menard, et al., 1999). The APCs of the immune system present GAD to T lymphocytes. This activates the T lymphocytes and they proliferate and attack the cells of the pancreas (Von Boehmer & Sarukhan, 1999). GAD has been the target of both humoral and cell-mediated immune responses (Zanone, et al., 1994), but the cell-mediated immune response is more closely connected with the onset of IDDM, whereas the humoral immune response does not seem to cause IDDM as frequently (Petersen, et al., 1999, Harrison, et al., 1993). According to one study performed by L.C. Harrison, et al. (1993), one or the other of the specific immune responses predominates in IDDM cases, not both. Immunity is led by one of the immune responses based on genetic makeup, nature and dose of antigens, and the type of APCs involved in the response (Harrison, et al., 1993). Therefore, the severity of IDDM, according to the aforementioned studies, depends upon which branch of the specific immune system is activated.

The CMI and humoral immune systems, then, must have specific characteristics in order for scientists to determine which is leading the immune system attack in IDDM. Low levels of GAD antibodies (GADAbs) and high levels of T cells against GAD characterize the cell-mediated immune response. The humoral response, on the other hand, has high levels of GADAbs and low levels of T cells against GAD (Harrison, et al., 1993). These findings confirm the fact that cell-mediated immune responses tend to cause IDDM, because the increased presence of autoreactive T cells is known to cause the destruction of the cells of the pancreas. Therefore, the presence of GADAbs may be a clinical marker for IDDM, but a better representation of the disease is found with the presence of autoreactive T cells (Harrison, et al., 1993). The fact that GADAbs are also present in low risk people follows the information that a humoral immune response produces a high number of antibodies, but a low incidence of IDDM (Daw & Powers, 1995). However, recent research still focuses on the GADAbs rather than the presence of T cells.

GADAbs are found in 70-80% of new onset and a significant percentage of long-term IDDM patients (Petersen, et al., 1994). They are also found in individuals who are at risk but never develop the disease. GADAbs are heterogeneous and differ in titer (amount), isoform specificity and epitope region reactivity (Ujihara, et al., 1994). The frequency of GADAbs varies in IDDM patients according to age, sex, race and sensitivity of antigen assays (tests) (Schmidli, et al., 1995). As mentioned before, GADAbs bind to conformational epitopes of GAD65 within the central portion of the molecule (amino acids 240-440) and at the -COOH terminal (amino acids 440-585) (Bonifacio, et al., 2000). The middle portion is bound most often, followed by the carboxy terminal, and finally the amino terminal (amino acids 11-16)(Bonifacio, et al., 2000). GADAbs do not bind denatured GAD, which is logical based on the information regarding the binding to conformational epitopes (Daw & Powers, 1995).

GAD has been positively implicated in the onset of diabetes, therefore it may also be an important target for treatment. A previously mentioned experiment, where an antisense GAD gene was expressed in islet cells (so no GAD would be produced in cells) showed that GAD is necessary for the onset of diabetes. Therefore, if GAD is not expressed, through a genetic treatment or incorporation of GAD-less islet cells, no diabetes will develop (Von Boehmer & Sarukhan, 1999). Other studies have administered GAD orally or nasally in order to sensitize the immune system to GAD. This is a dangerous treatment, however, because it may actually accelerate the onset of the disease (Zekzer, et al., 1998). It should be tested further in order to understand completely why the immune system is targeting GAD. Other forms of treatment for IDDM that may be related to its linkage with GAD involve drugs and environmental factors. One such drug, cyclosporin, inhibits the T cell-dependent immune responses and slows down cell loss--probably by keeping T cells from attacking self-antigens like GAD (Petersen, et al., 1994). Due to the severity of IDDM, further study of GAD will definitely be beneficial. It is obvious, however, that much study must be done in order to comprehend the workings of this complex disease.

LITERATURE CITED

Bonifacio, E., Lampasona, V., Bernasconi, L., and A.G. Ziegler. Maturation of the humoral autoimmune response to epitopes of GAD in preclinical childhood Type I diabetes. Diabetes. 49: 202, 2000.

Chao, C., Sytwu, H., Chen, E.L., Toma, J., and H.O. McDevitt. The role of MHC class II molecules in susceptibility to Type I diabetes: identification of peptide epitopes and characterization of the T cell repertoire. Proceedings of the National Academy of Sciences of the United States. 96: 9299-9304, 1999.

Daw, K., and A.C. Powers. Two distinct glutamic acid decarboxylase auto-antibody specificities in IDDM target different epitopes. Diabetes. 44: 216-221, 1995.

De Blasio, B.F., Bak, P., Pociot, F., Karlsen, A.E., and J. Nerup. Onset of Type I Diabetes. Diabetes. 48: 1677, 1999.

Harrison, L.C., Honeyman, M.C., DeAizpurua, H.J., Schmidli, R.S., Colman, P.G., Tait, B.D., and D.S. Cram. Inverse relation between humoral and cellular immunity to glutamic acid decarboxylase in subjects at risk of insulin-dependent diabetes. The Lancet. 341: 1365-1370, 1993.

Hawa, M.I., Fava, D., Medici, F., Deng, Y., Notkins, A.L., De Mattia, G., David, R. and G. Leslie. Antibodies to IA-2 and GAD65 in Type I and Type II diabetes. Diabetes Care. 23: 228, 2000.

Marieb, E. Human Anatomy and Physiology, 4 ed. (Menlo Park: Addison Wesley Longman, 1998), 758-786.

Menard, V., Jacobs, H., Jun, H., Yoon, J., and S.W. Kim. Anti-GAD Monoclonal Antibody Delays the Onset of Diabetes Mellitus in NOD Mice. Pharmaceutical Research. 16: 1059-1066, 1999.

Myers, M.A., Georgiou, H.M., Byron, S., and M.D. Espsoti. Inhibition of Mitochondrial Oxidative Phosphorylation Induces Hyper-Expression of Glutamic Acid Decarboxylase in Pancreatic Islet Cells. Autoimmunity. 30: 43-51, 1998.

O'Brien, B.A., Harmon, B.V., Cameron, D.P., and D.J. Allan. Apoptosis is the mode of beta-cell death responsible for the development of IDDM in the nonobese diabetic (NOD) mouse. Diabetes. 46: 750-758, 1997.

Petersen, J.S., Dyrberg, T., Karlsen, A.E., Molvig, J., Michelson, B., Nerup, J., and T. Mandrup-Poulsen. Glutamic acid decarboxylase (GAD65) autoantibodies in prediction of beta-cell function and remission in recent-onset IDDM after cyclosporin treatment. Diabetes. 43: 1291-1297, 1994.

Petersen, J.S., Kulmala, P., Clausen, J.T., Knip, M., Dyrberg, T., and Childhood Diabetes in Finland Study Group. Progression of Type I Diabetes is Associated with a Change I the Immunoglobulin Isotope Profile of Autoantibodies to Glutamic Acid Decarboxylase (GAD65). Clinical Immunology. 90: 276-281, 1999.

Schmidli, R.S., Colman, P.G., and E. Bonifacio. Disease sensitivity and specificity of 52 assays for glutamic acid decarboxylase antibodies: the Second International GADAb Workshop. Diabetes. 44: 636-641, 1995.

Schmidt, G.D. and L.S. Roberts. Foundations of Parasitology, 6^th ed. (Boston: McGraw Hill, 2000), 21-27.

Schrof, J. The Silent Killer. U.S. News and World Report. 126: 50, 1999.

Seppa, N. Cows' milk, diabetes connection bolstered. Science News. 155: 404, 1999.

Soghomonian, J., and D.L. Martin. Two isoforms of glutamate decarboxylase: why? Trends in Pharmacological Sciences. 19: 500-506, 1998.

Syren, K., Lindsay, L., Stoehrer, B., Jury, K., Luhder, F., Baekkeskov, S., and W. Richter. Immune Reactivity of Diabetes-Associated Human Monoclonal Autoantibodies Defines Multiple Epitopes and Detects Two Domain Boundaries in Glutamate Decarboxylase. The Journal of Immunology. 157: 5208-5214, 1996.

Ujihara, N., Daw, K., Gianani, R., Boel, E., Yu, L., and A.C. Powers. Identification of glutamic acid decarboxylase autoantibody heterogeneity and epitope regions in Type I diabetes. Diabetes. 43: 968-976, 1994.

Von Buehner, H., and Adelaida S. GAD, a Single Autoantigen for Diabetes. Science. 284: 1135, 1999.

Yoon, J., Yoon, C., Lim, H., Huang, Q., Kang, Y., Pyun, K., Hirasawa, K., Sherwin, R.S., and H. Jun. Control of Autoimmune Diabetes in NOD mice by GAD Expression or Suppression in Beta Cells. Science. 284: 1183, 1999.

Zanone, M.M., Petersen, J.S., Peakman, M., Mathias, C.J., Watkins, P.J., Dyrberg, T., and D. Vergani. High Prevalence of autoantibodies to glutamic acid decarboxylase in long-standing IDDM is not a marker of symptomatic autonomic neuropathy. Diabetes. 43: 1146-1152, 1994.

Zekzer, D., Wong, F.S., Ayalon, O., Millet, I., Altieri, M., Shintani, S., Solimena, M., and R.S. Sherwin. GAD-reactive CD4+ Th1 Cells Induce Diabetes in NOD/SCID Mice. Journal of Clinical Investigation. 101: 68-73, 1998.

Copyright © 2000 Cara Neese and Koni Stone