AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 6, Number 8, 1990 Mary Ann Liebert, Inc., Publishers

Cytotoxic Activity Against HIV-Infected Monocytes by

Recombinant Interleukin 2-Activated Natural Killer Cells

Downloaded by Pierre Marie Curie University from www.liebertpub.com at 08/28/18. For personal use only.

ROBERT J.

MELDER,1,4 R. BALACHANDRAN,3 CHARLES R. RINALDO,1,3 PHALGUNI GUPTA,3 THERESA L. WHITESIDE,1,4 and RONALD B. HERBERMAN1,2,4

ABSTRACT Natural killer (NK) cells have long been known to aid in the control of viral infections by killing virus-infected cells, including those infected with human immunodeficiency virus (HIV). Among the possible NK-susceptible target cells in an infected individual, the monocyte/macrophages are of special significance since they may serve as both a reservoir of HIV and aid in dissemination of the virus throughout the body. A new technique for the enrichment and cultivation of large numbers of recombinant interleukin 2 (rIL-2 (-stimulated NK cells has been developed which provides cells with high cytotoxic activity. These IL-2-activated NK cells, adherent lymphokine-activated killer cells (A-LAK), can kill monocytes infected with HIV for 24 h to 7 days, with optimal target sensitivity between 3 and 7 days. Recognition and killing of the infected monocytes did not appear to be restricted by the major histocompatibility complex (MHC) antigens and could be cold-target inhibited by tumor cell lines. A-LAK cells may be useful in newer therapeutic approaches to treatment of HIV infection.

INTRODUCTION

NATURAL significant models.1

(NK) CELLS HAVE BEEN shown to play a role in restricting viral pathogenesis in a variety of Several in vitro studies of human NK cells and virus-infected targets have shown preferential lysis of infected cells.24 More recently, NK cells have been found to kill T cells infected with human immunodeficiency virus (HIV)5,6 as well as an HIV-infected monocyte-like cell line.7 NK cells may also kill HIV-infected target cells by an antibody-dependent cellular cytotoxicity mechanism, through NK-associated Fc receptors and anti-HIV antibodies.8 In addition to other immunological abnormalities, patients with the acquired immunodeficiency syndrome (AIDS) or AIDS-related complex (ARC) also usually have abnormal levels of NK cell activity.9"11 It has recently become possible to generate cultures of highly enriched rat or human NK cells with high cytotoxic activity against both NK-sensitive and NK-resistant tumor target cells.12"14 Such cultures represent an expanded subpopulation of NK cells which become adherent to plastic surfaces when stimulated with rIL-2 and have been called adherent lymphokine-activated killer cells (A-LAK).13 It has been possible to KILLER

culture A-LAK cells on a large scale, thus facilitating their use in clinical adoptive immunotherapy trials in patients with certain tumors.14 The anti-HIV properties of these cells have been investigated to determine if A-LAK cells may be useful in adoptive immunotherapy of AIDS. Although NK cells stimulated with rIL-2 have been found to kill HIV-infected T lymphoblasts,5 their cytotoxic activity against infected monocytes has not been characterized. The HIV-infected monocyte/macrophage may represent a critical element in the development of AIDS since it may act as both a reservoir of virus and a means of disseminating the infection throughout different tissues.15-18 We have chosen therefore to use HIV-infected monocytes as possible targets for IL-2-activatedNK cells.

MATERIALS AND METHODS Cells

Peripheral blood mononuclear (PBMN) cells were obtained by Ficoll-hypaque gradient isolation of leukopheresed blood or

Departments of 'Pathology, 2Medicine, and infectious Diseases and Microbiology, University of Pittsburgh and 4Pittsburgh Cancer Institute, Pittsburgh, PA 15213. 1011

MELDER ET AL.

1012

heparinized venous blood from HIV seronegative normal donors collected cells were washed twice in CCM and the pellet was (Central Blood Bank, Pittsburgh, PA). Isolated cells were resuspended in 100 pJ of medium. A volume of 50 to 100 p.1 of

washed twice with RPMI-1640 medium (GIBCO, Grand Island, NY) and counted in a hemocytometer in the presence of trypan

blue dye. The cells were suspended at a concentration of 1 x 107/ml in complete culture medium (CCM) containing RPMI-1640, 2 mM L-glutamine, 100 U/ml penicillin, 100 p,g/ml of streptomycin (GIBCO), and 10% (v/v) heat-inactivated AB human serum (NABI, Miami, FL). Samples of heparinized venous blood were typed for major histocompatibility (MHC) antigens by the Tissue Typing Labo-

ratory, CLSI, Presbyterian University Hospital. A-LAK cell cultures Downloaded by Pierre Marie Curie University from www.liebertpub.com at 08/28/18. For personal use only.

Cultures of A-LAK cells

were

prepared

as

previously

described.1314 Briefly, PBMN cells in CCM were passed through a 30 cm3 Nylon wool column to remove monocytes. The Nylon wool columns were incubated at 37°C in 5% C02 for 24 h,

after which the nonadherent cells were eluted with excess medium, centrifuged, and resuspended at a cell concentration of 107/ml in CCM with 1,000 U/ml rIL-2. The cell suspension was incubated in Costar plastic tissue culture flasks for 24 h, the nonadherent cells were decanted, and the adherent cells were washed with warm RPMI-1640. The 24 h conditioned medium was separated from the nonadherent cells by centrifugation and added back to the adherent cells. A-LAK cells were cultured for 12-15 days, and the cells which remained adherent were collected from the cultures by washing with cold serum-free medium and dislodging with a rubber policeman. The detached cells were combined with the nonadherent culture phase and washed three times in medium.

51Cr (as sodium chromate in saline, NEN, Boston, MA)

Tumor cells Tumor lines K562 and Daudi were maintained in RPMI-1640 with 10% FCS and subcultured as needed. Cells in the log phases of growth were used for cytotoxicity assays. For cold target inhibition assays, the tumor cells were washed three times in fresh medium prior to use.

Cytotoxicity assays Adherent LAK cells were tested for cytotoxicity activity against HIV-infected monocytes (with either IIIB or monocytotropic strain virus) or against tumor target cells. Four-hour 51Cr release assays were performed with labeled cells at a concentration of 5 x 104/ml. Target cell suspensions were added at 100 u-l/well to 96-well U-bottom plates (Costar) in triplicate. Effector cells were added to the targets at ratios ranging from 50:1 to 1-5:1. During the assay, the plates were incubated at 37°C with 5% C02. Total 51Cr release was determined by the addition of

5% Triton X-100 and medium was added to the control wells to determine the amount of spontaneous release. Supernatants were collected from each well of the test plates and counted in a gamma counter. Specific lysis was determined according to the formula:

Monocytes obtained from Ficoll-hypaque isolated Monocytes PBMN cells by countercurrent elutriation using a Beckman elutriation system (Beckman Instruments, Palo Alto, CA) as previously described.14 Elutriated monocytes were resuspended in CCM and incubated for 18 h in plastic culture flasks (Costar) at 37°C in 5% C02. Nonadherent cells in CCM were decanted from the flask after incubation. The remaining adherent cells were detached from the flask by using 0.2% EDTA in phosphate-buffered saline (PBS) at 4°C for 15 min and were then gently dislodged with a cell scraper. The detached cells were washed three times in CCM and resuspended in fresh medium to a concentration of 1 x 10s cells/ml with 1 x 106 reverse transcriptase units of HIV. Either HIV-IIIB passaged in the H9 cell line or a monocytotropic strain of HIV (kindly donated by Dr. Howard Gendelman, Walter Reed Army Medical Center, Washington, DC), was used to infect the monocytes. Following 90 minutes of incubation at 37°C, the infected cells were diluted with 5 ml of CCM and placed into tissue culture flasks for further incubation. Samples of noninfected monocytes were prepared in the same manner and incubated under identical conditions. The purity of the monocyte preparations was checked by Wright's stain and flow cytometry with LeuMl monoclonal antibody. At the termination of the cultures, cells were collected by decanting the medium and adding 0.2% EDTA in PBS at 4°C. The adherent cells could then be gently dislodged mechanically. The

was

added to the cell suspension and incubated for 1 h at 37°C. The labeled cells were washed three times with medium and resuspended in RPMI-1640 with 10% fetal calf serum (FCS, GIBCO) to a concentration of 5 x 104/ml and immediately used in cytotoxicity assays.

(Experimental mean cpm Spontaneous release cpm)

were

—

(Maximum —

Spontaneous

Lytic et

units

were

X 100= %

cpm release cpm)

mean

calculated

according

Specific lysis

to the method

of Pross

al.19 Spontaneous release of 51Cr from labeled targets was in

the range 15-30%. Relative sensitivity of macrophage target cells to cytotoxic activity of effector cells was determined by the formula:

LU20 monocyte target LU20 Daudi target

X 1000

=

Relative units (RU)

compensated for daily variation in A-LAK cytotoxic activity by indexing the cytotoxic sensitivity of the macrophage targets to tumor cells.

This calculation

Cold target inhibition assay Tumor cells, either K562 or Daudi

were washed and resusin RPMI-1640 with 10% FCS at concentrations of 5 x 105andl x 106/ml. A volume of 10 p.1 of unlabeled tumor cell suspension was added to each well of the assay plate, along with the effector cells and 51Cr-labeled target cells, in order to achieve a 1:1 or 2:1 unlabeled (cold) to labeled (hot) ratio.

pended

1013

NK ACTIVITY AGAINST HIV-INFECTED MONOCYTES Table 1. Cytotoxic Activity of A-LAK Cells Against HIV-Infected Monocytes Tumor Cell Targets

and

A-LAK Cytotoxicity

Time

IIB

postinfection

Downloaded by Pierre Marie Curie University from www.liebertpub.com at 08/28/18. For personal use only.

24 h 3 days 7 days

Monocyte strain-infected

Noninfected

monocytes

monocytes

monocytes

Daudi

25a, 129" 41; 83 56; 78

14; 71 181; 364 60; 83

6; 30 4; 7 0;0

NAC; 5222 NA; 2008 NA; 1389

strain-infected

A-LAK cells cultured in rIL-2 for 14 days were tested for cytotoxicity activity against autologous monocytes infected with either HIV-1 strain IIIB or a monocytic strain at various times postinfection. The results are representative of three separate experiments. The variation among triplicates in the assay was less than 15%. Cytotoxicity data are presented in both lytic units (LU) and relative units (RU), which represent the ratio of LU with monocyte targets to LU with tumor targets, multiplied by 1,000 (scaling constant). Noninfected autologous were used as a control for nonviral-associated killing. monocytes a Relative units of cytotoxicity. bLU20/107 cells. CNA not applicable. =

Flow cytometry

Staining of A-LAK cells and monocyte targets was performed previously described.13 A-LAK were stained with fluorescein or phycoerythrin-labeled monoclonal antibodies Leul9, Leul 1, as

Leu4, and LeuMl (Becton Dickinson, Mountain View, CA) and

analyzed by two-color flow cytometry using a FACStar fluorocytometer. Monocytes were stained directly with LeuMl antibody or indirectly with a monoclonal anti-gpl20 (NIH reference reagent) and a fluorescein conjugated goat-antimouse Fab fragment (TAGO, Burlingame, CA) diluted 1:20, then examined microscopically and by flow cytometry. RESULTS Adherent LAK cells were phenotyped to determine their surface marker profile. All A-LAK preparations used in this study contained from 67% to 95% of CD56+ (Leu 19), 32 to

51%ofCD16+(Leulla),3to34%ofCD3+(Leu4),and0%of

Table 2. Autologous and

Donor

Infected

and

CD 15+ (Leu Ml). Morphologically, A-LAK cells were 7095% large granular lymphocytes with small eosinophilic granules. Isolated monocytes were 92% CD15+ by flow cytometry. Morphologically, the cells showed typically monocytic staining with Wright's stain and had ¡S 5% contaminating small lymphocytes. Monocytes stained with anti-gpl20 antibody after 7 days infection were 32% positive for gpl20 by flow cytometric analysis, although cells showed only very weak or no staining by direct microscopic observation. The cytotoxic activity of A-LAK cells against infected and noninfected macrophage targets, as well as a tumor cell target is shown in Table 1. The absolute cytotoxic activity of the A-LAK cells varied considerably over the 7-day culture interval, requiring normalization of the data in order to evaluate the relative sensitivity of the monocyte targets to lysis at different culture intervals. This was accomplished by comparing the degree of lysis of the monocyte targets to the Daudi target cells (assuming that the sensitivity of the Daudi cells to lysis remains constant over the 7-day culture period). The results indicated that al-

Allogeneic Killing of Infected

Noninfected Monocytes

Lytic units20/107 celb Noninfected

monocytes

monocytes

K562

Daudi

416 110 306

183 41 119

5120 3531 4723

3574 2180 3222

Cytotoxic activity of autologous (Donor 3) and allogeneic (Donors 1 and 2) A-LAK cells against HIV-infected and noninfected monocytes from Donor 3. Cultures of 15-day-old A-LAK cells were tested for cytotoxic activity against monocytes from unrelated, HIV seronegative donors which had been incubated for 7 days with HIV-1 (IIIB strain). These results are representative of three separate experiments. Variation of triplicate samples was less than 15%.

MELDER ET AL.

1014

Table 3. Cold Target Inhibition of HIV-Infected Monocytes by Tumor Cell Lines

Cold

target cells

Ratio

None K562

0:1 1:1 2:1 1:1 2:1

Daudi

Lytic unitS2o/107 cells (% inhibition) labeled target cells Infected Noninfected K562 Monocytes Monocytes 83(0) (36) (63) (28) (49)

53 31 60 42

0(0) ND ND ND ND

2620 (0) 1692 (35) 1010 (62) ND ND

Daudi

1389(0) ND ND 489 (65)

318(77)

Monocytes infected with the HIV-1 monocytes strain and cultured for 5 days were used as targets for 14 day-old autologous A-LAK cells. Unlabeled tumor target cells, K562 or Daudi, were added to the assay system at 1:1 and 2:1 ratios to the infected monocyte targets. The results have been expressed both as lytic units and as percent change in cytotoxicity compared to untreated control samples and are representative of three experiments performed (ND not =

Downloaded by Pierre Marie Curie University from www.liebertpub.com at 08/28/18. For personal use only.

done).

though killing of the infected monocyte targets may occur 24 h after exposure to the virus, optimum killing was achieved 3 to 7 days after infection. During this time interval, killing of the noninfected monocytes was low; however, beyond 7 days in culture, the lysis of uninfected monocytes by A-LAK cells increased, making it difficult to evaluate the effects of HIV

recognition of cells infected by cytomegalovirus by cytotoxic lymphocytes, despite a lack of ability to demonstrate infection using an antibody against the late antigen of CMV.20 Thus, the early recognition of virus-infected targets may be a characteristic of cytotoxicity by NK cells. In addition to viral infection, increased sensitivity of monocytes to lysis by rIL-2 stimulated infection. cells has been observed to occur with prolonged culture or Lysis of HIV-IIIB-infected monocytes could also be observed activation with granulocyte-macrophage colony-stimulating with allogeneic A-LAK cells (Table 2). The level of cytotoxic factor.21 This supports the possibility that a nonviral, endogeactivity observed did not correspond to the number of shared nous cellular antigen may be up-modulated in response to MHC class 1 or class 2 loci, but did parallel the level of cytotoxic exposure to HIV, thus rendering the monocyte/macrophage activity directed against the tumor target cells. Donor 3 shared 3 sensitive to lysis. In such a scenario, productive infection of the

MHC loci with Donor 1 (2 class I, 1 class II) and 4 loci with Donor 2 (2 class I and 2 class II). Similar results were observed when the monocytotropic strain of HIV was used (data not

shown).

Cold-target competition assays using both K562 and Daudi cells indicated shared specificities between HIV-infected monocytes and tumor cells (Table 3). Both K562 and Daudi cells were able to compete with HIV-infected monocytes for lysis by A-LAK cells in a dose-dependent manner. Similar degrees of blocking could be observed when targets were used to block labeled tumor cell killing. DISCUSSION The role of the monocyte/macrophage in the pathogenesis of AIDS has become an issue of great interest. HIV may be recovered from the monocytes of AIDS patients,16,18 although only a relatively small proportion of these cells appears to harbor HIV by immunofluorescent techniques.5,18 Such findings are in agreement with the weak staining of short-term infected monocytes with anti-gpl20 antibodies in this study. A-LAK cells, however, were able to distinguish HIV-exposed from nonexposed control cells very early after the monocytes were challenged with HIV, suggesting that either a very low concentration of viral product is required to trigger recognition of an infected cell, or that a nonviral product may be produced as a consequence of infection, which renders the infected monocyte susceptible to NK-mediated lysis. This may be analogous to the

not be needed to render the cell sensitive to NK-mediated lysis; contact with the virus may be sufficient to induce sensitivity to lysis in the target cell. Since only relatively brief incubation of infected cells following exposure to HIV is required to sensitize the cells to lysis, and both HIV-IIIB and the monocytotropic strain are able to induce these target cell changes, there is some experimental support to this model. Furthermore, only slight increases in reverse transcriptase levels were observed following incubation of monocytes with either strain of HIV (not shown). The early recognition of HIVinfected monocytes might play an important role in the control of the progress of infection, since the elimination of infected cells prior to the rapid production of virus would decrease the likelihood that the infection would be spread to the other cells or tissues. Cold-target competition assays indicate that the subset of cytotoxic cells that recognize the HIV-infected target cells are the same as those that kill tumor target cells. This finding implies that a novel subset of activated NK cells is not required to explain HIV-associated killing but that a generalized cytotoxic action of NK cells against both HIV-infected monocytes and tumor cells may occur. Adhesion-associated proteins, such as CD18, have been shown to be involved with the binding of NK cells to target cells.22 One possible mechanism for the increased sensitivity of HIV infected monocytes to NK-mediated killing may utilize the up-modulation of adhesion protein ligands, such as ICAM-1,23 although this has not yet been demonstrated to occur with HIV-infected cells. Further studies into the mechanism of recognition of HIV-infected monocytes by activated and

monocyte may

1015

NK ACTIVITY AGAINST HIV-INFECTED MONOCYTES

Downloaded by Pierre Marie Curie University from www.liebertpub.com at 08/28/18. For personal use only.

nonactivated NK cells may lead to more effective methods to direct cytotoxicity against cells actively producing virus as well as those providing a reservoir of virus following treatment with an antiviral agent. The lack of MHC restriction in the killing of HIV-infected monocytes by A-LAK cells makes it possible to measure cytotoxicity of cells from different individuals against a common allogeneic HIV-infected target cell, thus allowing a direct comparison of anti-HIV cytotoxicity among different individuals. This may be of importance in evaluating the effects of biological response modifiers of the anti-HIV cytotoxicity of lymphocytes obtained from these patients. The demonstration that IL-2-activated NK cells plays a role in the control of HIV infection lead to new therapeutic approaches to the treatment of HIV infection, as well as to greater insights into the mechanisms of the nonspecific immune response to viral infections.

ACKNOWLEDGMENT This work was supported in part by Public Health Service Contract N01-AI-72632 and the Pathology Education Research Fund of University of Pittsburgh.

REFERENCES 1. Welsh RM: Regulation of virus infections by natural killer cells. Nat Immun Cell Growth Regul 1986;5:169. 2. LopezC, Kirkpatrick D, Fitzgerald PA, Ching CY, Paha RN, Good RA, and Smithwick EM: Studies of the cell lineage of the effector cells that spontaneously lyse HSV-1 infected fibroblasts (NK (HSV-l)). I Immunol 1982;129:824. 3. Hayward AR, Herberger M, and Lazslo M: Cellular interactions in the lysis of varicella-zoster virus infected human fibroblasts. Clin Exp Immunol 1986;63:141. 4. Santoli D, Trinchieri G, and Lief F: Cell-mediated cytotoxicity against virus-infected target cells in humans. 1. Characteristics of the effector lymphocyte. I Immunol 1978;121:526. 5. Ruscetti FW, Mikovits JA, Kalyanaraman VS, Overton R, Stevenson H, Stromberg K, Herberman RB, Farrar WL, and Ortaldo JR: Analysis of effector mechanisms against HTLV-1 and HTLVIII/LAV-infected lymphoid cells. J Immunol 1986;136:3619. 6. Ojo-Amaize EA, Nashanian P, Keith DE, Houghton RL, Heitjan DF, Fahey JL, and Giogi JV: Antibodies to human immunodeficiency virus in human sera induce cell mediated lysis of human immunodeficiency virus-infected cells. J Immunol 1987;139: 2458. 7. Rappocciolo G, Toso JF, Torpey DJ III, Gupta P, and Rinaldo CR: Association of alpha interferon production with natural killer cell lysis of U937 cells infected with human immunodeficiency virus. J Clin Microbiol 1989;27:41. 8. Tyler DS, Nastala CL, Stanley SD, Mathews TJ, Lyerly HK, Bolognesi DP, and Weinhold KJ: GP120 specific cellular cytotoxicity in HIV-1 seropositive individuals. Evidence for circulating CD16+ effector cells armed in vivo with cytophilic antibody. J Immunol 1989;142:1177. 9. BrennerBG, Dascal A, Margolese RG, and Wainberg MA: Natural

killer cell function in patients with acquired immunodeficiency syndrome and related diseases. J Leuk Biol 1989;46:75. 10. Lew F, Tsang P, Solomon S, Selikoff IJ, and Bekesi JG: Natural killer cell function and modulation by alpha IFN and IL2 in AIDS patients and prodromal subjects. J Clin Lab Immunol 1984;14:115. 11. Reddy MM, Chinoy P, and Grieco HM: Differential effects of interferon-alpha 2 and interleukin 2 on natural killer cell activity in patients with acquired immune deficiency syndrome. J Biol Resp Modif 1984;3:379. 12. Vujanovic NL, Herberman RB, Maghazachi AA, and Hiserodt JC: Lymphokine activated killer cells in rats. III. A simple method for the purification of large granular lymphocytes and their rapid expansion and conversion into lymphokine activated killer cells.

JExpMedl988;167:15.

13. Melder RJ, Whiteside TL, Vujanovick NL, Hiserodt JC, and Herberman RB: A new approach to generating antitumor effectors for adoptive immunotherapy using human adherent lymphokine activated killer cells. Cancer Res 1988;48:3461. 14. Melder RJ, Rosenfeld CS, Herberman RB, and Whiteside TL: Large-scale preparation of adherent lymphokine activated killer (A-LAK) cells for adoptive immunotherapy in man. Cancer Immunol Immunother 1989;29:67. 15. Levy JA, Shimabukuro J, McHugh T, Casavant C, Stites D, and Oshiro L: AIDS-associated retroviruses (ARV) can productively infect other cells besides human helper T cells. Virology 1986; 147:441. 16. Ho DD, Rota TR, and Hirsch MS: Infection of monocyte/ macrophages by human T lymphotropic virus type III. J Clin Invest

1986;77:1712. S, Mills J, and McGrath MS: Quantitative immunocytoflu-

17. Crowe

orographic analysis of CD4 surface antigen expression and HIV infection of human peripheral blood monocyte/macrophages. AIDS

Res Human Retroviruses 1987; 3:135. 18. Gartner S, Markovits P, Markovits DM, Kaplan MH, Gallo RC, and Popovic M: The role of mononuclear phagocytes in HTLVIII/LAV infection. Science 1986;223:215. 19. Pross HF, Baines NG, Rubin P, Shragge P, and Patterson MS: Spontaneous human lymphocyte-mediated cytotoxicity against tumor target cells. IX. The quantitation of natural killer cell activity. J Clin Immunol 1981;1:51. 20. Bandyopadhyay S, Oh SH, Michelson S, Miller DS, Virelizier JL, and Starr SE: Natural killing of fibroblasts infected with lowpassage clinical isolates of human cytomegalovirus. Clin Exp Immunol 1988;73:11. 21. Djeu JY, Widen R, and Blanchard DK: Susceptibility of monocytes to lymphokine-activated killer cell lysis: effect of granulocytemacrophage colony-stimulating factor and interleukin 3. Blood

1989;73:1264. 22. Timonen T, Patarroyo M, and Gahmberg CG: CDlla-c/CD18 GP84 (LB-2) adhesion molecules on human large granular lymphocytes and their participation in natural killing. J Immunol 1986; 141:1041. 23. Makgoba MW, Sanders ME, Luce GE, Dustin M, Springer TA, Clark EA, Mannoni P, and Shaw S: ICAM-1, a ligand for LFA1-dependent adhesion of B, T and myeloid cells. Nature 1988; 331:86.

Address

reprint requests

to:

Robert J. Melder Department of Chemical Engineering Carnegie Mellon University Pittsburgh, PA 15213-3896