Cocaine or Δ9-tetrahydrocannabinol does not affect cellular cytotoxicity in vitro

  • Published on

  • View

  • Download


  • Int. J. lmmunopharmae., Vol. 11, No. 7, pp. 849-852, 1989. 0192-0561/89 $3.00 + .00 Printed in Great Britain. 1989 International Society for Immunopharmacology.



    Department of Pathology, University of Illinois at Chicago and West-side VA Hospital, Chicago, IL 60612, U.S.A.

    (Received 7 September 1988 and in final form 27 February 1989)

    Abst ract - - Cocaine or A9-tetrahydrocannabinol (THC), the major psychoactive component of marijuana, has no effect on in vitro cytotoxicity mediated by natural killer (NK) cells and cytotoxic T-lymphocytes (CTL) at concentrations similar to those observed in vivo.

    Cocaine and marijuana have become major substances of abuse. Various effects are implicated on different tissues and body functions (for review, see Nahas, 1984). NK cells can inflict lethal damage to target cells without prior immunization. Their recognization of antigens is not restricted by the major histocompatibility complex (MHC). In contrast, immunization generates cytotoxic lymphocytes, which mount a specific secondary response upon reintroduction of the antigen. The CTL reactivity is strictly MHC-restricted. Together these effector cells are thought to be important in host resistance against malignancy and virus-infected cells. However, few data are available regarding the influence of cocaine and THC on these subpopulations of the immune system. We, therefore, investigated the interactions of the compounds with these cytolytic cells.



    Six-week-old male C3H/HeJ(H-2k), C57BL/6(H-2 b) and A/J(H-2") were obtained from Biological Test- ing Branch, NCI (Frederick, MD).


    THC (10 mg/ml in ethanol) was purchased from Sigma. Four percent cocaine was obtained from the West-side VA Hospital, Chicago. Both drugs were diluted with RPMI-1640 medium to make stock solutions of 50 and 5/~g/ml.

    Spleen cell suspensions

    Spleen cell suspensions were prepared by mechanical disruption between frosted glass slides. The viability, as determined by trypan blue (0.4o70), always exceeded 957o.

    Effector cells

    To get NK cells, C3H/He mice were injected with polyinosinic-polycylidylic acid (Sigma) 18 h prior to removal of the spleens. Mixed lymphocyte cultures were acquired by culturing C57BL/6 spleen cells with irradiated (3000R) A/ J spleen cells (Duke, Cohen & Chervenak, 1986). To prepare peritoneal exudate lymphocytes (PEL), C57BL/6 mice received i.p. 2.5 x 106 Yac-1 tumor ceils (A/J origin) and then 4 x 107 Yac-1 cells on day 12. Peritoneal killer cells were harvested on day 18, some of which were purified with nylon wood (New England Nuclear, Boston, MA) column chromatography before assay.

    Culture medium

    RPMI 1640 medium (Gibco) was supplemented with 5% fetal calf serum, 100 U/ml penicillin, 100~g/ml streptomycin and 50taM 2-mercapto- ethanol.

    5~ Cr-release assay

    Four-hour cell-mediated cytotoxicity was determined as described by Duke et al. (1986). All tests were set up in triplicates. The percent specific release was calculated by the following formula:

    *Author to whom correspondence should be addressed.


  • 850 F. Lu and D. W. Ou


    "~ 30


    O lo

    zX9 - THC C b ) 40

    1 30 20 , . , . . . . . ; 25 50 7'5 100 200 400 600 1000 2000 ng/ml 25 50


    Concent ra t ion


    , , , , , , ,

    100 200 400 600 1000 2000 ng/ml 8OO

    Concent ra t ion

    Fig. 1. Effect of THC (,O) and cocaine (A,A) oh NK cytotoxicity. The means of the controls, which had no drugs but were otherwise the same, are the horizontal lines across the figure, with the standard deviations at the left ends of the lines.

    For some treatment groups, the standard deviations are less than the height of the symbols and not shown.


    .~- 50-

    x 40- 0 "5 30,


    o~ to-

    iX 9 - THC o)

    ~ o



    >, 60.

    -- 50

    40 ~" 30.


    (b ) COCAINE

    t ~ ~ ~ E/T 50

    ~ ' i ~ ~ ~' E/T=5

    " ~ 10

    . . . . o o ' ' 25 50 100 200 6 0 10 0 2000 ng /ml 25 50 100 200 400 600 1000 2000 ng /ml

    Concentration Concentration

    Fig. 2. Effect of THC (,O) and cocaine (A,A) on cytotoxic mixed lymphocyte cultures. The means and standard deviations of the controls are drawn as in Fig. 1.

    % specific release = counts/mi~xp- counts/min~pom counts/minm, _ counts/min~po, x 100.

    Maximal release (counts/min) was determined from supernatants of target cells that were lysed by addit ion of Triton X-100 (final concentrat ion 0.5%). Spontaneous release (counts/min~po,,,) was determined from target cells incubated without added effector cells and was always less than 16%. The maximal or spontaneous release was not affected by the presence of cocaine or THC at the concentrat ions used in this study. In each experiment, the control group did not contain cocaine or THC but was otherwise the same as the treatment groups. The data shown in the Results are one representative of three to five experiments.

    Statistical analysis

    Multivariate analysis of variance (MANOVA) using SAS software (SAS Institute Inc., Cary, NC) was applied. The mice (MOUSE) and the wells (WELL) of triplicates were independent variables, with WELL nested within MOUSE. The % specific release of control and that of each treatment group was compared pairwise. If P was greater than 0.05, the difference was not considered statistically significant. In the Results, P was invariably greater than 0.10.


    NK activity was assayed with 5'Cr-labeled Yac-I target cells (Fig. 1). At E/Tratio = 50, there was no significant change of cytotoxicity in the presence of THC or cocaine. Similar results were observed at a lower E/T ratio of 5. As another control, NK cells did not cause 5'Cr release from NK-resistant P815 cells.

    Mixed lymphocyte cultures generated cytotoxic T-lymphocytes in C57BL/6 responder cell populations. With THC or cocaine from 25 to 2000 ng/ml final concentrat ions (Fig. 2), the 5~Cr release was not statistically significantly different from the control groups, whether the E/T ratio was high or low.

    To further determine whether the drugs could affect cell lysis by in vivo induced lymphocytes, PEL were tested against the same target, Yac-I (Fig. 3). Neither drug exerted any effect on PEL, with or without nylon wool purif ication.


    Long-term CTL clones and cytolytic mixed lymphocyte cultures contain perforin and serine esterase activity, which are supposed to play a role in

  • 80


    :>, 60

    ._o 50"


    ~, 30 0




    Cocaine, THC and in vitro Cytotoxicity

    A9 - THC

    _ ~L._ _ _ ~ - ~ ~ -~ ~ ~ E/T=50 b)


    60- >,

    'o 50-



    o~ i0-

    100 200 400 600 1000 2000 ng/rnl

    Concentrat ion

    25 50 25


    50 100 200 600 1000 2000 ng/ml

    Concent ra t ion

    Fig. 3. Effect of THC and cocaine on PEL lytic activity. PEL were used with (O,A; E/Tratio = 0.5) or without (,A; E/T ratio = 50) nylon wool purification before cytotoxicity assay. The means of controls are the horizontal lines across

    the figure.

    their cytotoxicity. However, in vivo primed PEL do not have either molecule. Because of this, recent papers (Ostergaard, Kane, Mescher & Clark, 1987; Berke & Rosen, 1987) have suggested that PEL might be different from other groups of CTL and have a different killing pathway. In addition, PEL cells are Thy-1 + and H-2 specific (Dennert, Anderson & Prochazka, 1987) and are not adherent macrophages (Berke, Sullivan & Amos, 1972). Consequently, PEL were used as another source of CTL effector cells.

    Unlike other investigators who compared NK cells and CTL on the basis of different target ceils, Duke et al. (1986) used the same target cell so that the killer cell was the only variable. The same rationale was followed throughout this study.

    The concentrations of THC and cocaine chosen here were similar to those measured in human sera. Intranasal application of cocaine in man resulted in peak plasma concentrations of 120-474 ng/ml (Van Dyke, Barash, Jatlow & Byck, 1976). Peak plasma levels of THC immediately after a single intravenous dose are in the region of 100-500 ng/ml. The concentration falls rapidly, reaching the 5 - 10 ng/ml range at 1 h. Even though THC is fat-soluble and tends to accumulate thereafter in fat, lungs and spleen, the levels in these organs are very low. Plasma kinetics after marijuana smoking are similar to those observed following intravenous admini- strations (Nahas, 1984). Klein, Newton & Friedman

    (1987) showed that preincubation of murine cultured spleen cells with THC suppressed the NK activity. The effective concentrations however, more than 5/~g/ml, were well above in vivo blood and tissue concentrations.

    Our experiments indicated that in terms of target lysis per se, THC or cocaine affected neither mouse NK cells nor CTL, regardless of whether the latter were sensitized in mixed lymphocyte cultures or in vivo, because very little is known about the effects of these drugs on the induction of immune effector cells. One previous report demonstrated that cocaine at similar concentrations had no effect on human NK cytotoxicity in vitro (Van Dyke, Stesin, Jones, Chuntharapai & Seaman, 1986), which is consistent with our results. However, that study also found intravenous administrations of cocaine to humans increased both NK activity and cell numbers in peripheral blood. On the other hand, subchronic, but not acute, exposure of rats to THC resulted in suppression of splenic NK activity (Patel, Borysenko, Kumar & Millard, 1985). Our laboratory is currently exploring the in vivo effects of these drugs in mice.

    Acknowledgements -- The authors would like to thank Dr Richard Duke for providing mycoplasma-free Yac-1 and P815 cells as well as helpful discussion. This work was supported by the Veterans Administration.


    BERKE, G. & ROSEN, E. (1987). Mechanism and inhibition of target damage. Transplantn Proc., 19, 412-416. BERKE, G., SULLIVAN, K. A. & AMOS, B. (1972). Rejection of ascites tumor allografts -- 1. Isolation, characterization, and

    in vitro reactivity of peritoneal lymphoid effector cells from BALB/c mice immune to EL4 leukosis. J. exp. Med., 135, 1334- 1350.

  • 852 F. Lu and D. W. Ou

    DENNERT, G., ANDERSON, C. A. & PROCHAZKA, G. (1987). High activity of N-benzyl oxycarbonyl-L-lysine thiobenzyl ester serine esterase and cytolytic perforin in cloned cell lines is not demonstrable in in vivo-induced cytotoxic effector cells. Proc. natn. Acad. Sci., U.S.A., 84, 5004-5008.

    DUKE, R., COHEN, J. J. & CHERVENAK, R. (1986). Differences in the target cell DNA fragmentation induced by mouse cytotoxic T lymphocytes and natural killer cells. J. lmmun. , 137, 1442-1447.

    KLEIN, T. W., NEWTON, C. & FRIEDMAN, H. (1987). Inhibition of natural killer function by marijuana components. J. ToxicoL envir. Hlth, 20, 321- 332.

    NAHAS, G. G. (1984). Marihuana in Science and Medicine. Raven Press, New York. OSTERGAARD, H. L., KANE, K. P., MESCHER, M. F. & CLARK, W. R. (1987). Cytotoxic T lymphocyte mediated lysis

    without release of serine esterase. Nature, 330, 71 -72. PATEL, V., BORYSENKO, M., KUMAR, S.A. & MILLARD, W. J . (1985). Effect of acute and subchronic

    /~9-tetrahydrocannabinol administration on plasma catecholamine, /3-endorphine, and corticosterone levels and splenic natural killer cell activity in rats. Proc. Soc. exp. Biol. Med., 180, 400-404.

    VAN DYKE, C., BARASH, P. G., JATLOW, P. & BYCK, R. (1976). Cocaine: plasma concentrations after intranasal application in man. Science, 191, 859-861.

    VAN DYKE, C., STESJN, A., JONES, R., CHUNTHRAPAI, A. & SEAMAN, W. (1986). Cocaine increases natural killer activity. J. clin. Invest., 77, 1387- 1390.


View more >