INTRODUCTION

TOP ABSTRACT INTRODUCTION BALANCE BETWEEN Th1/Th2... WHAT ARE THE SIGNALS... REFERENCES

The immune system has the capacity to distinguish between antigens derived from pathogenic microorganisms, to which it is necessary to generate protective immunity, and harmless environmental antigens such as those encountered at mucosal surfaces, to which it should develop tolerance. In this review we focus on the mechanisms by which the immune system responds to antigens encountered in the lung. It is suggested that allergy and asthma arise as the result of a breakdown in the normal regulatory mechanisms that operate at mucosal surfaces. We examine the contribution experimental animal models have made in defining the primary events involved in allergen sensitization, the immunological mechanisms that limit the induction of pathogenic responses, and the role of the Notch signaling pathway in the regulation of peripheral immunity. As we come to a better understanding of the cellular events that regulate the decision between productive immunity and tolerance, this may reveal why some patients are susceptible to allergen-induced pulmonary inflammation while others appear to be protected.

	BALANCE BETWEEN Th1/Th2 IMMUNITY TO INHALED ANTIGENS

TOP ABSTRACT INTRODUCTION BALANCE BETWEEN Th1/Th2... WHAT ARE THE SIGNALS... REFERENCES

Responses to Aerosolized Soluble Proteins

Several rodent models have been useful in analyzing the early cellular events that occur in response to inhaled proteins antigens (1). These studies have revealed that inhalation of soluble proteins through the respiratory tract leads to activation of CD4⁺ "helper" T cells (Th) (3). Resident dendritic cells (DCs) located in the descending airway epithelium are thought to carry antigen to the draining lymph nodes, where it is presented to antigen-specific T cells (11, 12). Exposure to aerosolized antigen appears to favor the generation of Th2 cells that on activation secrete cytokines, predominantly interleukin 4 (IL-4), IL-5, and IL-13, which support the induction of humoral immunity (3). The early response to aerosolized allergen leads to a transient rise in allergen-specific IgE that is attenuated by continual antigen exposure (1, 2, 7). This phenomenon is referred to as inhalation tolerance and early studies suggested that antigen-specific CD8⁺ T cells play an important role in maintaining the tolerant state (1, 2). It was subsequently reported that these CD8⁺ T cells expressed the ⁺ T cell receptor (TCR) and their expansion was CD4⁺ T cell dependent (3, 13). These regulatory cells were extremely potent in adoptive transfer experiments and as few as 5,000 cells could induce IgE-specific tolerance (14). Other groups have also identified a role for CD8⁺ T cells after exposure to aerosolized proteins and these have been implicated in the control of airway hyperresponsiveness (AHR) through their capacity to secrete interferon  (IFN-) (15).

Naive CD4⁺ T cells default to a Th2 pattern of cytokine synthesis and this is consistent in studies of both rodent and human experimental systems (16). Therefore, on the basis of these findings it has been suggested that allergen exposure in early infancy leads to priming of CD4⁺ T cells that display a Th2 phenotype. This would support the production of allergen-specific IgE antibodies and, thus, predispose to allergic sensitization. However, on continual antigen exposure the CD4⁺ T cells switch their pattern of cytokine secretion from a Th2 to Th1 phenotype in a process termed immune deviation (3). However, the role of immune deviation in controlling allergen-specific IgE synthesis does not appear to be a unifying process. Several groups have failed to find evidence of immune deviation or a role for CD8⁺ T cells and IFN- secretion in the downregulation of IgE synthesis after immune responses to inhaled antigen (7, 8). Indeed, these more recent studies have indicated that the decrease in IgE synthesis in the rodent models is related to a functional inactivation by antigen-specific CD4⁺ T cells.

It is difficult to dismiss, however, the fact that CD8⁺ T cells may have some function in the control of immune responses to inhaled proteins. For example, in a model of autoimmune diabetes in nonobese diabetic (NOD) mice it was observed that aerosol delivery of insulin could protect susceptible mice from developing diabetes if delivered before the onset of clinical symptoms (17). The restoration of tolerance to insulin led to a reduction in both pancreatic islet pathology and disease incidence, with treated mice exhibiting increased circulating levels of antibodies to insulin and CD4⁺ T cells that were refractory to restimulation with the immunodominant T cell epitope derived from insulin chain B and to the autoantigen glutamic acid decarboxylase (GAD). Despite the absence of a proliferative response to antigen in vitro, T cells secreted IL-4 and IL-10 after stimulation with antigen. The ability of splenocytes from insulin-treated mice to suppress the adoptive transfer of diabetes to nondiabetic mice was mediated by CD8⁺ TCR⁺ cells (17). Thus two independent studies have identified a potential role for CD8⁺TCR⁺ cells after aerosolized delivery of protein antigen. How these CD8⁺ T cells become primed, and how they mediate their effects in vivo, are at present unclear and require further investigation. It raises the question of whether the TCR⁺ cells have evolved from a sentinel function at mucosal surfaces to one that allows them to interface with conventional TCR⁺ T cells and, thus, provide an efficient means of regulating immune responses at mucosal surfaces. Consistent with this idea is the finding by Lahn and co-workers demonstrating that CD8⁺⁺ T cells help protect mucosal surfaces from airway inflammation caused by ⁺ T cells (18). The function of CD8⁺⁺ T cells is not, however, completely clearcut, as there are contrasting findings showing that the ⁺ T cells can in fact exacerbate AHR in a murine model of allergic inflammation by secretion of cytokines such as IL-4 (19). It is hoped that further studies will resolve this issue.

Control of Responses to Intranasally Administered Peptide

Delivery of peptide or protein antigens intranasally is an effective way of inducing antigen-specific peripheral tolerance. Studies with house dust mite-derived antigens have revealed that intranasal delivery of a single immunodominant peptide from the allergen protein Der p 1 can induce tolerance not just to the peptide used in the treatment phase but to stimulation with the whole protein (20). The induction of tolerance to all epitopes on the antigen is a phenomenon termed linked suppression. It is dependent on the induction of regulatory CD4⁺ T cells specific for the inhaled peptide (Figure 1) but is observed only if the tolerant mice are immunized with the intact protein (21). The induction of tolerance to inhaled peptide is an active process that induces a strong immune transient CD4⁺ T cell response that peaks 4 d after the treatment schedule (22). Similar findings in terms of the strength and kinetics of the response and the nature of cytokines produced by the responding T cells to inhaled peptides have been published independently (23, 24). Thus, even though the induction of immune responses to the inhaled peptide bears many similarities to that induced during a productive immune response, the animals develop tolerance to the antigen. We have observed that while the house dust mite peptide can induce a primary antibody response, there is no switch to IgG subclasses, despite giving animals repeated intranasal challenges with peptide (21). This feature of peptide tolerance may reflect a lack of T cell help from the primed T cell population. Consistent with this idea is that T cells from tolerant mice show downregulation of secretion of all cytokines (Th1 and Th2) when stimulated with antigen and, moreover, the cells proliferate poorly in response to the antigen in vitro (20). Thus these cells may not produce enough of the relevant cytokines (IL-4, IL-6) in vivo to support B cell isotype switching and/or they may not be able to provide the right cognate signals (CD40L/CD40) to drive this process.

View larger version (22K): [in this window] [in a new window]   Figure 1.   Model of linked suppression. An animal is rendered tolerant via intranasal administration of peptide, which leads to induction of CD4+ Tr cells. On immunization of a tolerant animal with protein, the DC takes up the antigen and processes relevant epitopes. The Tr cell will be drawn to the surface of the APC at the time when naive T cells specific for other epitopes on the antigen are present. The Tr cell when activated may secrete inhibitory cytokines (e.g., IL-10, TGF-1) or may inhibit the growth of naive T cells through a contact-dependent mechanism.

Independent laboratories have reported that peptides derived from self-antigens can be used as tolerogens when delivered intranasally (25). These peptides can downregulate established disease or prevent the development of disease in naive animals. Because autoimmune diseases invariably involve immune responses generated to multiple antigens (i.e., known as determinant spreading), the successful abrogation of autoimmune diseases after intranasal tolerance mediated by a single peptide reflects the induction of strong regulatory control through a process termed bystander suppression. This phenomenon was first observed in studies of oral tolerance to self-antigens and reflects that action of a population of regulatory T cells generated against one antigen that can effectively control responses of T cells directed to unrelated antigens (29). Bystander suppression would require the copresentation of antigens specific for both the regulatory T cells and for the T cells specific for the irrelevant antigen. Bystander and linked suppression are commonly used by the immune system to control peripheral immunity and this has led to the proposal of various mechanisms by which these functional outcomes can be mediated.

	WHAT ARE THE SIGNALS THAT CAN GIVE RISE TO REGULATORY T CELLS?

TOP ABSTRACT INTRODUCTION BALANCE BETWEEN Th1/Th2... WHAT ARE THE SIGNALS... REFERENCES

A common theme emerging from studies of mucosal tolerance, whether by the intranasal or oral route, is that tolerance is associated with the activation of regulatory CD4⁺ T cells (Tr cells) (30, 31). Studies of oral tolerance identified a unique population of CD4⁺ T cells that displays a restricted pattern of cytokine secretion when activated; these have been termed Th3 cells and they can secrete high levels of IL-4, IL-10, and transforming growth factor 1 (TGF-1) (32, 33) (Table 1). The cytostatic function of TGF-1 in lymphocyte proliferation may provide a mechanistic explanation of how Tr cells can mediate bystander suppression.

Accumulating evidence indicates that the secretion of Th2 cytokines is important in the regulation of immune responses in the gut, such that mice deficient in cytokines such as IL-2, IL-4 and IL-10, or mice deficient in TCR⁺ cells, are all susceptible to developing intestinal inflammation such as colitis (30, 31). Several groups studying mucosal immunity have identified a population of CD4⁺CD45RB^lo cells that are important in controlling intestinal inflammation (34). The induction of these Tr cells is dependent on the presence of TGF-1 and IL-10 because mice lacking either of these the CD4⁺CD45RB^lo cells fail to develop a regulatory function. Similarly, treating mice with neutralizing antibodies to TGF-1 or IL-10 can block the induction of Tr cells in vivo and thus predispose susceptible animals to the development of colitis (35, 36). In support of these studies, Groux and co-workers have identified culture conditions that support the differentiation of CD4⁺ T cells into a Tr phenotype, termed Tr-1 cells (37). Expanding CD4⁺ T cells in vitro with anti-CD3/CD28 in the presence of IL-10 gave rise to a population of cells that display a regulatory function both in vitro and in vivo. The Tr-1 cells generated by in vitro culture protected susceptible BALB/c-SCID mice from developing colitis when adoptively transferred with a population of disease-inducing CD4⁺CD45RB^hi T cells. Furthermore, several groups have now identified a population of CD4⁺CD25⁺ T cells from the thymus and spleen that appears to exert immunosuppressive effects on other T cells both in vitro and in vivo through a cell contact-dependent process (31) (Table 2).

Studies of intranasal tolerance have also shown an important role for IL-10 in the induction of Tr cells (23). Neutralizing IL-10 could abrogate induction of tolerance to intranasally delivered peptide by preventing the development of peptide-specific Tr cells. Thus, taken together, accumulating data are beginning to support the view that the cytokines IL-10 and TGF-1 have an important influence on CD4⁺ T cell differentiation in vivo.

Notch Signaling in T Cells

The route, dose, and context of antigen administration are all known to influence the subsequent outcome of immune responses to antigens as illustrated by our own studies of intranasal tolerance mediated by Der p 1 peptides. The immunodominant Der p 1 peptide could elicit two distinct functional outcomes depending on the route of antigen administration in naive animals. Delivering the peptide in saline intranasally induced profound T cell tolerance through induction of Tr cells, whereas the same peptide delivered in adjuvant evoked productive immunity (38). Thus it appears that naive T cells have the capacity to undergo alternate cell fate decisions in the periphery after antigen recognition (Figure 2). Because the recognition of antigen by T cells is antigen-presenting cell (APC) dependent, signals delivered from the APC to the naive T cell must be critical in influencing the differentiation of T cells. It is possible that lymphocytes, like other cells of the body, are guided by growth control mechanisms that operate in other tissues in embryonic and/or adult life. A major signaling pathway that influences binary cell fate decision processes is that regulated by Notch (39). The Notch pathway is evolutionarily conserved and functions in regulating cell fate decisions. Through its ability to regulate cell growth and differentiation it therefore has the potential to regulate lymphocyte cell fate decisions. The first human Notch1 gene was isolated from a T cell leukemia (TAN-1) and since then there has been growing interest in the role Notch signaling plays in the immune system (42). At present Notch has been shown to have a role in hematopoietic development and appears to play a role at several stages during thymocyte development (43). Notch was first identified in Drosophila as a neurogenic gene and functions as a cell surface receptor that can bind to its ligands Delta and Serrate (39). In Drosophila, Notch has been associated with two different signaling processes: inductive signaling (e.g., limb development) and lateral inhibition (neurogenesis; see below). Four different Notch receptors are found in vertebrates together with two Delta and two Serrate (i.e., Jagged) homologs. Cells of the peripheral immune system express all these genes (39).

View larger version (26K): [in this window] [in a new window]   Figure 2.   Cell fate decisions for T cells. Depending on the signals received by the naive CD4+ T cell from the APC it may respond in one of two ways. Activation by recognition of peptide-MHC and costimulatory signals on the APC can lead to its differentiation as a Th cell, enabling it to participate in generating a productive immune response. Alternatively, the naive T cell may receive an additional signal through Notch from the APC expressing a Notch ligand. This would enable the cell to differentiate into a Tr cell and thus participate in suppressing immune responses to an antigen, giving rise to peripheral tolerance.

Notch Signaling in Peripheral T Cell Tolerance

Notch signaling might be important in the regulation of peripheral immunity because it may influence the ability of naive CD4⁺ T cells to choose between immunity and tolerance (Figure 2). We suggest that after intranasal delivery of high-dose peptide the naive T cell would receive TCR-derived and costimulatory signals together with a signal through Notch. The reception of signals through the TCR and costimulatory molecules together with ligation of a Notch receptor would enable naive T cells subsequently to differentiate as Tr cells. Moreover, it is possible that the induction of "linked suppression" observed in intranasal tolerance may arise through a cell-cell contact mechanism whereby Tr cells would directly interact with naive T cells, clustered around an APC at the time of antigen presentation. In this situation the Tr cell would begin to express a Notch ligand when reactivated, and through a cognate interaction with neighboring T cells clustered at the same APC interface, would signal to inhibit the growth of naive cells. Thus the effector activity of Tr cells could be independent of the secretion of inhibitory cytokines. Thus this process of linked suppression may be akin to the function of Notch in lateral inhibition as described previously in neurogenesis in Drosophila (46) (Figure 3). This is a process whereby neural progenitors are selected from a field of equipotential precursor cells (Figure 3). Although all cells within a field have the same developmental potential and express the Notch receptor, only one will be selected to become a neuron. This process is initiated through stochastic fluctuation in the expression of proneural genes (achaete/scute), which enables the neural progenitor cell to express the Notch ligand Delta at the cell surface. This results in an increase in expression of Notch on the surrounding cells, allowing the neural precursor to "send" the signal through the Notch ligand, while the neighboring cells "receive" the signal and subsequently diminish Notch ligand expression and upregulate expression of the Notch receptor. This process eventually gives rise to a mature neuron while the remaining cells are kept in an undifferentiated state as epidermal cells.

View larger version (17K): [in this window] [in a new window]   Figure 3.   Notch signaling and lateral inhibition. Lateral inhibition in neurogenesis allows the selection of a neural precursor from a field of equipotential cells (homotypic interactions). The neural progenitor begins to express the Notch ligand Delta and can send a signal via binding to Notch to inhibit other cells from proceeding to the same cell fate. In a manner akin to this, we hypothesize that the regulatory T cell, when reactivated by the APC, would begin to express a Notch ligand (Delta/Serrate) that could bind Notch on a neighboring cell. The consequence of such an interaction would be inhibition of growth of the naive T cell, giving rise to linked suppression. This model would not be dependent on the secretion of inhibitory cytokines by Tr cells to explain this phenomenon.

A Role for Notch Signaling in Peripheral Immune Tolerance

Splenic and lymph node DCs, B cells, as well as CD4⁺ and CD8⁺ T cells all express transcripts for Notch1, Notch2, and the ligands Delta1 and Serrate1. We observed that the Notch ligands Delta1 and Serrate1 are differentially expressed on CD4⁺ T cells during the induction phase of tolerance with intranasally administered peptide (47, 48). However, the expression of these genes is either unaltered or are downregulated on CD4⁺ T cells after immunization with the same peptide in adjuvant (Corsin-Jimenez, M., and J. R. Lamb, manuscript in preparation). Dendritic cells are regarded as the principal APCs of the immune system and, having observed expression of Serrate1 transcripts on freshly isolated DCs from mice, we were prompted to assess the potential function of this gene in peripheral DCs. To do this we used retroviral gene-mediated transfer to overexpress the Serrate1 gene in spleen-derived DCs. These studies revealed that presentation of antigen by Serrate1⁺ DCs could induce profound peripheral T cell tolerance (48). The induction of tolerance in vivo was characterized by a decrease in antigen-specific T cell proliferation and cytokine production and the development of linked suppression. Furthermore, tolerance could be transferred to naive recipient mice, suggesting that Serrate1-induced Notch signaling had induced CD4⁺ T cells to differentiate as Tr cells in vivo (48).

As mentioned above, IL-10 has been identified as a key regulator for the differentiation of CD4⁺ T cells into Tr cells in vitro and in vivo. At the present time it is not known whether Tr cells secrete IL-10 after Notch receptor-ligand interactions with APCs. However, it is possible that Notch may function synergistically with IL-10 to promote Tr cell differentiation in vivo. Findings arising from the study of Notch function in the thymus have revealed that one important function of Notch is to protect T cells from death by neglect, and from activation-induced cell death (AICD). This observation comes from the finding that the intracellular domain of activated Notch1 can bind to the nuclear steroid hormone receptor, Nur-77, which is an important effector in thymocyte apoptosis (49). Constitutive expression of activated Notch can protect thymocytes from glucocorticoid-induced apoptosis and can protect T cell hybridomas from AICD.

If the Notch signaling pathway were to have a function in the peripheral immune system similar to that observed in the thymus, then this pathway may have an important role during a temporal period of T cell activation that would protect the responding cell from AICD. This mechanism may provide a window of opportunity for the responding cell to receive and respond to extracellular cues (e.g., cytokines) and thus allow the cell to differentiate appropriately. When antigens are presented in the absence of inflammatory stimuli (e.g., through mucosal surfaces) Notch signaling on naive T cells would lead to the induction of Tr cells (Figure 2).

Although we have addressed the function of Serrate1 on DCs it will be interesting to explore the effect of this ligand on the effector function of other cell types of the immune system. It might be anticipated that Notch ligand signaling may have multiple effects through the course of an immune response, and the temporal expression and selective use of different Notch ligand genes interacting with specific Notch receptors may guide the functional outcomes. For example, other situations in which binary cell fate decisions occur in the peripheral immune system are in the differentiation of Th1 versus Th2 cells and in the decision between effector versus memory cell differentiation for both T and B cells.

The discovery of a family of evolutionarily conserved pattern recognition receptors belonging to the Toll-like receptor family has uncovered another level of interactions between the innate and adaptive immune systems (50). The conservation of the Notch and Toll-like receptor families over millions of years of evolution is unlikely to be an accident. Rather, these signaling pathways control fundamental biological processes and thus the cross-talk between them may have important consequences in how lymphocytes will respond to an antigen. Thus as our knowledge of the interactions between the innate and adaptive immune systems advances, we will begin to understand the pathophysiology of allergen-induced pulmonary inflammation.