Znt. J . Cancer: 23, 28-31 (1979)

CYTOLYTIC ACTIVITY OF CIRCULATING HUMAN MONOCYTES ON TRANSFORMED A N D UNTRANSFORMED HUMAN FIBROBLASTS

Albert0 MANTOVANI Ronald B. HERBERMAN

Laborutory of immunodiagnosis, National Cancer Institute, Bethesda, Md. 20014; and Department of fmmunobiology, Litton Bionetics, inc., Kensington, Md. 20795, USA

Aldo TAGLIABUE 2, Jack H. DEAN 3, Thomas R. JERRELLS and

fstituto Di Ricerche Farmacologiche Mario Negri, Milan, f faly ;

Monocytes from the peripheral blood of 10 normal adult human donors were tested for their cytolytic activity on untransformed or SV40-transformed, early passage, human fibroblasts. Two pairs of human fibroblast lines (2303 and SVIO-2303.2931 and SV40-2931) were used and lysis was assessed in terms of release of [JHIthymidine from prelabelled target cells over a 72-h incubation period. High levels of cytolytic activity were consistently observed on transformed fibroblasts a t attacker to target cell (A:T) ratios ranging from 5:l to 40:l. Low, but significant, lysis of untransformed cells was observed only with two donors at the highest (40:l) A:T ratio tested. Wi th all donors, the cytolytic activity on transformed lines was significantly greater than on untransformed fibroblasts. Thus natural human monocyte-mediated cytotoxicity is preferentially expressed on transformed, as compared to untrans- formed, cells although sparing of the latter is not absolute.

Rodent macrophages can inhibit growth and kill tumor target cells by a non-phagocytic mech- anism which, in most systems investigated, requires cell-to-cell contact (recently reviewed by Evans and Alexander, 1976; Hibbs, 1976; Keller, 1976~). The cytotoxic activity of rodent macrophages is not immunologically specific, but murine macrophages activated as a consequence of exposure to chemical or biological stimuli have been reported to be selectively cytotoxic against neoplastic cells as compared to normal cells (Hibbs, 1973; Holtermann et al., 1973; Kaplan et al., 1974; Cleveland et al., 1974; Meltzer et al., 1974). Resistance of normal cells to murine macrophage killing does not appear to be absolute, since significant cytolytic effects have been seen with “normal” cell lines (Keller, 1976b, 1978). On the basis of their selective cyto- toxicity for tumor cells, it has been suggested that macrophages represent a primitive surveillance mechanism against neoplasia (Hibbs, 1974; Alexander, 1976).

Although the cytotoxic activity of rodent macro- phages has been extensively investigated and some evidence has accumulated for a role of cells of the monocyte-macrophage series in the regulation of tumor growth and metastasis in experimental models (Birbeck and Carter, 1972; Eccles and Alexander, 1974; Russell and McIntosh, 1977; Wood and Gillespie, 1975; Zarling and Tevethia, 1973), the cytotoxic capacity of human mononuclear phago- cytes in relation to neoplasia has not been studied.

In the preceding paper, we have demonstrated that human monocytes, isolated from the peripheral blood of healthy adult donors, are spontaneously cytolytic and cytostatic on human and murine tumor cells in vitro (Mantovani et al., 1978). Here we present evidence that natural human monocyte- mediated cytotoxicity has considerable selectivity against transformed as compared to untransformed, early-passage, human fibroblasts.

MATERIAL AND METHODS

Heparinized venous blood (50 to 200 ml) was obtained from 10 normal, healthy, adult volunteers. The materials and methods employed to isolate adherent cells (>90 % moilocytes) and the cytolysis assay based on the release of methyl[3H]thymidine from prelabelled target cells are described in detail in an accompanying report (Mantovani et al., 1978) and are only briefly summarized here.

Monocytes from 2/10 donors (MM and HK) were isolated by adherence on microexudate- coated plastic and exposure to ethylenediamine- tetra-acetic acid (EDTA), whereas the remaining monocyte preparations were obtained by adherence on untreated plastic and scraping with a rubber policeman. More than 90% of the adherent cells were monocytes as assessed by morphologic and functional criteria and the two separation procedures yielded effector cells with similar in vitro cytotoxicity on tumor cells.

Target cells used in the present study were: the human fibroblast lines 2303, derived from a patient with acute myelogenous leukemia, and 2931, derived from a patient with xeroderma pigmentosum, and their simian virus 40 (SV40)-transformed counterparts (SV40-2303 and SV40-2931). The 2303, SV40-2303, 2931 and SV40-2931 cells were obtained through the courtesy of Dr. T. D. Lubiniecki, Meloy Lab., Springfield, Va., USA, at their 13th, 3rd, 10th and 15th passages, respectively. Each monocyte preparation was tested in the same experiment against both the transformed and untransformed line of each pair. The cell lines were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10 % fetal bovine serum and tested within four passages in our laboratory.

Received: September 12, 1978.

SELECTIVE HUMAN MONOCYTE CYTOTOXICITY 29

Non-confluent cultures in 75 cm2 tissue culture flasks (Falcon, Oxnard, Calif.) were incubated for 24 h with 25 ml of medium containing 0.5 pCi/ml of [3H]thymidine (6 Ci/molej Schwarz/Mann, Orangeburg, N.Y.). After exposure to 5 ml Trypsin- EDTA (Gibco, Grand Island, N.Y.) for 5 min at 37" C, the cells were washed twice with 50 ml of RPMI 1640 medium with 20% FBS. In the series of experiments reported here, mean isotope incorpor- ation of lo4 target cells was 7,365 cpm for 2303, 8,569 cpm for SV40-2303, 3,654 cpm for 2931 and 10,152 cpm for SV40-2931.

After resuspension in RPMI 1640 medium sup- plemented with 10% FBS, cells were used in the [3H]thymidine release assay. Target cells (104/0.3 ml) were incubated for 72 h, unless otherwise specified, with a range of attacker to target cell (A:T) ratios (from 40:l to 5 : l ) in flat-bottomed 6.4 mm culture wells (3596, Costar, Cambridge, Mass.). Tumor-cell growth was checked daily under an inverted micro- scope. Spontaneous isotope release in the super- natant never exceeded 25% of the incorporated radioactivity over a 72-h incubation period. Results were presented as percentage specific lysis (mean & SD) of three replicates per experimental group after subtraction of the spontaneous release. Statis- tical significance was assessed by one-way analysis of variance.

RESULTS

Table I summarizes data on the cytolytic activity against transformed and untransformed human fibroblasts of human peripheral blood monocytes of 10 donors after incubation for 72 h. The untrans- formed and transformed target cells showed different susceptibilities to monocyte-mediated cytotoxicity. Low, but significant, lysis of the untransformed 2303 and 2931 cell was observed with 2/10 monocyte preparations (RP and LV) at the highest (40:l) A:T ratio tested, and the cytotoxicity levels did not exceed 10 %. Under these experimental conditions, the remaining monocyte preparations did not show significant cytolytic capacity on untransformed target cells. In contrast, monocyte preparations from all donors were significantly cytotoxic against SV40-2303 and SV40-2931 cells, with specific isotope release ranging from 19.5 to 45.8% at an A:T ratio of 40:1. Interestingly, the two monocyte preparations (RP and LV) which showed significant lysis against untransformed 2303 and 2931 also showed the highest cytotoxicity levels (40.1 and 45.8%, respectively, at an A:T ratio of 40:l) of the donors tested against the SV40-transformed lines. With all of the donors, specific lysis of transformed cells was significantly (pt0.05) higher than that of the corresponding untransformed line of each pair. Preferential destruction of transformed as opposed to untransformed human fibroblasts was also evident microscopically.

When the kinetics of the cytolytic reaction was investigated (Table 11), significant lysis of SV40- transformed cells was detected only after 48 h of incubation, with maximal specific isotope release occurring at 72 h. In the tests with untransformed human fibroblasts, monocytes from donors RP and

LV showed similar, low levels of cytolysis at 48 and 72 h, and the other monocyte preparations were unreactive over the 72-h incubation period.

DISCUSSION

The results presented here show that human monocytes obtained from the peripheral blood of healthy adult volunteers are consistently cytolytic in vitro against early passages of SV40-transformed human fibroblasts. Thus, these data extend our previous observation that circulating human mono- cytes are cytotoxic in vitro against human and murine tumor target cells (Mantovani et al., 1978). In contrast, untransformed, early-passage human fibroblasts were relatively resistant to cytolysis by circulating human monocytes. Low, but significant, levels of lysis of untransformed cells were detected only with two donors who, among the 10 donors investigated, also showed the highest cytotoxic activity against the SV40-transformed cells. With each monocyte preparation tested, cytolytic activity against the transformed fibroblasts was significantly greater than against the corresponding untrans- formed line.

In a number of studies, activated rodent macro- phages have been reported to kill neoplastic cells as opposed to normal cells (Hibbs, 1973, 1974; Holtermann et al., 1973; Cleveland et al., 1974; Meltzer et al., 1974; Kaplan et al., 1974). However, more recently, the concept of the selectivity of murine macrophage cytotoxicity for transformed cells has been questioned (Keller, 19766, 1978). At least, part of the conflicting (Hibbs, 1973; Keller, 19766) or irreproducible (Hibbs, 1976) data can be attributed to the use of " normal " cell lines continuously maintained in vitro, such as 3T3 which has been recently demonstrated to be tumor- igenic in mice (Boone et al., 1976). In the present study, we approached the question of the selectivity of cytotoxicity by using two pairs of early-passage, untransformed or SV40-transformed, human fibro- blast cells. The results indicate that monocytes from normal human donors have preferential reactivity against transformed as compared to untransformed cells, but that sparing of untransformed fibroblasts is not absolute. Interestingly, in a parallel study on cytotoxicity by monocytes or macrophages from normal mice, we observed a similar preferential killing of transformed as compared to untrans- formed human fibroblasts (Tagliabue et al., in preparation).

The mechanism(s) responsible for the preferential cytotoxic activity of cells of the monocyte-macro- phage series against transformed cells is still un- known. In experimental models, it has been shown that differences in proliferative capacity between transformed and untransformed cells do not account for their different susceptibilities to macrophage cytotoxicity (Holtermann et al., 1973). This also seemed to be the case here, since the levels of [3H]thymidine incorporation by transformed and untransformed 2303 cells were very similar. It has been suggested that possible alterations in cell membrane properties play a role in the selective

30 MANTOVANI ET AL.

TABLE I

CYTOLYTIC ACTIVITY OF CIRCULATING HUMAN MONOCYTES ON TRANSFORMED AND UNTRANSFORMED HUMAN FIBROBLASTS

% specific lysis ( ~ s D ) Donor Target cells

40:l ' 20:l 1 O : l 5:1

J.S.

M.S.

B.F.

H.K.

R.P.

M.M.

A.M.

A.P.

A.T.

L.V.

2303

2303

2303

2303

2303

2303

293 1

293 1

2931

293 1

SV40-230.3

SV40-2303

SV40-2303

SV40-2303

SV40-2303

SV40-2303

SV40-293 1

SV40-293 1

SV40-2931

SV40-293 1

3.8 (2.3) 19.5 (1.4) * - -

3.6 (2.5) 38.0 (2.3) * 2.8 (2.7) 36.6 (0.6) * 9.3 (0.6) * 40.1 (1.7) *

-

0 25.1 (0.3) * 1.4 (1.8) 21.6 (2.2) * 8.7 (0.8) * 45.8 (3.5) *

1.7 (0.5) 18.0 (1.7) * - -

2.3 (1.5) 32.1 (1.4) * 3.3 (1.1) 23.1 (0.7) * 4.8 (3.4) 32.5 (5.7) * 0.9 (0.3) 25.8 (0.5) * 2.1 (0.4) 23.4 (2.2) * 1.2 (0.3) 19.9 (0.9) * 1.1 (0.3) 16.8 (0.4) * 0.4 (1.5) 33.5 (1 .8) *

0.1 (0.6) 15.0 (2.9) * 2.9 (0.2) 18.7 (0.8) * 0.9 (0.8) 20.1 (1.8) * 2.1 (0.7) 14.5 (2.7) * 3.5 (0.6) 16.7 (2.4) * - -

2.0 (0.3) 16.3 (1.6) * :' 0.3 (0.8) 14.5 (0.9) * 0.2 (0.5) 12.1 (1.9) * 0.2 (0.6) 19.3 (0.6) *

- 2

- 1.1 (0.4) 9.4 (0.9) * -

- 0

7.4 (0.8) *

' A:T ratio. - Signifi-antly (p,:0.05) ahove spontaneous release. In the present series of experiments. mean spontaneous release levelr in 72 h were

Not tested. - Significantly above (pCr0.05) specific lysis against the Corresponding untransformed fibroblasts.

10.1 for 2303, 24.5% for SV40-2303. 10.7% for 2931 and l3.8"( for SV40-2931.

cytotoxic capacity of murine macrophages against grown in virro under conditions which do not neoplastic cells (Hibbs, 1976) or virus-infected support cellular proliferation (Schiaffonati and normal cells (Goldman and Hogg, 1978). Untrans- Baserga, 1977). Thus, it is possible that the prefer- formed cells also have been reported to survive ential destruction of transformed cells may in better than their transformed counterparts when part depend on the higher susceptibility of these

TABLE I1

KINETICS OF THE CYTOLYTIC EFFECTS OF HUMAN PERIPHERAL BLOOD MONOCYTES ON TRANSFORMED AND UNTRANSFORMED HUMAN FIBROBLASTS

Incubation time (h)

24 48 72 Donor A:T ratio Target cells

J.S.

M.S.

B.F.

H.K,

R.P.

A.P.

A.T.

L.V.

20:l

5: l

40:l

40:l

40:l

40:l

40:l

40:l

2303

2303

2303

2303

SV40-2303

SV40-2303

SV40-2303

SV40-2303 2303 SV40-2303

SV40-2931 2931

293 1 SV40-2931 293 1 SV40-293 1

1.1 (0.4) ' 0

0.1 (0.8) 0

0 4.9 (0.2) * 0

0.9 (0.5) 2.0 (0.5) 30.8 (0.5) * 2.0 (0.3) 34.9 (3.5) * 8.4 (0.7) * 33.5 (13) * 1.7 ii.ij 11.9 (0.1) * 1.6 io.3j 5.8 (1.1) * 9.4 (I .8) * 26.1 (1.3) *

1.7 (0.5) 18.0 (1.7) * a 1.1 (0.4) 9.4 (0.9) *.3

3.6 (2.5) 38.0 (2.3) * 2.8 (2.7) 36.6 (0.6) * 9.3 (0.6) * 40.1 (1.7) *

0 25.1 (0.3) * 1.4 (1.8) 21.6 (2.2) * 8.7 (0.8) * 45.8 (3.5) * a

~~ ~~ ~~

' Results are percentage specific lysis ( ~ s D ) . - Significantly (p<O.O5) above specific lysis against the corresponding untransformed fibroblasts.

* Significantly above (p<O.OS) spontaneous release. Spontaneous release levels for 2303 were 5.1 % at 24 h, 6.8 at 48 h and 10.1 at 72 h; for SV40-2303 they were 3.1%. 12.3% and 24.5%: for 2931 they were 6.5% at 48 h and 10.7% at 72 h: and for SV40-2931 they were 8.6% at 48 h and 13.8% at 72 h.

Not tested. -

SELECTIVE HUMAN Mc 3NOCY TE CY TOTOXICITY 31

cells to develop lethal damage following inhibition of growth by monocytes or macrophages.

Extrapolation of in vitro cytotoxicity data to in vivo conditions requires caution. However, the evidence presented here that human monocytes from normal donors have preferential cytolytic activity against transformed target cells is consistent with the hypothesis that cells of the rnonocyte- macrophage series may play a role in surveillance against neoplasia.

ACKNOWLEDGEMENTS

Albert0 Mantovani was supported by a Yamagiwa- Yoshida study grant from the International Union Against Cancer, Geneva, Switzerland, and from CNR, Rome, Italy. Laboratory studies were sup- ported in part by contract N01-CB-63975 from the Division of Cancer Biology and Diagnosis of the National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA.

ACrIVITE CYTOLYTIQUE DES MONOCYTES HUMAINS ClRCULANTS SUR LES FIBROBLASTES HUMAINS TRANSFORMES O U NON

Des monocytes du sang peripherique de dix donneurs adultes normaux ont ete testes du point de vue de leur activite cytolytique sur des fibroblastes humains non transform6s o u transformes par le SV40, qui n’en etaient qu’i quelques pas- sages. Deux paires de lignees de fibroblastes hurnains (2303 et SV40-2303, 2931 et SV40-2931) ont Bt6 utilisees; la lyse a kt6 6valu6e d’apres le relargage de la thymidine tr itiee par les cellules cibles premarquees, au cours d’une p6riode d’incubation de 72 h. Une forte activite cytolytique sur les fibroblastes transformes a regulikrement t te observke lorsque les rapports entre attaquants et cibles (A:T) variaient entre 5:l et 40:l. On a not6 une lyse faible, mais significative, des cellules non transformees avec deux donneurs seulement, au rapport A:T le plus 61ev6 que I’on ait test&, c’est-8-dire 40: l . Avec tous les donneurs, I’activitt cytolytique siir les cellules transformtes etait significativement plus grande que sur les fibroblastes non transform& Donc, chez I’homme, la cytotoxicite naturelle 8 mediation monocytaire s’exprirne prkf6rentiellement sur les cellules trans- formkes, par comparaison avec les cellules non transformees, bien que ces dernibres ne soient pas totalement Bpargnees.

REFERENCES

ALEXANDER, P., Surveillance against neoplastic cells-is it mediated by macrophages? Brit. J. Cancer, 33, 344-347 (1976). UIRRECK, M. S. C., and CARTER, R. L., Observations on the ultrastructure of two hamster lymphomas with particular reference to infiltrating macrophages. Int . J . Cancer, 9, 249-257 (1972).

Vasoformative sarcomas alising from BALBI3T3 cells attached to solid substrates. Cancer Res., 36, 1626-1633 ( 1 976). CLEVELAND, R. P. , MELTZER, M. S., and ZBAR, B., Tumor cytotoxicity in vitro by macrophages from mice infected with Mycobacterium bovis strain BCG. J. nat. Cancer Inst.,

ECCLES, S. A., and ALEXANDER, P., Macrophage content of tumours in relation to metastatic spread and host immune reaction. Nature (Lond.), 250, 667-669 (1974). EVANS, R., and ALEXANDEH, P., Mechanisms uf extracellular killing nucleated mammalian cells by macrophages. In: D. S. Nelson (ed.), Immunobiology of-the mcrcrophage, pp. 535-576: Academic Press, New York (1976). GOLDMAN, R., and HOGG, N., Enhanced susceptibility of virus-infected fibroblasts to cytostasis mediated by peri- toneal exudate cells. J . Immunol. (in press). HIBBS, J. B., Macrophage nonimmunologic recognition: Target cell factors related to contact inhibition. Science,

HIBBS, J. B., Discrimination between neoplastic and non- neoplastic cells in vitro by activated macrophages. J. nut. Cancer Inst . , 53, 1487-1491 (1974). HIBBS, J. B., The macrophage as a tumoricidal effector cell : a review of in vivo and in virro studies on the mechanism of the activated macrophage nonspecific cytotoxic reaction. In: M. A. Fink (ed.), The Macrophage and neoplasia, pp. 83- 90, Academic Press, New York (1976). HOLTERMANN, 0. A,, KLEIN, E., and CASALE, G. P., Selective cytotoxicity of peritoneal leukocytes for iieoplastic cells. Cell. Immunol., 9, 339-322 (1973). KAPLAN, A. M., MORAIIAN, P. S., and REGELSON, W., Induction of macrophage-mediated tumor-cell cytotoxicity

BOONE, w. c., TAKEICHI, N., PARANIPE, M., and GILDEN, R.,

52, 1887-1894 (1974).

180, 868-870 (1973).

by pyran copolymer. J. nut. Cancer Inst., 52, 1919-1921 (1974). KELLER, R., Cytostatic and cytocidal effectors of activated macrophages. fn: D. S. Nelson (ed.), Immunokiology of the mncrophage, pp. 487-512. Academic Press, New York (1976~) . KELLER, R., Susceptibility of normal and transformed cell lines to cytostatic and cytocidal effects exerted by macro- phages. J. nut. Cancer Inst., 56, 369-374 (19766).

KELLER, R., Macrophage-mediated natural cytotoxicity against various target cells in vitro. I . Macrophages from diverse anatomical sites and different strains of rats and mice. Brit. J. Cancer, 37, 732-741 (1978).

MANTOVANI, A., JERRELLS, T. R . , DEAN, J . H., and HERRER- MAN, R. B., Cytolytic and cytostatic activity on tumor cells of circulating human monocytes. Int. J. Canrer, submitted for publication. MELTZER, M. S., TUCKER. R. W., SANFORD, K. K., and LEONARD, E. J., Interaction of BCG-activated macrophages with neoplastic and nonneoplastic cell lines in vitro: quanti- tation of the cytotoxic reaction by release of tritiated thymi- dine from prelabeled target cells. J. nar. Cancer Insr., 54,

RUSSELL, S. W., and MCINTOSH, A. T., Macrophagesisolated from regressing Moloney sarcomas are more cytotoxic than those recovered from progiessing sarcomas. A‘ature (Lond.),

SCHIAFFONATI, L., and BASERGA, R., Different survival of normal and transformed cells exposed to nutritional con- ditions nonpermissive for growth. Cancer Res., 37, 541-545 (1977). WOOD, G. W., and GILLESPIE, G. Y., Studies on the role of macrophages i n regulation of growth and metastasis of murine chemically induced fibrnsarconias. Int. J . Cancer,

ZARLING, J. M., and TEVETHIA, S . S., Tiansplantation immunity to simian virus 40-transformed cells in tumor- bearing mice. 11. Evidence for macrophage participation at the effector level of tumor rejection. J. nar. Cancer Insr..

1 177-1 184 ( 1975).

268, 69-71 (1977).

16, 1022-1029 (1975).

50, 149-157 (1973)