Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
c Markers of Inflammatory and Immune Response Inflammation in the respiratory tract can be caused by injury, immune response, or infection. All three routes of induc- tion of inflammation are of interest in studying the health effects of respirable pollutants. Inflammation due to injury from inhaled toxicants is a measure of the cytotoxicity of the pollutants; markers of the progress of the inflammation could potentially be used in a predictive fashion to detect the early stages of irreversible structural changes in the lung, such as fibrosis and emphysema. Many inhaled toxicants-such as isocyanates, cotton dust, and beryllium-induce an immune response that constitutes the major ad- verse health effect of exposure. The inflammatory response to infectious agents is not the topic of this report, but is of interest in relation to the potential for some inhaled pollutants to reduce the ability of the body to resist infection. In this chapter, we will discuss markers to detect and require the inflammation induced by all three types of agents. First is a discussion of markers of injury in- duced in the respiratory tract by inhaled toxic materials. That is followed by a discussion of the effect of pollutants on the infectivity of pathogens in the respiratory tract. Finally, a major portion of the chapter is devoted to a 91 discussion of the immune response of the lung, markers of this type of response and the use of memory cells of the immune system as markers of both exposure and adverse health effects in the respiratory tract. INFLAMMATORY RESPONSE TO INHALED TOXINS Epithelial cells and resident macro- phages in the respiratory tract are the points of first contact of the body with inhaled toxicants. The ensuing injury or death of those cells induces an in- flammatory response characterized by the release of cytoplasmic enzymes from the damaged or lysed cells and the recruitment of neutrophils to the site of injury. An increase in the permeability of the alveo- lar-capillary barrier is accompanied by the transudation of serum protein. Macro . . ~ phages may increase in number; it the toxic material is a particle, there will be a release of hydrolytic enzymes from the macrophages, either during phagocyto- sis, after lysis of the macrophage, or as an active secretory process. The inflammatory process provides many markers that can be used to detect and measure the response and to follow its progress toward resolution or chronic
92 inflammation. . . . Lactate dehydrogenase, a cytopiasm~c enzyme released from dam- aged or lysed cells, can be used as a marker of cytotoxicity. An increase in this en- zyme can be used to distinguish between toxic events and physiologic responses. The presence of neutrophils or increased serum protein in the epithelial lining fluid can be used as a marker of the inflam- mation induced by the injury. The activity of hydrolytic and proteolytic enzymes released by the phagocytic cells has been shown to correlate with the toxicity of particles (Beck et al., 1982; Henderson, 1988a,b). The advent of fiberoptic bronchoscopy has allowed sampling of fluids lining the respiratory tract for analyses of the above markers. The use of the technique is described fully in Chapter 6. The analy- sis of epithelial lining fluid (ELF) for markers of inflammation has some advan- tages over older methods of detecting in- flammation. First, a pulmonary inflamma- tory response can be detected with analysis of the bronchoalveolar fluids before it can be detected with radiography (Fahey et al., 1982~. In addition the use of the markers in the lining fluids allows meas- urement of the degree of inflammation. which is useful not only for determining the progress of an inflammatory response, but also for ranking inhaled compounds for toxicity in animal studies. In£1am- matory responses in the nose and in the upper respiratory tract can be detected by site-specific sampling of the lining fluids. The biochemical and cellular content of the ELF can also provide infor- mation on the type of inflammatory response and the stage of the disease process. Several examples of the use of ELF analysis for the detection of toxicant-induced respiratory tract inflammation are given in Chapter 6. The major uses of the markers has been in animal toxicity studies to rank a series of compounds for toxicity and to study the mechanisms of toxicant-induced lung disease. Studies in this field have been reviewed (Henderson, 1988a,b). MARKERS OF PULMONARY TOXICOLOGY INFLAMMATORY RESPONSE TO MICROBIAL INFECTIONS The respiratory tract constitutes the primary mammalian portal of entry for many pathogens. For some microbial infec- tions, the bronchopulmonary mucosa serves as a benign substrate for initial replica- tion events that lead to eventual systemic spread without producing any clinical disease locally. For other infectious agents, however, the respiratory tract is the principal target for the disease- producing potential. It is estimated that respiratory viruses are responsible for 5-6% of all deaths and about 60% of deaths related to respiratory disease (Ogre et al., 1984~. Ample evidence supports the conclusion that normal AMs can ingest and kill many types of microorganisms that gain en- trance to the respiratory tract (Jakab, 1984~. Some microorganisms, particularly the virulent intracellular parasites, can survive in normal macrophages; it is apparent that this class of parasite can be controlled only when the forces of ac- quired immunity orchestrate the macro- phage system into antimicrobial action that is more potent, both qualitatively and quantitatively. Various chemicals in the environment and workplace affect the immune response, as determined by one or more of the many tests available to measure various com- ponents of the immune system (Faith et al., 1980; Sharma, 1981; Luster et al., 1988~. Extensive evidence from animal lung infectivity models points to the det- rimental effects of air pollution on vari- ous defense mechanisms of the lung (Neiman et al., 1977; Fauci et al., 1984). Results of epidemiologic studies indicate that living in urban areas increases the in- cidence of airway infections (Hong, 1976; Penn, 1978), although direct correlations of infection with specific pollutants have not been found. In vivo and in vitro stud- ies, using mainly animal models, have shown that ozone impairs AM phagocytic function (U.S. Environmental Protection Agency, 1984~. The success of AMs in digesting organic particles and organisms depends on lysosomal hydrolytic enzymes, includ
INFLAMAL4TORYAND IMMUNE RESPONSE . sing acid phosphatase, cathepsins, lyso- zyme, beta- glucuronidase, beta- galac - tosidase, arylsulfatase, and beta-gluco- saminidase. Exposure to ozone in viva and in vitro depressed the intracellular activity of lysozyme, beta-glucuronidase, and acid phosphatase from rabbit AMs (Hurst etal., 1970~. Kimura and Goldstein ~ 1981 ~ demonstrat- ed a decrease in the bactericidal enzyme lysozyme in AMs from rabbits exposed to ozone at 0.25 ppm. That finding paralleled the increased susceptibility of those animals to various infectious agents (Coffin et al., 1968, Ehrlich et al., 1977; Miller et al., 1978~. Other products of the AMs that are important in phagocytosis and have antibacterial activity are su- peroxide anions (O2-), hydroxyl radical (OH.), and H2O2. In viva exposure of rats to ozone at 0.9-3.2 ppm resulted in a dose- dependent decrease in O2- production by AMs from the exposed animals (Amoruso et al., 1981; Witz et al., 1983~. Thus, the studies in animal models have shown that ozone depresses many components of AM function that are important in phagocy- tosis and in protecting the lung from microorganisms or particles present in the environment and that can be measured and therefore can serve as markers of exposure. In vitro studies can also be used to im- prove understanding of in viva phenomena. Phagocytosis is a well-organized process, made up of several integrated steps, many of which are adversely affected by ozone. McAllen et al. (1981) have noted that AMs obtained from rats after in viva exposure to ozone at 1 ppm exhibit decreased mobili- ty. AMs in ravage fluid from the lungs of rodents that had been briefly exposed to ozone at 0.5-1.0 ppm display a lower ability to engulf bacteria than AMs from control animals. However, if the animals are exposed to ozone at 0.8 ppm for a longer period before lung ravage, the ability to incorporate carbon-coated latex micro- spheres of isolated AMs is increased. The data suggest that phagocytic activity is impaired soon after exposure to ozone, but that AM function recovers if the insult persists. Whether those changes are due to the influx of unexposed AMs into the 93 lung or to adaptation by resident AMs is not clear. Several investigators have recently started to examine the effect of inhaled pollutants on human subjects' ability to resist infections. Frampton et al. (1987) exposed normal human subjects to NO2 in controlled chamber conditions. AMs were obtained from the subjects by BAL 3.5 hours after exposure and exposed in vitro to influenza virus. It was ob- served that the AMs obtained from subjects exposed to NO2 at 0.6 ppm continuously (3.5 hours) were able to inactivate the virus significantly less than those ob- tained from subjects exposed to clean air. No major changes in the cell numbers in the BAL fluid from the exposed subjects were observed, but several biochemical changes indicated that NO2 did induce in- flammation in the exposed subjects. In another study, Kulle et al. (1987), exposed normal human subjects to NO2 at different concentrations, and then a live attenuated cold-adapted influenza A virus was ad- ministered intranasally to all subjects. Infection was determined by virus recov- ery and a 4-fold or greater increase in antibody titer. The results suggest that subjects that were exposed to NO2 at 1 or 2 ppm for 2 hours/day for 3 days and inhaled the virus on day 2 had small, reproducible signs of infectivity. The approaches taken by the different investigators were dif- ferent-Frampton et al. exposed subjects to the pollutant in viva and studied infec- tivity in vitro, and Kulle et al. exposed subjects to both the pollutant and the infectious agent-but the results collec- tively suggest a decrease in host defense against the viral infection. In both studies, spirometrically measured pul- monary functions were unchanged by ex- posure to the pollutants. IMMUNE RESPONSE The human respiratory tract contains a complex array of host defenses-anatom- ic barriers, mucociliary clearance, phag- ocytic cells, and various components of cellular and humoral immunity-that col- lectively cleanse inhaled air and inac- tivate infectious and other injurious
94 agents that are inhaled (Reynolds, 1979; Reynolds and Merrill, 1981~. In particu- lar, the mucosal lining of the small air- ways and alveolar airspaces contains many components of the immune system that are important in providing protection of the normal lung. However, some of the compo- nents also play an important role in im- munologic lung disease. This section re- views the general features of the immune system and how it operates in the lung. Antigen-antibody complexes are the basis of immune response. The afferent phase of the immune response usually be- gins with antigen processing by phago- cytes, such as macrophages. That includes degradation of foreign substances and exposure of lymphocytes to antigens, which stimulates the production of anti- body, sensitized cells, or both (Figure 5- 1~. Interactions of macrophages stimu- lated by antigens with cells in lymphoid tissue result predominantly in a cellular or humoral immune response. Cellular im- mune responses (delayed hypersensitivity) are mediated by thymus-derived lympho- cytes-T cells. Antigen interaction with T cells usually leads to their prolifera- tion. It is now recognized that T-cell proliferation and the generation of effec- tor cells occur separately. Antigens interact with macrophages, and inter- leukin- 1 (I1- 1 ) stimulates resting T cells. The T cells then can respond to a growth factor called interleukin-2 (I1- 2~. Among the progeny of antigen-stim- ulated T cells are memory cells, which M'4R=RS OF PULMONARY TOXICOLOGY respond quickly to later challenge with the original antigen; killer cells (or natural killer cells, NK cells), which destroy alien cells; and effecter T cells, which produce molecules called lympho- kines. 11-2 signals T cells to produce more T cells and effecter T cells. Lympho- kines can play an important role in the generation of an inflammatory response, particularly one involving cell-mediated immunity. For example, initiation and development of granulomas are thought to arise from the secretion of lymphokines that influence macrophage motility, ac- tivation, and function. Humoral responses are the end result of antigen interaction with marrow-de- rived or bursar-cell-equivalent lympho- cytes (B cells). B-cell function is regu- lated by at least two subpopulations of T cells: helper T cells (Th cells) are re- quired for optimal production of anti- body to most antigens, and suppressor T cells (Ts cells) are required for inhibi- tion or modulation of the humoral re- sponse once it is initiated. T-cell subsets in humans have been shown to express distinct differentiation anti- gens, which can be identified with mono- clonal antibodies to T cells. At each step in the sequence of immune stimulation, immunocompetent cells are activated and liberate soluble mediators. For example, when activated by immune stimulation, macrophages can produce various potentially injurious agents, including arachidonic acid metabolites, ~> ~ Ceil Antigen 111/' \\ \~J immunity ~ ~ J ( Helper ~(Suppresso) ~\ ~ ~ Macrophage \ ~\ \~ ( ~Ant~body Plasma cell FIGURE ~1 Cellular interactions involved in generation of immune response. Antigen presentation leads to stimulation of T-cell or B-cell systems. Factors involved in T-cell system include interleukin-1, which stimulates T cells to acquire receptor for T-cell growth factor called interleukin-2 (I1-2~; same subpopulation of T cells can also secrete 11-2.
INFLAMMATORY AND IMMUNE RESPONSE free oxygen radicals, and growth factors capable of initiating abnormal growth and metabolism of fibroblasts. Similarly, activated lymphocytes can produce soluble mediators that can act on immunocompetent cells and other cells, including inter- feron-gamma capable of activating macro- phages and chemotactic factors capable of activating other phagocytic cells, particularly neutrophils. Thus, although immune stimulation is initiated by a speci- fic antigen, secondary inflammatory medi- ators can appear as the cascade of cells and inflammatory signals progresses. Many of the cells, particularly macrophages, can be stimulated by nonspecific agents, and the appearance of soluble mediators does not necessarily indicate that an im- mune response has been initiated. The potential use of some of the inflammatory mediators as specific biologic markers of injury must be viewed in light of the fact that they can be initiated by both specific and nonspecific stimuli. The pulmonary immune cells are hetero- geneous, and external stimuli, such as pollutants, can modulate their behavior both qualitatively and quantitatively. The changes can be used to assess effects of exposure. Characteristics that can be monitored, such as cell numbers and cell-surface markers, can be considered biologic markers. Some of the changes can be transient (reversible); others can be chronic or irreversible. Examples of modulation in the immune cells in the lung include changes in the proportions of sub- populations of immune cells (e.g., the TH/TS ratio), in the extent of expression (density) of cell-surface markers, in the cytolytic capacity of T cells or nat- ural killer cells, and in the ability of phagocytes to ingest particles. Various components of lung fluid that are associated with the inflammatory re- sponse can serve as markers. The fiberop- tic bronchoscope has made access to the trachea and major airways routine, and this versatile instrument has contributed enormously to the care and diagnosis of patients with lung diseases (Sackner et al., 1972~. Since the late 1960s, BAL has been incorporated into fiberoptic bron- choscopy and has proved to be safe and 95 reliable for sampling airway and alveolar fluid and cellular components in normal lungs (Reynolds and Newball, 1974) and diseased lungs (Reynolds et al., 1977; Crystal et al., 1981~. BAL has made pos- sible extensive study of the pathogenic roles played by immune and inflammatory reactions in the respiratory tract and important advances in understanding the pathogenesis of various forms of obstruc- tive, inflammatory, and interstitial lung disease. Much of the information reviewed in this chapter has been obtained with BAL (Reynolds, 1987~. In ravage of the lungs of a normal non- smoker, approximately 10-15 million res- piratory cells typically are recovered. The number of cells obtained in BAL can vary widely. For instance, the number of cells is increased by a factor of ap- proximately 4-5 in ravage fluid from a cigarette-smoking patient and is cor- related roughly with the intensity of smok- ing. Approximately 90% of the cells in ravage fluid are alveolar macrophages (AMs). Most of the remaining cells are lymphocytes, and usually only a few poly- morphonuclear leukocytes (PMNs) are found. Eosinophils and basophils are rare- ly detected. In smokers, the cell yield far exceeds that of nonsmokers. Therefore, although the percentage of lymphocytes is diminished, the absolute number is ac- tually increased. Analysis of the differ- ential counts and more recently the use of monoclonal antibodies and flow cytomet- ry of the cells in BAL specimens have been found to provide useful markers of a vari- ety of pulmonary disorders, particularly the granulomatous and nongranulomatous interstitial lung diseases, and of effects of exposure to pollutants. AMs are the principal phagocytic cell in the airways and seem to play a pivotal role in initiating and modulating the pul- monary immune response (Merrill et al., 1982~. Phagocytosis of microorganisms and other foreign particles in the alveoli by macrophages is an important defense mechanism of the lung against Ins Eaton and other forms of external assault by inhalation. AMs interact with other cells and foreign material in the lung via mem- brane receptors. Surface receptors for
96 the Fc portion of IgG and for complement fragments Cab and Cad have been identified (Reynolds et al., 1975~. The activity of the phagocytic system generally de- pends on the specific immunoglobulins IgA and IgG-IgA probably functions main- ly as an antitoxin and in the neutraliza- tion of viral infectivity, whereas IgG is also important in promoting phagocytosis. Although complement concentrations are low in the airways, complement can be an important participant in infections when the inflammatory response promotes transduction of complement, as well as of serum immunoglobulins. Once phago- cytosis is completed, killing of an in- gested organism is mediated by phagosome- lysosome fusion, degranulation, and the elaboration of digestive enzymes and toxic oxygen species. Present evidence suggests that AMs use metabolically generated H2O2 in conjunction with some type of oxidase to kill microorganisms. The hydrolases in the lysosomes probably play a major role in digesting a phagocytosed microbial carcass. It is also likely that mobilized and activated lymphocytes and macrophages can produce an exterior milieu that is adverse for at least some microorganisms. The various specific immunologic mechan- isms probably act in concert with nonspeci- fic factors in mucus, such as lysozyme and other nonantibody antimicrobial agents; some events of inflammation, as well as the mucociliary approaches, also are im- portant contributions to the overall de- fense of the lung. AMs participate further in the regula- tion of the inflammatory and immune proc- ess in the lungs by secreting a variety of soluble mediators, including products of the arachidonic acid pathway, which seem to play an important role in inflamma- tion (Hunninghake et al., 1980a; Slauson, 1982~. AMs secrete chemotactic factors for neutrophils that cause influx of these cells into the lung parenchyma and alveo- lar space, where they can participate ac- tively in the inflammatory response. AMs have a wide range of other secretory capabilities and have been shown to se- crete such products as colony-stimulating factor, superoxide anions, and various M,4R=RS OF PULMONARY TOXICOLOGY enzymes, including collagenase, neutral protease, and elastase. Among the most potent inflammatory substances produced by AMs are products of arachidonic acid (Slauson, 1982~. Some of the better-known mediators of the inflammatory response include prostaglandin F2 (PGF2 alpha) and the chemotactic factor LTB4. Those arachi- donic mediators and their occurrence are described below. AMs also play a unique role in the development of an immune re- sponse to a novel antigen by presenting bound or ingested antigen to T lympho- cytes. About 8-10% of the cells in the BAL fluid recovered from normal human lungs are lymphocytes. As in blood, three subpopulations of lung lymphocytes (T. B. and NK cells) can be discerned on the basis of their differing surface mark- ers. The proportions of T and B lympho- cytes in ravage fluid from the airway lumi- na have been found to approximate closely those in peripheral blood (Hunninghake and Crystal, 1981a). Whereas T lymphocytes make up about 60-70% of the lymphocytes found in normal BAL fluid, only about 10- 15% of the lymphocytes have surface im- munoglobulin and can be identified as B cells. A portion of the lymphocyte popula- tion that cannot be classified by classi- cal T- and B-cell markers has been iden- tified as NK cells. PMNs are not usually found in large num- bers in BAL fluid from normal lungs (Reynolds, 1987~. In ravage fluid from the lungs of smokers (Young and Reynolds, 1984) or subjects that were exposed to ozone (Seltzer et al., 1986; Koren et al., 1989), an increased percentage and greatly increased absolute numbers of neutrophils are found. As participants in the inflam- matory response in lung tissue, PMNs mi- grate into the lung from the blood under the influence of one or several chemotactic factors, including complement component C5a and soluble factors secreted by AMs. They might be considered as a secondary line of phagocytic defense of the lungs, which can be recruited into the airspaces in response to exposure to microbial agents or other inhaled materials. Some of the observed changes have been shown to have a predictive value; others represent more progressive biologic changes.
INFL4MMATORYAND IMMUNE RESPONSE Acquired Local Immune Response In defining the role of the lungs' im- mune system in the generation of specific biologic markers of injury or disease, it is important to consider the central feature of an immunologic response-that it is a specific reaction to an antigenic stimulus and is capable of distinguishing proteins that differ by only a few amino acids. A substance that is inhaled and can act as a unique and foreign antigen elicits an immune response that is mani- fested predominantly in the form of anti- bodies or sensitized cells. (This mani- festation is the cellular immune re- sponse.) As discussed below, local immune responses in the lung operate in concert with other mechanisms (such as mucocil- iary clearance) in recognizing, trans- porting, and eliminating inhaled foreign agents. Obviously, local immune responses are very important; their failure can result in injury and tissue damage in the lung. With respect to local immune system as a generator of markers, two underlying possibilities or conditions need to be considered: the presence of antigen-spe- cific antibodies or cells indicates that the host has been exposed to an antigen at some time, even if no longer harboring it; and specific immune responses might indicate the presence and persistence of an antigen that produces a chronic inflam- matory response that leads to tissue in- jury. Thus, the products of the immune system can be used as markers of a host's exposure to an antigen that might or might not be responsible for tissue injury and disease. In an operational model, the lung's im- mune system can be viewed as having three distinct compartments, each containing immunocompetent cells (lymphocytes and macrophages): the bronchoalveolar air- spaces, the submucosal or secretory an- tibody system lying beneath the lamina propria of the tracheobronchial tree, and a network of lymphatic vessels and lymph nodes lining the tracheobronchial tree (Daniele, 1980~. In each of these compartments, the potential exists for lymphocyte-macrophage interaction and the generation of immune responses. 97 In the last 5 years, most of our knowledge about cell-mediated immune responses in the lung has come from studies involving cells recovered from bronchoalveolar airspaces with ravage in humans or in ex- perimental animals (Daniele et al., 1985~. Until the advent of the flexible fiber- optic bronchoscope, little was known ab- out the cells and secretions in the bron- choalveolar airspaces of the human lung. It has since been observed that, in non- smoking adults, cell yields equal 10- 15 x 106 cells/100 ml of ravage fluid (Dauber et al., 1979~; AMs are the predominant cell type (80-90%), lymphocytes con- stitute about 10% of the cells (Reynolds et al., 1977; Dauber et al., 1979), and neutrophils, eosinophils, and basophils constitute less than 1 % of the cells. In smokers, the cell yield is some 4 times as great; macrophages usually account for 90% or more of recovered cells, and lymphocytes for 1-5% (Daniele et al., 1977b; Reynolds et al., 1977; Dauber et al., 1979~; there can also be a slightly higher proportion of neutrophils ~ 1 -4%~. The distribution of lymphocyte subpopu- lations in the ravage fluid is similar to that in blood, with T cells accounting for 60-70% of the lymphocytes and B cells 5-10%. The ratio of Th cells to Ts cells is 1.6:1 (Daniele et al., 1975; Dauber et al., 1979; Hunninghake et al., 1979b; Hunninghake and Crystal, 1981 a). The major soluble constituents in la- vage fluid are IgG and IgA; their concen- trations reflect rates of active trans- port across the bronchial epithelium (Reynolds et al., 1977; Low et al., 1978; Hunninghake et al., 1979b). Little or no IgM is present. Components of both the classical and alternative pathways have been identified in ravage fluid, but CS appears to be absent (Robins et al., 1982~. Other inflammatory derivatives have been detected in ravage fluid, including al- phal-antitrypsin. Much needs to be learned about the role of lymphocytes in the normal human lung, particularly with respect to their initi- ation and development of immune responses. The evidence is more substantial in experi- mental animals (Daniele, 1980~. Lympho- cytes recovered with lung ravage from guinea pigs and rabbits respond to
98 antigens introduced into the respiratory tract by producing antibody and lympho- kines (such as macrophage-migration in- hibition factor, MIF). Furthermore, lung lymphocytes can demonstrate an anamnestic response to airborne antigens and, de- pending on the type and dose of antigen, exhibit a capacity to respond that is inde- pendent of systemic lymphoid tissue. Thus, results of animal experiments indi- cate that localized immune responses can occur in the bronchoalveolar airspaces. Alternatively, it has been proposed that the cells and secretions in the bron- choalveolar airspaces are deployed so as to prevent entry of antigenic particles beyond the mucosal barrier and to deter antigen interaction with organized lung lymphoid tissue. According to the notion of "immune exclusion," the primary func- tion of AMs is to ingest particles and re- move them from the lung, rather than to transport them to submucosal and tracheo- bronchial lymph nodes, where lymphocytes and tissue macrophages might interact. Which of the two hypotheses is correct remains to be settled. The two hypotheses might not be mutually exclusive. The nonspecific activities of AMs and the mucociliary blanket might be entirely adequate for expelling some inhaled inert substances. Nonspecific clearance mechanisms would not suffice for other antigens, such as microorgan- isms with capsular membranes that resist phagocytosis, and the aid of specific anti- body and cells in bronchial secretions would be required for effective phagocyto- sis, killing, and clearance. The genera- tion of a specific immune response consist- ing of either antibody or cells in bronchi- al secretions requires, however, that the inciting antigen in some way penetrate the mucosal barrier and stimulate submuco- sal lymphoid cells. That condition is also required for any inhaled particles (e.g., allergens and organic particles) that result in local immoral and cellular immune responses. It should also be emphasized that initially only a relatively small fraction of the inhaled antigenic load might be required for stimulating submuco- sal lymphoid tissue. Once initiated, the secretion of antibody or the appearance AL4RKERS OF PULMONARY TOXICOLOGY of sensitized cells in the airspaces would greatly increase the exclusion of the same or similar inhaled antigens on later challenges. The degree to which nonspecific de- fenses interact with specific immune re- sponses in the lung remains ill defined. It probably depends on the size of the par- ticle, the antigenic load, and the physico- chemical characteristics of the particle, which are related to its antigenicity, toxicity, and, perhaps most important, biologic properties (e.g., type of virus and capsulated bacteria). Those are some of the variables that determine whether inhaled particles and microorganisms are contained or eliminat- ed or result in lung injury and disease. Perhaps equally or more important are the unique genetic properties of the host, especially the immune responses that are linked and controlled by the immune-re- sponse genes. The latter consideration is particularly relevant for two immuno- logic diseases, hypersensitivity pneumo- nitis and chronic berylliosis, that are discussed below. In both, only a minority of persons equally exposed to the airborne agents develop disease. Examination of cells and secretions in BAL fluid from patients with immunolo- gic lung diseases has provided important insights into pathogenesis. First, the lung can be the site of a com- partmentalized inflammatory response (Daniele, 1980), as in hypersensitivity pneumonitis, in which the disease is re- stricted to the lung. In other systemic disorders, the inflammatory response that evolves in the lung might not be reflected in the peripheral blood (Daniele et al., 1980~. The reason for the difference is unclear; one hypothesis is that the lung, when it is involved, acts as a selective target for acute (neutrophils) or chronic (lymphocytes and monocytes) inflammatory cells, which are increased in the pulmon- ary parenchyma as well as in the ravage fluid (Crystal et al., 1981~. Second, pulmonary ravage has establish- ed the existence of two predominant types of chronic inflammatory response in the lung, one involving neutrophils and mac- rophages (idiopathic pulmonary fibrosis)
INFL4MMATORYAND IMMUNE RESPONSE and the other involving lymphocytes and macrophages (hypersensitivity pneumoni- tis and berylliosis). Finally, several laboratories have found in studies of pneumonitis a height- ened state of activation of these inflam- matory cells (Daniele et al., 1980; Crys- tal et al., 1981~. Lymphocyte activation probably reflects immune stimulation in cases of hypersensitivity pneumonitis and berylliosis. In summary, the ability to detect sen- sitized cells or antibodies that are spe- cifically reactive to large complex or- ganic antigens (as in hypersensitivity pneumonitis) or simple elements that be- have as haptens (as in berylliosis) can serve as a useful paradigm for investigat- ing other inhalational diseases in which an immunologic response is predominant in pathogenesis. The presence of specific responses in the lung indicates that the subject has been exposed to a foreign anti- gen; it does not necessarily mean that the antigen is causing disease. for ex- ample, in both hypersensitivity pneumoni- tis and chronic berylliosis, it is still unclear whether the presence of sensitiz- ed cells or antibodies in BAL fluid indi- cates that a patient has or will have dis- ease related to the foreign substances found. Acquired Antigen-Specific Immune Response In Vivo Challenge Testing for immune response has often included testing of whole animals or hu- mans. With such testing, the interaction of several components of the immune system can be tested at once, and actual body re- sponse can be measured, so that one need 99 not rely on extrapolation from results obtained in vitro. However, in viva chal- lenge has several difficulties: the risk of a serious adverse reaction, including anaphylaxis; the difficulty of separating · an- ~ · ~ a nonspec~t arc ~ rom a specie ic response; the difficulty of interpreting whether a response in one area reflects a response in another area; and the difficulty of purifying an antigen to be specific enough for testing and suitable for administra- tion without causing nonspecific damage. The immune response has been divided into four groups summarized in Table 5- 1 (Bellanti, 1985~. Skin testing usually elicits Type I reactions, although Type IV reactions can be detected in skin. Cell-mediated immunity is tested by ex- amining the skin site 24-48 hours after - injection; this can be done to determine whether a subject has been infected with tuberculosis-as with the PPD skin test (Snider, 1 982-or to determine whether a patient is allergic (not reacting to any of the common antigens, such as those of tetanus or mumps). Testing for granuloma formation can use the Kveim antigen (sar- coid tissue antigen) (Chase, 1961~. Skin testing for delayed reactions has not been routinely used for detecting sensitivity to pulmonary toxicants. With further sophistication, a chal- lenge might be graded not only by the am- ount of visible inflammation present, but by other factors, such as the influx of inflammatory cells and the presence of inflammatory mediators, including histamine, immunoglobulins, and immune complexes. Skin Testing A standard method for testing for reac- tion to a possible pollutant is skin test TABLE 5-1 Immunologic Mechanisms of Tissue Injury - Type Manifestations Mediators II III IV Immediate hypersensitivity reactions Antibody-directed reactions Formation of antigen-antibody complexes Mainly IgG IgE and other immunoglobulins IgG and IgM Delayed hypersensitivity (cell-mediated) Sensitized T lymphocytes reactions
100 ing. Allergists have used skin testing extensively to identify substances to which a patient is allergic (Norman, 1980~. Its potential use in environmental studies in toxicology is based mostly on its simplicity of application and inter- pretability. The procedure is relatively safe, although some subjects are allergic to antigens and anaphylaxis has been re- ported after skin testing in a few sub- jects. Usual precautions in skin testing include testing with the lowest possible dose and observing subjects for some time after testing. Methods of skin testing include prick testing, scratch testing, and intradermal injection, in order of increasing dose. The skin prick test is the safest, in that a very small amount of antigen is injected. Reactions to a prick test are read at 10 minutes; a reaction indicates an im- mediate type of sensitivity. Patients with dermatographism will have a false- positive wheel; otherwise, the test is readily assessed. Pepys's group has used the prick test for many years to evaluate exposure of platinum workers (Cleare et al., 1976~. Platinum salts can be highly reactive, and systemic reactions can oc- cur even to scratch tests, which therefore were considered too risky for general surveillance testing. Nevertheless, the authors also found that the prick test was better for differentiating between con- trols and reactive subjects. A major difficulty with skin testing has involved the preparation of a suffi- ciently reactive antigen. Most sub- stances studied are haptens and become antigenic on combination with high-molec- ular-weight carriers. That can happen at the injection site. For example, phthalyl acid anhydride is an essential reagent in the manufacture of epoxy resins and some paints. It is highly reactive, and skin testing can be performed direct- ly. Positive skin tests have been seen in documented cases of asthma induced by phthalyl anhydride (Maccia et al., 1976~. Most substances do not induce responses by themselves and have to be conjugated to human serum albumin before testing. Some have been found to be reliable skin- test reagents for particular environmen- tal pollutants (Zeiss et al., 1983~. MARKERS OF PULMONARY TOXICOLOGY Another difficulty with skin testing has involved the need to relate the find- ings with the pulmonary symptoms of the subjects. A positive skin prick test in platinum refiners is a more specific and sensitive index of disease than are some clinical symptoms (Dally et al., 1980~. For example, skin prick tests of mouse and rat urine extracts in laboratory-ani- mal workers have yielded a sensitive meas- ure of asthma, but not of rhinitis or urti- caria (Newman Taylor et al., 1981~. Work with skin testing has extended be- yond the routine measurement of size and character of skin reaction. As mentioned above, skin biopsies are routinely used to examine for the presence of granulomas after a Kveim test; studies are underway to characterize the earlier stages of the inflammation (Mishra et al., 1986~. The studies have included examination of in- flammatory cell population and mediators in the biopsies of skin lesions during various phases of immediate skin reaction and have led to a better understanding of early pathologic response. During the late-phase reaction, skin biopsies can show neutrophil and lymphocyte influx (Felarca and Lowell, 1971~. A novel method is the injection of anti- gen into bullae in the skin. This particu- lar challenge allows one to measure the influx of mediators, including histamine, into the site of a skin reaction (Warner etal., 1986~. In summary, skin testing has several advantages, including low cost, wide ap- plicability, ready acceptance by pa- tients, and relative safety. Its major drawbacks include difficulty in assessing observations regarding skin reactions and in correlating reactions with symp- toms in other organs and identifying prop- er antigens for testing. Nasal Challenge The upper airways, especially the nose, are a major target of toxic damage. Rhini- tis is a common complaint after exposure to toxicants; but research into rhinitis has been limited, because it is not asso- ciated with substantial morbidity and its relationship with lower respiratory symptoms is not clear. Nasal challenges
INFLAMMATORYAND IMMUNE RESPONSE do provide information that might not be obtainable with any other method and thus should always be considered when examin- ing new ways of studying toxicants are being examined. Nasal challenges date at least to 1873, when Blakley placed grass pollen in the noses of allergic patients and induced the signs and symptoms of al- lergic rhinitis (Naclerio et al., 1983~. The method of intranasal challenge var- ies. Again, identification of the correct antigen is difficult. One method is to study patients with known intradermal reactions. The specific known antigens are then delivered by nebulizer (Naclerio et al., 1983) or by direct application of an extract (Naclerio et al., 1985~. Methods of assessing inflammatory reac- tions after nasal challenge also vary. They include measurement of airway resis- tance in the challenged nostril (McLean et al., 1976), measurement of mucus pro- duction (Maim et al., 1981), subjective assessment (Cornell, 1979; Naclerio et al., 1985), and objective assessment of hyperemia and stenosis (Naclerio et al., 1983~. Nasal challenge is fairly safe. The usual symptom is rhinitis, and an oc- casional patient develops wheezing. Un- like skin tests, it has not been used in large populations. But there is little to suggest that it could not be performed in a similar manner, with the patient ob- served for some period after challenge. The cost would depend on the extent of as- sessment. For example, if measurement of mediators in nasal washes were the goal, the assays could become expensive and thus impractical for screening large popula- tions. Observation for the presence of edema would be simple, although difficult to measure. Nasal airway resistance can be measured by anterior rhinomanometry, which is relatively simple and inexpen- sive (McLean et al., 1976; Naclerio et al., 1983~. In summary, nasal challenge has dis- tinct advantages over skin testing, be- cause it uses a mucosal surface. Direct observation can be used to assess inflam- matory response, so it might be appropri- ate for screening large populations. In addition, when more objective data are 101 required, nasal airway conductance is easily measured. The best method for assessing airway response to an antigen would be direct observation. The antigen is chosen on the basis of intradermal response. The antigen dose, described in protein nitro- gen units (1 unit is the amount that causes a 4 x 4-mm wheel after intradermal injection), is determined. Intrabron- chial challenge is then begun at one-hun- dredth of that unit. Intrabronchial chal- lenge is usually complemented by BAL in the contralateral lung and in the edema- tous bronchus after challenge. Bronchoscopy is performed in the usual manner. Subjects are premedicated with atropine, metaproterenol (a beta agonist), and topical Xylocaine. The bronchoscope is advanced to a subsegmen- tal bronchus, the initial dose of antigen is injected through the bronchoscope, and the bronchus is observed for 3 minutes. If there is no change in the bronchus, the dose is increased. A recordable response consists of blanching, edema, or narrow- ing of the airways. The major advantage of Intrabronchial challenge is its specificity for identi- fying an inflammatory response in the bronchus. Visualization lasts for only 3-5 minutes, so it would detect only an immediate response. However, repeat bronchoscopy has been done 2-3 days after bronchial challenge to assess persistent changes, and persistent abnormalities in the cell population have also been ob- served in BAL fluid (Metzger et al., 1987~. Patients challenged to date have been challenged only with antigen to which they have a good skin response. Patients were usually far more sensitive to intrabron- chial than to intradermal exposure. Of 11 patients studied by Metzger et al. (1987), nine responded to less than one- twentieth of the intradermal dose. Intrabronchial challenge presents many problems, mostly because it is relatively new. Although the bronchial changes are visually dramatic, there is little objec- tive measure of response. Because of prob- lems with parallax from a flexible fiber
102 optic bronchoscope, it is difficult to determine size in the bronchus without a reference object at the same plane as the area one wishes to measure. That is commonly provided by touching the area with an open biopsy forceps or attempting to pass a bronchoscope or a bronchoscopy brush through a narrowed bronchus (Zavala, 1978~. Obviously, touching the walls of the bronchus that one wishes to evaluate for edema can lead to local trauma and in- correct interpretation of edema. In addi- tion, accurate estimation of airway nar- rowing might not be possible when a bron- chus has responded to an antigen. Intrabronchial challenge poses a sub- stantial risk in some persons. The risk associated with bronchoscopy in asthmatic subjects is dealt with in the section on BAL; the risk associated with intrabron- chial challenge conceivably is even high- er. In the studies reported so far, pa- tients have been carefully selected, many precautions have been observed, and es- tablished guidelines have been followed (NHLBI, 1985~. Patients have been observ- ed closely for evidence of bronchospasm. In one study (Metzger et al., 1987), three of 11 asthmatic subjects developed wheez- ing; two were treated with local epine- phrine, and the other with aerosol ther- apy. Pulmonary function of all asthmatics returned to normal within 15 minutes of the procedure. A final problem in intrabronchial chal- lenge is cost. With the current system, including close observation, studies are expensive and require highly trained med- ical and technical assistants. In conclusion, the utility of intra- bronchial challenge as a screening tool for identifying patients sensitive to pulmonary toxicants seems limited. In studies to date, only patients who were highly responsive to skin tests responded to intrabronchial challenge. In most of the reported studies, patients were chal- lenged with an antigen clearly associated with pulmonary symptoms. Although the research data obtained after intrabron- chial challenge are considerable, their application to a large group of subjects remains questionable. AL9RKERS OF PULMONrARY TOXICOLOGY In Vitro Challenge Proliferation of lymphocytes exposed to antigen in vitro is an indication of sensitization. In general, lymphocyte proliferation requires the participation of accessory cells and products of Type I or Type II histocompatibility antigens expressed on accessory cell surfaces. Accessory cells are usually macrophages, but dendritic cells, B cells, and perhaps other cells (such as fibroblasts) can act as accessory cells. The exact relation- ships between lymphocytes and accessory cells in the lung remain to be defined. The pulmonarylymphocytes obtained with BAL are functionally competent-they can proliferate and produce lymphokines when exposed to antigens to which they are sensitized (Schuyler et al., 1978; Moore et al., 1980; Pinkston et al., 1983~. The exact population of lymphocytes re- sulting from proliferation and the level of lymphokine secretion are not known. Proliferation of antigen-induced and mitogen-induced BAL lymphocytes is lower than proliferation of peripheral blood lymphocytes. Increases in the percentage and number of BAL lymphocytes are characteristic of granulomatous lung diseases, such as hypersensitivity pneumonitis, sarcoidos- is, berylliosis, and tuberculosis (Reynolds et al., 1977; Rossman et al., 1978; Godard et al., 1981, Epstein et al., 1982~. Recent reports indicate that pulmonary lymphocytes from patients with sarcoido- sis spontaneously secrete interleukin- 2 (Pinkston et al., 1983), which provides a signal for responsive lymphocytes to proliferate. There is evidence that I1- 2 secretion by pulmonary lymphocytes from patients with sarcoidosis is secondary to an altered milieu in the lung, rather than being a reflection of changes of the constitutive properties of T lymphocytes (Muller-Quernheim et al., 1986~. Pulmonary lymphocytes from patients with hypersensitivity pneumonitis are sensitized: they proliferate and produce lymphokines on exposure to the appropri- ate antigen (Schuyler et al., 1978; Moore et al., 1980~. There is evidence that
INFL4MMATORYAND LINE RESPONSE cells that suppress lymphocyte prolifera- tion are present in asymptomatic exposed persons, but not in symptomatic exposed persons (Amrein et al., 1970~. Therefore, lack of suppressor cells in some subjects could be associated with development of symptoms of hypersensitivity pneumonitis after the same amount of systemic exposure that does not cause clinical symptoms in subjects with suppressor cells. In general, lymphocytes from patients with berylliosis, but not from control populations, proliferate when exposed to beryllium salts. The results with sub- jects exposed to beryllium but without apparent disease are controversial (Hani- fin et al., 1970; Deodhar et al., 1973; Epstein et al., 1982; Williams and Wil- liams, 1982, 1983; Rom et al., 1983; Bargon et al., 1986~. The relationship of lymphocyte proliferation and beryl- liosis is complex. Beryllium salts have multiple effects on lymphocytes in cul 103 sure: at high concentrations, beryllium is toxic to lymphocytes and decreases pro- liferation; at low concentrations, it increases mitogen- and antigen-induced proliferation (Williams and Williams, 1982~. Lymphocyte proliferation in peri- pheral blood has been found to correlate with beryllium exposure in a beryllium plant (Rom et al., 1983) and thus might be a good marker of a population's exposure to beryllium. Although proliferation of lymphocytes from peripheral blood has been studied most extensively, there is preliminary evidence that bronchoalveo- lar lymphocytes from a patient with beryl- liosis also proliferate when exposed to beryllium (Epstein et al., 1982~. In summary, lymphocyte proliferation seems to be an index of exposure to envir- onmental agents and in some instances a marker of disease. The relationship of lymphocyte proliferation and pathogen- esis in humans is unknown.
