Pig Lungs. chial cell infiltration, and tissue necrosis are. of host resistance during virus infection is mediated

Transcription

1 INFECTION AND IMMUNITY, Jan. 1973, p American Society for Microbiology Vol. 7, No. 1 Printed in U.S.A. Effect of Hypersensitivity Pneumonitis on the Pulmonary Defense Mechanisms of Guinea Pig Lungs GEORGE J. JAKAB AND GARETH M. GREEN The Pulmonary Unit, Department of Medicine, and the Specialized Center of Research in Pulmonary Disease, University of Vermont College of Medicine, Burlington, Vermont Received for publication 16 August 1972 Many edemagenic and consolidating inflammatory diseases, such as virus pneumonias, of the lung are complicated by bacterial infection. Previous literature has stressed that edema and consolidation may promote bacterial proliferation by interfering with phagocytosis. To test that hypothesis, lung defense mechanisms were studied in guinea pigs with tuberculin-induced hypersensitivity pneumonitis, a noninfectious edemagenic, and consolidating inflammatory disease. Pulmonary bactericidal activity and particle clearance were measured with a mixed aerosol of 32P-labeled Staphylococcus aureus and 35S-labeled Proteus mirabilis. Hypersensitivity pneumonitis enhanced the bactericidal activity of the lung but had no effect on particle clearance despite the presence of consolidation and edema. These data indicate that altered host resistance to bacterial infection in acute inflammatory lung diseases can not be attributed to edema, inflammation, consolidation, changes in lung weight, etc., per se and that causes must be sought in functional changes in the bactericidal system of the lung rather than in specific histopathological changes. A variety of edemagenic and consolidating pulmonary lesions are complicated by bacterial infection, presumably because of impaired host antibacterial defenses in the lung. Viral pneumonia (5, 13), oxygen toxicity (12), chemicallyinduced pulmonary edema (10, 11; M. LaForce et al. Clin. Res. 19:324, 1971), and kerosene pneumonitis (G. Huber et al. Clin. Res. 19: 741, 1971) induce fluid accumulation and consolidation and impair pulmonary antibacterial activity. Because of this relationship, it has been generally accepted that the anatomic lesion, particularly the edema, may promote bacterial multiplication by interfering with phagocytosis (11; G. Huber et al. Clin. Res. 19:741, 1971). Nonlethal influenza virus (5), reovirus (14), or Sendai virus (13) infection of mice suppresses pulmonary antibacterial activity 6 to 10 days after inoculation of the virus. This impairment of bactericidal activity is not associated in time with virus proliferation but rather with the period of rapid decline in pulmonary virus titers and with the appearance of serum antibody. Pulmonary histological changes, however, marked by epithelial cell destruction, peribron- 39 chial cell infiltration, and tissue necrosis are prominent during the period of peak clearance suppression (5, 13). A recent finding from this laboratory (13) indicates that this impairment of host resistance during virus infection is mediated through a defect in in situ bactericidal mechanisms rather than a transport defect in the mucociliary train or alveolar transport system. Depressed pulmonary bactericidal activity in the presence of virus-induced lesions, however, has been restored by immunization with the homologous challenge organism (G. Jakab and G. M. Green, Abst. Annu. Meet. Amer. Soc. Microbiol., p. 207, 1972). To determine whether these defects in bactericidal activity were due to the pathology per se (e.g., fluid accumulation, edema, etc.) as attributed by others, or due to functional changes in the bactericidal system not necessarily related to the edemagenic lesions, a model comparable in pathology was induced by hypersensitivity mechanisms and studied in similar fashion. The results (presented in part at a meeting, Recent Advances in Infectious Diseases, Boston, 1972) prove that pathological lesions characterized by edema, cellular infil-

2 40 JAKAB AND GREEN INFECT. IMMUNITY tration, and consolidation do not of themselves interfere with the host defense mechanisms of particle transport and phagocytosis. MATERIALS AND METHODS Immunization. Male Hartley guinea pigs, weighing 350 to 400 g, were actively immunized against tuberculin protein by a single intramuscular injection of 0.5 ml of complete Freund adjuvant (Difco). Four weeks later, the animals were tested for tuberculin hypersensitivity by the intradermal injection of 1.0 jig of purified protein derivative (PPD, Parke-Davis) into the shaved dorsal skin. Only animals with strong positive reactions were used for subsequent tuberculin challenge studies. Preparation of bacterial aerosols. Staphylococcus aureus (coagulase-positive FDA strain 209P, phage type 42D) and Proteus mirabilis were labeled with 32P and 35S, respectively, by the methods of Green and Goldstein (6) and Green and Green (9). In brief, S. aureus and P. mirabilis were incubated in 70 ml of phosphorus-free and sulfur-free culture medium containing 1.0 mci of either 32P or 35S. After 18 hr of growth at 37 C in a rotating shaker water bath, the labeled cells were centrifuged separately, washed twice in 30 ml of phosphate buffer (ph 7.4) to remove all unattached label, and suspended in 4 ml of Trypticase soy (TS) broth. Animal exposure. For inhalation challenge with PPD, 15 ml of 0.5 mg of PPD per ml were aerosolized into a chamber previously described (13) which contained the sensitized and normal guinea pigs. Twentyfour hours later, normal, sensitized, normal PPDtreated, and sensitized PPD-treated animals were challenged by aerosol for 30 min with a mixed bacterial suspension containing 3"S-labeled P. mirabilis and 32P-labeled S. aureus. The chamber air was sampled for bacterial concentration during the middle 10 min of the exposure period with an Andersen sampler (1). Petri plates containing 6 ml of TS broth were inserted into each of the six stages of the sampler. After sampling of the aerosol, 2-ml fractions from each of the petri plates in the lower three stages were pooled, and three samples were assayed quantitatively for culturable bacteria and radioactivity. The lower three stages contained over 97% of the infectious droplets (3.5 jam in diameter or less) and radiotracer activity. Bacteriological and radioassay procedures. The guinea pigs were killed by intraperitoneal injection of sodium pentobarbital and exsanguinated by cardiac puncture at 0, 4, and 24 hr after the bacterial infection. The lungs of each animal were removed aseptically, washed with sterile phosphate buffer, and dissected into the individual lobes. Each lobe was then weighed and homogenized in 3 ml of TS broth. Samples (1 ml) of the homogenates and aerosol samples were cultured quantitatively in quadruplicate on petri X-plates by the standard dilution pour plate technique. Separation of the two bacterial species for colony counting was accomplished by using phenylethanol agar to select for S. aureus and bismuth sulfite agar to select for P. mirabilis. Colony counts were performed after 48 hr of incubation at 37 C and were expressed as the average colony-forming units per milliliter of material assayed. Quantitative measurement of radioactivity of each lobe was performed on 1-ml fractions of each homogenate and aerosol sample. Samples were prepared for liquid scintillation counting by methods described previously (13). In brief, 1 ml of each homogenate was digested in 2 ml of hyamine hydroxide (10x) overnight at room temperature, and 5 ml of absolute ethanol and 10 ml of liquid scintillation solution were added to each sample. The samples were assayed in a Beckman liquid scintillation spectrometer (LS 150 series). The two isotopes were differentiated by using the discriminator ratio method of Okita et al. (23). After correction for background, quench, and dilution, radioactivity was expressed as counts per, minute per milliliter of material assayed. Calculation of pulmonary antibacterial activity. Pulmonary bactericidal activity, defined as the change in the proportion of viable to total bacteria in each individual lobe, was calculated by the radioactive ratio method of Green and Goldstein (6). The ratio of bacterial to radioactive counts was determined for the organism in the aerosol and in the lungs. Bactericidal activity was calculated from the change that occurred between the two ratios and is expressed as a percentage of the ratio in the aerosol as follows: Percent bacteria remaining = (bacterial counts [lung] per tracer counts/bacterial counts [aerosol] per tracer counts) x 100 = (ratio [lung] per ratio [aerosol]) x 100. This method calculates bactericidal activity of the lung as a function independent of the number of inhaled organisms. Calculation of physical transport activity. Transport of 3IS-labeled P. mirabilis and 32P-labeled S. aureus from the lungs was determined by following the decline in radiotracer activity. The radiotracer counts are expressed as the percentage of the mean counts obtained from the 0-hr sacrifices. Histology. Unless otherwise noted, the left upper lobe of each test guinea pig was utilized for study of pulmonary histology. All lobes were fixed in 10% buffered Formalin, sectioned at 5 jm, and stained with hematoxylin and eosin (HE). Statistical determinations. The data were analyzed by determining the arithmetic means, standard deviations, and standard errors. Comparisons were made according to the paired t test. RESULTS Pulmonary changes. Macroscopic examination of the lungs of sensitized guinea pigs 24 hr after inhalation of PPD showed plum-colored consolidation involving, in most cases, an estimated 80% of the surface area of the lung. One animal, in which the pneumonic process involved the entire lung, presumably died of hypoxia just prior to aerosol challenge with the bacterial mixture. Microscope examination 24 hr after PPD exposure showed peribronchial and perivascular infiltration of mononuclear

3 VOL. 7, 1973 HYPERSENSITIVITY PNEUMONITIS 41 cells, mostly lymphocytes (Fig. 1). The alveolar septae and spaces were filled with the mononuclear infiltrate, and edematous inflammatory changes were observed in some heavily damaged areas (Fig. 2). These pathological changes were observed in an estimated 60 to 80% of the area in each histological section. No such macroscopic or microscopic changes were seen in the lungs of sensitized guinea pigs not exposed to inhalation of PPD, or unsensitized animals exposed to PPD, or untreated control animals. Lung weights of guinea pigs, expressed as percentage of body weight, are presented in Table 1. In experiment 1, the entire lung was utilized in these determinations, whereas in experiment 2 the upper left lobe was used for histology, and the subcardiac lobes were omitted from the calculations for convenience. In experiment 1, lungs of control animals were only 0.68% of total body weight as compared with 1.20% of those with hypersensitivity pneumonia (P < 0.01). In experiment 2, the lung weight-body weight values for untreated control guinea pigs, those sensitized without PPD inhalation, and unsensitized animals exposed to PPD were 0.38, 0.43, and 0.42%, respectively, as compared with 1.08% for those animals with hypersensitivity pneumonitis (P < 0.01 between the former three groups and the latter group). The higher value obtained from control animals in experiment 1 (0.68%) as compared with control animals in experiment 2 (0.38%) is obviously because the entire lung was not used for the determinations in the latter experiment; however, pulmonary weight of animals with hypersensitivity pneumonitis did not differ sig- FIG. 1. Photomicrograph of a section of lung from a guinea pig with hypersensitivity pneumonitis. Note the area of dense consolidation (left) as compared with interstitial infiltration accompanied with patches of intraalveolar edema (right). HE stained, x 250. FIG. 2. Higher power photomicrograph of a section of lung from a guinea pig with hypersensitivity pneumonitis showing greater detail of interstitial infiltration of mononuclear cells and intraalveolar edema. HE stained, x 600. nificantly between the experiments even though less lung volume was used in experiment 2. Histological examinations showed that consolidation was more extensive in those animals of experiment 2 (one even died presumably of hypoxia), hence more weight per unit volume. Pulmonary deposition and physical transport. Table 2 compares radiophosphorus and radiosulfur activity in the lungs of guinea pigs at 0, 4 and 24 hr after aerosol challenge with 35S-labeled P. mirabilis and 32P-labeled S. aureus. At 0 hr, 32P and 35S activity recovered from the lungs of animals with hypersensitivity pneumonitis (columns B and D, Table 2) was counts/min and 199 ± 77 counts/min, respectively, which was significantly lower (P < 0.01) than the counts/min and counts/min recovered from normal lungs (columns A and C, Table 2). Radiotracer deposition was also inversely proportional to the amount of consolidation present. At 0 hr, animal no. 3 with hypersensitivity pneumonia had less consolidation per lobe (average of 1+ on a 0 to 4+ scale) than animals no. 1 and 2 which had approximately 3+ consolidation. From the animal with lesser consolidation, three times as much 35S was recovered than from the other two animals with a greater degree of consolidation. Table 2 also compares clearance of pulmonary radiophosphorus and radiosulfur from lungs of control guinea pigs and guinea pigs with hypersensitivity pneumonitis at 4 and 24 hr after aerosol exposure. At these time periods, clearance of 32P from lungs of animals with hypersensitivity pneumonitis was % and % as compared with 79 i 8% and 58 i 3% in normal animals. At the same time

4 42 JAKAB AND GREEN INFECT. IMMUNITY TABLE 1. Lung weights of guinea pigs with and without hypersensitivity pneumonitis Group Total body wt (%) in expta la 2 Nonhypersensitive, without PPD ± 0.05 (12) 0.38 ± 0.02 (6) Hypersensitive, with PPD.1.20 ± 0.06 (12) 1.08 ± 0.10 (6) Nonhypersensitive, with PPD.NT 0.43 ± 0.04 (6) Hypersensitive, without PPD... NT (6) a Numbers in parentheses indicate the number of animals tested. NT, Not tested. b Entire lung utilized. c Upper left lobe and subeardial lobes omitted from calculations. TABLE 2. Radiophosphorus and radiosulfur clearance from lungs of normal control guinea pigs and guinea pigs with hypersensitivity pneumonitis Time after exposure Animal Nonsensitive Hypersensitive Nonsensitive Hypersensitive without PPD with PPD without PPD with PPD (A) (B) (C) (D) 0 Hour 1 3, a , ,245 ± 620 1,420 ± t ± 63 Grand mean ± SE 3,438 ± 234b 670 ± ± i 77 4 Hour 1 2,610 ± ± ± , , , Grand mean ± SE 2,738 ± ± ± ± 27 % of O-hr grand mean ± 8 75 ± 8 94 ± 13 32p 24 Hr 1 2, , ± ,949 ± ± ± ± ,848 ± ± Grand mean + SE 2, ± ± ± 29 % of0-hrgrand mean ± ± a Determined by calculating the mean and standard error (SE) of radiotracer counts in all five individual lobes in each animal. Determined by calculating the mean and SE of radiotracer counts of all lobes in each group. periods, radiosulfur declined to 94 ± 13% and % in hypersensitive pneumonitic lungs as compared with 75 i 8% and 41 4% in normal lungs. The probability of statistical significance of differences in clearance of each radiotracer at 4 and 24 hr between normal animals and animals with hypersensitivity pneumonitis was greater than Bactericidal activity. Table 3 contains detailed data from the first of two experiments performed to determine the effect of hypersensitivity pneumonia on the bactericidal activity of the guinea pig lung. In this initial study, only normal control animals were compared with those with hypersensitivity pneumonitis. Examination of pulmonary bactericidal values in guinea pigs at 0, 4, and 24 hr after mixed bacterial challenge shows a rapid decline of both viable staphylococci and P. mirabilis in both animals and in animals with hypersensitivity pneumonitis. In the control group, S. aureus declined from 79.6 i 5.7% to 12.0 ± 0.6% at 4 hr and % at 24 hr; by comparison, a significant enhancement of bactericidal activity was demonstrated by lower values of 28.5 ± 6.0%, 2.2 ± 0.2%, and 0.16 i 0.05% at the same time periods in the group with hypersensitivity pneumonitis (P < 0.01 at 0 hr and P < at 4 hr). The same pattern was observed with P. mirabilis. At these time periods, control values declined from i 6.4 to % and finally to 0.40 ± 0.05% as compared to 7.5 i 1.8%, 0.4 ± 0.1%, and % in the pneumonitis group (P < at 0 hr and P < 0.05 at 4 hr). The second experiment, concentrated on ap- 35S

5 VOL. 7, 1973 HYPERSENSITIVITY PNEUMONITIS TABLE 3. Effect of hypersensitivity pneumonia on intrapulmonary killing of Staphylococcus aureus and Proteus mirabilis in guinea pig lungs Bacteria remaining (%) 43 Time after exposure Animal Staphylococcus aureus Proteus mirabilis no. Nonsensitive Hypersensitive Nonsensitive Hypersensitive without PPD with PPD without PPD with PPD 0 Hour ± 7.5a 40.3 ± i ± ± ± ± ± ± ± ± Grand mean ± SE b 28.5 ± ± ± Hour ± ± ± ± ± ± ± ± i i ± ± ± ± ± ± 0.1 Grand mean + SE 12.0 ± ± ± ± Hour ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.1 Grand mean + SE 0.10 ± ± ± ± 0.5 a Determined by calculating the mean and standard error (SE) of bactericidal values of all five individual lobes in each animal. The individual bactericidal values were first calculated by the formula appearing in Materials and Methods. bdetermined by calculating the mean and SE of bactericidal values of all lobes in each group. The individual bactericidal values were first calculated by the formula appearing in Materials and Methods. propriate controls, as hypersensitive guinea pigs without PPD challenge and normal animals with PPD challenge were not included in the initial experiment. For convenience, this latter experiment included only animals sacrificed 4 hr after aerosol exposure to the mixed bacterial suspension. S. aureus in normal animals declined to % as compared to % (Table 4) in guinea pigs with hypersensitivity pneumonitis (P < 0.01). Pulmonary bactericidal values in suitable controls (immunized only, 7.7 i 1.2%; and PPD alone, %) did not differ significantly from the two experimental groups. The same pattern was observed with P. mirabilis except that the values of animals with hypersensitivity pneumonitis ( %) were not lower than the values of normal animals ( %). DISCUSSION The decline of culturable organisms in the lung without comparable loss of an attached physical marker means that in situ killing of the bacteria precedes removal and is in fact separable as a defense mechanism. This in situ pulmonary bactericidal function has been localized to pulmonary phagocytes, with alveolar macrophages playing a predominant role in the resting lung (7). There is no evidence that bacterial replication (balanced or exceeded by killing) occurs in this system in healthy animals when organisms of low virulence are used. The bactericidal activity of the lung is a logarithmic function in normal animals and is independent of the number of bacteria deposited (15). A variety of experimental treatments alter the deposition rate of inhaled aerosols apparently by affecting ventilation patterns (rate and depth of respiration) and, in the case of consolidative lesions, by reduction of lung space available for particle deposition. The inflammatory process accompanying the hypersensitivity reaction presumably reduces particle deposition by these mechanisms. However, the method of assaying bactericidal activity used in these studies is independent of the number of organisms initially deposited since it uses the radiotracer as a stable denominator against which the numerator of viability is quantified (6). Thus, although the initial deposition rate was less in the hypersensitive animals, the method used and the characteristics of the model system nullify the significance of that difference as far as bactericidal activity is concerned. Gram-negative and staphylococcal pneumonias are common respiratory infections which occur during periods of lowered host resistance. P. mirabilis and S. aureus were studied

6 44 JAKAB AND GREEN INFECT. IMMUNITY TABLE 4. Effect of hypersensitivity pneumonia on intrapulmonary killing of Staphylococcus aureus and Proteus mirabilis in guinea pigs Bacteria remaining (%) Organism Time exposure after Anim~al no. Nonsensitive Nonsensitive Hypersensitive Hypersensitive without with PPD without PPD with PPD _PPD S.aureus 4 Hour ± OPa 5.0 i ± ± ± ± ± ± ± ± ± ± ± X ± ± ± ± ± ± ± ± ± 4.1 Grand mean 13.1 ± ± ± ± 0.9 ± SE P. mirabilis 4 Hour ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 0.8,5 6.8 ± ± ± ± Grand mean 4.6 ± ± ± ± 0.9 I I _ a Determined by calculating the mean and standard error (SE) of bactericidal values of three individual lobes of each animal. The individual bactericidal values were first calculated by the formula appearing in Materials and Methods. b Determined by calculating the mean and SE of bactericidal values of all lobes in each group. The individual bactericidal values were first calculated by the formula appearing in Materials and Methods. simultaneously as the lung does not respond to all organisms in the same way (8) but does react independently to each bacterium in mixed infection (9). For example, hypersensitivity pneumonitis was associated with a small but significant degree of enhancement of destruction of staphylococci, but had no effect on decline of P. mirabilis (Table 4). Impairment of pulmonary antibacterial activity is temporally associated with a variety of edemagenic and consolidative lesions. Because of this relationship, it has been suggested and generally accepted that the accumulation of cells, cell debris, and fluid in the lung furnishes culture medium for bacterial proliferation, interferes with removal of bacteria, or interferes with phagocytosis (10, 11). The finding that spontaneous (G. Jakab and G. M. Green, manuscript submitted for publication) or immunologically induced consolidation with edema causes no defect in pulmonary microbicidal mechanisms and that depressed pulmonary bactericidal activity in the presence of virusinduced lesions can be restored by immunization with the challenge organism (G. Jakab and G. M. Green, Abstr. Annu. Meet. Amer. Soc. Microbiol., p. 207, 1972) indicates that suppression of bactericidal activity in the lung is mediated by a mechanism other than the classical fluid and cellular debris hypothesis. 'l'he effect of acquired cellular resistance on phagocytosis is well documented. Acquired cellular resistance is mediated by specifically sensitized lymphocytes (16) and expressed through mononuclear phagocytes (21). During the course of infection with intracellular organisms such as Mycobacterium or Listeria, the host macrophages become activated (16), a process that involves changes in morphology (2), enzyme content (4), and an enhanced capacity to inactivate ingested organisms (3). The immune state and the triggering of macrophage activation are specific, but once they are activated their effect is nonspecific (17). In vitro experiments indicate that both phagocytic uptake and cytopepsis are enhanced (16). The development of this cellular resistance runs parallel with the development of delayed hypersensitivity (18) which, in this model, reaches a maximum approximately 24 to 48 hr after inhalation of PPD (20). Cell-mediated immunity in the respiratory tract against Listeria monocytogenes has been demonstrated by experiments of Truitt and Mackaness (24). In normal mice, aerosol-administered L. monocytogenes can survive ingestion by alveolar macrophages and multiply in the lung; however, in mice actively immunized by intravenous infection with sublethal doses of the live organism, L. monocytogenes proliferated for only 24 hr after aerosol challenge, after which microbicidal mechanisms

7 VOL. 7, 1973 HYPERSENSITIVITY PNEUMONITIS caused rapid elimination of the microbial population. During the 24-hr period, the cellular composition of the lungs assumed the appearance of a delayed hypersensitive reaction characterized by infiltration of blood-borne mononuclear cells. Their studies indicate that existing pulmonary bactericidal mechanisms of the alveolar macrophages were not effective against the organism. The lung defenses had to wait recruitment of cells of another type. In contrast, the present study did not use the homologous organism for induction, elicitation of delayed hypersensitivity, and aerosol challenge to quantify pulmonary bactericidal activity. Instead, the delayed hypersensitivity reaction in lungs of tuberculin-sensitized guinea pigs was induced by inhalation of PPD. These animals, subsequently challenged with a mixture of staphylococci and P. mirabilis, showed no delay in pulmonary antibacterial activity. Instead, guinea pigs with hypersensitivity pneumonitis showed enhanced bactericidal activity toward staphylococci at 0 and 4 hr after aerosol challenge. By 24 hr, however, the number of viable bacteria recovered from lungs of animals with hypersensitivity pneumonitis did not differ significantly from the number recovered from control guinea pigs. These results indicate that the overall antibacterial success is not greater but only faster. Accelerated phagocytic uptake with no enhanced intracellular killing may explain the absence of a long-term antibacterial enhancement in lungs of guinea pigs with hypersensitivity pneumonitis. These data demonstrate that anatomical lesions of the lung, as measured by consolidation, fluid accumulation, and lung-body weight ratios, are not necessarily correlated with impairment of host antibacterial defenses and may in fact be associated with accelerated host defense responses. ACKNOWLEDGEMENTS The writers thank Linda Pfeiffer, Jeanne Lisbon, and Lorraine Lindsay for their competent technical assistance. This work was aided by Public Health Service grants and LITERATURE CITED 1. Andersen, A. A New sampler for the collection, sizing, and enumeration of viable airborne particles. J. Bacteriol. 76: Blanden, R. V Modification of macrophage function. J. Reticuloendothel. Soc. 5: Blanden, R. V., G. B. Mackaness, and F. M. Collins Mechanisms of acquired resistance in mouse typhoid. J. Exp. Med. 124: Dannenberg, A. M Cellular hypersensitivity and cellular immunity in the pathogenesis of tuberculosis: specificity, systemic and local nature of associated macrophage enzymes. Bacteriol. Rev. 32: Green, G. M Patterns of bacterial clearance in murine influenza. Antimicrob. Ag. Chemother. 1965, p Green, G. M., and E. Goldstein A method for quantitating intrapulmonary bacterial inactivation in individual animals. J. Lab. Clin. Med. 68: Green, G. M., and E. H. Kass The role of the alveolar macrophage in the clearance of bacteria from the lung. J. Exp. Med. 119: Green, G. M., and E. H. Kass The influence of bacterial species on pulmonary resistance to infection in mice subjected to hypoxia, cold stress, and ethanolic intoxication. Brit. J. Exp. Pathol. XLVI: Green, L. H., and G. M. Green Differential suppression of pulmonary antibacterial activity as the mechanism of selection of a pathogen in mixed bacterial infection of the lung. Amer. Rev. Resp. Dis. 98: Harford, C. G., and M. Hara Pulmonary edema in influenzal pneumonia of the mouse and the relation of fluid in the lung to the inception of pneumococcal pneumonia. J. Exp. Med. 91: Harford, C. G., V. Leidler, and M. Hara Effect of the lesion due to influenza virus on the resistance of mice to inhaled pneumococci. J. Exp. Med. 89: Huber, G. L., and F. M. LaForce Comparative effect of ozone and oxygen on pulmonary antibacterial defense mechanisms. Antimicrob. Ag. Chemother. 1970, p Jakab, G. J., and G. M. Green The effect of Sendai virus infection on bactericidal and transport mechanisms of the murine lung. J. Clin. Invest. 151: Klein, J. O., G. M. Green, J. G. Tilles, E. H. Kass, and M. Finland Effect of intranasal reovirus infection on antibacterial activity of mouse lung. J. Infect. Dis. 119: Laurenzi, G. A., L. Berman, M. First, and E. H. Kass A quantitative study of the deposition and clearance of bacteria in the murine lung. J. Clin. Invest. 43: Mackaness, G. B Cellular resistance to infection. J. Exp. Med. 116: Mackaness, G. B The immunological basis of acquired cellular resistance. J. Exp. Med. 120: Mackaness, G. B The relationship of delayed hypersensitivity to acquired cellular resistance. Brit. Med. Bull. 23: Mackaness, G. B The influence of immunologically committed lymphoid cells on macrophage activity in vivo. J. Exp. Med. 129: Miyamoto, T., J. Kabe, M. Noda, N. Kobayashi, and K. Miura Physiologic and pathologic respiratory changes in delayed type hypersensitivity reaction in guinea pigs. Amer. Rev. Resp. Dis. 103: North, R. J The relative importance of blood monocytes and fixed macrophages to the expression of cellmediated immunity to infection. J. Exp. Med. 132: North, R. J Cellular kinetics associated with the development of acquired cellular resistance. J. Exp. Med. 130: Okita, G. T., J. J. Kabara, F. Richardson, and G. V. LeRoy Assaying compounds containing H3 and C". Nucleonics 15: Truitt, G. L., and G. B. Mackaness Cell-mediated resistance to aerogenic infection of the lung. Amer. Rev. Resp. Dis. 104: