subpopulations of multiparous rat lymph-node cells cytotoxic for rat tumour cells and capable of...

Int. J. Cancer: 15, 762-772 (1975)

SUBPOPULATIONS OF MULTIPAROUS RAT LY MPH-NODE CELLS CYTOTOXIC FOR RAT TUMOUR CELLS

AND CAPABLE OF SUPPRESSING CYTOTOXICITY I N VZTRO

R. C . REES, Judith BKAY, R . A. ROBINS and R. W. BALDWIN Cancer Research Campaign Laboratories, The University of Nottingham, University Park,

Nottingham NG7 2RD, England

Lymph-node cells ( L N C ) from multiparous pregnant rats were separated on columns prepared from nylon wool, and tested for cytotoxicity against target tumour cells. Reactivity of L N C towards hepatoma 0 2 3 and mammary carcinoma AAF.57 was demonstrated in cell populations retained on the nylon wool, and not with cells eluted from the column. Although only 25% of the samples of unfractionated L N C were cytotoxic for tumour cells, retained cell fractions were cytotoxic in I 1 out of 12 tests ( p = <0.05). Similarly retained L N C were also cytotoxic for 15-day-old embryo cells but not for normal adult rat fibroblasts. Using multiparous rat serum it was shown that the reactivity of the retained L N C population could be abrogated in eight out of 11 tests ( p <0.05). The L N C population recovered from the nylon wool constituted 28 to 3.5 of the original L N C preparation, and consisted of 60-70 '%, Ig- bearing cells together with a subpopulation of cells responding to soluble PHA. Separation of multiparous L N C on glass beads coated with rat Ig and then rabbit anti-rat Ig ( in excess) also demonstrated the retained cell population to be cytotoxic against tumour cells. Approximately 17-20 % of the original cell population was recovered froin cells retained on the column, and consisted of an enriched Ig-bearing cell population (6.5- 80% Ig-bearing cells) and L N C responsive to PHA. Carbonyl iron treatment oJ' multiparous rat L NC was found to remove detectable cytotoxicity from multiparous rat L NC preparations. The cytotoxicity of multiparous rat L N C retained on nylon wool was also abolished following incubation with carbonyl iron. Definite conclusions as to the nature of the effector cell cannot be drawn from this test, since carbonyl iron treatment was found to remove not only phagocytic cells from L N C preparations but also a proportion of other cell populations including Ig-bearing lymphocytes. In addition to detecting a cytotoxic L N C population reactive towards tumour-associated embryonic antigens (retained fraction from nylon-wool column separation), a subpopulation o j multiparous rat L N C was demonstrated in cell fractions eluted from the nylon wool which was shown to suppress the cytotoxicity of the retained multiparous L N C population. The exact nature of this subpopulation of L N C and the mechanism of action is at present not known.

Tumours induced by 4-dimethylaminoazo- Barker, 1967a, 19676; Baldwin et at., 1971, benzene and 3-methylcholanthrene express 19736; Baldwin and Embleton, 1971 ). In addition tumour-specific cell surface antigens, which re-expressed, cross-reacting embryonic antigens, can be demonstrated using in vivo tumour which are distinct from the tumour-specific rejection experiments and in vitro lymphocyto- component, are commonly detected on toxicity and serum antibody tests (Baldwin and chemically-induced and spontaneously arising

Received : January 21, 1974.

762

SUBPOPULATIONS OF CYTOTOXIC CELLS

neoplasms (Baldwin et al., 1974a; Baldwin and Embleton, 1974). Although some workers have shown immunization with embryonic tissue to be effective in eliciting a tumour-rejection response (Le Mevel and Wells, 1973; Coggin et al., 1970), experiments in this laboratory indicate that multiparous or embryo-immunized rats do not consistently reject tumours known to express embryonic antigens at their cell surface (Baldwin et al., 1974b; Shah et al., in preparation). Embryo-sensitized rats do, however, possess lymph-node cells capable of reacting with tumour cells in vitro. Thus, using microcytotoxicity tests, lymph-node cells prepared from multiparous rats and from rats immunized with irradiated (5,000 R) embryo cells have been shown to be cytotoxic for plated tumour target cells (Rees et al., 1974; Shah et al., in preparation).

Using multiparous pregnant rats as a source of sensitized lymph-node cells, and cell fractionation techniques to separate subpopulations of cells, the effector cell mechanism of in vitro cytotoxicity was investigated. In these studies subpopulations of lymph-node cells were shown with both cytotoxic and cytotoxic-inhibitory activity for several tumour-cell lines.

MATERIAL AND METHODS

Rats and tumours

Hepatomas were induced in syngeneic Wistar rats by oral administration of 4-dimethyl- aminoazobenzene (Baldwin and Barker, 1967a), sarcomas were induced by subcutaneous injec- tion of 3-methylcholanthrene and a mammary carcinoma (AAF57) by oral administration of 0.4% of the carcinogen in the diet (Baldwin and Embleton, 1969). In vivo tumour lines were maintained by serial transplantation of the tumour tissue into rats of the same sex as the original tumour donor.

Cell cultures

Cell cultures were derived from in vivo trans- planted tumour lines or normal adult rat lungs by trypsinization of tissue with 0.25 % trypsin (Difco 1:250), and cultured in medical flats in Eagle's MEM with penicillin (100 IU/ml) and streptomycin (200 ,ug/ml) and 10% calf serum. Embryo cells derived from 14- to 15-day-

old rat embryos were prepared in a similar manner, and grown in Waymouth's tissue culture medium, supplemented with penicillin (100 IU/ml), streptomycin (200 ,ug/ml) and 20% foetal bovine serum. The age of embryos was estimated according to size and development (Witchi, 1956). When confluent, monolayers were harvested with trypsin and subcultured.

Serum and Iyrnphoid donors

Serum and lymph-node cells (LNC) were derived from multiparous pregnant rats having had three or more pregnancies or from nulli- parous female rats. Blood samples were collected by cardiac puncture and the serum was separated and stored at -20°C in 1- to 2-ml volumes. LNC were prepared from the cervical, axillary and mesenteric lymph nodes. Lymph nodes removed aseptically were finely chopped and pressed through a 120 gauge stainless steel wire mesh. The single-cell suspension was then cen- trifuged at 120xg for 5 min, washed three times with Eagle's MEM, and used in cytotoxicity tests.

Microcytotoxicity tests

LNC cytotoxicity tests were performed according to the method previously described (Baldwin et al., 1973~). Cells derived from in vitro cell cultures were plated into wells of microtest plates (Falcon No. 3040) at 100-200 cells per well in appropriate growth medium. Plates were incubated at 37" C for 24 h to allow adherence of the cells, and used in cytotoxicity assays. Growth medium was removed from the wells, and LNC prepared from normal or multiparous rats were added (4 x lo5 to 5x104 LNC in 0.2 ml volume per well) in Eagle's MEM. After 60 min, foetal calf serum was added to a final concentration of 10% (v/v). The plates were incubated for 48 h at 37" C, washed with saline, fixed in methanol and stained with 0.2% (w/v) crystal violet, and the number of surviving cells counted.

Serum blocking o j multiparous rat lymph-node cell cytotoxicity

Serum from either normal or multiparous female rats was heat-inactivated at 56" C for 30 min, diluted 1 in 3 in Eagle's MEM and added to wells in microtest plates containing target cells (0.1 ml of serum dilution per well).

763

REES ET AL.

After incubation for 45-60 min at 37" C, the serum was removed and either normal or multiparous rat LNC added to each of the eight wells treated with either normal female (control) or multiparous rat serum. The plates were incubated for 48 h at 37"C, and the number of surviving target cells determined.

Cell separation

Nylon-wool column separatiorr. LNC separations were performed according to the procedure of Julius et al. (1973). Before use in fractionation procedures, the nylon-wool (Fenwal Labora- tories, Morton Grove, Ill., USA) was washed in 0.2 N HCI, rinsed in tap water, and finally in glass double-distilled water. The columns were prepared with 0.6 g of dried nylon-wool. Each column was then washed with 25 ml of Eagle's MEM (5% foetal bovine serum) and incubated at 37" C for 60 min. Cells were loaded onto columns (5.0 x lo7 to 1 .O x 10* LNC) in a 1 ml volume and the column was then incubated at 37" C for 45 min. Cells were eluted from the column with 15 ml of medium at a constant flow rate of 1 ml per minute (eluent fraction). The nylon adherent cell population (retained fraction) was recovered by consecutive teasing of the wool in 5 ml volumes of medium (four washings). The cells from both eluent and retained fractions were washed twice with Eagle's MEM, and tested for cytotoxicity against target cells.

Rat Ig-rabbit anti-rat Ig coated glass-bead column separation. Glass beads of 200 pm diameter (Superbrite 100-5005, Minnesota Mining and Mfg. Co., St. Paul, Minn., USA) were coated with ammonium-sulphate-precipi- tated rat Ig and then with rabbit anti-rat Ig (in excess), and columns prepared according to the method previously described (Lamon et al., 1973b). Glass beads and rabbit antisera were obtained from Dr. H. Wigzell, Depart- ment of Tumor Biology, Karolinska Institute, Stockholm, Sweden.

Before loading, columns were washed with 1 x25 ml physiological saline and then with 25 ml of Eagle's MEM (10% foetal bovine serum) at 20" C. Cells (1 x108 to 2x108 LNC in 1 ml volume) were run through the column and eluted with 15 ml of medium at a constant flow rate of 1 ml per minute (eluent fraction). The beads were poured from the column and

washed four times with 5 .ml of medium (retained fraction). Both LNC fractions were then washed twice in medium and tested for reactivity against target tumour cells in microcytotoxicity tests.

Carbonyl iron treatment. Carbonyl iron (GAF, Great Britain Ltd., Wythenshawe, Manchester, England) was washed three times with Eagle's MEM (10% foetal bovine serum) immediately prior to use. LNC from normal and niulti- parous rats were incubated at 37" C for 1 h ( 3 ~ 1 0 ~ LNC per ml) together with carbonyl iron at 4 mg per ml in medium containing 10% foetal bovine serum, with continuous shaking. Carbonyl iron was then separated at 4 ° C with a magnet and the remaining LNC tested for cytotoxicity against target tumour cells.

Determination of Ig-bearing lymphocytcs

LNC preparations were tested before and after column fractionation or carbonyl iron treatment for cells bearing 1g on the surface membrane. Fluorescein-labelled rabbit anti-rat Ig (0.1 ml volume) was obtained from Wellcome Reagents Ltd., Beckenham, Kent, England, and used at a 1 in 5 dilution to stain 5 .: 10" LNC. Following subsequent washing procedures the percentage of cells staining was estimated.

Mitogenic stimulation

Cultures of 2x106 LNC (1 ml volume) in Eagle's MEM (10% foetal bovine serum) were set up in triplicate in 3 x '/2 inch glass tubes, with loose-fitting metal caps. PHA (Wellcome Reagents Ltd., Beckenham, Kent, England) was added (10 pg per tube), and tubes were incubated at 37°C for 72 h in an atmosphere of 5% C0,/95% air; then 0.5 i c Ci of Sodode- oxyuridine lZ5I (Radiochemical Centre, Amer- sham, Bucks., England) was added per tube. Uptake of IUdR was assessed after incubation at 37" C for 18 h (Toivanen and Toivanen, 1973), and the results expressed as the stimulation index.

RESULTS

Nylon-wool column separation

Normal and multiparous rat LNC preparations were separated on nylon-wool columns, and retained and eluent cell fractions tested for cytotoxicity at LNC:tumour-cell ratios of 4,OOO:l and 2,oOO:l. The results presented in

764


Table I, using LNC at 4 x lo5 cells per well, indicate that the cytotoxic activity for hepatoma D23 and mammary carcinoma AAF57 cells was retained on the nylon-wool (Exps. 1-4). Similarly retained multiparous LNC were shown to be cytotoxic for 15-day-old rat embryo fibroblasts (Table I, Exp. 5), but not adult rat fibroblasts derived from rat lung tissue (Table I, Exp. 6). In none of the experiments performed were cells eluted from the column found to be cytotoxic for tumour cells, embryo cells or adult fibroblasts. Similarly, in each experiment, cells were used at a 2,OOO:l ratio with tumour cells and results consistent with those obtained using a 4,000: 1 LNC: target-cell ratio, were obtained. The results of 12 subsequent nylon- wool fractionations are summarized in Table 11; the cytotoxicity towards tumour cells was always associated with cells retained on the column. In only one experiment presented in Table I, and in three out of 12 experiments shown in Table 11, were unfractionated LNC preparations found to be cytotoxic for tumour cells.

The cells recovered from the nylon-woal column (retained cell fraction) constituted 28-35 % of the original cell population (Table 111) and contained an enriched Ig-bearing cell population (60-70 % Ig-bearing cells). The retained fraction also contained cells capable of responding to mitogen stimulation with PHA; such activity was not associated with eluent cell populations.

Rat Zg-rabbit anti-rat Zg glass-bead column separation

Following incubation of normal and multiparous LNC on rat Ig-rabbit anti-rat Ig-coated glass beads, eluent and retained cell fractions were obtained and tested for cytotoxicity towards hepatoma D23 and sarcoma Mc57 cells. Cyto- toxicity against hepatoma D23 of 48% was observed with unfractionated cells tested at 4 x lo5 cells per well (Table IV, Exp. 1). Foltowing separation, 57% cytotoxicity was shown with multiparous LNC with the retained cell fraction; the eluent cells were not found to be cytotoxic for target tumour cells. In a subsequent

TABLE I

DETERMINATION OF THE CYTOTOXIC EFFECTOR CELL FOLLOWING SEPARATION OF MULTIPAROUS RAT LYMPH-NODE CELLS ON NYLON WOOL

Exp. No. Target cells

No. cells remainindwell % cell

reduction (meanfs~)

LNC fraction LNC Multiparous

LNC

1 Hepatoma D23

2 Hepatoma D23

Unfractionated 7 1 1 9 6 4 h 4 Eluted fraction 75f 9 87& 3 -

Retained fraction 60& 8 2 8 5 3

10 -16 53**

Unfractionated 98*12 9 5 i 1 1 3 Eluted fraction 166k17 183h23 -10 Retained fraction 80& 8 20* 7 75***

3 Mammary carcinoma AAF57 Unfractionated 3 0 1 3 26& 2 13 Eluted fraction 3 6 1 4 41k 3 -14 Retained fraction 3 4 5 5 1 1 + 2 68***

Eluted fraction 200h27 230513 -15 Retained fraction 1801 9 1 3 2 i 6 27***

Mammary carcinoma AAF57 Unfractionated 137&11 104+ 9 24* 4

5 15-day rat embryo Unfractionated 4 7 1 7 27* 2 43* Eluted fraction 5 7 f l l 47+ 7 18 Retained fraction 3 7 1 4 16+ 2 57***

6 Normal rat fibroblasts Unfractionated 1 3 1 1 1 1 * 1 I5 Eluted fraction 1 9 1 1 2 1 1 4 -11 Retained fraction 1 1 * 1 1 4 1 1 -27

LNC fraction added at 4 x lo6 cells per well (0.2 rnl volume). * * p = < O . O l ; ** p = 0.0025; *** p = <0.0005.

765

REES ET AL.

h

00 m 0

m r g CI

Y d

O I

- d N

0 0 0 -

Y m

00 d

h p' d

N r-" c r- m

O I

0

m m +

3

I

9 I

v

h

2 I 0

m o rl

CI

0 ,N

I

h m

I 0 CI

r - 2 I v m 3

I

test, significant cytotoxicity was recovered in the retained cell fraction despite there being no detectable activity in the unfractionated preparation (Table IV, Exp. 2). Using sarcoma Mc57 cells as targets, similar results were obtained; the cytotoxicity being associated with the retained cell fraction (Table IV, Exps. 3 and 4). LNC recovered from the cell populations retained on the glass-bead column represented 17-20% of the original LNC population, and consisted of 6540% Ig-bearing cells (Table 111). The retained cell fraction was also capable of stimulation by PHA.

Carbonyl iron treatment

Treatment of multiparous rat LNC with carbonyl iron was found to remove detectable cytotoxicity from the cell preparation (Table V, Exps. 1 and 2). In both experiments cytotoxicity of 55 % and 47 % respectively was demonstrated, with multiparous LNC tested at 4 x lo5 cells per well; this cytotoxicity was abolished following carbonyl iron treatment. Similar results were obtained using LNC at 2 x 1 0 cells per well when cytotoxicity was detectable in the untreated cell preparation (Table V, Exp. 2).

Retained cell fractions obtained from nylon- wool column separations were treated with carbonyl iron and tested for cytotoxicity. This treatment was found to remove detectable cytotoxicity for tumour target cells. With retained LNC at 4 x lo4 cells per well the cytotoxicity was reduced from 29 % to 4 (Table V, Exp. 3) and from 45% to --27% (Table V, Exp. 4).

With this test it is not possible to remove selectively only phagocytic cells. I n addition to removing adherent cells staining with acridine orange (3-6 % of unfractionated LNC population) approximately 50% of other .cells are also removed (Table 111). These may constitute cell subpopulations other than those with phagocytic activity which may be responsible for the cytotoxicity.

'" * .$ Serum blocking of nylon-wool-retained rellfiaction - 0 s

2 5 2 In order to demonstrate that cells retained 2 E on nylon-wool columns were specifically sen- ?:* sitized to tumour-associated embryonic antigens, " a 3 blocking tests using multiparous rat serum and - tumour target cells were performed, the results

of which are presented in Table VI. Using

P ? ? Q

5;E

766


TABLE 111

COMPOSITION AND PERCENTAGE RECOVERY OF LNC BEFORE AND AFTER COLUMN FRACTIONATION AND TREATMENT WITH CARBONYL IRON

LNC preparation from

Test Nylon-wool column fractionation Glass-bead column fractionation Carbonyl iron

Eluent Retained Eluent treatment Unfractionated Retained fraction fraction fraction fraction

FA test rabbit anti-rat Ig 20-287; 60-70% 1-3 % 65-80% 1-2% 18-20 % PHA response >20 > 20 <3 >21 <3 NT Percentage recovery - 28-35 % 44-58 % 17-20 % 50-72 % 40-60 %

CPM obtained with cells plus PHA CPM obtained with cells only

' Stimulation index =

NT 2 not tested.

'

hepatoma D23 cells and mammary carcinoma AAF57 cells, significant blocking of cytotoxicity with MP serum was observed in eight out of 12 tests (Table VI, Exps. 1-4).

Suppression of cytotoxicity using nylon-wool eluent cell fractions

Tests carried out in this laboratory have shown that not all multiparous LNC preparations are cytotoxic towards tumour cells (Rees, unpublished results), and in the present study

only 25% of all multiparous LNC preparations were found to be cytotoxic before column fractionation (Tables I, I1 and IV). However, following separation on either nylon-wool or rat Ig-rabbit anti-rat Ig glass-bead columns, cells were demonstrated in all LNC preparations with cytotoxic activity towards tumour cells (Tables I, I1 and IV). This cytotoxicity was associated with cell populations retained on the columns and not with eluent cell fractions. It was therefore thought possible that cells

TABLE IV

DETERMINATION OF THE CYTOTOXIC EFFECTOR CELL FOLLOWING SEPARATION ON RAT Ig-RABBIT ANTI-RAT Ig-COATED CLASS-BEAD COLUMNS OF MULTIPAROUS RAT LYMPH-NODE CELLS

% cell No. cells remaining/well ( m e a n 5 ~ ~ )

Normal LNC Multiparous LNC reduction Exp. No. Target cells LNC fraction

0 Hepatoma D23 Unfractionated 81k 8 42f 4 48** Eluted fraction 118* 9 145115 - 23 Retained fraction 82113 35+ 9 57*

Eluted fraction 248117 267131 -8 Retained fraction 151 f 1 2 7 0 1 9 54***

Eluted fraction 100118 91$ 7 9 Retained fraction 63+13 20+ 5 68*

4 Sarcoma Mc57 Unfractionated 721- 5 71f 9 1 Eluted fraction 100118 72110 28 Retained fraction 634~13 17% 3 73 *

2 Hepatoma D23 Unfractionated 133h13 131% 5 2

3 Sarcoma Mc57 Unfractionated 72& 5 6 0 1 5 17

LNC from unfractionated and fractionated preparations added at 4 x LOs LNC per well (0.2 ml volume). * p = < O . O l ; ** p = 10.0025; *** p = <0.0005.

767

REES ET AL.

TABLE V

DETERMINATION OF MULTLPAROUS LYMPH-NODE CELL CYTOTOXICITY TOWARDS HEPATOMA D23 CELLS FOLLOWING TREATMENT WITH CARBONYL IRON

No. cells remaining/well (mean&tS~)

Normal Multioarous reductiol' ' "(, cell LNC "N",s. LNC preparation concentration lNC 'Ieatment

LNC LNC

1 Unfract ionated 4 x 104 4x104 2x104 2x104

2 Unfractionated 4 x lo4 4 x lo4 2 x 1 0 4

2 x 104

3 Retained fraction 4 x lo4 Nylon-wool column 4 x 1 0 4

4 Retained fraction 4 x 10' Nylon-wool column 4 x 104

- Carbonyl iron

Carbonyl iron

Carbonyl iron

Carbonyl iron

Carbony1 iron

Carbonyl iron

-

-

-

-

-

1 3 0 t l 5 61 i 5 7 2 1 6 130t 6

419116 76k17

399124 117122

17 I 1 2 2 1 I

4 5 t 6 1 9 t 2

581 3 691 4 631 3

114 t 5

222 t 26 78 t 18

141 t 2 1 131 I 9

121 2 251 2

251 2 25-t 4

55** 13 12 12

47** 3

65** 12

29' 14

45* 27

' * p 7 <0.025; ** p = 10.0005.

TABLE VI

BLOCKING BY MULTLPAROUS RAT SERUM OF THE CYTOTOXICITY OF MULTIPAROUS RAT LYMPH-NODE CELLS RETAINED O N NYLON WOOL

Exp. No. Target cells

No. cells remaining/well cell '>; abrogation (mean&%)

Normal Multiparous reduction of cylotoxicity Serum

LNC LNC

I Hepatoma D23 NS I 5 8 1 2 3645 38*** MPS 6045 57L4 6016 -- 5 I oo* * MPS 6075 5 0 1 1 43 *2 14 63** MPS 6055 57*5 54+3 5 87*

MPS 7595 3311 21 k 3 36*** 45* MPS 7659 4 2 k 8 20*1 52** 21 MPS 7658 46+7 19+2 59*** I I

MPS 7574 1211 . 101-1 17 60* * MPS 7660 12+1 12&1 0 loo*** MPS 7595 1 3 1 1 13411 0 loo***

2 Hepatoma D23 NS 3 2 1 4 l l i l 66* * *

3 Hepatoma D23 NS 143 1 8 tl 43***

4 Mammary NS ' 106111 48 1 4 55*** carcinoma AAF57 MPS 7595 101 $12 65*6 36* 35

MPS 7659 128*6 8 0 j 3 38 31 MPS 7658 10218 9 8 ~ 1 0 4 93**

' NS - normal rat serum MPS multiparous rat serum.

' * p ~ 0.05; ** p - <0.01; *** p - <0.005

768


present in unfractionated multiparous LNC preparations were suppressing the cytotoxicity of cytotoxic effector LNC. Experiments were therefore designed to test for a suppressor cell population.

Table VII shows the results obtained with unfractionated multiparous LNC preparations used at different LNC: target-cell ratios. Signifi- cant cytotoxicity against hepatoma D23 cells was demonstrated only when multiparous LNC were used at a concentration of 5 x lo4 cells

TABLE VII

CYTOTOXICITY OF MULTIPAROUS RAT LYMPH-NODE CELLS TESTED AT VARIOUS LYMPH-NODE CELL:

TUMOUR-CELL RATIOS

LNC concentration per well ' 4x106 2 x 1 0 5 1x106 5x10 '

"N",9' Target cells

1 Hepatoma D23 17 - 3 -5 21* 2 Hepatoma D23 -24 18 35** N T 3 3 Sarcoma Mc7 17 -20 30* 12 4 Sarcoma Mc7 2 -10 47**60***

' LNC concentration in 0.2 ml volume added per well in microtest plates, using 100 plated target cells per well. ' * p = <0.025; ** p = 10 .005; *** p = .<0.0005. * NT = not tested.

per well (Table VII, Exp. l), and not with LNC at higher concentrations. I n three subsequent experiments, and with other tumour cells as targets, significant cytotoxicity could not be demonstrated with multiparous LNC at 4x105 or 2x105 cells per well, but was detected with a concentration of 1 x105 LNC per well (Table VII, Exps. 2, 3 and 4). In order to establish the presence of LNC capable of inhibiting cytotoxicity, normal control and multiparous LNC were fractionated on nylon- wool columns and tested at ratios of 4,OOO:l and 2,OOO:l for cytotoxicity towards tumour cells. In all experiments, results similar to those shown in Table I were obtained, showing the cytotoxicity to be present in column-retained LNC fractions. In none of the experiments performed were unfractionated LNC found to have cytotoxic activity. Normal and multiparous eluent cell fractions (2 x lo5 cells/O.l ml) were added to either normal or multiparous retained cell fractions (2x105 cells/O.l ml) in equal

volumes; 0.2 ml of the mixture was then added to each of eight wells containing tuniour target cells. The cytotoxicity of multiparous retained cells in the presence of either normal or multiparous eluent cells is shown in Table VTII. Significant abrogation of multiparous retained- cell cytotoxicity was observed in the presence of multiparous eluent cell populations (p = <:0.05). When hepatoma D23 cells were used as targets, the percentage abrogation of cytotoxicity with multiparous eluent LNC was 40% and 100% in two separate experiments (Table VIII, Exps. 1 and 2). Similar results were obtained in two further experiments using sarcoma Mc7 cells and mammary carcinoma AAF57 cells as targets. These results, together with the results presented in Table VII showing that unfractionated multiparous rat LNC are more cytotoxic when used at lower cell concentrations, indicate the presence of a cell population capable of inhibiting cytotoxicity of specifically sensitized cytotoxic eEector cells.

DISCUSSION

The nature of the effector cells from multiparous pregnant rats reactive against tumour cells in vitro has been investigated by means of various cell fractionation procedures. Using columns prepared from nylon-wool or glass beads coated with rat Ig and then rabbit anti- rat Ig (in excess), cytotoxic LNC were always recovered from cell fractions retained on the column. These cells constituted approximately 30% (nylon-wool column) and 20% (rat Ig-rabbit anti-rat Ig column) of the original LNC population, and consisted of 60 to 80 Ig-bearing lymphocytes. In addition, responsiveness to soluble PHA was associated with LNC retaind on the columns, and so suggests that the retai,hed- cell population may contain a subpopulatioh of T-cells. This is further indicated by results showing that 3 % of rat thymus cells are retained on nylon-wool columns, and that these cells also respond to soluble PHA (Bray, unpublished results); no PHA response was demonstrated in the eluent cell population. Experiments are Tat present in progress to determine whether the cytotoxicity observed with multiparous LNC is due to either Ig-bearing cells or a subpopula$ion, of retained T-cells. Using similar separation procedures, and LNC and spleen cells from

769

REES ET AL.

TABLE VIII

DEMONSTRATION OF A CELL SUBPOPULATION, ISOLATED FROM MULTIPAROUS RAT LYMPH-NODES AND NOT RETAINED ON NYLON WOOL, CAPABLE OF ABROGATING

MULTIPAROUS RAT LYMPH-NODE CELL CYTOTOXICITY

Target cells No.

No. cells remaining/well ( m e a n + s ~ ) ;, ce l l ;<abrogation

o f cyto- Normal MP retained ' toxicity

Eluent cell fraction from

retained LNC LNC

1 Hepatoma D23 Normal female rat 98&6 49 t 2 50*** -

2 Hepatoma D23 Normal female rat 4 4 5 5 241 3 45*** -

3 Sarcoma Mc7 Normal female rat 138+6 90 1 5 35*** -

4 Mammary carcinoma Normal female rat 143&7 112 t 3 22' * ' -

Multiparous rat 74*7 52 1 3 30* 40* *

Multiparous rat 36+3 361 4 0 loo**

Multiparous rat 77 t 9 8 2 t 1 0 -6 loo***

AAF57 Multiparous rat 12218 133 i 9 -9 loo***

Eluent lymph-node cells from nylon-wool columns added to multiparous rat lymph-node cells retained on nylon wool in i t I :I

* p 0.05: ** p 0.01; *** p 0.005. ratio and tested for cytotoxicity at a final concentration o f 4 x lo5 LNC per well.

multiparous pregnant BALB/c mice, i t has been shown that multiparous LNC cytotoxic towards tumour cells are eluted from Sephadex GlOO columns to which rabbit anti-mouse IgG has been bound (Hellstrom and Hellstrom, 1975). In addition, treatment with a n t i 4 antiserum and complement abolished the cytotoxicity, suggesting the effector cells to be thymus- derived. These results contrast with those of the present study, and also with those obtained following fractionation of tumour-immune rat LNC (Bray et al., manuscript in preparation). The use of nylon-wool columns to fractionate tumour-immune LNC has revealed cytotoxicity in the eluent LNC fractions as well as in the retained fraction. Further study has shown that cytotoxicity towards the embryonic component is present in the retained cell fraction, indicating that the reactivity towards the embryonic antigen may be due to a different subpopulation of LNC than those reactive against the tumour-specific component.

In syngeneic murine sarcoma virus-induced tumour systems, evidence indicative of T-cell involvement, obtained by a chromium-release assay, has been presented (Plata et al., 1973), and contrasts with results yielded by the microcytotoxicity tests demonstrating the anti-tumour

immune mechanism to be B-cell dependent (Lamon et a/., 1973~). These contradicting results may be due to differences in the tests used for assay or, as has recently been proposed, by different effector cells being present at different times during oncogenesis (Lamon et a/., 19736). The complexity of the immune response to cell- surface antigen(s) is further evidenced by work relating to the involvement of macrophages (Evans and Alexander, 1970), and the demonstration of antibody-dependent lymphocytotoxicity mechanisms (Maclennan et al., 1969). The absence of adherent cells in retained LNC fractions (as measured by glass adherence and acridine orange staining) does not rule out the possibility that precursor phagocytic cells are present, which may develop into mature macrophages during the course of the test. Evidence for macrophage maturation has recently been reported (Seeger and Owen, 1974). Thus, in the present study, the involvement of niacro- phages in cytotoxicity against tumour-associated embryonic antigens cannot be excluded.

Previous studies have established the presence in tumour-bearers of factors capable of abrogation of lymphocytotoxicity at the level of the target cell and the lymphocyte (Baldwin et a / . , 1973~). Using rat hepatoma systems, the

770


embryonic antigen component has been isolated from the tumour-cell surface as a relatively homogeneous preparation (Baldwin et al., 1974c), and shown to inhibit the cytotoxicity of multiparous rat LNC for hepatoma target cells at the level of the lymphocyte (Rees et al., 1974). In addition to serum factors, the present study shows that multiparous rat LNC include a subpopulation capable of abrogating lym- phocytotoxic reactions against the embryonic antigen component. This is shown by the fact that multiparous eluent cell fractions obtained from nylon-wool column separation, when added to the multiparous retained LNC fraction, reduce the cytotoxicity compared with

controls where normal eluent LNC preparations were added to multiparous retained-cell fractions. Whether or not suppressor mechanisms operate in vivo in preventing responses against the growing tumour has yet to be established.

ACKNOWLEDGEMENTS

The authors wish to acknowledge the skil- ful technical assistance of Mrs. M. E. Marshall and Miss J. McVeagh. This work was supported by the Cancer Research Campaign. One of us (J. B.) was in receipt of a Medical Research Council Studentship during the course of the work.

SOUS-POPULATIONS DE CELLULES DE GANGLIONS LYMPHATIQUES DE RATES MULTIPARES CYTOTOXIQUES

POUR LES CELLULES TUMORALES DE RAT ET CAPABLES DE SUPPRIMER LA CYTOTOXICITE I N VITRO

Des cellules de ganglions lymphatiques ( L NC) de rates multipares gravides ont ttP sipartes sur des colonnes de laine de nylon et leur cytotoxicitk pour les cellules- cibles tumorales a Ptt testte. La rtactivitk des L NC envers I’hipatome 0 2 3 et le carci- nome mammaire AAF57 a kte dtmontrte avec les populations retenues sur la laine de nylon, mais pas avec les cellules klukes. Bien que 25 % seulement des prkparations de LNC non fractionntes soient cytotoxiques pour les cellules tumorales, les fractions cellulaires retenues Ptaient cytotoxiques dans 11 tests sur 12 ( p = <:0.05). Elles Ctaient Pgalement cytotoxiques pour les cellules d’embryons de 15 jours, mais pas pour les fibroblastes normaux de rats adultes. Lorsque l’on a utilise‘ du sPrum de rate multi- pare, la rgactivitk des L N C retenues a Ctt bloquie dans 8 tests sur 11 ( p = <0.05). La population de LNC rtcupirie sur la laine de nylon constituait 28 a 35% de la prt- paration initiale et comptait 60 a 70% de cellules porteuses d’lg ainsi qu’une sous- population de cellules rPagissant a la PHA soluble. La stparation des L N C de rattes multipares sur des billes de verre enduites d’lg de rat puis d’lg de lapin anti-Ig de rat (en exckdent) a tgalement montrt que la population retenue est cytotoxique pour les cellules tumorales. Environ 17 a 20 % de la population initiale ont Ctt rkcupkrts parmi les cellules retenues; il s’agissait d’une population riche en cellules porteuses d’Ig (6580%) et de L N C riagissant a la PHA. Le traitement des LNC de rates multipares au fer-carbonyle supprime toute cytotoxicitt dkcelable dans ces prtparations. La cytotoxiciti des L N C retenues sur la laine de nylon disparait Pgalement aprPs incubation avec le fer-carbonyle. I1 est impossible de tirer de ce test des conclusions prPcises quant a la nature des cellules effectrices, car il est apparu que le traitement au fer-carbonyle supprime non seulement les phagocytes des prtparations de L N C mais aussi une partie des autres populations, y compris les lymphocytes porteurs d’lg. Les auteurs ont dtcele‘ une population de LNC cytotoxiques qui rPagissent aux antigenes embryonnaires associts a la tumeur (fraction retenue apres stparation sur laine de nylon), mais ils ont aussi mis en Cvidence, dans les fractions cellulaires tlutes, une sous- population de L N C de rates multipares qui supprime la cytotoxicitt des L N C retenues par la laine de nylon. On ne connait pas encore la nature exacte de cette sous-population de L NC ni le mkcanisme de son action.

771

REES ET AL.

REFERENCES

BALDWIN, R. W., and BARKER, C. R., Tumor- specific antigenicity of aminoazo-dye-induced rat hepatomas. Int. J. Cancer, 2, 355-364 ( 1 9 6 7 ~ ) .

BALDWIN, R . W., and BARKER, C. R., Demonstration of tunlour-specific humoral antibody against aminoazo-dye-induced rat hepatomas. Brit. J. Cancer, 21, 793-800 ( I 9676).

BALDWIN, R. W., BARKER, C. R . , EMBLETON, M. J . , GLAVES, D., MOORE, M., and PIMM, M. V., Demonstration of cell surface antigens on chemically induced tumours. Ann. N.Y. Acad. Sci.,

BALDWIN, R. W., BOWEN, J . G . , EMBLETON, M . J . , PRICE, M. R., and ROBINS, R. A,, Serum factors modifying lymphocyte cytotoxicity for tumour cells. Advanc. Biouci., 12, 539-559 ( 1 9 7 3 ~ ) .

BALDWIN, R. W., and EMBLETON, M. J. , Immunology of 2-acetylaminofluorene-induced rat mammary adenocarcinomas. Int. J . Cancer, 4, 47-53 (1969).

BALDWIN, R. W., and EMBLETON, M. J. , Demonstra- tion by colony inhibition methods of cellular and humoral immune reactions to tumour-specific antigens associated with aminoazo-dye-induced rat hepatomas. Int. J . Cancer, 7 , 17-25 (1971).

BALDWIN, R. W., and EMBLETON, M. J . , Neoantigens on rat tumours defined by in vitro lymphocytotoxicity assays. Int . J. Cancer, 13, 433-443 (1974).

BALDWIN, R. W., EMBLETON, M. J . , PRICE, M. R., and VOSE, B. M., Embryonic antigen expression on experimental rat tumours. Transplant. Rev., 20,

BALDWIN, R . W., EMBLETON, M . J., and ROBINS, R. A., Cellular and humoral immunity to rat hepatonia- specific antigen correlated with tumour status. Int. J . Cancer, 11, 1-10 (19736).

BALDWIN, R . W., GLAVES, D., and VOSE, B. M., lmmunogenicity of embryonic antigen associated with chemically induced rat tumours. Int. J . Cancer,

BALDWIN, R. W., PRICE, M. R., and ROBINS, R. A., Inhibition of hepatoma immune lymph-node cell cytotoxicity by tumour-bearer serum, and solu- bilized hepatoma antigen. Int . J . Cancer, 11,527-535 ( I 973c).

COGGIN, J . H. , AMBROSE, K. R. , and ANDERSON, N. G., Fetal antigen capable of inducing transplantation immunity against SV40 hamster tumour cells. J. Immunol., 105, 524-526 (1970).

177, 268-279 (1971).

77-99 (1974~) .

13, 135-142 (19746).

EVANS, R. , and ALEXANDER, P., Co-operation of immune lymphoid cells with macrophages in tumour immunity. Nature (Lond.), 228, 620 (1970).

effect of lymphocytes from pregnant mice on cultivated tumour cells. I . Specificity of effector cells, blocking by serum. Int. J . Cancer, in press.

JULIUS, M. H., SIMPSON, E., and HERZENBERG, L. A,, A rapid method for the isolation of functional thymus-derived murine lymphocytes. Europ. J . Immunol., 3 , 645-649 (1973).

LAMON, E. W., WIGZELL, H. , ANDERSSON, B., and KLEIN, E., Antitumour activity in vitro deDendent

HELLSTROM, I., and HELLSTR~M, K. E., Cytotoxic

on immune B lymphocytes. -Nature New Biol., 244, 209-210 ( 1 9 7 3 ~ ) .

LAMON, E. W., WIGZELL, H . , KLtIN, E., ANDtRS- SON, B. , and SKURZAK, H. M., The lymphocyte response to primary Moloney sarcoma virus tumours in BALB/c mice. Definition of the active subpopulations a t different times after infection. J. exp. Med., 137, 1472-1493 (19736).

LE MEVEL, B. P., and WELLS, S. A., Foetal antigens cross-reactive with tumour-specific transplantation antigens. Nature (Lond.), 244, 183-184 (1973).

MACLENNAN, I. C. M., LOEWI, G . , and HOWARD, A,, A serum immunoglobulin with specificity for certain homologous target cells which induces target cell damage by normal human lymphocytes. Immunology, 17, 897-910 (1969).

PLATA, F., GOMARD, E., LECLERC, J.-C., and LFVY, J . P., Further evidence for the involvement of thymus- processed lymphocytes in syngeneic tumour cell cytolysis. J . Immunol., 1 1 1 , 667-671 (1973).

REES, R. C., PRICE, M. R., BALDWIN, R . W., and SHAH, L. P., Inhibition of multiparous rat lymph node cell cytotoxicity by hepatoma-associated embryonic antigen. Nature (Lond. ) , 252, 75 1-753

SEEGER, R. C., and OWEN, J . J . T., Antituniour cytotoxic effects mediated by minor and major cell populations of lymph nodes. Nature (Lond.),

TOIVANEN, P., and TOIVANEN, A., Selective activation of chicken T lymphocytes by concanavalin A. J . Immunol., 111, 1602-1603 (1973).

WITSCHI, E., Mammals. In: Development of verteb- rates, Chap. 22, p. 386-414, W. B. Saunders Co., London (1956).

( I 974).

252, 420-421 (1974).

772

subpopulations of multiparous rat lymph-node cells cytotoxic for rat tumour cells and capable of...

Documents