View larger version (28K): [in this window] [in a new window]   Figure 1. Asthma hospitalization rates per 1000 children in the United States who are 0 to 17 years old. Symbols represent observed rates; lines represent predicted trends based on log-linear regression analysis. From the National Center for Health Statistics, National Hospital Discharge Survey. Reproduced from Gergen and Weiss [4], with permission from the publisher.

Researchers estimate that in 1990, direct medical expenditures for asthma, such as those for medications and inpatient and outpatient hospitalization, totaled $3.6 billion, and that indirect costs, such as those for time lost from work, totaled $2.6 billion [7]. Because more than 10 million persons in the United States may have asthma [8], clearly the disease is worthy of our attention. More important, in the past decade, our knowledge has increased dramatically and we now understand that asthma is really an inflammatory disease of the airways. Although theories may differ, investigators seem to be focusing on studies to understand the mechanisms by which inflammation occurs. Whether it is specific and immunologically induced or nonspecifically induced, inflammation is a primary factor. Progress in understanding the pathways by which antigens and other stimuli cause their ill effects presumably will lead to a specific therapy and to the means to prevent asthma.

This conference provides current information on specific immunoregulatory elements (cytokine production and regulation) and mediators produced from mast cell interaction, reviews the modern approach to treatment, and discusses a planned intervention for inner-city children with asthma.

Determination of the Lymphokine-Producing Phenotype of CD4+ T Cells: A Potential Control Point in Allergic and Infectious Diseases

Dr. William E. Paul (Laboratory of Immunology, NIAID, NIH): I describe some of the critical regulatory steps that determine whether an immune response will be dominated by IgE production and therefore presumably contribute to allergic and asthmatic diseases or, alternatively, will be mediated by the production of lymphokines such as interferon- and will be particularly valuable in eliminating intracellular pathogens such as Mycobacterium tuberculosis. These distinct types of responses primarily represent alternative outcomes of an encounter with an immunogen.

Recently recognized sets of lymphokines play the central role in the development of a response dominated by IgE production. Allergic diseases are characterized by production of immunoglobulins of the particular isotype that elicits an array of inflammatory products that cause allergic symptoms. Further, in the striking IgE responses that occur in parasitic infections, lymphokine interleukin-4 (IL-4) production is required. A simple experiment illustrates this point. Bagg albino/c is the gene for albinism in BALB/c mice infected with the helminthic parasite Nippostrongylus brasiliensis. Thirteen days after infection with the parasite, these mice were bled and their serum concentrations of total IgE were measured (Figure 2). In a well-maintained animal colony, BALB/c mice have very low levels of IgE; in this case they were less than 0.1 µg/mL. In mice infected with N. brasiliensis, the increase in serum IgE concentration is spectacular; in this case it rose to 33 µg/mL, a more than 300-fold increase. Treating these animals with a control monoclonal antibody had no effect. If, however, they are treated with an antibody that can neutralize IL-4, that striking increase in IgE production is completely inhibited [9].

View larger version (30K): [in this window] [in a new window]   Figure 2. Treatment with anti-interleukin-4 (IL-4) antibody prevents increases in serum IgE in mice infected with Nippostrongylus brasiliensis. Groups of five BALB/c mice were infected (or not infected) with 600 third-stage N. brasiliensis larvae and treated with varying amounts of ascitic fluid containing monoclonal anti-IL-4 antibody, during inoculation and 7 days later. Thirteen days after infection, the mice were bled and serum IgE levels were measured. Data adapted from Finkleman and associates [9], with permission from the publisher.

Therefore, IL-4 must be available if an IgE response is to develop. By inference, immune responses that produce large amounts of IL-4 would be associated with development of responses dominated by IgE production. Responses in which IL-4 is not made will preclude production of IgE. These results have been confirmed in recent studies with so-called "knock-out" mice, in which the IL-4 gene has been disabled, making the mouse unable to produce IL-4. Such mice make no IgE, either in resting states or after receiving various stimuli [10, 11].

We want to understand what determines whether an immune response will be dominated by the production of T cells that make IL-4. First, we must know that IL-4 and its congener lymphokines, of which there are a growing number, are largely produced by the subset of T cells designated CD4⁺ cells. Second, the naive cell's immediate response to stimulation either by its cognate antigen or by polyclonal stimulants is to produce IL-2 but very little else [12]. We will concentrate on two critical lymphokines, IL-4 (the lymphokine necessary for IgE production and important in all forms of antibody production) and interferon- (the principal regulator of cellular immunity). Neither of these molecules is produced in any substantial amounts by naive T cells. However, naive T cells can differentiate in such a way that they can produce either, or sometimes both, of these molecules. In the last several years, investigators determined that the differentiation of CD4⁺ cells is primarily polar, either to cells that produce largely IL-4 and a set of associated molecules or to cells that produce largely interferon- as their dominant cytokine, but that also make IL-2. The former are TH₂-like cells and the latter are TH₁-like cells [13].

We have been interested in studying what determines whether a given immune response will be dominated by the differentiation of CD4⁺ cells into IL-4 producers, therefore facilitating the development of allergic disorders or, alternatively, into a response dominated by interferon- production. In the simplest approach, we would take naive T cells that will make only IL-2 when challenged, prime them in vitro, and study that which regulates that priming. The difficulty is that the frequency of T cells specific for any individual antigen in a healthy human or mouse is so low that such an experiment is very difficult to perform. The recent availability of mice transgenic for genes specifying T-cell receptor and ß chains provides a system in which this experiment can be done. In this mouse, all of the CD4⁺ T cells have the same receptor. Therefore, it is simple to follow the process by which these T cells change from naive cells that can produce only IL-2 into highly differentiated cells that can produce IL-4, interferon-, or both.

We developed a system involving a "priming culture" meant to mimic the situation in which cells first encounter their antigen, and a "challenge culture," in which the primed cells are restimulated with antigen and their lymphokine-producing potential can be evaluated. We relied on transgenic mice whose T cells bear a receptor specific for a peptide fragment derived from the protein cytochrome c [14]. The T-cell receptor recognizes the peptide bound to an allelic form of a class II major histocompatibility complex molecule. Naive cells are derived from these mice and cultured with antigen-presenting cells. This step is essential because T cells recognize antigen only on the surface of specialized antigen-presenting cells that express both the peptide derived from the particular antigen and a class II major histocompatibility complex molecule. We can vary the type of antigen-presenting cell, the amount of the antigen used, and, most importantly, lymphokines or cytokines themselves. Although various lymphokines are measured, our primary interest is in production of IL-4, interferon-, and, to some extent, IL-2.

An experiment described in Figure 3 shows one of the most striking and, in a sense, unanticipated results. T cells from a naive, transgenic donor were primed in vitro for 4 days with dendritic or antigen-presenting cells, a peptide derived from cytochrome c, and IL-2. When these cells were restimulated 4 days later, the result was clear and reproducible. No IL-4 was produced, but substantial amounts of interferon- were made [14]. Thus, cells primed under these conditions tend to produce interferon- and no IL-4. If a source of IL-4 is introduced into the priming culture, however, the opposite result is achieved. The priming for interferon- production is completely suppressed, and striking priming occurs for IL-4 production. Thus, an autocatalytic effect occurs whereby IL-4, which determines the production of IgE, is also critical in determining the differentiation of naive T cells into cells that can produce IL-4.

View larger version (24K): [in this window] [in a new window]   Figure 3. Interleukin-4 (IL-4) is required for priming for IL-4 production and inhibits priming for interferon- production. CD4+ T cells (106) from mice transgenic for the and ß chains of T-cell receptors specific for a cytochrome C peptide (88-104), in association with I-Ek, were cultured for 4 days with dendritic cells (DC) from B10.A mice (2.5 x 104), peptide 88-104 (0.1 µmol/L), IL-2 (10 U/mL) with or without IL-4 (1000 U/mL). The cells were then washed and restimulated with peptide and dendritic cells. Interleukin-4 and interferon- concentrations were measured 36 hours later. 11B11 is a specific monoclonal anti-IL-4 antibody. Adapted from Seder and coworkers [14], with permission from the publisher.

A secondary effect is also important. Interferon- can diminish priming for IL-4 production even in the presence of IL-4, although usually only when suboptimal amounts of IL-4 are present. Thus, if a limited amount of IL-4 (about 0.4 x 10^–11 mol/L) is added to the priming culture, measurable but modest priming for IL-4 production ensues. This priming is almost completely abrogated by adding interferon- to the culture. These results imply that two factors are critical in determining whether IL-4-producing cells will appear in culture: Interleukin-4 is essential for such priming, and interferon- can regulate or diminish the priming. Thus, the interplay of these two cytokines, critical effectors of the system, will also determine whether cells that can make IL-4 will be produced. That finding leads us to an initial conclusion. For a naive CD4⁺ T cell to develop into an IL-4 producer, IL-4 must be present, and interferon- diminishes this priming, particularly when IL-4 is not present in optimal amounts at the outset.

We are still uncertain about how T cells develop into interferon- producers. One possibility considered initially was that if cells were not positively stimulated to become IL-4 producers, the "default" condition would be that they developed into interferon- producers; that is, the production of interferon- is the normal, natural state of development and, unless altered, is the anticipated outcome. Additional results have indicated that the situation is more complex. Recently, a product of activated macrophages, called IL-12, was described [15]. In contrast to IL-4, which is largely a product of T cells, basophils, and mast cells, IL-12 is made by a series of cell types. An important source is macrophages that have been recently activated by infection with intracellular parasites. If IL-12 is added to the culture in which T-cell receptor transgenic mouse cells are primed, the degree of priming for interferon- is strikingly enhanced. This increase is approximately 10-fold [16, 17]. Although IL-12 dramatically enhances priming for interferon- production, it is not essential. Our work indicates that T cells cultured in the presence of neutralizing anti-IL-12 antibodies will display measurable, although modest, priming for interferon- production but no priming for IL-4 [17]. Although IL-12 is not essential for priming for interferon- production, IL-4 appears to be essential for priming for IL-4 production.

My final point addresses mutual regulation of priming. Earlier results showed that if IL-4 was added to a priming culture lacking IL-12, not only was priming for IL-4 production induced but priming for interferon- production was fully suppressed. What is the effect of a mixture of IL-4 and IL-12 in the system? Which of these two will dominate in these priming cultures? If IL-4 and IL-12 exist simultaneously, a response with mixed properties will ensue. Thus, no priming for interferon- production occurs when naive T cells are stimulated with peptide, antigen-presenting cells, and IL-4, with no IL-12 added. As IL-12 is titrated into the culture, priming for interferon- production increases, but no diminution in priming for IL-4 production occurs. With IL-12 and IL-4, a mixed response occurs. Indeed, mixed responses are almost certainly the normal physiologic outcome of most types of environmental immunization. Responses dominated by IL-4 or interferon- production probably arise from very polarized stimulation or, alternatively, from genetic predispositions to make responses largely dominated by one lymphokine or the other.

One such genetic predisposition that has a striking effect in experimental animals and may have a correlate in humans is an excellent example. It comes from the study of infection with the protozoan parasite Leishmania major. Most strains of mice, if infected with L. major, contract a transient disease—the equivalent of a delayed hypersensitivity response—characterized by swelling of the infected area. The infection resolves and the animal is cured. In these mice, T-cell responses to Leishmania are dominated by interferon- production; very little IL-4 is made [18]. But in some strains, especially BALB/c mice, a vigorous immune response is dominated by IL-4 production. These mice, rather than having a transient infection, develop a progressive infection and finally die of the disease. Thus, their "bad judgment" in making an IL-4-dominated response leads to their deaths.

We know this distinction is based on genetics because we can cross these mice to show a degree of genetic control of these alternative outcomes. The gene or genes that determine the production of very large amounts of IL-4 in response to this immunization are not known, but identifying them could be useful. It would not be surprising if the same genetic factor helps determine the dominance of IL-4 production in humans. Determining the nature of the gene or genes may reveal why some persons have responses dominated by IL-4 and thus develop allergies and asthmatic diseases.

Mediators Relevant in Asthma and Allergic Diseases

Dr. Dean D. Metcalfe (Allergic Diseases Section, Laboratory of Clinical Investigation, NIAID, NIH): Mast cells and eosinophils are the two principal inflammatory cells associated with asthma, and both promote various mediators. To develop rational intervention strategies, we must understand how the mediators function. Many mediators are involved in asthma, and they can be pleiomorphic and redundant.

Mast Cells

Although mast cells are present throughout the lung parenchyma, they are concentrated along blood vessels and epithelial basement membranes [19]. They are also identified in bronchoalveolar fluid. The surface membrane is often interdigitated with other cells, such as stromal cells. Each mast cell has about 1000 granules containing preformed mediators (histamine, proteases) that are instantly released when mast cells are exposed to certain stimuli, primarily those that are IgE mediated. In the past few years, researchers confirmed that mast cells are hematopoietically derived, arising from cells known to be CD34⁺ within human bone marrow [20] (Figure 4). Initially their development requires IL-3 [21], and mast cell precursors circulate through the blood and lymphatic systems. These precursor cells are also CD34⁺ and Fc_eRI and are not granulated. They resemble large lymphocytes. Then, through some stimulus, mast cell precursors are drawn into specific tissues, where they develop into one of two mast cell phenotypes. The mucosal phenotype is most common in the lungs [22], and the connective tissue phenotype, which is heparin-rich [23], stains more easily, and has a different array of mediators, is most common in connective tissues.

View larger version (44K): [in this window] [in a new window]   Figure 4. Mast cell development. Human mast cells arise from CD34+ pluripotential progenitor cells under the influence of interleukin-3 (IL-3), stem cell factor (SCF), and local tissue factors. G-CSF = granulocyte colony-stimulating factor; GM-CSF = granulocyte macrophage colony-stimulating factor; IFN- = interferon-; FceRI = high-affinity IgE[4 receptor; T Mast Cell = mucosal mast cell; TC Mast Cell = connective tissue mast cell.

The mucosal and connective tissue phenotypes may be separated based on ultrastructural anatomy. Mucosal mast cells tend to have scroll patterns in their granules [24], and protease distribution is characteristic. Mucosal mast cells contain tryptase, whereas connective tissue mast cells contain both tryptase and chymotryptase. The important observation is that mast cells, depending on which population is activated, release a different pattern of mediators that may affect local inflammation.

Mast cells target specific tissues after activation by IgE-mediated stimuli or growth factors, such as stem cell factor, which is a growth and maturation factor for mast cells [21]. Activated mast cells express certain receptors for components of matrix, such as laminin, vitronectin, and fibronectin [25]. These receptors allow mast cells to target such sites as the human lung. Indeed, when the lung is inflamed, mast cell numbers increase slightly within the tissues associated with the inflammatory response [26]. The same principle, of course, targets other inflammatory cells to the site of inflammation. In fact, the proximity of mast cells to blood vessels is important in the pathogenesis of asthma and in the initiation of inflammation by mast cells with mediators that alter vessel permeability and thus cause edema and other tissue changes.

Several stimuli in addition to those requiring antigen-specific IgE degranulate mast cells. Because asthma is often an allergic disease, the most common stimulus to degranulate mast cells and introduce inflammation in this context is IgE mediated. But other stimuli, including C3a, C5a, cytokines, and neuropeptides, also activate mast cells. Some of these are components of environmental stimuli that can activate, and perhaps even recruit, mast cells through a different mechanism.

As noted previously, mast cells produce various mediators [27], which can be divided into three groups: secretory granule preformed mediators, cytokines, and lipid mediators (Table 1). Of the secretory granule preformed mediators, the most notable is histamine. Histamine release is one measure of mast cell activation. Proteoglycans help form the granule matrix. Heparin is an anticoagulant and has a protease-stabilizing function in that certain proteases are held in an active form when bound to proteoglycan [28]. The proteases also have biological functions. We now know that at least some proteases induce the secretion of mucus [29] within the lung and increase the reactivity of smooth muscle [30].

Cytokines with many diverse biological functions are also produced by mast cells [27]. Several cytokines are produced by murine mast cells, including tumor necrosis factor- (TNF-), IL-1, IL-3, IL-4, and IL-6 [27]. Human mast cells have been reported to produce IL-4, IL-5, IL-6, and TNF- [31-33]. The third group, leukotrienes, are produced in tissues after mast cell degranulation and are important mediators of inflammation [34]. Leukotrienes alter vascular permeability and cause changes in blood vessel permeability that lead to tissue edema. Platelet-activating factor is another lipid-derived mediator that can increase the response of airways' smooth muscle and alter the permeability and tone of blood vessels [35].

Investigators recently reported the presence of endothelin, both message and product, in murine mast cells [36]. This finding adds yet another mediator with various biological functions to the mediators produced by mast cells. Endothelin has not yet been shown to be produced by human mast cells.

Mast cell activation leads to both microvascular and cellular responses, causing contractions and rapid edema of smooth muscles and mobilizing other inflammatory cells to enter into the tissues (Figure 5). In asthma, inflammation is particularly associated with the eosinophil, and responses are organ specific. In the lung, for example, bronchospasm and secretion of mucus are unique, important components of the inflammatory response. Activated mast cells set in motion what appears to be a biphasic inflammatory response. Degranulation of mast cells results in release of histamine, synthesis of leukotrienes, and release and synthesis of cytokines such as TNF- [31], which collectively cause immediate inflammation. Thus, if a person inhales a pollen to which he or she is sensitive and if the lungs are predisposed to react, immediate wheezing occurs. This reaction is followed by the ingress of other cell types and the release of further cytokines. The result is tissue inflammation of varying degrees, a late-phase reaction. If repeated, these events may cause chronic lung inflammation.

View larger version (21K): [in this window] [in a new window]   Figure 5. Mast cell-dependent immediate and late-phase reactions. Ag = antigen.

The clinical correlate of these mast cell-initiated events is the biphasic response in the lung [37]. Thus, if you challenge a sensitized person with antigen, there is an immediate decrease in lung function as measured by forced expiratory volume in 1 second (FEV₁), a recovery, and then a decrease in FEV₁ occurring 4 to 8 hours later, a so-called late-phase reaction. If certain anti-inflammatory compounds, such as disodium cromoglycate, are given, this biphasic response can be prevented experimentally after bronchus provocation with allergen. Corticosteroids do not ablate the early response but do ablate the late one.

In asthma, there is a substantial redundancy of mediators that cause specific pathologic problems. For instance, histamine, leukotrienes, and platelet-activating factor cause tissue edema. In every case, many effects observed in asthma may be related to more than one mediator, usually belonging to different classes. Thus, in general and in developing therapies to treat asthma, we cannot think in terms of blocking the action of a single compound and thus preventing inflammation. A more broadly based conceptual approach is usually required. Although some mediators seem to be important and although there may be a measurable value in blocking a specific mediator, blocking the action of one mediator alone is not sufficient.

Certain mediators are associated in specific patterns with immediate and late reactions. For instance, the late phase in nasal tissues is associated with both leukotriene C₄ (LTC₄) and histamine but not prostaglandin D₂, which is produced by mast cells [38]. This suggests that basophils have a function in the late inflammatory response. It does not appear, however, that basophils are the major cell responsible for inducing asthma, although a few are found in chronically inflamed pulmonary tissues.

Eosinophils

Eosinophils, like mast cells, may be activated by multiple stimuli [39] (Figure 6). Several growth factors prime eosinophils [40], which may then release several preformed components, including eosinophil peroxidase and major basic protein, causing problems such as toxicity to epithelial cells in pulmonary tissues [41]. Thus, like mast cells, eosinophils show several biological effects that can modulate inflammatory cellular reactions and promote further inflammation. The preformed mediators in eosinophils include several proteins, such as major basic protein, which is cytotoxic to epithelial cells; both eosinophils and major basic protein have been identified in areas of bronchial denudation. Major basic protein may also be recovered from bronchoalveolar lavage fluid in persons with symptomatic asthma [42]. This clinical evidence supports the concept that the eosinophil has major importance in causing inflammation in asthma.

View larger version (23K): [in this window] [in a new window]   Figure 6. Eosinophil priming and activating stimuli and resultant mediators of inflammation. IL-3 = interleukin-3; GM-CSF = granulocyte macrophage colony-stimulating factor; IL-5 = interleukin-5; C3b = complement protein 3b; IgG = immunoglobulin G; IgA = immunoglobulin A; PAF = platelet-activating factor; EGF's = epidermal growth factors; MBP = major basic protein; ECP = eosinophil cationic protein; EPO = eosinophil peroxidase; EDN = eosinophil-derived neurotoxin; LTC4 = leukotriene C4; PGE2 = prostaglandin E2; O2 Radicals = oxygen radicals; TGF- = transforming growth factor-; TGF-ß = transforming growth factor-ß; TNF- = tumor necrosis factor-; MIP-1 = macrophage inflammatory peptide-1 .

Eosinophils also synthesize other mediators, including metabolites of arachidonic acid, particularly LTC₄ [43]. Eosinophils can thus contribute to bronchoconstriction, mucus secretion, and changes in vasopermeability. Again, the common theme is that many of these cells have redundant effects. For instance, platelet-activating factor, which is a potent bronchoconstrictor and a chemoattractant for eosinophils, is also synthesized by eosinophils, which is perhaps more important in terms of eosinophil than mast cell function [44]. As might be expected, eosinophils also have receptors for and make cytokines. Interleukin-5 and IL-3 receptors are present on eosinophils, as are granulocyte macrophage colony-stimulating factor receptors [45]. Cytokine messenger RNAs (mRNAs) are found in eosinophils as well [46-49]. The percentage of eosinophils positive for mRNA and for IL-5 is high, and so eosinophils appear to contribute to inflammation through cytokine generation.

Other Cells

Other cells, including lymphocytes and neutrophils, are also involved in chronic lung inflammation. Convincing evidence exists for infiltration of inflamed pulmonary tissues with lymphocytes of the TH₂ profile, that is, those preferentially synthesizing IL-4 and IL-5. Indeed, lymphocyte responses may be an important determinant in the genesis of chronic asthma. Similarly, neutrophils clearly infiltrate inflamed pulmonary tissues and contribute to the immunopathologic basis of asthma.

A unifying concept of inflammation integrates this wealth of information about the number of cytokines, lipid-derived mediators, and preformed substances that can be released by many cell types. We have not even addressed the possibility of activating the kinin or the complement systems, fibrin deposition, or the damage caused by reactive free radicals. What appears is a type of inflammatory chaos generated after inflammation has been initiated and potentiated. It has several key components. First is the stimulus [50] (Figure 7); in terms of allergic asthma, it is an IgE-mediated stimulus following sensitization to an environmental allergen, usually pollen. This stimulus must activate some resident cell, often the mast cell. Second are other stimuli, such as direct toxicity, which may be produced by environmental pollutants on the epithelial cell in the airways. Third is recruitment of other inflammatory cells, including eosinophils, neutrophils, and lymphocytes. This process then becomes cyclic. The initial stimulus can be removed but chronic inflammation continues. The interruption of this process is discussed in the next section. Conceptually, it deals with removing the stimulus, preventing activation, or providing therapy that prevents this cycle of inflammation. Steroids, for example, are an effective strategy for asthma treatment because they target multiple cell types within the inflammatory response.

View larger version (39K): [in this window] [in a new window]   Figure 7. The inflammatory cycle. An inflammatory stimulus (chemical, physical, infectious, or other) activates a primary effector system that releases chemical mediators and initiates cell-independent effector systems (for example, complement) that directly produce inflammation. This activation also results in the recruitment of a secondary effector system that amplifies the inflammatory process. Anti-inflammatory systems (arrows with negative signs) act at several sites within the cycle to suppress and contain inflammation. Reproduced from Raphael and Metcalfe [50] with permission from the publisher.

Rationale and Current Approach to Treating Asthma

Dr. William W. Busse (University of Wisconsin, Madison, Wisconsin): In the past decade, new information emerged indicating that asthma is the consequence of elaborate interactions among inflammatory cells, mediators, cytokines, epithelium, and airway smooth muscle [51]. In this section, three areas are addressed: airway inflammation in asthma, the relevance of airway inflammation to paradigms of treatment, and evidence that anti-inflammatory therapy may influence not only physiologic components of asthma, such as airflow obstruction, but also pathophysiologic features of the disease.

Airway Inflammation in Asthma

Asthma is characterized by 1) airway obstruction that is reversible either spontaneously or as a consequence of treatments; 2) airway inflammation; and 3) airway hyper-responsiveness to various stimuli—methacholine, histamine, exercise, and cold air [52]. Bronchospasm, which is an important component of asthma and airflow obstruction, is only one feature of the pathophysiologic characteristics of this disease. Mucosal edema, mucus, and bronchial inflammation also contribute to bronchial narrowing. Airway inflammation in particular was identified as an essential feature of asthma and a possible link to a bronchial hyper-responsiveness [53]. For example, Beasley and colleagues [54] compared mucosal biopsy specimens from patients with allergic asthma with those of persons without asthma. The airway epithelium was denuded, collagen deposition under the basement membrane increased, and inflammatory cells in the airway epithelium, particularly eosinophils, were present. Table 2 summarizes the histopathologic features of asthma, even mild disease.

In addition to mast cells, which were reviewed by Dr. Metcalfe, mucosal eosinophilia is also prominent in asthma. Eosinophils are easily recognized by their prominent granules [55], which contain a series of basic proteins, such as major basic protein. Eosinophil major basic protein damages bronchial epithelium, promotes airway hyper-responsiveness, and contracts airway smooth muscle. When activated, eosinophils generate sulfidopeptide leukotrienes such as LTC₄, which can contract bronchial smooth muscle, cause vasodilation, and increase vascular permeability. Finally, the eosinophil can express and release various proinflammatory cytokines [55] to further potentiate and promote allergic inflammation.

The lymphocyte, a rich source of proinflammatory cytokines, has emerged as an important and perhaps pivotal cell in asthma [56, 57]. In addition to IL-4, other cytokines, such as IL-5, are produced by CD4⁺ lymphocytes; IL-5 promotes eosinophilopoesis and can prime eosinophils to enhance their inflammatory potential. Recent observations in patients strongly support the important role of lymphocytes in asthma. Hamid and colleagues [58] used immunohistochemical techniques and in situ hybridization on airway bronchial biopsy samples from patients with asthma. Bronchial tissue showed increased numbers of CD4⁺/CD25⁺ T lymphocytes, a marker of T-cell activation [59]. Further, mucosal eosinophils stained positively with an EG2 antibody, suggesting cell activation. Finally, IL-5 was expressed in airway cells, and the level of expression correlated with the number of activated lymphocytes and eosinophils. Robinson and associates [60] analyzed cells in bronchial lavage fluid for cell type and cytokine profile. A characteristic pattern emerged: Airway T cells from asthma patients had increased expression of mRNA for IL-2, IL-3, IL-4, and IL-5. Increased expression of IL-4 and IL-5, in particular, by these cells suggests a TH₂ phenotype in allergic asthma [61] and implies that the airway lymphocyte phenotype may be especially well suited to cause allergic inflammation in persons with asthma.

Changing Concepts in Asthma Treatment

The National Asthma Education Program developed guidelines, dividing asthma into three levels of severity—mild, moderate, and severe—based on symptoms and objective means of airflow obstruction. Because severity of asthma may change, specific treatment must be viewed as a dynamic process (Figure 8). Therapeutic compounds available for asthma treatment can be divided into two general classes, bronchodilators and anti-inflammatory agents (Table 3). Three classes of bronchodilators are available for asthma therapy: ß-adrenergic agonists, methylxanthines (theophylline), and anticholinergics, with the first two used most commonly. Beta-adrenergic agonists exert their effect by combining with ß₂ receptors on airway smooth muscle to cause relaxation. Although ß-agonists may modulate inflammatory cell function in vitro [62], there is little evidence that they exert such actions in vivo [63]. Nonetheless, they are an indispensable component of therapy for all patients. In the inhaled version, ß-agonists cause rapid bronchodilation and few side effects, and they are the preferred medication to treat acute asthma and will also prevent exercise-induced asthma [64].

View larger version (38K): [in this window] [in a new window]   Figure 8. Overview of asthma therapy. All therapy must include patient education about prevention, including environmental control when appropriate. PEFR = peak expiratory flow rate. Reproduced from the National Heart, Lung, and Blood Institute [52] with permission.

The usual recommendation of long-term daily administration of inhaled ß-agonists is being re-evaluated. Several recent studies have questioned whether regular use of inhaled ß-agonists may promote bronchial responsiveness, cause deterioration of asthma control [65], or increase deaths [66]. These concerns have not been fully resolved. However, most clinicians agree that excessive daily use of inhaled ß-agonists indicates poor asthma control and the need for additional therapy. Further, because ß-agonists are unlikely to modify bronchial inflammation, combination therapy with anti-inflammatory agents is essential.

Theophylline, like ß-agonists, is an effective bronchodilator. It can be administered orally and is effective for 12 to 24 hours [67]. Its mechanism of action has yet to be fully established, and its precise role in asthma therapy is also being re-evaluated [68]. Methylxanthines are used mainly in patients with moderate to severe asthma. They also have a special role in nocturnal asthma, possibly because their action peaks between midnight and 6 a.m. Corticosteroids are the most effective anti-inflammatory medications to treat asthma [69]. They modify many cell activities in the allergic inflammatory response: lymphocyte function, eosinophil activities, inflammatory cell migration, and leukotriene synthesis, to name a few [70]. The severity of asthma dictates whether corticosteroids are used parenterally, orally, or by aerosol. For severe episodes that require hospitalization (status asthmaticus), aggressive administration of corticosteroids is essential [64]. Although precise guidelines for optimal doses have not been determined, most clinicians recommend 0.5 to 1.0 mg/kg methylprednisolone (or equivalent) given intravenously every 6 hours. As the attack resolves, doses of parenteral corticosteroids are reduced and oral formulations are substituted.

Inhaled corticosteroids, in contrast, are the mainstay of treatment for patients with chronic, persistent asthma. Inhaled formulations provide safe, effective therapy for many patients. Juniper and colleagues [71] evaluated the long-term effect of inhaled corticosteroids. Although the patients being studied had mild disease, the researchers made several interesting observations. Those who received inhaled budesonide, 400 µg/d, had a gradual improvement in their asthma severity, with maximal effect after 9 months of use (Figure 9). Although the dose of budesonide was generally low (the usual dose is 800 to 1600 µg/d), the principle that continued use of anti-inflammatory therapy leads to progressive improvement is relevant to current concepts of therapy. In addition, patients receiving inhaled corticosteroids also had reduced bronchial responsiveness to methacholine (Figure 10). Changes in bronchial responsiveness to methacholine-induced airway smooth muscle contraction generally were small but consistent. The reduction in bronchial responsiveness was interpreted as an indication of the anti-inflammatory actions of corticosteroids. However, budesonide did not improve bronchial hyper-responsiveness to the levels of participants without asthma.

View larger version (27K): [in this window] [in a new window]   Figure 9. Effect of budesonide on asthma questionnaire severity score. Closed circles represent budesonide, and open circles represent placebo (P < 0.009). S.E.M. = standard error of the mean. Reproduced from Juniper and colleagues [71] with permission from the publisher.

View larger version (28K): [in this window] [in a new window]   Figure 10. Effect of budesonide on airway responsiveness to methacholine expressed as a –2-foldto +4-fold change from baseline. Closed circles represent budesonide, and open circles represent placebo (P < 0.0001). PC20 = provocation concentration that causes a 20% decrease in forced expiratory volume in 1 second; S.E.M. = standard error of the mean. Reproduced from Juniper and colleagues [71] with permission from the publisher.

Djukanovic and colleagues [72] determined the effect of inhaled corticosteroids (beclomethasone) on asthma symptoms, pulmonary functions, and airway histopathologic features. In the patients studied, inhaled corticosteroid therapy improved lung function, decreased symptoms and the need for inhaled ß-agonists, and made airways less responsive. Further, bronchial biopsy specimens were obtained before and after corticosteroid therapy. Post-therapy biopsy samples showed fewer T cells in the submucosa and fewer mast cells and eosinophils in the submucosa but no change in the extent of mast cell and eosinophil degranulation. Thus, a link has begun to be established between the reduction in airway injury, albeit incomplete, and parameters of asthma severity.

Cromolyn sodium is a nonsteroidal anti-inflammatory compound that effectively treats some patients [73]. Although originally classified as a mast cell "stabilizer," cromolyn has other anti-inflammatory actions, but its mechanism of action must be determined. In clinical trials, cromolyn reduced asthma symptoms and diminished bronchial responsiveness [74]. Internists might find cromolyn useful as prophylaxis before animal exposure. Its effectiveness is greatest in children, and it is safe. Nedocromil sodium is a pyranoquinoline derivative that has many anti-inflammatory properties, including modification of the immediate and late airway response to inhaled antigen [75]. Nedocromil sodium was just released for clinical use. With further experience, a more precise determination of its usefulness in treating asthma will be established.

Rationale and Early Results from the NIAID National Cooperative Inner-City Asthma Study

Dr. Elena R. Reece (Allergy and Immunology Section, Howard University Medical Center, Washington, D.C.): Deaths due to asthma appear to be increasing worldwide. The death rate has increased from fewer than 2000 per year in the United States in 1978 to more than 4500 per year [76]. Why this is happening, given our understanding of the inflammatory nature of asthma and modern, effective treatment, remains perplexing. Among the associated factors are suboptimal approaches to therapy combined with delay, by both patients and physicians, in recognizing the severity of the illness.

The burden of asthma on the U.S. population is also increasing; among the subpopulations at highest risk are racial and ethnic minorities that are both poor and residing in certain urban environments, particularly the children of these minorities [77]. In addition, recent studies have suggested that the high rates of death and hospitalization due to asthma are not evenly distributed among urban neighborhoods: Very high rates are concentrated in very small areas of high poverty, commonly called "the inner city" [78].

Despite mounting evidence that excessive asthma-related morbidity is strongly correlated with poverty and inner-city residence, few studies have helped elucidate the causal pathways through which these factors operate. In 1991, the NIAID launched the National Cooperative Inner-City Asthma Study. Eight institutions in seven cities (Baltimore, Cleveland, Chicago, Detroit, New York, St. Louis, and Washington, D.C.) are involved. This cooperative study was designed first to identify specific determinants of morbidity and death from asthma in this population, to understand their inter-relationships, and then, based on the findings, to develop and evaluate comprehensive therapeutic and environmental intervention strategies.

This study assumes that risk factors are identifiable, and that some can be modified and others probably cannot. It targets those that can. It also seeks to understand subjective factors within a person that help control asthma, specifically the information that persons have that allows them to control their asthma, their attitudes toward disease, their care givers, and their education.

Initially, many factors are being studied: the role of ambient allergen levels in the indoor environment and their identification; the role of allergen sensitization and exposure, which involves measuring a patient's reactivity with skin testing; the role of irritants, such as exposure to smoke, and the level of smoking in the household, as well as evidence that patients are exposed; psychosocial factors and their role; and the possible relation to over- or under-treatment. Although we will focus on the home environment, other areas clearly affect the patient, particularly the school environment. We are assembling the data on these factors and preparing to develop an intervention strategy that we hope will change the morbidity and mortality rates of asthma in these urban populations.