Ultrastructural Study - Infection and Immunity

Vol. 29, No. 2

INFECTION AND IMMUNITY, Aug. 1980, p. 692-703 0019-9567/80/08-0692/12$02.00/0

Interaction of Pneumocystis carinii with Host Lungs: an Ultrastructural Study KOKICHI YONEDA AND PETER D. WALZER* Veterans Administration Medical Center, and Department of Pathology and Division of Infectious Diseases, Department ofMedicine, University of Kentucky College ofMedicine, Lexington, Kentucky 40507

Pneumocystis carinii pneumonia was produced in rats by the administration of corticosteroids, low (8%) protein diet, and tetracycline in the drinking water. The rats were sacrificed at weekly intervals, and their lungs were examined by electron microscopy. For the first 6 weeks, few alterations were noted in host pulmonary tissue, except a close attachment of P. carinii trophozoites to the type I pneumocytes. At 7 to 8 weeks, when the infection reached the peak intensity on light microscopy, degenerative changes occurred in the type I pneumocyte, beginning with subepithelial bleb formation and followed by denudation of the basement membrane. This denuded surface appeared to be the site both of exudation of serum and tissue fluid into the alveolar space and of spread of P. carinii into the interstitium. There was hypertrophy of type II pneumocytes, which also occurred in uninfected control rats ingesting tetracyclines. With tapering of the corticosteroid dose, P. carinii was slowly cleared from the lungs, but latent infection persisted for at least 21 weeks. The host response to the corticosteroid dose tapering included increased prominence of alveolar macrophages and progressive interstitial lymphocytic infiltrate and fibrosis. Thus, P. carinii interacts with, and is associated with damage to, specific host cells. This interaction is important in the host-parasite relationship in this infection. Pneumocystis carinii resides in the lungs of humans and a variety of other animals in nature. Under the conditions of immunosuppression, P. carinii causes serious and (if untreated) fatal pneumonia in humans (35). Although a similar disease can be produced in rats and other animals by the administration of corticosteroids (12, 27, 36), the basic host-parasite relationship in P. carinii infection remains poorly understood. There have been numerous light and electron microscopic studies of P. carinii in humans, experimental animals, and tissue culture (3, 4, 14, 16, 23, 26, 28, 29, 31, 33, 41). Interest has focused mainly on the morphology and life cycle of P. carinii, with little attention being devoted to the host. Recently, we conducted long-term light microscopy and quantitative studies of the sequential development of P. carinii pneumonia in rats administered full standard doses of corticosteroids and in rats whose corticosteroid dose had been tapered (38). In the present study we examine the ultrastructural changes in these rats with primary emphasis on host lung cells. We have also employed special histochemical stains to learn more about the P. carinii life-cycle and interaction with specific host cells. MATERIALS AND METHODS Experimental design. Seventy adult male Sprague-Dawley rats weighing about 250 g were di-

vided into groups outlined in Table 1. Group 1 rats received the standard treatment regimen of cortisone acetate (25 mg) injected subcutaneously twice weekly, low (8%) protein diet, and tetracycline (1 mg/ml) in the drinking water for 8 weeks, as described previously (38). Group 2 rats received the standard treatment regimen for 4 weeks; then a regular diet was instituted, and the cortisone dose was tapered to 0 over the next 3 weeks. These rats were followed for a total of 21 weeks. Rat groups 3 to 8 received variations of the standard treatment regimen and served as controls. These variations were chosen to account for the possible effects of substances used in the regimen (e.g., corticosteroids, diet, tetracycline) or other microorganisms on the observed ultrastructural changes. Group 9 rats, which were part of another study of the effects of oxytetracycline on the lungs (15), were included here as additional controls for comparison. The study period for all rat groups (except group 2 rats) comprised 8 weeks, because few group 1 rats on the standard treatment regimen survived beyond this time. Most of the rats were housed in standard plastic cages without filter tops in a conventional rat colony. Groups 7 and 8 consisted of germfree and defined bacterial flora rats raised and maintained in germfree isolators throughout the study. The rats ate food and drank tap water ad libitum, and were weighed at weekly intervals. The drinking water was changed twice weekly. Four group 1 and one to three group 2 rats were sacrificed each week by exsanguination under halothane anesthesia; members of control groups were sacrificed in a similar manner at varying intervals throughout the study. Morphological studies. At autopsy, the thorax 692

P. CARINII AND HOST LUNGS

VOL. 29, 1980

693

TABLE 1. Rat treatment regimens Group

Type of rat

Cortisone acetate'

was opened by a midline incision, and the trachea and lungs were excised en bloc. Some lung specimens were fixed by infusion of the fixative (5% glutaraldehyde in 0.2 M cacodylate buffer) through the canula in the trachea at a pressure of 10 cm of water; other specimens were immediately diced into small pieces and immersed in the fixative. The upper and lower lobes of the right lung and upper half of the left lung were bisected sagitally. A portion of the bisected lobe was processed for light microscopy and embedded in paraffin. Sections were stained with hematoxylin-eosin and Gomori methenamine silver stains. For electron microscopy, the tissue fixation was carried out for 2 h at 4VC. After several washings in the same buffer, the tissue blocks were postfixed in 1% osmium tetraoxide in 0.2 M cacodylate buffer. Some of the blocks were stained with ruthenium red for the glycocalyx (6, 20) and other blocks were stained with periodic acid-silver (9, 17). The tissue blocks were stained en bloc with 0.5% uranyl acetate in 0.1 M barbital buffer. The blocks were dehydrated through graded ethanols and embedded in epoxy resin. Sections (1 pum thick) were cut with glass knives and stained with toluidine blue. Selected blocks were sectioned with a diamond knife for electron microscopy. A Porter-Blum ultramicrotome was used for sectioning. The sections were stained with lead citrate and examined with a Philips 300 electron microscope. All the electron micrographs are those of sections stained with uranyl acetate and lead citrate unless designated otherwise. In this study the following forms of P. carinii have been defined. "Trophozoites" are small (1 to 4 pm), thin-walled, pleomorphic forms. "Cysts" are large (5 to 7 pum), thick-walled structures with up to eight daughter forms, termed "sporozoites."

Antibiotic'

Diet

Yes Low protein 1 Conventional 2 Conventional Yes - No Low protein 3 Conventional Yes Regular No Low protein 4 Conventional Conventional No 5 Regular 6 Conventional No Regular 7 Low protein Yes GF/DBFc No 8 DF/DBFc Regular No 9 Conventional Regular aTwenty-five milligrams injected subcutaneously twice weekly. b Concentration of antibiotics in drinking water was 1 mg/ml. Germfree and defined bacterial flora.

regular

Tetracycline Tetracycline Tetracycline Tetracycline Tetracycline None Tetracycline Tetracycline Oxytetracycline

peak by 7 to 8 weeks of corticosteroids. On electron microscopy Pneumocystis organisms were first easily recognizable at 2 weeks. P. carinii trophozoites were much more numerous than the cysts; both forms of the organism increased in number over the next several weeks, gradually filling alveoli (Fig. 1). The trophozoites lined up along the alveolar wall, especially on the surface of type I pneumocytes. Ruthenium red stain showed heavy stain deposits on the alveolar side of trophozoites and type I pneumocytes which were not covered by trophozoites (Fig. 2). No stain deposits were observed between trophoites and type I pneumocytes where they were in close contact, indicating a rather tight apposition of the cell membranes. The contact of trophozoites was on the smooth cell surface, and we did not observe any instance in which the filopodia of the organism were tightly attached to the type I pneumocyte. On the attached surface, the type I pneumocyte cytoplasm interdigitated with the membrane of the organism and extended between the organisms. At this stage, capillary endothelial cells showed increased pinocytotic vesicles. Beginning at 7 weeks, morphological changes became prominent in capillary endothelial cells and type I pneumocytes. Endothelial cells showed focal intracellular and subcellular vacuoles (Fig. 3). Although endothelial cell change was constantly present in the experimental animals at the peak of infection (7 to 8 weeks), a complete detachment of the endothelium from the basement membrane was not observed RESULTS throughout the experiment. On the alveolar epChanges in alveolar type I pneumocytes. ithelial surface, the first change observed was a (i) Rats on the standard treatment regimen bleb formation beneath the type I pneumocyte (group 1). Group 1 rats became chronically ill which was covered by trophozoites (Fig. 3). The and wasted. As described in detail previously bleb was not a complete detachment of cell (37, 38), the intensity of P. carinii infection on cytoplasm from the basement membrane, but light microscopy was judged by the number of portions of cytoplasm remained attached to the organisms in enzyme-digested lungs and a scor- basement membrane. After this, the detached ing system of lung sections. The intensity pro- portions of type I pneumocyte cytoplasm were gressively increased over time and reached its absorbed into the intraalveolar aggregates of the

694

YONEDA AND WALZER

r

_i

t

R/

+

INFECT. IMMUN.

Wi

j

ir4

5'

At -'7

,

#4s iti

-

yr...~~~~~~~~~~~. 44

_7

:A

~

.,

7

~r

FIG. 1. Pulmonary alveolus at 7 weeks of corticosteroids. Alveolar space is filled by trophozoites and cysts. Some trophozoites (arrow) line up against the type Ipneumocyte. x2,000.

B.9 r.

w:rI -*

w* 'N.? '7~

__ 1,rf

4. i.%t --

..

~ 4i ~ ~ ~ , ,5 =

*

*s

ii..

tV

'

..

FIG. 2. Alevolar surface at 7 weeks. Heavy stain deposits of ruthenium red are seen on the alveolar surface of trophozoite and type I pneumocyte (arrow). Notice the absence of stain between the trophozoites and type I pneumocytes where they are attached (double arrow). Also notice the extension of type I pneumocyte cytoplasm between the trophozoites. Ruthenium red stain. xl15,000.

;

695


VOL. 29, 1980

AS

f~~~~~~

p.

C

W?_i '__

fW

2

FIG. 3. Alveolar surface at 7 weeks. Notice the vacuolation of the type I pneumocyte cytoplasm (arrow) beneath the attached P. carinii trophozoite. AS, alveolar space; PC, P. carinii; CAP, capillary; END, endothelial cell. x12,000.

organisms, and the alveolar surface appeared partially denuded in a fenestrated fashion (Fig. 4). In the alveolar space, the trophozoites and cysts of P. carinii were embedded into two kinds of material (Fig. 5). One was tubulomyelin figures of 4-nm periodicity which are seen in normal mammalian lungs and considered to be pulmonary surfactant (13). The other was an irregular aggregate of unit membrane indicating degenerative cell membranes of host and organism tissues. By 8 weeks, the alveolar surface was partially denuded of epithelium, and the trophozoites were directly attached to the alveolar basement membrane (Fig. 6). The alveolar interstitium showed proliferation of fibroblasts and increased collagen fibers. A few trophozoites with intact filopodia were observed in interstitium in all group 1 rats (Fig. 7). (ii) Control Rats (groups 3 to 9). Mild-

moderate P. carinji infection was found in group 3 rats by 8 weeks of corticosteroids and regular diet, but this infection never attained the level of intensity found in group 1 rats, and degenerative changes in type I pneumocytes were not observed. Most other control rats raised in the conventional colony (groups 4 to 6 and 9) exhibited no P. carinii infection. Scattered foci of Pneumocystis organisms were found in a few of these iats, consistent with the subclinical carrier state in nature, but no degenerative changes in type I cells occurred. Rats maintained in isolates (groups 7 and 8) exhibited neither P. caring nor alterations in type I pneumocytes. Changes in other alveolar cells. Type II pneumocytes showed a steady increase in size of the lamellar bodies and cellular volume over time in all groups of rats except group 6, which did not receive a tetracycline in the drinking water. This type II cell hypertrophy, which was most pronounced in rats ingesting oxytetracy-

696

YONEDA AND WALZER

INFECT. IMMUN.

FIG. 4. Alveolar surface at 7 weeks. The changes here appear to follow the changes shown in Fig. 3. The alveolar surface is partially covered by the cytoplasm of the type I pneumocyte (arrow). AS, alveolar space; PC, P. carinii; CAP, capillary; END, endothelial cell. x12,000.

dine (group 9), has been described in detail previously (15). Alveolar macrophages were easily recognizable in group 1 rats and appeared to increase slightly in number over time. The morphology of the macrophages was unremarkable despite the fact that numerous P. carinii organisms were present adjacent to them. Macrophage organelles were normal in shape and number. The few phagosomes present contained membranous structures, suggesting digestion of organism cell membrane. Changes with clearance of P. carini from the lungs. Group 2 rats were indistinguishable from group 1 rats during the first 4 weeks of the study; with tapering of corticosteroid dose and institution of a regular diet, group 2 rats slowly regained their original weight and cleared P. carinii from the lungs. P. carinji cysts remained easily detectable until 13 weeks (i.e., 6 weeks off all corticosteroids); after this they could only be found with great difficulty. Although trophozoites decreased markedly in numbers, they could still be easily found attached to type I cells at 21 weeks, when the study was terminated.

Degenerative changes in type I pneumocytes were not observed, but there was hypertrophy of type II pneumocytes. Alveolar macrophages became considerably more numerous and prominent in the phagocytosis of P. carinii (Fig. 8). The macrophages contained P. carinji cysts within phagosomes. The phagosomes showed attached osmophilic granules, suggesting fusion of lysosomes with phogosomes. Changes in the pulmonary interstitium were characterized by the progressive development of a lymphocytic infiltrate and fibrosis, which reached peak intensity at 15 to 21 weeks (Fig. 9). Studies of P. carinii. On light microscopy at peak intensity of infection, the alveoli of methenamine silver-stained lung sections were filled with P. caring cysts and smaller, much more numerous, dark spherical bodies (Fig. 10). Electron microscopic studies were performed with periodic-acid silver stain to better define the relationship of P. carinii to the spherical bodies (Fig. 11). Heavy silver deposits were found in the walls of cysts, but no deposits were found intracellularly within cysts or in smaller trophozoites. The spherical bodies represented intra-

VOL. 29, 1980


697

FIG. 5. Pulmonary alveolus at 8 weeks. The alveolar space contains many trophozoites and cysts. Tubulomyelin figures (arrow) and aggregrates of the unit membrane (double arrow) are also seen.

cellular silver deposits in the electron-lucent areas of larger trophozoites. Infection with other organisms. As reported elsewhere (22), bacteria and fungi could be cultured from bronchial lavage fluid but were uncommon causes of pneumonia, as judged histopathologically by the presence of organisms or of acute exudative inflammation. Neither viral inclusions nor mycoplasmas were observed on electron microscopy. DISCUSSION Previous ultrastructural studies of P. carinii pneumonia have been almost exclusively concerned with the organism. Limited available data on the host indicate that Pneumocystis organisms line up along the alveolar wall during

the course of infection (4, 31). P. carinii maintains an intimate relationship with the alveolar lining cells, and it has been suggested that the organism obtains essential nutrients from these cells by metabolizing low-molecular-weight substances (4). The nature of the attachment of P. carinii to host cells is unknown. Although invasion of the alveolar lining cells has been suggested (28, 33), most authors have found that P. carinii remains extracellular, and, except for some nonspecific pinocytotic changes of capillary endothelial cells (31), produces no morphological alterations in host cells. The extracellular existence of P. carinji has also been confirmed by in vitro tissue culture as well as macrophage monolayer studies (21, 23, 24, 32). This study was primarily concerned with the

698

YONEDA AND WALZER

'L -n

INFECT. IMMUN.

-

FIG. 6. Alveolar septum at 8 weeks. Trophozoites cover the denuded alveolar surface (arrow). A trophozoite (double arrow) is seen in the interstitium adjacent to the capillary (CAP). x8,500.

host response to P. carinii. The corticosteroidtreated rat was chosen because it is the standard animal model for the infection and has been extensively used in the ultrastructural studies in the past. Yet, we have had only variable success with the standard regimen of cortisone acetate and antibiotics in producing Pneumocystis pneumonia in rats. The addition of a low (8%) protein diet (19) has improved the reproducibility of this regimen, shortened the time to reach peak infection, and increased the intensity of the infection. Protein depletion, however, has limited longterm studies, because few rats survived beyond 8 weeks. In the present study, the first interaction observed between the organism and host cells is

organism attachment. Although the filopodia have been proposed as an organ of attachment (4), we did not observe tight attachment of fiopodia to the alveolar lining cells. Instead, the smooth cell surface of trophozoite was closely attached to the type I pneumocyte, as revealed by the glycocalyx stain ruthenium red. Attachment to host cells appears to be important in the pathogenesis of less invasive organisms such as Neisseria gonorrheae (39). Degeneration of the type I pneumocyte followed this attachment of the organism, but the mechanism of degeneration is not clear. Beginning at 7 weeks, changes were seen in both endothelial and epithelial cells. The damaged endothelial cells appeared to regenerate

VOL. 29, 1980

FIG. 7. High magnification of the trophozoite in the interstitium. Notice a few filopodia (arrow).

x26,000.

rapidly, whereas the epithelial cell changes were more extensive. Of particular interest were the sequential changes of the type I pneumocyte. The first change was a subepithelial bleb, which led to partial denudation of the basement membrane. This denuded surface was covered by trophozoites. The foamy eosinophilic material seen on light microscopy in P. carinii pneumonia appeared to represent serum protein, degenerative cell membrane of the host and organism, and myelin-figures considered to be pulmonary surfactant (13). The presence of P. carinii trophozoites in the pulmonary interstitium raises the question of tissue invasion, presumably through the denuded basement membrane. Spread of P. carinii has rarely been reported in humans, mainly in patients with primary immune deficiency diseases (34). The frequency and mechanism of this dissemination in humans or experimental animals are unknown. In the present study, the fact that invasion was only observed late in infection is consistent with the low pathogenicity of P. carinii. This is in contrast to organisms such as Entamoeba histolytica, which can penetrate an


699

intact epithelial layer (30). Despite a careful search, we did not find P. carinii in pulmonary capillaries or lymphatics. A recent report suggests the presence of P. carinji in pulmonary capillaries (33); yet, we believe that their electron micrograph more likely represents part of a capillary endothelial cell cytoplasm rather than P. carinii. The importance of P. carinji in the ultrastructural changes in the type I pneumocyte noted here must be interpreted with caution, because this animal model is based on reactivation of latent infection rather than exogenous organism challenge. The spectrum of P. carinii infection in control rats ranged from a subclinical carrier state in otherwise healthy animals to moderate levels of infection induced by corticosteroids and regular diet. Yet, no such changes in the type I pneumocyte occurred in any of the controls. The fact that rats on the standard treatment regimen had greater numbers of Pneumocystis organisms in their lungs, the temporal association of the degenerative changes in type I pneumocytes with peak intensity of P. carinii, and the lack of evidence for parenchymal infection with other microorganisms all suggested a role for P. carinji in the production of alveolar cell injury. It is tempting to attribute the hypertrophy of type II pneumocytes (a typical host response to alveolar damage) to the changes in type I pneumocytes; however, the presence of type II cell hypertrophy in uninfected control rats which ingested tetracyclines precluded such an interpretation. The substitution of other antibiotics in the drinking water might overcome this problem. With the tapering of corticosteroid dose, P. carinii was slowly cleared from the lungs. On light microscopy, moderate levels of P. carinii infection detected by methenamine silver stain (which selectively stains the cell walls of cysts) persisted for 13 weeks and focal clusters of cysts could be found at 21 weeks (i.e., 14 weeks off corticosteroids) (38). In the present electron microscopy study the trophozoite form could also be detected at 21 weeks. These data suggest that P. carinii remains present as a latent infection long after normal immune function has been restored. Recently, prolonged treatment with trimethoprim-sulfamethorazole has failed to eradicate P. carinii from the lungs (18). The response of host alveolar cells to tapering of the corticosteroid dose was characterized primarily by increased prominence of macrophages, which became active in phagocytosis of P. carinii. The precise role of alveolar macrophages in host defense against P. carinii is unclear: some in vitro studies have suggested opsonizing antibody is necessary for phagocytosis of P. carinii

'

A

'. '

FIG. 8. Alveolar macrophage after corticosteroids had been tapered (week 12 of study). Notice intracellular P. carinii cyst. x8,500.

FIG. 9. Pulmonary alveolus after corticosteroids had been tapered (week 15 of study). Notice the focal thickening of the alveolar septum by collagen fibers. Trophozoites (arrow) are still recognizable, attached to the type Ipneumocyte. x8,500. 700

a..,.^_

f

rat

* fth

FIG. 10. Light microscope figure of methenamine silver-stained lung section. In addition to P. carinii cysts, notice the spherical bodies (arrow). x2,000.

I

I4.

a

14

I e-.

St-

S

'V

Saa

-'

U..

.%.4.

.1

.i .e. ,

'-'

I;

FIG. 11. Electron micrograph of lung section stained with periodic acid-silver. Silver deposits the cell wall of cysts and in intracellular locations in the trophozoites (arrow). xlI1,000. 701

A. ',

"A

h ~~~~~~~~~~~0.24' are seen in

702

YONEDA AND WALZER

(21), but others have not (32). The lack of degenerative changes in type I pneumocytes in these rats probably reflects the lighter degree of P. carinii infection achieved than in rats maintained on the standard treatment regimen for a full 8 weeks. The development of interstitial lymphocytic infiltrate and fibrosis with corticosteroid tapering supports the findings made on light microscopy; the significance of these findings has been discussed previously (38). Although this study was primarily concerned with the host-parasite relationship in P. carinii infection, some of the data obtained may be relevant to the life cycle of the organism. Deposits of silver stain were found by electron microscopy only within the walls of cysts and intracellularly within large trophozoites; this suggests that some cell wall constituents might be synthesized or stored at specific locations in the maturation process from trophozoite to cyst. Since the periodic-acid silver stain stains a number of chemical radicals (e.g., sulfhydryl groups, aldehyde groups, etc.) (40), further histochemical studies are needed to elucidate the nature of these silver deposits. Finally, the changes in the type I pneumocyte observed here may have relevance in other pulmonary conditions. Degeneration of the type I pneumocyte is generally considered to be a primary and nonspecific reaction to injury, including the use of oxygen, noxious gases, drugs, radiation, and bacterial infections (1, 2, 8, 10, 11, 25). In the present study, however, the type I pneumocyte degeneration appears to be secondary to subepithelial bleb formation and detachment from the basement membrane, presumably due to tissue edema in the presence of organism attachment (40). It would be of interest to determine whether this pattern of type I pneumocyte degeneration applies to other conditions but has not been observed because of the rapidity of the process. If this pattern of degeneration is a generalized process, experimental P. carinii pneumonia could provide an excellent model for the study of lung injury because of the slow evolution of the changes. ACKNOWLEDGMENTS This work was supported by a grant from the Medical Research Service of the Veterans Administration, an American Cancer Society Institutional Research Grant, and Public Health Service Biochemical Research Support Grant RR05374, Division of Research Facilities and Resources, National Institutes of Health. We gratefully acknowledge the excellent technical assistance of Carolyn S. Richey and Mary E. Rutledge.

LITERATURE CITED 1. Adamson, I. Y. R., and P. H. Bowden. 1974. The pathogenesis of bleomycin induced pulmonary fibrosis in mice. Am. J. Pathol. 77:85.

INFECT. IMMUN. 2. Aso, Y., K. Yoneda, and Y. Kikkawa. 1976. Morphological and biochemical study of pulmonary changes induced by bleomycin in mice. Lab. Invest. 35:558-568. 3. Barton, E. G., and W. G. Campbell. 1967. Further observations on the ultrastructure of pneumocystosis. Arch. Pathol. 83:527-534. 4. Barton, E. G., and W. G. Campbell. 1969. Pneumocystis carinii in the lungs of rats treated with cortisone acetate. Am. J. Pathol. 54:209-236. 5. Brody, A. R., P. C. Kelleher, and J. E. Craighead. 1978. A mechanism of exudation through intact alveolar epithelial cells in the lungs of cytomegalo-virus infected mice. Lab. Invest. 39:281-288. 6. Brooks, R. E. 1969. Ruthenium red stainable surface layer on lung alveolar cells: electron microscopic interpretation. Stain Technol. 44:173. 7. Campbell, W. G. 1972. Ultrastructure of pneumocystis in human lungs: life cycle of human pneumocystosis. Arch. Pathol. 93:312-324. 8. Coalson, J. J., J. J. Beller, and L. J. Greenfild. 1971. Effects of 100 per cent oxygen ventilation on pulmonary ultrastructure and mechanics. J. Pathol. 104:267-273. 9. Dhir, S. P., and E. S. Boatman. 1972. Location of polysaccharide on Chlamydia psittaci by silver methenamine staining and electron microscopy. J. Bacteriol. 111:267-271. 10. Evans, M. J., L. J. Cabla, R. J. Stephens, and G. Freeman. 1973. Renewal of alveolar epithelium in the rat following exposure to NO2. Am. J. Pathol. 70:175190. 11. Faulkner, C. S., and K. S. Connolly. 1973. The ultrastructure of COw radiation pneumonitis in rats. Lab. Invest. 28:454-463. 12. Frenkel, J. K., J. T. Good, and J. A. Schultz. 1966. Latent pneumocystis infection of rats, relapse and chemotherapy. Lab. Invest. 15:1559-1577. 13. Gil, J., and E. R. Weibel. 1969. Improvements in demonstration of lining layer of lung alveoli by electron microscopy. Respir. Physiol. 8:13-36. 14. Gold, J., P. L'Henreux, and I. P. Debner. 1977. Ultrastructure in the differential diagnosis of pulmonary histiocytosis and pneumocystosis. Arch. Pathol. Lab. Med. 101:243-247. 15. Gottschall, J. L., P. D. Walker, and K. Yoneda. 1979. Morphological changes of the rat type II pneumocyte induced by oxytetracycline. Lab. Invest. 41:5. 16. Ham, E. H., S. D. Greenberg, R. C. Reynolds, and D. B. Singer. 1971. Ultrastructure of Pneumocystis carinii. Exp. Mol. Pathol. 14:362-372. 17. Hayat, M. A. 1975. Positive staining for electron microscopy, p. 124-146. Van Nostrand Rheinhold Co. New York. 18. Hughes, W. T. 1979. Limited effect of trimethoprimsulfamethoxazole prophylaxis on Pneumocystis carinii. Antimicrob. Agents Chemother. 16:333-335. 19. Hughes, W. T., R. A. Price, F. Sisko, W. S. Havron, A. Kefatos, M. Schonland, and P. M. Smythe. 1974. Protein caloric malnutrition. Am. J. Dis. Child. 128:4453. 20. Luft, J. H. 1964. Electron microscopy of cell extraneous coats as revealed by ruthenium red staining. J. Cell. Biol. 23:54A. 21. Masur, H., and T. C. Jones. 1978. The interaction in vitro of Pneumocystis carinii with macrophages and Lcells. J. Exp. Med. 147:157-170. 22. Milder, J. E., P. D. Walker, J. C. Coonrod, and M. E. Rutledge. Comparison of histological and immunological techniques for the detection of Pneumocystis carinii in rat bronchial lavage fluid. .J. Clin. Microbiol. 11: 409-417. 23. Murphy, M. J., L. L. Pifer, and W. T. Hughes. 1977. Pneumocystis carinii in vitro. Am. .J. Pathol. 88:387402. 24. Pifer, L., W. T. Hughes, and M. J. Murphy. 1977.

VOL. 29, 1980

25.

26. 27.

28. 29.

30.

31.

32. 33.

Propagation of Pneumocystis carinii in vitro. Pediatr. Res. 11:305-316. Pine, J. H., W. R. Richter, and J. R. Esterly. 1973. Experimental lung injury. I. Bacterial pneumonia, ultrastructural, autoradiographic and histochemical observations. Am. J. Pathol. 23:115-130. Price, R. A., and W. T. Hughes. 1974. Histopathology of Pneumocystis carinii infestation and infection in malignant disease in childhood. Human Pathol. 5:737. Sheldon, W. H. 1959. Experimental pulmonary Pneumocystis carinii infection in rabbits. J. Exp. Med. 110: 147-160. Shivley, J. N., K. K. Moe, and R. W. Dellers. 1974. Fine structure of spontaneous Pneumocystis carinii pulmonary infection in foals. Cornell Vet. 64:72-88. Sueishi, D., S. Hisano, A. Sumiyoshi, and K. Tanaka. 1977. Scanning and transmission electron microscopic study of human pulmonary pneumocystosis. Chest 72: 213-216. Takeuchi, H., and B. P. Phillips. 1975. Electron microscopic studies of experimental Entamoeba histolytica infection in the guinea pigs. I. Penetration of the intestinal epithelium by trophozoites. Am. J. Trop. Med. Hyg. 24:34-48. Vavra, J., and K. Kucera. 1970. Pneumocystis carinii Delanoe, its ultrastructure and ultrastructural affinities. J. Protozool. 17:463-483. Von Behren, L. A., and E. L. Pesant. 1978. Uptake and depredation of Pneumocystis carinii by macrophages in vitro. Am. Rev. Respir. Dis. 118:1051. Vossen, M. E. M. H., P. J. A. Beckers, J. H. T. U. Menwissen, and A. M. Stadboulders. 1978. Devel-


703

opmental biology of Pneumocystis carinii, and alternative view on the life cycle of the parasite. Z. Parasitenkd. 55:101-118. 34. Walzer, P. D. 1977. Pneumocystis carinii infection: a review. South. Med. J. 70:1130-1337. 35. Walzer, P. D., D. P. Perl, D. J. Krogstad, P. G. Rawson, and M. D. Shultz. 1974. Pneumocystis carinii pneumonia in the United States. Ann. Intern. Med. 80: 83-93. 36. Walzer, P. D., R. D. Powell, and K. Yoneda. 1979. Experimental Pneumocystis carinii pneumonia in different strains of cortisonized mice. Infect. Immun. 24: 939-947. 37. Walzer, P. D., M. E. Rutledge, K. Yoneda, and B. J. Stahr. 1979. A new method of separating Pneumocystis carinii from infected lung tissue. Exp. Parasitol. 47:356368. 38. Walzer, P. D., K. Yoneda, R. D. Powell, M. E. Rutledge, and J. E. Milder. 1980. Growth characteristics and pathogenesis of experimental Pneumocystis carinit pneumonia. Infect. Immun. 27:928-937. 39. Watt, P. J., and M. E. Ward. 1977. The interaction of gonococci with human epithelial cells, p. 354. In R. B. Roberts (ed.), The gonococcus. John Wiley and Sons, New York. 40. Wang, N. S., S. N. Huang, H. Sheldon, and W. M. Thurlbeck. 1971. Ultrastructural changes of Clara and type II alveolar cells in adrenaline-induced pulmonary edema in mice. Am. J. Pathol. 62:237-246. 41. Wang, N. S., S. W. Huang, and W. M. Thurlbeck. 1970. Combined Pneumocystis carinii and cytomegalovirus infection. Arch. Pathol. 90:529-535.