EXPERIMENTAL

AND

MOLECULAR

lmmunogenetic

PATHOLOGY

Influences

54, 172-180 (19%)

on Skin Granuloma in Mice

Formation

TOSHIHIRO IIDA, NAOYOSHI SATO, KIMIE FUKUYAMA, DANIEL T. LAU,’ AND WILLIAM L. EPSTEIN Department of Dermatology and ‘Animal Care Facilities, University of California, San Francisco, San Francisco, California 94143-0536 Received October 1, 1990; and in revised form February 7, 1991 Genetic influence on the development of granulomatous tissue reaction was investigated in C57BL/6 mice. Granulomas developed in the skin of euthymic C57BL/6 mice by transplantation of lyophilized hepatic granulomas were excised and lyophilized. The tissue mass free of parasite egg antigen and living cells was grafted into the skin of euthymic, athymic (n&u), and beige (bglbg) C57BL/6 mice. Histological changes at the skin sites were studied weekly by light microscopy, and cells in newly developed granulomas at 6 weeks after grafting were examined by electron microscopy. Granulomatous inflammation occurred in all the variants but morphometric analysis showed that granulomatous inflammation was the most extensive in beige mice and least in athymic mice. The differences in the degree of tissue reaction were also quantified by measuring angiotensin converting enzyme and prolyl endopeptidase. Statistically significant differences among the animals with varying genetic background were confirmed by the marker enzyme activity. The findings confirm that initiation of a granulomatous response does not require T cells but T cell function is important for full expression of the reaction, while NK cell activity seems to suppress granuloma fOmatiOn.

0 1991 Academic

Press, Inc.

INTRODUCTION Organized granulomas are a pathological tissue reaction characteristic of certain chronic diseases, such as sarcoidosis, tuberculosis, and leprosy. Primary cells in the lesions come from the monocyte/macrophage lineage, but other cells have been considered to modulate this unique inflammatory condition. In Schistosoma munsoni-infected mice, Chensue and Boros (1979) and Colley (1976) demonstrated that a decrease of granulomatous lesions is governed by T suppressor cells. These and other reports on modulating effects of T cells and their products on granulomatous inflammation led to the concept that organized granulomatous reactions are a form of delayed hypersensitivity (Adams, 1983; Boros and Warren, 1970; Kunkel et al., 1989). However, the role of T helper/inducer cells in the initiation of granulomatous inflammation was not adequately investigated until athymic mice became available. Hsu et al. (1976) and Phillips et al. (1977) detected small, poorly organized schistosome egg granulomas of athymic mice. We (Epstein et al., 1979) confirmed the presence of activated macrophages and the occasional appearance of epithelioid cells in the tissue reaction. The findings suggested that cell-mediated immunity might not totally dictate development of granulomatous tissue reactions, particularly during the initiation stages. The use of athymic mice for the studies of systemic infections has drawbacks and disadvantages since the nude mice often become ill and may die of the infection, and cellular functions which participate in the local tissue reaction surely are altered and difficult to interpret. In fact, Byram and von Lichtenberg (1977) failed to produce granulomas in athymic mice and reported the lesions to be necrotic, simulating viral hepatitis. 172 0014-4800/91 $3.00 Copyright All rights

Q 1991 by Academic Press. Inc. of reproduction in any form reserved.

GENETIC

INFLUENCES

ON

SKIN

GRANULOMAS

173

Recently we established a noninfectious murine skin model for organized granulomas by transplanting developing skin granulomas into the skin of naive mice (Iida et al., 1991). The host cells which respond to the noninfectious granulomagenie inoculum accumulate in the skin sites, proliferate, and differentiate into organized granulomas with activated macrophages and epithelioid cells. The advantage of this model is that the skin reaction is a response of healthy host mice not suffering from any other illnesses. Thus, they remain able to express the full range of inflammatory reactions determined by their genetic phenotypic background. In this study we investigated granuloma formation in C57BL16 mice, because athymic nude and beige mice, which congenitally lack NK cells, are both commerically available for this strain. While the athymic variant demonstrates the effects of T helper/inducer on the granulomatous tissue reaction, the beige mutant is expected to show how NK cells are involved in the same tissue reaction. Hashimoto et al. (1991) reported that the suppression of NK cell activity by an injection of monoclonal antibody to NK cells increased the granuloma size in the skin of mice. MATERIALS

AND METHODS

Mice Female mice 5-6 weeks of age at the beginning were used in all experiments. C57BLW background euthymic (+/+) and beige (bglbg) mice were purchased from Jackson Laboratories (Bar Harbor, ME) while athymic nude (C57BL/6 nul nu) mice were obtained from Taconic (Germantown, NY). They were housed separately in a temperature-controlled room. Athymic mice were maintained in sterilized quarters in filter-top cages and fed sterilized water and chow supplied ad libitum . Skin Granuloma

Model

Euthymic C57BL/6 mice 54 weeks old were infected with the Puerto Rican strain of S. mansoni by the subcutaneous injection of 75 cercariae/mouse. The livers were removed 9 weeks later, and the granulomas which developed were isolated from the hepatic tissue by the method of Okamoto et al. (1987). Granulomas from 10 mice were frozen in liquid nitrogen, lyophilized, and kept at - 70°C until use. A group of about 20 euthymic mice (54 weeks old) was inoculated subcutaneously in the back skin with 50 mg each of the freeze-dried granulomas moistened with distilled water. The grafted tissues became replaced by newly formed granulomas consisting of host cells, as previously observed (Iida et al., 1991). The skin granulomas were excised from 20 euthymic mice 5 weeks after the grafting of hepatic granulomas and weighed (the usual yield was about 1 g of tissue from 20 mice). They were minced, freeze-dried, and kept at -70°C. After hydrating with distilled water, about 50 mg was transplanted into the skin of a naive mouse. The recipients of the second passage were euthymic, beige, or athymic mice. Light and Electron

Microscopy

Grafted skin sites excised weekly up to 7 weeks were either fixed in 4% formalin and prepared for pa&l-in sections or fixed in 3% glutaraldehyde and postfixed

174

IIDA

ET AL.

with 2% osmium tetroxide and embedded in a mixture of plastic. The paraffin sections were stained with hematoxylin and eosin. The plastic-embedded tissues either were cut 1 urn thick and stained with toluidine blue for light microscopy or were cut at 60 nm thick and stained with uranyl acetate and lead citrate for electron microscopy. The percentage area of grafted tissue replaced by newly developed granulomatous inflammation and the mean area (mm2) covered by each granuloma unit were morphometrically analyzed. Assay for Angiotensin Converting Enzyme (ACE) and Prolyl Endopeptidase (PSE) Activity Approximately 50 mg of tissue from newly developing granulomas in the back skin was homogenized in 1 ml of 0.02 h4 Tris-HCl, pH 7.8, containing 0.25 M sucrose with a Polytron homogenizer (Model PT-10; Brinkmann, Switzerland) for 1 min. The normal apparent ventral skin was also excised and similarly homogenized after removing the epidermis and subcutaneous tissues to measure the enzyme activity in the normal dermis. The homogenate was adjusted to 2 ml by adding 2% Triton X-100 to a final concentration of 0.25% (v/v) in the same buffer and stored at 4°C for 90 min. The homogenate was centrifuged at 1OOOgfor 10 min and the supernatant was used for enzyme assay. ACE (EC 3.4.15.1) activity was measured by the method established by Hara et al. (1981a) using hippurylL-histidyl-L-leucine as a substrate. PSE (EC 3.4.21.26) was assayed by the method of Yoshimoto et al. (1979) which measures hydrolysis of iV-(benzyloxycarbonyl)glycyl-L-prolyl-4-methylcoumarinyl amide (Z-Gly-Pro-MCA). The fluorescence of 7-amino-4-methylcoumarin was quantified at 370 nm of excitation and 440 nm of emission. One unit of ACE and PSE was defined as the amount of enzyme that hydrolyzed 1 pmole of each substrate per minute at 37” and 25°C respectively. Protein concentration was determined by the method of Lowry et al. (1951) using bovine serum albumin as the standard. Statistical Analysis The values were calculated as means + standard error. Statistical was determined using the Student’s t test.

significance

RESULTS Light and Electron Microscopy Biopsies taken weekly from the second passage graft sites showed an initial degeneration of grafted tissue and the subsequent repopulation by mononuclear cells as described previously in both the first and second passage granuloma grafts (Nishimura et al., 1985; Okamoto et al., 1989; Iida et al., 1991). The general time course of the cellular response in skin was virtually identical in all euthymic, athymic, and beige mice, but the degree of cellular infiltration starting 3-4 weeks after grafting and the mode of subsequent cellular organization differed distinctly among the mutants. The number of cells accumulating in the skin sites of athymic mice was considerably limited even 7 weeks after grafting (Fig. IA). The infiltrating cells appeared loosely clustered around egg shells and other unidentified particles. In euthymic and beige mice, large numbers of cells entered the grafted tissue from the edge and formed well-organized granulomas (Fig. 1B). The morphometrical analyses showed that granulomas developing in the skin of beige mice were larger than euthymic mice, whereas those in athymic animals were smaller.

GENETIC

INFLUENCES

ON

SKIN

GKANULOMAS

175

FIG. 1. Histological comparison of second passage skin granulomas formed in athymic versus beige C57BLI6 mice. (A, B) Pa&in-embedded specimens stained with hematoxylin and eosin. The moderate degree of mononuclear cell infiltration seen in athymic mice (A) is not organized, while the exuberant collection of mononuclear cells appearing in beige mice (B) is organized about an egg fragment magnification. x500. Bars = 10 pm. (C-F) Plastic-embedded l-km sections stained with toluidine blue. Large mononuclear cells and epitheloid cells form loosely arranged units in athymic mice (C, E), while they are much better organized as granulomas in beige mice (D, F) egg shell (*). Note also that eosinophils (+) form clusters within granulomas of beige mice (D, F). Magnification X800, Bars = IO pm in C, D; x2OlW. Bars = 5 urn in E, F.

By percentage area (Table I) the reaction in beige mice was greater than either euthymic (P < 0.005) or athymic (P < 0.005) mice, and that in euthymic was greater than in athymic (P < 0.025) mice. By measuring mean areas (Table II), beige mice also formed the largest granulomas but were not significantly larger than in euthymic mice. However, both beige and euthymic mice produced significantly larger granulomas than did athymic mice (P < 0.005).

176

IIDA

ET AL.

TABLE I Percentage Area of Skin Graft Replaced by Newly Formed Granulomatous

Response

Mice

% Area

C57BL/6 euthymic (6)” C57BL/6 beige (3) C57BL/6 athymic (3)

34.3 -+ 10.6’ 68.3 2 3.6 14.5 2 3.1

u Number of biopsies examined. b Mean k standard error.

The use of plastic-embedded l-pm-thick sections provides better assessment of cell types accumulating in the skin. Despite differences in cellular arrangement, the intlammatory cell foci showed pathological changes which are classified as granulomatous tissue reaction, and the majority of cells in the graft sites were activated macrophages mixed with epithelioid cells (Figs. IC-1F). The individual epithelioid cells in beige, euthymic, and athymic mice as observed by transmission electron microscopy were indistinguishable from each other. The nuclei were large and oval with chromatin marginated along the nuclear lamina and prominent nucleoli were seen. The cytoplasm contained secretory and phagocytic organelles and the plasma membrane interdigitated with neighboring cells (Fig. 2). Most of these epithelioid cells are classified as type EC-I cells, but there are also type EC-II cells which primarily show expanded arrays of rough endoplasmic reticulum without phagocytic vacuoles or dense bodies. In addition, cells with multilobular nuclei were present in the granulomas of euthymic and beige (Fig. 1D) mice. By eIectron microscopy these cells were identified as eosinophils with typical granules in the cytoplasm. ACE and PSE Activity ACE activity in the freeze-dried inoculum prepared from granulomas developed in the first passage skin was 38.10 + 2.11 mU/mg protein. Kanazawa et al. (1987) and Iida et al. (1991), who assayed ACE activity during reformation of granulomas in the skin, reported that the activity in the inoculum decreased immediately after tissue grafting, but that the activity gradually elevated as new granulomas formed. Since a limited number of athymic and beige mice were available, we only measured ACE activity at 6 weeks after grafting in the present study. The activity in granulomas of beige mice was the highest and that of athymic mice was the lowest (Table III). The differences, as compared to euthymic mice, were statistically significant (P < 0.01). All normal skin showed a lower activity than any of the TABLE II Mean Area of Individual Granulomas without Central Nidus Mice

mm2

C57BL/6 control (25) C57BL/6 beige (11) C57BL/6 athymic (15)

200.4 + 35.8’ 344.3 2 181.0 75.2 d 25.0

a Number of histological sections examined. b Mean k standard error.

GENETIC

INFLUENCES

ON SKIN

177

GRANULOMAS

FIG. 2. Electron microscopy of epithelioid cells in beige mice show polysomes, arrays of rough endoplasmic reticulum, and dense lysosomal bodies (-+) in the cytoplasm. Note also the interdigitation of plasma membranes, Magnification x 18,000. Bar = 0.5 pm.

granulomas studied and there were no significant differences (P > 0.05) among the variants. PSE activity in the lyophilized skin granulomas used for initiation of the second passage granulomas was 5.70 ? 0.32 mU/mg protein. Skin excised from euthymic, athymic, and beige mice at 6 weeks after grafting showed elevations of PSE activity in the same order as that seen for ACE activity. Table IV summarizes statistically significant differences (P < 0.01) in PSE activity among granulomatous tissue from the three variants while insignificant differences (P > 0.05) were determined in their normal skin. TABLE III Angiotensin Converting Enzyme Activity” from Granulomatous

Tissue in Skin and Normal Dennis

Mice

Granulomas developed 6 weeks after grafting

Normal dermis

C57BIJ6 euthymic (5)” C57BIJ6 beige (3) C57BU6 athymic (3)

SO.99 2 4.81’ 65.11 ” 6.99 32.92 2 1.92

8.23 + 3.10 6.21 f 4.16 5.87 f 0.08

0 Activity expressed as mU/mg protein. b Number of mice. c Mean activity 2 standard error.

178

IIDA ET AL. Prolyl Endopeptidase

TABLE IV Activity” from Granulomatous

Tissue in Skin and Normal Dermis

Mice

Granulomas developed 6 weeks after inoculation

Normal dermis

C57BLl6 euthymicb C57BLl6 beige C57BLi6 athymic

6.76 + 0.56’ 10.12 * 0.41 3.14 k 0.33

1.43 + 0.32 1.21 2 0.43 1.28 2 0.01

a Activity expressed as mU/mg protein. b Number of mice. c Mean activity 2 standard error.

DISCUSSION The granulomatous tissue reaction which occurs after grafting a freeze-dried granuloma implant devoid of parasite egg antigen (Pelley et al., 1976) and living cells was compared in the skin of euthymic C57BL/6 mice with mutant, athymic, and beige mice of the same strain. The degree of tissue response, as assessed by morphological and biochemical analyses, showed significant differences according to genetic background. By taking advantage of the unique experimental model, we were able to elicit a T cell-independent tissue reaction in healthy athymic mice without any systemic involvement (Fujioka et al., 1990). The situation differs considerably from experiments with S. munsoni-infected mice (Hsu et al., 1976; Phillips er al., 1977; Epstein et al., 1979) and with euthymic mice severely immunosuppressed (Pincelli et al., 1988; Fujioka et al., 1989) since the mice remain in general good health throughout the observation period. Accumulation of activated macrophages and differentiation into epithelioid cells occurred in the skin of athymic mice, confirming our earlier proposal that the initiation of the granulomatous tissue reaction is T cell independent. However, the cellular infiltrate appeared as a poorly organized cellular mass even at 7 weeks after grafting, suggesting that the process of granuloma organization requires T cell modulation. This point could be proved by transfusing the athymic mice with competent T cells. In contrast, granulomas which grew in the beige mouse skin were the best organized and showed the greatest tissue reaction. Beige mice have a severe impairment of NK cell activity, while T cell and macrophage functions remain intact (Roder and Duwe, 1979; Talmage et al., 1980). The up-regulated macrophage differentiation is thus considered to be directly related to NK cell activity. This mutant strain is more susceptible to tumor growth and spread (Thiel et al., 1988; Talmage et al., 1980) and has fostered the proposal of “NK cell-dependent” surveillance against infections and cancer. Since beige mice suffer other defects and since many factors are involved in NK cell activity (Merrill, 1986), the specific mechanisms by which granulomatous inflammation is modulated in beige mice remain unknown. In beige mice with pulmonary granulomas, Remick et al. (1988) suggested that NK cells may act as static suppressor cells of the tissue reaction. In addition, a reverse relationship between macrophage growth and NK cell activity seems to exist. Gradual reduction in NK cell activity, without changing NK cell numbers in the spleen and lymph node of the mice infected with S. munsoni, was detected by Hashimoto et al. (1991). NK cell activity was also reported to be reduced in some other chronic diseases, i.e., lupus erythematosus (Gaspar et al., 1988) and multiple sclerosis (Hauser er al., 1981), so that down-

GENETIC

INFLUENCES

ON

SKIN

GRANULOMAS

179

modulation of NK cell activity may be a relevant event in the chronicity of disease in general. In this study we measured PSE activity in addition to the well-established enzyme marker, ACE (Liberman, 1975; Silverstein et al., 1976; Hara et al., 1981b), for assessment of granulomatous tissue reaction. PSE catalyzed hydrolysis of the small peptide on the carboxyl side of a proline residue (Walter and Yoshimoto, 1978; Yoshimoto et al., 1981). The enzyme activity increased 3.65 fold in hepatic granulomas of mice, and the immunohistochemically monospecific antibody to PSE reacted in macrophages accumulated within granulomas (Nozaki et al., 1990). PSE hydrolyzes both angiotensin I and II at k,,JK,,, of 70 mM-’ set-’ and 15 mM-’ set-‘, respectively. Thus the enzyme may regulate the angiotensin cascade. Angiotensin II was shown to alter macrophage function (Foris et al., 1983) and granulomatous tissue reaction (Weinstock and Kassab, 1984; Weinstock and Blum, 1985). Elevation of the enzyme activity was the highest in beige mice, while relatively low but more than normal dermal levels were detected in athymic mice. The data confirm an earlier suggestion (Iida et al., 1991) that PSE may prove useful as a marker enzyme for granulomatous tissue reaction, in addition to ACE activity. ACKNOWLEDGMENT This work was supported by Grant AR31853 from the National Institutes of Health.

REFERENCES ADAMS, D. 0. (1983). “Pathology of Granulomas” (H. L. Ioachim, Ed.), pp. l-20. Raven Press, NY. BOROS, D. L., and WARREN, K. S. (1970). Delayed hypersensitivity-type granuloma formation and dermal reaction induced and elicited by a soluble factor isolated from Schistosoma mnnsoni eggs. J. Exp.

Med.

132, 488-507.

BYRAM, J. E., and VON LICHTENBERG, F. (1977). Altered schistosome granuloma formation in nude mice. Am. J. Trop. Med. Hyg. 26, 944-956. &ENSUE, S. W., and BOROS, D. L. (1979). Modulation of granulomatous hypersensitivity. I. Characterization of T lymphocytes involved in the adoptive suppression of granuloma formation in Schistosoma mansoni-infected mice. J. Immunol. 123, 1409-1414. COLLEY, D. G. (1976). Adoptive suppression of granuloma formation. J. fip. Med. 143, 696-700. EPSTEIN, W. L., FUKUYAMA, K., DANNO, K., and KWAN-WONG, E. (1979). Granulomatous inBammation in normal and athymic mice infected with Schistosoma mansoni. J. Pathol. 127, 207-215. Foam, G., DEZSO, B., MEDGYESI, G. A., and FOST, G. (1983). Effect of angiotensin II on macrophage functions. Immunology 48, 529-535. FUJIOKA, A., PINCELLI, C., HASHIMOTO, A., FUKUYAMA, K., and EPSTEIN, W. L. (1989). Immunosuppression an in vivo T cell function in granulomatous intlammation. Zni. Arch. Allergy Appl. Immunol. 90, 313-319. FUJIOKA, A., SUYA, H., P~NCELLI, C., HASHIMOTO, A., FUKUYAMA, K., and EPSTEIN, W. L. (1990). Local effects of granulomatous inflammation on functional activation of T cells in athymic mice. Nat.

Immun.

Cell Growth

Regul.

9, 342-352

GASPAR, M. L., ALVAREZ-M• N,M., and GUTIERREZ, C. (1988). Role of interleukin 2 in inducing normalization of natural killer activity in systemic lupus erythematosus. Clin. Zmmunol. Zmmunopathol. 49, 20&214. HARA, A., FUKUYAMA, K., and EPSTEIN, W. L. (1981a). Angiotensin-converting enzyme measured in mouse tissue by inhibition of histidyl-leucine peptidase. Biochem. Med. 26, 199-210. HARA, A., FUKUYAMA, K., and EPSTEIN, W. L. (1981b). Angiotensin-converting enzyme and other enzymes in lives of mice with experimental schistosomiasis. Exp. Mol. Pathol. 35, 199-210. HASHIMOTO, A., PINCELLI, C., FUJIOKA, A., FUKUYAMA, K., and EPSTEIN, W. L. (1991). Relationship between NK cells and granulomatous inflammation in mice. Clin. Lab. Immunol., in press. HAUSER, S. L., AULT, K. A., LEVIN, M. J., GAROVOY, M. R., and WEINER, H. L. (1981). Natural killer cell activity in multiple sclerosis. J. Zmmunol. 127, 1114-l 117.

180

IIDA

ET AL.

Hsu, C.-K., Hsu, S. H., WHITNEY, R. A., JR., and HANSEN, C. T. (1976). Immunopathology of schistosomiasis in athymic mice. Nature 262, 397-399. IIDA, T., NOZAKI, Y., FUKUYAMA, K., and EPSTEIN, W. L. (1991). An improved noninfectious murine skin model of organized granulomatous inflammation. Experientia 47, in press. KANAZAWA, K., HIGUCHI, M., OKAMOTO, M., FUKUYAMA, K., and EPSTEIN, W. L. (1987). Induction of granuloma-dependent angiotensin-converting enzyme and eosinophil chemotactic factor in the skin of athymic nude mice. J. Cell. Biochem. 34, 61-69. KUNKEL, S. L., CHENSUE, S. W., STRIETER, R. M., LYNCH, J. P., and REMICK, D. G. (1989). Cellular and molecular aspects of granulomatous inflammation. Am. J. Respir. Cell Mol. Biol. 1,

439-447. LIEBERMAN, J. (1975). Elevation of serum angiotensin-converting-enzyme (ACE) level in sarcoidosis. Am. J. Med. 59, 365-372. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., and RANDALL, R. J. (1951). Protein measurement with the folin phenol reagent. J. Biol. Gem. 193, 265-275. MERRILL, J. E. (1986). In “Immunobiology of Natural Killer Cells” (E. Lotzova and R. B. Herberman, Eds.), pp. 75-81. CRC Press, Boca Raton, FL. NISHIMURA, M., HIGUCHI, M., FUKUYAMA, K., and EPSTEIN, W. L. (1985). Autoradiographic studies of cell dynamics in immunogenic granulomas transplanted into the skin of athymic nude mice. Arch. Dermatol. Res. 278, 61-67. NOZAKI, Y., IIDA, T., SASAKI, N., SATO, N., and EPSTEIN, W. L. (1990). Purification of proline specific endopeptidase from organized granulomas. Clin. Res. 38, 601A OKAMOTO, M., EPSTEIN, W. L., SUYA, H., KANAZAWA, K., and FUKUYAMA, K. (1987). Transfer of granulomatous inflammation with nonviable preparations of schistosome granulomas in naive mice. Exp. Cell Biol. 55, 173-178. PELLEY, R. P., PELLEY, R. J., HAMBURGER, J., PETERS, P. A., and WARREN, K. S. (1976). Schistosoma mansoni soluble egg antigens. I. Identification and purification of three major antigens, and the employment of radioimmunoassay for their further characterization. J. Zmmunol. 117, 15531559. PHILLIPS, S. M., DICONZA, J. J., GOLD, J. A., and REID, W. A. (1977). Schistosomiasis in the congenitally athymic (nude) mouse. I. Thymic dependency of eosinophilia, granuloma formation, and host morbidity. J. Zmmunol. 118, 594-599. PINCELLI, C., FUJIOKA, A., HASHIMOTO, A., FUKUYAMA, K., and EPSTEIN, W. L. (1988). T-cell depletion by monoclonal antibodies does not prevent granuloma formation in mice. Exp. Cell Biol.

56, 229-235. REMICK, D. G., KUNKEL, S. L., HIGASHI, G. I., and HISERODT, J. C. (1988). Suppression of natural killer cytolytic activity in mice undergoing pulmonary granulomatous inflammation. J. Zmmunol. 140, 2225-2230. RODER, J., and DUWE, A. (1979). The beige mutation in the mouse selectivity impairs natural killer cell function. Nature 278, 451453. SILVERSTEIN, E., FRIEDLAND, J., LYONS? H. A., and GOURIN, A. (1976). Markedly elevated angiotensin converting enzyme in lymph nodes containing non-necrotizing granulomas in sarcoidosis. Proc. Natl. Acad. Sci. USA 73, 2137-2141. TALMAGE, J. E., MEYERS, K. M., PRIEUR, D. J., and STARKEY, J. R. (1980). Role of NK cells in tumour growth and metastasis in beige mice. Nature 284, 622-624. THIEL, K., GENOVESI, E. V., IGLEHART, J. D., BOLOGNESI, D. P., and WEINHOLD, K. J. (1988). Loss of tumorigenicity following in vitro MuLV infection is associated with induction of peritoneal natural killer cell activity. Adv. Exp. Med. Biol. 239, 169-183. WALTER, R., and YOSHIMOTO, T. (1978). Postproline cleaving enzyme: Kinetic studies of size and stereospecificity of its active site. Biochemistry 17, 4139-4144. WEINSTOCK, J. V., and BLUM, A. M. (198.5). Effects of granuloma modulation induced by regulatoryT-lymphocyte activity on angiotensin II/III production by granuloma macrophages in murine Schistosomiasis mansoni. Cell. Zmmunol. 94, 558-567. WEINSTOCK, J. V., and KASSAB, J. T. (1984). Functional angiotensin II receptors on macrophages from isolated liver granulomas of murine Schistosoma mansoni. J. Zmmunol. 132, 2598-2602. YOSHIMOTO, T., OGITA, K., WALTER, R., KOIDA, M., and TSURU, D. (1979). Synthesis of a new fluorogenic substrate and distribution of the endopeptidase in rat tissues and body fluids of man. Biochim Biophys. Acta 569, 184-192. YOSHIMOTO, T., SIMMONS, W. H., KITA, T., and TSURU, D. (1981). Post-proline cleaving enzyme from lamb brain. J. Biochem. 90, 325-334.