- Email: [email protected]
A hybridoma producing human monoclonal antibody specific for glycoprotein 120kDa of human immunodeficiency virus (HIV-1)
Life Sciences, Vol. 45, pp. iii-x Printed in the U.S.A.
Pergamon Press
AIDS RESEARCH CO~UNICATIONS:
A HYBRIDOMA PRODUCINGHUMANMONOCLONALANTIBODY SPECIFIC FOR GLYCOPROTEIN 120kDa OF HUMANIMMUNODEFICIENCY VIRUS (HIV-1) Douglas Lake', Toru Sugano^, Yoh-ichi Matsumoto^, Yasuhiko Masuho^, Eskild A. Petersen", Paul FeorinoO and Evan M. Hersh" "Section of Hematology and Oncology, "Section of Infectious Diseases Department of Internal Medicine and Health Sciences, University of Arizona Cancer Center, Tucson, Arizona 85724. ^Teijin Institute for Biomedical Research, Tokyo, Japan. oCenters for Disease Control, Atlanta, Georgia. (Received in final form August 4, 1989)
Summary A stable hybridoma producing anti-HIV human monoclonal antibody (HMCA) was generated by fusing CD3-depleted human splenic lymphocytes from an HIV sero-positive donor with the mouse myeloma cell line P3x63AgU1. The resultant hybridoma has been secreting IgG1, lambda chain for over nine months at a rate of 2.5 ug/lOBcells/day. The HMCA shows specific reactivity in ELISA using HIV-infected cell lysates. Immunofluorescence tests have indicated that this HMCAbinds specifically to the surface of H9 and C3 HIV/HTLVIIIb infected cells, HIV/NIT infected CEM cells and to MoT cells infected with an HIV clinical isolate. Western blotting revealed recognition of glycoproteins 120 and 160kDa of HIV by the HMCA. Although this HMCA demonstrated no neutralizing activity, the production of an anti-HIV HMCA specific for glycoprotein 120kDa indicates the possibility that a neutralizing HMCAcan be developed as further fusions with lymph nodes and spleens from HIV positive donors are performed. Human Immunodeficiency Virus (HIV) has been identified as the etiologic agent of Acquired Immune Deficiency Syndrome (AIDS) (1,2). The disease is characterized by a depletion of the helper/inducer subset of T-lymphocytes bearing the CD4 molecule. Depletion is in part related to the facts that the virus causes destruction of CD4 positive cells (3-5). While serum antibodies to the gag gene product, p24 are usually detected near the i n i t i a l stages of HIV infection, antibodies to glycoprotein 160kDa (gp160) and glycoprotein 120kDa (gp120) that are encoded by the env gene have been most commonly detected somewhat later in the course of infection and persist in AIDS patients (6). Howeverin experimental infections (HIV and SIV) of non human primates, antibodies to env can precede antibodies to gag. Although their t i t e r is usually low, persons infected with HIV have been shown to possess neutralizing antibodies to virus envelope antigens (7-9). Their antibodies have also been shown to mediate antibody dependent cellular cytotoxicity (ADCC) (10). In addition, gp120 has been identified as the viral envelope antigen necessary for HIV i n f e c t i v i t y (11-13). A neutralizing epitope on HIV which inhibits infection has been indentified as Please address requests for reprints to Evan M. Hersh, Section of Hematology and Oncology. Arizona Cancer Center, 1501N. Campbell Ave, Tucson, AZ 85724. 0024-3205/89 $3.00 + .00 Copyright (c) 1989 Pergamon Press plc
iv
Human Monoclonal Antibody Against HIV-I
Vol. 45, No. 16, 1989
amino acid sequence 254-274 of the env gene product (14). Also, neutralizing antibodies have been raised to the R loop region of gp120 which is centered around amino acid residue 310 (15). Thus, a monoclonal antibody specific for this epitope of gp120 could be developed which might neutralize HIV. Such an antibody might have prophylactic or therapeutic potential in patients with HIV infection. In addition, the development and characterization of human monoclonal antibodies to HIV might provide useful information on the interaction HIV with the human immune system. In this paper, we report the development and characterization of a human monoclonal antibody designated 13.10, which is specific for gp120, an envelope glycoprotein of HIV. Methods
Cells and Virus Four cell lines and three strains of HIV-I were used for assay purposes. All cells were propagated in RPMI 1640 containing 10% fetal calf serum, 2mM L-glutamine, lmM sodium pyruvate and 50ug/ml gentamycin. C3 cells (C3 cells were generously provided by Dr. Robinson, Vanderbilt University) (16) and H9 cells (17) were infected either with cell free supernate from HIV/HTLVIIIb persistently infected H9 cells or cell free supernate from HIV/NIT persistently infected CEM (CRIO) cells (18) (CEMand CRIO/NIT cells were kindly provided by David Volsky, Columbia University). Three to four days after innoculation with HIV, 90-100% of C3 and H9 cells expressed viral antigens by indirect immunofluorescence (I.F.) using AIDS patients' sera. We were also able to infect MoT cells (19) with a clinical isolate coded #20 provided by one of us (P.F.). Preparation of Viral Antigen HIV infected cell membrane lysates were used for the Enzyme-Linked Immunosorbant Assay (ELISA). Uninfected C3 cells or infected C3 cells were grown to a total of 108 cells, harvested and frozen at -70oc until use. The cells were thawed, swelled hypotonically and sonicated. Nuclei and cell debris were pelleted by low speed centrifugation. This supernate was removed and centrifuged again at 154,000xg. Triton X-tO0 (Sigman Chemicals, St. Louis, MO) was added at 0.1% to the resultant pellet which was sonicated using a Valet Cell Disrupter for 3 minutes and centrifuged at 100,O00xg on a sucrose gradient. The viral antigen-containing layer was harvested and stocked at -70oc for ELISA. Cell Fusion Lymphocytes from the spleen of a HIV sero-positive, idiopathic thrombocytopenic purpura (ITP) patient were treated with 5ug/ml of OKT3 antibody (Ortho Diagnostics, Raritan, NJ) plus low t o x i c i t y , 3-4 week old rabbit complement (Pel Freez) to deplete the T-cell population. After OKT3 treatment, 1.35 x 108 lymphocytes were fused in the presence of 35% polyethylene glycol (PEG) and 7.5% dimethyl sulfoxide to the same number of the mouse myeloma cell line P3x63AgUI (20). The hybridomas were selected in the media above using Hypoxanthine (9SUM) Aminopterin (O.4uM) Thymidine (16uM) (HAT) as selective additives. The fused lymphocytes were seeded at a density of 1.35 x 105 cells per well in 96 well micro-culture plates (Costar, Cambridge, MA) coated with mouse peritoneal macrophages at 104 cells/well as a feeder layer. The hybridomas were assayed 14 days later by ELISA.
Screening/Cloning Hybridoma supernates were screened by ELISA for anti-HIV human monoclonal antibody production using cell membrane lysates of uninfected C3
Vol. 45, No. 16, 1989
Human Monoclonal Antibody Against HIV-I
cells, HIV/NIT, and HIV/HTLVIIIb infected C3 cells coated onto 96 well plates (Falcon 3912 Oxnard, CA). ELISA plates were coated at a protein concentration of 7.5ug/ml as determined by a Pierce BCA protein assay and blocked with 3% bovine serum albumin (Sigma Chemicals). Briefly, hybridoma supernates were added to the above coated ELISA plates, incubated at room temperature for one hour then washed. Goat-anti-human IgG, alkaline phosphatase conjugated second antibody (Tago, Inc. Burlingame, CA) was added, incubated and washed as above. P-nitrophenyl phosphate (Sigma Chemicals) was used as the substrate. Development was read on a Titertek Multiskan at 405nm. Positive anti-HIV supernates were reconfirmed by indirect immunofluorescence (I.F.) using acetone-fixed infected and uninfected H9 cells. After a positive reconfirmation the appropriate hybridomas were cloned by limiting dilution at either 2 or 5 cells per well. Two weeks later the hybridoma supernates were screened again using the above methods. We screened and cloned positive hybridomas at least 4 times to ensure monoclonality and to stabilize the hybridoma (21).
Immunofluorescence Viable Cell Immunofluorescence: HIV infected cells or uninfected cells were incubated with hybridoma supernates and the appropriate controls at 40C in the dark for one hour. The cells were washed and incubated with Goatanti-human IgG, FITC conjugated (Tago, Inc.) and f i n a l l y washed free of excess second antibody. Surface immunofluorescence was analyzed on a Becton-Dickinson FACSkan Analyzer (22). Purification and Characterization Anti-HIV HMCAsupernate from mass culture was ammonium sulfate precipitated then purified using a Protein A-Sepharose column (Pharmacia, Sweden). IgG1,2,3,&4 single radial immunodiffusion (SRID) plates (ICN, Lisle, IL) were employed to determine the isotype of our HMCA. Kappa and lambda chain specific second antibodies were used in ELISA, as above, to identify the l i g h t chain subtype. Bio-Rad's HTLVIIIb Immunoblotting k i t was used to determine the viral antigen recognized by the HMCA. Briefly, hybridoma supernate or serum was incubated with the viral antigen blotted nitrocellulose strip for 30 minutes and washed. Then Bio-Rad's second antibody solution was added, incubated with the strip and washed. The substrate was added and precipitates formed where the antibodies reacted with the viral bands. Neutralization A 3H thymidine uptake assay was employed to detect inhibition of viral infection in C3 cells (23). Three-fold dilutions of 250ug/ml anti-HIV HMCA or an appropriate control HMCAwere carried out and incubated with 103 or 104 TCIDs0 of HIV for one hour at 37oc. The HIV/HMCA mixture was then incubated with uninfected C3 cells for two hours. The cells were washed free of the HIV/HMCA mixture and incubated in RPMI 1640 containing 10% FCS. The cells were observed for cytopathic effect each day for 6 days. On day 7, 50ul containing I uCi of 20ci/mMol 3H thymidine was added to each well and the cells were incubated for a final 12 hours. The plates containing the cells were frozen and thawed and the cell nuclei were harvested for 3H thymidine uptake using a PhD Cell Harvester. The 3H thymidine incorparation was measured using a Packard liquid s c i n t i l l a t i o n counter. Neutralization was calculated as follows: [HMCA + HIV + C3 c e l l s (cpm)] - [C3 c e l l s (cpm)] [C3 c e l l s (cpm)] - HIV + C3 c e l l s (cpm) X 100% = % N e u t r a l i z a t i o n .
Abbott Labs' HIV antigen capture assay was also used to test for neutralization by 13.10. The Abbott assay uses p24 as a marker antigen to
v
vi
Human Monoclonal Antibody Against HIV-I
Vol. 45, No. 16, 1989
detect the presence of HIV. Briefly, 103TCIDso from H9/HTLVIIIb infected cells were incubated for 1 hour at 37oc with 4-fold dilutions of 500ug/ml 13.10. Cell-free supernates of infected Hg and uninfected H9 cells were used as controls. After 1 hour the HMCA/HIV mixture was added to uninfected MoT cells and incubated at 37oc for 1 hour. Then the MoT cells were washed free of the HMCA/HIV mixture and were incubated in RPMI 1640 for 3 days. On day 3, supernate was taken from the MoT/HIV-MCA and the control cell cultures and added to Abbott's bead coated with anti-p24 antibody. From this point a standard enzyme-linked immunoassay was performed with the beads using Abbott's reagents and a bead-adapted ELISA plate.
Antlbody-Dependent Cell Mediated Cytotoxiclty ADCC was performed and the results were calculated using a modification of the procedure of Blumberg et. al. (24). Basically, a 5 hour Chromium-51 release assay was done. Healthy donor's peripheral blood lymphocytes were used as effector cells. Hg/HTLVIIIb persistently infected cells and uninfected H9 cells were labeled with 51Cr and used as target cells. Target cells were plated at 10,000 cells/well along with 5-fold dilutions of 500ug/ml (25ug/well) 13.10. The effector cells were added at 25,5, and 1:1 effector:target ratio. HIV positive serum was used as a positive control. After a 5 hour incubation, lOOul of supernate was removed from the wells and counted on a Packard gamma counter. Results The human anti-HIV monoclonal antibody, designated 13.10, was derived from OKT3 treated splenic lymphocytes fused with the mouse myeloma P3x63AgU1 and was selected from 170 o r i g i n a l l y positive (ELISA), but either nonspecific or unstable anti-HIV HMCAproducing hybridomas. The hybridoma secretes 2.Sug IgG/IOB cells/day. SRID showed 13.10 to be human IgG1 isotype. I t had a lambda l i g h t chain as determined by ELISA. The binding capabilities of 13.10 were determined by viable cell surface immunofluorescence (Fig. 1), and HIV immunoblotting. The fact that HMCA 13.10 bound to the available laboratory HIV strains including NIT (CEMinfected strain), HTLVIIIb, and also a clinical isolate with which we successfully infected MoT cells may indicate that i t recognizes a group specific epitope on HIV. However, the divergence of the above HIV isolates is not known. Our HMCAdid not bind to uninfected cells nor did i t crossreact with HTLVI or HTLVII infected cells including MT-4, MoT an C3 cell lines. HMCA 13.10 recognized gp120 and gp160 in western blotting analysis (Fig. 2). The recognition of both gp120 and gp160 indicates that 13.10 binds to the region of gp160 that is cleaved to form gp120 and is associated with the external portion of the viral envelope (7,25). An ADCC assay was performed to check the a b i l i t y of 13.10 to mediate ADCC. HMCA13.10 did not mediate ADCCon H9/HTLVIIIb infected cells or the control uninfected H9 cells. A 3H thymidine uptake neutralization test to detect the a b i l i t y of 13.10 to protect against HIV infection was performed. Although gp120 has been identified as the viral envelope binding site to T4+cells, our anti-gp120 HMCAdid not i n h i b i t viral infection or syncytium formation when compared to uninfected cells that show no syncytia. Abbott's antigen capture assay was also employed to evaluate neutralization. While positive controls (patient sera) showed neutralization of HIV in this assay, HMCA13.10 did not i n h i b i t HIV infection. Finally, HMCA13.10 was not neutralizing when examined for syncytia inhibition by light microscopy.
Vol. 45, No. 16, 1989
Human Monoclonal Antibody Against HIV-I
C3
C3/HTLVIIIB
% POSITIVE.
1.8
FITC INTENSITY
FITC INTENSr ~F
~
vii
% POSITIVE.g9.5
% POSITIVE=0.7 c~
B :I
FITC '.NTENSITY
Fr'C INTENSITY
% POSITIVE.1.1
C
d
FITC INTENSI~Y
FrTC INTENSITY
% POSITIVE.1,1
% POSITrVE.2.1
D
lo
Io
ic~
I
I
I
4
FITC INTENSITY
FITC INTENSITY
Fig. 1. Indirect immunofluorescence analysis demonstrates specific reactivity of 13.10 with the surface of C3/HTLVIIIb infected cells (right column) and not with uninfected C3 cells ( l e f t column). Immunofluorescence was performed as described in Methods. Results show: A, Human, HIV-negative serum at a 1:100 dilution; B, Human, HIV-positive serum at a 1:100 dilution; C, HMCA 13.10 at 50ug/ml; D, Anti-Varicella Zoster virus HMCAat 50ug/ml.
viii
Human M o n o c l o n a l
I
Antibody
Against
HIV-I
2
Vol.
45, No.
16,
1989
4
p65 p55 ~ gp41
p32
-~
p24
p17
Fig. 2. Western blot r e a c t i v i t y of HMCA 13.10. Western was performed as described in Methods. Results show: Lane I : Human, HIV-positive serum (1:100), Lane 2: Human HIV-negative serum (1:100), Lane 3: HMCA 13.10 at 50ug/ml, Lane 4: Anti-Varicella Zoster virus HMCAat 50ug/ml. Discussion
We have established the f i r s t human x mouse hybridoma, designated 13.10, that secretes human IgGl, lambda chain and binds s p e c i f i c a l l y to the gpl20 and gp]60 HIV antigens. Previously our group has produced a human x mouse hybridoma y i e l d i n g human IgG MCA to gpl60 and gp41 (26). Only a few reports of human IgG MCA to HIV have appeared (27-2g). Most of these were made by Epstein Barr Virus transformation and t h e i r s t a b i l i t y was not defined. In contrast the current HMCA has been stably secreting reasonable amounts of IgG for over 9 months. The lack of suitable human myeloma fusion partners, convinced us to use the mouse myeloma cell l i n e P3x63AgUI. I t has been reported that human x mouse hybridomas are very unstable (30). We encountered some i n s t a b i l i t y problems, however, through repeated sub-cloning our human x mouse hybridoma s t a b i l i z e d and continues to secrete human IgG. Our previous work has indicated that in v i t r o stimulation of sensitized lymphocytes with a v a r i e t y of viruses increases the number of antigens p e c i f i c B-lymphocytes before fusion (31-33). However, we did not observe t h i s phenomenon with HIV and in v i t r o stimulation with HIV antigens a c t u a l l y suppressed the antibody response (26). We attempted in v i t r o stimulation in t h i s study, as before, but the lymphocytes, for reasons we did not determine, f a i l e d to survive the combination of both in v i t r o stimulation and fusion processes. Therefore, our current opinion is that the best approach to generating human monoclonal antibodies to HIV is d i r e c t fusion without antigenic stimulation using a suitable fusion partner. Although our HMCA bound to gp120, no n e u t r a l i z i n g a c t i v i t y was observed by 3H thymidine uptake, antigen capture or ADCC methods used to attempt to detect these b i o l o g i c a l a c t i v i t i e s . However, i t has been reported (34) that gp120 is r a p i d l y shed from the v i r a l envelope creating a gp120 antigenemia which may be more severe than the p24 antigenemia previously reported (35).
Vol. 45, No. 16, 1989
Human Monoclonal Antibody Against HIV-I
ix
I f there is free gp120 in the blood, i t may bind the CD4 molecule on uninfected cells causing them to fuse or causing the destruction of uninfected cells by cytotoxic T-cells. In this respect HMCA13.10 may have therapeutic potential by clearing gp120 antigenemia. We have, as our major objective to develop neutralizing HMCAto HIV from the lymphoid cells of HIV sero-positive donors. We speculate that this may be d i f f i c u l t for several reasons. These include the facts that the Blymphocyte population may already be polyclonally activated, that the specific B-lymphocyte population may be depleted late in the disease and that the regulation by T-cell populations of B-cell function in AIDS may be greatly perturbed (36). In addition, HIV infected patients often have low levels of neutralizing antibodies (7,8) suggesting that there are relatively few B lymphocytes committed to this pathway of differentiation or that the neutralizing epitopes are immuno-recessive. Virologically, HIV exhibits great antigenic shift resulting in many divergent strains that cannot be neutralized by type specific antibodies or autologous serum (37). Because of the great divergence among HIV isolates, a cocktail of neutralizing, type specific human monoclonal antibodies might be used as passive immunotherapy. In spite of these considerations, the fact that we can readily develop stable hybridomas secreting HMCAsto several HIV antigens (gp160,120, and 41) suggests that further efforts may yield biologically active and therefore clinically useful human monoclonal antibodies to HIV. Acknowledgement This work was supported by funds from Teijin Ltd., Tokyo, Japan. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
D. KLATZMAN, F. BARRE-SINOUSSI, M.T. NUGEYRE, C. DAUGUET, , E. VILMER, C. GRISCELLI, F. BRUN-VEZINET, C. ROUZIOUS, J.D. GLUCKMANand L. MONTAGNIER, Science 225 59-62 (1984). L. MONTAGNIER, J. GRUEST, S. CHAMARET, S. DAUGUET, C. AXLER, D. GUETARD, M.T. NUGEYRE, F. BARRE-SINOUSSl, J.C. CHERMANN, J.B. BRUNET, D. KLATZMAN and J.C. GLUCKMAN, Science 225 63-66 (1984). J.D. LIFSON, G.R. REYES, M.S. McGRATH, B.S. STEIN and E.G. ENGLEMAN, Science 232 1123-1127 (1986). J.D. LIFSON, M.B. FEINBERG, G.R. REYES, L. RABIN, B. BANAPOUR, S. CHAKRABARTI, B. MOSS, F. WONG-STAAL, K.So STEIMER and E.G. ENGLEMAN, Nature 323 725-728 (1986). J. SODROSKI, W.C. GOH, C. ROSEN, K. CAMPBELL and W.A. HAZELTINE, Nature 322 470-474 (1986). F. BARIN, M.G. McLANE, J.S. ALLAN, T.H. LEE, J.E. GROOPMANand M. ESSEX, Science 228 1094-1096 (1985). R.A. WEISS, P.R. CLAPHAM, R. CHEINGSONG-POPOV, A.G. DALGLEISH, C.A. CARNE, I.V.D. WELLER and R.S. TEDDER, Nature 316 69-71 (1985). M. ROBERT-GURNOFF, M. BROWNand R.C. GALLO, Nature 316 72-74 (1985). R.A. WEISS, P.R. CLAPHAM, J.N. WEBER, A.G. DALGLEISH, L.A. LAKSY and P.W. BERMAN, Nature 324 572-575 (1986). A.H. ROOK, G.C. LANE, T. FOLKS, S. McCOY, H. ALTER and A.S. FAUCl, J. Immunol 138 1064-1067 (1987). S.D. PUTNEY, T.F. MATTHEWS, W.G. ROBEY, D.L. LYNN, M. ROGERT-GURNOFF, W.Y. MUELLER, A.J. LANGLOIS, J. GHRAYEB, S.R. PETTEWAY JR. and K.J. WEINHOLD, Science 234 1392-1395 (1986). T.J. MATTHEWS, A.J. LANGLOIS, W.G. ROBEY, N.T. CHANG, R.C. GALLO, P.J. FISCHINGER and D.P. BOLOGNESI, Proc. Natl. Acad. Sci. USA 83 9709-9713 (1986).
x
Human Monoclonal Antibody Against HIV-I
13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.
Vol. 45, No. 16, 1989
W.G. ROBEY, L.O. ARTHUR, T.J. MATTHEWS, A. LANGLOIS, T.D. COPELAND, N.W. LERCHE, S. OROSZLAN, D.P. BOLOGNESI, R.V. GILDEN and P.J. FISCHINGER, Proc. Natl. Acad. Sci. USA 83 7023-7027 (1986). D.D. HO, J.C. KAPLAN, I.E. RACKAUSKASa~-d M.E. GURNEY, Science 239 1021-1023 (1988). T.J. MATTHEWS, H.K. LYERLY, K.J. WEINHOLD, A.J. LANGLOIS, J. RUSCHE, S.D. PUTNEY, R.C. GALLO and D.P. BOLOGNESI, AIDS Res. Hum. Retroviruses 3 197-206 (1987). D.C. MONTEFIORI and W.M. MITCHELL, Virology 155 725-731 (1986). M. POPOVlC, M.G. SARNGADHARAN, E. READ and R.T. GALLO, Science 224 497500 (1984). D. CASAREALE, M. STEVENSON, K. SAKAI and D.J. VOLSKY, Virology 156 4049 (1987). A. SAXON, R.H. STEVENSand D.W. GOLDE, Ann. Intern. Med. 88 323-326 (1978). D.E. YELTON, Lymphocyte Hybridomas, B.A. Diamond, S-P. Kwan and M.D. Scharff, 1-7, Springert-Verlag, New York (1985). S.SAWADA, T. KAWAMURAand Y. MASUHO, J. Gen. Microbiol. 133 3588-3590 (1987). J.D. LIFSON, D.T. SASAKI and E.G. ENGLEMAN, J. Immunol. Meth. 86 143149 (1986). S. HARADA, D.T. PURTILO, Y. KOYANAGI, J. SONNABENDand N. YAMAMOTO, J. Immunol. Meth. 92 177-187 (1986). R.S. BLUMBERG, T. PARADIS, K.L. HARTSHORN,M. VOGT, D.D. HO, M.S. HIRSCH, J. LEBAN, V.L. SATOand R.T. SCHOOLEY, J. Infect. Dis. 156 878884 (1987). J.S. ALLAN, J.E. COLIGAN, F. BARIN, M.F. McLANE, J.G. SODROSKI, C.A. ROSEN, W.A. HAZELTINE, T.H. LEE and M. ESSEX, Science 228 1091-1093 (1985). T. SUGANO, Y. MASUHO, Y. MATSUMOTO, D. LAKE, C. GSCHWIND, E.A. PETERSEN and E.M. HERSH, Biochem. Biophys. Res. Comm. 155 1105-1112 (1988). L.A. EVANS, J.M. HOMSY, W.J.W. MORROW, I. GASTON, C.D. SOOYand J.A. LEVY, J. Immunol. 140 941-943 (1988). B. BANAPOUR, K. ROSENTHAL, L. RABIN, V. SHARMA, L. YOUNG, J. FERNANDEZ, E. ENGELMAN, M. McGRATH, G. REYES and J.D. LIFSON, J. Immunol. 139 4027-4033 (1987). M.K. GORNY, V. GIANAKAKOS and S. ZOLLA-PAZNER, The fourth annual AIDS meeting [abstract 1596] Sweden, (1988). C.M. CROCE, M. SHANDER, J. MARTINIS, L. CICUREL, G.G. D'ANCONA and H. KOPROWSKI, Eur. J. Immunol. I0 486-488 (1987). S. FUJINAGA, T. SUGANO, Y. MA~SUMOTO, Y. MASUHOand R. MORI, J. Infect. Dis. 155 45-53 (1987). T. SUGANO, Y. MATSUMOTO, C. MIYAMOTOand Y. MASUHO, Eur. J. Immunol. 17 359-364 (1987). Y. MATSUMOTO, T. SUGANO, C. MIYAMOTOand Y. MASUHO, Biochem. Biophys. Res. Comm. 137 273-280 (1986). R.R. REDFIELD and D.S. BURKE, Sci. American 259 90-99 (1988). J.S. McDOUGAL, M.S. KENNEDY, J.K.A. NICHOLSON, T.J. SPIRA, H.W. JAFFE, J.E. KAPLAN, D.B. FISHBEIN, P. O'MALLEY, C.H. ALOISO, C.M. BLACK, M. HUBBARD and C.B. REINER, J. Clin. Invest. 80 316-324 (1987). J. GOUDSMIT, M.A. LANGE, D.A. PAUL and J. DAWSON, J. Infect. Dis. 155 558-560 (1987). J.A. LEVY, The f i f t h annual AIDS conference [abstract M.C.O.I] Montreal, Canada (1989).
Pergamon Press
AIDS RESEARCH CO~UNICATIONS:
A HYBRIDOMA PRODUCINGHUMANMONOCLONALANTIBODY SPECIFIC FOR GLYCOPROTEIN 120kDa OF HUMANIMMUNODEFICIENCY VIRUS (HIV-1) Douglas Lake', Toru Sugano^, Yoh-ichi Matsumoto^, Yasuhiko Masuho^, Eskild A. Petersen", Paul FeorinoO and Evan M. Hersh" "Section of Hematology and Oncology, "Section of Infectious Diseases Department of Internal Medicine and Health Sciences, University of Arizona Cancer Center, Tucson, Arizona 85724. ^Teijin Institute for Biomedical Research, Tokyo, Japan. oCenters for Disease Control, Atlanta, Georgia. (Received in final form August 4, 1989)
Summary A stable hybridoma producing anti-HIV human monoclonal antibody (HMCA) was generated by fusing CD3-depleted human splenic lymphocytes from an HIV sero-positive donor with the mouse myeloma cell line P3x63AgU1. The resultant hybridoma has been secreting IgG1, lambda chain for over nine months at a rate of 2.5 ug/lOBcells/day. The HMCA shows specific reactivity in ELISA using HIV-infected cell lysates. Immunofluorescence tests have indicated that this HMCAbinds specifically to the surface of H9 and C3 HIV/HTLVIIIb infected cells, HIV/NIT infected CEM cells and to MoT cells infected with an HIV clinical isolate. Western blotting revealed recognition of glycoproteins 120 and 160kDa of HIV by the HMCA. Although this HMCA demonstrated no neutralizing activity, the production of an anti-HIV HMCA specific for glycoprotein 120kDa indicates the possibility that a neutralizing HMCAcan be developed as further fusions with lymph nodes and spleens from HIV positive donors are performed. Human Immunodeficiency Virus (HIV) has been identified as the etiologic agent of Acquired Immune Deficiency Syndrome (AIDS) (1,2). The disease is characterized by a depletion of the helper/inducer subset of T-lymphocytes bearing the CD4 molecule. Depletion is in part related to the facts that the virus causes destruction of CD4 positive cells (3-5). While serum antibodies to the gag gene product, p24 are usually detected near the i n i t i a l stages of HIV infection, antibodies to glycoprotein 160kDa (gp160) and glycoprotein 120kDa (gp120) that are encoded by the env gene have been most commonly detected somewhat later in the course of infection and persist in AIDS patients (6). Howeverin experimental infections (HIV and SIV) of non human primates, antibodies to env can precede antibodies to gag. Although their t i t e r is usually low, persons infected with HIV have been shown to possess neutralizing antibodies to virus envelope antigens (7-9). Their antibodies have also been shown to mediate antibody dependent cellular cytotoxicity (ADCC) (10). In addition, gp120 has been identified as the viral envelope antigen necessary for HIV i n f e c t i v i t y (11-13). A neutralizing epitope on HIV which inhibits infection has been indentified as Please address requests for reprints to Evan M. Hersh, Section of Hematology and Oncology. Arizona Cancer Center, 1501N. Campbell Ave, Tucson, AZ 85724. 0024-3205/89 $3.00 + .00 Copyright (c) 1989 Pergamon Press plc
iv
Human Monoclonal Antibody Against HIV-I
Vol. 45, No. 16, 1989
amino acid sequence 254-274 of the env gene product (14). Also, neutralizing antibodies have been raised to the R loop region of gp120 which is centered around amino acid residue 310 (15). Thus, a monoclonal antibody specific for this epitope of gp120 could be developed which might neutralize HIV. Such an antibody might have prophylactic or therapeutic potential in patients with HIV infection. In addition, the development and characterization of human monoclonal antibodies to HIV might provide useful information on the interaction HIV with the human immune system. In this paper, we report the development and characterization of a human monoclonal antibody designated 13.10, which is specific for gp120, an envelope glycoprotein of HIV. Methods
Cells and Virus Four cell lines and three strains of HIV-I were used for assay purposes. All cells were propagated in RPMI 1640 containing 10% fetal calf serum, 2mM L-glutamine, lmM sodium pyruvate and 50ug/ml gentamycin. C3 cells (C3 cells were generously provided by Dr. Robinson, Vanderbilt University) (16) and H9 cells (17) were infected either with cell free supernate from HIV/HTLVIIIb persistently infected H9 cells or cell free supernate from HIV/NIT persistently infected CEM (CRIO) cells (18) (CEMand CRIO/NIT cells were kindly provided by David Volsky, Columbia University). Three to four days after innoculation with HIV, 90-100% of C3 and H9 cells expressed viral antigens by indirect immunofluorescence (I.F.) using AIDS patients' sera. We were also able to infect MoT cells (19) with a clinical isolate coded #20 provided by one of us (P.F.). Preparation of Viral Antigen HIV infected cell membrane lysates were used for the Enzyme-Linked Immunosorbant Assay (ELISA). Uninfected C3 cells or infected C3 cells were grown to a total of 108 cells, harvested and frozen at -70oc until use. The cells were thawed, swelled hypotonically and sonicated. Nuclei and cell debris were pelleted by low speed centrifugation. This supernate was removed and centrifuged again at 154,000xg. Triton X-tO0 (Sigman Chemicals, St. Louis, MO) was added at 0.1% to the resultant pellet which was sonicated using a Valet Cell Disrupter for 3 minutes and centrifuged at 100,O00xg on a sucrose gradient. The viral antigen-containing layer was harvested and stocked at -70oc for ELISA. Cell Fusion Lymphocytes from the spleen of a HIV sero-positive, idiopathic thrombocytopenic purpura (ITP) patient were treated with 5ug/ml of OKT3 antibody (Ortho Diagnostics, Raritan, NJ) plus low t o x i c i t y , 3-4 week old rabbit complement (Pel Freez) to deplete the T-cell population. After OKT3 treatment, 1.35 x 108 lymphocytes were fused in the presence of 35% polyethylene glycol (PEG) and 7.5% dimethyl sulfoxide to the same number of the mouse myeloma cell line P3x63AgUI (20). The hybridomas were selected in the media above using Hypoxanthine (9SUM) Aminopterin (O.4uM) Thymidine (16uM) (HAT) as selective additives. The fused lymphocytes were seeded at a density of 1.35 x 105 cells per well in 96 well micro-culture plates (Costar, Cambridge, MA) coated with mouse peritoneal macrophages at 104 cells/well as a feeder layer. The hybridomas were assayed 14 days later by ELISA.
Screening/Cloning Hybridoma supernates were screened by ELISA for anti-HIV human monoclonal antibody production using cell membrane lysates of uninfected C3
Vol. 45, No. 16, 1989
Human Monoclonal Antibody Against HIV-I
cells, HIV/NIT, and HIV/HTLVIIIb infected C3 cells coated onto 96 well plates (Falcon 3912 Oxnard, CA). ELISA plates were coated at a protein concentration of 7.5ug/ml as determined by a Pierce BCA protein assay and blocked with 3% bovine serum albumin (Sigma Chemicals). Briefly, hybridoma supernates were added to the above coated ELISA plates, incubated at room temperature for one hour then washed. Goat-anti-human IgG, alkaline phosphatase conjugated second antibody (Tago, Inc. Burlingame, CA) was added, incubated and washed as above. P-nitrophenyl phosphate (Sigma Chemicals) was used as the substrate. Development was read on a Titertek Multiskan at 405nm. Positive anti-HIV supernates were reconfirmed by indirect immunofluorescence (I.F.) using acetone-fixed infected and uninfected H9 cells. After a positive reconfirmation the appropriate hybridomas were cloned by limiting dilution at either 2 or 5 cells per well. Two weeks later the hybridoma supernates were screened again using the above methods. We screened and cloned positive hybridomas at least 4 times to ensure monoclonality and to stabilize the hybridoma (21).
Immunofluorescence Viable Cell Immunofluorescence: HIV infected cells or uninfected cells were incubated with hybridoma supernates and the appropriate controls at 40C in the dark for one hour. The cells were washed and incubated with Goatanti-human IgG, FITC conjugated (Tago, Inc.) and f i n a l l y washed free of excess second antibody. Surface immunofluorescence was analyzed on a Becton-Dickinson FACSkan Analyzer (22). Purification and Characterization Anti-HIV HMCAsupernate from mass culture was ammonium sulfate precipitated then purified using a Protein A-Sepharose column (Pharmacia, Sweden). IgG1,2,3,&4 single radial immunodiffusion (SRID) plates (ICN, Lisle, IL) were employed to determine the isotype of our HMCA. Kappa and lambda chain specific second antibodies were used in ELISA, as above, to identify the l i g h t chain subtype. Bio-Rad's HTLVIIIb Immunoblotting k i t was used to determine the viral antigen recognized by the HMCA. Briefly, hybridoma supernate or serum was incubated with the viral antigen blotted nitrocellulose strip for 30 minutes and washed. Then Bio-Rad's second antibody solution was added, incubated with the strip and washed. The substrate was added and precipitates formed where the antibodies reacted with the viral bands. Neutralization A 3H thymidine uptake assay was employed to detect inhibition of viral infection in C3 cells (23). Three-fold dilutions of 250ug/ml anti-HIV HMCA or an appropriate control HMCAwere carried out and incubated with 103 or 104 TCIDs0 of HIV for one hour at 37oc. The HIV/HMCA mixture was then incubated with uninfected C3 cells for two hours. The cells were washed free of the HIV/HMCA mixture and incubated in RPMI 1640 containing 10% FCS. The cells were observed for cytopathic effect each day for 6 days. On day 7, 50ul containing I uCi of 20ci/mMol 3H thymidine was added to each well and the cells were incubated for a final 12 hours. The plates containing the cells were frozen and thawed and the cell nuclei were harvested for 3H thymidine uptake using a PhD Cell Harvester. The 3H thymidine incorparation was measured using a Packard liquid s c i n t i l l a t i o n counter. Neutralization was calculated as follows: [HMCA + HIV + C3 c e l l s (cpm)] - [C3 c e l l s (cpm)] [C3 c e l l s (cpm)] - HIV + C3 c e l l s (cpm) X 100% = % N e u t r a l i z a t i o n .
Abbott Labs' HIV antigen capture assay was also used to test for neutralization by 13.10. The Abbott assay uses p24 as a marker antigen to
v
vi
Human Monoclonal Antibody Against HIV-I
Vol. 45, No. 16, 1989
detect the presence of HIV. Briefly, 103TCIDso from H9/HTLVIIIb infected cells were incubated for 1 hour at 37oc with 4-fold dilutions of 500ug/ml 13.10. Cell-free supernates of infected Hg and uninfected H9 cells were used as controls. After 1 hour the HMCA/HIV mixture was added to uninfected MoT cells and incubated at 37oc for 1 hour. Then the MoT cells were washed free of the HMCA/HIV mixture and were incubated in RPMI 1640 for 3 days. On day 3, supernate was taken from the MoT/HIV-MCA and the control cell cultures and added to Abbott's bead coated with anti-p24 antibody. From this point a standard enzyme-linked immunoassay was performed with the beads using Abbott's reagents and a bead-adapted ELISA plate.
Antlbody-Dependent Cell Mediated Cytotoxiclty ADCC was performed and the results were calculated using a modification of the procedure of Blumberg et. al. (24). Basically, a 5 hour Chromium-51 release assay was done. Healthy donor's peripheral blood lymphocytes were used as effector cells. Hg/HTLVIIIb persistently infected cells and uninfected H9 cells were labeled with 51Cr and used as target cells. Target cells were plated at 10,000 cells/well along with 5-fold dilutions of 500ug/ml (25ug/well) 13.10. The effector cells were added at 25,5, and 1:1 effector:target ratio. HIV positive serum was used as a positive control. After a 5 hour incubation, lOOul of supernate was removed from the wells and counted on a Packard gamma counter. Results The human anti-HIV monoclonal antibody, designated 13.10, was derived from OKT3 treated splenic lymphocytes fused with the mouse myeloma P3x63AgU1 and was selected from 170 o r i g i n a l l y positive (ELISA), but either nonspecific or unstable anti-HIV HMCAproducing hybridomas. The hybridoma secretes 2.Sug IgG/IOB cells/day. SRID showed 13.10 to be human IgG1 isotype. I t had a lambda l i g h t chain as determined by ELISA. The binding capabilities of 13.10 were determined by viable cell surface immunofluorescence (Fig. 1), and HIV immunoblotting. The fact that HMCA 13.10 bound to the available laboratory HIV strains including NIT (CEMinfected strain), HTLVIIIb, and also a clinical isolate with which we successfully infected MoT cells may indicate that i t recognizes a group specific epitope on HIV. However, the divergence of the above HIV isolates is not known. Our HMCAdid not bind to uninfected cells nor did i t crossreact with HTLVI or HTLVII infected cells including MT-4, MoT an C3 cell lines. HMCA 13.10 recognized gp120 and gp160 in western blotting analysis (Fig. 2). The recognition of both gp120 and gp160 indicates that 13.10 binds to the region of gp160 that is cleaved to form gp120 and is associated with the external portion of the viral envelope (7,25). An ADCC assay was performed to check the a b i l i t y of 13.10 to mediate ADCC. HMCA13.10 did not mediate ADCCon H9/HTLVIIIb infected cells or the control uninfected H9 cells. A 3H thymidine uptake neutralization test to detect the a b i l i t y of 13.10 to protect against HIV infection was performed. Although gp120 has been identified as the viral envelope binding site to T4+cells, our anti-gp120 HMCAdid not i n h i b i t viral infection or syncytium formation when compared to uninfected cells that show no syncytia. Abbott's antigen capture assay was also employed to evaluate neutralization. While positive controls (patient sera) showed neutralization of HIV in this assay, HMCA13.10 did not i n h i b i t HIV infection. Finally, HMCA13.10 was not neutralizing when examined for syncytia inhibition by light microscopy.
Vol. 45, No. 16, 1989
Human Monoclonal Antibody Against HIV-I
C3
C3/HTLVIIIB
% POSITIVE.
1.8
FITC INTENSITY
FITC INTENSr ~F
~
vii
% POSITIVE.g9.5
% POSITIVE=0.7 c~
B :I
FITC '.NTENSITY
Fr'C INTENSITY
% POSITIVE.1.1
C
d
FITC INTENSI~Y
FrTC INTENSITY
% POSITIVE.1,1
% POSITrVE.2.1
D
lo
Io
ic~
I
I
I
4
FITC INTENSITY
FITC INTENSITY
Fig. 1. Indirect immunofluorescence analysis demonstrates specific reactivity of 13.10 with the surface of C3/HTLVIIIb infected cells (right column) and not with uninfected C3 cells ( l e f t column). Immunofluorescence was performed as described in Methods. Results show: A, Human, HIV-negative serum at a 1:100 dilution; B, Human, HIV-positive serum at a 1:100 dilution; C, HMCA 13.10 at 50ug/ml; D, Anti-Varicella Zoster virus HMCAat 50ug/ml.
viii
Human M o n o c l o n a l
I
Antibody
Against
HIV-I
2
Vol.
45, No.
16,
1989
4
p65 p55 ~ gp41
p32
-~
p24
p17
Fig. 2. Western blot r e a c t i v i t y of HMCA 13.10. Western was performed as described in Methods. Results show: Lane I : Human, HIV-positive serum (1:100), Lane 2: Human HIV-negative serum (1:100), Lane 3: HMCA 13.10 at 50ug/ml, Lane 4: Anti-Varicella Zoster virus HMCAat 50ug/ml. Discussion
We have established the f i r s t human x mouse hybridoma, designated 13.10, that secretes human IgGl, lambda chain and binds s p e c i f i c a l l y to the gpl20 and gp]60 HIV antigens. Previously our group has produced a human x mouse hybridoma y i e l d i n g human IgG MCA to gpl60 and gp41 (26). Only a few reports of human IgG MCA to HIV have appeared (27-2g). Most of these were made by Epstein Barr Virus transformation and t h e i r s t a b i l i t y was not defined. In contrast the current HMCA has been stably secreting reasonable amounts of IgG for over 9 months. The lack of suitable human myeloma fusion partners, convinced us to use the mouse myeloma cell l i n e P3x63AgUI. I t has been reported that human x mouse hybridomas are very unstable (30). We encountered some i n s t a b i l i t y problems, however, through repeated sub-cloning our human x mouse hybridoma s t a b i l i z e d and continues to secrete human IgG. Our previous work has indicated that in v i t r o stimulation of sensitized lymphocytes with a v a r i e t y of viruses increases the number of antigens p e c i f i c B-lymphocytes before fusion (31-33). However, we did not observe t h i s phenomenon with HIV and in v i t r o stimulation with HIV antigens a c t u a l l y suppressed the antibody response (26). We attempted in v i t r o stimulation in t h i s study, as before, but the lymphocytes, for reasons we did not determine, f a i l e d to survive the combination of both in v i t r o stimulation and fusion processes. Therefore, our current opinion is that the best approach to generating human monoclonal antibodies to HIV is d i r e c t fusion without antigenic stimulation using a suitable fusion partner. Although our HMCA bound to gp120, no n e u t r a l i z i n g a c t i v i t y was observed by 3H thymidine uptake, antigen capture or ADCC methods used to attempt to detect these b i o l o g i c a l a c t i v i t i e s . However, i t has been reported (34) that gp120 is r a p i d l y shed from the v i r a l envelope creating a gp120 antigenemia which may be more severe than the p24 antigenemia previously reported (35).
Vol. 45, No. 16, 1989
Human Monoclonal Antibody Against HIV-I
ix
I f there is free gp120 in the blood, i t may bind the CD4 molecule on uninfected cells causing them to fuse or causing the destruction of uninfected cells by cytotoxic T-cells. In this respect HMCA13.10 may have therapeutic potential by clearing gp120 antigenemia. We have, as our major objective to develop neutralizing HMCAto HIV from the lymphoid cells of HIV sero-positive donors. We speculate that this may be d i f f i c u l t for several reasons. These include the facts that the Blymphocyte population may already be polyclonally activated, that the specific B-lymphocyte population may be depleted late in the disease and that the regulation by T-cell populations of B-cell function in AIDS may be greatly perturbed (36). In addition, HIV infected patients often have low levels of neutralizing antibodies (7,8) suggesting that there are relatively few B lymphocytes committed to this pathway of differentiation or that the neutralizing epitopes are immuno-recessive. Virologically, HIV exhibits great antigenic shift resulting in many divergent strains that cannot be neutralized by type specific antibodies or autologous serum (37). Because of the great divergence among HIV isolates, a cocktail of neutralizing, type specific human monoclonal antibodies might be used as passive immunotherapy. In spite of these considerations, the fact that we can readily develop stable hybridomas secreting HMCAsto several HIV antigens (gp160,120, and 41) suggests that further efforts may yield biologically active and therefore clinically useful human monoclonal antibodies to HIV. Acknowledgement This work was supported by funds from Teijin Ltd., Tokyo, Japan. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
D. KLATZMAN, F. BARRE-SINOUSSI, M.T. NUGEYRE, C. DAUGUET, , E. VILMER, C. GRISCELLI, F. BRUN-VEZINET, C. ROUZIOUS, J.D. GLUCKMANand L. MONTAGNIER, Science 225 59-62 (1984). L. MONTAGNIER, J. GRUEST, S. CHAMARET, S. DAUGUET, C. AXLER, D. GUETARD, M.T. NUGEYRE, F. BARRE-SINOUSSl, J.C. CHERMANN, J.B. BRUNET, D. KLATZMAN and J.C. GLUCKMAN, Science 225 63-66 (1984). J.D. LIFSON, G.R. REYES, M.S. McGRATH, B.S. STEIN and E.G. ENGLEMAN, Science 232 1123-1127 (1986). J.D. LIFSON, M.B. FEINBERG, G.R. REYES, L. RABIN, B. BANAPOUR, S. CHAKRABARTI, B. MOSS, F. WONG-STAAL, K.So STEIMER and E.G. ENGLEMAN, Nature 323 725-728 (1986). J. SODROSKI, W.C. GOH, C. ROSEN, K. CAMPBELL and W.A. HAZELTINE, Nature 322 470-474 (1986). F. BARIN, M.G. McLANE, J.S. ALLAN, T.H. LEE, J.E. GROOPMANand M. ESSEX, Science 228 1094-1096 (1985). R.A. WEISS, P.R. CLAPHAM, R. CHEINGSONG-POPOV, A.G. DALGLEISH, C.A. CARNE, I.V.D. WELLER and R.S. TEDDER, Nature 316 69-71 (1985). M. ROBERT-GURNOFF, M. BROWNand R.C. GALLO, Nature 316 72-74 (1985). R.A. WEISS, P.R. CLAPHAM, J.N. WEBER, A.G. DALGLEISH, L.A. LAKSY and P.W. BERMAN, Nature 324 572-575 (1986). A.H. ROOK, G.C. LANE, T. FOLKS, S. McCOY, H. ALTER and A.S. FAUCl, J. Immunol 138 1064-1067 (1987). S.D. PUTNEY, T.F. MATTHEWS, W.G. ROBEY, D.L. LYNN, M. ROGERT-GURNOFF, W.Y. MUELLER, A.J. LANGLOIS, J. GHRAYEB, S.R. PETTEWAY JR. and K.J. WEINHOLD, Science 234 1392-1395 (1986). T.J. MATTHEWS, A.J. LANGLOIS, W.G. ROBEY, N.T. CHANG, R.C. GALLO, P.J. FISCHINGER and D.P. BOLOGNESI, Proc. Natl. Acad. Sci. USA 83 9709-9713 (1986).
x
Human Monoclonal Antibody Against HIV-I
13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.
Vol. 45, No. 16, 1989
W.G. ROBEY, L.O. ARTHUR, T.J. MATTHEWS, A. LANGLOIS, T.D. COPELAND, N.W. LERCHE, S. OROSZLAN, D.P. BOLOGNESI, R.V. GILDEN and P.J. FISCHINGER, Proc. Natl. Acad. Sci. USA 83 7023-7027 (1986). D.D. HO, J.C. KAPLAN, I.E. RACKAUSKASa~-d M.E. GURNEY, Science 239 1021-1023 (1988). T.J. MATTHEWS, H.K. LYERLY, K.J. WEINHOLD, A.J. LANGLOIS, J. RUSCHE, S.D. PUTNEY, R.C. GALLO and D.P. BOLOGNESI, AIDS Res. Hum. Retroviruses 3 197-206 (1987). D.C. MONTEFIORI and W.M. MITCHELL, Virology 155 725-731 (1986). M. POPOVlC, M.G. SARNGADHARAN, E. READ and R.T. GALLO, Science 224 497500 (1984). D. CASAREALE, M. STEVENSON, K. SAKAI and D.J. VOLSKY, Virology 156 4049 (1987). A. SAXON, R.H. STEVENSand D.W. GOLDE, Ann. Intern. Med. 88 323-326 (1978). D.E. YELTON, Lymphocyte Hybridomas, B.A. Diamond, S-P. Kwan and M.D. Scharff, 1-7, Springert-Verlag, New York (1985). S.SAWADA, T. KAWAMURAand Y. MASUHO, J. Gen. Microbiol. 133 3588-3590 (1987). J.D. LIFSON, D.T. SASAKI and E.G. ENGLEMAN, J. Immunol. Meth. 86 143149 (1986). S. HARADA, D.T. PURTILO, Y. KOYANAGI, J. SONNABENDand N. YAMAMOTO, J. Immunol. Meth. 92 177-187 (1986). R.S. BLUMBERG, T. PARADIS, K.L. HARTSHORN,M. VOGT, D.D. HO, M.S. HIRSCH, J. LEBAN, V.L. SATOand R.T. SCHOOLEY, J. Infect. Dis. 156 878884 (1987). J.S. ALLAN, J.E. COLIGAN, F. BARIN, M.F. McLANE, J.G. SODROSKI, C.A. ROSEN, W.A. HAZELTINE, T.H. LEE and M. ESSEX, Science 228 1091-1093 (1985). T. SUGANO, Y. MASUHO, Y. MATSUMOTO, D. LAKE, C. GSCHWIND, E.A. PETERSEN and E.M. HERSH, Biochem. Biophys. Res. Comm. 155 1105-1112 (1988). L.A. EVANS, J.M. HOMSY, W.J.W. MORROW, I. GASTON, C.D. SOOYand J.A. LEVY, J. Immunol. 140 941-943 (1988). B. BANAPOUR, K. ROSENTHAL, L. RABIN, V. SHARMA, L. YOUNG, J. FERNANDEZ, E. ENGELMAN, M. McGRATH, G. REYES and J.D. LIFSON, J. Immunol. 139 4027-4033 (1987). M.K. GORNY, V. GIANAKAKOS and S. ZOLLA-PAZNER, The fourth annual AIDS meeting [abstract 1596] Sweden, (1988). C.M. CROCE, M. SHANDER, J. MARTINIS, L. CICUREL, G.G. D'ANCONA and H. KOPROWSKI, Eur. J. Immunol. I0 486-488 (1987). S. FUJINAGA, T. SUGANO, Y. MA~SUMOTO, Y. MASUHOand R. MORI, J. Infect. Dis. 155 45-53 (1987). T. SUGANO, Y. MATSUMOTO, C. MIYAMOTOand Y. MASUHO, Eur. J. Immunol. 17 359-364 (1987). Y. MATSUMOTO, T. SUGANO, C. MIYAMOTOand Y. MASUHO, Biochem. Biophys. Res. Comm. 137 273-280 (1986). R.R. REDFIELD and D.S. BURKE, Sci. American 259 90-99 (1988). J.S. McDOUGAL, M.S. KENNEDY, J.K.A. NICHOLSON, T.J. SPIRA, H.W. JAFFE, J.E. KAPLAN, D.B. FISHBEIN, P. O'MALLEY, C.H. ALOISO, C.M. BLACK, M. HUBBARD and C.B. REINER, J. Clin. Invest. 80 316-324 (1987). J. GOUDSMIT, M.A. LANGE, D.A. PAUL and J. DAWSON, J. Infect. Dis. 155 558-560 (1987). J.A. LEVY, The f i f t h annual AIDS conference [abstract M.C.O.I] Montreal, Canada (1989).