Znt. J. Cancer: 19, 441-449 (1977)

CHARACTERIZATION OF HUMAN LYMPHOCYTE SUBPOPULATIONS FOR CYTOTOXICITY AGAINST TUMORDERIVED MONOLAYER CULTURES Tibor BAKACS,Peter GERGELY I, Santoso CORNAIN and Eva KLEIN Department of Tumor Biology, Karolinska Institutet, S-104 01 Stockholm 60, Sweden

Lymphocyte fractions of three healthy donors were tested for cytotoxicity in a 48-h assay on several tumor-derived lines. Analysis of surface markers indicated that a non-T non-B fraction comprising about SOo/, Fc-receptor-positive cells was most potent. The cell yield in this fraction was on the average 6.5% of the non-fractionated population. Elimination of Slg positive cells did not influence the cytotoxic potential. Pure T cells isolated with E-rosetting subsequent to passage on nylon wool column had no or very low cytotoxicity. W e had indications that cells without conventional markers -6' null " cells-were also cytotoxic.

Documentation of non-disease-related or natural cytotoxicity (NSC) of human blood lymphocytes towards a variety of target cells (Takasugi et al., 1973, 1974, 1975; Skurzak et al., 1973) raised doubts concerning the validity of the cytotoxic assay in human tumor immunology. It is essential to define and characterize the cells responsible for this effect since meaningful studies in search of disease-relatedness may still be conducted after elimination of the cells responsible for such nondisease-related cytotoxicity. Besides the importance for human tumor immunology, such studies are motivated for functional characterization of lymphocyte subsets and elucidation of the possible biological significance of the presence of such killer cells in certain individuals. Non-disease-related activities may still be selective if different individuals react selectively with targets in a panel of lines. In previous studies, non-disease-related effector cells were claimed to be monocytes (Holtermann et al., 1974), activated T cells (Hersey et al., 1975), non-T cells with complement receptors (de Vries et al., 1974), complement/Fc-receptors (Jondal and Pross, 1975; Pross and Jondal, 1975) or Fc-receptors (Peter et al., 1975; Kiuchi and Takasugi, 1976). Since in previous studies disease-related effects were revealed after elimination of the natural cytotoxic cells in infectious mononucleosis, our aim was to characterize such cells in the monolayer cytotoxic systems (Svedmyr and Jondal, 1975) also. The strategy of the experiments was as follows: is it possible to concentrate the cytotoxic cell to a particular lymphocyte fraction ? Is the same lympho-

cyte subset responsible for activity in different individuals ? Is the cytotoxicity reproducible with the lymphocytes of one individual from experiment to experiment?

MATERIAL AND METHODS

Lymphocyte source

Three healthy blood donors from Karolinska Hospital Blood Bank Center were selected for repeat tests since their lymphocytes showed cytotoxicity against several target cells. One hundred ml heparinized blood were drawn at 3-weekly intervals. Target cell cultures

2T and 393T osteogeneic sarcoma cell lines (received from K. Nilsson, Wallenberg Laboratories, Uppsala), and QDHT derived from nasopharyngeal carcinoma (received from Dr. Gordon Sato) were kept in 10% FCS containing Dulbecco's modified MEM. MCF7, derived from mammary carcinoma (received from Dr. Marvin Rich, Michigan Cancer Foundation) was kept in RPMI 1640 containing FCS as above and 260 IE Insulin/1000 ml culture medium. The lines were fed every 3rd day and passaged every 6th day. Effector cells

The Ficoll-Isopaque method of Boyum (1968) was used to isolate lymphocytes from the blood. The yield was about 1 x lo6 lymphocytes /ml of blood. Macrophages were removed by carbonyl iron treatment (Jondal, 1974), This suspension was designated as " unfractionted lymphocytes " : population A. Subsequent processing of the cells was carried out in the following steps: cells recovered after adherence to nylon wool: population B. Passage through a nylon wool column: population C. Subsequent removal of Received: December 21, 1976. Present address: I1 Department of Medicine, Semmelweis University, Budapest, Hungary.

BAKACS ET

442

the residual Fc receptor-positive cells by EA rosette depletion: population F. Recovery of the EA rosette-forming cells: population G. Recovery of T cells by E-rosetting : population E. Recovery of non-T non-B cells by nylon wool column passage and E rosette depletion: population D. In some experiments E- or EA-rosettes were recovered directly after iron treatment. The expected composition of cell populations and the lymphocyte yield after different separation steps are shown in Table I. From the recoveries it can be calculated that 17 % is lost during nylon wool passage and recovery of the attached cells; 20% in dividing the nylon-woolpassed cells into E-rosetted cells and the rest, and 12% when collecting the EA rosetted cells and the rest. Zg-anti Ig-coated column. In some experiments, columns were prepared according to Wigzell et al. (1972) with some modification. Glass beads were incubated with 0.5 % rabbit globulin in 0.15 M PH 7.4 phosphate-buffered saline (PBS) at 4" C overnight. Subsequently, the beads were poured into a glass column (1.5 x 30 cm, Pharmacia, Uppsala, Sweden) and washed with 100ml buffer. The column was filled with 1 :4 diluted goat anti-rabbit immunoglobulin serum (kindly provided by Dr. B. Anderson) and allowed to stand at 4" C for at least 2 h. Thereafter, the columns were washed once again with 100 ml PBS. Ten to 20 x lo6lymphocytes in 3 ml medium were passed through the columns at a flow rate of approximately 2ml/min. The effluent cells were collected, washed and counted. This fraction was designated as Ig-IgC.

AL.

Nylon wool column passage. (Population C ) the method of Julius et 01. (1973) was applied with a slight modification (Cornain et al., 1975). The nylonwool-adherent cells (population B) were eluted by rinsing the wool with FCS at 37" C. Removal and recovery of Fc-receptor positive lymphocytes. EA-rosettes were formed according to the method of Hallberg et al. (1973). 7 s rabbit anti ox RBC (kindly provided by Dr. B. Anderson) was used for sensitization of a 2.5 % washed ox RBC to form EA cells. Equal volume of lymphocytes (4-8 x 106ml) and 0.5% suspension of EA cells were incubated at 37" C for 15 min, pelleted at lOOg for 8 min and further incubated for 30 min at 0"C. The pellet was resuspended and centrifuged on FicollIsopaque gradient to remove the EA rosetteforming cells. The cells in the interface were collected and washed (population F). The EA rosetted cells were recovered (population C ) from the pellet by lysis of the erythrocytes with tris-buffered isotonic 0.83 % NH&l for 10 min at room temperature. Removal and recovery of T cells. This was performed after nylon wool passage by E rosette buoyant density fractionation. E rosettes were formed at +4" C by the method of Jondal et al. (1972). The separation and recovery of the pelleted SRBCrosetting cells (population E) were performed as described above for the EA rosettes. The cells in the interface were also collected and designated as non-T non-B fraction (population D). Test for surface markers. Each cell fraction was tested for surface markers : surface immunoglobulin (SIg) was detected by direct immunofluorescence.

TABLE I EXPECTED COMPOSITION OF LYMPHOCYTE SUBPOPULATIONS

Subpopulation. fractionation steps

A; iron treated B; eluted from nylon wool C ; nylon wool passed

Yield expressed as % of the total lymphocyte number

Lymphocytes

Present

'

62.542.8 8.3*1.6 23.0k1.2

Absent

B, T, Fc, 0 B, Fc T, Fc, 0

T, 0 B

Enrichment of SIg-positive cells No elimination, only reduction of Fc-positive cells

Fc, 0

B, T

Efficient enrichment of Fc-positive cells Efficient elimination of Slg and Fc-positive cells Efficient elimination of Slg and Fc-positive cells Efficient elimination of Slg and Fc-positive cells

thereafter D; E rosettes depleted or

6.5k0.6

E; E rosettes recovered

23.8*11.5

T

B, Fc, 0

F; EA rosettes depleted

19.8h3.3

T, 0

Fc, B

Fc

B, T, 0

G ; EA rosettes recovered

6.1&1.8

Mean of seven experiments performed with two donors.

Remarks concerning our results

443

NON-DISEASE-RELATED CYTOTOXICITY

FITC labelled goat anti-human Ig (Hyland Travenol Laboratories, Los Angeles, Calif., USA) was used at 1 :30 dilution. The cells were stained for 45 min at 4" C. The proportion of cells with surface immunoglobulin was determined by scoring 200 cells. The Fc-receptor-bearing cells were detected both by the EA rosetting technique as described above, and by aggregated human gamma globulin (HAG) binding (Dickler and Kunkel, 1972). In each test at least 200 cells were scored. Lymphocytes binding three or more erythrocytes were counted as rosettes. Cells showing clear patchy fluorescence were considered positive.

Lierbyen, Norway) on to glass fibre filter paper and transferred into scintillation vials. Three ml of scintillator solution (5.5 g Permablend 111, Packard, per 1 toluene) was added and counted in a Packard Tri-Carb Scintillation spectrometer. Each assay was done in six replicates. The percentage of the target cell reduction was calculated on the basis of the values in the wells containing no lymphocytes (medium control). The statistical significance of the cytotoxicity tests was evaluated by the two-tailed Student's test. RESULTS

Cytotoxic assay

Monolayers with a degree of 70 to 90% confluency were washed three times with Earle's medium without proline and labelled with SH proline as described by Bean et al. (1973). Four ml of Earle's medium with 10%FCS, penicillin, streptomycin and 50 pCi of 3H-proline (L-5-3H/proline 18 Ci/mmol, Radiochemical Centre, Amersham, England) per ml were added and the cultures incubated for 18-20 h. Subsequently, the cells were washed with RPMI 1640 medium, detached from the surface with 0.25 % trypsin and 0.002 % EDTA, washed and resuspended in RPMI 1640 with FCS (10%). Twenty p1 of the target cell suspension with 3,000 labelled cells were added into each well of a Falcon plastic microplate (No. 3040) containing 0.1 ml culture medium with 10% FCS. After 8 h incubation at 37" C, approximately 2,000 cells were attached. Effector cells were tested in two effector target cell ratios of 100:l and 50:l when the cell numbers were sufficient. The plates were incubated at 37" C in 5 % COz for 48 h, then washed with PBS, and thereafter the cells were removed by trypsin (0.25 %) treatment and collected by a multiple sample harvester machine (Skatron,

Two to four experiments were done at 3-weekly intervals with the lymphocytes of each donor. The results are presented in Tables 11-IV. We shall point out certain details in connection with each donor and give an overall conclusion thereafter. On the whole, repeat experiments gave similar results with few exceptions. Subject 1 (Table 11). In the first experiment the effects of Ig-anti Ig and nylon wool column treatment are shown in comparison. There was no major difference between the two types of separation (this was the experience in many previous experiments also, not shown here). Both columns removed the majority of Slg-carrying cells, reduced the proportion of Fc-receptor positive cells and did not abolish cytotoxicity. In the following experiments the nylon wool column was used. Subsequent depletion of Fc receptor-positive cells (F) removed all cytotoxicity against the 2T cell line, whereas 393T was still sensitive. In the second experiment, performed 3 weeks later, we attempted to eliminate Fc-positive cells directly by EA rosetting. However, this procedure did not

TABLE I1

CYTOTOXICITY OF THE VARIOUS LYMPHOCYTE FRACTIONS WITH BLOOD OF SUBJECT 1 Target cells Effector population ZT

A ; iron treated Ig-IgC C; nylon wool passed F; EA-rosette-depleted

Direct EA rosette-depleted G ; EA-rosettes recovered B; nylon wool eluted

I2

I1

31 23 53 -6 -

22

5'

-

*

Composition of the population: % of cells

3931

-

64 59 13 19 12

'

I

I1

78 81 75

82 84 86 83 50 63

70

35& -

SIg

QDHT

-

HAG

I1

1

I1

9 19 51 11 26 9

8 0.2 1 -

8 0.5 1.5 9 5 44

-

1

16 8.5 10 -

-

-

-

EA-rosettes

I1

I

I1

8.5 9.5 1.5 6.5 16 9

16 10 8.7 -

9 5.5 -

-

1 The results are calculated from the cpm given as percentage of reduction in relation to the cpm in the medium control wells, negative values designate stimulation; effector:target ratio was 1OO:l; the values for 50:l are not given. With the lower ratios correspondingly lower effects were obtained. Experiment number. - Not done. - a 50:l effector target ratio. Significance: p

Characterization of human lymphocyte subpopulations for cytotoxicity against tumor-derived monolayer cultures.

Znt. J. Cancer: 19, 441-449 (1977) CHARACTERIZATION OF HUMAN LYMPHOCYTE SUBPOPULATIONS FOR CYTOTOXICITY AGAINST TUMORDERIVED MONOLAYER CULTURES Tibor...
734KB Sizes 0 Downloads 0 Views