Znt. J. Cancer: 19, 97-106 (1977)

GENETIC ASSOCIATION OF THE HUMORAL AND CELLULAR IMMUNE RESPONSES OF RATS TO MOLONEY SARCOMAS

Bruce C. VEIT, Joe M. JONES, Glenn A. MILLER and Joseph D. FELDMAN Department of lmmunopathology, Scripps Clinic and Research Foundation, La Jolla, California 92037, USA

Brown Norway (BN), Lewis (Le), F, hybrids of Lex B N (LBN) and parent-to-LBN backcross rats were tested for cellular and humoral responses t o a B N Moloney sarcoma. Regardless of AgB phenotype, B N backcrosses produced low levels of cell-mediated cytotoxicity (CMC) that were comparable t o those of B N parents. Le backcrosses developed high levels of CMC similar t o those produced in Le parents. An inverse relationship between levels of CMC and serum antibodies (cytotoxic for tumor cells and anti-p30 of MuLV) was observed; B N parents and backcrosses produced high levels of serum antibodies whereas levels in Le parents and backcrosses were low. L B N hybrids developed relatively high levels of CMC and serum antibodies. An additional finding was that the CMC response in Le parents and back- crosses was directed primarily against tumor- associated antigens rather than histocompatibility antigens expressed on the tumor cells. The results suggest that humoral and cellular responses t o Moloney sarcoma in rats are not determined solely by the major AgB histocompatibility locus but do have a genetic association. This genetic association was detected with a 51Cr release assay which detects 1-cells, suggesting that select populations of effector T-cells may begenetically regulated.

Major histocompatibility antigens in rats are primarily associated with genetic variations in the region known as AgB or H-1 (see Palm, 1970, for review of nomenclature). Like H-2 antigens in mice, AgB antigens in rats play an important role in homograft survival (Silvers and Billingham, 1970). This locus, like its H-2 counterpart, appears to be linked to immune response (Zr) genes that control the level of responsiveness to various antigens (Armerding et al., 1974). In mice, studies of virus- induced oncogenesis have shown an association between H-2 histocompatibility type and suscepti- bility or resistance (Lilly et al., 1964; Tennant and Snell, 1968; Chesebro et al., 1974). Resistance is sometimes determined by H-2 linked l r genes (Oldstone et al., 1973; Lilly and Pincus, 1973; Sato et al., 1973). Recently, Colombatti et al. (1975) described a genetic basis for the susceptibility or resistance of mice to Moloney sarcoma virus (MSV) oncogenesis. In this case, l r genes which control this function are apparently not associated with the H-2 complex. Possible associations of tumor resistance with AgB phenotype have not been thoroughly studied.

We undertook to determine if there is a genetic basis for the immune response to tumor-associated

antigens expressed on rat Moloney sarcoma cells. A previous report (Jones and Feldman, 1975a) indicated that Le rats were highly resistant to in vivo growth of a BN allogeneic tumor cell line, and that these animals may have rejected their tumors on the basis of factors other than AgB antigens. In the present report, we provide substantial evi- dence in support of this concept.

MATERIAL AND METHODS

Animals and tumor cells

Tumor cells used in this study were MST or BM2. Both cell lines were derived from a Moloney sarcoma of BN rats and expressed AgB3 histocompatibility antigens. Their characteristics have been previously described (Jones et al., 1974; Veit and Feldman, 1975). BM2 cells are a subline derived from MST after in vivo passage. The growth patterns and anti- genicities of BM2 and MST were qualitatively the same. MST was used in all experiments except those related to CMC. BM2 cells were used in studies of CMC because they elicited higher levels of CMC in syngeneic recipients and allowed more accurate measurements. Cells to be injected into rats were cultured as suspensions in Spinner's minimal essential medium (MEM) containing 10% fetal calf serum (FCS). Those cells used for in vitro assays were cultured in 250-ml flasks as monolayers and har- vested by trypsinizing for 5 min at 37" C with 5 ml of 0.5 g trypsin+0.2 g EDTA per liter.

BN, Le and LBN rats were obtained from the Scripps breeding colony. Backcrosses were produced by mating BN or Le females with LBN males. Weanling progeny were grouped according to sex and tested for AgB phenotype as described below. Groups of rats 3 to 6 months of age were inoculated subcutaneously (SC) with various numbers of tumor cells as described in the text and legends. Tumor sizes were determined by measuring the largest and the smallest axes with Vernier calipers and recording the mean of the two values.

Received: July 23, 1976 and in revised form September 8, 1976.

98 VEIT ET AL.

Backcross typing

AgB phenotypes of backcross rats were deter- mined by binding assays with 1251-labelled allo- antibody (Jones and Feldman, 1975~). Hyperimmune Le anti-BN (anti-AgB3) and BN anti-Le (anti-AgB1) that was adsorbed to and eluted from spleen cells of BN or Le rats respectively were tested with washed peripheral red and white blood cells of each backcross animal.

Skin grafting

Donor skin grafts approximately 1 cm square and trimmed free of subcutaneous tissue were sutured into place in the axillary region of the recipient with 5-0 silk, 6-8 sutures per graft. Grafts were examined daily, and rejection was scored when the graft became mummified and loosened from its bed.

51Cr release assay for CMC

Two x lo7 BM2 cells were incubated for 30 min at 37" C with 250 pCi of Na, 51Cr04 (ICN, Los Angeles, California) in 0.5 ml of supplemented MEM con- taining 5 % FCS (MEM-S). When peritoneal exudate cells (PEC) were used as targets, BN or Le rats were injected intraperitoneally with 8 ml of proteose-peptone broth 3 days prior to assay. PEC were harvested, washed and labelled as described above.

Cytotoxic effector cells were obtained from spleens and draining lymph nodes of tumor-bearing rats by teasing the organs into cold MEM-S and filtering through fine nylon mesh bags (Tetko, lnc., Elmsford, New York) to remove cell clumps. Viability of cell suspensions were determined by trypan blue exclusion and the desired concentrations of viable cells then prepared in RPMI 1640 (Flow Laboratories, Rockville, Maryland) containing 5 % FCS, anti- biotics-antimycotics (Gibco, Grand Island, New York) and 25 mM Hepes buffer.

Suspensions of 5 x loE cytotoxic eRector cells and 5 x lo4 51Cr-labelled target cells (100:l ratio) were added to flat-bottomed Microtest I1 plates in a volume of 0.25ml. Plates were covered and incubated at 37" C in 95 % air-5 % CO, on a rocking platform at 7 cycles/minute for 10 h (PEC targets) or 16 h (tumor targets). Cells were then resuspended and spun at 1,000 g for 5 min. One-tenth ml supernatant fluid was removed and counted in an automatic gamma counter. Total releasable counts were determined by incubating target cells in 10% NPdO. This value ranged from 88-91 % of the input counts. Spontaneous release from target cells incu- bated in medium alone was 21-23 % of input counts for tumor targets and 17-22% for PEC targets. Percent specific release was calculated as follows:

% Specific release =

CPM target cells CPM target cells +immune cells +norm2 cells

x loo. CPM target cells CPM target cells

t l O % NP,, +lo% NP4,

1251UdR microcytotoxicity assay

The assay was modified according to the method of Seeger et al. (1974), and was based on release of radiolabelled target cells from the surface of a tissue culture well. Subconfluent cultures of MST growing in 25 cm2 tissue culture flasks (lo8 cells/ml in a volume of 15 ml) were pulsed for 24 to 48 h with 7.5 pCi of 1251-Iododeoxyuridine (1a66LUdR, 0.1 pCi/ pl, specific activity 100 pCi/pg). Labelled target cell monolayers were rinsed with medium and cells were removed by incubation with 0.25 % trypsin- EDTA (Flow Laboratories, Rockville, Maryland). Cells were washed three times with complete medium containing 10% FCS and 20 mM Hepes buffer and adjusted to lo8 cells/ml. Plating efficiency in this system ranged from 75% to 95% and remained so for 72 h. Also, tumor cells failed to proliferate under these labelling conditions.

Two ml of medium were then pipetted into the wells of Linbro 16-mm TC trays (FB-l6-24-TC, Linbro Chemical Co., New Haven, Connecticut) and lo5 target cells in 0.1 ml were added. Cells were allowed to adhere overnight. Five million immune or non-immune spleen or lymph-node cells prepared as described for the %r release assay for CMC were added to triplicate wells to a volume of 2.1 ml, and the plates were incubated for 24 and 48 h at 37" C. At the end of the assay, unattached target cells were rinsed from each well. One ml of 1.5% sodium lauryl sulfate in distilled water was added to dissolve attached cells, and the plates were incubated for 30 min at room temperature. The sodium lauryl sulfate solution was then trans- ferred to tubes and counted in an automatic gamma scintillation spectrometer. Percentage tumor cell kill was calculated as follows :

% Tumor kill =

CPM in well with immune spleen or lymph node

CPM in well with non-immune spleen or lymph node

1- x 100%.

This method of calculation negates any cytotoxic activity exhibited by non-immune spleen cells after 24 and 48 h incubation.

Serum cytotoxic antibody assay

Tumor cells were labelled with W r as described above. Serial 0.1 ml dilutions of sera were made

GENETICS OF RESPONSE TO TUMOR IN RATS 99

in MEM-S in microtiter plates (u bottom) and 50y1 containing 1 x lo5 W r labelled tumor cells were added. Plates were incubated for 30min at 37" C and then 50 pl guinea pig serum diluted 1/3 was added. No prior absorptions of the guinea- pig serum were necessary, since it was not cytotoxic to the tumor cells. Incubation was continued for 60 min at 37" C. Cells were then resuspended and spun at 1,OOOg for 5 min. One-tenth ml of super- natant was counted in an automatic gamma counter. A standard reference serum obtained from BN rats 13 days after tumor inoculation was included in each assay. At a 1/8 dilution, this serum caused release of 75 to 81 % of the releasable counts obtained with 10% NP,,. Results are expressed as the dilution releasing 50% of the counts that were released by the reference serum diluted 1/8 (50% titers).

Virus inoculation MSV (Lot No. MSV-B-123) obtained from tumors

of BALB/c mice was generously supplied by Dr. Jack Gruber, National Cancer Institure, Bethesda, Maryland. Virus was diluted in MEM and injected subcutaneously into NB, Le, and LBN rats 24 h after birth. Animals were inspected daily for 32 days and any tumors that appeared were examined histo- logically.

Immunodiffusion and radioimmunoassay for viral p30 Double immunodiffusion (Ouchterlony) tests for

p30 were made in 0.75% agar in borate buffer PH 8.6. Sera (1Oy1) were tested undiluted against a standard mouse p30 extract containing 2 y g p30 per well. A standard rabbit anti-mouse p30 was included in each test (Jones et al., 1974).

FIGURE 1 Rate of tumor growth in rat

strains. Groups of 7 to 11 (A) or 3-7 (B) rats were injected with 5 x lo7 BM2 cells SC. All points represent mean of tumor diameters in all rats, none of which died during the 20-day observation period. Vertical bars depict standard errors. By Student's t-tests, the strains that differed significantly from BN (P<O.Ol) at three or more time intervals were Le, (Lex LBN)1/3, (Le x LBN)l/'; those differing from Le at three or more time intervals were LBN, (BN x LBN)3/3 and BN.

2 4 6 8 10 12 14 16 18 2 0 Days

100 VEIT ET AL.

For co-precipitation radioimmunoassays (Jones and Feldman, 1975b), a standard antibody curve was generated. One-tenth to 50yg of a y globulin fraction of hyperimmune BN anti-MST (total protein concentration adjusted to 50 pg with normal BN 71 globulin) was incubated with 0.1 to 0.2 ng (10,000-20,000 cpm) 125L-p30 at room temperature for 30min. Rat immunoglobulin with bound p30 was then precipitated by adding 2.5 mg (excess) goat anti-rat y globulin. Precipitates were pelleted at 1,000g and washed twice with MEM. Values were corrected for amount of non-specific trapping by 50pg normal BN y globulin (1-5%). Routinely, the amount of 1251-p30 specifically precipitated by 0.1 yg BN anti-MST was approximately 10% and the amount precipitated by 50 yg was 80%. The relative amount of antibody in a standard quantity of y globulin from a test serum could then be determined by comparing its co-precipitating activity with the standard antibody curve. The purified 1251-labelled p30 used in these tests was kindly provided by Dr. Steve Kennel and Ms. Pat Mc- Conahey.

RESULTS

Tumor growth and serum cytotoxic antibody activity in parental, Fl and backcross rats

Groups of from three to 11 BN, Le, LBN, (BN x LBN)ll3 (AgB1/3), (BN x LBN)3/3 (AgB3/3), (Le x LBN)1/3 (AgB1/3), and (Le x LBN)'/* (AgB1/1) rats were injected with 5 x lo7 BM2 tumor cells SC and tumor growth was followed for 20 days post- inoculation; all rats survived for this length of time. Figure 1~ shows tumor growth in BN, Le, and LBN rats. Tumors grew at similar rates in all three groups up to day 8 (tumor growth in Le recipients appeared to have levelled off at this time). After day 8, tumors in Le hosts were rapidly rejected and were undetectable by day 16. Tumors in BN rats grew more progressively than in the other groups, and in other studies were found to cause death for most BN animals by 50 days. Growth rates in LBN rats were somewhat intermediate between those of Le and BN recipients. Figure 1~ shows that, of the four backcross groups tested, tumor growth in (BN x LBN)3/3 was most progressive and similar in time-course to that of BN parents. By contrast,

15- A

10- - 7 0

7T 5- - VI L FIGURE 2

Cytotoxic antibody response of I- rat strains described in Figure 1 . u Each point represents the mean x 50% cytotoxic titers of 7 to 11 (A)

or 3 to 7 (B) sera f standard errors. Strains differing significantly from BN (p<O.Ol) at three or more time intervals were Le, (Le x LBN)'i3, s (Lex L B N ) * / ' ~ ~ ~ ( B N X L B N ) / ~ / ~ ; strains differing from Le were BN, LBN, (BNxLBN)lIS and (BNx LBN)3/3.

W c .- .- 0 c

Backcrosses 0 rn

Days

GENETICS OF RESPONSE TO TUMOR I N RATS 101

TABLE I

ANTI-p30 ACTIVITY IN SERA OF TUMOR-BEARING BN, Le, LBN HYBRIDS AND F,-TO-PARENT BACKCROSS RATS

pg equivalent anti-p30 antibody * Day I Day 15 Day 25 Day 31

Host AgB phenotype

BN 313 0 0.6 .t0.2 3.5&0.8 5.3k 1.7

LBN 113 0 0.5h0.1 2.4k0.4 7.531.9

BN x LBN 313 0 0.410.1 2.5k0.3 5.7h1.1 BN x LBN 1!3 0 0.5h0.1 2.8 h0.7 5.2f 1.7

Le x LBN 113 0 0.310.1 1.0*0.2 2.9k1.5 Le x LBN 111 0 0.550.2 0.8h0.2 1.6k0.1

0.410.2 Le 111 0 0.1 *0.05 0.4h0.1

' Determined by binding assays with whole peripheral blood cells as described in " Methods ". - * 50 pg y globulin was tested and antibody equivalent to the indicated amount of hyperimmune BN anti-MST ( ~ S E ) was detected.

tumor growth in (Le x LBN)'13 and (Le x LBN)'I1 rats was less progressive and quite similar to that of Le parents. Tumor growth in (BNXLBN)'/~ resembled that in LBN rats.

Cytotoxic antibody activity was measured in the sera of the above groups of tumor recipients. Figure 2 shows that peak 50% cytotoxic titers occurred at day 13 for most groups. At the time of peak activity, BN, LBN, (BNxLBN)ll3 and (BN x LBN)3/3 had three to six times as much cytotoxic antibody activity as did Le, (Le x LBN)l13 or (Le x LBN)'I1. By day 20, activity in all groups had declined to 50% titers of 375 or less with BN, LBN and BN backcross titers remaining significantly elevated over those of Le and Le backcrosses.

Anti-p30 response of parental, Fl and backcross rats The sera of parent, F, and backcross recipients

of MST tumor were tested for their response to the major viral core antigen, p30. Table I shows pg equivalent anti-p30 activity in the sera of the various groups on days 7, 15, 25 and 31 post tumor inocu- lation. Anti-p30 antibodies were not detected until day 15. At that time, BN and LBN sera had signifi- cantly higher quantities of anti-p30 than did Le sera. This relationship was also observed on days 25 and 31 when anti-p30 activity was increasing in BN and LBN sera but not in Le sera. Sera of BN x LBN backcrosses contained anti-p30 antibody levels that were comparable to those of BN parents whereas sera of Le x LBN backcrosses contained considerably lower levels of anti-p30 that were comparable to or slightly elevated over those of Le parents.

Since tumors grew to a larger size and remained larger for a longer period of time in BN recipients than in Le recipients (Fig. l), the possibility existed that increased antigenic load in BN rats accounted for their greater response to p30 as compared to Le rats. Therefore. an exDeriment was designed

so that anti-p30 activity could be measured under conditions in which tumor growth was greater in Le recipients than in BN recipients. Groups of four Le and BN rats were injected SC with 1 x lo8 and 5 x los MST cells, respectively. Tumor growth and the number of animals responding to p30 were determined. Le rats injected with 1 x lo8 tumor cells developed tumors that were larger (25 mm YS 15 mm maximum diameter) than those produced in BN rats injected with 5 x lo6 cells. Despite the increased tumor mass in Le recipients, none of these animals developed anti-p30 antibodies that could be detected by Ouchterlony analysis. Response to p30 was detected in the sera of all BN rats IS days after tumor inoculation.

Oncogenicity of MSV in parental and Fl rats Groups of newborn BN, Le and LBN rats were

injected SC with lo3, lo4, or lo5 focus-forming units (FFU) of MSV. Development of tumors in these three groups of rats is shown in Table 11. Le rats developed tumors only when given lo5 FFU and at this dose only 2/10 of the animals were positive. BN rats appeared to be more susceptible to tumor formation since all developed tumors when given lo5 FFU. Tumors also developed in some of the BN

TABLE I1

ONCOGENICITY OF MOLONEY SARCOMA VIRUS I N RATS

Virus dose (FFU) and reaction * Strain

106FFU 104FFU 108FFU Total

BN 717 219 116 10122 LBN 3/3 116 013 4/12 Le 2/10 0/10 019 2/29

Focus-forming units; given SC to rats 24h after birth. - - Number of rats developing tumorslnumber injected.

102 VEIT ET AL.

TABLE 111

CMC ACTIVITY IN TUMOR-BEARING BN, Le, LBN HYBRID AND Fl-TO-PARENT BACKCROSS RATS

% specific "Cr releasefsE a

Day 6 Day 8 Day 10 Host ' Phenotype Organ

BN LBN Le

BN x LBN BN x LBN Le x LBN Le x LBN

BN LBN Le

BN X LBN BN x LBN Le x LBN Le x LBN

Spleen Spleen

313

1/1 Spleen

Spleen Spleen

313

1 /3 Spleen 111 Spleen

313 Lymph node 113 Lymph node l / l Lymph node

313 Lymph node 113 Lymph node 113 Lymph node 1/1 Lymph node

113

113

0 121.0.6 19410.5

0 0

11 1 0 . 6 9 1 0

14h0.6 19f1.0 43&2.1

121-0.4 1610.6 39h1.2 4 2 k 1.6

17+0.6 2711.0 321.1.2

24f0.8 2211.0 51 k2 .6 401.1.7

231 1.0 5311.2 641-3.1

24k1.0 1811.0 4411.7 501.1.3

6+0.3 34k1.6 51 1 2 . 3

I3 1 0 . 6 2010.8 5412.7 54k1.9

2050.6 3451.5 26+0.9

21 50 .9 610.6

51 2.2.0 2510.8

Groups of three rats given 5 x 10' BM2 SC; equal numbers of cells from each of three rats were pooled and tested in triplicate, 5 x 1 0 4 effector cells: 5 x lo4 target cells. - *Percentage specific "Cr release rt SEM of triplicate determinations; effector cells incubated with BM2 target cells at a ratio of 1OO:l for 16 h.

rats given lo4 or lo3 FFU. LBN recipients appeared to be as susceptible as BN rats to tumor formation. All tumors that developed occurred at the site of injection, appearing 10-20 days after injection of virus.

CMC in spleens and draining lymph nodes of parental, Fl and backcross rats

Results of CMC measured by a 51Cr release assay after injection of BM2 cells are shown in Table 111. CMC in spleens of BN, (BNxLBN)l13 an ( B N X L B N ) ~ ~ ~ reached peaks of 17 to 24% on day 8. By contrast, peak CMC activity of spleens from Le, LBN, (LexLBN)'l3 and (LexLBN)'l1 occurred on day 10 and ranged from 34 to 54%.

BN, (BN x LBN)1/3 and (BN x LBN)3/3 lymph- node cells produced peak levels of CMC activity that ranged from only 18% to 24%, whereas those of Le, LBN, (LexLBN)'I3 and (LexLBN)'l1 were considerably higher (51% to 64%). There was a correlation between the magnitude of CMC responses and the nature of the F,-to-parent back- cross, since progeny of BN x LBN rats gave low CMC responses while progeny of Le x LBN crosses gave high CMC responses. The tumor was histo- compatible with respect to AgB antigens in all of the above backcrosses except (Le x LBN)l/l.

CMC in animals given skin grafts Since (Le x LBN)ll3 backcrosses were AgB com-

patible with BM2, it was of interest to determine

TABLE IV CMC IN Le RATS AND PROGENY OF Le x LBN BACKCROSSES GIVEN BN SKIN ALLOGRAFTS

%specific &'Cr release+sE a

Day 6 Day 7 Day 8 Day 10 AgB Organ phenotype Host

Le 111 Spleen 3911.1 54f2.0 46*2.3 2511.1 Le x LBN 1/1 Spleen 18h0.9 62h3.1 441.2.1 1 110.4 Le x LBN 113 Spleen 0 250.1 15k0.4 11 1 0 . 3

Le l / l Lymph node 59&2.2 68h2.9 43 3: 1.9 21h1.6 Le x LBN 111 Lymph node 45 12 .4 70h3.6 41 k2 .2 1911.1 Le x LBN 113 Lymph node 1 kO.1 9&0.3 I1 k0 .5 9k0.4

Groups of four rats were grafted with BN skin; grafts were rejected in all Le and (Le x LBN)'/' recipients between days 6 and 8 after Percentage 61Cr released after 10 h graft placement; (Le x LBN)'/* grafts were intact and healthy in three of four recipients on day 10. -

incubation of 5 x LOa effector cells and 5 x lo4 BN PEC target cells. ~ S E M of triplicate determinations.

GENETICS OF RESPONSE TO TUMOR IN RATS 103

TABLE V

CMC IN BN AND Le RATS GIVEN RECIPROCAL SKIN GRAFTS ' ~ _ _ _ _ _ _ ~ _____

% specific O'Cr releasefsE a Host Donor Organ

Day 6 Day 7 Day 8

Le BN Spleen 3451.1 54*1.6 39k1.0 BN Le Spleen 6k0.2 15k0.6 2k0.1 Le BN Lymph node 66h2.1 46h1.8 27 :k 1 .O BN Le Lymph node 9*0.4 710 .3 2k0 .1

Groups of eight to 10 rats were studied for graft survival; mean graft survival for BN recipients with Le grafts was 6.7f0.8 days; that for Le recipients with BN grafts was 7.4f0.6 days. - Groups of four rats were tested for CMC; pooled equal numbers of Le effector cells were incubated with BN PEC target cells ( 5 x 103 and BN effector cells (5 x loa), with Le PEC target cells (5 x lo4). - a Percentage "Cr released after a 10 h incubation period &EM of triplicate determinations.

if the high CMC responses in these rats might be directed against non-AgB histocompatibility anti- gens. Groups of Le, (Le x LBN)l" and (Le x LBN)1(3 rats were grafted with BN skin and the results of testing for CMC against normal BN PEC targets on days 6 to 10 after grafting are shown in Table IV. Peak activity in spleens and draining lymph nodes of Le and (Le x LBN)lI1 was 54 to 70%. Activity of spleen and lymph-node cells of (Le x LBN)lI3 rats were 15 and 11 %, respectively. The low response of the latter group contrasts with the high response observed with animals immunized with BM2 and tested with tumor targets (Table 111).

CMC in BN and Le rats given reciprocal skin grafts Groups of four BN rats were grafted with Le skin

and Le rats were grafted with BN skin. CMC activity was assayed in spleens and draining lymph nodes 6, 7 and 8 days after graft placement. Mean graft survival for BN recipients with Le skin was

6.7*0.8 days and that for Le rats with BN skin was 7.4f0.6 days. As seen in Table V, Le recipients responded to BN skin grafts much more vigorously as measured by CMC than did BN recipients to Le skin.

Microcytotoxicity of spleen and lymph-node cells from parental, Fl and backcross rats

Since the 61Cr release assay for CMC measures primarily T cell killer activity (Veit and Feldman, 1975; Cerottini and Brunner, 1974), it was of interest to utilize another type of cytotoxicity assay. Groups of four rats were inoculated with 2x10' MST, and 10 days later spleen and lymph-node cells from these and from untreated rats were tested individually in triplicate. As shown in Table VI, cells from BN tumor recipients showed much lower levels of CMC activity at 24 and 48 h than all other combinations tested. CMC of draining lymph-node cells was significantly less than that of spleen cells.

TABLE VI

MICROCYTOTOXICITY OF BN, Le, LBN AND LBN TO PARENT BACKCROSSES WHEN INCUBATED WITH la61-MST

% t W o r Cell killfSE AgB

Host Phenotype 24 h incubation 48 h incubation

Spleen Lymph node Spleen Lymph node

BN 313 32&6 6 4 2 7+5 8&3 BN x LBN 313 73 h 7 4 2 ~ 2 7 2 5 6 14&6 BN x LBN 113 81 I t4 l0hl 80&2 13&7 LBN 113 79&3 10*3 83&2 1 5 h 1 Le x LBN 1 /3 9 5 h 2 7*3 93 +4 12&4 Le x LBN 1 I1 83 k 4 10+2 84510 I l k 1 Le 111 64&13 14&3 8 4 5 5 17&2

CPM in well with immune spleen or L node % tumor cell = 1 - [ CPM in well with non-immune spleen or node] x 100%. - a Groups of four rats were given 2 X 10' MST

SC 10 days earlier, except in the case of LBN where three rats were tested. Cells from immune and non-immune rats were tested individually in triplicate; effector: target ratio = 50:l.

104 VEIT ET AL.

DISCUSSION

Results of our studies of the cellular and humoral immune responses of rats to Moloney sarcomas suggest a genetic orientation for the intensity of these responses but non-linkage to the AgB histo- compatibility locus. The level of T-cell CMC, as measured by 51Cr release, appeared to correlate inversely with development of serum anti-tumor antibodies in all groups with the exception of LBN hybrids which exhibited high responses in both assays. Peak CMC activity in BN, (BNxLBN)'/~ and ( B N x L B N ) ~ ~ ~ rats was quite low when com- pared to that of Le, (Le x LBN)'13 and (Le x LBN)l/l rats. However, peak levels of serum cytotoxic antibodies or anti-p30 antibodies were high in the former group as compared to those of the latter group. Tumor size (antigen dose) apparently did not correlate with CMC activity of different groups since at the time of peak CMC activity, tumors in all groups were approximately the same size. We have observed a correlation between antibody responses to p30 and susceptibility to tumor growth (Jones and Feldman, 1975u), but anti-p30 antibody levels were not determined solely by tumor size since increasing the tumor-cell inoculum given to Le recipients did not alter their low-level response to p30. Other factors, such as virus replication in the host, may also influence responses to p30 and studies of these factors are under way.

The above findings suggest that cellular and humoral responses to tumor antigens may be regulated by different genetic influences (Kolsch, 1975) and they could be directed at different antigenic determinants. It is possible that each of these responses is under the genetic control of Ir genes (McDevitt and Benacerraf, 1969) not linked to the AgB histocompatibility locus. Since the low- respander or high-responder status of cellular or antibody rzsponses was not transmitted to backcross rats according to AgB phenotype, we concluded that AgB-linked gene products did not dictate this status. An alternative explanation is that cellular immunity in this system, whether linked or not linked to AgB, is not under genetic control that segregates among the strains examined, but that only antibody formation is regulated by a hypo- thetical non-AgB Ir gene product. In this case, high responders (BN) might produce blocking antibodies whose activity would lead to suppressed cellular immunity. Low responders (Le) might not develop blocking antibodies and, therefore, would exhibit high levels of cellular immunity. At the time of peak CMC, circulating antibody levels in all strains were low or undetectable, but it was not known how much antibody was bound to tumor components or other cellular elements. Studies of the antibody response to sheep erythrocytes in Le

and BN hosts indicate that both strains respond equally well (unpublished information). This suggests that the antibody response to tumor-associated antigens may be under a genetic control different from that which regulates the response to hetero- logous erythrocytes.

Newborn Le rats were found to be much more resistant to tumor development after MSV injection than were BN or LBN rats. Whether susceptibility to oncogenesis as seen in the present study was related to differences in immune responsiveness was not determined. Colombatti et at. (1975) have recently reported that susceptibility of mice to MSV oncogenesis was under genetic control that was not linked to H-2. They suggested that these regulatory genes, despite their location outside of the IX linkage group, may still be Ir genes.

The possibility remains that genetic control of F cell responses to tumor antigens may be revealed only by in vitro assays. Recently, it has been demon- strated (Nielsen and Koch, 1975) that BN T cells respond less well in various in vitro tests than those of other rat strains. In vim, however, BN rats were as capable as other strains of rejecting third- party grafts and inducing graft-versus-host reactions in F, hybrids. In addition, studies with congenic rat strains suggested that AgB-linked loci did not determine in vifro responsiveness (Nielsen and Koch, 1975). In our studies, BN rats rejected Le skin grafts as rapidly as Le rats rejected BN skin grafts, despite the large difference in CMC activity of the two strains as measured in vitro. In addition, some influence of AgB on the rate of tumor growth in uivo was observed since animals carrying the AgBl allele exhibited higher regression rates than animals typing as AgB 3/3 (Fig. 1) .

The finding that (Le x LBN)lI3 backcross rats rejected tumors as quickly as did Le recipients and likewise developed high levels of CMC suggested that their response was directed against either non- AgB histocompatibility antigens or tumor-associated antigens. The possibility that Le rats may develop a more vigorous response to tumor-associated antigens than do BN recipients has been previously discussed (Jones and Feldman, 1975~) . The present report provides additional evidence in support of this concept. Since the tumor was syngeneic in (Le x LBN)'13 with respect to AgB histocom- patibility antigens, high CMC activity could only be directed against non-AgB histocompatibility antigens or tumor-associated antigens. It was found that (Le x LBN)lI3 rats developed relatively low levels of CMC when stimulated with BN skin grafts which expressed both AgB and non-AgB antigens. This finding indicated that high levels of CMC activity in tumor-bzaring (Le x LBN)'13 rats were probably directed against tumor-associated antigens.

GENETICS OF RESPONSE TO TUMOR I N RATS 105

Since the level of CMC activity in (Le x LBN)’I3 and Le rats was comparable, it is likely that Le recipients also rzsponded to tumor-associated antigens in addition to any histocompatibility antigens that were exprsssed on the tumor cells. The possibility that the histocompatibility antigens of tumor cells wzrz more effective immunogens than skin grafts was not supported by our studies (Jones and Feldman, 1975a; Veit and Feldman, 1975).

In addition to the 51Cr releise assay, we utilized a microcytotoxicity-type assay to examine the cell- mediated immune response in parents, F, hybrids and F,-to-parent backcrosses. Although the responses to tumor in BN recipients were much lower than in Le recipients, the backcross rats all responded at a level comparable to that of Le rats. This was in marked contrast to results obtained with the 51Cr release assay in which BN x LBN backcross responses were comparable to those of BN parents and Le x LBN backcross responses were comparable to those of Le parents. Furthermore, the responses in draining lymph nodes as measured by micro- cytotoxicity were very low. By 51Cr release, draining lymph-node cells, when compared to spleen cells, produced equally high or higher levels of CMC. Lymphoid cells from tumor-bearing rats were assayed by both methods at the time of peak response; therefore, time of assay would not explain the differences. Microcyfotoxicity assays have been

shown to detect activity of cell types other than, and in addition to, T cells (Lamon et al., 1972). Indeed, Evans and Alexander (1972) showed that macrophages from peritoneal exudates of tumor- bearing mice were cytotoxic when measured in a growth inhibition assay. Our results suggest that two different cytotoxic cell populations were dis- tinguished by the use of 51Cr release and micro- cytotoxicity assays and that T-cell cytotoxicity, as measured by 51Cr release, appears to be regulated by genetic factors not associated with the response detected by microcytotoxicity. The possibility that macrophage-mediated cytotoxicity is detected in our microcytotoxicity assay is currently under investigation.

ACKNOWLEDGEMENTS

The authors acknowledge the skilled technical assistance of Mrs. Jill Patch, Ms. Pat Sharp and Ms. Valentina Russack.

This is Publication No. 11 32 from the Department of Immunopathology, Scripps Clinic and Research Foundation, La Jolla, California 92037. The work was supported by United States Public Health Service Contract NO1-CB-43874, ERDA Contract AT(04-3)-779, USPHS Grant AI-07007, USPHS Fellowship Award 1 F22CA01028-02 and USPHS Training Grant 5TOlGM00683-15.

ASSOCIATION GCNCTIQUE DES RCPONSES IMMUNITAIRES HUMORALES ET CELLULAIRES DES RATS A U X SARCOMES DE MOLONEY

Des rats Brown Norway (BN), Lewis (Le), des hybrides F, de Lex B N (LBN) e t des rats r6trocroises (parents/LBN) ont Cte utilises dans une etude des r6ponses cellulaires e t humorales P un sarcome de Moloney des rats BN. Quel que soit le ph6notype AgB, les rats r6trocroisCs B N x LBN ont produit une faible cytotoxicit6 P mediation cellulaire (CMC) qui Ctait comparable P celle que I’on avait observ6e chez les parents BN. Chez les rats retrocroises LexLBN, la CMC etait forte, comme chez les parents Le. On a constate qu’il existe un rapport inverse entre le niveau de la CMC et celui des anticorps seriques (cyto- toxiques pour les cellules tumorales e t anti-p30 du MuLV); les t i tres d’anticorps seriques Ctaient eleves chez les parents B N e t les B N x LBN etfaibles chez les parents Le e t les Lex LBN. Les hybrides LBN avaient des niveaux relativement 61ev6s de CMC et d’anticorps s6riques. On a Bgalement constate que chez les parents Le e t les Lex LBN, la r6ponse CMC 6tait dirig6e contre les antighnes associes a l a tumeur plutBt que contre les antigenes d’histocompatibilite exprimes sur les cellules tumorales. Ces resultats conduisent P penser que les reponses humorales e t cellulaires au sarcome de Moloney chez le ra t ne sont pas unique- ment dCtermin6es par le principal locus d’histocompatibilite AgB, mais qu’il existe une association gene- tique. Cette association a 6th dCtectee lors d’une epreuve de liberation du C P , qui permet de mesurer I’activite des cellules T; il semble donc que certaines populations de cellules T effectrices puissent &re g6nCtiquement contrbl6es.

REFERENCES

ARMERDING, D., KATZ, D. H., and BENACERRAP, B., Immune CHESEBRO, B., WEHRLY, K. , and STIMPFLING, J., Host genetic response genes in inbred rats. 11. Segregation studies of the control of recovery from Friend leukemia virus induced GT and GA genes and their linkage to the major histocom- splenomegaly. Mapping of a gene within the major histo- patibility locus. Immunogenetics, 1, 340-351 (1974). compatibilty complex. J. exp. Med. , 140, 1457-1467 (1974).

COLOMBATTI, A., COLLAVO, D., BIAS, G., and CHIECO- CEROTTINI, J -C , and BRUNNER, K. T., Cell mediated cyto- BIANCHI, L., Genetic control of oncogenesis by murine toxicity, allograft rejection, and tumor immunity. Advanc. sarcoma virus Moloney pseudotype. I. Genetics of resistance Immunol., 18, 67-132 (1974). in AKR mice. In?. J . Cancer, 16, 427-434 (1975).

106 VEIT ET AL.

EVANS, R., and ALEXANDER, P., Role of macrophages in tumor immunity. I. Cooperation between macrophages and lymphoid cells in syngeneic tumor immunity. Immunology,

JONES, J. M., and FELDMAN, J. D., Alloantigen expression of a rat Moloney sarcoma. J. nut. Cancer Inst., 55, 995-999 (1 975a).

23, 615-626 (1972).

JONES, J. M., and FELDMAN, J. D., Immunogenic and anti- genic properties of a rat Moloney sarcoma. Behring Inst. Mitt., 56, 14-18 (1975b). JONES, J. M.. JENSEN. F.. VEIT, B., and FELDMAN, J. D. . In vivo growth. and antigenic properties of a rat sarcoma induced by Moloney sarcoma virus. J. nut Cancer Inst., 52, 1771-1777 (1974). K~LSCH, E., The genetic control of T-cell mediated immunity against the DBA/2 mastocytoma P-815. 11. Low respon- siveness in T-cell mediated cytotoxicity accompanied by the inability to produce antibodies in a secondary response. Europ. J. Immunol., 5, 527-532 (1975). LAMON, E. W., SKURZAK, H. M., KLEIN, E., and WIGZELL, H., In vitro cytotoxicity by a nonthymus processed lymphocyte population with specificity for a virally determined tumor cell surface antigen. J. exp. Med., 136, 1072-1079 (1972). LILLY, F., BOYSE, E. A., and OLD, L. J., Genetic basis of susceptibility to viral leukemogenesis. Lancet, ii, 1207-1209 (1964). LILLY, F., and PINCUS, T., Genetic control of murine viral leukemogenesis. Advanc. Cancer Res., 17, 231-277 (1973). MCDEVITT, H. O.,, and BENACERRAF, B., Genetic control of specific immune responses. Advanc. Immunol., 11, 31-74 (1969).

NIELSEN, H. E., and KOCH, C., Genetic control of the in vitro responses of rat blood lymphocytes. I. Comparison of in vitro and in vivo responses. Scand. J. Immunol., 4, 31-36 (1975).

OLDSTONE, M. B. A., DIXON, F. J., MITCHELL, G. F., and MCDEVITT, H. O., Histocompatibility-linked genetic control of disease susceptibility. Murine lymphocytic choriomenin- gitis virus infection. J . exp. Med., 137, 1201-1212 (1973).

PALM, J., " Ontogeny " of the major histocompatibility locus in rats-a problem in nomenclature. Transplantation, 9, 161- 163 (1970).

SATO, H., BOYSE, E. A., AOKI, T., IRITANI, C., and OLD, L. J., Leukemia-associated transplantation antigens related to murine leukemia virus. The X-1 system: Immune response controlled by a locus linked to H-2. J. exp. Med., 138, 593- 606 (1973). SEEGER, R. C., RAYNER, S. A., and OWEN, J. J. T., An analysis of variables affecting the measurement of tumor immunity in vitro with 1a61-iododeoxyuridine-labelled target cells. Studies of immunity to primary Moloney sarcomas. Int. J . Cancer, 13, 697-713 (1974). SILVERS, W. K. , and BILLINGHAM, R. E., Contributions of the rat to the immunobiology of tissue transplantation. Transplant. Proc., 11, 152-161 (1970). TENNANT, J., and SNELL, G., The H-2 locus and viral leukemogenesis as studied in congenic strains of mice. J. nat. Cancer Inst., 41, 597-604 (1968). VEIT, B. C., and FELDMAN, J. D., Cell mediated cytotoxicity to Moloney sarcoma in syngeneic and allogeneic rats. Int. J. Cancer, 15, 367-376 (1975).