12ème Symposium International sur le VIH
et les maladies infectieuses émergentes
Toulon juin 2002

2.c. The anti-beta 2 microglobulin (b2m) cellular AIDS vaccine revisited:
Molecular homology between
b2m and HIV-1 pol p51 and gag p17 (HGP 30) viral epitopes.

2.c. The anti-beta 2 microglobulin (b2m) cellular AIDS vaccine revisited:
Molecular homology between b2m and HIV-1 pol p51 and gag p17 (HGP 30) viral epitopes.

EXTENDED VERSION /VERSION LONGUE

* Association Positifs, BP 230, 75865 Paris 18, France. E-mail : caprani@ccr.jussieu.fr, positifs@positifs.org.
¹ University Paris-Sud, Faculty of Pharmacy, Therapeutic Chemistry, 92290 Chatenay Malabry, France.
Correspondence : 31, Av du Bois 92290 Chatenay Malabry, France. E-mail : mkg_tran@yahoo.fr.
² Centre Médical Europe, 44, rue d'Amsterdam, 75009 Paris, France.

HIV VACCINE CHALLENGE

HIV-1 vaccine remains a very difficult challenge, owing to the fantastic variability of the virus. A single patient contains a flurry of pseudotypes in his blood, and the number of different strains can reach the hudge number of a million of different sequences. Such a challenge is impossible to reach, if pharmaceutical companies intend to design 1 million of vaccines for only one illness. We focused primarily on this major difficulty, namely to look for conserved or highly conserved sequences in HIV-1 genome, with the hope that such an approach would limit the infectivity of HIV-1. The second help is provided by naturally observed resistance to HIV-1, whether with exposed uninfected seronegative prostitutes or long-term non progressors. HIV-1 can escape the immune response of the patient by many of the following mechanisms:

The absence of anti-CD4 binding site (Cordonnier A., 1989) antibodies, or their presence at a later stage (too late to avoid intra-cellular infection) at too low levels or affinity [the epitope is a " canyon "-like (Wang J.H. , 1990), similar to rhinovirus receptor site, instead of a protruding spike easily recognized by the immune system ; only camel antibodies recognized " canyon " epitopes(Desmyter A., 1996) ; human antibodies did not].

The absence of neutralizing antibodies against the V3 loop of the actual infectious, already an escape mutant virus (Schreiber M., 1997) (instead, the anti-V3 loop antibodies are directed to the preceeding strain, so the immune system is always one war late).
The enhancing properties of facilitating anti-V3 loop (Levy J.A., 2000) and anti-gp41 (Robinson W.E. Jr, 1991) antibodies render deleterious any approach using these epitopes, whether or not included in an HIV-1 larger complete protein.

For cytotoxic T lymphocytes (CTLs), the down-regulation of major histocompatibility (MHC) class I (except HLA-C, which allows HIV-1 to escape also the natural killer NK cells) from the cell membrane by HIV-1 Nef (Piguet V., 1999) renders the objective of CTL efficiency dramatically problematic. If Nef is continuously active [Nef is by far the most abundant transcript in infected cells (H9 cells, lymphocytes, macrophages) making 80% amount of mRNA, versus 18% for Rev and only 2% for Tat (Smythe J.A., 1992)], all efforts to obtain CTL and NK killer cell responses are condemned per advance to fail in pure perte. Thus an absolute requirement is to simultaneously neutralize Nef to avoid HLA down-regulation and CTL inefficiency.

The preliminary hopes are the sommation of a few epitopes in a coalition of, for example, 5 or 6 or more if possible short sequences, all of them being highly conserved, to avoid HIV-1 mutation escape, and create a simultaneous attack at many fronts against the virus. If an additive effect is obtained (i.e. ~ 5%+30%+20%+15+É ~ optimistically 70%), the result will attain a level high enough to be considered satisfying (~70% protection), although not able to cure completely the patient (sterilizing vaccine). The advantage of small (~6 residues long) selected epitopes, instead of a big (~hundreds of residues) protein, is the absence of nocivity (no enhancement by facilitating antibodies). However, the haptenic character of these epitopes makes them non immunogenic, if not covalently linked to a carrier. Our objective is to determine what is the functional role of each sequence (gp41 is an Interferon-beta) (Tran M.K.G., 2002), and its implication in the pathology observed (Gag p24 and aphthous ulcer) (Tran M.K.G., 2002). The following sequences are selected for a preliminary partial vaccine:

Sequence 1: Lopalco's gp41 sequence, in exposed uninfected seronegative prostitutes (mucosal IgA) (Pastori C., 2000 ; Mazzoli. S., 1999).
Sequence 2: Lopalco L. found also a cellular CCR5 epitope, that we found mimetic of a viral post-V3 loop (Tran M.K.G., 2001)
Sequence 3: Belec's sequence,. in exposed unifected prostitutes (mucosal IgA)
Sequence 4: Harrer's sequence, in a long-term non progressor (CTL)
Sequence 5: Rowland-Jones' sequences, in exposed uninfected prostitutes (CTL).
Sequence 6: Chermann's b2-microglobulin-like (modified) sequence, in non-progressors (IgG).

A cellular AIDS vaccine with a b2-microglobulin epitope called R7V 3-RTPKIQV-9 was proposed by Galea P. and Le Contel C. (1999) ; its rationale was that HIV-1, during its budding, entrapped various membrane proteins in its envelope, such as the MHC class I and II, as well as b2-microglobulin. So they proposed to vaccine with b2-microglobulin itself, present in 100% HIV-1 strains, with the speculation that the virus hypervariability would be avoided. Interestingly, anti-R7V antibodies neutralized primary strains. By an ELISA test, antibodies against R7V were found in 70% of non progressors (n = 94), versus 35% of progressors (n = 142) and 22% in controls (n = 270). Rabbits immunized with R7V. We studied here particularly the work on beta 2 microglobulin linked to keyhole limpet develop in their sera antibodies with neutralizing and precipitating properties. No deleterious auto-immunisation was found, assessing of the apparent safety of these anti-b2-microglobulin antibodies.

OBJECTIVE.

We try to demonstrate here that another concept could explain the presence of neutralizing anti-R7V antibodies. This concept is molecular mimicry between b2-microglobulin and HIV-1 proteins. What is molecular mimicry ? A brief rappel is necessary :

MOLECULAR MIMICRY.

There are antecedents of molecular mimicry between members of the HLA family and viruses:

1°) For example, between HIV-1 envelope gp120 sequence VVSTQLLLNG and HLA-class II beta chain sequence VVSTxLIxNG (x is a conservative change) (Young J.A.T., 1988)
2°) Another example of mimicry is the common heptapeptide: LGRPDED, shared by human Cytomegalovirus (HCMV) and HLA-DP3 beta (Fujinami R.S., 1988; Tran M.K.G., 1996).
3°) or between HCMV and MHC class I (Beck S., 1988), explaining HCMV binding to b2-microglobulin. Interestingly, b2-microglobulin is elevated in HCMV infections, whether or not HIV-1 associated.
4°) In the pathogenesis of spondyloarthropathies (Reiter's syndrome, dysenteria, ankylosing spondylitis) HLA-B27 and microbial pathogens [Klebsiella Pneumoniae (Schwimmbeck P., 1987), Shigella flexneri, Yersinia Enterocolitica (Yop1) (Oldstone M.B.A., 1998)] share a common epitope.

We personnally found an additional pathogen: Chlamydia trachomatis (outer membrane protein, omp) (Zanelli E., 1997), to this list, confirming the importance of this region. Curiously, we succeeded with the omp sequence TRLIDER read in the reverse sense. (Tran M.K.G., unpublished) : As can be seen here, Chlamydia REDILRT is identical (one gap) to HLA-B27.1 sequence REDLRT. There is one chance out of 207 (= 1,280,000,000) to be a pure chance, which is highly significant.

Galéa P. and Le Contel C. tried to eliminate this hypothesis (of molecular mimicry between HIV-1 and b2-microglobulin) by a standard Western-Blot, but their negative results (no details available) are not convincing at all, because standard Western-Blot is not a sensitive test for this precise purpose (HIV-1 proteins were huge molecules which could completely hide the PK motif inside their folding). Instead, it would be preferable to use a sensitized Western-Blot with only short, small peptides containing the PK (Pro Lys ; proline lysine) motif, which is then completely exposed. For example, in their analysis of Parvovirus B19 peptides, Zuffi E. (2001) used such a methodology.

So we were intrigued by the possibility that b2-microglobulin would mimick an HIV-1 sequence, and decided to screen in the Los Alamos databank (Kuiken C., 2000) with the PK motif ; our hypothesis is that if PK was found somewhere in HIV-1 proteins, it may be an immunodominant epitope, [Williams R.C. (1993) for example found that the sequence (87-97) LSQPKIVKWDR of b2-microglobulin containing PK is a major reactive antigenic site in rabbits] because proline is a residue protruding in the environment and easily recognized by antibodies, and thus induce antibodies with large degenerate cross-reactivity with any highly homologous sequence, such as R7V. In other words, the correct interpretation would be that the so-called anti-b2-microglobulin antibody is nothing else than a cross-reactive anti-HIV-1 antibody directed against an HIV-1 peptide containing PK. Such a result would be much more satisfying and classical (viral vaccine) than the so-called revolutionnary concept of cellular vaccine.

METHODS.

We use the amino acid (AA) sequences comparison between the various b2-microglobulins and the HIV-1 protein sequences in the Los Alamos databank (Kuiken C., 2000). We analyzed 38 b2-microglobulin sequences [on Internet : NCBI/Protein] from various species: 1 human, 23 primates (1 chimpanzee, 1 gorilla, 1 orang-outang, 1 ateles, 7 callithrix, 5 callicebus, 4 saimiri, 3 aotus) and many mammals (1 pig, 1 dog, 1 horse, 1 cow), including 7 rodents (4 mice, 1 rat, 1 guinea pig, 1 rabbit) and 3 fishes. Human and the 3 major primates (chimpanzee, gorilla and orang-outang) b2-microglobulins contain the same sequence 3-RTPKIQV-9.

RESULTS.

We found in HIV-1 Polymerase Pol (Reverse Transcriptase RT p51) the sequence RTPKFRL (strain Eli), which shares with R7V a tetrapeptide RTPK (with the immunodominant proline P) and 3 semi-conserved AA (83,3 % homology for 6 AA). RTP is functionnal, as it is a phosphorylation motif. In rat and mouse b2m, R3 is replaced by K3, so we could also align it with Pol RT p51 of the majority (65/102= 65%) of HIV-1 strains 540-KTPKFRL-546 (or K545 in 37/102= 37%): 

Another molecular homology was found between b2m epitope E7E (44-ERIEKVE-50) [target of the anti-b2m monoclonal antibody BBM1 (Parham P., 1983) which neutralized HIV-1 (Devaux C., 1990)] and Gag p17 ERIEVKD [= region 90-QRIEIKD-96 of HGP 30 vaccine (Naylor P.H., 1987)]. Anti-Gag p17 antibodies were low in AIDS aggressive states, and, in children, correlated with neutralization in a cell fusion assay (Lange J.M., 1987).

DISCUSSION.

The role of a concomittent HCMV co-infection, which elevated per se beta2-microglobulin levels, in progressors has not been assessed.
In fact, if we reasonned, the anti-R7V antibodies are not found in 100% as claimed and expected (if beta2-microglobulin is stolen during HIV-1 budding, as presupposed, it would be present in 100% cases), but in only 70% or even less (because control are 22% positive, thus the 70-22%= 48% is nearer to the reality than 70%).
HIV-1 is not the only budding virus. There are too many enveloped budding viruses (Influenza, HCMV, all the retroviruses: lenti-, onco-, spumavirus, etc.). Obviously, this would means that any budding virus would benefit from a beta2-microglobulin vaccine ; This makes such a vaccine a tremendously universal vaccine against practically all viruses. Such a huge speculation is by far too prematurous and no data other than for HIV-1 and R7V has until now be claimed by nobody in the litterature.
During budding, there is no reason for a preferential protective property of anti-beta2-microglobulin antibodies for only 2 epitopes, among all the many possible. It is purely arbitrary to state that only these 2 are " revealed ".
What about the innumerable other proteins also catched by budding ? HLAs, Integrins (CD11a, CD18, CD29), GPI anchored (CD48, CD55, CD59), CD2, CD3, CD54, L-selectin, CD44, etcÉ ? Why would not they be also protective ?
It can be seen that the cellular vaccine concept is very complicated to explain and contains in itself numerous unexplored lacunes. No currently used vaccine in the world is a cellular one, all are classical: bacterial, viral or toxoid.
Our opinion is that the experimental results are correct, but not their interpretation, which seems more simplist than simple.

CONCLUSION.

Chermann's b2m vaccine heptapeptides seem to be molecular homologues of HIV-1 conserved epitopes in Pol RT p51 and Gag p17. We propose to confirm this with a sensitive Western-Blot (Zuffi E., 2001) and replace b2m R7V epitope by Pol RT p51, to obtain a better vaccine fittness and less auto-immune reaction. Conceptually, a conserved viral epitope is also much more acceptable than a cellular one.
Gag p17 epitope seems to be a known vaccine (HGP 30) and could replace E7E.

---

BIBLIOGRAPHY.

Beck S. & Barrell B.G., Human cytomegalovirus encodes a glycoprotein homologous to MHC CLASS6I antigens.Nature 1988, 331:269-72.
Bélec L., Ghys P.D., Becquart P. et al., Cervicovaginal secretory antibodies to HIV that block viral transcytosis through tight epithelial barriers, in sexually-exposed HIV-seronegative African women. Workshop on " HIV/AIDS vaccine development ". Pasteur Institute, Paris (France). May 5-6, 2000: p 47.
Cordonnier A., Montagnier L. & Emerman M.Single amino-acid changes in HIV envelope affect viral tropism and receptor binding.Nature 1989, 340:571-4.
Desmyter A., Transue T.R., Ghahroudi M.A. et al.Crystal structure of a camel single-domain VH antibody fragment in complex with lysozyme.Nature Struct. Biol. 1996, 3:803-11.
Devaux C., Boucraut J., Poirier G. et al., Anti-b2- microglobulin monoclonal antibodies mediate a delay in HIV1 cytopathic effect on MT4 cells.Res. Immunol. 1990, 141:357-72.
Fujinami R.S., Nelson J.A., Walker L. et al., Sequence homology and immunologic cross-reactivity of human cytomegalovirus with HLA-DR beta chain: A means for graft rejection and immunosuppression.
Galea P. et al.J. Virol. 1988, 62:100-5.
Galéa P., Le Contel C., Coutton C., Chermann J.-C., Rationale for a vaccine using cellular-derived epitope presented by HIV isolates.Vaccine 1999, 17:1700-5.
Galéa P., Le Contel C., Chermann J.C., A novel epitope R7V common to all HIV-1 isolates is recognized by neutralizing IgG found in HIV-infected patients and immunized rabbits.Vaccine 1999, 17:1454-61.
Harrer T., Bachmann O., Schmitt M. et al., Development of an immunotherapeutic vaccine for CTL induction in HIV-infected patients. Sixth Europ. Conf. Clinical Aspects & Treatment of HIV-infection. Hamburg, Germany, Oct. 11-15, 1997. P344.
Kuiken C., Foley B., Hahn B. et al., HIV sequence compendium 2000. Los Alamos Nat. Lab., Los Alamos, New Mexico, USA. 592 p.
Lange J.M., De Wolf F., Krone W.J. et al., Decline of antibody reactivity to outer viral core protein p17 is an earlier serological marker of disease progression in human immunodeficiency virus infection than anti-p24 decline.Aids 1987, 1:155-9.
Levy J.A., Potential problems involved in vaccine development.Workshop on " HIV/AIDS vaccine development ". Pasteur Institute, Paris (France). May 5-6, 2000:p 11.
Lopalco L., Barassi C., Pastori C. et al., Neutralizing IgA of exposed-uninfected and HIV-seropositive individuals recognize different epitopes of gp41. Workshop on " HIV/AIDS vaccine development ". Pasteur Institute, Paris (France). May 5-6, 2000:p 45.
Mazzoli S., Lopalco L., Salvi A.et al., Human immunodeficiency virus (HIV)-specific IgA and HIV neutralizing activity in the serum of exposed seronegative partners of HIV-seropositive persons.J. Inf. Dis. 1999, 180:871-5.
Naylor P.H., Naylor C.W., Badamchian M. et al., Human immunodeficiency virus contains an epitope immunoreactive with thymosin alpha1 and the 30-amino acid synthetic p17 group-specific antigen peptide HGP-30.Proc. Nat. Acad. Sci. USA 1987.
Oldstone M.B.A., Molecular mimicry and immune-mediated diseases. FASEB J. 1998, 12:1255-65.
Parham P., Androlewicz M.J., Holmes N.J., Rothenberg B.E. Arginine 45 is a major part of the antigenic determinant of human beta 2-microglobulin recognized by mouse monoclonal antibody BBM.1.J. Biol. Chem.1983, 258:6179-86.
Pastori C., Barassi C., Piconi S. et al., HIV neutralizing IgA in exposed seronegative subjects recognise an epitope within the gp41 coiled-coil pocket.J. Biol. Regul. Homeost. Agents 2000, 14:15-21.
Piguet V. and Trono D., A structure-function analysis of the Nef protein of primate lentiviruses. In : Kuiken C. et al. Human Retroviruses and AIDS. Los Alamos Nat. Lab., Los Alamos, New Mexico USA, 1999, 790 pages.pp 448-59.
Robinson W.E. Jr, Gorny M.K., Xu J.Y. et al., Two immunodominant domains of gp41 bind antibodies which enhance human immunodeficiency virus type 1 infection in vitro.J. Virol. 1991, 65:4169-76.
Rowland-Jones S.L., Dong T., Fowke K.R. et al., Cytotoxic T cell responses to multiple conserved HIV epitopes in HIV-resistant prostitutes in Nairobi.J. Clin. Invest. 1998, 102:1758-65.
Schreiber M., Muller H., Wachsmuth C. et al., Escape of HIV-1 is associated with lack of V3 domain-specific antibodies in vivo. Clin. Exp. Immunol. 1997, 107:15-20.
Schwimmbeck P., Yu D.T.and Oldstone M.B.A., Auto-antibodies to HLA-B27 in sera of patients with ankylosing spondylitis and Reiter's syndrome.J. Exp. Med.1987, 166:173-181.
Smithe J.A. (Smythe J.A.) and Reitz M.S. Points to ponder on the function of Nef.Res. Virol. 1992, 143:47-9.
Tran M.K.G., Kirkiacharian S., Maurisson G., Caprani A., Aids vaccine for HLA-A2 & B14 Caucasians with Gag inducing cytotoxic lymphocytes (CTL). Homology with Hand, Foot & Mouth Disease virus (HFMDV): Role of Gag in aphthous ulcers. XIV Int. AIDS Conf., July 7-12 2002, Barcelone, Spain. (Late Breaker submission).
Tran M.K.G., Kirkiacharian S., Maurisson G. & Caprani A., Thailand and Africa vaccines with HIV-1 gp41 LQAR (Leu Gln Ala Arg): Thailand strains are the most homologous to interferon-beta2 (IFN-beta2). 12th Int. Symp. HIV & Emerging Infectious Dis., June 2002, Toulon, France. Track B: P.P.B.6.
Tran M.K.G., Maurisson G., Richert A. & Caprani A., HIV-1 vaccine: The auto-immune protective anti-CCR5 antibody targets the motif AQWD shared with HIV-1 envelope gp120 post-V3 loop. 8th Europ. Conf. Clin. Aspects & Treat. HIV-infection. Athens, Greece Oct 28-31 2001. Late Breaker. 
Tran M.K.G., Molecular mimicry between cytomegalovirus IE2 and HLA-DP3. Médecine/Sciences 1996, n° spécial, 12: (S239).
Wagner R., Leschonsky B., Harrer H. et al., Molecular and functional analysis of a conserved CTL epitope in HIV-1 p24 recognized from a long-term nonprogressor: Constraints on immune escape associated with targeting a sequence essential for viral replication.J. Immunol. 1999, 162:3727-34.
Williams R.C., Malone C.C. Antigenic epitopes on beta 2-microglobulin reacting with monoclonal antibodies.J. Lab. Clin. Med. 1993, 121:805-20.
Young J.A.T., HIV and HLA similarity. Nature 1988, 333:215.
Zanelli E., Krco C.J., David C.S., The motif " DERAA " of HLA-DRB1*0402 is essential for DQ8-restricted Tcell response in transgenic mice: Implication for rheumatoid arthritis predisposition. In: Charron D. HLA. Genetic diversity of HLA. Functional and medical implication. 1997, 778p. EDK, Med. & Scient. Int. Pub., Paris. (Ed.). pp 631-3.
Zuffi E., Manaresi E., Gallinella G. et al., Identification of an immunodominant peptide in the parvovirus B19 VP1 unique region able to elicit a long-lasting immune response in humans.Viral Immunol 2001, 14:151-8.