Growth Characteristics and Pathogenesis of ... - Europe PMC

2 downloads 0 Views 2MB Size Report
RUTLEDGE,1'2 AND JAMES E. MILDER" 2 ..... We thank Ward Bullock and J. Donald Coonrod for review- ... Lim, S. K., W. C. Eveland, and R. J. Porter. 1973.
INFECTION AND IMMUNITY, Mar. 1980, p. 928-937 0019-9567/80/03-0928/10$02.00/0

Vol. 27, No. 3

Growth Characteristics and Pathogenesis of Experimental Pneumocystis carinii Pneumonia PETER D. WALZER,l* RALPH D. POWELL, JR.,'2 KOKICHI YONEDA,"' MARY ELLEN RUTLEDGE,1'2 AND JAMES E. MILDER" 2 Veterans Administration Medical Center, 1 and the Division of Infectious Diseases, Departments of Medicine and Pathology, University of Kentucky College of Medicine,2 Lexington, Kentucky 40507

Pneumocystis carinii pneumonia was produced in two groups of rats by the administration of corticosteroids, a low-protein (8%) diet, and tetracycline in the drinking water. A third group not on corticosteroids or a low-protein diet served as controls. Members of the first group were sacrificed weekly for 8 weeks, and lungs were examined. A highly significant correlation was found between the histopathological assessment of the intensity of P. carinii infection and the number of cysts counted in enzyme-digested lungs. P. carinii progressively filled alveoli, and cyst counts increased from -104 to 109 cysts/g of lung at peak intensity of infection at 7 to 8 weeks. The second group of rats was placed on a regular diet and tapering doses of corticosteroids after week 4, and they were sacrificed at varying intervals for up to 21 weeks. P. carinii was not cleared from the lungs until after week 13 (more than 6 weeks after discontinuation of all steroids). Histologically, there was an increased prominence of alveolar macrophages and the progressive development of interstitial mononuclear cell infiltrate and fibrosis. Thus, P. carinii grows slowly in vivo and interacts with specific host cells. The resulting changes may be important in the pathogenesis of the infection and in the clearance of the organism from the lung after immunocompetence has been restored. Pneumocystis carinii, an organism of low virulence, is an important cause of pneumonia in immunocompromised patients (30). The hostparasite relationship in this infection is poorly understood. Animal models of P. carinii pneumonia have traditionally been based on the reactivation of latent infection by the administration of corticosteroids. Rats and, to a lesser extent, rabbits and mice have been used for this purpose (5, 24, 31, 35, 36). Recently, exogenous P. carinii infection has been transmitted to athymic (nude) mice (26, 33); however, the light intensity of infection so far achieved suggests that this animal model might be more useful in studies of epidemiology rather than of pathogenesis. Most studies of experimental Pneumocystis

infection in rats administered corticosteroids and a low-protein diet with an emphasis on the following: (i) developing methods to quantitate Pneumocystis organisms in lungs; (ii) correlating this quantitation with pathological changes in host lungs; and (iii) studying the long-term effects in the host after the steroid dose has been tapered. MATERIALS AND METHODS Rats. Adult male Sprague-Dawley rats weighing about 250 g obtained from ARS Sprague Dawley

(Madison, Wis.) and Harlan Industries (Indianapolis, Ind.) were pooled for use in this study. The rats were divided into the following groups. Group A rats (controls) received no corticosteroids, ate a regular diet, and drank tap water with or without tetracycline (1 mg/mil). Group B rats received the standard treatment regimen of cortisone acetate (Cortone [Merck Sharp & Dohme, West Point, Pa.]), 25 mg, injected subcutaneously twice weekly, low protein (8%) diet, and tetracycline in their drinking water for 8 to 9 weeks or until death. Group C rats received the standard treatment regimen for 4 weeks; then a regular diet was substituted and the steroid dose was tapered to twiceweekly cortisone injections of 12.5 mg during week 5, 6.25 mg during week 6, 3 mg during week 7, and none during week 8. These rats were observed for a total of 21 weeks.

infection in cortisonized rat models have focused on the morphology, antigenic characteristics, and life cycle of P. carinii (1, 7, 13, 16, 17, 27, 29) or on the development of new treatment modalities (5, 8, 9, 14, 37). Little is known about the quantitative aspects of the infection or growth characteristics of the organism. Scant attention has also been paid to changes in the host or to certain host factors (e.g., protein-calorie malnutrition) which enhance the development of P. carinii infection (10). We have reexamined experimental P. carinii 928

VOL. 27, 1980

EXPERIMENTAL PNEUMOCYSTIS CARINII PNEUMONIA

The rats were weighed at weekly intervals. Group A rats were sacrificed by exsanguination under deep halothane anesthesia at varying intervals throughout the study. At least six group B and one to three group C rats were sacrificed at weekly intervals in the same manner. At autopsy the lungs were aseptically removed and processed for light microscopy. The remaining portions of lungs were stored at -70oC. Bronchial lavage was performed by the method of Masur and Jones (18); lavage fluids were processed for histological staining and bacterial and fungal cultures as described elsewhere (J. E. Milder, P. D. Walzer, J. D. Coonrod, and M. E. Rutledge, J. Clin. Microbiol., in press). Mice. We have recently reported the development of Pneumocystis pneumonia in different strains of steroid-treated normal mice (31). The same procedures for histological examination and quantitation of P. carinii in lungs have been used for rats and mice. The mice are included here to validate the procedures

developed. Light microscopy. Lung sections prepared by standard techniques were stained with methenamine silver and hematoxylin and eosin and, in selected instances, by Masson trichrome methods. A standardized procedure was devised for grading the intensity of P. carinii infection in rats and mice. At least three blocks of lung were removed for histopathological examination: one each from the upper and lower portions of the right lung and one from the middle portion of the left lung. The lung sections were stained with methenamine silver, coded, and read blindly. The following scoring system for the infection was used: 0, no P. carinii found; 0.5+, minimal infection, 75% alveoli involved. Quantitation of P. carinii Procedures for separation and quantitation of P. carinii in lungs have been previously described in detail (32). Briefly, portions of both lungs of a rat or mouse were pooled and cut up into small pieces, air dried in a laminar flow hood at room temperature for 1 h, weighed, ground through a no. 60 mesh-wire screen with Teflon pestle, washed with Hanks balanced salt solution, digested with 0.2% collagenase and 0.2% hyaluronidase, and then washed again and suspended in 1 ml of Hanks balanced salt solution. Two 1-pl drops of this material were placed on a slide, air dried, and then stained with cresyl echt violet (2). All cysts in each drop were counted and the mean of the counts was taken as the final cyst count (per microliter). If there were too many cysts in drops for accurate counting, an appropriate dilution was made. The following calculations were performed: (number of cysts/microliter) X dilution x i03 = number of cysts/milliliter; (number of cysts/milliliter)/weight of lung tissue (milligrams) = number of cysts/milligram; (number of cysts/milligram) x 103 = number of cysts/gram of lung tissue

929

Cresyl echt violet stained the cell wall of P. carinii pinkish-purple against a yellow-white (unstained) background of host cells and tissues. Variation in the staining intensity of P. carinii was sometimes encountered. This variation was minimized by careful selection of the source of the stain and the use of freshly prepared reagents. To test the accuracy and reproducibility of quantitation, portions of lungs of a heavily infected rat were cut up, pooled, and then divided into four samples. Each fraction was processed, and cysts were quantitated as described above. The mean ± standard error cyst count for the four fractions was 8.61 ± 0.60 x 108 cysts/g of lung. One of the fractions was then counted 11 different times. The mean ± standard error cyst count for this fraction was 9.32 ± 0.77 x 108 cysts/g of lung. Quantitation by Giemsa, which stained both trophozoite and cyst forms of P. carinii, was unsatisfactory. The trophozoites were small, tended to clump, and were often confused with platelets or debris; thus, they could not be counted accurately. Whereas mature cyst forms were usually easily recognizable, intermediate forms were difficult to distinguish from other cells. Giemsa stain yielded an underestimation of cysts when compared with cresyl echt violet stain. Statistics. The nonparametric Kendall Tau test (4, 25) was used to determine the correlation between the histological assessment of the intensity of P. carinii infection and the quantitative cyst count.

RESULTS Clinical course. Figure 1 depicts the serial weight changes in different groups of rats. The control group (group A) rats appeared healthy and had increased weights by 40 to 45% after week 8. Rats on the standard treatment regimen (group B) became chronically ill and wasted and lost about 35% of their original weight. Few of these rats survived beyond week 8. Group C rats also lost weight on the standard treatment regimen for the first 4 weeks; their weights plateaued when the regular diet was substituted and the steroid dose was tapered; they began to gain weight after week 8 and had reached their original weights by week 11. Growth characteristics of P. carinii The growth of P. carinii is shown in Fig. 2 and 3. Some control rats (group A) had minimal infection at the beginning of the study. Control rats sacrificed at varying intervals up to week 17 in the colony did not show any greater tendency to acquire the infection over time. The presence of tetracycline in the drinking water did not enhance susceptibility to P. carinii. The intensity of P. carinii infection in the group of rats on the standard treatment regimen (group B) slowly increased over time (Fig. 2); however, the intensity of the infection varied somewhat among these rats sacrificed at any

930

INFECT. IMMUN.

WALZER ET AL.

given time interval. P. carinii infection was light to moderate in most rats by week 4, moderate to heavy at week 6, and at peak intensity at weeks

7 to 8. P. carinii infection in group C rats beginning at week 5 is shown in Fig. 3. Although no steroids were administered after week 7, the intensity of P. carinii infection of group C rats did not diminish and perhaps increased slightly. These rats remained moderately infected through week 13 (i.e., 6 weeks after discontinuation of all steroids). After this time, the infection was classified as minimal because the Pneumocystis organisms were few in number and had usually lost their characteristic morphology. Yet even at 21 weeks rare clusters of well-preserved organisms could be found. Quantitation of P. carinii. Quantitation studies were performed in 34 rats and 29 mice, and the results are outlined in Fig. 4. There was a highly significant correlation between the histological assessment of the intensity of P. carinii infection and the quantitative cyst count for rats (P < 0.001), mice (P < 0.001), and both groups overall (P < 0.001). The minimum detectable cyst count was 104/g (i.e., 10/mg) of lung tissue. There was some overlap in cyst counts among different scores of the intensity of the infection. Median cyst counts (per gram of lung) were: 0.5+, 4.27 x 105; 1+, 1.20 x 107; 2+, 2.10 x 107; 3+, 4.90 x 107; 4+, 3.14 x 108. Histopathology during steroid administration. The progress of P. carinii infection in group B rats on the regular treatment regimen was followed best with methenamine silverstained lung sections. In minimal or light infection, Pneumocystis organisms were found individually or in small groups along the walls of the alveoli (Fig. 5A). As the infection progressed, more alveoli became involved and began to be filled with clumps of organisms (Fig. 5B and C). The degree of alveolar involvement varied somewhat in different areas of the lungs. At peak intensity of the infection (weeks 7 to 8) alveoli were almost completely filled with P. carinii cysts and smaller, much more numerous, dark

400 K A

.-3' H..

z w LI)

8 10 12 14 16 18 20 WEEKS FIG. 1. Sequential changes in weights of rats over time. Each point represents the mean ± 1 standard error of the mean weight of the group of rats. Group A, controls. Group B, standard treatment regimen of corticosteroids (represented by hatch marks) low-protein diet, and tetracycline in the drinking water. Group C, standard treatment regimen for 4 weeks, and then on a regular diet and tapering doses of steroids (hatch marks). 2

4

6

4+

,,

3+

0 (

Fr

2+ 1+ _

C-,I

X: 0 5+

_

0-00

0

00

9

10

0 0

2

3

4

5

6

7

8

11

12

WEEKS

FIG. 2. Sequential changes in intensity of P. carinii infection. 0, Group B rats on standard treatment regimen. 0, Group A (control) rats which drank plain tap water. A, Group A (control) rats which drank water with tetracycline.

EXPERIMENTAL PNEUMOCYSTIS CARINII PNEUMONIA

VOL. 27, 1980

931

4+ r w 0

3+

A

2+

AA

LA

A

M.

0

-_AA

A

A

C,)

z

1+

a:

0.5+_

A

A

AL

A

0

/liZW7 -1

!-

5

6

8

7

9

11 Iu WEEKS

12 ".

,,, .1. .. 13 15

e,

..; 17

",

I19

-V

21

FIG. 3. Sequential changes in intensity ofP. carinii infection. A, Group C rats on tapering doses ofsteroids (represented by hatch marks). 0 and A, Group A (control) rats as in Fig. 2. 1010

x x

109

K

x

x j1o0

xx

x x

x x

o

Io

x

In hematoxylin- and-eosin-stained lung sections, there were few noticeable changes in alveoli early in the infection. Later, alveoli became filled with typical foamy, eosinophilic material which was largely acellular. This coincided with the presence of masses of P. carinii in methenamine silver-stained sections and was accompanied by hypertrophy of alveolar lining cells (Fig. 6). Alveolar macrophages were first noticeable at week 3 and became somewhat more numerous over the next several weeks. A mild mononuclear cell interstitial infiltrate was present throughout administration of the standard treatment regimen.

Histopathology during tapering of steroid dose. With group C rats, alveolar macrophages became considerably more prominent, 3 10? o and some multinucleated forms were seen. By CP weeks 8 to 10 (i.e., 1 to 3 weeks after discontin"I-. 0 8 uation of all steroids), the alveolar eosinophilic cn material appeared to be composed predomiCO _° x nantly of macrophages with foamy vacuolated 106 cytoplasm. Alveolar cell hypertrophy persisted. The most striking changes in this group of o rats were the development of an interstitial mononuclear cell infiltrate and fibrosis. The x mononuclear cell infiltrate was both perivascular and peribronchiolar, it was quite prominent at weeks 9 to 10 and reached peak intensity at x x weeks 15 to 21 (Fig. 7). The infiltrate was composed primarily of lymphocytes. Occasionally the infiltrate assumed a nodular character composed of macrophages, polymorphonuclear leu104 '___________________________ .kocytes, and proliferating epithelial cells. Rarely was there evidence of focal necrosis. I Interstitial fibrosis was evident by both he4+ 4+ 2 0 0.5I.5 3 matoxylin and eosin and trichrome stains (Fig. P FIG. 44. Correlation between histopathological 8). The fibrosis first appeared at weeks 9 to 10 evaluatiLon of the intensity of P. carinii infection and and was focal; it became more prominent and quantityitive cyst count for rats (x) and mice (0). The widespread over time. In rats with the nodular inflammatory response, the fibrosis tended to be horizonstal bars represent median cyst counts. adjacent to the nodules and was accompanied sphericial bodies (Fig. 5D). These findings were by hyalinization and bronchiolectasis. Infection with other microorganisms. most prrenounced in rats which died from their Cultures of bronchial lavage fluids grew a variety infection n. Cn z

w__

0

*1

bI

4

A, 0 St

A 0

SoI*

AWt

AVtV-

Oh

-L

-t

-'p

aftM

:*

_t

r

*0 t:if

Wo..

r,

W-

7;

a

It

-.; .11 1k:

t

.-12

i

r,

.-

I

. -

f

B

'.IP

.0

4k.

.

.

-%

It

S. .*

r at

r

I}

a

W.

I.

-,

p

FIG. 5. Sequential histological changes in P. carinii infection. Methenamine silver stain. (A) Minimal infection at about week 1 of steroid administration. Note that P. carinii cysts line up along walls of alveoli (x1250). (B) Light-moderate P. carinii infection at week 4 of steroids (x500). (C) Moderate-heavy P. carinii infection at week 6 of steroids (x500). (D) Very heavy P. carinii infection in rat which died at week 8 of steroids. Note that alveoli are virtually filled by masses of organisms (x500). 932

:,~ ib.*

933

EXPERIMENTAL PNEUMOCYSTIS CARINII PNEUMONIA

VOL. 27, 1980

;i

*

A

"'ai

4'

t

i _,

6

,

t1

-;_.

_

Y.* 14

&

v

eV'i

;

.

~~~~~~~~~

~

#-S^

r,

fi

V

'

V2,,

A

I

..t

1

ut .

Sf.A -u-,

P 9

*f .1. _

%-

j it

,,

It

.

CO ~ ~,*~ ~ ~

~~4~

J.

.0,

.

;

*is

Il, C.

A614

A'

., \

40

"

g

0'

,

As

,0, C)

'r

'A

S.

9.z41 4,

FIG. 5 C-D.

..

934

WALZER ET AL.

Iw ;_

M.,'.'i $. i. 1

ft.

410

AO

.I

4w.

^~ w

,JI

0

it

I~~~4

t4E

^

INFECT. IMMUN.

4

e

S

*

..41t

S

*~~

At tI

F.

T A:

,X

J

FIG. 6. Hypertrophic alveolar lining cells prominent along alveolar walls of rat sacrificed after week 9 of full doses of corticosteroids. Within alveolar lumens are foamy exudate (long arrow) and alveolar macrophages (short arrow). (Hematoxylin and eosin stain, x620.)

of bacteria, and detailed results have been published elsewhere (Milder et al., in press). The predominant organism was Flavobacterium meningosepticum. However, bacterial pneumonia was uncommon and was accompanied by an acute inflammatory exudative response. Fungi (e.g., Penicillium sp.) could also be cultured from bronchial lavage fluids but rarely caused parenchymal lung infection. When present on methenamine silver-stained lung sections, fungi could easily be distinguished from P. carinii. Viral inclusions were not observed on light microscopy.

DISCUSSION The principal animal model for P. carinii infection has been the corticosteroid-treated rat. A major problem with this system has been the lack of data about the number of Pneumocystis organisms in infected lung tissue. Most authors have used descriptive terms (e.g., light, heavy) or an arbitrary numerical scoring system (e.g., 1-4+) to evaluate the intensity of P. carinii infection (3, 9, 14). Although these methods may accurately reflect the status of the infection in a group of lungs, it has been difficult to compare

directly the results of one study with those of another. In the present study there was a highly significant correlation between the histopathological assessment of the intensity of P. carinii infection and the quantitative cyst count in rat and mouse lungs. Cyst quantitation underestimates the number of P. carinii organisms present in tissue preparations because the trophozoite form is much more numerous than the cyst form. Nevertheless, on the basis of technical difficulties involved in counting trophozoites, we and others (20) believe that cyst quantitation by the use of selective cell wall stains (e.g., methenamine silver, toluidine blue, cresyl echt violet) currently is the most reliable marker for the number of Pneumocystis organisms. These stains are not technically difficult to perform, but considerable skill is necessary to avoid counting errors resulting from variations in the intensity of the staining. Our approach to quantitating P. carinii in lungs also differed from that used to quantitate bacteria or fungi, which can be readily grown on standard culture media. Since we were interested in correlating histology with cyst quantitation in the same animal, we sampled portions

EXPERIMENTAL PNEUMOCYSTIS CARINII PNEUMONIA

VOL. 27, 1980

*,O

I

*

e

it a

935

I..

if

i &

e~~~~~~f

,

A_

.'

f~~~~~~~~~~~~~~~~~~~~~~A

$ pi4, f

At

,

40.

,

*

.

;.Y #

#

*o ,Ip.

. t

-No,

P. *Ad

V

*

e0 6

K'~

4;14.\A*

?4*

I

1* 4 A-*-

,.

I 7r FIG. 7. Mononuclear cell infiltrate in interstitial space surrounding small vessels of rat sacrificed after week 15. (Hematoxylin and eosin stain, x400.)

of lungs and processed them in a uniform manner. Studies of experimental coccidiomycosis in mice have used whole lung homogenates for cultures and included other factors (e.g., changes in lung weight over time) in analysis of the results (11). The intensity of P. carinji infection in our rats on steroids and low-protein diets slowly and steadily progressed over time. Cyst counts increased from s104 cysts (the minimum level of detection) to 109 cysts/g of lung at peak intensity of the infection by weeks 7 to 8. Ogino et al. have also recently noted an increase in P. carinii cyst counts over time in steroid-treated rats, but peak counts were 10-fold lower than those in our rats (12, 19). These authors used a different quantitative technique and did not include a low-protein diet. Thus, it may be possible to develop crude in vivo growth curves for the organism. Some work in this regard has been performed in tissue culture (15, 20, 21), but only short-term cultivation of P. carinii has been achieved. Growth kinetics of experimental systemic Mycobacterium lepraemurium infection in mice have been studied based on microscopic quantitation of acid-fast bacilli (3). Histopathologically, P. carinii maintained an

intimate relationship with alveolar lining cells, type I pneumocytes, but induced no cellular damage detectable by light microscopy. In electron microscopic studies detailed elsewhere (K. Yoneda and P. D. Walzer, manuscript submitted for publication), degenerative changes occurred in type I cells at peak intensity of the infection. The alveolar cell hypertrophy was due to hypertrophy of type II cells; since we had previously found type II cell hypertrophy in uninfected control rats ingesting tetracyclines (6), its relationship to P. carinii is unclear. The intensity of P. carinii infection initially did not diminish with the institution of steroid dose tapering and a regular diet; moderate levels persisted for at least 6 weeks after corticosteroids had been discontinued. The similar delay experienced by the rats in regaining weight with steroid dose tapering is further evidence of the profound effects of the combination of steroids and a low-protein diet. In the only comparable pathological study, Frenkel et al. administered steroids and regular diets to rats for 5 weeks and then abruptly discontinued the steroids; after week 3, P. carinii had been almost completely cleared from the lungs(5). The increased prominence of alveolar macro-

936

WALZER ET AL

An. ~ 4

INFECT. IMMUN.

~

J

i

v

FIG. 8. Interstitial fibrosis (arrows) involving peribronchial and periarterial spaces and alveolar walls of rat sacrificed at 15 weeks. (Trichrome stain, x400.)

phages in our rats with steroid dose tapering suggests that these cells are important in removing P. carinji from alveoli. Yet P. carinii continued to proliferate despite the presence of macrophages throughout the period of administration of steroids and a low-protein diet. In vitro studies have shown no differences in the interaction between P. carinii and alveolar macrophages from steroid and non-steroid-treated rats: phagocytosis of the organism occurs only under certain culture conditions or in the presence of opsonizing antibody (18, 28). More data about the interaction of steroids, protein-calorie malnutrition, B and T lymphocytes, and macrophages are needed to help clarify the specific host defense mechanisms in P. carinii infection. The most dramatic changes with the tapering of steroid doses occurred in the pulmonary interstitium. The mononuclear cell response appeared more extensive than that reported by Frenkel et al. (5). In a few instances, cessation of corticosteroid treatment has been followed by the development of serum antibodies to P. carinii (5, 17). However, the nature of these antibodies or of other host humoral immune responses to the organism has not been studied. The development of interstitial fibrosis in our rats is important because it suggests a greater

degree of host-parasite interaction in this infection than has been generally appreciated. Interstitial fibrosis has been observed in a number of patients with Pneumocystis pneumonia (22, 24, 38) and in a few cortisonized rats treated with pentamidine (5). Recent pathological studies have emphasized a variety of atypical manifestations (e.g., granulomas) in humans and, rarely, in animals with P. carinji pneumonia (26, 34). Since patients often receive a variety of insults to the lung (e.g., radiation, oxygen, cancer chemotherapy) or may have coexistent infection, the precise role of P. carinii in any of these changes is unclear. In the cortisonized rat model, the influence of steroids themselves or of latent infection with other microorganisms on the development of fibrosis cannot be ruled out. Nevertheless, the increase in fibrosis in our rats over time and its correlation with other host inflammatory responses and clearance of the organisms support the argument that recovery from P. carinii pneumonia can be associated with interstitial fibrosis. ACKNOWLEDGMENTS This work was supported by the Veterans Administration; by an American Cancer Society Institutional Research Grant; and by Biomedical Research Support grant RR05374, Division

VOL. 27, 1980

EXPERIMENTAL PNEUMOCYSTIS CARINII PNEUMONIA

of Research Facilities and Resources, National Institutes of Health. This work was presented in part at the National Meetings of the Association of American Physicians, American Society for Clinical Investigation, and American Federation for Clinical Research, Washington, D.C., May 1979. We thank Ward Bullock and J. Donald Coonrod for reviewing the manuscript.

LITERATURE CITED 1. 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. 2. Bowling, M. D., L. M. Smith, and S. L. Wescott. 1973. A rapid staining procedure for Pneumocystis carinii. Am. J. Technol. 39:267-268. 3. Brown, L. N., and H. N. Krenzien. 1976. Systemic Mycobacterium lepraemurium infection in mice: difference in doubling time in liver, spleen, and bone marrow, and a method for measuring the proportions of viable organisms in an inoculum. Infect. Immun. 13:480-486. 4. Conover, W. J. 1960. Practical nonparametric statistics. John Wiley & Sons, Inc., New York. 5. Frenkel, J. K., J. T. Good, and J. A. Schultz. 1966. Latent Pneumocystis infection of rats, relapse, and chemotherapy. Lab. Invest. 15:1559-1577. 6. Gottschall, J. L., P. D. Walzer, and K. Yoneda. 1979. The morphologic changes in the rat type II pneumocyte induced by oxytetracycline. Lab. Invest. 41:5-12. 7. Hendley, J. O., and T. H. Weller. 1971. Activation and transmission in rats of infection with Pneumocystis carinii. Proc. Soc. Exp. Biol. Med. 137:1401-1404. 8. Hughes, W. T., H. K. Kim, and R. A. Price. 1973. Attempts at prophylaxis for murine Pneumocystis carinii pneumonitis. Curr. Ther. Res. Clin. Exp. 15:581587. 9. Hughes, W. T., P. C. McNabb, and T. D. Makres. 1974. Efficacy of trimethoprim and sulfamethoxazole in the prevention and treatment of Pneumocystis carinii pneumonitis. Antimicrob. Agents Chemother. 5:289293. 10. 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. 128:4452. 11. Huppert, M., S. H. Sun, I. Gleason-Jordan, and K. R. Vukovich. 1976. Lung weight parallels disease severity in experimental coccidiomycosis. Infect. Immun. 14: 1356-1368. 12. Ikai, T., Y. Yoshida, K. Ogino, S. Takeuchi, and M. Yamada. 1977. Studies on Pneumocystis carinii and Pneumocystis carindi pneumonia. II. Method for concentration and quantitation of P. carinii cysts. Jpn. J. Parasitol. 13. Kim, J. K., W. T. Hughes, and S. Feldman. 1972. Studies of morphology and immunofluorescence of Pneumocystis carinii. Proc. Soc. Exp. Biol. Med. 141: 304-309. 14. Kluge, R. M., D. M. Spaulding, and J. A. Spain. 1978. Combination of pentamidine and trimethoprim-sulfamethoxazole in the therapy of Pneumocystis carindi pneumonia in rats. Antimicrob. Agents Chemother. 13: 975-978. 15. Latorre, C. R., A. T. Sulzer, and L. G. Norman. 1977. Serial propagation of Pneumocystis carinii in cell line cultures. Appl. Environ. Microbiol. 33:1204-1206. 16. Lim, S. K., W. C. Eveland, and R. J. Porter. 1973. Development and evaluation of a direct fluorescent antibody method for the diagnosis of Pneumocystis carinii infections in experimental animals. Appl. Micro-

937

biol. 26:666-671. 17. Uim, S. K., R. H. Jones, and W. C. Eveland. 1971. Fluorescent antibody studies in experimental Pneumocystis. Proc. Soc. Exp. Biol. Med. 136:675-679. 18. Masur, H., and T. C. Jones. 1978. The interaction in vitro of Pneumocystis carinii with macrophages and Lcells. J. Exp. Med. 147:157-170. 19. Ogino, K. 1978. Pneumocystis carinii: experimental pulmonary infection in rats. Jpn. J. Parasitol. 27:77-89. 20. Pifer, L., W. T. Hughes, and M. J. Murphy. 1977. Propagation of Pneumocystis carinii in vitro. Pediatr. Res. 11:305-316. 21. Pifer, L. L., D. Woods, and W. T. Hughes. 1978. Propagation of Pneumocystis carinii in Vero cell culture. Infect. Immun. 20:66-68. 22. Price, R. A., and W. T. Hughes. 1974. Histopathology of Pneumocystis carinii infestation and infection in malignant disease in childhood. Human Pathol. 5:737752. 23. Rosen, P., D. Armstrong, and C. Ramos. 1972. Pneumocystis carinii pneumonia. Am. J. Med. 53:428-436. 24. Sheldon, W. H. 1959. Experimental pulmonary Pneumocystis carinii infection in rabbits. J. Exp. Med. 110: 147-160. 25. Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill Book Co., New York. 26. Ueda, K., U. Goto, S. Yamazaki, and K. Eugiwara. 1977. Chronic fatal pneumocystosis in nude mice. Jpn. J. Exp. Med. 47:475-482. 27. Vavra, J., and K. Kucera. 1970. Pneumocystis carinii in Delanoe, its ultrastructure and ultrastructural affinities. J. Protozool. 17:463-483. 28. Von Behren, L. A., and E. L. Pesant. 1978. Uptake and degradation of Pneumocystis carinii by macrophages in vitro. Am. Rev. Respir. Dis. 118:1051-1059. 29. Vossen, M. E. M. H., P. J. A. Beckers, J. H. E. T. H. Meuwissen, and A. M. Stadboulders. 1978. Developmental biology of Pneumocystis carinii, an alternative view on the life cycle of the parasite. Z. Parsitenkd.

55:101-118. 30. Walzer, P. D., D. P. Perl, D. J. Krogstad, and M. G. Schultz. 1974. Pneumocystis carinii pneumonia in the United States. Ann. Intern. Med. 80:83-93. 31. 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. 32. Walzer, P. D., M. E. Rutledge, K. Yoneda, and B. J. Stahr. 1979. Pneumocystis carinii: new separation method from lung tissue. Exp. Parasitol. 47:356-368. 33. Walzer, P. D., V. Schnelle, D. Armstrong, and P. P. Rosen. 1977. Nude mouse: a new experimental model for Pneumocystis carinii infection. Science 197:177179. 34. Weber, W. R., F. B. Askin, and L. P. Dehner. 1977. Lung biopsy in Pneumocystis carinii pneumonia: a study of typical and atypical features. Am. J. Clin. Pathol. 67:11-19. 35. Weller, R. W. 1955. Zur Erzeugung von Pneumocystosen im Tierversuch. Z. Kinderheilk. 76:366-378. 36. Weller, R. W. 1956. Weitere Untersuchungen uber experimentelle Rattenpneumocystose im Hinblick auf die interstitielle Pneumonie der Fruhgebornen. Z. Kinderheilkd. 78:166-176. 37. Western, K. A., N. Norman, and A. F. Kaufman. 1975. Failure of pentamidine isethionate to provide chemoprophylaxis against Pneumocystis carinii infection in rats. J. Infect. Dis. 131:273-276. 38. Whitcomb, M. E., M. I. Schwarz, and M. A. Charles. 1970. Interstitial fibrosis after Pneumocystis carinii pneumonia. Ann. Intern. Med. 73:761-765.