ancer mmunology mmunotherapy

Cancer Immunol Immunother (1983) 16:59-64

© Springer-Verlag 1983

Cytotoxic T lymphocytes in DBA/2 mice harboring L5178Y cells in a tumor-dormant state* Mark A. Marsili, Michael K. Robinsovl*, Gary A. Truitt, and E. Frederick Wheelock Department of Pathology and Laboratory Medicine, Hahnemann University, 245 North 15th Street, NCB, Philadelphia, PA 19102, USA

Summary. Previous experiments have demonstrated a temporal relationship between the decline of cy~'otoxic T lymphocyte (CTL) activity in the peritoneal cavity of DBA/2 mice harboring L5178Y cells in a tumor-dormant state and the appearance of ascitic tumors. Some tumor-dormant mice remain clinically normal for many weeks after the decline of CTL activity, and this activity can be rapidly restimulated by an IP inoculation of irradiated L5178Y cells. We report here that the peritoneal cells from many tumor-dormant mice can be stimulated to cytolytic activity in vitro when cultured for 4 days either with or without the addition of irradiated L5178Y cells. Peritoneal cell populations which cannot be stimulated in vitro can suppress the generation of CTL in those populations which can be stimulated. The tumor-dormant state may terminate when suppressor cells in the peritoneal cavity of tumor-dormant mice inhibit the generation of CTL activity and permit tumor cells to produce an ascitic tumor.

Introduction A tumor-dormant state is established in DBA/2 mice by subcutaneous (SC) immunization and intraperitoneal (IP) challenge with syngeneic L5178Y lymphoma cells [21, 22]. This tumor-dormant state is characterized by the long-term persistence of small numbers of L5178Y cells in the peritoneal cavity of clinically normal mice. A number of host cells have been implicated in the maintenance of the tumor-dormant state, including cytotoxic T lymphocytes (CTL) [21], cytotoxic macrophages [15], and noncytotoxic macrophages, which combine with memory T lymphocytes to produce a synergistic cytotoxic effect [16]. Cytotoxic lymphocytes are responsible for the reduction of the initial tumor burden during establishment of the L5178Y cell tumor-dormant state [21 ], with CTL activity gradually declining to background levels by 4 0 - 7 0 days post-L5178Y cell challenge. We have recently found that CTL activity in tumor-dormant mice can be repeatedly and rapidly re-elicited by IP inoculations of irradiated L5178Y cells [12], and that this re-elicitation is associated with elimination of tumor cells from approximately 50% of mice. The in vitro Reprint requests should be addressed to: Dr. E. F. Wheelock * Present address: Department of Immunochem[stry Research, Evanston Hospital, Evanston, IL 60201, USA. Abbreviations used in this paper: C', complement; CTL, cytotoxic thymus-derived lymphocyte; PC, peritoneal cells; DPC, clays post challange, NAD, nonadherent; SC, subcutaneous; IP, intraperitoneal

experiments described herein are an extension of these observations and were undertaken to detect and identify host cells in the peritoneal cavity of tumor-dormant mice which can rapidly develop CTL activity against L5178Y cells. We report here that many mice which harbor L5178Y cells in a tumor-dormant state contain T lymphocytes in t h e i r peritoneal cavity which can be rapidly stimulated to cytolytic activity when cultured in vitro. The peritoneal cell populations of some tumor-dormant mice cannot be stimulated in vitro, and these populations can suppress the generation of CTL activity in vitro in those populations which can be stimulated.

Materials and methods Mice. Female DBA/2 mice at 8 weeks of age were obtained from the Jackson Laboratories, Bar Harbor, Maine, USA. Tumor cell lines. The methylcholanthrene-induced DBA/2 T cell lymphoma was obtained from Dr Lionel Manson of the Wistar Institute, Philadelphia, PA, USA. These cells were used in all immunization, challenge, and CTL restimulation protocols. The FLC-745 cell is a transplantable Friend leukemia virus-induced erythroleukemia line syngeneic to DBA/2 mice, and was obtained from the Camden Institute for Medical Research. Cell lines were maintained by weekly IP passage of 5 x 106 tumor cells in normal DBA/2 mice and as suspension cultures in RPMI 1640 (Grand Island Biological Co,, Grand Island, NY, USA) containing 10% fetal bovine serum (FBS), 100 ~g/ml streptomycin, 100 ~g/ml penicillin, and 50 ~tg/ml gentamycin. This medium is referred to as 'culture medium'. All cell cultures were incubated at 37° C in a humidified atmosphere containing 5% CO2 in air. Immunization and challenge protocol for establishment of the tumor-dormant state. The L5178Y tumor-dormant state was established in DBA/2 mice as previously described [21]. Briefly, mice received an SC implant of 106 viable L5178Y cells on the mid-ventral surface and 10 days later the resulting 1-cm tumor nodules were surgically excised. The mice were challenged with a single IP inoculation of 5 x 104 viable L5178Y cells 7 days after tumor nodule excision. The immunized and challenged mice resist the rapid outgrowth of the L5178Y cell challenge inoculum and maintain small numbers of L5178Y cells in the peritoneal cavity for prolonged periods.

60

Clinical normalcy. Immunized and challenged mice were judged to be clinically normal if the total body weight did not exceed 30 g and the abdomen was not distended. These clinically normal mice were identified as tumor-dormant if a partial or complete peritoneal lavage revealed <- 5 x 104 tumor cells in the peritoneal cavity. Complete peritoneal lavage and enumeration of tumor cells. Mice were killed by cervical dislocation and the PC removed in two 5-ml peritoneal washes with PBS. The PC were pelleted, resuspended in 3 ml MEM, and counted by hemacytometer. This technique recovers > 99% of peritoneal cells (12). The tumor cells were enumerated by serial end-point dilution analysis (SEPD) as previously described [21]. Four replicate sets of dilutions of each sample were made. A known number of PC was placed in the first well of a 96-well flat-bottomed microtiter plate and serially diluted two-fold through 24 wells. After 2 weeks in culture at 37 ° C the last well in each dilution series that was positive for tumor cell growth was recorded and the mean number of tumor cells in the first well was calculated. From this number, the number of tumor cells per 106 PC was determined and used to calculate the total tumor burden in the mouse. Using the SEPD technique the plating efficiency of L5178Y cells assayed either alone or in the presence of L5178Y-immune PC has consistently been 100% in our laboratory [12].

Preparation of effector cells. Mice were killed by cervical dislocation. Peritoneal cells (PC) were obtained by two 5-ml peritoneal lavages with PBS. These cells were washed once by centrifugation and resuspended in culture medium for use in in vitro assays. Nonadherent cells were prepared by adherence of PC to plastic tissue culture dishes for 45 min followed by gentle pipetting of the nonadherent cells.

Anti- Thy 1.2 treatment of effector cells. Anti-Thy 1.2-producing hybridoma cells were obtained from Dr T . T . MacDonald, Jefferson Medical College, Philadelphia, PA. These cells were cultured in complete RPM11640 with 10% fetal calf serum in a humidified 5% CO2 atmosphere at 37 ° C for 7 days. The cell culture supernates were harvested after centrifugation and were filtered through a 0.45 ~tm nalgene filter (Sybron Corp., Rochester, NY). Treatment of peritoneal cells with the anti-Thy 1.2 and complement was performed as follows: The peritoneal cells at 5 x 106 cells/ml were incubated at 4 ° C for 30 rain with anti-Thy 1.2 antibody at a dilution of 1 : 27, which was the terminal dilution to eliminate the lyric activity of allogeneic CTL and to lyse > 95% of syngeneic thymocytes in the presence of complement. Unbound antibody was then removed by a single wash with culture medium. Freshly thawed guinea pig serum was used as a complement source and was added at a final concentration of 1 : 10. The cells were incubated in a 37° C water bath for 45 rain to allow complement fixation and lysis of Thy-l.2-bearing cells. After this incubation, cells were washed three times in culture medium and used as effector cells.

Labeling of tumor cell targets. Tumor cells (5 x 106) obtained from log phase suspension cultures were suspended in 1.0 ml culture medium and incubated at 37°C in a humidified 5% CO2 atmosphere for 90rain with 250 ~tCi of Na2-51CrO4 (specific activity 200-500 mCi/mg, Amersham Corporation, Arlington Heights, IL). After incubation, the cells were

washed four times and resuspended in culture medium at appropriate concentrations for use as target cells.

In vitro generation of cytolytie T lymphocytes. PC were obtained from normal or tumor-dormant DBA/2 mice and counted in a hemacytometer with viability determined by erythrosin dye exclusion. The cells were diluted in culture medium to a final concentration of 1.0 x 107/mI. Mixed lymphocyte tumor cell cultures (MLTC) were prepared by incubating 1 x 106 nonadherent PC in a volume of 0.2 ml with or without 104 irradiated L5178Y cells in microtiter wells (Costar, Cambridge, MA). Culture medium for the MLTC included 5 x 10 .5 M 2-mercaptoethanol. After a 4-day in vitro culture period at 37 ° C in a humidified 5% CO2 atmosphere, effector cells were harvested using a Pasteur pipette. The viability of recovered PC varied from 80% to 100%. The recovered cells were washed two times in culture medium and suspended at a concentration of 5 x 106 cells/ml for use as effector cells in in vitro assays.

Assay for suppressor cell activity in PC from tumor-dormant mice. PC pools were obtained from tumor-dormant mice between 55 and 65 DPC or from normal mice, and the N A D subpopulations were prepared. On the day of MLTC inception, the N A D - P C pools were plated at 1 x 106 cells/well in triplicate in microtiter wells. A part of each pool was tested for its ability to be stimulated to CTL activity by co-culture with 104 irradiated L5178Y cells. Another part of each pool at 5 x 105 PC per well was mixed with 5 x 105 PC of each other pool, to test for suppression of CTL stimulation of one pool by another. Irradiated L5178Y cells were added and after 4 days' incubation the cells were harvested as above and used as effector cells against SlCr-labeled L5178Y target cells in an 18-h assay. The number of tumor cells in each culture was determined at the end of the incubation period to rule out suppression of CTL lyric activity by competition from unlabeled L5178Y cells that might have proliferated during the culture period.

Unlabeled target cell competition. The ability of unlabeled L5178Y cells or unlabeled FLC-745 cells to competitively inhibit CTL lysis of 51Cr-labeled L5178Y target cells was measured using a modification of the cold target inhibition assay of Oritz de Landazuri and Herberman [14]. Briefly, various numbers of unlabeled tumor cells were added to microtiter plate wells containing 5 x 105 restimulated CTL and 5 x 10351Cr-labeled L5178Y target cells. Plates were incubated for 18 h at 37°C in a humidified 5% CO2 atmosphere, after which specific lysis was calculated. CLT assays. A n 18-h 51Cr-release assay, as previously described [21], was used to measure CTL activity against 51Cr-labeled target cells. The values listed in Tables 1 - 6 represent mean % 51Cr-release from triplicate cultures. The standard deviation of the experimental CPM release values were always < 10% of the mean values. The formula used to calculate the percentage specific 51Cr release was: % 51Crrelease = CPM in experimental culture - CPM spontaneous release x 100. Max. CPM release - CPM spontaneous release

61 Results

Table 2. L5178Y cell specificity of the afferent and efferent arms of the in vivo-restimulated cell-mediated cytolytic response

Demonstration of L5178Y cell-specific immunologic memory in tumor-dormant DBA/2 mice

Inoculum a

A n I P i n o c u l a t i o n of 2 x 106 irradiated L5178Y cells to t u m o r - d o r m a n t mice rapidly elicits high levels of L5178Y cell-specific C T L activity in t h e p e r i t o n e a l cavity [12]. T a b l e 1 shows t h a t this is a s e c o n d a r y i m m u n e r e s p o n s e a n d is associated with i n c r e a s e d n u m b e r s of P C (mostly l y m p h o i d ) in t h e p e r i t o n e a l cavity. T h e a n t i g e n specificity of t h e a f f e r e n t a n d e f f e r e n t a r m of the C T L r e s p o n s e to a n I P i n o c u l a t i o n of irradiated L5178Y cells to t u m o r - d o r m a n t mice is s h o w n in T a b l e 2. FLC-745 cells, a F r i e n d l e u k e m i a v i r u s - t r a n s f o r m e d t r a n s p l a n t a b l e cell line o r g i n a t i n g in D B A / 2 mice, did n o t elicit cytolytic activity in L5178Y cell t u m o r - d o r m a n t mice a n d were n o t lysed by L5178Y cell-induced cytolytic cells. T h e FLC-745 cells were s h o w n previously to b e n o n - c r o s s - r e a c t i v e with L5178Y cells b u t susceptible to lysis by a l l o i m m u n e C T L [20].

Demonstration of anti-L5178Y memory T lymphocytes in the peritoneal cavity of tumor-dormant mice W e n e x t s e a r c h e d for m e m o r y T lymphocytes in t u m o r - d o r m a n t mice b y e v a l u a t i n g t h e ability of p e r i t o n e a l cells to d e v e l o p C T L activity w h e n c u l t u r e d with i r r a d i a t e d L5178Y cells in vitro. In t h e first e x p e r i m e n t , p e r i t o n e a l cells were h a r v e s t e d f r o m t u m o r - d o r m a n t mice 55 days post-L5178Y challenge. T h e cells w e r e p o o l e d a n d t h e n o n a d h e r e n t p e r i t o n e a l cell p o p u l a t i o n ( N A D - P C ) p r e p a r e d . T h e lytic activity of N A D - P C against L5178Y cells was m e a s u r e d in a 51Cr-release assay a n d t h e n u m b e r of t u m o r cells in the N A D - P C was d e t e r m i n e d by t h e serial e n d - p o i n t dilution t e c h n i q u e . T h e r e m a i n i n g p e r i t o n e a l cells were p l a t e d in m i c r o t i t e r wells at a density of 106 p e r well, a n d 104 i r r a d i a t e d L5178Y cells were a d d e d to o n e half of the wells in the plate. A f t e r 4 days of i n c u b a t i o n , t h e cytolytic activity a n d t h e n u m b e r of t u m o r cells w e r e again d e t e r m i n e d . A s s h o w n in T a b l e 3, p e r i t o n e a l cells f r o m t u m o r - d o r m a n t b u t n o t n o r m a l mice d e v e l o p e d cytolytic activity after 4 days of culture with

PC/mouse b x 106

% 51Cr-releaseC L5178Y

2 x 106 irr. L5178Y cells

17.0

2 x 106 irr. FLC-745 cells

8.7

PBS

7.2

FLC-745

100:1

50:1

60.4

42.6

- 0.6

0.1

-

100:1

50:1

2.0

-

0.9

- 1.7

- 2.3

1.6

-

1.5

- 0.2

- 2.0

PBS or 2 x 106 irradiated L5178Y or FLC-745 cells were inoculated IP into each of three tumor-dormant mice 50 DPC and the PC collected 4 days later b The PC were removed by complete peritoneal lavage and counted in a hemocytometer chamber o The % 51Cr_release determined in triplicate using labeled L5178Y or FLC-745 cells in an 18-h in vitro assay. The numbers presented represent the mean values from three mice per treatment group a

Table 3. In vitro generation of cytolytic activity in pooled peritoneal cells from tumor-dormant mice Cell culture contents a

Day 0

Responder PC

Tumor cells

Tumordomant mice

Normal mice

Tumor antigen source

Day 4 % 51Cr release c

Tumor cells

/106

1106

pC b

PC

% 5aCr release

Peritoneal L5178Y + irr. L5178Y

280

- 3.4

256

56.2

Peritoneal L5178Y

280

- 3.4

512

44.3

0

2.2

Irr. L5178Y

0

0

PC were collected from normal or 42 DPC tumor-dormant mice and the nonadherent cells cultured for 4 days in microtiter wells at a density of 106 cells per well with and without 1 0 4 irradiated L5178Y cells b The tumor cell number was determined by serial end-point dilution as described previously [21] c % 51Cr_release was determined in triplicate in an 18-h assay at an effector-to-target ratio of 100:1 a

Table 1. Generation of secondary cell-mediated anti-L5178Y cytolytic responses in tumor-dormant mice after IP inoculation of irradiated L5178Y cells Mice

Tumor dormant Normal

IP inoculum a

PC per mouse (x 10-6) b

% 5iCr release ° 100:1

50:1

25:1

2 x 106 irr. L5178Y PBS

14.5

7'6.2

64.6

48.6

3.6

0.1

0.8

0.6

2 x 106 irr. L5178Y PBS

4.1

6.1

1.6

-

0.1

3.6

0

2.3

-

3.4

-

a 2 x 1 0 6 irradiated L5178Y cells or PBS were inoculated IP into each of four normal and four tumor-dormant mice 42 days post-L5178Y cell challenge, and the PC collected 4 days later b The number of PC per mouse represents the mean of four mice per group. The standard error for each value given was less than 5% of the mean c % 51Cr_release was determined in triplicate using labeled L5178Y cells in an 18-h assay at various effector:target cell ratios. The standard deviation of the experimental CPM release values was always < 10% of the mean value

Table 4. Effect of anti-Thy 1.2 + C' on the in vitro-generated cytolytic activity of peritoneal cells from tumor-dormant mice Treatment a

Medium alone C' alone Anti-Thy 1.2 + C'

% SlCr-releaseb

(E: T = 100: 1)

Added irr. L5178Y cells

No added irr. L5178Y cells

55.4 41.8 - 1.6

27.2 25.1 - 0.9

Anti-Thy 1.2 + C' treatment was performed on cells at the end of the 4-day culture period as described in Materials and Methods b c~o 51Cr_release at an effector: target cell ratio of 100: 1, based on pretreatment PC number, was determined in triplicate after a 1-day culture. The mean preculture % 51Cr-release value by untreated PC was - 3 . 4 a

62 Table 5. In vitro stimulation of cytotoxic activity in peritoneal cells from tumor-dormant mice Mousea

Tumor cells/106 PC b Day 0

% 51Cr-releaseC Day 4

Day 0

1 2 3 4

0 0 0 0

0 0 0 0

1.8 -4.5 -4.9 - 1.7

5 6 7 8

32 8,192 1,024 1,024

ND 20,480 9,306 20,480

2.1 4.5 3.1 0.3

9 10 11 12 13 14 15

64 10 64 64 2,048 256 1,025

0 0 0 0 27,306 5,818 14,222

4.9 - 3.8 0.2 - 4.1 2.3 4.7 0.1

A Cytolysis

Ratio of cold/hot tumor cells in day 4 CTL assay

Day 4 4.5 6.9 3.1 0.3

2.7 0 1.8 1.4

< 1:1 <1:1 <1:1 < 1:1

9.0 7.8 5.1 1.8

0 3.3 2.0 1.5

ND 2:1 1: 1 2:1

73.4 81.5 61.5 53.4 27.8 44.4 35.5

68.5 85.3 63.3 59.5 25.5 39.7 35.4

-

< < < <

1:1 1:1 1:1 1:1 3:1 < 1:1 2:1

a The nonadherent population of PC from individual tumor-dormant mice, 50-72 DPC were cultured with irradiated L5178Y cells for 4 days at a responder: stimulator ratio of 100 : 1 b Tumor cell enumeration was performed at the beginning and the end of the 4 day culture period using the serial dilution end point culture technique described previously [21] c % 51Cr.release was determined in triplicate in an 18-h assay at an effector : target ratio of 100 : 1. The standard error for each value given was less than 5% of the mean

60-

t .....

u~ 5O < ul Ju~ E 40

. ...........

}._____~.. . . . . .

Mouse source of PC pool (s)

~3 L)

Table 6, Suppression of in vitro generation of CTL activity in stimulatable PC pools by non-stimulatable pools from tumor-dormant mice

30-

u-

% 51Crreleaseb Day 0

Day 4

u~ 2 0 -

10-

Normal N 0 RATIO

1:1

5:1

10=1 20:1

OF NON L A B E L E D z L A B E L E D

TARGET

100:1 CELLS

Fig. 1. Inhibition of anti-L5178Y cytolytic activity of in vitro stimulated P C from tumor dormant mice by various numbers of unlabeled L5178Y ( ) or FLC-745 ( - - - ) cells. The effector PC were obtained from cultures of 106 nonadherent peritoneal cells incubated for 4 days in the presence of 104 irradiated L5178Y cells. The % SlCr-release was measured in triplicate wells in an 18-h in vitro assay using 5 x 105 effector PC, 5 x 10 351Cr-labeled L5178Y target cells per well and various numbers of unlabeled L5178Y or FLC-745 cells per well. The values are expressed as means +_ SEM

irradiated t u m o r cells. A n u n e x p e c t e d observation was that the p e r i t o n e a l cells f r o m t u m o r - d o r m a n t mice also d e v e l o p e d cytolytic activity w h e n cultured in the absence of irradiated L5178Y cells. H o w e v e r , t h e s e cultures did contain small n u m b e r s of viable L5178Y t u m o r cells which h a d b e e n p r e s e n t in the p e r i t o n e a l cavity and which could have acted as an antigen source in vitro. T h e anti-L5178Y cytolytic activity that d e v e l o p e d in the p r e s e n c e or absence of a d d e d irradiated t u m o r cells was c o m p l e t e l y ablated by t r e a t m e n t with anti-Thy 1.2 + C (Table 4). T h e s e results indicate that t u m o r - d o r m a n t mice, but n o t n o r m a l mice, contain resident p e r i t o n e a l

Tumordormant A A + N B B B C C C C

5.7

0.5

9.4

11.0 9.1

7.7 + N + A 11.7 + N + A + B

% SupTumor pression cells per of stimulat- 106 PC able pool (Day 0) by nonstimulatable pool

0

N.A. N.A.

46.2 50.9 13.3 5.2

0 71.1 88.7

N.A.

662

< 1: 1 2:1

2,619

1: 1 < 1:1 2:1

85

< 1: 1 < 1: 1 2:1 < 1: 1

N.A. c

6.2 4.8 7.1

Tumor cell cold hot ratio in SlCr assay (Day 4)

a Pools were comprised of 3 - 5 normal or 38-50 DPC tumor-dormant mice. The PC pools were mixed at a 1:1 ratio b Tumor cell enumeration and % 51Cr-release were performed as in Table 3 c Not applicable since there was no stimulatable pool in the pool mixture

m e m o r y T l y m p h o c y t e s which develop into C T L after antigenic stimulation in vitro. T h e lack of a m o n o c l o n a l antibody specifically reactive with L5178Y cells p r e v e n t e d us f r o m selectively r e m o v i n g the L5178Y cells f r o m the p e r i t o n e a l cell p o p u l a t i o n and evaluating the antigen-specificity of the

63 afferent arm of the CTL response. However, the efferent CTL activity in L5178Y cell-stimulated peritoneal populations was antigen-specific, since the restimulated CTL could lyse L5178Y cells, but not FLC-745 cells (data not shown). We found that some PC pools from tumor-dormant mice did not develop cytolytic activity against L5178Y cells during in vitro culture. To explore this observation further, we evaluated individual tumor-dormant mice 50 days after L5178Y cell immunization and challenge. The results obtained in 15 representative mice are shown in Table 5. The NAD-PC from seven of these 15 mice developed cytolytic activity during 4 days' in vitro culture with irradiated L5178Y cells, and all seven contained L5178Y cells in their peritoneal cavity. Of the eight mice whose NAD-PC did not develop cytolytic activity, four did not contain recoverable tumor cells in their peritoneal cavity. The inability to detect CTL activity in the NAD-PC of four of the tumor-dormant mice which contained tumor cells ( # 5 - # 8 in Table 5) could not have been due to competition for lysis of the labeled cells by the unlabeled tumor cells, since the numbers of these unlabeled cells (expressed as ratios of unlabeled: labeled tumor cells) were less than that which would produce cold target competition (20 : 1) (Fig. 1). These mice, like the four mice which were devoid of tumor cells ( ~ 1 - # 4 ) , may have either contained no memory T cells in their peritoneal cavity or contained suppressor cells which blocked the generation of CTL.

Demonstration of suppressor cell activi(y in nonstimulatable PC populations from tumor-dormant mice To search for suppressor cell activity in the peritoneal cavity of mice whose NAD-PC could not be stimulated to cytolytic activity in vitro, we prepared one NAD-PC pool from normal mice (N), and three pools from tumor-dormant mice (A, B and C), with 3 - 5 mice per pool. The cytolytic activity and tumor cell numbers in each pool were determined, and the pools were then cultured individually and mixed 'with each other (see Materials and Methods for details). Irradiated L5178Y cells were added to all cultures, and after an incubation period of 4 days each culture was evaluated for the number of tumor cells per 106 recovered cells and for anti-L5178Y CTL activity. As indicated in Table 6, the NAD-PC from pool C but not pools N, A, or B, became cytolytic during the 4-day culture. In the mixed pool groups, both nonstimulatable pools A and B prevented the stimulation of pool C. This was not a dilution effect, since pool N did not prevent stimulation of pool C. Also, as in the previous experiment, cold target cell competition by unlabeled L5178Y cells could not have been the cause of the absence of cytolytic activity in the nonstimulated cultures, since insufficient numbers of L5178Y cells were recovered from these cultures to achieve the unlabeled-to-labeled target cell ratios needed to reduce cytolytic activity in the 18-h 51Cr-release assay. The results from this set of experiments indicate that some tumor-dormant mice contain peritoneal cells which can either directly or indirectly inhibit the stimulation of memory T lymphocytes to become CTL.

Discussion

We have demonstrated that memory T lymphocytes are present in the peritoneal cavity of many DBA/2 mice which harbor L5178Y cells in a tumor-dormant state. These cell populations did not express cytolytic activity on removal from

the peritoneal cavity, but developed cytolytic activity when cultured in vitro (Table 3). There are several reasons for the ability of these PC to develop detectable cytolytic activity in vitro but not in vivo: (1) stimulation of memory T cells [2, 11] may in fact occur in vivo, but at a microenvironmental level undetectable when tested directly from the mouse in an in vitro cytolytic assay; (2) the L5178Y cells that are maintained in a tumor-dormant state in vivo may not express sufficient tumor-associated antigen to stimulate memory T lymphocytes, possibly because they are in a nonproliferative phase, either arrested in Go or partially or completely differentiated. When these tumor cells are placed in culture, they could resume cycling and express antigen; cell cycle effects on antigen expression have been well documented [1, 3]; (3) there may be insufficient contact between memory T lymphocytes and L5178Y cells in vivo during the tumor-dormant state to stimulate these memory cells, whereas the high cell density and environmental conditions of in vitro culture permit such contact and stimulation to occur; and (4) only the NAD-PC population was cultured in these experiments; recent experiments reveal that many tumor-dormant mice contain macrophages which have an increased capacity to suppress the generation of CTL in mixed lymphocyte-tumor cell cultures (M. K. Robinson et al., 1983, Cancer Res, in press). These suppressor macrophages may inhibit the generation of CTL activity in vivo and their removal from the in vitro cultures may permit the NAD-PC to develop CTL activity. The inability to stimulate cytolytic activity in the peritoneal cell populations of certain tumor-dormant mice in vitro can be explained in a number of ways. These peritoneal cell populations may not have contained memory T cells or they may have contained suppressor cells which prevented stimulation of the memory T cells to cytolytic activity. These suppressor cells may have been either T lymphocytes or macrophages contaminating the NAD population [5-19, 18-20]. The ability to stimulate CTL activity in the peritoneal cell population of some tumor-dormant mice in vitro is at variance with the observation that all tumor-dormant mice develop CTL activity in their peritoneal cavity in response to an IP inoculation of irradiated L5178Y cells [12]. Tumor-dormant mice with in vitro=nonstimulatable PC may harbor memory T cells outside the peritoneal cavity and have these cells drawn into the peritoneal cavity in response to tumor antigen. Alternatively, suppressor cells may be present in the peritoneal cavicty and the large inocuhim of irradiated tumor cells may have overwhelmed these cells and stimulated CTL activity. Immune-mediated restraint on tumor cell growth has been implicated in a number of tumor-dormant animal systems [4, 13, 23, 24]. In mice harboring L5178Y cells in a tumor-dormant state memory T lymphocytes may be present in the peritoneal cavity as a reservoir of potential cytolytic cells which regulate the tumor-dormant state by differentiating into CTL in response to proliferating L5178Y cells. These CTL may then lyse the tumor cells and prevent them from proliferating into an ascitic tumor. This could occur at a microenvironmental level and thereby escape detection in in vitro assays. Suppressor cells, if present, could inhibit the generation of cytolytic activity in memory T cells and thereby permit the tumor cells to proliferate and produce an ascitic tumor. Recent experiments have revealed that tumor-dormant mice harboring suppressor cells in their peritoneal cavity will soon develop ascitic tumors (M. K. Robinson et al., 1983, Cancer Res., in press). The L5178Y cell tumor-dormant model may thus

64 represent an in vivo kinetic system of reciprocal conversion of C T L to m e m o r y T cells to CTL, driven by tumor antigen expressed on proliferating L5178Y cells and susceptible to immunoregulatory influences.

12.

Acknowledgements. This research was supported by grant CA 32577

13.

awarded by the National Cancer Institute, Department of Health and Human Services. MKR was the recipient of National Research Service Award IF32 CA06940.

14.

References 1. Barlogie B, Drewinko B (1977) Growth kinetics and biochemical regulation of normal and malignant cells. In: Drewinko B, Humphrey RM (eds) Williams & Wilkins, Baltimore, p 315 2. Burton RC, Fortin JL, Russell PS (1980) T-cell responses to alloantigens. I. Studies of in vivo and in vitro immunologic memory and suppression by limit dilution analysis. J Immunol 124 : 2936 3. Cikes MS, Friberg S (1971) Expression of H-2 and Moloney leukemia virus-transformed cell surface antigens in synchronized cultures of a mouse cell line. Proc Natl Acad Sci USA 66 : 566 4. Eccles SA, Alexander P (1975) Immunologically mediated restraint of latent tumor metastases. Nature 257:52 5. Eggers AE, Wunderlich JR (1975) Suppressor cells in tumor-bearing mice capable of non-specific blocking of in vitro immunization against transplant antigens. J Immunol 114:1554 6. Fujimoto S, Greene MI, Sehon AH (1976a) Regulation of the immune response to tumor antigens. I. Immunosuppressor cells in tumor-bearing hosts. J Immunol 116:791 7. Fujimoto S, Greene MI, Sehon AH (1976b) Regulation of the immune response to tumor antigens. II. The nature of immunosuppressor cells in tumor-bearing hosts. J Immunol 116:800 8. Gorczynski RM, Norbury C (1975) Suppressor cells in murine sarcoma virus induced tumors. In: Sighal SK, Sinclair NRStC (eds) Suppressor cells in immunity (international symposium). University of Western Ontario Press, London Ontario, p 149 9. Kirchner H, Wuchmore AV, Chused IM, Holden HT, Herberman RB (1975) Inhibition of proliferation of lymphoma cells and T-lymphocytes by suppressor cells from spleens of tumor bearing mice. J Immunol 114:206 10. Kung JT, Brooks SB, Jakway JP, Leonard LL, Talmage DW (1977) Suppression of in vitro cytotoxic response by macrophages due to induced arginase. J Exp Med 146:665 11. MacDonald HR, Enger HD, Cerottini JC, Brunner KT (1974) Generation of cytotoxic T-lymphocytes in vitro. II. Effect of

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

repeated exposure to alloantigens on the cytotoxic activity of long term mixed leukocyte cultures. J Exp Med 140:718 Marsili MA, Robinson MK, Truitt GA, Wheelock EF (1982) Elimination of L5178Y cells from tumor dormant DBA/2 mice by specific active immunotherapy. Cancer Res 43:15 Minato N, Bloom BR, Jones C, Holland R, Reid LM (1979) Mechanism of rejection of virus persistantly infected tumor cells by athymic nude mice. J Exp Med 149:1117 Ortiz de Landarzuri MD, Herberman RB (1972) Specificity of cellular immune reactivity to virus-induced tumors. Nature New Biol 283 : 13 Robinson MK, Wheelock EF (1981) Identification of macrophage-mediated cytolytic activity as a tumor suppressive mechanism during maintenance of the L5178Y-tumor dormant state in DBA/2 mice. J Immunol 126:673 Robinson MK, Wheelock EF (1982) Synergistic cytolytic activity by combined populations of peritoneal T-lymphocytes and macrophages during the L5178Y cell tumor-dormant state in DBA/2 mice. Cell Immunol 73:230 Takei F, Levy JG, Kilburn DG (1977) Characterization of suppressor cells in mice bearing syngeneic mastocytoma. J Immunol 118:412 Treves AJ, Carnaud C, Trainin N, Feldman M, Cohen IR (1974) Enhancing T lymphocytes from tumor-bearing mice suppress host resistance to a syngeneic tumor. Eur J Immunol 4:722 Umiel T, Trainin N (1974) Immunological enhancement of tumor growth by syngeneic thymus-derived lymphocytes. Transplantation 18 : 244 Weinhold KJ, Wheelock EF (1982) Cross-reacting antigens on L5178Y cells which serve as targets for CTL lysis during establishment of the tumor-dormant state. Cancer Res 42 : 3607 Weinhold KJ, Goldstein LT, Wheelock EF (1979a) The tumor-dormant state: Quantitation of L5178Y cells and host immune responses during the establishment and course of dormancy in syngeneic DBA/2 mice. J Exp Med 149:732 Weinhold KJ, Miller DA, Wheelock EF (1979b) The tumor-dormant state: Comparison of L5178Y cells used to establish dormancy with those that emerge following its termination. J Exp Med 419 : 745 Wheelock EF, Caroline NL, Moore RD (1969) Suppression of established Friend virus leukemia by Statolon. I. Demonstration of a latent infection in clinically normal mice. J Virol 4:1 Wheelock EF, Weinhold KJ, Levich J (1981) The tumor-dormant state. Adv Cancer Res 34:107

Received November 29, 1982/Accepted June 6, 1983