c mice which prevent myocarditis

CELLULAR

98, 104- 113 ( 1986)

IMMUNOLOGY

Demonstration of Suppressor Cells in Coxsackievirus Group B, Type 3 Infected Female Balb/c Mice Which Prevent Myocarditis’ L. P. JOB,* D. C. LYDEN, AND S. A. HUBERT Department of Pathology, University of Vermont, College ofMedicine, Burlington, Vermont 05405 Received September 30, 1985; accepted October 7, 1985 Coxsaclcievirus group B type 3 (CVB3) induces myocarditis in male Balb/c mice but produces little cardiac injury in females. Males develop cytolytic T lymphocytes (CTL) reactive to heart antigens which primarily cause the inflammation and cardiac injury observed in the disease. Infected female mice lack this CTL response because they rapidly produce suppressor cells inhibiting both cellular immunity and cardiac inflammation. Four lines of evidence demonstrate suppressor cells in females. First, females develop myocarditis when treated with low-dose cyclophosphamide under conditions known to preferentially eliminate suppressor cells but not other immune cells. Second, lymphocytes obtained from females at various times after infection prevent myocarditis when adoptively transferred into CVB3-infected males. Virus concentrations in the hearts of males receiving immune female cells and control males were equivalent. Thus protection did not result from accelerated virus elimination in recipient males. Third, CTL from CVB3 infected male mice could induce myocarditis in infected T-lymphocyte depleted but not in intact females suggesting the presence of an inhibitory T cell in the intact animals. Finally, male lymphocytes cultured on heart cell monolayers for 5 days generate significant cytolytic activity to myocyte targets. CTL generation could be inhibited by co-culture of the male cells with immune female lymphocytes. Nonimmune female cells were not inhibitory. 0 1986 Aeademie RUZ,IDC.

INTRODUCTION Myocarditis, a disease characterized by either diffuse or focal inflammation of the myocardium, is usually associated with recent viral infections (1). Clinical evidence strongly suggests cardiac damage results from immunological mechanisms, not virusinduced injury. This concept receives further support from studies using a murine model of the disease (2-6). Coxsackievirus group B, type 3 (CVB3) induces severe myocarditis in male BaIb/c mice, but animals depleted of T lymphocytes prior to infection usually experience little cardiac injury even though virus concentrations in the hearts of intact and immunocompromised mice are equivalent (2). Murine and human myocarditis share several interesting characteristics. The disease primarily occurs in males and pregnant females while non-pregnant females develop minimal cardiac damage (1, 7). Myocarditis in males is closely associated with the generation of autoimmune cytolytic T lymphocytes (CTL) directed toward myocyte ’ Supported by Public Service Grants HL/AI 28480, HL 28833, and HL 33256 horn the National Institutes of Health and by an American Heart Association Established Investigatorship to S.A.H. ’ Change of address: Department of Physical and Life Sciences, University of Portland, Portland, Oregon. ’ To whom all correspondence should be addressed. 104 0008-8749/86 $3.00 l3pyhgMo 1986byAcademicRss,1nc. Au rights of rep-n

in any form -cd.

SUPPRESSOR

CELLS

IN

VIRAL

MYOCARDITIS

105

cell surface antigens (6). This CTL response is noticeably absent in the virgin females (5) because females either lack appropriate T-cell precursors or rapidly produce sup pressor cells which actively inhibit cellular immunity. The present study clearly shows that suppressor cells are generated in infected females which protect the animals from severe heart disease. MATERIALS

AND

METHODS

Animals. Balb/c mice were purchased originally from Cumberland Farms, Clinton, Tennessee. Neonates and 6- to &week-old adult animals were obtained from colonies of these mice maintained at this institution. Virus preparation and purification. A myocarditic strain of CVB3M (Nancy strain) was used. The procedures for growth and purification of the viruses on HeLa cultures have been reported in detail earlier (5). Infection of mice. Each animal was infected by intraperitoneal (ip) inoculation of 0.1 ml PBS containing 6 X lo4 plaque forming units (PFU) of CVB3M. Animals were sacrificed 7 days after infection. Cyclophosphamide (CY) administration. CY (Sigma Chemical Co., St. Louis, MO.) was dissolved in sterile PBS immediately before use. Animals were injected intraperitoneally with 50 mg CY/kg body wt. Virus titration. Virus titers in the heart were obtained by homogenization of the tissue in minimal essential medium (MEM), centrifugation of the cellular debris, and titration of virus by plaque formation on monolayers of HeLa cells (5). Indirect enzyme-linked immunoabsorbent assay (ELZSA) for the detection of Coxsackievirus B-3 spec$c ZgG and ZgA4 antibodies. PFU CVB3M (8 X lo6 in 50 ~1 PBS) were added to wells of polystyrene microtiter plates (Cooke MicroELISA substrate plates, Dynatech Labs., Alexandria, Va.) incubated overnight at 4°C the plates were washed six times with PBS containing 0.05% Tween. Twofold dilutions of serum obtained from CVB3M-infected mice were added in 50-~1 amounts to the wells, and incubated 90 min at room temperature. The wells were washed six times with PBS/Tween, and 50 ~1 of an appropriate dilution of P-galactosidase linked goat anti-mouse IgG (gamma chain specific) and anti-mouse IgM (mu chain specific) (Zymed, San Francisco, Calif.) was added for another 90 min at room temperature. After washing with. PBS/Tween, 100 ~1 of o-nitrophenyl-&Dgalactopyranoside (ONPG), was added for 45 min at 37°C. The reaction was halted by the addition of 50 ~1 of 1 M Na2C03. The optical density change at 4 10 nm was determined with a MicroELISA minireader (Dynatech) and compared to a control containing no enzyme. The ELISA antibody titer was defined as the reciprocal of the highest serum dilution giving 0.06 absorbance units above control levels. Preparation, culture, and characterization of myocytes. The procedure for the preparation and characterization of primary cultures of mouse myocytes has been described in detail previously (8, 9). Briefly, hearts from neonatal mice less than 72 hr old were minced then subjected to stepwise enzymatic digestion with 0.4% collagenase (Worthington Biochemical Corp. Freehold, N.J.). The isolated cells were washed in basal medium Eagle’s (BME), depleted of endothelial cells and fibroblasts by two sequential 1-hr adsorptions to plastic and the nonadherent myocardial cells were dispensed into 6-mm tissue culture wells (Linbro Chemical Co., Hamden, Conn.) in BME containing penicillin, streptomycin, 0.2 mg crystalline insulin, 20 m&f Hepes, 10% horse serum

106

JOB, LYDEN,

AND HUBER

(GIBCO), and 5% FCS (GIBCO) at a concentration of 3 X lo5 viable cells/ml. Ten microliters of the myocyte suspension was dispensed into 1-mm diameter wells of 60well tissue culture plates (Falcon Plastics, Oxnard, Calif.). After incubation in a humidified, 6% CO2 incubator for 48 hr, the myocytes were used in the cytotoxicity assay. This procedure results in a cell population comprised of greater than 90% myocytes. Preparation of immune lymphocytes. Inguinal and mesenteric lymph nodes from normal and CVB3M infected mice were removed 7 days after infection of the animals. The lymphocytes were obtained by pressing the nodes through a fine mesh screen, and washing the cells in BME-5% FCS. Viability of the lymphocyte preparation was determined by trypan blue exclusion. Preparation of female spleen T-cell populations. Spleens were removed from uninfected animals and mice 3, 6, 9, and 12 days after infection with CVB3M. Red blood cells, macrophages, and B cells were removed by centrifugation on Ficoll-Paque (Pharmacia Fine Chemical Co., Piscataway, N.J.) and passaged through nylon wool column as described previously (9). T cells were suspended in PBS at a concentration of 10’ viable nucleated cells per milliliter. In vitro generation of cytotoxicity. The procedure has been described in detail earlier (3 1). Briefly, approximately 1O6heart cells were cultured in 25 cm* tissue culture flasks for 2 days, washed with medium, and either infected for 1 hr at 37°C with lo8 PFU CVB3 or cultured under similar conditions with medium alone. The monolayers were washed, incubated with 0.3% glutaraldehyde (Sigma Chemical Co.) for 30 set, and again washed thoroughly with medium. Either 5 or 10 X 10’ lymphocytes were added to the monolayers in 10 ml BME-S%FCS supplemented with 5 X 10e5 M 2-mercaptoethanol and the cultured were incubated for 5 days at 37°C in a humidified 5% CO2 incubator. Viable cells were recovered by centrifugation on Ficoll-Paque. Cell mediated cytotoxicity assay. Myocyte target cells were incubated with 3 X lo5 PFU CVB3M at 37°C for 4-6 hr. The target cells were washed and infected and uninfected cells were incubated with 2 pCi/well “Cr (Na5’Cr04, 1 mCi/ml, New England Nuclear) for 1 hr at 37°C. After washing three times, the target cells were overlaid with 4.5 X lo5 lymphocytes in BME-5% FCS or BME-5% FCS alone. The cultures were incubated for 18 hr at 37°C. After incubation, the amount of 5’Cr in the supernatnat and pellet was determined using an Intertechnique CG 4000 gamma counter. Percentage “Cr release was calculated using the formula: cpm in supematant x 100. cpm in supematant + cpm in cells Cytotoxicity

was expressed as the percentage of lysis calculated from the formula:

Av% 51Cr release from test group - Av% “Cr released from media group Av% “Cr released after freeze-thaw - Av% 5’Cr released from media group ’ The percentage of specific lysis represented the percentage of lysis of sensitized lymphocytes minus the percentage of lysis by normal lymphocytes. Spontaneous “Cr release refers to the average percentage released by the media treated group. A negative percentage of lysis occasionally occurred when “Cr release in the presence of nonimmune lymph node cells was less than that of spontaneous “Cr release.

SUPPRESSOR CELLS IN VIRAL

MYOCARDITIS

107

T cell depletion offemale Bulb/c mice. T-Lymphocyte deficient mice were animals which had been thymectomized, lethally irradiated, and reconstituted with syngenic bone-marrow cells (TXBM) (2). At 3 weeks of age, Balb/c female mice were anesthesized and thymectomized by opening the anterior mediastinum through an incision in the neck and sternum of the second rib. The anterior ribs were spread apart to enable complete aspiration of the thymus with a pipet. One week later the animals were exposed to 850 R using a Theratron Jr. gamma irradiator. On the same day 5 X lo6 syngeneic sex matched bone marrow cells were injected into the animals intraveneously through the tail vein. The bone marrow cells were obtained by flushing the tibial and femoral bones of syngeneic donor mice. The cells were treated with antiThy 1.2 serum (New England Nuclear, Boston, Mass.) and guinea pig complement (GIBCO, Grand Isle, N.Y.) to remove any residual T lymphocytes. After reconstitution with bone marrow cells, the animals were maintained on drinking water containing tetracycline (7.6 g/liter) and rested for 5 weeks prior to use in the experiments. To ensure T-cell depletion the animals were injected with 0.5 ml rabbit anti-mouse thymocyte serum (Microbiological Associates, Bethesda, Md.) 1 week prior to use in the experiment. Adoptive transfer of lymphocytes. TXBM mice were inoculated with 6 X lo4 PFU CVB3 and 3 days later injected intravenously through the tail vein with 5 X lo6 lymphocytes from Day 6 CVB3 immune male animals. The transfused cells were enriched for T lymphocytes by passage over nylon wool as described in detail elsewhere. The recipients were sacrificed 7 days after infection. Histology. The hearts were fixed in 10% buffered Formalin and sectioned laterally approximately midway between the apex and atria resulting in cross-sections of both ventricles. The sections were stained with hematoxylin and eosin, coded and projected onto paper where the total area of the myocardium and the areas of inflammation were outlined. The percentage area of myocardium undergoing inflammation and necrosis was determined by image analysis using an Optomax image analyzer and Apple computer (30). Evaluation of inflammation and necrosis was done by a single individual (N.G.) without prior knowledge of the identity of the specimens. Statistical analysis. The Student t test was used to analyze the significance of differences between groups in the cytotoxicity assays. The Wilcoxon score test was used for evaluation of histologic features. RESULTS Eflect of CY on CVB3-Induced

Myocarditis

in Female Mice

Low doses of CY preferentially eliminate suppressor T-cell activity in treated mice ( lo- 12). The following experiment determined whether cyclophosphamide treatment of CVB3 infected female mice could enhance myocarditis. Groups of 6-10 female Balb/c mice were given a single dose of 50 mg/kg CY on various days before and after inoculation with 6 X lo4 PFG CVB3. Control animals were infected but not given CY. All animals were sacrificed 7 days after receiving virus. Animals treated with CY on the same day or within 24 hr after viral inoculation developed significantly more myocarditis than control mice and animals given CY at other times (Fig. 1). Generally, CY treatment had no significant effect on either virus concentrations in the heart or virus specific antibody titers (Fig. 2). Thus, the enhanced disease could not be explained by ineffective control of virus infection and replication in the heart.

108

JOB, LYDEN,

No CY

-3

AND HUBER

o

Cyclophosphamide on Days Relative

I

3

5

Administration to Infection

FIG. 1. Effect of cyclophosphamide (CY) on CVB3M infection in Balb/c female mice. Female mice were inoculated with CVB3 and given a single injection of 50 mg/kg cyclophosphamide intraperitoneally on different days relative to infection. The animals were sacrificed 7 days after virus inoculation and hearts were examined for percentage myocardium inflamed using an Optomax Image Analyzer. (*) Myocarditis was significantly greater than in females not receiving CY at P < 0.05 level. The number of animals in each group is given in parentheses.

Demonstration of Active Suppressionof Cellular Immunity in Female Mice Infected with CVB3M The above experiment suggests but does not prove the presence of suppressor cells in females. We therefore determined whether immune male cells could induce myo-

6) [3 None q

Cyclophosphamldo

moo 7ooo 6ooo 5ooo 4ooo aooo 2000

L = It g : 8 t a cn 8 *’

1000 0 Daya

FIG. 2. Serum (B) and hearts (A) from animals in Fig. 1 were analyzed for virus-specific antibody and virus concentrations, respectively. Virus concentrations represent mean PFU X lo6 k SEM/heart. Antibody titers are given as reciprocal of arithmatic mean + SEM. (*) Value exceeds that of untreated females at P -c 0.05 level.

SUPPRESSOR CELLS IN VIRAL

109

MYOCARDITIS

carditis in infected females. If females generate T suppressor cells preventing CTL production, the male lymphocytes may not induce heart disease in immunologically intact females, but might cause cardiac injury in T-cell depleted females. Intact and TXBM female mice were inoculated with 6 X lo4 PFU CVB3M and transfused 3 days later with 5 X lo6 immune lymph node cells from syngeneic male mice inoculated 7 days earlier with the virus. The immune lymphocytes were given to the females after virus to prevent possible abrogation of infection by transferred immune cells or antibody. Control groups consisted of infected intact and TXBM female mice not given male cells, and infected males. All mice were sacrificed 7 days after virus inoculation and the hearts were examined by image analysis for myocarditis (Fig. 3). Infected males developed severe myocarditis (4.7% of the myocardium inflamed), while intact and TXBM females developed little inflammation. Cardiac inflammation in intact females given immune male cells was not significantly different than that in the intact female control animals. However, the same male donor cells did induce enhanced disease in the infected TXBM females (2.9%), suggesting that a subpopulation of inhibitory T cells was absent in the immunocompromised female mice. Suppression of Myocarditis

in Male Mice by Transfer of Immune

Female T Cells

The next experiment demonstrates the ability of immune female lymphocytes to suppress myocarditis induction in CVB3 infected male animals. T lymphocytes were obtained from the spleens of uninfected female mice and animals 3,6,9, and 12 days after CVB3M inoculation. Cells (5 X 106) were injected intravenously into male mice 3 days after CVB3 and the recipients were sacrificed 7 days after infection (Fig. 4). Transfer of normal, Day 3 and Day 12 immune female T cells had little effect on myocarditis. However, infected males receiving Day 6 and 9 immune female T cells showed significant reductions in myocarditis compared to control male animals.

0 m

Male

Female

Untreated l Immune d T Cells

TXEM Female

PK. 3. Adoptive transfer of immune male T cells into CVB3M-infected &lb/c female mice. Mice were inoculated with 6 X lo4 PFU CVB3. On the third day after infection, intact and TXBM femaks received 5 X lo6 T lymphocytes from male mice immunized 7 days earlier to the virus. All recipient animals were sacrificed 7 days after infection and the hearts were examined for inflammation. Results are given as the mean percentage of inflammation + SEM. The number of animals in each group is given in parentheses. (:) Inflammation exceeded that in infected, intact female mice at P < 0.05 level.

110

JOB, LYDEN,

Untreated male

0

AND HUBER

3 Female

6

9

Lymphocytes Obtained Days Post-infection

12 on

FIG. 4. Adoptive transfer of immune female T cells into CVB3M-infected Balb/c male mice. Male Balb/c mice were inoculated intraperitoneally with CVB3, given 5 X lo6 syngeneic T lymphocytes from nonimmune and 3 to 12 day CVB3 infected Balb/c female mice, and sacrificed 7 days after infection. The hearts were examined for myocarditis by image analysis. The results represent the mean percentage area inflammation f SEM. Myocarditis was significantly different from male animals receiving nonimmune (Day 0) female T lymphocytes at (t) P < 0.05 and (**) P < 0.01 level. The number of animals in each group is given in parentheses.

Suppressionof CTL Generation in Vitro In vitro culture of male lymphocytes on glutaraldehyde-fixed monolayers of cardiocytes produces CTL reactive to heart antigens and capable of inducing myocarditis in vivo (3 1). We therefore determined the ability of female spleen cells to inhibit male CTL generation. Females were inoculated with CVB3 and sacrificed at various times thereafter. T-Cell enriched female cells were mixed with equal numbers of 7 day immune male cells and cultured for 5 days on either CVB3 infected or uninfected cardiocytes. Control cultures consisted of male cells cultured with nonimmune female cells or atone on the monolayers. Surviving lymphocytes were assayed for cytolytic activity in the “Cr release assay (Table 1). Cultures of male cells lysed approximately 60 and 100% uninfected and virus-infected targets. Co-culture of male and nonimmune female cells resulted in similar activity. However, addition of immune female cells reduced CTL generation by as much as 60% (Day 3) indicating the presence of suppressor cells. DISCUSSION Male and female Balb/c mice differ markedly in myocarditis induced after CVB3M infection. Males develop severe cardiac inflammation but myocarditis in virgin females is either minimal or absent (1,4, 5, 7). Interestingly, the general protection of female mice from heart disease does not persist during pregnancy when CVB3M-induced myocarditis can be more severe than that observed in males (7). Sex-associated hormones are important in determining whether an animal will develop myocarditis. After castration, CVB3M-induced myocarditis in males is reduced to levels normally observed in virgin females, while castration of females slightly enhances subsequent cardiac disease (5). Exogenous administration of testosterone and progesterone in

SUPPRESSOR CELLS IN VIRAL

111

MYOCARDITIS

TABLE 1 Inhibition of Generation of Male Cytolytic T Cells by Female T Cells in Vitro Cultured on uninfected monolayers and assayed on uninfected heart cell@

Cultured on infected monolayers and assayed on infected heart ceIIsb

None

59.6 f 2.3

104 + 3.1

Nonimmune

12.5 + 2.9

80.1 + 2.5

Day 3

24.1 + 3.8

(66)’

52.1 f 6.8 (35)

Day 6

41.8 f 3.8 (34)

58.5 + 2.6 (271

Day 9

62.5 + 2.2 (141

64.3 + 2.3

57.5 k 1.4

50.1 + 4.4 (37)

Female” cells

Day 12

(21)

(20)

a Female mice were inoculated with 6 X lo4 PFU CVB3. b 5 X 10’ lymph node cells from Day 7 immune males were cultured alone and with an equal number of nonimmune or immune female lymphocytes for 5 days on monolayers of glutaraldehyde-fixed uninfected or CVB3-infected heart cells. The surviving cells were assayed for cytotoxicity to uninfected and CVB3 infected cells at an E/T of 100: 1 and an incubation time of 18 hr. Results are given as mean percentage lysis zk SEM. ’ Percentage inhibition of cytolytic activity generation in the presence of immune female cells compared to nonimmune female cells.

physiological concentrations to castrated animals is highly effective in inducing severe myocarditis (5, 7) while similar administration of estrogens can be protective (5). It is not clear how the sex-associated hormones influence pathophysiological disease processes. Females are generally more resistant to infectious diseases than males ( 13- 15) either because of a more vigorous humoral immune response ( 1,16) or because androgen treated cells are more permissive toward virus infections [( 17), unpublished observations]. Frequently, enhanced immunity in females is associated with decreases in suppressor cell activity. Thus, estrogens have been reported to preferentially increase helper T-cell activity (18, 19) while testosterone supposedly stimulates suppressor cells (18-22). The present system is unusual in that the relative protection of female mice from myocarditis appears to depend upon the preferential generation of suppressor cells in this sex and an inhibition of the cellular immunity generally seen in susceptible animals. The evidence that suppressor cells are generated in infected female mice includes the demonstration of enhanced myocarditis after administration of CY, a drug known to preferentially eliminate suppressor T cells when given at low concentrations (50 mg of CY/kg body wt) (23). Neither cardiac virus concentrations nor virus-specific antibody titers were affected in CY-treated females developing aggravated heart disease, but the treated animals demonstrated slightly enhanced cytotoxicity to infected myocytes (data not shown). This observation confirms that the CY regimen used did not either inhibit humoral immunity or cause aggravated disease by poorly controlling the virus. The timing of CY administration can be critical. Accumulated evidence on suppressor

112

JOB, LYDEN,

AND HUBER

cells suggests that they are rapidly dividing cells and are generated very soon after antigen sensitization (24, 25). As an alkylating agent affecting DNA replication, CY should maximally inhibit these early cells (26,27). Furthermore, the effects of CY are short-lived, since administration of the drug before virus inoculation apparently had little effect on the immune response. Thus the optimal time of CY action was either on the day of virus inoculation or within 24 hr of this event. Interestingly, inflammation is only minimally enhanced when CY is given 3 to 5 days after infection even though both virus-specific antibody and CTL activity are elevated. Immunological effector cells may arise in the lymph nodes 2 to 4 days following CY administration, but migration from the peripheral lymphoid organs to the heart does not occur. During the more extended period, both induction and migration of the effector cells could be completed. The experiments employing CY give strong suggestive evidence that virgin females are actively protected from developing myocarditis by the generation of suppressor cells. However, the experiments employing adoptive transfer of the female cells into otherwise susceptible male animals more directly proves both the existence of the suppressor cells and their role in myocarditis. Myocarditis was suppressed in virusinfected males when the animals received isolated immune female T cells. While some apparent suppression could be noted with spleen cells obtained 3 days after infection of the females, the suppressive effect was maximal with Day 9 immune female T cells, suggesting the presence of populations of mature suppressor cells at this time. Suppressor cells have been tentatively identified by other investigators studying CVB3-induced myocarditis (28). Specifically, suppressor cells were identified in male mice infected with a nonmyocarditic variant of CVB3. This variant has been shown to infect the heart, producing virus concentrations equivalent to that observed in CVB3M infected mice, but no autoimmunity to myocyte antigens is observed. Therefore, suppressor cells may be a common mechanism responsible for preventing this disease and only those individuals lacking the specific suppressor activity may be at risk for myocarditis. ACKNOWLEDGMENTS We thank Ned Gutman and Marianne Feran for technical assistance and Mrs. Jackie Marceau for help in preparing the manuscript.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

Woodrutf, J. F., Amer. J. Pathol. 101,425, 1980. Woodruff, J. F., and Woodruff, J. J., J. Immunol. 113, 1726, 1974. Wang, C. Y., Woodruff, J. J., and Woodruff, J. F., J. Immunol. 118, 1159, 1977. Wong, C. Y., Woodruff, J. J., and Woodruff, J. F., J. Immunol. 119, 591, 1977. Huber, S. A., Job, L. P., and Auld, K. R., Cell. Immunol. 67, 173, 1982. Huber, S. A., and Lodge, P. A., Amer. J. Pathol. 116,21, 1984. Lyden, D. C., and Huber, S. A., Cell. Immunol. 87, 462, 1984. Wang, C. Y., Woodruff, J. J., and Woodrutf, J. F., J. Immunol. 118, 1165, 1977. Huber, S. A., Job, L. P., and Woodrut?‘, J. F., Amer. J. Pathol. 98,68 1, 1980. Debre, P., Walter&au.&, C., Dot-f M. E., and Benacerraf, B., J. Exp. Med. 142, 1436, 1976. Sym, S. N., Miller, S. D., and Claman, H. N., J. Zmmunol. 119,240, 1977. 12. Chiorazzi, N., Fox, D. A., and Katz, D. H., .I. Immunol. 118,48, 1977. 13. Tompson, D. J., Gezon, H. M., Hatch, T. F., Rycheck, R. R., and Rogers, K. D., N. Engl. J. Med. 269, 337, 1963.

SUPPRESSOR CELLS IN VIRAL 14. IS. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.

MYOCARDITIS

113

Washburn, T. C., Medeares, J. L., and Childs, B., Pediatrics 35, 5 1, 1965. Friedman, S. G., Grota, L. J., and Glasgow, L. A., Infect. Immunol. 5,637, 1972. Cohn, D. A.,Arthritis Rheum. 22, 1218, 1979. Kalter, S. H., Smohn, J., McElhaney, J. M., and Tepperman, J., J. Exp. Med. 93, 529, 1951. Weinstein, Y., and Berhovich, Z., J. Immunol. 126, 998, 1981. Weinstein, Y., Ran, S., and Segal, S., J. Immunol. 132, 656, 1984. O’Heam, M., and Stiles, D. P., Cell. Immunol. 76, 340, 1983. Huston, D. P., Smathers, P. A., Reeves, .I. P., and Steinberg, A. D., Clin. Immunol. Immunopathol. 26, 91, 1983. Roubinian, J. R., Talal, J. S., Greenspan, J. R., Goodman, J. R., and Siiteri, P. K., J. Exp. Med. 147, 1568, 1978. Berebaum, M. C., Immunology 7,65, 1964. Taylor, R. B., and Basten, A., Brit. Med. Bull. 32, 152, 1976. Pierce, C. W., and Kapp, J. A., Tohuku J. Exp. Med. llS(Suppl9), 1976. Mitsuoka, A., Morikawa, S., Baba, M., and Harada, T., J. Exp. Med. 1018, 1979. Stevens, J. G., and Cook, M. L., J. Exp. Med. 133, 19, 197 1. Gaunt&C. J., Trot&ale, M. D., LaBadie, D. R. L., Paque, R. E., and Nelson, T., J. Med. Viral. 3,207, 1979. Huber, S. A., and Job, L. P., Infect. Immunol. 39, 1419, 1983. Adler, K. B., Brody, A. R., and Craighead, J. E., Proc. Sot. Exp. Biol. Med. 166, 96, 1981. Huber, S. A., Job, L. P., and Woodruff, J. F., Infect. Immun. 43, 567, 1984.