Assessment of in situ host immunity to syngeneic tumors utilizing the multicellular spheroid model

CELLULAR

IMMUNOLOGY

85,

340-350 (1984)

Assessment of in Situ Host Immunity to Syngeneic Utilizing the Multicellular Spheroid Model’ EDITHM.LORD~ANDGINA

Tumors

BURKHARDT

Department of Microbiology, Division of Immunology, Cancer Center, university of &&ester, 601 Elrnwood Avenue, Rochester, New York 14642 Received August 8. 1983; accepted January 20, 1984 In situ host immunity to the EMT6/Ro mammary sarcoma tumor was evaluated by implanting multicellular spheroids of this tumor into the peritoneal cavity of syngeneic BALB/cKa mice and determining the kinetics of host cell infiltration and tumor cell killing. Spheroids grown in vitro and implanted into unsensitized mice continued to grow resulting in peritoneal tumor masses and eventual death of the animal. However, in mice previously sensitized with a single injection of heavily irradiated EMT6/Ro cells, spheroids implanted intrapcritoneally were rapidly infiltrated by host immune cells (macrophages, lymphocytes, and granulocytes), tumor cell killing was detectable within 1 day and by Day 6 essentially no clonogenic tumor cells were recoverable. Despite this marked loss of both total and clonogenic tumor cells, there was little decrease in the diameter of the spheroids recovered during this time period. Physical size thus does not provide a reliable estimation of tumor cell killing. The tumor cell killing was immunologically specific in that little killing was observed when EMT6/Ro spheroids were implanted into mice sensitized with other allogeneic or syngeneic tumor cells. Host cells from within the spheroids were found to be cytotoxic for EMT6/Ro tumor cells in a “Cr release assay. A major portion of these cytotoxic cells appear to be T lymphocytes. However, other host cell types may also be involved in the in vivo tumor cell killing.

INTRODUCTION The role of the host immune response in combating neoplastic growth has been extensively studied over the past several years. It has been possible to demonstrate specific humoral and cellular antitumor immunity in tumor-bearing animals and patients (1, 2). The majority of these studies have utilized serum antibodies, or cells from the spleen, regional lymph nodes, or the peripheral blood and demonstrated with in vitro assays that these have cytostatic or cytotoxic effects on tumor cells (1, 3). The assumption is that host cells within the tumor mass are similar to those present in the periphery and that in vitro cell reactivity is a correlate of or at least indicative of the in vivo situation. Only recently, have investigators begun studying the host-tumor cell interactions within the tumor itself. Histological studies have been used in several animal tumor ’ This investigation was supported by Grant CA-28332 and by the Core Support Grant, University of Rochester Cancer Center CA-I 1198-10. * Reprint requests should be sent to Dr. Edith M. Lord, University of Rochester Cancer Center, Box 704, 601 Elmwood Avenue, Rochester, N.Y. 14642. 340 0008-8749184 $3.00 Copyright 0 1984 by Academic Rcs, Inc. AU rights of reproduction in any form rraemd.

IN SITU HOST IMMUNITY

IN A MODEL

TUMOR

SYSTEM

341

systems (4-7) as well as in human tumors (8, 9) to identify the types of host cells which infiltrate solid tumors. In addition, infiltrating host cells are now being isolated by dispersing solid tumors into single-cell suspensions. This has been accomplished for both primary (lo- 12) and transplanted ( 13- 15) tumor cell lines in mice and rats. A variety of human tumors (2, 16) have also been dissociated and the composition of the host cell infiltrates analyzed. A major problem with the studies has been the difficulty in isolating tumor-infiltrating cells in a functionally viable state. The use of combinations of enzymes such as trypsin, collagenase, and DNase has proven relatively effective with tumors such as the Maloney sarcoma (7). Despite the improvement in cell yields using this and other similar methods, it is still not possible to completely dissociate a tumor and the severe mechanical and enzymatic procedures required for dissociation makes it difficult to accurately assessthe functional capability of the recovered cells. Because of these difficulties in working with solid tumors we have chosen to study in situ host-tumor cell interactions using the multicellular spheroid tumor model originally developed by Sutherland and colleagues (17). These spherical structures which are composed solely of tumor cells and which have many of the properties and characteristics of in vivo solid tumors (18) can be implanted into mice, where they are infiltrated by host cells. A distinct advantage of this model system is that the spheroids can be completely dissociated by a mild trypsinization procedure without any apparent loss of cells. This model has previously been used by MacDonald et al. ( 19, 20) and ourselves (2 1, 22) to study allograft reactions. In this study we have implanted multicellular spheroids of EMT6/Ro sarcoma cells into syngeneic BALB/ cKa mice and analyzed the kinetics of host cell infiltration, tumor cell killing, and the development of cytotoxic host cells. MATERIALS

AND

METHODS

Mice. BALB/cKa female mice were purchased from BioBreeding Laboratories of Canada, Ltd, Ottawa, Ontario, Canada and used at lo-12 weeks of age. Growth of tumor cells and spheroids. The EMT6 tumor, which arose from a spontaneous hyperplastic nodule in a BALB/cKa, was originally obtained from Dr. R. Kallman and has been maintained as frozen stocks from several alternate in vivo to in vitro passages. These cells designated EMT6/Ro (Rochester) have been used for all of the experiments reported here. Multicellular tumor spheroids (MTS) of EMT6/Ro cells were grown in vitro as previously described (22). Briefly, spheroids were initiated by placing 2.5 X lo4 exponentially growing cells into 60-mm nontissue culture petri dishes (Labteks), whose nonadherent surface favors intracellular contact and results in small spheroids (loo150 pm in diameter) after 3-4 days. These were then transferred to 300-ml spinner flasks (100 rpm) with 125 ml of Eagle’s basal medium (GIBCO, Grand Island, N.Y.) supplemented with 15% fetal bovine serum (FBS)3, penicillin (100 units/ml), and streptomycin ( 100 &ml). After 4 more days of growth when the spheroids had reached a diameter of 300-400 pm, they were sorted to obtain a homogeneous sized population, and 500 spheroids were placed in spinner flasks (190 rpm) with 200 ml 3 Abbreviations used: BME, Eagle’s basal medium; BSS, Hanks’ balanced salt solution; C, complement; CFE, colony-forming efficiency; FBS, fetal bovine serum; HEPES, N-2-hydroxyethel piperazine-N-2ethenesulfonic acid; MTS, multicellular tumor spheroid, PC, peritoneal cells; SAC, spheroid-associated cells.

342

LORD

AND

BURKHARDT

of medium which was replenished daily. Spheroids of 700-900 pm diameter obtained after 2-3 weeks growth were used. Size was determined by examining 20-40 spheroids with an inverted microscope having a calibrated reticle in the eyepiece. Two diameters, a, and b, at 90” were measured for each spheroid and the diameter calculated from (ab)‘12. Immunizations. Mice were immunized 15 days prior to spheroid implantation with a single intraperitoneal (ip) injection of 1 X 10’ heavily irradiated (5000 r) EMT6/ Ro cells from in vitro monolayer cultures. Spheroid implantation and recovery. As described earlier (22), 60 spheroids/mouse were implanted into the peritoneal cavity using an 1&gauge needle. At varying times later, the mice were sacrificed and the spheroids were recovered by repeatedly flushing the peritoneal cavity with BSS. The spheroids, separated from the peritoneal cells (PC) by gravity sedimentation, were washed three times with BSS and dissociated into a single cell suspension by trypsinization with 0.05% trypsin (Sigma Chemical Co., St. Louis, MO.) for 12 min at 37°C. Cell counts of the trypsinized spheroidassociated cells (SAC) were obtained using an electronic particle counter set for optimal counting of particles of the size of EMT6 cells and by parallel hemocytometer counts. The peritoneal cells, which contained very few tumor cells, were washed by centrifugation and counted using a hemocytometer. Colony-forming eficiency (CFE) assay. The CFE of the SAC and of cells dissociated from control spheroids maintained in vitro were determined by plating the cells at four dilutions in 60-mm petri dishes (five plates/dilution) in 5 ml of BME supplemented with 15% FBS. After 10 days incubation the plates were stained with methylene blue and scored for colonies. Cytotoxicity assay. The cytotoxic potential of SAC and PC were determined using a “Cr release assay. Exponentially growing EMT6/Ro cells (5 X 106) were harvested from monolayer cultures, washed, and resuspended in 1 ml of RPM1 1640 medium supplemented with 5% FBS, antibiotics, and 10 mM HEPES buffer (assay medium). We added 200 &i Na25’Cr04 (Amersham/Searle Corp., Arlington Heights, Ill.) and the cells were incubated at 37°C for 1 hr. After three washings the labeled target cells were resuspended in assay medium and 1 X lo4 cells placed in each well of Linbro IS-FB-96 TC microtiter culture plates in a total volume of 0.2 ml/well. After an additional 2 hr incubation to allow cell attachment, the target cells were gently washed with BSS and twofold dilutions of the test effector cells added at effector:target ratios from 1OO:l to 6.25:1 with triplicate wells at each dilution. Following incubation of the plates at 37°C for 20 hr, one half the culture supernatant (0.1 ml) was removed from each well and radioactivity was counted using a well-type gamma counter. Spontaneous release controls contained target cells but no lymphoid cells. Spontaneous release was routinely 30-33% over the 20-hr culture periods. Maximum release was determined by incubating target cells in 1.0 N HCl. Percentage specific lysis was calculated using the following formula: experimental “Cr release - spontaneous release x 100. maximal release - spontaneous release The calculated specific lysis was then plotted versus the effector:target ratio, parallel lines were fitted to the linear portion of the resulting curves using a computer program, and lytic units/ lo6 cells calculated from these lines. One lytic unit (LU) is defined

IN SITU HOST IMMUNITY

IN A MODEL

TUMOR

SYSTEM

343

as the number of effector cells required to cause 30% 51Cr release under these conditions (3). Morphological analysis. Slides were prepared by centrifuging 0.5 ml of a single cell suspension containing 5 X lo5 cells/ml of the SAC or PC for 5 min at 500 t-pm on a Shandon-Elliott cytocentrifuge. The slides were air-dried, stained with Wright’s Giemsa stain, and differential counts made on at least 500 cells/slide. RESULTS Kinetics of Spheroid Tumor Cell Killing

The fate of spheroids implanted into the peritoneal cavity of unsensitized syngeneic BALB/cKa mice or mice which had been previously sensitized with heavily irradiated EMT6 cells was followed by recovering the spheroids at daily intervals and determining the CFE of the dissociated tumor cells. The results of a representative experiment are shown in Fig. 1. The CFE of tumor cells from spheroids implanted into unsensitized mice remained fairly constant for at least 10 days and was only slightly lower than that of spheroid cells maintained in vitro (Fig. 1A). Similar results were obtained when these data were plotted as colonies/spheroid (CFE times the number of cells

I

UNSENSITIZED

UNSENSITIZED

SENSITIZED

SENSITIZED

2 DAYS

AFTER

4

6

6

IO

IMPLANTATION

FIG. 1. Kinetics of spheroid tumor cell hilling in syngeneic mice. EMT6 spheroids (diameter +- SD: 848 pm + 45) were implanted into groups of sensitized or unsensitized BALB/c mice (60/mouse). At daily intervals, the spheroids were recovered, dissociated, and assayed for clonogenic tumor cells. The data for control spheroids maintained in vitro (circles), unsensitized (squares), and sensitized (triangles) groups are expressed as colony forming efficiency (A) and as absolute number of clonogenic EMT6 tumor cells per recovered spheroid (B).

344

LORD AND BURKHARDT

recovered/spheroid) as shown in Fig. 1B. Expression of the data in this fashion insures that one is seeing actual tumor cell killing and not just dilution of the tumor cells by infiltrating host cells. In contrast to unsensitized mice, when spheroids were implanted into sensitized mice there was an immediate decline in the CFE of the recovered tumor cells which was detectable by the first day. The CFE decreased steadily so that by Day 9 it was less than 0.01% (Fig. 1A). This decline was even more dramatic when expressed as colonies/spheroid from the sensitized mice indicating that some tumor cells are being completely lysed during the 9 days.

Kinetics of Host Cell Injiltration Infiltration of the spheroids by host cells was monitored daiy by doing differential counts on cytocentrifuge preparations of the dissociated SAC. In spheroids implanted in unsensitized mice the total number of tumor cells increased slowly over the 9-day period that followed (Table 1). Granulocytes, which were the predominant host cell present, were at high numbers within 1 day and stayed at a high level. Macrophages, which were also present in large numbers (approximately half that of granulocytes) reached these levels by Day 2 and then remained roughly constant. Lymphocytes, a third type of host cell present never reached any appreciable number and at no time comprised more than l-4% of the total cells. TABLE

1

Kinetics of Host Cell In&ration into Spheroids Implanted into the Peritoneal Cavity of Syngeneic Mice”

Mice Unsensitized

Sensitized

Time after implantation (days) I 2 3 4 5 6 7 9

Total cells

Tumor cells

Lymphocyt=

Macrop&es

Granulocyt=

Cells/spheroid (X l@)b 38.6 43.4 34.3 32.9 32.6 50.6 57.1 42.0

10.8 9.6 10.7 12.6 11.2 15.0 16.0 14.5

1.5 1.7 1.0 0.5 0.6 1.0 0.5 0.8

4.1 11.1 9.2 9.5 7.9 12.6 11.9 1.9

22.3 21.0 13.4 10.2 12.9 22.0 29.1 18.8

35.6 42.0 37.1 38.3 25.4 19.4 18.8 10.3

9.1 5.4 3.6 3.4 1.1 1.0 1.0 0.9

0.4 1.6 1.7 2.0 2.4 2.7 3.0 2.9

4.7 8.8 11.3 22.0 15.3 13.3 12.3 6.1

21.4 26.1 20.4 11.0 6.6 2.4 2.4 0.4

4 EMT6 spheroids (mean diameter ;+ SD: 848 rt_45 am) were implanted in each mouse at time 0. b Single cell suspensions were prepared by. trypsinization of the recovered spheroids, ditferential counts were. done on Wrightaiesma-stained cytocentrifuge preparations and the numbers of different types of cells calculated on the basis of viable cell counts.

IN SITU HOST IMMUNITY

IN A MODEL

TUMOR

SYSTEM

345

A quite different pattern of host cell infiltration was observed when spheroids were implanted into mice previously sensitized to the EMT6 tumor cells (Table 1). At 1 day postimplantation, similar numbers of tumor cells were present as in spheroids in unsensitized mice, but at each subsequent day there was a steady decrease in these cells so that by Day 9 there had been a lo-fold decrease in the number of cells scored morphologically as tumor cells. As with unsensitized mice, granulocytes were a predominant cell type during the first few days but in contrast to the unsensitized mice declined thereafter to low levels. The number of macrophages appeared to be almost inversely related to the granulocyte population in that macrophages were present in low numbers initially and then increased to high and fairly stable levels. As with unsensitized mice, lymphocytes were detected in much smaller numbers than the other host cell populations. However, in the sensitized mice, the absolute number of lymphocytes increased at a slow but steady rate throughout the time period observed, and the relative number increased from 1% at Day 1 to greater than 25% of the total cells by Day 9. Kinetics of Spheroid Size Change In an attempt to correlate visible changes in the spheroids with the extent of tumor cell killing, the spheroids recovered from the mice were sized by measuring the diameters of at least 20 spheroids. During the first 3 days following implantation, the diameter of spheroids implanted in both unsensitized and sensitized mice increases similar to that of spheroids maintained in vitro (Fig. 3) and reflects the total number of cells present within the spheroids (Table 1). At later times the spheroids in unsensitized mice continue to grow at rates similar or greater than that of in vitro controls and is probably due to both increased number of tumor cells and the large numbers of host macrophages and granulocytes which are present. In sensitized mice, after the first few days the spheroids do decrease in size relative to in vitro controls but size does not accurately reflect the fate of the tumor cells. For instance at Day 5 postimplantation the spheroids are still as large as those originally implanted (Fig. 2) even though there has been a IO-fold decrease in the total number of tumor cells (Table 1) and a loo-fold decrease in clonogenic cells/spheroid (Fig. 1B). Despite the maintenance of their overall diameter, visible alterations with time in the appearance of the spheroids implanted in sensitized mice were readily apparent. The spheroids became slightly elongated and appeared almost translucent. Histological examination of these spheroids revealed a very narrow rim one to two tumor cells thick with the remainder being an almost hollow sac containing loosely packed host cells. Thus, despite detectable changes, accurate quantitation of host effects by merely following alterations in spheroid size is not possible. Cytolytic Capacity of the Spheroid Injiltrating

Cells

In order to assessthe functional activity of the spheroid infiltrating cells as well as the cells free in the peritoneal cavity, these cells obtained from sensitized mice were used as effector cells in an in vitro assay against “Cr-labeled EMT6 target cells. The kinetics of the “Cr release assay were first established by incubating the effector and target cells together for various times from 6 to 22 hr. Minimal cytolytic activity was detectable at 6 hr but this increased reaching a plateau at 18-20 hr (Fig. 3). The kinetics were similar for both the PC and SAC effector cells. The kinetics of lysis and

346

LORD AND BURKHARDT

0.6 0

I

2

3 DAYS

4

5

AFTER

6

7

8

9

IMPLANTATION

FIG. 2. Kinetics of spheroid size alteration in syngeneic mice. Sixty EMT6 spheroids were implanted into the peritoneal cavity of each BALB/c mouse. At daily intervals the spheroids were recovered from 3 unsensitized (squares) or 3 sensitized (triangles) mice or from in vitro culture flasks (circle) and sized. Each point represents the mean diameter of at least 20 spheroids from each of 3 mice.

5’Cr release in this syngeneic system are thus much longer than those observed previously in an allogeneic system (20). A 20-hr incubation period was used for all subsequent assays in which the cytolytic activity of PC and SAC recovered at daily intervals following spheroid implantation was determined. Activity was detectable in

4.0

\ 2

2.a

I-

0 c >

I.0

I-

12

6 LENGTH

OF

18 ASSAY

20

22

(HRS.)

FIG. 3. Kinetics of the %r release assay using syngeneic EMT6 target cells. 5’Cr-labeled EMT6 target cells were incubated for varying lengths of times with varying ratios of effector cells from spheroids (SAC) or peritoneal cells (PC) recovered 5 days atIer implantation in sensitized BALB/c mice.

IN

HOST IMMUNITY

SITU

IN A MODEL

TUMOR

SYSTEM

347

both the PC and SAC from previously sensitized mice within 1 day (Fig. 4). The activity of the PC did not increase markedly and after Day 4 decreased to very low levels. In contrast the activity of the SAC rose sharply reaching a maximum at Day 6. This demonstration of in vitro cytolytic activity by the host cells within the recovered spheroids led to the concern that our results on tumor cell viability assessed by the CFE assay might be influenced by the presence of cytolytic host cells. Control experiments were thus designed to directly test this possibility. EMT6 tumor cells dissociated from control in vitro maintained spheroids were combined at varying ratios with SAC or PC from sensitized mice as a source of cytolytic host cells and the CFE for these defined mixtures determined. We were unable to demonstrate any diminution in the colony forming efficiency of EMT6 cells in the presence of equal or greater numbers of host cells (data not shown). These differences between the assays would, at first, appear to be contradictory, however, it is important to take into account the differences in cell densities. In the CFE assay a maximum of 1 X lo4 cells are plated in 60-mm petri dishes having a surface area of 2827 mm2 while in the 5’Cr assay lo4 EMT6 tumor cells and up to lo6 host cells are plated in microtiter wells having a surface area of 32 mm2. Since cell contact is crucial for at least T cell mediated cytotoxicity the lesser probability of cell contact in the CFE assay than in the 5’Cr release assay or in the tumor itself may well explain the lack of cytotoxicity. Specificity of Spheroid Tumor Cell Killing The immunological specificity of tumor cell killing in the multicellular spheroid model was confirmed by implanting spheroids into BALB/cKa mice which were either unsensitized, sensitized with allogeneic (H-2b) tumor cells (EL4), another H-2d tumor (P8 15) or with syngeneic EMT6 tumor cells. The number of clonogenic cells/spheroid

6.0

y _1 w 0 IO 0\ 2 z 3 ”

-

5.0.

4.0-

30-

2.0-

r 5

I oPC

I 0

I

2 DAYS

3 AFTER

4

5

6

7

IMPLANTATION

FIG. 4. Kinetics of cytolytic activity of host cells from sensitized syngeneic mice. EMT6 spheroids were recovered and dissociated at daily intervals after implantation in sensitized BALB/c mice. The resulting SAC (m) and PC (0) were used at varying ratios as effector cells against EMT6 target cells in a 20 hr assay.

348

LORD AND BURKHARDT

of the spheroids recovered 5 days postimplantation from either the EL4 or P8 15sensitized mice was not significantly different from that of unsensitized mice (Fig. 5). In contrast almost no clonogenic cells were recovered from the specifically sensitized mice. This specificity is also reflected in the type and number of host and tumor cells in the spheroids recovered at Day 5 from each of the four groups of mice. In the EL4 and P8 15-sensitized mice, as in the unsensitized mice, large numbers of tumor cells were present and the predominant host cells were macrophages and granulocytes while only very low numbers of lymphocytes were observed (Fig. 6). In the specifically sensitized (EMT6) mice, the spheroids contain very few tumor cells and the host cell infiltrate is almost devoid of granulocytes and instead consists primarily of macrophages and a sizable number of lymphocytes. DISCUSSION Many investigators have examined the immunological response to tumor cells with immune cells from lymphoid organs such as the spleen or peripheral blood, but much less is known about what occurs within the tumor itself. Studies on the host cells from the tumor have been more limited, in part due to the difficulty in isolating such cells in a quantitative manner while maintaining their functional capabilities. In this study we have overcome this major problem by utilizing multicellular tumor spheroids,

PSI5 SENSITIZING

EMT6

CELLS

FIG. 5. Immunological specificity of in viva spheroid tumor celi killing. The number of clonogenic tumor cells/spheroid is shown for EMT6 spheroids recovered 5 days after implantation into BALB/c mice which were either unsensitized or sensitized with EL4, P8 IS, or EMT6 tumor cells.

IN SITU HOST IMMUNITY MICE

SENSilTlfING

CELL TYPE

S

I

IN A MODEL

TUMOR

TUMOR

CELLS/SPH;ROID I

SYSTEM ‘6”10-4

I

I.

)

8

CELLS

LYMPHOCYTES MACROPHAGES

LYMPHOCYTES MACROPHAGES GRANULOCYTES

I

TUMOR

BALE/c

CELLS

b

I

EMT6

FIG. 6. Effect of sensitization on host cell infiltration into in vivo implanted EMT6 spheroids.

which unlike in vivo solid tumors can be easily dissociated into single suspensions without any apparent loss of viable cells. As a model system, spheroids are artificial in the sense that they lack a vascular system. They do, however, have the diffusionlimited microenvironments present within solid tumors at distances from the vascular supply as well as the obvious geometric similarities. These easily manipulable structures thus allow a quantitative assessment of the ability of host cells to both infiltrate and function within tumor-like microenvironments. Spheroids of EMT6 mammary sarcoma cells implanted into syngeneic BALB/cKa mice were rapidly infiltrated by host cells composed primarily of macrophages and granulocytes with a much smaller number of lymphocytes. Distinctly different patterns of host cell infiltration were observed between spheroids implanted into unsensitized mice and those in mice previously sensitized to the EMT6 tumor cells. In sensitized mice, the lymphocyte infiltrate, although small, increased steadily while in unsensitized mice lymphocytes failed to increase with time following implantation. Similarly, granulocytes which were present in high numbers initially in both groups of mice remained at high levels in spheroids in unsensitized mice but decreased markedly in the sensitized mice. Macrophages comprised a large proportion of the infiltrates with both types of mice. Because the spheroids are grown in vitro and are composed solely of tumor cells, there is no doubt that the lymphoid cells present within the recovered spheroids were due to infiltration by the host’s cells. Similarly, large host cell infiltrates in in vivo solid tumors have been reported by several investigators in both experimental animal tumors ( 10, 12- 15,23) and human neoplasms (2, 24, 25). In most of the tumors examined T lymphocytes (2, 10, 24, 25) or macrophages (10, 13, 14,23) have been the predominant host cell type identified. Other investigators examining several types of human tumors have described highly variable compositions of host cell infiltrates (26). The presence of large numbers of host cells within tumors implies a functional interaction with the tumor cells. We have been able to demonstrate with the multicellular spheroid system that in specifically sensitized mice, host cells are able to infiltrate and kill tumor cells as evidence by the loss of clonogeneic tumor

350

LORD AND BURKHARDT

cells (Fig. 1). Furthermore, we have shown direct cytotoxicity of EMT6 cells by host cells recovered from the spheroids (Fig. 4) in an in vitro long-term (20 hr) 51Cr release assay. Unlike the previously studied allogeneic system (11, 18), little activity was detectable in short-term (4-6 hr) assays. The difference in the kinetics of this in vitro assay for allogeneic and syngeneic effector cells is in agreement with those recently reported by Huber and Lucas (27) for cytolysis by host cells in a rat adenocarcinoma tumor system. Our preliminary experiments indicate that T lymphocytes are responsible for a large percentage of the activity measurable in the 5’Cr release assay, but other host cell types may also participate. Additional studies are currently in progress to characterize the cytolytic host cells and determine their specificity. REFERENCES I. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 2 1. 22. 23. 24. 25. 26. 27.

Hellstrom, K. E., and Hellstrom, I., Immunology 18, 209, 1974. Hersh, E. M., Mavligit, G. M., Gutterman, J. U., and Barsales, P. B., Med. Ped. Oncol. 2, 1, 1976. Cerottini, J. C., and Brunner, T., Adv. Immunol. 18, 67, 1974. Birbeck, M. S. C., and Carter, R. L., Int. J. Cancer 9, 249, 1972. Hanna, M. G., Jr., Snodgrass, J. M., Zbar, B., and Rapp, H. J., Natl. Cancer Inst. Monogr. 35, 345, 1972. Kihxh, K., Ishii, Y., Ueno, H., and Koshiba, H., Ann. N.Y. Acad. Sci. 276, 188, 1976. Russell, S. W., Doe, W. F., Hoskins, R. G., and Cochrane, C. G., J. Zmmunol. 116, 164, 1976. Black, M. M., Barelay, T. H. C., and Hankey, B. F., Cancer 36, 2048, 1975. Mansell, P. A., Ichinase, H., Reed, R. J., Krementz, E. T., McNamee, R., and DiLuzie, N. R., J. Natl. Cancer Inst. 54, 571, 1975. Blazar, B. A., and Heppner, G. H., J. Immunol. 120, 1881, 1978. Holden, H. T., Haskill, J. S., Kirchner, H., and Herberman, R. B., J. Immunol. 117, 440, 1976. Wood, G. W., and Gallahan, K. A., Cancer Res. 38, 1857, 1978. Evans, R., Transplantation 14, 468, 1972. Haskill, J. S., Practor, J. S., and Yamamura, Y., J. Natl. Cancer Inst. 54, 387, 1975. Russell, S. W., Doe, W. F., Hoskins, R. G., and Cochrane, C. G., Znt. J. Cancer 18, 322, 1976. Blomgren, H., Glas, U., Franzen, S., and Granberg, P. O., Acta. Radiol. (Ther.) 12, 434, 1973. Sutherland, R. M., McCredie, J. S., and Inch, W. R., J. Natl. Cancer Inst. 46, 113, 197 1. Sutherland, R. M., and Durand, R. E., Curr. Top. Rad. Res. Quart. 11, 87, 1976. MacDonald, H. R., and Howell, R. L., Transplantation 25, 136, 1978. MacDonald, H. R., Howell, R. L., and McFarlane, D. L., Transplantation 25, 141, 1978. Lord, E. M., and Nardella, G., Transplantation 29, 119, 1980. Lord, E. M., Penney, D. P., Sutherland, R. M., and Cooper, R. A., Virchows Arch. B. Cell Pathol. 31, 103, 1979. Evans, R., in “The Macrophage in Neoplasia” (M. Fink, Ed.), pp. 27-42. Academic Press, New York, 1976. Klein, E., Becker, S., Suedmyr, E., Jondal, M., and Vanky, F., Ann. N.Y. Acad. Sci. 276, 207, 1976. Husby, G., Hoagland, P., Strickland, R., and Williams, R. C., J. C/in. Invest. 57, 1471, 1976. Hayry, P., and Totterman, T. H., Eur. J. Immunol. 8, 866, 1978. Huber, S. A., and Lucas, Z. J., J. Immunol. 121,2270, 1978.