Role of Corynebacterium parvum in the activation of peritoneal macrophages

CELLULAR

IMMUNOLOGY

76, 49-57 (1983)

Role of Corynebacterium Activation of Peritoneal

pan/urn in the Macrophages

II. Identification of Distinguishable Anti-tumor Activities by Macrophage Subpopulations’ STEPHEN

K. CHAPES~ AND STEPHEN

HASKILL

Cancer Research Center, Department of Obstetrics and Gynecology, and Department of Bacteriology and Immunology, University of North Carolina, Chapel HiiI, North Caroka 27514 Received August 24. 1982; accepted November 19, 1982 Two different mechanisms of murine macrophage (MP) antitumor activity are described in this report. C. parvum-activated peritoneal MPs were tested for cytotoxic and cytostatic activity 4 days after ip immunization. Cytotoxic activity could be distinguished from cytostatic activity using two different assay protocols. When MPs were separated by lg velocity sedimentation, cytotoxic MPs were confined to high velocity fractions. In contrast, cytostatic MPs were found in cell fmctions with velocities as low as 5.2 mm/hr. These two MP activities were also distinguishable by culturing at 37°C for 24 hr. Cytotoxicity was abrogated when MPs were incubated in MEM, or MEM supplemented with lymphokine (LK) or indomethacin. In contrast, cytostasis remained at high levels when the cells were incubated with LK or indomethacin. Cytotoxicity was not retained after overnight culture even if LPS was present, or if various spleen or noaadherent peritoneal exudate cells were cocultured with the cytotoxic effector cells. Assays done to determine the presence of suppressor cells failed to find any inhibitory cell type. The phagocytic index, acid phosphatase activity, and H20z secretion were also measured before and after overnight culture. Acid phosphatase and phagocytic activities did not decline whereas HzO, secretion declined significantly. Them data indicate that in response to C. parvum, at least two different effector cell types with distinct antitumor activities are generated. Cytotoxicity, like the ability of cells to secrete H20z, is found to be a short-lived function of CP stimulated MPs. In contrast, cytostasis is a function retained longer by MPs in culture. INTRODUCTION

The generation and description of tumoricidal macrophages (MP@ has been a major interest of tumor immunologists in recent years. The ability to attract and activate MPs with biological response modifiers gave promise that MPs could be utilized as effector cells in immunotherapeutic protocols (1, 2). In more recent years, a vast heterogeneity within the monocyte-macrophage lineage has been demonstrated. Macrophage subpopulations with distinct functional activities (3-5), along with different biochemical activities (6-8) have been isolated. ’ This investigation was supported by National Cancer Institute Grant 1-POlCA29589. ’ SKC is supported by National Institutes of Health, National Service Award CA09156 from the National Cancer Institute. ’ Abbreviations used: CP, C. parvum; MP(s), macrophage(s); LK, lymphokine; PEC, peritoneal exudate cell; VSG, velocity sedimentation gradient; LPS, lipopolysaccharide; H-S3, Hela S-3. 49 0008-8749183 $3.00 Copyright Q 1983 by Academic Press. Inc. All rigbtr of reproduction in any form reserved.

50

CHAPES AND HASKILL

The ways MPs exert their tumoricidal influences are under intense discussion at this time. The secretion of cytolytic factor (9), the utilization of oxygen intermediates ( 10, 1 1), direct cell-cell contact mechanisms ( 12- 14), and even the ingestion of tumor cells has been reported (15). With the description of such diverse mechanisms of tumoricidal activity and with the cytochemical and physical heterogeneity found with macrophages, it is not di5cult to hypothesize different tumoricidal mechanisms in different MP subpopulations. We have previously described that there is a strong correlation between large, C. parvum (CP)-containing MPs and the cytotoxic activity against tumor cells ( 16). This laboratory has also established that C. pamum-activated MPs can be both cytolytic and cytostatic to tumor cells, depending upon the stage of the tumor cell growth cycle ( 17). We describe here that the different antitumor activities are associated with different MP subpopulations. These subpopulations were identified by both cell fractionation and in vitro culture methods. MATERIALS

AND

METHODS

Animals. AB6Fr mice, 6 to 8 weeks old, were obtained from the Trudeau Institute (Saranac Lake, N.Y.). Immunization of animals. Eight- to twelve-week-old animals were immunized in our studies. Corynebacterium parvum (C. parvum, Burroughs Wellcome Co., Research Triangle Park, N.C.), at a concentration of 700 CLgIO.1 ml, was injected ip to activate MPs. Cell cultures. HeLaS3 (H-S3) cells were used as target cells in the MP cytotoxicity and cytostasis assays. Cells were grown in minimum essential medium (MEM, Gibco, Grand Island, N.Y .) supplemented with L-glutamine, 20 &ml gentamycin, and 10% FBS (KC Biological Co., Kansas City, MO.). Peritoneal exudate cells (PEC). The PEC were harvested from mice killed by cervical dislocation. Cells were obtained by washing the peritoneal cavity with lo-20 ml ice cold PBS and were washed in MEM. Macrophage harvest. Macrophages were recovered from the PEC as described above, 4 days after ip injection of CP. The percentage of MPs in the PEC was determined by differential staining using Wright’s stain. The PEC were counted in a hemocytometer and were diluted to a concentration of 2.0 X lo6 MPs/ml to be used in a cytotoxicity assay. Target cell preparation for cytotoxicity assays. H-S3 cells were seeded into 25-cm2 flask 2 to 3 days before use. Before the assay, medium was poured off and fresh medium containing 5 @/ml of [3H]thymidine (Schwartz/Mann, Spring Valley, N.Y.) was added. The cells were then incubated 24 to 30 hr. Radiolabeled cells were dispersed from the tissue culture flask by washing the cells from the plastic with strong jets of media from a Pasteur pipet. Cells were washed, counted, and adjusted to a concentration of 5 X lo4 viable radiolabeled cells/ml. Target cell viability, determined by trypan blue exclusion, was always greater than 95% after dispersion. Target cells, 1 X lo4 per well were used in each MP cytotoxicity assay. Adherent macrophage cytotoxicity assay. The MP cytotoxicity assay was performed in 96-well microtiter plates employing [3H]thymidine-prelabeled targets. Macrophages were added to microtiter plate wells in a volume of 0.1 ml at a concentration

DISTINGUISHABLE

MACROPHAGE

ANTITUMOR

ACTIVITIES

51

of 20 X lo5 MPs/ml. Cells were allowed to adhere for 1 to 2 hr and wells were washed vigorously to leave only adherent cells (>85% MPs). Target cells were added to wells in a volume of 0.2 ml at a concentration of 5 X lo4 cells/ml. The assay plates were incubated 18 to 20 hr at 37°C in a humidified COZ incubator. After incubation, the percentage specific 3H was calculated by removing 0.1 ml of supematant from each well and counting the amount of activity in a scintillation counter (Packard Instruments Co., Rockville, Md.). Maximum release was determined by placing 1 X lo4 cells directly into Aquasol and counting. Spontaneous release was determined by incubating target cells in media for the duration of the assay and ranged between 7 and 15% for these assays. The % specific 3H release for each sample was calculated as follows: % specific release =

% release with MPs - % spontaneous release x loo ’ % maximum release - % spontaneous release

Cytostasis assays. Unlabeled H-S3 cells, 1 X 104, were added to adherent MP effector cells in flat-bottomed 96-well tissue culture plates. After a 24-hr incubation, 0.5 &i [3H]thymidine was added to each well. The assay was harvested on an automated cell harvester 6 hr later. Growth inhibition was determined as follows: % cytostasis = 100 -

100 X

cpm target cells with MPs - cpm MPs alone CPM target cells alone

This laboratory has previously found a lack of cold thymidine artifacts using this assay with CP-activated MP by comparing it to flow cytometric analysis of growth inhibition ( 17). Velocity sedimentation gradients (VSG). The VSG was used previously, to separate C. parvum-stimulated MPs by this laboratory ( 16) and is described in detail (18). Flow cytometry. Acid phosphatase enzyme was detected using flavone-3diphosphate substrate. This procedure has been described by this laboratory previously (19). Phagocytosis was detected by using isatoic anhydrideconjugated, human AB serumopsonized zymosan particles. This procedure has also been described previously by this laboratory (19). Macrophages analyzed for fluorescence (acid phosphatase or zymosan ingestion) were analyzed on an Ortho ICP-22 flow cytometer. A TRS-80 Model I Computer was utilized for data analysis (see 17). Macrophage HzOz secretion. Analysis of H202 secretion was done using the procedure described previously by Root et al. (20, 21). MPs, 1 X lo6 were incubated for 60 min at 37°C with 10 IrM scopolotin or scopolotin and horseradish peroxidase (12 mg/ml) or pulsed with 10 pg/ml PMA and scopolotin and horseradish peroxidase in Hanks’ balanced salt solution. Total reaction volume was 1.5 ml. One milliliter was harvested after 60 min and was diluted with 1.5 ml Hanks’. Samples were assayed in a 1 X l-cm cuvette on a fluorescence spectrophotometer set for 350-nm excitation and emission set at 460 nm. A standard curve was used to determine H202 concentration by adding a 1.6 A4 hydrogen peroxide standard at appropriate dilutions to the reaction mixture. Spleen cells. Spleens were aseptically removed from normal or CP-immunized mice. Spleens were disrupted by expression through Nytex fabric. Cells were washed, centrifuged on Picoll/Hypaque, and mononuclear cells were adhered two times on

52

CHAPES AND HASKILL

,.t 4

FIG.

, 5

6 7 VELOCITY

8 9 (mdhr)

10

II

12

1. Cytotoxicity and cytostasis of lg separated macrophage subpopulations.

tissue culture plastic. Nonadherent cells >98% lymphocytes were then added to MP monolayers for overnight culture. Miscellaneous.LPS was obtained from Sigma (Escherichiacoli serotype No. 055:B5). Macrophages were incubated overnight in media with LPS at a concentration of 100 &ml. Lymphokine was obtained by pulsing normal mouse spleen cells with concanavlin A (1 &ml in MEM) for 24 hr. Supematant was spun at 2000g to remove cells and debris and was frozen at -20°C until use. RESULTS

Comparison of MP Cytotoxic and Cytostatic Activity We have previously shown that MP cytotoxic activity is closely associated with cells containing intracellular CP 4 or 5 days postinjection. To determine if other MP antitumor functions followed this same correlation, we analyzed MP-mediated growth inhibition of H-S3 cells using the [3H]thymidine incorporation postlabeling technique. Macrophage cytostasis was found in all fractions isolated using lg velocity sedimentation, whereas the cytolytic MP activity was only present in the largest cells that we previously showed contained CP (16) (see Fig. 1).

TIME ka~rs) FIG. 2. Cytostasis (A and B) and cytotoxicity (C and D) of freshly harvested macrophages or macrophages cultured in MEM or MEM containing lymphokine or 10e6 M indomethacin.

DISTINGUISHABLE

MACROPHAGE TABLE

ANTITUMOR

53

ACTIVITIES

1

Ability of Various Treatments to Retain Cytotoxic Macrophages after 24-hr Culture Percentage specific ‘H release Experiment

Treatment

24-hr Culture

Fresh

None C. parvum (0.7 jkg) Lymphokine None Indomethacin ( 10e6 M) +Nonadherent PEC Lymphokine None LPS Lymphokine None +Normal SPC +Immune SPC

46.8 f 52.3 f 75.7 f 34.8 + -

5.6

1.1 3.1 1.7 12.2 13.6 6.1 11.9 6.4 9.6 7.0 14.0 12.5 17.4

6.5

2.2 1.9

f + f f f + f + + f f + +

1.0 2.2 0.9 1.3 1.3 0.8 1.3 2.4 3.2 2.6 2.5 3.3 3.5

Influence of Overnight Culture on Cytotoxic and CytostaticActivity To further define these different effector cells, CP-activated MPs were incubated at 37°C overnight. Cytolytic activity was abolished when MPs were incubated either in MEM or MEM supplemented with lymphokine (see Fig. 2 C and D). In several experiments, the cytostatic activity of MPs also decreased following incubation in MEM overnight, although this was always less than the loss of cytolytic activity (see Fig. 2, A and B). In contrast, cytostatic activity remained close to or as high as in fresh MPs following incubation in lymphokine-containing MEM (see Fig. 2, A and B). We were concerned that the generation of prostaglandins during the culture period inhibited the expression of cytotoxicity as has been described previously (3 1). In order TABLE 2 Phagocytic and Biochemical Activity of Freshly Harvested and Cultured C. parvum-Activated Macrophages Activity MP C. parvum C. parvum C. parvum C. parvum

Assay Cytotoxicity Acid phosphatase Zymosan ingestion HZ02 Secretion

Fresh

24-hr Culture

68.4” 23.2’ 39.4b 17.5’

-0.3 23.4 44.6 7.5

’ Numbers represent mean specific ‘H release of triplicates. b Values shown are the weighted means of fluorescence intensity for cells assayed for acid phosphatase (AP) using fluorescent AP substrate or phagocytosis using fluorescent isatoic anhydride conjugated, op son&d zymosan. ’ HzOz secretion by 106 MPs after pulse of 10 &ml PMA. Units are nm per lo6 MPs.

54

CHAPES AND HASKILL

VELOCITY

kt-mdhr)

FIG. 3. Cytotoxicity of freshlyharvested(0) or 24-hr-cultured(0) lg separatedmacrophagesubpopulations.

to investigate this possibility, CP-activated MPs were also assayed for cytotoxicity and cytostasis after overnight culture in the presence of 10e6 M indomethacin. With this treatment (Fig. 2, C and D), cytotoxicity decreased as with MEM-incubated MPs. Cytostasis, however, remained at levels comparable to lymphokine-incubated MPs. Treatments Used to Influence Retention

of Cytolytic Activity after Overnight Culture

To examine the loss in cytolytic activity during the 24-hr in vitro culture, MPs were exposed to various factors and cell types throughout the incubation period. Addition of lop6 M indomethacin to the cultures had no influence on loss of cytotoxicity. Similarly, macrophages incubated overnight with additional C. parvum, LPS, and various combinations of peritoneal exudate cells, normal spleen cells, and immune spleen cells failed to prevent the loss in cytotoxicity (see Table 1). Measurement

of A4P Phagocytic and Biochemical

Activity after 24 Hr Culture

To determine how overnight culture affects other MP activities, phagocytic and acid phosphatase levels were measured in freshly harvested and 24-hr-cultured MPs. We found no change in the level of acid phosphatase after culture (Table 2). In the representative experiment shown, the mean values of fresh and cultured MPs were 23.2 and 23.4 units, respectively. Phagocytic activity increased from 39.4 to 44.6 units. Thus, both phagocytic and acid phosphatase activity were not lost with culture. In contrast, when PMA-induced Hz02 secretion was measured, we found a loss of activity. Freshly harvested MPs produce greater than twice the amount of H202, as did 2Chr-cultured MPs. Assay for Suppressor Cells Since indomethacin did not reverse the loss of cytolytic activity with 24 hr culture (see Table 1, Fig. 2) other mechanisms of suppression were examined. Macrophages were fractionated into subpopulations by lg VSG. These MP subpopulations were cultued overnight and tested for cytotoxicity. We hoped the fractionation of MPs might remove possible suppressor populations from the cytotoxic MPs. Figure 3 shows that although MPs >7.0 mm/hr were cytolytic when assayed immediately after harvest, these same MPs became significantly less cytotoxic with overnight culture. Since it was possible that the suppressor cells might be isolated in the same subpopulations as the killer MPs, we used another protocol to assessthe possible presence

DISTINGUISHABLE

MACROPHAGE

ANTITUMOR

55

ACTIVITIES

TABLE 3 Coculture of Fresh Cytotoxic C. parvum Macrophages with Incubated Normal or C. parvum Macrophages Percentage specific “H release Cultured 24-hr Normal

Fresh C. parvum

Normal

No. MPS cultured

None

+Fresh CP-MP

None

+Fresh CP-MP

None

+Fresh CP-MP

4x 2x 1x 0.5 x 0.1 x

3.9” 1.7 1.0 1.8 NT’

65.86 60.4 52.6 72.6 NT

16.4 14.4 4.4 1.8 0.6

62.1 15.8 66.6 63.3 66.0

3.9 2.1 1.8 0.9 2.8

52.1 52.2 66.3 58.1 71.8

10’ lo5 lo5 lo5 105

DNumbers represent mean of triplicate wells, 20-hr assay. b Cytotoxicity of C. parvum MPs alone = 42.8. 2 X lo5 fresh CP-MPs assayed for cytotoxicity. ’ Not tested.

of suppressor MPs (see Table 3). In these experiments, various numbers of CPactivated MPs were cultured overnight. Freshly harvested CP-stimulated MPs were then added to the cultured MP monolayers and were assayed for cytotoxicity. If a suppressor cell was present in the cultured CP-MPs, we expected suppression of our freshly harvested MPs. We found (see Table 3) no suppression of freshly harvested CP-MPs in experiments using this protocol. DISCUSSION Previous studies in our laboratory have established that there are two different antitumor activities among C. parvum-stimulated macrophages. Four days following injection, the presence of both cytolytic as well as growth inhibitory cells could be detected in a flow cytometric assay (17). This past study, while suggesting that both mechanisms could be present at the same time, did not address the question of a single or multiple effector cell origin of these activities. In the present report, we have attempted to study this. We have been able to isolate MPs with distinct antitumor activities. Cytotoxic MPs can be distinguished from cytostatic cells using two different assay protocols. Whereas cytotoxic MPs are limited to the large, CP-containing fractions of lg velocity sedimentation-fractionated cells, cytostatic MPs were isolated in fractions as small as 5.2 mm/hr. Macrophages expressing these two activities were also separated by culturing cells overnight at 37°C. Cytostatic cells remain active when cultured in media containing lymphokine or 10m6M indomethacin, but macrophage cytotoxic activity was lost under similar conditions. This demonstration of functional heterogeneity among CP-activated MPs is consistent with previous studies that separated MPs into different functional subpopulations (3-5,22). These studies were able to separate cytolytic cells from noncytolytic and/or helper MPs or separated MPs with differing phagocytic activities. Several groups have published results which can be contrasted with our data. Lee

56

CHAPES

AND

HASKILL

et al. (23) found both growth inhibitory and cytotoxic MPs were limited to the large cell fractions. Weinberg et al. (24) reported that the largest cells separated on BSA discontinuous gradients were the most cytostatic of the MP subpopulations. Becker and Haskill (25) however, described that cytotoxic and cytostatic activities were unseparable by lg VSG with MSV-induced MPs. It is possible that some of these observed differences result from different target cell sensitivities. Previous studies have suggested that some target cells are susceptible only to the cytostatic effects of macrophages while other targets may be growth inhibited and/or lysed (26-28). Lee et al. (23) needed a longer assay to find both cytolytic and cytostatic MPs against P8 15 tumor cells. For others, the expression of MP cytotoxicity depended on the MP, the treatment of the target cell, and assay time (29). We have found that the time MPs are tested postinjection with CP can influence the assayable cytotoxicity of CP-stimulated cells. The association between bacteria, size, and cytotoxicity is best at 4 or 5 days. At 8 days, most large bacteria-containing MPs have disappeared and cytotoxicity is much lower by the smaller, remaining MPs (Chapes and Haskill, unpublished observation). The loss of CP-MP cytotoxicity with overnight culture is consistent with previous reports (11, 30, 3 1). Furthermore, Meltzer et al. (30) were also unable to restimulate MP cytotoxicity with additional or continued lymphokine treatment of their cells. In contrast, the addition of 10d6 A4 indomethacin or endotoxin to the culture media in the systems of Taffet and Russell (3 1) or Cohen et al. ( 1 1), respectively, maintained cytolytic activity of MPs. Treatment of MPs from MSV-induced tumor-bearing animals with LPS was also found to retain all or some cytotoxic activity of 24-hrcultured cells (32, 33). In our hands, these same treatments did not maintain MP cytotoxicity. The complexity of MP activation makes it difficult to account for these apparent differences. It is possible that the cellular source and the different procedures by which each group activates their MPs determines the cell’s functional expression. We employed a single C. purvum immunization to induce cytotoxicity, whereas other groups used lymphokine, LPS, multiple immunizations of BCG, or assay MPs from tumor-bearing animals. Since there was no apparent suppressor cell interference with the expression of CP-MP cytotoxicity and we could not retain cytolytic macrophages by using various culture conditions, we examined other measurable parameters of macrophage function and activation. We found that phagocytic and acid phosphatase activities remained high after overnight culture, whereas the inducible levels of secreted H202 were quite low. This suggested that the loss of ability to secrete H202 might be associated with loss of cytotoxicity. However, Sorrel1 et al. (34) demonstrated that the cytotoxic activity of CP stimulated macrophages does not seem to depend upon oxygen intermediates. Cohen et al. (11) found only a modest decline of H202 secretory activity was associated with the loss of cytotoxicity of BCG activated MPs following overnight culture. It is possible that the relevance of the loss of H202 secretory ability in our system may be related to bacteriocidal processes. In conclusion, we have shown that sedimentation velocity separations can be used to distinguish cytotoxic from cytostatic MPs in CP-stimulated animals. This confirms our previous conclusion that more than one mechanism of antitumor activity can be active at the same time (17). The two simultaneously occurring antitumor mechanisms described here also help explain the description of so many possible means of MP killing in the recent literature.

DISTINGUISHABLE

MACROPHAGE

ANTITUMOR

ACTMTIES

57

ACKNOWLEDGMENTS We thank Dr. M. Cohen, Department of Medicine, University of North Carolina at Chapel Hill, for helping us set up the H20z assay. In addition, we appreciate the excellent laboratory help of Mrs. Willa Bell and secretarial assistance of Mrs. Linda McAlister.

REFERENCES 1. 2. 3. 4. 5. 6. I. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.

Alexander, P., Annu. Rev. Med. 27, 207, 1976. Bast, R. C., Bast, B. S., and Rapp, H. J., Ann. N. Y. Acad. Sci. 277, 60, 1976. Miller, G. A., Campbell, M. W., and Hudson, J. L., J. Reticuloendothel. Sot. 27, 167, 1980. Walker, W. S., J. Reticuloendothel. Sot. 20, 57, 1976. Chapes, S. K., and Tompkins, W. A. F., J. Reticuloendothel. Sot. 30, 5 17, 198 1. Fishman, M., and Weinberg, D. S., Cell. Immunol. 45, 437, 1979. Moore, K., and McBride, W. H., Int. J. Cancer 26, 609, 1980. Beelen, R. H., van’t Veer, M. B., Fluitsma, D. M., and Hoefsmit, E. C. M., J. Reticuloendothel. Sot. 24, 351, 1978. Adams, D. O., Kao, K., Farb, R., and F&o, S. V., J. Immunol. 124, 293, 1980. Nathan, C., and Cohn, Z., J. Exp. Med. 152, 198, 1980. Cohen, M. S., Taffet, S. M., and Adams, D. O., J. Immunol. 128, 1781, 1982. Cabilly, S., and Gallily, R., Immunology 42, 149, 1981. Koren, H. S., Meltzer, M. S., and Adams, D. O., J. Immunol. 126, 1013, 1981. Hibbs, J. B., Science 184, 468, 1974. Key, M., and Haskill, S., J. Nat. Cancer Inst. 66, 103, 1981. Chapes, S. K. and Haskill, S., Ceil. Immunol. 70, 65, 1982. Haskill, S., J. Reticuloendothel. Sot. 31, 85, 1982. Haskill, S., In “Manual of Macrophage Methodology” (H. B. Harcowitz et al., Fds.), pp. 43-50. Dekker, New York, 1981. Haskill, S., and Becker, S., J. Reticuloendothel. Sot. 32, 273, 1982. Root, R. K., and Metcalf, J. A., J. Clin. Invest. 60, 1266, 1977. Root, R. K., Metcalf, J. A., Oshino, N., and Chance, B., J. Clin. Invest. 55, 945, 1975. Campbell, M. W., Sholley, M. W., and Miller, G. A., Cell. Immunol. 50, 153, 1980. Lee, K. C., Kay, J., and Wong, M., Cell. Immunol. 42, 28, 1979. Weinberg, D. S., Fishman, M., and Veit, B. C., Cell. Immunol. 38, 94, 1978. Becker, S., and Haskill, S., Int. J. Cancer 25, 535, 1980. Keller, R., Brit. J. Cancer 30, 401, 1974. Krahenbuhl, J. L., Lambert, L. H., and Remington, J. S., Cell. Immunol. 25, 279, 1976. Krahenbuhl, J. L., Cancer Res. 40, 4622, 1980. Chapes, S. K., and Tompkins, W. A. F., J. Immunol. 123, 303, 1979. Meltzer, M. S., Ruco, L. P., and Leonard, E. J., Advan. Exp. Med. Biol. 121, 381-398, 1980. TalIet, S. M., and Russell, S. W., J. Immunol. 126, 424, 1981. Russell, S. W., Gillespie, G. Y., and McIntosh, A. T., J. Immunol. 118, 1574, 1977. Taniyama, T., and Holden, H. T., Int. J. Cancer 24, 15 1, 1979. Sorrell, T. C., Lehrer, R. I., and Cline, M. J., J. Immunol. 120, 347, 1978.