During the past five to ten years, an evolution of thinking has taken place for scientists involved in autoimmunity research. The foundation of that new vision is this: Autoimmunity, at some level, exists in everyone. The human immune system does not eliminate all the potential cells that can attack the body that made them. Moreover, the immune system does this on purpose, to generate a system that has a broad repertoire for detecting infectious agents and cancer cells, and one that can self-regulate unwanted responses.

In most people, the immune system has a number of checks and balances that regulate self-reactive cells through a series of intrinsic and extrinsic factors - at least, most of the time. One class of T cells, called regulatory T cells, is designed to see self-antigen and control the potentially damaging cells.¹ Consequently, immunologists adjusted their focus from trying to determine why autoreactive cells escape deletion to exploring how the immune system maintains a homeostatic balance and why it sometimes fails. It seems clear that understanding this balance - a peaceful coexistence of protective immunity without self-destructing - will reveal how the immune system really works and how autoimmunity is triggered.

The challenge resides in the complexity of the biochemical and cellular pathways involved in tolerance, which is the immune system's ability to distinguish self from non-self, allowing the body to recognize and destroy invading viruses, bacteria, and other pathogens while ignoring the body's own tissue and organs. Moreover, genetic and environmental influences that we don't fully understand multiply the complications of the pathways. The discovery and characterization of the regulatory pathways, however, will lead to new drugs and cell therapies for autoimmune diseases.^2-4

More than 80 diseases with an autoimmune etiology have been identified. Clinicians divide autoimmune diseases into multiple categories based on systemic versus organ-specific autoimmunity, distinct roles of pathogenic antibodies versus pathogenic T cells, and complex mechanisms of action in the different disease settings. In some autoimmune diseases, an antibody acts against cell-surface or matrix antigens. This mechanism appears in Graves' disease, insulin-resistant diabetes, myasthenia gravis, pemphigus vulgaris, and other autoimmune diseases. In Graves disease, for example, autoreactive T cells promote the development of autoantibodies that then bind to thyroid cells that express the receptor for thyroid stimulating-hormone. The cells are destroyed by a combination of complement-mediated lysis (a cascade of proteins that poke holes in cells) and antibody-dependent cellular cytotoxicity (in which an antibody forms a bridge between cytotoxic and target cells), causing cell lysis. Together these mechanisms of cell destruction lead to hypothyroidism.

Some autoimmune diseases, such as systemic lupus erythematosus (SLE) and thrombocytopenia, involve immune complexes of autoantigens bound to antibodies. These multimeric complexes can attach to a cell's surface leading to the deposition of complement, an immune mediator that causes inflammation and destruction of cells and tissues. For example, autoantibody-antigen complexes deposited in various organs such as the kidneys prove especially damaging in SLE.

The largest group of autoimmune diseases consists of T cell-mediated autoimmune diseases, including type 1 diabetes (T1D), rheumatoid arthritis (RA), psoriasis, inflammatory bowel disease, and multiple sclerosis (MS), to name a few. In these diseases, T cells - including helper T cells and cytotoxic T cells, which express the CD4 and CD8 surface protein, respectively - generate cytokines and cytolytic granules that directly destroy the targeted tissues. There are multiple pathways to this tissue destruction. The CD8⁺ T cells can interact directly with a target cell, such as an islet or a neuron, and lyse the cell. Moreover, cytokines including tumor necrosis factor-α (TNF-α), also destroy tissues by triggering a suicide event (apoptosis) with the target cell. Apoptosis also can be triggered in tissues by way of Fas ligand (FasL), another pro-apoptotic membrane protein.

Consider the mechanism behind T1D: T cells attack beta cells, the insulin producers, in islets of the pancreas. Once those cells are destroyed, the production of insulin is dramatically compromised and patients lose control of blood-glucose levels, resulting in the complications of high blood sugar. This disease, like many autoimmune diseases, is a slow, progressive disease with relapses and remissions that often take years to manifest. This is especially true in T1D, as each individual has a great excess of insulin-producing islets of Langerhans. Thus, more than half of the 1,000,000 islets must be destroyed or inactivated before the disease is manifested.

T cells are often considered the conductor of the immune orchestra. Like the individual sections of woodwinds, strings, and percussion, T cells come in a number of subtypes. The CD8⁺ T cells recognize antigens in the context of class I major histocompatibility complex (MHC) molecules and largely recognize proteins made within the cells; CD4⁺ T cells recognize processed antigens, picked up exogenously and presented in the context of class II MHC molecules. Together, these T cells can recognize proteins picked up by the antigen presenting cells (APCs) - dendritic cells, macrophages, and B cells - as well as directly infected cells. The T-helper lymphocytes (Th1) primarily produce interferon-γ (IFN-γ) and interleuken-2 (IL-2), which promote the development of CD8⁺ T cells and intracellular immunity. Th2 cells, on the other hand, produce IL-4, IL-5, and IL-13, which drive humoral immunity. Finally, the newly defined Th17 subset produces IL-17, which promotes inflammatory immunity, especially in the gut. Together these processes combat infections through a combination of cellular immunity to destroy infected cells and antibody production that mops up the circulating infectious agents, such as viruses and bacteria.

The immune system battles invaders in several general ways. First, innate mechanisms can block microbes with epithelial barriers or engulf them with phagocytes. Within a day of an invasion, adaptive mechanisms also start. B cells can recognize antigens from microbes, which leads to B cells producing antibodies that mark invaders for attack. Likewise, some cells that engulf microbes break up the antigens and display parts of them on their surface, thereby becoming antigen-presenting cells. These cells activate T cells that help kill infected cells.

Recent attention finally has been paid to a small population of CD4⁺ T cells that express the high-affinity IL-2-receptor-achain, CD25.⁵ These cells express a specific transcription factor, forkhead box p3 (Foxp3), which has been associated with a homeostatic activity focused on regulating an ongoing immune response and, as will be described below, the pathogenic autoreactive cells crucially involved in autoimmunity. The majority of CD4⁺ CD25⁺ Foxp3⁺ regulatory T cells (Tregs) develop in the thymus in response to self-antigens. However, Tregs with a similar phenotype can develop from CD4⁺ CD25^- cells in the periphery. It is interesting that unique transcription factors - Tbet for Th1 cells, GATA-3 for Th2 cells, ROR a for Th17 cells, and FoxP3 for Tregs - control the individual subsets of T cells. Finally, this T-cell orchestra is greatly influenced by a number of critical cell-cell interactions with antigen presenting cells (i.e., dendritic cells, macrophages and B cells), which help target foreign proteins to the naive T cells and promote differentiation down the individual T-cell pathways.

Autoimmunity arises from a combination of genetic predisposition and environmental factors that lead to a failure of the immune system's tolerance mechanisms. Given the right constellation of events - perhaps stress, a viral infection, or general deterioration of the regulatory pathways - CD4 ⁺ T cells can start responding to self-tissues, initiating an autoimmune cascade. Those CD4 ⁺ T cells support the development of autoreactive B cells, which make autoantibodies that attack the self-tissues and organs. The CD4 ⁺ T cells also produce pro-inflammatory and cytotoxic cytokines that directly destroy tissues, as well as growth and differentiation factors that stimulate the development of CD8 ⁺ cytotoxic T cells, inflammatory macrophages, and autoantibody production by B cells. In combination, the autoantibodies and cytotoxic cells damage tissue, release more autoantigens, and exacerbate the autoimmune response.

Autoimmune diseases could develop in most anyone, but they don't. Historically, a simple explanation - that the immune system eliminates all autoimmune cells during T- and B-cell development through "central" tolerance - has been the dominant paradigm in the field. This elimination takes place when the autoreactive T cells that develop in the thymus undergo "negative selection," a process that deletes the majority of potentially autoreactive cells. Autoreactive B cells undergo negative selection in the bone marrow, where receptor editing replaces an autoantigen-reactive receptor with one that is poised to react with foreign proteins rather than self-proteins.

It is now clear, however, that large numbers of potentially autoreactive cells escape negative selection and clonal elimination, for several reasons. For example, the self-antigen might get expressed only in specialized organs, and not when the cells are "seeing" self-antigens in the bone marrow or thymus. In addition, autoreactive cells develop as part of the critical Treg subset, both inside and outside the thymus to control autoreactivity. This and other "peripheral tolerance" mechanisms exist to control these unwanted T- and B-cell reactivities.

Peripheral-tolerance mechanisms fall into three broad categories: cell inactivation (anergy) and cell elimination (apoptosis), both due to suboptimal activation; activation-induced cell death (AICD), which occurs as a consequence of overactivation; and active suppression mediated by regulatory cells and cytokines.

Jeffrey Bluestone

Courtesy of Jeffrey Bluestone

The induction of anergy or apoptosis occurs as a consequence of a weak or altered T-cell-receptor signal or an ineffective costimulatory signal. When T cells encounter antigen, the T-cell receptor (TCR) interacts with a specific antigenic peptide/MHC molecule. This leads to a series of intracellular phosphorylation events, enzyme activation, and gene transcription that initiate the primary activation events (called signal one). This process is promoted in a number of ways, including the use of altered antigenic peptides that deliver an ineffective TCR signal or blockade by monoclonal antibodies, which prevents the TCR from interacting effectively with peptide-MHC on professional APCs.

Even if the TCR signal is effectively delivered, full T-cell activation and differentiation require a second, "costimulatory" signal that enhances the TCR signal (signal one) and promotes unique biochemical signals to the T cells, or B cells in a similar fashion. In this step, B7 proteins (also called CD80 and CD86) on the APC bind to the T cell's CD28 receptor, which leads to a second biochemical-signaling cascade (called signal two). Among other things, this costimulatory signal upregulates CD40L, a ligand for CD40. When CD40L (also called CD154) binds with CD40 on an APC, this enhances the expression of the B7 proteins, which creates a positive-feedback loop between CD28 and CD40. This combination of signals results in: effective signal transduction, which leads to increased cell-cycle progression; induction of survival factors that preempt apoptosis; and production of essential cytokines, which are essential for a productive immune response.

Costimulation plays an essential role in autoimmunity. In mice, for example, blocking CD40:CD154 costimulation shuts down a variety of autoimmune diseases, including T1D in nonobese diabetic mice, experimental autoimmune encephalitis (the mouse equivalent of MS), and mouse models of RA. Similarly, blocking CD28/CD86 pathways can inhibit a number of autoimmune syndromes in small animals, as well as in humans. Together the TCR and costimulatory antagonist have developed as part of a new drug arsenal for the treatment of autoimmune diseases. (For more information, see "Fine Tuning Our Defenses.")

Although the above costimulatory pathways lead to positive immune activation, some cell-surface interactions can negatively regulate immunity by turning off T and B cells. These negative regulatory pathways include: receptor-ligand interactions, such as CTLA-4/CD80⁶; programmed death (PD)-1/PD-L1; and B and T lymphocyte attenuator (BTLA)-4. Thus, during an immune response, these negative regulators are expressed on the activated T cells and result in shut down of the functional activity. Moreover, the genetic disruption of these molecules leads to immune dysregulation. For example, dysregulation of CTLA-4 leads to death of a mouse in less than one month due to a massive autoimmune response.

Peripheral tolerance can also come from AICD. Here, a peptide-MHC on a professional APC binds a TCR inducing the expression of Fas. The Fas protein can bind to FasL on various cell types, which activates a cysteine aspartate protease, caspase-8, that triggers a cell-death cascade. The Fas-FasL interaction can also eliminate autoreactive B cells through apoptosis. Thus, it is not surprising that mutations in Fas or FasL can prevent AICD, leading to both T- and B-cell mediated autoimmunity.

Perhaps the most potent regulation of autoimmune diseases is mediated by a specialized class of regulatory T cells, the CD4 ⁺ Tregs that control unwanted immune responses.⁷ These cells can actively suppress ongoing autoimmunity and restore self-tolerance in patients with autoimmune diseases. How Tregs suppress immune responses, though, remains poorly understood, but a variety of properties appear to give these cells their suppressive capabilities. For example, many in vitro and in vivo studies show that Tregs can function in the lymph nodes as well as in the peripheral tissues. In vivo, Tregs act through dendritic cells to limit autoreactive T-cell activation, which prevents their differentiation and acquisition of effector functions. Moreover, CD4⁺ FoxP3⁺ Tregs act through cell-cell contact. This might involve the negative regulator, CTLA-4, or other cell-surface molecules. Finally, CD4⁺ Tregs suppress an immune response through a cytokine-dependent mechanism where both IL-10 and transforming growth factor β (TGF-β) have been implicated. Sister regulatory T cells - Tr1 and Th3 cells - also mediate suppression by cytokine-dependent pathways.

By limiting the supply of activated pathogenic cells, the autoantigen-specific Tregs prevent or slow down the progression of autoimmune diseases. Thus, precisely the same antigenic self-proteins that have been implicated in causing autoimmunity can be recognized by this specialized regulatory T-cell population to shut it down. This raises the obvious question: Why is this protective mechanism insufficient in autoimmune-prone individuals to control the pathogenic T- and B-cell responses? Is it due to a shortage of Tregs cells, the development and accumulation of Treg-resistant pathogen T cells, or the production of other cytokines or innate immune responses, such as those elicited during an infection that compromise Treg activity? Defining these parameters has important implications for enhancing Treg activity in patients with disease, but more importantly, it may help advance research into the use of cell-based therapies (either Tregs or regulatory dendritic cells). Such therapies could be used to expand Treg cells for reinfusion of this robust, homeostatic T-cell subset to restore self-tolerance in patients with autoimmune diseases such as MS, T1D, and RA. However, it should be noted that successful regulatory cell therapy will undoubtedly require a concomitant debulking step to remove pathogenic B and/or T cells.

CTLA-4lg impacts several biochemical pathways. It can interact with B7 receptors on dendritic cells, and that leads to the production of indoleamine 2,3-dioxygenase (IDO), which suppresses the activation of T cells (1). It can also block B7-CD28 binding, and that blocks cell differentiation and leads to anergy and an increase in cell death (2). CTLA-4lg also can increase the immune response in two ways: It can block binding between B7 and CTLA-4 on already activated helper or cytotoxic T cells (3); and it can prevent B7 from binding CD28 on regulatory T cells (4). Redrawn from JA Bluestone et al. "CTLA-4Ig: Bridging the basic immunology with clinical application," Immunity, 24: 233-8, 2006.

Traditional treatments for autoimmune disease essentially blindfold the immune system as a whole. For example, nonspecific immunosuppressants such as cyclosporin A and anti-inflammatory agents such as steroids can inhibit immune responses in general, reducing the impact of autoimmune diseases in the process. Unfortunately, these treatments have transient effects and so must be used for the lifetime of the patient. In addition, suppressing the entire immune system makes patients vulnerable to infections. Worse still, overall immune suppression increases the risk of cancer. On the other hand, an explosion of novel immunotherapies to treat autoimmunity take advantage of the various aforementioned pathways of peripheral immune regulation to suppress disease symptoms and potentially induce a state of immune tolerance.

One of these advances in autoimmune treatments came in December, 2005, when the US Food & Drug Administration approved a new Bristol-Myers Squibb drug, Orencia (abatacept) for the treatment of RA. This protein, CTLA-4Ig, is created through recombinant technology, connecting human CTLA-4 (a CD28 antagonist) to human immunoglobulin G1 (IgG1). Specifically, the drug inhibits costimulation via the CD28 receptor on T cells by binding the CD28 ligands (CD80 and CD86) on APCs, thus preventing effective CD28-signal transduction and leading to anergy, apoptosis, and tolerance in some systems. Other interesting drugs in this same class of costimulation blockers include monoclonal antibodies that block the CD154-CD40 pathway, the inducible-costimulator pathway, and the B-cell-specific (Baff) pathway.⁸

Another encouraging class of new immune therapies relies on cell depletion as a means of debulking the pathogenic cells. For example, the anti-CD20 monoclonal antibody therapy, Rituxan (rituximab), which Genentech originally developed to deplete CD20+ tumor cells, has been successfully used to treat RA, MS, and lupus. By depleting B cells, the drug eliminates some autoantibody-producing cells and a major population of APCs that are essential for maintaining autoimmune T-cell activation in these and perhaps other diseases. Importantly, this illustrates the potential for these depleting agents and others, such as thymoglobulin and alemtuzumab (Campath-1H), to cross over - bringing new approaches to autoimmunity that are already in use for other diseases.

Despite the growing knowledge of autoimmune mechanisms and how to treat these diseases, many avenues of research need much more work. For example, scientists must find the genetic elements that determine the propensity for autoimmune disease. In addition, we need biomarkers that reveal when autoimmune diseases exist, even at the earliest stages, and that show when therapies work or fail. Scientists studying autoimmunity strive to better understand the homeostasis in the immune system, including balance within and between central and peripheral tolerance.⁹ Armed with such understanding, we can leverage this knowledge for developing new treatments, essentially mimicking what the body already does on its own. Only then can science develop powerful yet specific means for keeping our immune system in harmony to fight off pathogens but not attack the very system designed to protect us.

Jeffrey A. Bluestone is the A.W. and Mary Clausen Distinguished Professor of Medicine, Pathology, Microbiology, and Immunology, and the director of the Diabetes Center at the University of California, San Francisco. He is also director of the Immune Tolerance Network.