Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 6, Number 8, 1990 Mary Ann Liebert, Inc., Publishers
Full-Length Recombinant CD4 and Recombinant gpl20 Inhibit Fusion Between HIV Infected Macrophages and Uninfected CD4-Expressing T-Lymphoblastoid Cells S.M.
CROWE,1,2 J. MILLS,2 J. KIRIHARA,2 J. BOOTHMAN,1
J.A.
MARSHALL,1
and M.S.
McGRATH2
ABSTRACT Human immunodeficiency virus- (HIV) infected monocyte-macrophages may contribute to the pathogenesis of HIV-associated immune deficiency and dysfunction by acting as a target and potential reservoir for the virus in vivo, and by functioning abnormally following infection. We have shown that HIV-infected macrophages fuse with uninfected CD4-expressing lymphoid cells in vitro; this may provide an additional mechanism for CD4 lymphocyte depletion in vivo. We report here the inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing T-lymphoid cells by monoclonal antibody S3.5, directed against an epitope of CD4 involved in binding HIV gpl20, by a recombinant protein that comprises the full-length extracellular domain of the CD4 molecule, and by recombinant full-length HIV envelope glycoprotein, gpl20. These results indicate that both molecules (gpl20 and CD4) are critical to the fusion process, and suggest that gpl20 is expressed on the surface of HIV-infected monocyte-macrophages.
INTRODUCTION
THE increasing
system, and the investigators have found functional impairment HIV infection, including altered phagocytosis, chemotaxis, and cytokine production.1113 HIV-infected lym-
following
PATHOGENESIS OF
ACQUIRED IMMUNODEFICIENCY
SYN-
(AIDS) remains incompletely understood, with evidence that human immunodeficiency virus (HIV)
DROME
infection of CD4-expressing lymphocytes alone is insufficient to explain all of the abnormal immune phenomena. There is an increase in virus production, as detected by plasma viremia and serum p24 antigen production, which occurs when the CD4 lymphocyte number is significantly depleted. ' 2 The role played by HIV-infected macrophages requires further elucidation, although these cells have been implicated in the development of the immune deficiency and dysfunction associated with HIV infection. Monocyte-macrophages express the CD4 molecule on their surface3,4 and can be infected with HIV in vitro and in vivo.4-9 HIV infection of CD4-expressing lymphocytes results in rapid formation of syncytia with subsequent cell death.10 There is much less cytopathology associated with HIV infection of monocyte-macrophages in vitro, and infected cells survive for at least several months with continuing virion production,4 suggesting that these cells provide a reservoir of the virus in vivo. The macrophage occupies a central position within the immune
phoid cells can fuse with other CD4-expressing lymphoid cells form giant syncytia resulting in cellular death in vitro.10 We have found that HIV-infected macrophages fuse with uninfected CD4-expressing cells, thus providing an additional mechanism for CD4 lymphocyte depletion in vivo.14 This report addresses the ability of several proteins to inhibit this in vitro fusion process, including a monoclonal antibody directed against CD4, a recombinant protein representing the full-length extracellular to
domain of the CD4 molecule, recombinant soluble CD4 (rsCD4), and recombinant protein representing the full-length, glycosylated HIV envelope glycoprotein gpl20.
MATERIALS AND METHODS
Preparation of monocytes Monocytes were isolated from the buffy coats of HIV seronegative blood donors (provided by Stanford Blood Bank, Palo Alto, CA, and Red Cross Blood Bank, South Melbourne,
Macfarlane Burnet Centre for Medical Research, Fairfield Hospital, Melbourne, Australia. 2San Francisco General Hospital, San Francisco, CA 94110.
1031
1032
CROWE ET AL.
in
Table 1. Neutralization of HIV-DV Infectfvity by Recombinant Soluble CD4
Monocyte-Macrophages
Experiment
Supernatant p24 antigen (pglml)
Additive NIL
62960 3976 72130 4070 904100 17165
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
rsCD4 NIL rsCD4 NIL
rsCD4
% Inhibition 93.7 94.4 98.0
Macrophages were infected with HIV-DV at a multiplicity of approximately one, and cultured in the presence or absence of 20 aglml of rsCD4 for 10 days. HIV p24 antigen in culture supernatants was assessed by Abbott electron immunoassay.
as previously described.4 Briefly, peripheral blood mononuclear cells were collected by Ficoll-hypaque (Pharmacia Fine Chemicals, Carlton, Victoria, Australia) gradient centrifugation, washed, and purified by glass adherence in the presence of 20% fetal calf serum (Flow Laboratories, North Ryde, NSW, Australia) in calcium- and magnesium-free phosphate-buffered saline (Flow Laboratories). Adherent monocytes were gently scraped from the petri dishes using a disposable cell scraper (Costar, Cambridge, MA), washed, and suspended in RPMI1640,10% human AB 4- serum (Gibco, Glen Waverly, Victoria, Australia) 50 p.g/ml gentamicin and 2 mM glutamine (Flow Laboratories, Irvine, Scotland). Monocytes were distributed to
Australia)
perfluoralkoxy (similar to Teflon) jars (Savillex, Minnetonka, MN)
at a concentration of 1 X
106 cells/ml. Media
were
changed weekly. HIV
infection
Monocyte-macrophages were infected at a multiplicity of infection of approximately one as previously described4 using a stock of HIV-DV strain for all infections. This isolate infected both monocyte-macrophages and T-lymphoblastoid cells equally well. Virus stocks were prepared in the CD4-expressing T-lymphoma cell line, VB (kindly provided by Dr. J. Lifson, GeneLabs, Redwood City, CA). Stocks of HIV were stored in small volumes at -70°C, and thawed immediately prior to use (Table 1).
Quantitation of HIV infection HIV infection was quantitated by immunofluorescent staining
using HIV p24 monoclonal antibody (Olympus, Lake Success, NY); an irrelevant monoclonal antibody of same class and isotype, the mouse myeloma protein MOPC21 (Litton Bionetics, MD) served as the control. The proportion of infected cells was quantitated by flow cytometry (FACSStar Plus, Becton Dickinson, Mountain View, CA). Infection was confirmed by assaying culture supernatants for HIV p24 antigen using the
Abbott HIVAg-1 enzyme immunoassay and p24 antigen quantitation panel (Abbott Laboratories, North Chicago, IL).
Fusion inhibition assays
Varying concentrations of monoclonal antibody directed against CD4 *S3.5, kindly provided by Dr. E. Engelman,
Stanford Blood Bank) were incubated with uninfected VB cells for 30 minutes at 37°C. These cells were then incubated with infected macrophages at a ration of 1:1, following restoration of the original concentration of anti-CD4 monoclonal antibody, at 37°C for 48 h in Labtek slide chambers. Cells were examined under phase contrast microscopy (Olympus CK2). Syncytia were scored in one of 20 low-power fields for each culture, with multinucleated giant cells being defined as having four or more nuclei contained within a common membrane. Experiments were performed in parallel in which VB cells were incubated with an irrelevant monoclonal antibody of same class and isotype prior to addition of infected macrophages. Further controls consisted of H9 cells chronically infected with HIVHTLV-HIB (HXB, kindly provided by Dr. R. Gallo) cultured with uninfected VB cells as a positive control, and uninfected macrophages incubated with uninfected VB cells as a negative control. Recombinant soluble CD4 (rsCD4) (kindly provided by Dr. R. Sweet, Smith Kline/Beecham, King of Prussia, PA) was incubated with HIV-infected macrophages in varying concentrations for 30 min at 37°C. These cells were then mixed with uninfected VB cells, the original concentration of rCD4 was restored, and experimental protocol was performed as above. Controls to which no protein was added were included. Recombinant full-length gp 120 (rgp 120) prepared in a Drosophila cell line (kindly provided by Dr. C. deBouck, Smith Kline/Beecham, King of Prussia, PA) was incubated at varying concentrations with uninfected VB cells for 30 min at 37CC. Cells were then incubated with HIV-infected macrophages, and the original concentration of rgp 120 was restored. The fusion assay was performed as outlined above. Recombinant protein
1033
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
INHIBITION OF HIV FUSION
representing the N terminal one third of HIV gpl20 (kindly HIV infection provided by Dr. C. deBouck, Smith Kline/Beecham) was HIV-infected macrophage cultures were assayed for HIV p24 included as an additional control. antigen expression by immunocytofluorography 9-14 days after infection. Cultures in which greater than 25% of macrophages Electron microscopy expressed HIV p24 were used in fusion experiments. HIV p24 expression of cultures used were in the range 26-74% Representative samples were selected for electron micros- antigen (mean 39%). copy. First, 2 x 106 cells were fixed with 2% glutaraldehyde/ 0.1 M Sorensen phosphate buffer, pH 7.3 for 2 h at 4°C. Following overnight rinsing in buffer, cells were further treated Inhibition offusion by S3.5 with 1 % osmium tetroxide in Sorensen buffer for 1 h at 4°C prior S3.5 is a monoclonal antibody directed against an epitope of to embedding in LX112 resin (Ladd Research Industries, BurlCD4 shown to be essential for fusion between HIV-infected and ington, VT). Specimens were examined using a Philips EM301 uninfected CD4-expressing lymphocytes.13 S3.5 inhibited fuelectron microscope. sion between HIV-infected macrophages and uninfected lymphoid cells in a dose-dependent fashion (Fig. 1), with 20 u-g/ml RESULTS
Purity of monocyte-macrophage cultures Using LeuM3 and LeuM5 monoclonal antibodies for immunofluorescent staining and flow cytometric analysis, as well as phagocytosis of Candida albicans, the purity of macrophage cultures was in the range 85-95%. Antibodies containing T lymphocytes (representing less than 5% of the cultures) were removed by treatment with OKT3 and complement.
% Inhibition of 100
resulting in > 99% inhibition. MOPC21, an irrelevant monoclonal antibody of same class and isotype used at the same concentration, did not inhibit fusion. No toxicity was observed with either antibody (Fig. 2).
Inhibition
offusion by recombinant gpllO
Full-length rgpl20 inhibited fusion in a dose-dependent man(Fig. 3), with 10 u.g/ml resulting in complete inhibition of fusion, with no effect on cell viability as assessed by Trypan blue ner
syncytltim formation
80
60
40
20
50
20
0.1
10
recombinant soluble CD4
(ug/ml)
FIG. 1. Inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing lymphoblastoid cells by rsCD4. Controls were cultures to which no recombinant protein was added. Mean percentage inhibition is given for a minimum of four experiments.
CROWE ET AL.
1034
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
100
% inhibition oí syncytium formation
control
s3.5 monoclonal antibody
(ug/ml)
FIG. 2. Inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing T-lymphoblastoid cells (VB) by monoclonal antibody S3.5 directed against CD4. Mean percentage inhibition for six experiments is given. Control cultures to which an irrelevant monoclonal antibody of the same class and isotype (MOPC21) was added at 20 u-g/ml showed no inhibition of fusion.
% inhibition 100
10
rgpl20
1 rgpl20
recombinant
0.1
rgpl20
gpl20 (ug/ml)
10
rgpl20-N
FIG. 3. Inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing T-lymphoblastoid cells (VB) by full-length recombinant gpl20 (rgpl20) and recombinant N terminal gpl20 (rgpl20-N). Controls were cultures to which no recombinant protein was added. The mean percentage inhibition for three experiments is given.
INHIBITION OF HIV FUSION
1035
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
>
í>'w,%
W\?~
HP •
:%
FIG. 4.
*TS'' :*
„
»'
(A) Electron micrograph depicting early fusion between an HIV-infected macrophage and lymphoblastoid cell, showing a region of loss of individual cell membranes ( x 7000). (B) At high magnification HIV can be seen budding from the macrophage surface (x 34,000). (C) The majority of retroviral particles are contained within intracytoplasmic vacuoles (x 15,000).
1036
CROWE ET AL.
exclusion method. Fusion was not inhibited by a recombinant protein representing the N-terminal on third of gpl20.
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
Electron
microscopy
When cultures showing extensive fusion between uninfected VB cells and HIV-infected macrophages were examined by electron microscopy, fusion of cells with loss of cell membranes separating macrophages and lymphoid cells was observed. Typical retroviral particles were visualized within vesicles within the multinucleated giant cells, and, less commonly, budding from the macrophage surface (Fig. 4).
ACKNOWLEDGMENTS The authors would like to acknowledge the expert technical assistance of Nancy McManus, Anne Col vin, and loann Vennari; Patricia Lekas for flow cytometric analyses, and Steven DiTomaso for preparation of the manuscript. This study was supported in part by grants from Grant AI25329 to Dr. John Mills from the U.S. NIH; and Commonwealth of Australia AIDS Research Center, NH&MRC, and the Cancer Research Institute of New York (SC).
REFERENCES AE, Phelan MA, Wells MA, Vujcic LK, Epstein JS, Lane HC, Quinnan GV: Detection of human immunodeficiency virus
1. Wittek
DISCUSSION HIV-infected macrophages can fuse with uninfected CD4expressing lymphoid cells, and this fusion may provide an additional explanation for the inexorable depletion of CD4expressing T lymphocytes in vivo. The demonstration of multinucleated giant cells is unlikely to be due to a phagocytic process, as fusion of cellular membranes can be demonstrated by electron microscopy, and the process can be specifically inhibited by blocking CD4-gpl20 interactions between the two cell populations by recombinant proteins (rCD4 and rgp 120) and a monoclonal antibody directed against the region of CD4 involved in HIV binding (@3.5). This proves that the CD4 molecule and gpl 20 are essential for the fusion process between these cell populations, but does not exclude the possibility of other molecules playing a role in this process. Although multinucleated giant cell formation is commonly
encountered in HIV-infected cultures in vitro, the relevance of this phenomenon in vivo is unclear. Much evidence supports the in vivo relevance of fusion in the pathogenesis of HIV infection. Multinucleated giant cells have been reported in brains1517 and lymph nodes18 of HIV-infected individuals. In patients with HIV-related subacute encephalopathy, histochemical staining indicates that these syncytia are composed of macrophages, and the presence of HIV within these cells has been documented.19 Similar multinucleated giant cells, composed of HIV-infected macrophages, have also been demonstrated within the spinal cords of patients with myelopathy.20 Thus our observations relating to syncytium formation between HIV-infected macrophages and uninfected CD4-expressing lymphoid cells, as well as the demonstration of inhibition of multinucleated giant cell formation by recombinant CD4 and gpl20 proteins are likely to be clinically relevant. There is currently no antiretroviral agent which will eradicate HIV, and the drugs in use clinically (reverse transcriptase inhibitors) are ineffective in chronically infected cells21 and only marginally effective in patients. To improve efficacy we must develop therapy targeted toward other sites in the replicative cycle of HIV. Compounds inhibiting attachment of virus cells and cell-cell fusion such as soluble CD4 are likely to be of most benefit when combined with a compound acting later in the cycle, such as reverse transcriptase inhibitors.
core
protein
in
plasma by
enzyme
immunoassay.
Ann Int Med
1987;107:286-292. 2. Ho DD, Moudgil T, and Alam M: Quantitation of human immunodeficiency virus type 1 in the blood of infected persons. N Engl J Med 1989;321:1621-1625. 3. Stewart SJ, Fujimoyo J, and Levy R: Human T lymphocytes and monocytes bear the same leu3 (T4) antigen. J Immunol 1986; 136:3773-3778. 4. Crowe SM, Mills J, and McGrath MS: Quantitative immunocytofluorographic analysis of CD4 surface antigen expression and HIV infection in human peripheral blood monocyte-macrophages. AIDS Res Human Retroviruses 1987;3:135-145. 5. Gartner S, Markovits P, Markovitz DM, Kaplan MH, Gallo RC, Popovic M: The role of mononuclear phagocytes in HTLVIII/LAV infection. Science 1986; 233:215-219. 6. Ho DD, Rota TR, and Hirsch MS: Infection of monocyte-macrophages by human T lymphotropic virus type III. J Clin Invest 1986;
77:1712-1715. 7. Nicholson JKA, Cross GD, Callaway CS, McDougal JS: In vitro infection of human monocytes with human T lymphotropic virus type III/lymphadenopathy associated virus (HTLVIII/LAV). J Immunol 1986;137:323-329. 8. Salahuddin SZ, Rose RM, Groopman JE, Markham PD, Gallo RC: Human T lymphotropic virus type III infection of human alveolar macrophages. Blood 1986;68:281-284. 9. Perno CF, Yarchoan R, Cooney DA, Hartman NR, Webb DS, Hao Z, Mitsuya H, Johns DG, Broder S: Replication of human immunodeficiency virus in monocytes. J Exp Med 1989;169:933939. 10. Lifson JD, Feinberg MB, Reyes G: AIDS retrovirus-induced cytopathology: giant cell formation and involvement of CD4 antigen. Science 1986;234:1123-1127. 11. Locksley RM, Crowe S, Sadick MD, Heinzel FP, Gardner KD, McGrath MS, Mills J: Release of interleukin-1 inhibitory activity (contra IL-1) by human monocyte-derived macrophages infected with human immunodeficiency virus in vitro and in vivo. J Clin Invest 1988;82:2097-2105. 12. Smith PD, Ohura K, Masur H, et al: Monocyte function in the acquired immune deficiency syndrome. J Clin Invest 1984; 74:2121-2125. 13. Tas M, Drexhage HA, and Goudsmit J: A monocyte chemotaxis inhibiting factor in serum of HIV infected men shares epitopes with the HIV transmembrane protein gp41. J Exp Immunol 1988; 71:13-18. 14. Crowe SM, Mills J, and McGrath MS: Removal of CD4 positive lymphocytes through a cell fusion process. Ill International Conference on AIDS, Washington, Abstract #F9.3 (1987).
1037
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
INHIBITION OF HIV FUSION 15. Sharer LR, Eun-Sook C, and Epstein LG: Multinucleated giant cells and HTLV-III in AIDS encephalopathy. Human Pathol 1985; 16:760. 16. Koenig S, Gendelman H, Orenstein JM, DalCanto MC, PezeshkpourGH, Yungbluth M, Janotta F, Aksamit A, Martin MA, Fauci AS: Detection of AIDS virus in macrophages of brain tissues from AIDS patients with encephalopathy. Science 1986;233:10891093. 17. Popovic M, Meliert W, Eifle V, Gartner S: Role of mononuclear phagocytes and accessory cells in human immunodeficiency virus type 1 infection of the brain. AnnNeurol 1988;23(suppl):s74-s77. 18. Byrnes RK, ChanWC, SpiraTJ, etal: Value of lymph node biopsy in unexplained lymphadenopathy in homosexual medicine. JAMA
1983;250:1313-1315.
19. Gartner S, Markovits P, Markovitz D, Betts RF, Popovic M: Virus isolation from and identification of HTLV-III/LAV-producing cells in brain tissue from a patient with AIDS. JAMA 1986:256:23652371.
20. Eilbott DJ, Peress N, Burger H, Weiser B, LaNeve D, Seidman R: Human immunodeficiency virus type-1 in spinal cords of acquired immunodeficiency syndrorne patients with myelopathy: expression and replication in macrophages. Proc Nati Acad Sei (USA)
1989;86:3337-3341.
21. Crowe S, McGrath MS, Elbeik T, Kirihara J, Mills J: Comparative assessment of antiretrovirals in human monocyte-macrophages and lymphoid cell lines acutely and chronically infected with the human immunodeficiency virus. J Med Virol 1989;29:176-180.
Address
reprint requests to:
Suzanne Crowe
Head, AIDS Research Unit
Macfarlane Burnet Centre for Medical Research
Yarra Bend Road
Fairfield, Australia 3078
AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 6, Number 8, 1990 Mary Ann Liebert, Inc., Publishers
Full-Length Recombinant CD4 and Recombinant gpl20 Inhibit Fusion Between HIV Infected Macrophages and Uninfected CD4-Expressing T-Lymphoblastoid Cells S.M.
CROWE,1,2 J. MILLS,2 J. KIRIHARA,2 J. BOOTHMAN,1
J.A.
MARSHALL,1
and M.S.
McGRATH2
ABSTRACT Human immunodeficiency virus- (HIV) infected monocyte-macrophages may contribute to the pathogenesis of HIV-associated immune deficiency and dysfunction by acting as a target and potential reservoir for the virus in vivo, and by functioning abnormally following infection. We have shown that HIV-infected macrophages fuse with uninfected CD4-expressing lymphoid cells in vitro; this may provide an additional mechanism for CD4 lymphocyte depletion in vivo. We report here the inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing T-lymphoid cells by monoclonal antibody S3.5, directed against an epitope of CD4 involved in binding HIV gpl20, by a recombinant protein that comprises the full-length extracellular domain of the CD4 molecule, and by recombinant full-length HIV envelope glycoprotein, gpl20. These results indicate that both molecules (gpl20 and CD4) are critical to the fusion process, and suggest that gpl20 is expressed on the surface of HIV-infected monocyte-macrophages.
INTRODUCTION
THE increasing
system, and the investigators have found functional impairment HIV infection, including altered phagocytosis, chemotaxis, and cytokine production.1113 HIV-infected lym-
following
PATHOGENESIS OF
ACQUIRED IMMUNODEFICIENCY
SYN-
(AIDS) remains incompletely understood, with evidence that human immunodeficiency virus (HIV)
DROME
infection of CD4-expressing lymphocytes alone is insufficient to explain all of the abnormal immune phenomena. There is an increase in virus production, as detected by plasma viremia and serum p24 antigen production, which occurs when the CD4 lymphocyte number is significantly depleted. ' 2 The role played by HIV-infected macrophages requires further elucidation, although these cells have been implicated in the development of the immune deficiency and dysfunction associated with HIV infection. Monocyte-macrophages express the CD4 molecule on their surface3,4 and can be infected with HIV in vitro and in vivo.4-9 HIV infection of CD4-expressing lymphocytes results in rapid formation of syncytia with subsequent cell death.10 There is much less cytopathology associated with HIV infection of monocyte-macrophages in vitro, and infected cells survive for at least several months with continuing virion production,4 suggesting that these cells provide a reservoir of the virus in vivo. The macrophage occupies a central position within the immune
phoid cells can fuse with other CD4-expressing lymphoid cells form giant syncytia resulting in cellular death in vitro.10 We have found that HIV-infected macrophages fuse with uninfected CD4-expressing cells, thus providing an additional mechanism for CD4 lymphocyte depletion in vivo.14 This report addresses the ability of several proteins to inhibit this in vitro fusion process, including a monoclonal antibody directed against CD4, a recombinant protein representing the full-length extracellular to
domain of the CD4 molecule, recombinant soluble CD4 (rsCD4), and recombinant protein representing the full-length, glycosylated HIV envelope glycoprotein gpl20.
MATERIALS AND METHODS
Preparation of monocytes Monocytes were isolated from the buffy coats of HIV seronegative blood donors (provided by Stanford Blood Bank, Palo Alto, CA, and Red Cross Blood Bank, South Melbourne,
Macfarlane Burnet Centre for Medical Research, Fairfield Hospital, Melbourne, Australia. 2San Francisco General Hospital, San Francisco, CA 94110.
1031
1032
CROWE ET AL.
in
Table 1. Neutralization of HIV-DV Infectfvity by Recombinant Soluble CD4
Monocyte-Macrophages
Experiment
Supernatant p24 antigen (pglml)
Additive NIL
62960 3976 72130 4070 904100 17165
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
rsCD4 NIL rsCD4 NIL
rsCD4
% Inhibition 93.7 94.4 98.0
Macrophages were infected with HIV-DV at a multiplicity of approximately one, and cultured in the presence or absence of 20 aglml of rsCD4 for 10 days. HIV p24 antigen in culture supernatants was assessed by Abbott electron immunoassay.
as previously described.4 Briefly, peripheral blood mononuclear cells were collected by Ficoll-hypaque (Pharmacia Fine Chemicals, Carlton, Victoria, Australia) gradient centrifugation, washed, and purified by glass adherence in the presence of 20% fetal calf serum (Flow Laboratories, North Ryde, NSW, Australia) in calcium- and magnesium-free phosphate-buffered saline (Flow Laboratories). Adherent monocytes were gently scraped from the petri dishes using a disposable cell scraper (Costar, Cambridge, MA), washed, and suspended in RPMI1640,10% human AB 4- serum (Gibco, Glen Waverly, Victoria, Australia) 50 p.g/ml gentamicin and 2 mM glutamine (Flow Laboratories, Irvine, Scotland). Monocytes were distributed to
Australia)
perfluoralkoxy (similar to Teflon) jars (Savillex, Minnetonka, MN)
at a concentration of 1 X
106 cells/ml. Media
were
changed weekly. HIV
infection
Monocyte-macrophages were infected at a multiplicity of infection of approximately one as previously described4 using a stock of HIV-DV strain for all infections. This isolate infected both monocyte-macrophages and T-lymphoblastoid cells equally well. Virus stocks were prepared in the CD4-expressing T-lymphoma cell line, VB (kindly provided by Dr. J. Lifson, GeneLabs, Redwood City, CA). Stocks of HIV were stored in small volumes at -70°C, and thawed immediately prior to use (Table 1).
Quantitation of HIV infection HIV infection was quantitated by immunofluorescent staining
using HIV p24 monoclonal antibody (Olympus, Lake Success, NY); an irrelevant monoclonal antibody of same class and isotype, the mouse myeloma protein MOPC21 (Litton Bionetics, MD) served as the control. The proportion of infected cells was quantitated by flow cytometry (FACSStar Plus, Becton Dickinson, Mountain View, CA). Infection was confirmed by assaying culture supernatants for HIV p24 antigen using the
Abbott HIVAg-1 enzyme immunoassay and p24 antigen quantitation panel (Abbott Laboratories, North Chicago, IL).
Fusion inhibition assays
Varying concentrations of monoclonal antibody directed against CD4 *S3.5, kindly provided by Dr. E. Engelman,
Stanford Blood Bank) were incubated with uninfected VB cells for 30 minutes at 37°C. These cells were then incubated with infected macrophages at a ration of 1:1, following restoration of the original concentration of anti-CD4 monoclonal antibody, at 37°C for 48 h in Labtek slide chambers. Cells were examined under phase contrast microscopy (Olympus CK2). Syncytia were scored in one of 20 low-power fields for each culture, with multinucleated giant cells being defined as having four or more nuclei contained within a common membrane. Experiments were performed in parallel in which VB cells were incubated with an irrelevant monoclonal antibody of same class and isotype prior to addition of infected macrophages. Further controls consisted of H9 cells chronically infected with HIVHTLV-HIB (HXB, kindly provided by Dr. R. Gallo) cultured with uninfected VB cells as a positive control, and uninfected macrophages incubated with uninfected VB cells as a negative control. Recombinant soluble CD4 (rsCD4) (kindly provided by Dr. R. Sweet, Smith Kline/Beecham, King of Prussia, PA) was incubated with HIV-infected macrophages in varying concentrations for 30 min at 37°C. These cells were then mixed with uninfected VB cells, the original concentration of rCD4 was restored, and experimental protocol was performed as above. Controls to which no protein was added were included. Recombinant full-length gp 120 (rgp 120) prepared in a Drosophila cell line (kindly provided by Dr. C. deBouck, Smith Kline/Beecham, King of Prussia, PA) was incubated at varying concentrations with uninfected VB cells for 30 min at 37CC. Cells were then incubated with HIV-infected macrophages, and the original concentration of rgp 120 was restored. The fusion assay was performed as outlined above. Recombinant protein
1033
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
INHIBITION OF HIV FUSION
representing the N terminal one third of HIV gpl20 (kindly HIV infection provided by Dr. C. deBouck, Smith Kline/Beecham) was HIV-infected macrophage cultures were assayed for HIV p24 included as an additional control. antigen expression by immunocytofluorography 9-14 days after infection. Cultures in which greater than 25% of macrophages Electron microscopy expressed HIV p24 were used in fusion experiments. HIV p24 expression of cultures used were in the range 26-74% Representative samples were selected for electron micros- antigen (mean 39%). copy. First, 2 x 106 cells were fixed with 2% glutaraldehyde/ 0.1 M Sorensen phosphate buffer, pH 7.3 for 2 h at 4°C. Following overnight rinsing in buffer, cells were further treated Inhibition offusion by S3.5 with 1 % osmium tetroxide in Sorensen buffer for 1 h at 4°C prior S3.5 is a monoclonal antibody directed against an epitope of to embedding in LX112 resin (Ladd Research Industries, BurlCD4 shown to be essential for fusion between HIV-infected and ington, VT). Specimens were examined using a Philips EM301 uninfected CD4-expressing lymphocytes.13 S3.5 inhibited fuelectron microscope. sion between HIV-infected macrophages and uninfected lymphoid cells in a dose-dependent fashion (Fig. 1), with 20 u-g/ml RESULTS
Purity of monocyte-macrophage cultures Using LeuM3 and LeuM5 monoclonal antibodies for immunofluorescent staining and flow cytometric analysis, as well as phagocytosis of Candida albicans, the purity of macrophage cultures was in the range 85-95%. Antibodies containing T lymphocytes (representing less than 5% of the cultures) were removed by treatment with OKT3 and complement.
% Inhibition of 100
resulting in > 99% inhibition. MOPC21, an irrelevant monoclonal antibody of same class and isotype used at the same concentration, did not inhibit fusion. No toxicity was observed with either antibody (Fig. 2).
Inhibition
offusion by recombinant gpllO
Full-length rgpl20 inhibited fusion in a dose-dependent man(Fig. 3), with 10 u.g/ml resulting in complete inhibition of fusion, with no effect on cell viability as assessed by Trypan blue ner
syncytltim formation
80
60
40
20
50
20
0.1
10
recombinant soluble CD4
(ug/ml)
FIG. 1. Inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing lymphoblastoid cells by rsCD4. Controls were cultures to which no recombinant protein was added. Mean percentage inhibition is given for a minimum of four experiments.
CROWE ET AL.
1034
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
100
% inhibition oí syncytium formation
control
s3.5 monoclonal antibody
(ug/ml)
FIG. 2. Inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing T-lymphoblastoid cells (VB) by monoclonal antibody S3.5 directed against CD4. Mean percentage inhibition for six experiments is given. Control cultures to which an irrelevant monoclonal antibody of the same class and isotype (MOPC21) was added at 20 u-g/ml showed no inhibition of fusion.
% inhibition 100
10
rgpl20
1 rgpl20
recombinant
0.1
rgpl20
gpl20 (ug/ml)
10
rgpl20-N
FIG. 3. Inhibition of syncytium formation between HIV-infected macrophages and uninfected CD4-expressing T-lymphoblastoid cells (VB) by full-length recombinant gpl20 (rgpl20) and recombinant N terminal gpl20 (rgpl20-N). Controls were cultures to which no recombinant protein was added. The mean percentage inhibition for three experiments is given.
INHIBITION OF HIV FUSION
1035
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
>
í>'w,%
W\?~
HP •
:%
FIG. 4.
*TS'' :*
„
»'
(A) Electron micrograph depicting early fusion between an HIV-infected macrophage and lymphoblastoid cell, showing a region of loss of individual cell membranes ( x 7000). (B) At high magnification HIV can be seen budding from the macrophage surface (x 34,000). (C) The majority of retroviral particles are contained within intracytoplasmic vacuoles (x 15,000).
1036
CROWE ET AL.
exclusion method. Fusion was not inhibited by a recombinant protein representing the N-terminal on third of gpl20.
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
Electron
microscopy
When cultures showing extensive fusion between uninfected VB cells and HIV-infected macrophages were examined by electron microscopy, fusion of cells with loss of cell membranes separating macrophages and lymphoid cells was observed. Typical retroviral particles were visualized within vesicles within the multinucleated giant cells, and, less commonly, budding from the macrophage surface (Fig. 4).
ACKNOWLEDGMENTS The authors would like to acknowledge the expert technical assistance of Nancy McManus, Anne Col vin, and loann Vennari; Patricia Lekas for flow cytometric analyses, and Steven DiTomaso for preparation of the manuscript. This study was supported in part by grants from Grant AI25329 to Dr. John Mills from the U.S. NIH; and Commonwealth of Australia AIDS Research Center, NH&MRC, and the Cancer Research Institute of New York (SC).
REFERENCES AE, Phelan MA, Wells MA, Vujcic LK, Epstein JS, Lane HC, Quinnan GV: Detection of human immunodeficiency virus
1. Wittek
DISCUSSION HIV-infected macrophages can fuse with uninfected CD4expressing lymphoid cells, and this fusion may provide an additional explanation for the inexorable depletion of CD4expressing T lymphocytes in vivo. The demonstration of multinucleated giant cells is unlikely to be due to a phagocytic process, as fusion of cellular membranes can be demonstrated by electron microscopy, and the process can be specifically inhibited by blocking CD4-gpl20 interactions between the two cell populations by recombinant proteins (rCD4 and rgp 120) and a monoclonal antibody directed against the region of CD4 involved in HIV binding (@3.5). This proves that the CD4 molecule and gpl 20 are essential for the fusion process between these cell populations, but does not exclude the possibility of other molecules playing a role in this process. Although multinucleated giant cell formation is commonly
encountered in HIV-infected cultures in vitro, the relevance of this phenomenon in vivo is unclear. Much evidence supports the in vivo relevance of fusion in the pathogenesis of HIV infection. Multinucleated giant cells have been reported in brains1517 and lymph nodes18 of HIV-infected individuals. In patients with HIV-related subacute encephalopathy, histochemical staining indicates that these syncytia are composed of macrophages, and the presence of HIV within these cells has been documented.19 Similar multinucleated giant cells, composed of HIV-infected macrophages, have also been demonstrated within the spinal cords of patients with myelopathy.20 Thus our observations relating to syncytium formation between HIV-infected macrophages and uninfected CD4-expressing lymphoid cells, as well as the demonstration of inhibition of multinucleated giant cell formation by recombinant CD4 and gpl20 proteins are likely to be clinically relevant. There is currently no antiretroviral agent which will eradicate HIV, and the drugs in use clinically (reverse transcriptase inhibitors) are ineffective in chronically infected cells21 and only marginally effective in patients. To improve efficacy we must develop therapy targeted toward other sites in the replicative cycle of HIV. Compounds inhibiting attachment of virus cells and cell-cell fusion such as soluble CD4 are likely to be of most benefit when combined with a compound acting later in the cycle, such as reverse transcriptase inhibitors.
core
protein
in
plasma by
enzyme
immunoassay.
Ann Int Med
1987;107:286-292. 2. Ho DD, Moudgil T, and Alam M: Quantitation of human immunodeficiency virus type 1 in the blood of infected persons. N Engl J Med 1989;321:1621-1625. 3. Stewart SJ, Fujimoyo J, and Levy R: Human T lymphocytes and monocytes bear the same leu3 (T4) antigen. J Immunol 1986; 136:3773-3778. 4. Crowe SM, Mills J, and McGrath MS: Quantitative immunocytofluorographic analysis of CD4 surface antigen expression and HIV infection in human peripheral blood monocyte-macrophages. AIDS Res Human Retroviruses 1987;3:135-145. 5. Gartner S, Markovits P, Markovitz DM, Kaplan MH, Gallo RC, Popovic M: The role of mononuclear phagocytes in HTLVIII/LAV infection. Science 1986; 233:215-219. 6. Ho DD, Rota TR, and Hirsch MS: Infection of monocyte-macrophages by human T lymphotropic virus type III. J Clin Invest 1986;
77:1712-1715. 7. Nicholson JKA, Cross GD, Callaway CS, McDougal JS: In vitro infection of human monocytes with human T lymphotropic virus type III/lymphadenopathy associated virus (HTLVIII/LAV). J Immunol 1986;137:323-329. 8. Salahuddin SZ, Rose RM, Groopman JE, Markham PD, Gallo RC: Human T lymphotropic virus type III infection of human alveolar macrophages. Blood 1986;68:281-284. 9. Perno CF, Yarchoan R, Cooney DA, Hartman NR, Webb DS, Hao Z, Mitsuya H, Johns DG, Broder S: Replication of human immunodeficiency virus in monocytes. J Exp Med 1989;169:933939. 10. Lifson JD, Feinberg MB, Reyes G: AIDS retrovirus-induced cytopathology: giant cell formation and involvement of CD4 antigen. Science 1986;234:1123-1127. 11. Locksley RM, Crowe S, Sadick MD, Heinzel FP, Gardner KD, McGrath MS, Mills J: Release of interleukin-1 inhibitory activity (contra IL-1) by human monocyte-derived macrophages infected with human immunodeficiency virus in vitro and in vivo. J Clin Invest 1988;82:2097-2105. 12. Smith PD, Ohura K, Masur H, et al: Monocyte function in the acquired immune deficiency syndrome. J Clin Invest 1984; 74:2121-2125. 13. Tas M, Drexhage HA, and Goudsmit J: A monocyte chemotaxis inhibiting factor in serum of HIV infected men shares epitopes with the HIV transmembrane protein gp41. J Exp Immunol 1988; 71:13-18. 14. Crowe SM, Mills J, and McGrath MS: Removal of CD4 positive lymphocytes through a cell fusion process. Ill International Conference on AIDS, Washington, Abstract #F9.3 (1987).
1037
Downloaded by Tufts University package NERL from online.liebertpub.com at 07/20/17. For personal use only.
INHIBITION OF HIV FUSION 15. Sharer LR, Eun-Sook C, and Epstein LG: Multinucleated giant cells and HTLV-III in AIDS encephalopathy. Human Pathol 1985; 16:760. 16. Koenig S, Gendelman H, Orenstein JM, DalCanto MC, PezeshkpourGH, Yungbluth M, Janotta F, Aksamit A, Martin MA, Fauci AS: Detection of AIDS virus in macrophages of brain tissues from AIDS patients with encephalopathy. Science 1986;233:10891093. 17. Popovic M, Meliert W, Eifle V, Gartner S: Role of mononuclear phagocytes and accessory cells in human immunodeficiency virus type 1 infection of the brain. AnnNeurol 1988;23(suppl):s74-s77. 18. Byrnes RK, ChanWC, SpiraTJ, etal: Value of lymph node biopsy in unexplained lymphadenopathy in homosexual medicine. JAMA
1983;250:1313-1315.
19. Gartner S, Markovits P, Markovitz D, Betts RF, Popovic M: Virus isolation from and identification of HTLV-III/LAV-producing cells in brain tissue from a patient with AIDS. JAMA 1986:256:23652371.
20. Eilbott DJ, Peress N, Burger H, Weiser B, LaNeve D, Seidman R: Human immunodeficiency virus type-1 in spinal cords of acquired immunodeficiency syndrorne patients with myelopathy: expression and replication in macrophages. Proc Nati Acad Sei (USA)
1989;86:3337-3341.
21. Crowe S, McGrath MS, Elbeik T, Kirihara J, Mills J: Comparative assessment of antiretrovirals in human monocyte-macrophages and lymphoid cell lines acutely and chronically infected with the human immunodeficiency virus. J Med Virol 1989;29:176-180.
Address
reprint requests to:
Suzanne Crowe
Head, AIDS Research Unit
Macfarlane Burnet Centre for Medical Research
Yarra Bend Road
Fairfield, Australia 3078
