- Email: [email protected]
Epitope-vaccines: A New Strategy to Induce High Levels of Neutralizing Antibodies Against HIV-1
Immunobiology (1999/2000) 201, 323-331 © 2000 Urban & Fischer Verlag
http://www.urbanfischer.de/journals/immunobiol
lLaboratory of Immunology, Research Center for Medical Research and School of Life Science and Engineering, Tsinghua University, Beijing, P.R. China, 2Institute of Hygiene, Ludwig-BoltzmannInstitute for AIDS-Research, University Innsbruck, Innsbruck, Austria
Epitope-vaccines: A New Strategy to Induce High Levels of Neutralizing Antibodies Against HIV- 1 YI XIAOI, MAOFU LIAOI, YUN LuI, MANFRED
P. DIERICH2, and YING-HUA CHEN l
Received June 1, 1999 . Accepted in revised form September 13, 1999
Abstract Based on the experimental evidence that gp120 subunit vaccine did not protect individuals from HIV-1 infection, we suggested that epitope-vaccines of HIV-1 gp41 may be a new strategy to induce high levels of neutralizing antibodies against HIV-1, and characterised immunogenicity of epitope-vaccines. Two epitopes, RILAVERYLKD-epitope (aa586-596) on the N-domain and ELDKWA-epitope (aa669-674) on the C-domain of gp41, were demonstrated by us and others to induce protective activity. After vaccination course, the RILAVERYLKD-dimer epitope-vaccine [C(RILAVERYLKDG)2-BSA] induced strong epitope-specific antibody response by about 1:25,600 dilution, and the ELDKWA-tetramer epitope-vaccine [C-(ELDKWAG)4-BSA] could yet induce strong antibody response to ELDKWA-epitope by 1:12,800-25,600 dilution of antisera in mice, while rgp41 subunit vaccine induced very weak antibody response to both epitopes (1:400). In rabbit experiments, the titres of ELDKWA-epitope-specific antibody induced by ELDKWA-epitope-vaccine [C-(ELDKWAG)4-BSA] reached to 1:6,400, while rgp41 subunit vaccine induced very weak antibody response to this epitope and to P1 and P2 peptides (1 :400). Moreover, the ELDKWA-epitope-specific antibodies in mice and rabbit antisera induced by epitope-vaccine could very strongly interact with P2 peptide sequence-corresponding to the Cdomain of gp41 (dilution by 1:25,600), and the RILAVERYLKD-epitope-specific antibodies in mice antisera induced by epitope-vaccine could also very strongly interact with P1 peptide sequence-corresponding to the N-domain of gp41 (dilution by 1:102,400). All these results provided experimental evidence that epitope-vaccine may be a new general strategy to induce high levels of neutralizing antibodies against HIV-1 or other viruses.
Introduction
The recombinant gp120 (rgp120) of HIV-1 has been used to develop and antibody-mediated subunit vaccine against HIV-1 in several laboratories. Unfortunately, CONNOR and co-workers provided direct experimental evidence that rgp 120 subunit vaccine did not protect individuals from HIV-l infection (1). 0171-2985/99-00/201/03-04-323 $ 12.00/0
324 . Y. XIAO et al.
reviewed problems of gp120 subunit vaccine, found that the failure of this vaccine results from inducing very weak neutralization activity against representative primary viral isolates, and indicated that a new strategy for developing an effective vaccine must be chosen to break through these problems (2, 3). To break through these problems, we suggest epitope-vaccine as a new strategy against HIV-l infection (4). Recent studies revealed that gp41 plays a very important role in HIV-l entry into the target cells and should be a potential component for developing an effective vaccine to protect against HIV infection. The extra-membrane structure of HIV-l gp41 can be divided into three parts, a fusion domain at the amino terminus and two regions (N- and C-domains) with a hydrophobic hepatic repeat in the middle. Recent studies suggest a new model of HIV-entry (5). The binding of gp120 to CD4 and chemokine receptor induces the conformational changes of gp41. These changes create a fusion intermediate, and then cause gp41 N - and Cdomains to snap closed, and mediate fusion of the viral membrane with cell membrane. Therefore, gp41 (especially the N- and C-domains) in principle should be a potential component for developing an effective vaccine to inhibit HIV entry. Over the last few years, it was demonstrated that two peptides (DPI07 corresponding to the N-domain and DP178 to the C-domain of gp41) have been shown to be potent inhibitors of HIV-l. The binding of DP178 to the Ndomain prevents the formation of the fusion intermediate (6). Studies by us and others provided experimental evidences that gp41 by the N - and C-domains bound a putative cellular receptor (7-9), which could be help to understand protective mechanism of the N- and C-domain-based vaccine. It was observed by us that an epitope on the N-domain could induce protective activity (10). Interestingly, the monoclonal antibody (mAb) 2F5 recognizing an epitope (ELDKWA) on the C-domain could inhibit the binding of gp41 to lymphocytes and monocytes, and neutralize HIV lab strains and even recently separated African, Asian, American and European strains from clades A, B, and E. Most of the investigated viruses were neutralized by 90% (11, 12), indicating that the neutralizing epitope recognized by 2F5 is generally exposed. Besides, the strong immunogenicity of SIV transmembrane protein gp32 (especially both domains) was demonstrated, and antibodies against both regions (homologous with the N - and C-domains on gp41) of gp32 could protect macaques from SIV infection (13), indicating that both domains could induce protective potency. Recent studies have provided experimental evidence that a fusion-competent vaccine with broad neutralization of primary isolates of HIV is associated with these fusion intermediates, which suggests that the C- and N -domains of gp41 in the fusion intermediates could contribute to the broad neutralization activity (14). Several groups demonstrated that HIV-l particles in blood of patients could be cleaned principally by the neutralizing antibodies (15). Three human mAbs (2F5, 2G12, and IgGlb12) were found to broadly neutralize primary viruses from subtypes A through E, and with quite respectable potencies. Used in combination, these mAbs are especially effective at suppressing virus replication in vitro. Unfortunately, these antibodies are rare in infected humans, and have never yet been raised by a vaccine (16). Gp120 subunit vaccine induced poor MOORE
Epitope-vaccines induce high levels of antibodies . 325
neutralization activity against representative primary viral isolates. Besides, HIV-l genetic variation is also a problem in developing an effective subunit vaccine. A multivalent SIV gp160 vaccine could protect monkeys from an autologous, clonal virus but not from viruses with a very minor sequence variant (Reviewed in Ref. 3). To break through these problems, we suggest epitope-vaccine as a new strategy against HIV-l (4). Here, we report that epitope-vaccines as a new strategy against HIV-l could induce high levels of neutralizing antibodies in animal experiments.
Materials and Methods HIV-l proteins, peptides and antibodies The RILAVERYLKD-dimer-peptide [(Pl)2-peptide: EnvlIIB aa586-596, C-(RILAVERYLKDG)2], ELDKWA-tetramer-peptide [EnvlIIB aa669-674; C-(ELDKWAG)4], Pl-peptide (a part of N-domain) [EnvlIIB aa583-599: (C-LQARILAVERYLKDQQL), P2-peptide (C-domain) (aa646-674: C-TSLIHSLIEESQNQQEKNEQELLELDKWA) of HIV-l and a control peptide [(KGGG)7-K] were commercially synthesised in Genemed Synthesis Inc. (California, USA). Peroxidase-conjugated rabbit immunoglobulins (Ig) to mouse Ig (P0260) and goat anti-rabbit Ig (P0448) were obtained from Dako (Denmark). The recombinant gp41 (rgp41, aa539-684) of HIV-l lIIB was obtained from Biotest in Germany (described in Ref. 7). The antisera to rgp41 were obtained from mice and rabbit immunized using rgp41 (10 pg/injection in mice and 80 pg/injection in rabbit) with Freunds adjuvant for 4 times, and tested in ELISA-assay. Construction of immunogen and immunization
Peptides were chemically linked to carrier protein BSA (bovine serum albumin; obtained from Sigma, USA) using different methods, including by MBS (17). After conjugation process, the precipitates were clearly observed. Each group of mice (n = 5) and rabbit (n = 3) were immunized subcutaneously in the foot pad of mice with 10 pg peptide and of rabbit with 80 pg peptide (in the conjugate) in Freund's adjuvant (CFA) (1:1 ratio) at a final volume of 50 pI. Boosters were given in incomplete Freund's adjuvant (IFA) on days 14, 28 and 42. Sera were separated. Pre-immune sera were collected before immunization. Measurement of anti-peptide antibodies
Peptide-specific antibodies in sera were detected in the enzyme-linked immunosorbent assay (ELISA). The peptides (5 pg/ml) were coated overnight on a microtiter plate at 4 ac. Nonspecific binding was blocked by incubation with 1% BSA or 0.3 % gelatin in PBS. After washing three times with PBS-Tween 20 (0.1 % Tween 20), antisera or normal serum (pre-immune serum) with different dilution were added and incubated for 1 hour at room temperature. After washing, peroxidase-conjugated rabbit anti-mouse Ig or goat anti-rabbit Ig were add. After further washing, freshly prepared 2,2-azino-di-(3-ethylbenzthiazoline sulfonate)-peroxide solution was added and the optical density was measured.
Results and Discussion Epitope-vaccine is principally based on the «principal neutralizing epitope» (PND) of antigens, and belong to a special type of synthetic peptide-vaccine. An immunodominant domain (Env aa598-609) located between N- and c-domains
326 · Y. XIAO et al. Ii (P1)2+AS1
1.8
IIP1+AS1 DCP+AS1 iii (P1)2+NS .P1+NS IICP+NS
1
1.4
(P1)2+AS1
P1+AS1
1.2
CP+AS1 (P1)2+NS
P1+NS CP+NS
~ ; W) ~
ci
a
1.3
0.8 0.6
0.4
:i c o II) ~
C
o
0.8
0.3
-0.2 .........
1:400
1:1600
1:6400
1:25600
---1
dilution Figure 1. Induction of antibodies recognizing the RILAVERYLKD-epitope and N-domain (Plpeptide) in mouse sera by immunization with C-(RILAVERYLKDG)2-BSA epitope-vaccines. ASl: antisera induced by immunization with C-(RILAVERYLKDG)2-BSA epitope-vaccine; NS: pre-immune mouse sera. (Pl)2: C-(RILAVERYLKDG)2-peptide; PI: IQARILAVERYLKDQQL-peptide; CP: control peptide [(KGGG)7-K]. The RILAVERYLKD-epitopespecific antibodies in AS1 were identified in ELISA-assay. Results from five mice per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
of gp41 was found by epitope-mapping (18). Up to now, two epitopes on gp41, the RILAVERYLKD-epitope (aa586-596) on the N-domain and ELDKWA-epitope (aa669-674) on the C-domain, proved to have significant potency to induce protective activity (4, 11, 12). To examine whether high levels of epitope-specific neutalizing antibodies against HIV-1 can be induced by epitope-vaccines, we characterized immunogenicity of both epitope on the epitope-vaccines. The RILAVERYLKD-dimer peptide [C-(RILAVERYLKDG)2J and the ELDKWA-tetramer peptide [C-(ELDKWAG)4J were synthesized and conjugated with BSA by MBS (17). After the vaccination course, antibody responses to these immunogens in mice or in rabbit were induced and evaluated by ELISA.
Epitope-vaccines induce high levels of antibodies . 327
1.4 1.2
EJ (2F5)4+AS2 IIP2+AS2 DCP+AS2 ED (2F5)4+NS .P2+NS IICP+NS
_(2F5}4+AS2
0.8
--+--P2+AS2
0.7
"""'--CP+AS2 -D-(2f5}4+NS
0.6
~ o
--$--P2+NS
0.5
-tr-CP+NS
~ 0.4
ci
d
1
0.3
:E 0.8 c 0 10
~
C
0
0.6 0.4 0.2
0 1:400 -0.2
1:1600
1:6400
1:25600
dilution
Figure 2. Induction of antibodies recognizing the ELDKWA-epitope and C-domain (P2-peptide) in mouse sera by immunization with C-(ELDKWAG)4-BSA epitope-vaccines. AS2: antisera induced by immunization with C-(ELDKWAG)4-BSA epitope-vaccine; NS: pre-immune mouse sera. (2FS)4: C-(ELDKWAG)4-peptide; P2: P2-peptide; CP: control peptide. The ELDKWA-epitope-specific antibodies in AS2 were identified in ELISA-assay. Results from five mice per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
The RIVALAVERYLKD-dimer epitope-vaccine [C-(RILAVERYLKDG)2-BSAJ induced very strong epitope-specific antibody response in mice by about 1:25,6000 dilution, and these antisera could recognize the P1-peptide up to 1:102,400 dilution (Fig. 1). The ELDKWA-tetramer epitope-vaccine [C-(ELDJWAG)4-BSAJ could yet induce strong antibody response to ELDKWA-epitope by 1:12,800-25,600 dilution, and these mouse antisera recognized the P2-peptide up to 1:25,600 (Fig. 2). The pre-immune sera did not recognize these peptides (Fig. 1 and 2). A strong recognition of control peptides by antisera can be seen in Fig. 1, and a weak recognition can be observed in Figure 2. Up to now, we can not give a clear explanation. Nevertheless, the reactivity of the antibodies in the
328
Y. XIAO et al.
0.9
II (2F5}4+AS3 11 (P1 )2+AS3 DP1+AS3 IiIP2+AS3 .CP+AS3
_C2F5}4+AS3 0.8
--'-P1+AS3
~CP+AS3
~
c o
0.4
o
0.2
.,... ci
0.7
0
_P2+AS3
0.6
It)
:E c
-+-CP1)2+AS3
0 1:4])
0.5
1:18XJ
1:&0)
1:256X)
'().2
dilution
10
~
C
0
0.3
0.1
1:400 -0.1
1:1600
1:6400
1:25600
dilution
Figure 3. Induction of antibodies recognizing the RILAVERYLKD- and ELDKWA-epitope, N -domain (Pi-peptide) and C-domain (P2 peptide) in mouse sera by immunization with rgp4i subunit vaccine. AS3: antisera induced by immunization with rgp4i subunit-vaccine. (Pi )2: (Pi)2-peptide; (2FS)4: C-(ELDKWAG)4-peptide; Pi: Pi-peptide; P2: P2-Peptide; CP: control peptide. The epitope-specific antibodies in AS3 were identified in ELISA-assay. Results from five mice per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
antisera with the HIV-peptides can be very clearly observed at the dilution 1:6,400 and even at 1:25,600 in comparison with the control peptide. These results indicate that both epitope-vaccines could induce high levels of antibodies recognizing neutralizing-epitope on N - and C-domains of gp41. Rgp41 subunit vaccine induced very weak antibody response to both epitopes (1:400), and the anti-rgp41 sera could bind PI and P2 weakly (Fig. 3), which suggests a very weak immunogenicity of the ELDKWA-epitope on the rgp41. The N- and Cdomains showed also little immunodominant by epitope-mapping (18). In rabbit experiments, the titres of ELDKWA-epitope-specific antibody induced by ELDKWA-epitope-vaccine [C-(ELDKWAG)4-BSA] reached to
Epitope-vaccines induce high levels of antibodies · 329
2.3
• (2F5)4+AS4 II P2+AS4 o (2F5)4+AS5 mp2+AS5 • (2F5)4+NS IIP2+NS
I
1.5
:E1 C
~
1.8
(2F5)4+AS4 P2+AS4 (2f5)-4+ASS P2+ASS
(2F5)4+NS P2+NS
~
ci 5 do.
:E c 1.3 o It) ~
C
o
0.8
0.3
-0.2
1:100
1:400
1:1600
1:6400
dilution
Figure 4. Comparison of the ELDKWA-epitope-specific recognizing antibody responses in rabbits by immunization with C-(ELDKWAG)4-BSA epitope-vaccine and rgp41 subunit vaccine. AS4: rabbit sera induced by immunization with C-(ELDKWAG)4-BSA epitope-vaccine; ASS: rabbit sera induced by immunization with rgp41 subunit-vaccine; NS: pre-immune sera (pooled). (2FS)4: C-(ELDKWAG)4-peptide;P2: P2-Peptide; CP: control peptide. The epitopespecific antibodies in AS4 and ASS were identified in ELISA-assay. Results from three rabbits per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
1.6,500, while rgp41 subunit vaccine induced very weak antibody response to this epitope (1.400). Moreover, the ELDWKWA-epitope-specific antibodies in rabbit sera could strongly interact with the P2-peptide (dilution by 1:25,600) (Fig. 4). All these results provided experimental evidence that epitope-vaccine may be a new general strategy to induce high levels of neutralizing antibodies against HIV-1 or other viruses. Previous studies by us revealed that gp41 by two binding sites located in the N- and C-domains binds to T, B lymphocytes and monocytes (7,19). The mAb 2F5 recognizing the epitope ELDKWA in the C-domain could inhibit the binding of gp41 to T, B lymphocytes and monocytes (19), which indicates that the
330 · Y. XIAO et al.
broadly neutralizing potencies of mAb 2FS could be associated with blocking the interaction between gp41 and the putative cellular receptor. Recent studies by us demonstrated that gp41 by C-domain binds the putative cellular receptor protein P62 (20). Sequence analysis of primary isolates suggests that the major determinant of mAb 2FS binding corresponds to the amino-acid sequence LDKW. Naturally occurring and in vitro selected neutralization-resistant viruses contained changes in the D and K positions of the ELDKWA motif, and the amino-acid changes from D top N, D to E, and K to N caused abrogation of 2FS-binding to the ELDKWA-epitope (12). The restricted antigenic variability of the ELDKWA-epitope recognized by the neutralizing mAb 2FS enable ELDKWA-epitope to be developed as an effective epitope-vaccine against HIV-l. We suggest epitope-vaccine as a new strategy against HIV-l (4). Ideal epitopevaccine, in our strategy, will use a combination of several «principal neutralizing epitopes». The efficacy of 100 percent neutralization of heterologous primary isolates may be achieved by inducing several neutralizing antibodies in a body, if these neutralizing epitopes, or in part, are presented on these primary isolates. In the fact, in vitro, triple combination of human mAbs (2FS, 2G12, and IgGlb12) can neutralize primary HIV-l strains at concentrations around 1,000-fold lower than a polyclonal antibody preparations till being considered for passive immunotherapy. More importantly, the speed to synthesize peptides is more quick than the speed of viral mutation. Hence, epitope-vaccine as a new strategy may be used to develop an effective vaccine to induce high levels of neutralizing antibodies against HIV-l or other viruses. Acknowledgements This work was supported by the Ministry of Education, the NSFC-39770696 and NSFC39880043, the 863-Program of China (1998-164), and the Kingdream Ltd.
References 1. CONNOR, R. I., B. T. M. KORBER, B. S. GRAHAM. 1998. Immunological and Virological Analyses of Persons Infected by Human-Immunodeficiency-Virus Type-l while Participating in trials of Recombinant gp120 Subunit Vaccines. J. Virol. 72: 1552-1576. 2. MOORE, J. P. 1996. Back to primary school. Nature 376: 115. 3. MOORE, J. P., and D. R. NURTON. 1999. HIV-l neutralizing antibodies: How full is the bottle? Nature Medicine 5: 142-144. 4. CHEN, Y. H., Y. XIAO, T. Yu, M. P. DIERICH. Epitope-vaccine: a new strategy against HIV1. Immunology Today (in press). 5. CHAN, D. C., D. FASS, and J. M. BERGER. 1997. Core structure of gp41 from the HIV envelope glycoprotein. Cell 89: 263-273. 6. FURUTA, R. A., C. T. WILD, Y. WENG, and C. D. WEISe 1998. Capture of an early fusionactive conformation of HIV-l gp41. Nature Structural Biology 5: 276-279. 7. CHEN, Y. H., C. EBENBICHLER, and R. VORNHAGEN. 1992. HIV-l gp41 contains two sites for interaction with several proteins on the helper T-Iymphoid cell line, H9. AIDS 6: 533-539. 8. HENDERSON, L. A., and M. N. QURESHI. 1993. A peptide inhibitor of human immunodeficience virus infection binds to novel human cell surface polypeptides. J. BioI. Chern. 268: 15291-15297.
Epitope-vaccines induce high levels of antibodies · 331 9. CHEN, Y. H., A. CHRISTIANSEN, G. BOCK, and M. P. DIERICH. 1995. HIV-2 transmembrane protein gp36 like HIV-l gp41 binds to human lymphocytes and monocytes. AIDS 9: 1193-1194. 10. CHEN, Y. H., and M. P. DIERICH. 1998. A common epitope on gp41, IFN-a and IFN-b induces protective activity. Immunology Today 19: 586-587. 11. MUSTER, T., F. STEINDL, and M. PURTSCHER. 1993. A conserved neutralizing epitope on gp41 of human immunodeficiency virus type I. J. Virol. 67: 6642-6647. 12. PURTSCHER, M., A. TRKOLA, and A. GRASSAUER. 1996. Restricted antigenic variability of the epitope recognized by the neutralizing gp41 antibody 2F5. AIDS 10: 587-593. 13. SHAFFERMAN, A., P. B. JAHRLING, and R. E. BENVENISTE. 1991. Protection of macaques with a simian immunodeficiency virus envelope peptide vaccine based on conserved human immunodeficiency virus type 1 sequences. Proc. Natl. Acad. Sci. USA 88: 7126-7130. 14. LACASSE, R. A., K. E. FOLLIS, M. TRAHEY, J. D. SCARBOROUGH, D. R. LITTMAN, and J. H. NUNBERG. 1999. Fusion-competent vaccines: broad neutralization of primary isolates of HIV. Science 283: 357-362. 15. IGARASHI, T., C. BROWN, A. AZADEGAN, N. HAIGWOOD, D. DIMITROV, M. A. MARTIN, and R. SHIBATA. 1999. Human immunodeficiency virus type 1 neutralizing antibodies accelerate clearance of cell-free virions from blood plasma. Nature Med. 5: 211-216. 16. MONTEFIORI, D. C., K. A. REIMANN, M. S. WYAND, K. MANSON, M. G. LEWIs, R. G. COLLMAN, J. G. SODROSKI, D. P. BOLOGNESI, and N. L. LETVIN. 1998. Neutralizing antibodies in sera from macaques infected with chimeric simian-human immunodeficiency virus containing n the envelope glycoproteins of either a laboratory-adapted variant or a primary isolate of human immunodeficiency type 1. J. Virol. 72: 3427-3431. 17. CHONG, P., N. CHAN, and A. KANDIL. 1997. A strategy for rational design of fully synthetic glycopeptide conjugate vaccines. Infection and Immunity 65: 4918-4925. 18. GNANN, J. W. J. R., J. A. NElSON, and M. B. A. OLDSTONEE. 1987. Fine mapping of an immunodominant domain in the transmembrane glycoprotein of human immunodeficiency virus. J. Virol. 61: 2639-2641. 19. CHEN, Y. H., and M. P. DIERICH. 1996. Identification of a second site in HIV-l gp41 mediating binding to cells. Immunology Letters. 52: 153-156. 20. CHEN, Y. H., Y. XIAO, W. Wu, J. YANG, S. SUI, and M. P. DIERICH. 1999. The C domain of HIV-l gp41 binds the putative cellular receptor protein P62. AIDS 13: 1021-1024. Prof. Dr. YING-HuA CHEN, Laboratory of Immunology, Research Center for Medical Research and School of Life Science and Engineering, Tsinghua University, Beijing, P.R. China, Fax: +8610-62 78 55 05, e-mail: [email protected]
http://www.urbanfischer.de/journals/immunobiol
lLaboratory of Immunology, Research Center for Medical Research and School of Life Science and Engineering, Tsinghua University, Beijing, P.R. China, 2Institute of Hygiene, Ludwig-BoltzmannInstitute for AIDS-Research, University Innsbruck, Innsbruck, Austria
Epitope-vaccines: A New Strategy to Induce High Levels of Neutralizing Antibodies Against HIV- 1 YI XIAOI, MAOFU LIAOI, YUN LuI, MANFRED
P. DIERICH2, and YING-HUA CHEN l
Received June 1, 1999 . Accepted in revised form September 13, 1999
Abstract Based on the experimental evidence that gp120 subunit vaccine did not protect individuals from HIV-1 infection, we suggested that epitope-vaccines of HIV-1 gp41 may be a new strategy to induce high levels of neutralizing antibodies against HIV-1, and characterised immunogenicity of epitope-vaccines. Two epitopes, RILAVERYLKD-epitope (aa586-596) on the N-domain and ELDKWA-epitope (aa669-674) on the C-domain of gp41, were demonstrated by us and others to induce protective activity. After vaccination course, the RILAVERYLKD-dimer epitope-vaccine [C(RILAVERYLKDG)2-BSA] induced strong epitope-specific antibody response by about 1:25,600 dilution, and the ELDKWA-tetramer epitope-vaccine [C-(ELDKWAG)4-BSA] could yet induce strong antibody response to ELDKWA-epitope by 1:12,800-25,600 dilution of antisera in mice, while rgp41 subunit vaccine induced very weak antibody response to both epitopes (1:400). In rabbit experiments, the titres of ELDKWA-epitope-specific antibody induced by ELDKWA-epitope-vaccine [C-(ELDKWAG)4-BSA] reached to 1:6,400, while rgp41 subunit vaccine induced very weak antibody response to this epitope and to P1 and P2 peptides (1 :400). Moreover, the ELDKWA-epitope-specific antibodies in mice and rabbit antisera induced by epitope-vaccine could very strongly interact with P2 peptide sequence-corresponding to the Cdomain of gp41 (dilution by 1:25,600), and the RILAVERYLKD-epitope-specific antibodies in mice antisera induced by epitope-vaccine could also very strongly interact with P1 peptide sequence-corresponding to the N-domain of gp41 (dilution by 1:102,400). All these results provided experimental evidence that epitope-vaccine may be a new general strategy to induce high levels of neutralizing antibodies against HIV-1 or other viruses.
Introduction
The recombinant gp120 (rgp120) of HIV-1 has been used to develop and antibody-mediated subunit vaccine against HIV-1 in several laboratories. Unfortunately, CONNOR and co-workers provided direct experimental evidence that rgp 120 subunit vaccine did not protect individuals from HIV-l infection (1). 0171-2985/99-00/201/03-04-323 $ 12.00/0
324 . Y. XIAO et al.
reviewed problems of gp120 subunit vaccine, found that the failure of this vaccine results from inducing very weak neutralization activity against representative primary viral isolates, and indicated that a new strategy for developing an effective vaccine must be chosen to break through these problems (2, 3). To break through these problems, we suggest epitope-vaccine as a new strategy against HIV-l infection (4). Recent studies revealed that gp41 plays a very important role in HIV-l entry into the target cells and should be a potential component for developing an effective vaccine to protect against HIV infection. The extra-membrane structure of HIV-l gp41 can be divided into three parts, a fusion domain at the amino terminus and two regions (N- and C-domains) with a hydrophobic hepatic repeat in the middle. Recent studies suggest a new model of HIV-entry (5). The binding of gp120 to CD4 and chemokine receptor induces the conformational changes of gp41. These changes create a fusion intermediate, and then cause gp41 N - and Cdomains to snap closed, and mediate fusion of the viral membrane with cell membrane. Therefore, gp41 (especially the N- and C-domains) in principle should be a potential component for developing an effective vaccine to inhibit HIV entry. Over the last few years, it was demonstrated that two peptides (DPI07 corresponding to the N-domain and DP178 to the C-domain of gp41) have been shown to be potent inhibitors of HIV-l. The binding of DP178 to the Ndomain prevents the formation of the fusion intermediate (6). Studies by us and others provided experimental evidences that gp41 by the N - and C-domains bound a putative cellular receptor (7-9), which could be help to understand protective mechanism of the N- and C-domain-based vaccine. It was observed by us that an epitope on the N-domain could induce protective activity (10). Interestingly, the monoclonal antibody (mAb) 2F5 recognizing an epitope (ELDKWA) on the C-domain could inhibit the binding of gp41 to lymphocytes and monocytes, and neutralize HIV lab strains and even recently separated African, Asian, American and European strains from clades A, B, and E. Most of the investigated viruses were neutralized by 90% (11, 12), indicating that the neutralizing epitope recognized by 2F5 is generally exposed. Besides, the strong immunogenicity of SIV transmembrane protein gp32 (especially both domains) was demonstrated, and antibodies against both regions (homologous with the N - and C-domains on gp41) of gp32 could protect macaques from SIV infection (13), indicating that both domains could induce protective potency. Recent studies have provided experimental evidence that a fusion-competent vaccine with broad neutralization of primary isolates of HIV is associated with these fusion intermediates, which suggests that the C- and N -domains of gp41 in the fusion intermediates could contribute to the broad neutralization activity (14). Several groups demonstrated that HIV-l particles in blood of patients could be cleaned principally by the neutralizing antibodies (15). Three human mAbs (2F5, 2G12, and IgGlb12) were found to broadly neutralize primary viruses from subtypes A through E, and with quite respectable potencies. Used in combination, these mAbs are especially effective at suppressing virus replication in vitro. Unfortunately, these antibodies are rare in infected humans, and have never yet been raised by a vaccine (16). Gp120 subunit vaccine induced poor MOORE
Epitope-vaccines induce high levels of antibodies . 325
neutralization activity against representative primary viral isolates. Besides, HIV-l genetic variation is also a problem in developing an effective subunit vaccine. A multivalent SIV gp160 vaccine could protect monkeys from an autologous, clonal virus but not from viruses with a very minor sequence variant (Reviewed in Ref. 3). To break through these problems, we suggest epitope-vaccine as a new strategy against HIV-l (4). Here, we report that epitope-vaccines as a new strategy against HIV-l could induce high levels of neutralizing antibodies in animal experiments.
Materials and Methods HIV-l proteins, peptides and antibodies The RILAVERYLKD-dimer-peptide [(Pl)2-peptide: EnvlIIB aa586-596, C-(RILAVERYLKDG)2], ELDKWA-tetramer-peptide [EnvlIIB aa669-674; C-(ELDKWAG)4], Pl-peptide (a part of N-domain) [EnvlIIB aa583-599: (C-LQARILAVERYLKDQQL), P2-peptide (C-domain) (aa646-674: C-TSLIHSLIEESQNQQEKNEQELLELDKWA) of HIV-l and a control peptide [(KGGG)7-K] were commercially synthesised in Genemed Synthesis Inc. (California, USA). Peroxidase-conjugated rabbit immunoglobulins (Ig) to mouse Ig (P0260) and goat anti-rabbit Ig (P0448) were obtained from Dako (Denmark). The recombinant gp41 (rgp41, aa539-684) of HIV-l lIIB was obtained from Biotest in Germany (described in Ref. 7). The antisera to rgp41 were obtained from mice and rabbit immunized using rgp41 (10 pg/injection in mice and 80 pg/injection in rabbit) with Freunds adjuvant for 4 times, and tested in ELISA-assay. Construction of immunogen and immunization
Peptides were chemically linked to carrier protein BSA (bovine serum albumin; obtained from Sigma, USA) using different methods, including by MBS (17). After conjugation process, the precipitates were clearly observed. Each group of mice (n = 5) and rabbit (n = 3) were immunized subcutaneously in the foot pad of mice with 10 pg peptide and of rabbit with 80 pg peptide (in the conjugate) in Freund's adjuvant (CFA) (1:1 ratio) at a final volume of 50 pI. Boosters were given in incomplete Freund's adjuvant (IFA) on days 14, 28 and 42. Sera were separated. Pre-immune sera were collected before immunization. Measurement of anti-peptide antibodies
Peptide-specific antibodies in sera were detected in the enzyme-linked immunosorbent assay (ELISA). The peptides (5 pg/ml) were coated overnight on a microtiter plate at 4 ac. Nonspecific binding was blocked by incubation with 1% BSA or 0.3 % gelatin in PBS. After washing three times with PBS-Tween 20 (0.1 % Tween 20), antisera or normal serum (pre-immune serum) with different dilution were added and incubated for 1 hour at room temperature. After washing, peroxidase-conjugated rabbit anti-mouse Ig or goat anti-rabbit Ig were add. After further washing, freshly prepared 2,2-azino-di-(3-ethylbenzthiazoline sulfonate)-peroxide solution was added and the optical density was measured.
Results and Discussion Epitope-vaccine is principally based on the «principal neutralizing epitope» (PND) of antigens, and belong to a special type of synthetic peptide-vaccine. An immunodominant domain (Env aa598-609) located between N- and c-domains
326 · Y. XIAO et al. Ii (P1)2+AS1
1.8
IIP1+AS1 DCP+AS1 iii (P1)2+NS .P1+NS IICP+NS
1
1.4
(P1)2+AS1
P1+AS1
1.2
CP+AS1 (P1)2+NS
P1+NS CP+NS
~ ; W) ~
ci
a
1.3
0.8 0.6
0.4
:i c o II) ~
C
o
0.8
0.3
-0.2 .........
1:400
1:1600
1:6400
1:25600
---1
dilution Figure 1. Induction of antibodies recognizing the RILAVERYLKD-epitope and N-domain (Plpeptide) in mouse sera by immunization with C-(RILAVERYLKDG)2-BSA epitope-vaccines. ASl: antisera induced by immunization with C-(RILAVERYLKDG)2-BSA epitope-vaccine; NS: pre-immune mouse sera. (Pl)2: C-(RILAVERYLKDG)2-peptide; PI: IQARILAVERYLKDQQL-peptide; CP: control peptide [(KGGG)7-K]. The RILAVERYLKD-epitopespecific antibodies in AS1 were identified in ELISA-assay. Results from five mice per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
of gp41 was found by epitope-mapping (18). Up to now, two epitopes on gp41, the RILAVERYLKD-epitope (aa586-596) on the N-domain and ELDKWA-epitope (aa669-674) on the C-domain, proved to have significant potency to induce protective activity (4, 11, 12). To examine whether high levels of epitope-specific neutalizing antibodies against HIV-1 can be induced by epitope-vaccines, we characterized immunogenicity of both epitope on the epitope-vaccines. The RILAVERYLKD-dimer peptide [C-(RILAVERYLKDG)2J and the ELDKWA-tetramer peptide [C-(ELDKWAG)4J were synthesized and conjugated with BSA by MBS (17). After the vaccination course, antibody responses to these immunogens in mice or in rabbit were induced and evaluated by ELISA.
Epitope-vaccines induce high levels of antibodies . 327
1.4 1.2
EJ (2F5)4+AS2 IIP2+AS2 DCP+AS2 ED (2F5)4+NS .P2+NS IICP+NS
_(2F5}4+AS2
0.8
--+--P2+AS2
0.7
"""'--CP+AS2 -D-(2f5}4+NS
0.6
~ o
--$--P2+NS
0.5
-tr-CP+NS
~ 0.4
ci
d
1
0.3
:E 0.8 c 0 10
~
C
0
0.6 0.4 0.2
0 1:400 -0.2
1:1600
1:6400
1:25600
dilution
Figure 2. Induction of antibodies recognizing the ELDKWA-epitope and C-domain (P2-peptide) in mouse sera by immunization with C-(ELDKWAG)4-BSA epitope-vaccines. AS2: antisera induced by immunization with C-(ELDKWAG)4-BSA epitope-vaccine; NS: pre-immune mouse sera. (2FS)4: C-(ELDKWAG)4-peptide; P2: P2-peptide; CP: control peptide. The ELDKWA-epitope-specific antibodies in AS2 were identified in ELISA-assay. Results from five mice per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
The RIVALAVERYLKD-dimer epitope-vaccine [C-(RILAVERYLKDG)2-BSAJ induced very strong epitope-specific antibody response in mice by about 1:25,6000 dilution, and these antisera could recognize the P1-peptide up to 1:102,400 dilution (Fig. 1). The ELDKWA-tetramer epitope-vaccine [C-(ELDJWAG)4-BSAJ could yet induce strong antibody response to ELDKWA-epitope by 1:12,800-25,600 dilution, and these mouse antisera recognized the P2-peptide up to 1:25,600 (Fig. 2). The pre-immune sera did not recognize these peptides (Fig. 1 and 2). A strong recognition of control peptides by antisera can be seen in Fig. 1, and a weak recognition can be observed in Figure 2. Up to now, we can not give a clear explanation. Nevertheless, the reactivity of the antibodies in the
328
Y. XIAO et al.
0.9
II (2F5}4+AS3 11 (P1 )2+AS3 DP1+AS3 IiIP2+AS3 .CP+AS3
_C2F5}4+AS3 0.8
--'-P1+AS3
~CP+AS3
~
c o
0.4
o
0.2
.,... ci
0.7
0
_P2+AS3
0.6
It)
:E c
-+-CP1)2+AS3
0 1:4])
0.5
1:18XJ
1:&0)
1:256X)
'().2
dilution
10
~
C
0
0.3
0.1
1:400 -0.1
1:1600
1:6400
1:25600
dilution
Figure 3. Induction of antibodies recognizing the RILAVERYLKD- and ELDKWA-epitope, N -domain (Pi-peptide) and C-domain (P2 peptide) in mouse sera by immunization with rgp4i subunit vaccine. AS3: antisera induced by immunization with rgp4i subunit-vaccine. (Pi )2: (Pi)2-peptide; (2FS)4: C-(ELDKWAG)4-peptide; Pi: Pi-peptide; P2: P2-Peptide; CP: control peptide. The epitope-specific antibodies in AS3 were identified in ELISA-assay. Results from five mice per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
antisera with the HIV-peptides can be very clearly observed at the dilution 1:6,400 and even at 1:25,600 in comparison with the control peptide. These results indicate that both epitope-vaccines could induce high levels of antibodies recognizing neutralizing-epitope on N - and C-domains of gp41. Rgp41 subunit vaccine induced very weak antibody response to both epitopes (1:400), and the anti-rgp41 sera could bind PI and P2 weakly (Fig. 3), which suggests a very weak immunogenicity of the ELDKWA-epitope on the rgp41. The N- and Cdomains showed also little immunodominant by epitope-mapping (18). In rabbit experiments, the titres of ELDKWA-epitope-specific antibody induced by ELDKWA-epitope-vaccine [C-(ELDKWAG)4-BSA] reached to
Epitope-vaccines induce high levels of antibodies · 329
2.3
• (2F5)4+AS4 II P2+AS4 o (2F5)4+AS5 mp2+AS5 • (2F5)4+NS IIP2+NS
I
1.5
:E1 C
~
1.8
(2F5)4+AS4 P2+AS4 (2f5)-4+ASS P2+ASS
(2F5)4+NS P2+NS
~
ci 5 do.
:E c 1.3 o It) ~
C
o
0.8
0.3
-0.2
1:100
1:400
1:1600
1:6400
dilution
Figure 4. Comparison of the ELDKWA-epitope-specific recognizing antibody responses in rabbits by immunization with C-(ELDKWAG)4-BSA epitope-vaccine and rgp41 subunit vaccine. AS4: rabbit sera induced by immunization with C-(ELDKWAG)4-BSA epitope-vaccine; ASS: rabbit sera induced by immunization with rgp41 subunit-vaccine; NS: pre-immune sera (pooled). (2FS)4: C-(ELDKWAG)4-peptide;P2: P2-Peptide; CP: control peptide. The epitopespecific antibodies in AS4 and ASS were identified in ELISA-assay. Results from three rabbits per group were expressed as the mean. The figure with bars and curves shows the data from one of three separate experiments.
1.6,500, while rgp41 subunit vaccine induced very weak antibody response to this epitope (1.400). Moreover, the ELDWKWA-epitope-specific antibodies in rabbit sera could strongly interact with the P2-peptide (dilution by 1:25,600) (Fig. 4). All these results provided experimental evidence that epitope-vaccine may be a new general strategy to induce high levels of neutralizing antibodies against HIV-1 or other viruses. Previous studies by us revealed that gp41 by two binding sites located in the N- and C-domains binds to T, B lymphocytes and monocytes (7,19). The mAb 2F5 recognizing the epitope ELDKWA in the C-domain could inhibit the binding of gp41 to T, B lymphocytes and monocytes (19), which indicates that the
330 · Y. XIAO et al.
broadly neutralizing potencies of mAb 2FS could be associated with blocking the interaction between gp41 and the putative cellular receptor. Recent studies by us demonstrated that gp41 by C-domain binds the putative cellular receptor protein P62 (20). Sequence analysis of primary isolates suggests that the major determinant of mAb 2FS binding corresponds to the amino-acid sequence LDKW. Naturally occurring and in vitro selected neutralization-resistant viruses contained changes in the D and K positions of the ELDKWA motif, and the amino-acid changes from D top N, D to E, and K to N caused abrogation of 2FS-binding to the ELDKWA-epitope (12). The restricted antigenic variability of the ELDKWA-epitope recognized by the neutralizing mAb 2FS enable ELDKWA-epitope to be developed as an effective epitope-vaccine against HIV-l. We suggest epitope-vaccine as a new strategy against HIV-l (4). Ideal epitopevaccine, in our strategy, will use a combination of several «principal neutralizing epitopes». The efficacy of 100 percent neutralization of heterologous primary isolates may be achieved by inducing several neutralizing antibodies in a body, if these neutralizing epitopes, or in part, are presented on these primary isolates. In the fact, in vitro, triple combination of human mAbs (2FS, 2G12, and IgGlb12) can neutralize primary HIV-l strains at concentrations around 1,000-fold lower than a polyclonal antibody preparations till being considered for passive immunotherapy. More importantly, the speed to synthesize peptides is more quick than the speed of viral mutation. Hence, epitope-vaccine as a new strategy may be used to develop an effective vaccine to induce high levels of neutralizing antibodies against HIV-l or other viruses. Acknowledgements This work was supported by the Ministry of Education, the NSFC-39770696 and NSFC39880043, the 863-Program of China (1998-164), and the Kingdream Ltd.
References 1. CONNOR, R. I., B. T. M. KORBER, B. S. GRAHAM. 1998. Immunological and Virological Analyses of Persons Infected by Human-Immunodeficiency-Virus Type-l while Participating in trials of Recombinant gp120 Subunit Vaccines. J. Virol. 72: 1552-1576. 2. MOORE, J. P. 1996. Back to primary school. Nature 376: 115. 3. MOORE, J. P., and D. R. NURTON. 1999. HIV-l neutralizing antibodies: How full is the bottle? Nature Medicine 5: 142-144. 4. CHEN, Y. H., Y. XIAO, T. Yu, M. P. DIERICH. Epitope-vaccine: a new strategy against HIV1. Immunology Today (in press). 5. CHAN, D. C., D. FASS, and J. M. BERGER. 1997. Core structure of gp41 from the HIV envelope glycoprotein. Cell 89: 263-273. 6. FURUTA, R. A., C. T. WILD, Y. WENG, and C. D. WEISe 1998. Capture of an early fusionactive conformation of HIV-l gp41. Nature Structural Biology 5: 276-279. 7. CHEN, Y. H., C. EBENBICHLER, and R. VORNHAGEN. 1992. HIV-l gp41 contains two sites for interaction with several proteins on the helper T-Iymphoid cell line, H9. AIDS 6: 533-539. 8. HENDERSON, L. A., and M. N. QURESHI. 1993. A peptide inhibitor of human immunodeficience virus infection binds to novel human cell surface polypeptides. J. BioI. Chern. 268: 15291-15297.
Epitope-vaccines induce high levels of antibodies · 331 9. CHEN, Y. H., A. CHRISTIANSEN, G. BOCK, and M. P. DIERICH. 1995. HIV-2 transmembrane protein gp36 like HIV-l gp41 binds to human lymphocytes and monocytes. AIDS 9: 1193-1194. 10. CHEN, Y. H., and M. P. DIERICH. 1998. A common epitope on gp41, IFN-a and IFN-b induces protective activity. Immunology Today 19: 586-587. 11. MUSTER, T., F. STEINDL, and M. PURTSCHER. 1993. A conserved neutralizing epitope on gp41 of human immunodeficiency virus type I. J. Virol. 67: 6642-6647. 12. PURTSCHER, M., A. TRKOLA, and A. GRASSAUER. 1996. Restricted antigenic variability of the epitope recognized by the neutralizing gp41 antibody 2F5. AIDS 10: 587-593. 13. SHAFFERMAN, A., P. B. JAHRLING, and R. E. BENVENISTE. 1991. Protection of macaques with a simian immunodeficiency virus envelope peptide vaccine based on conserved human immunodeficiency virus type 1 sequences. Proc. Natl. Acad. Sci. USA 88: 7126-7130. 14. LACASSE, R. A., K. E. FOLLIS, M. TRAHEY, J. D. SCARBOROUGH, D. R. LITTMAN, and J. H. NUNBERG. 1999. Fusion-competent vaccines: broad neutralization of primary isolates of HIV. Science 283: 357-362. 15. IGARASHI, T., C. BROWN, A. AZADEGAN, N. HAIGWOOD, D. DIMITROV, M. A. MARTIN, and R. SHIBATA. 1999. Human immunodeficiency virus type 1 neutralizing antibodies accelerate clearance of cell-free virions from blood plasma. Nature Med. 5: 211-216. 16. MONTEFIORI, D. C., K. A. REIMANN, M. S. WYAND, K. MANSON, M. G. LEWIs, R. G. COLLMAN, J. G. SODROSKI, D. P. BOLOGNESI, and N. L. LETVIN. 1998. Neutralizing antibodies in sera from macaques infected with chimeric simian-human immunodeficiency virus containing n the envelope glycoproteins of either a laboratory-adapted variant or a primary isolate of human immunodeficiency type 1. J. Virol. 72: 3427-3431. 17. CHONG, P., N. CHAN, and A. KANDIL. 1997. A strategy for rational design of fully synthetic glycopeptide conjugate vaccines. Infection and Immunity 65: 4918-4925. 18. GNANN, J. W. J. R., J. A. NElSON, and M. B. A. OLDSTONEE. 1987. Fine mapping of an immunodominant domain in the transmembrane glycoprotein of human immunodeficiency virus. J. Virol. 61: 2639-2641. 19. CHEN, Y. H., and M. P. DIERICH. 1996. Identification of a second site in HIV-l gp41 mediating binding to cells. Immunology Letters. 52: 153-156. 20. CHEN, Y. H., Y. XIAO, W. Wu, J. YANG, S. SUI, and M. P. DIERICH. 1999. The C domain of HIV-l gp41 binds the putative cellular receptor protein P62. AIDS 13: 1021-1024. Prof. Dr. YING-HuA CHEN, Laboratory of Immunology, Research Center for Medical Research and School of Life Science and Engineering, Tsinghua University, Beijing, P.R. China, Fax: +8610-62 78 55 05, e-mail: [email protected]