WO1996020006A1

WO1996020006A1 - Vaccine against aids comprising a peptide sequence of hiv

Info

Publication number: WO1996020006A1
Application number: PCT/GB1995/003009
Authority: WO
Inventors: Sarah Rowland-Jones; Frances Gotch; Andrew James Mcmichael
Original assignee: Medical Research Council
Priority date: 1994-12-23
Filing date: 1995-12-21
Publication date: 1996-07-04
Also published as: GB9426153D0; AU4270896A

Abstract

Disclosed is a vaccine against AIDS, comprising a peptide sequence of HIV capable of generating a protective cytotoxic T lymphocyte (CTL) response in humans.

Description

C1052.02/M

Title: Vaccine against AIDS compri sing a peptide sequence of HIV

Field of the Invention

This invention relates to certain peptides and to vaccines comprising said peptides.

Background of the Invention

A critical requirement in the rational design of a prophylactic vaccine against HIV is to establish whether or not protective immunity can occur following natural infection. The immune response to HIV infection is characterised by very vigorous HIV-specific cytotoxic T lymphocyte (CTL) activity. The inventors identified several HIV-1 and HIV-2 cross-reactive peptide epitopes, presented to CTL from HlV-infected Gambians by HLA- B35 (the most common Gambian class I HLA molecule). These peptides were used to elicit HIV-specific CTL from three out of six repeatedly exposed but HIV-seronegative female prostitutes with an HLA-B35 haplotype. These women remain seronegative with no evidence of HIV infection by PCR or viral culture. Their CTL activity appears therefore to represent protective immunity against HIV infection.

The production of an effective prophylactic vaccine against human immunodeficiency virus (HIV) infection would be greatly assisted by an understanding of the requirements for protective immunity. One step towards this is to determine if such immunity can develop naturally following exposure to the virus. HIV can usually be isolated from seropositive subjects at any stage after infection¹ and cohort studies suggest that most people with HIV- antibodies will ultimately develop disease². Thus there is little evidence that HIV seropositive people can eliminate their virus. However, some individuals remain seronegative despite definite exposure to HIV and several studies suggest that cellular immune responses to HIV can occur in the absence of persistent antibody in such cases. These include T-helper (Th) cell proliferation and Interleukin-2 (IL-2) secretion in response to envelope peptides^{3 6}, and inhibition of HIV replication by their CD8+ T cells⁷. It is possible, however, that these responses might reflect exposure merely to defective virus particles or protein antigens rather than replication-competent virus. The induction of major histocompatibility complex (MHC) class I-restricted CTL may be a more reliable indication of exposure to replicating virus^{8 9}, and the finding of HIV-specific CTL in a small number of apparently uninfected children born to HIV-infected mothers is believed to indicate exposure to live virus^10"12. The most important subjects for study are those repeatedly exposed to HIV who remain uninfected and in whom persistent seronegativity might truly represent resistance to HIV infection. Preliminary studies of uninfected sexual partners of HIV-infected people suggest that they may have an increase in ne/-specifιc CTL precursors¹³. We have extended these studies by looking for the presence of HIV- specific CTL in a group of repeatedly exposed but persistently seronegative female prostitutes in The Gambia, West Africa.

HIV-specific CTL activity can be readily detected in asymptomatic HIV-infected donors, often directly from unstimulated peripheral blood mononuclear cells (PBMCs)^{14 15} suggesting that there is a high frequency of circulating CTL in response to continuing viral replication^16,17. In contrast, CTL activity in the absence of persistent infection might be transient and temporally related to exposure, as seen in several of the babies born to infected mothers. We therefore selected for study a group of women who are likely to have been (and continue to be) repeatedly exposed to both HIV-1 and HIV-2. Approximately 35 % of the female sex-workers in Gambian towns are HIV-infected and this proportion has been increasing rapidly^{18 19}. Whereas initially HIV-2 was predominant, most recent infections are with HIV-1. The seronegative women in this study have worked as prostitutes for more than five years, use condoms infrequently with clients and only rarely with their regular partners²⁰ and have a high incidence of other sexually- transmitted diseases, making it less likely that they have escaped exposure to HIV simply by chance. Although the prevalence of HIV infection in their clients is not known, levels of infection in the general population are consistent with their exposure to at least one HIV-infected man every month.

HIV-specific CTL from infected donors are usually detected using their endogenous virus to restimulate the CTL in vitro²¹, but this method would not be appropriate to elicit CTL from uninfected donors. One approach that the present inventors tried was to use exogenous HIV to stimulate CTL in culture, but this may damage the CD4+ T cells needed to initiate the response and was not successful in these studies. Virus-specific CTL recognise viral antigens in the form of naturally-processed peptides eight to ten amino acids in length, which are bound in the cleft of class I MHC molecules on the surface of the infected cell and presented to the T-cell receptors of the CTL²². In order to detect CTL in the women in this study we first defined peptide epitopes from HIV-1 and -2, recognised by CTL from infected donors in association with the Human Leucocyte Antigen (HLA) Class I molecule occurring most frequently in The Gambia, HLA-B 3501 (present in 32% of the major ethnic group²³). These peptides showed evidence of cross-reactivity between HIV-1 and HIV-2 i.e. CTL from donors infected with one strain of HIV frequently recognised the corresponding peptide from the other virus strain. The peptides were then used directly to stimulate CTL from the PBMCs of seronegative donors, a protocol which has previously been shown to be an efficient method of generating secondary specific (or memory) CTL responses in vitro²*^'26. Using this strategy, specific CTL activity against one or more HIV peptides was detected in three out of six highly exposed but apparently uninfected women with HLA-B35, but not in a panel of 19 volunteers with no history of HIV exposure.

Summary of the Invention

In a first aspect the invention provides a vaccine against AIDS, comprising a peptide sequence of HIV capable of generating a protective cytotoxic T lymphocyte (CTL) response in humans.

The present inventors have identified a number of peptide fragments of HIV which are the target of a HLA-B35 restricted CTL response in humans (i.e. are CTL epitopes). In particular, peptide epitopes have been identified which are the targets of a CTL response in a group of women who, despite almost certain, repeated exposure to HIV, remain seronegative, such that the CTL reponse to these epitopes appears to have a genuinely protective effect. Thus, a vaccine designed to stimulate an immune response to these peptides should equally have a genuinely protective effect. This is a truly significant finding: numerous CTL epitopes have been identified in HIV -encoded proteins, but so far none has been convincingly demonstrated to stimulate a protective CTL response.

The inventors found that probably the most significant CTL epitope was that corresponding to a fragment of HIV-1 pol protein, having the sequence: HPDIVIYQY. However, the proline residue at position 2, and the C-terminal tyrosine residue, are thought to be important in allowing the peptide to bind to the HLA-B35 Class I antigen. Thus alteration of these residues should allow the peptide to be presented by Class I antigens other than HLA-B35, so that a general consensus sequence HXDIVIYQZ (Seq ID No. 1) may be defined, which might be expected to act as an epitope for CTLs restricted by human leukocyte antigens other than B35. The letters X and Z represent any amino acid, and both may represent the same residue in a particular peptide.

In addition to the epitope defined above, the inventors found that the exposed, but seronegative, individuals also had transient CTLs which recognised epitopes corresponding to peptide fragments of HIV /ιe/ (same sequence for HIV-1 and -2 isolates), HIV-1 pl7 and p24 polypeptides, and HIV-2 pl7 and p24 polypeptides. The epitopes had the following sequences: (tie ) VPLRMPTY; (HIV-1 pl7) NSSKVSQNY; (HIV-1 p24) PPIPVGDIY; (HIV-2 pl7) PPSGKGGNY; and (HIV-2 p24) NPVPVGNIY respectively. As above, the epitopes generally have a proline residue at position 2, and all have a C-terminal tyrosine residue, which are believed to facilitate binding to the B35 class I HLA antigen, rather than binding to the T-cell receptor. Thus, consensus sequences for the epitopes may be defined as follows: VXLRMPTZ (Seq ID No. 2); NXSKVSQNZ (Seq ID No. 3); PXIPVGDIZ (Seq ID No. 4); PXSGKGGNZ (Seq ID No. 5); and NXVPVGNIZ (Seq ID No. 6); where X and Z may be any amino acid residue (and may both be the same in any peptide). Whilst the strongest and most persistent CTL response was against the HIV-1 pol epitope, responses to one or more of the other epitopes may also have a protective effect.

The vaccine will preferably comprise a number of peptides, each of which is recognised by CTL restricted by a different respective human leukocyte antigen. In this way, the vaccine will be capable of generating a protective CTL response in a large number of people, not just those having an HLA-B35 haplotype. The vaccine may conveniently comprise different variants of one or more (preferably all) of the consensus sequences defined above, and/or may comprise different CTL epitopes. A large number of CTL epitopes have so far been determined in HIV proteins, and many of these are disclosed by McMichael & Walker (1994 AIDS 8 (Supplement 1), S155-S173), including peptides having the consensus sequences defined above. These may be usefully included in the vaccine, but it should be remembered that, unlike the peptide epitopes having the consensus sequences defined above, those disclosed in the prior art have not been demonstrated thus far to be capable of generating a protective CTL response.

The vaccine can be given prophylactically to healthy, seronegative individuals to protect them against HIV infection. In addition, the vaccine may be useful when given therapeutically to seropositive patients, with or without symptoms, in order to help reduce the severity of disease.

The vaccine will preferably also comprise one or more further HIV-encoded components to stimulate CTL responses restricted by HLA antigens other than B35, and/or components to stimulate other arms of the immune response (e.g. T helper cells, serum/secretory antibody).

The vaccine is preferably capable of conferring protection after just one inoculation, but it may be necessary to administer several inoculations to an individual in order to elicit a suitably protective immune response.

Numerous vaccine compositions are known to those skilled in the art. For example, peptide vaccines could be co-administered parenterally with adjuvants (e.g. aluminium salts), or in recombinant bacterial (e.g. Salmonella sppJ or viral (e.g. retro virus) vectors. Such vectors could be killed or living organisms, expressing the peptide component of the vaccine for example as a fusion protein. Synthesis of fusion proteins is well-known to those skilled in the art, and involves insertion of the nucleic acid sequence encoding the peptide(s), in frame with the nucleic acid sequence encoding another (possibly unrelated) polypeptide. Particularly suitable fusion proteins may be made by fusion of the peptide(s) with targeting or immunomodulatory moieties. Alternatively, the peptide(s) of the invention could be incorporated into inert complexes such as ISCOMs or liposomes. The vaccine will typically also comprise a sterile physiologically acceptable carrier substance, such as phosphate buffered saline (PBS).

In a second aspect, the invention provides a peptide fragment having the general consensus sequence NXDVILIQX (Seq ID No. 7), capable of stimulating a CTL response against HIV in humans. The present inventors found that this peptide constitutes a CTL epitope, not previuosly recognised in the prior art. This peptide may conveniently be included in the vaccine of the first aspect of the invention. In particular, the peptide is a B35- restricted epitope in which X is proline and Z is tyrosine (such that the peptide corresponds to a fragment of HIV-2 pol protein), but other residues at the X and Z positions (which may both be the same residue) should give rise to a CTL response being restricted by other human leukocyte antigens.

Other alterations to the peptide sequences defined in relation to the first or second aspects of the invention could be made. For example, cysteine residues could be incorporated to allow the peptides to be covalently bound to other peptides or polypeptides via disulphide bridges. Equally, one or two conservative substitutions could be introduced without destroying the immunogenic properties of the peptides.

The present inventors have made some surprising findings concerning the above epitopes. CTL generated against a peptide derived from HIV-1 cross-react with the corresponding peptide derived from HIV-2, and vice versa. This is not so surprising in the case of the nef peptide fragment, which has precisely the same sequence in both HIV-1 and HIV-2 isolates (see Table 1), but this could not have been predicted for the other peptides which contain several amino acid differences. Whilst the present inventors would not wish to be bound by any theory, it is possible that this cross-reactivity is important for the protective effect of the epitopes.

In a further aspect, the invention provides a method of prophylactic or therapeutic treatment against AIDS, comprising administering to an individual a composition comprising an effective amount of one or more of the peptides defined above.

An "effective amount" is the amount of a peptide required to be given to an average human adult in order to induce a substantial CTL response to the peptide, either upon initial immunisation (as is preferred) or upon subsequent exposure to HIV. Typically, an effective amount will vary between 1-500 mgs of each peptide present in the vaccine.

In a further aspect the invention provides for use of a peptide in making a vaccine against AIDS capable of generating a protective CTL response against HIV in humans, wherein the peptide has a sequence in accordance with any one of the following consensus sequences: HXDIVIYQZ; VXLRMPTZ; NXSKVSQNZ; PXIPVGDIZ; PXSGKGGNZ; NXVPVGNIZ; and NXDVILIQX; where X and Z may be any amino acid residue, and where X may be the same as Z.

Thus the invention also provides a method of making a vaccine against AIDS capable of generating a protective CTL response to HIV, comprising mixing with a physiologically acceptable carrier substance an effective amount of a peptide having a sequence in accordance with any one of the following consensus sequences: HXDIVIYQZ; VXLRMPTZ; NXSKVSQNZ; PXIPVGDIZ; PXSGKGGNZ; NXVPVGNIZ; and NXDVILIQX; where X and Z may be any amino acid residue, and where X may be the same as Z.

The various aspects of the invention will now be further described by way of illustrative example and with reference to the drawings, of which:

Figures la and lb are bar charts showing lysis of target cells by HIV-specific CTL;

Figure lc is a graph of percent lysis against peptide concentration;

Figure 2 shows four bar charts (one per donor) demonstrating lysis of peptide-pulsed target cells; Figure 3a shows a bar chart demonstrating lysis of recombinant vaccinia virus- infected cells by CTL from two donors;

Figure 3b is a graph of percent lysis against peptide concentration; and

Figure 3c is a graph of percent lysis against length of time in culture.

Identification of peptides presented by HLA-B35 to CTL from infected donors.

The amino acid sequence motif of peptides binding to HLA-B35 has previously been determined and used to identify HLA-B35-restricted epitopes from P. falciparum²⁵. In a similar way the consensus sequences of HIV-1 and 2 gag, pol and nef proteins were scanned for possible B35-binding epitopes, i.e. octamers or nonamers with proline or serine at position two of the peptide and tyrosine at the C-terminus. Candidate peptides were synthesised and screened for binding to HLA-B35, using the previously-described T2-B35 assembly assay²⁵. Those peptides which bound (15 out of 19) were tested for recognition by bulk CTL cultures from HIV (1 or 2)-infected Gambians, identified as having HLA-B-3501 by serological typing and isoelectric focusing²⁷.

The majority of HIV-infected donors with an HLA-B35 haplotype had B35-restricted CTL recognising one or more peptide from gag (pi 7 and p24), pol and nef (Table 1). The nef peptide, which is conserved between HIV-1 and -2, has previously been identified as an eptitope for CTL from HIV- 1 -infected donors²⁸ and was also recognised by CTL from HIV-2-infected donors (Fig. la). For the other peptides, CTL from most HIV-2-infected donors recognised both the HIV-2 peptide and the corresponding peptide from HIV-1 (Fig. lb), suggesting that donors with a strong B35 -restricted CTL response against HIV-2 may show cross-immunity to HIV-1. The HIV-1 pol and p24 peptides were subsequently confirmed as epitopes for B35-restricted CTL from HIV- 1 -infected donors (both European and Gambian), and also showed cross-reactivity between HIV-1 and -2 peptides (Fig. lc). This is in contrast to the only previously described peptide epitope from HIV-2, where HLA-B53-restricted CTL from HIV-2-infected donors fail to recognise the corresponding HIV-1 gag sequence²⁹.

CTL studies in exposed seronegative donors

The study population consisted of 20 women - 14 had been prostitutes for more than five years and reported little condom usage with clients or regular partners²⁰ and six were long- term sexual partners of HIV-infected men. Cells from those donors clearly typed by serology as having HLA-B35 or B53 were stimulated in vitro with each of the candidate B35-restricted peptides from HIV-1 and -2 or the previously identified B53-restricted peptide from HIV-2 (ref. 29). Peptide stimulated cultures were also set up from eight known seropositive donors with HLA-B35 or B53, and from a control group of volunteers at low-risk of HIV infection¹⁹ with HLA-B35 (12 Gambian and seven European) and two Gambians with HLA-B53. For those donors without either HLA-B35 or B53, CTL cultures were established using autologous phytohaemaglutinin (PHA) blasts infected with HIV-1 (IIIB) and a Gambian strain of HIV-2 (ref. 30). Similar cultures were set up from four previously identified seropositive donors.

The results of these studies are shown in Table 2. Using the virus-stimulation protocol, CTL could only be generated in cultures from known HIV-infected donors and not in the exposed seronegatives. However, specific CTL activity against one or more peptides was repeatedly detected after 10-14 days in the peptide-stimulated cultures from three of the six high-risk seronegative women with HLA-B35 (OXll , 12 and 14, in Fig. 2), but not in their three counterparts with HLA-B53 nor in any of the low-risk volunteers (Table 2). These CTL grew in culture and continued to show specific activity without further restimulation for over four weeks. Of the responding seronegatives, donors OXll and OX12 generated CTL to most of the HIV peptides tested and OX14 responded only to the HIV-1 pol peptide. The strongest responses were generated towards the HIV-1 pol peptide, which lies close to d e active site of reverse transcriptase³¹, and to the nef peptide, which is conserved between HIV-1 and -2 in a region thought to be important as a protein kinase phosphorylation site³². Three months later, further samples were taken from the six exposed seronegative women with HLA-B35, and persistent CTL activity was detected against the HIV-1 pol peptide in cultures from OXll and OX12, but not against any of the other HIV peptides. These CTL killed target cells infected with recombinant vaccinia virus expressing the HIV-1 pol protein and recognised the peptide at 1 x 10^~6M (Fig. 3a,b).

CTL cultures were generated from both HIV-exposed and unexposed donors, using the same protocol, to an HLA-B35-restricted influenza matrix peptide (Influenza A matrix protein 128-135) and an HLA-B35-restricted EBV peptide (EBNA 3a 458-466, Lee, S et al., in press). The kinetics of these secondary in vitro responses against EBV and influenza were virtually identical to those elicited by the HIV peptides in the exposed seronegative women (Fig. 3c).

Occult HIV infection excluded

To exclude occult HIV infection in the women with detectable CTL, samples from each donor were tested for HIV-1 and -2 DNA by means of the polymerase chain reaction (PCR). Nested PCR was performed on two samples taken three months apart and was consistently negative. Attempts to isolate virus, by co-culture of cells stored from each bleed with uninfected PHA blasts, monitoring for reverse transcriptase (RT) activity with a highly sensitive assay capable of detecting single virions³³, showed no evidence of HIV infection. HIV-specific CTL in seronegative subjects could potentially be a response to acute HIV infection, before the development of antibodies³⁴, but the women were still seronegative and virus-culture negative three months after the CTL were first detected, making recent infection extremely unlikely.

Discussion

The most probable explanation for the finding of HIV-specific CTL, able to kill virus- infected cells, in apparently uninfected but repeatedly HIV-exposed women is that they have been immunised by exposure to HIV. This could result from exposure to live or defective virus particles, but live virus would be the more efficient in inducing CTL. Other possible explanations are less likely. Occult HIV infection or delayed seroconversion is improbable in the absence of detectable viraemia using the most sensitive methods available. Cross-reactivity from CTL primed by epitopes of similar sequence in another pathogen would be unlikely to lead to responses against more than one HIV peptide, and the recognition by the CTL of target cells pulsed with low concentrations of peptide or infected with virus is much more in keeping with a high affinity specific response than a low affinity cross-reaction. A related possibility is that the CTL responses represent cross-reactive alloreactive CTL³⁵ induced by pregnancies, but the donor with the strongest responses (OXll) is nulliparous whereas the Gambian control population included several multiparous women. Finally, these are unlikely to be primary in vitro CTL responses given that no responses to HIV peptides could be elicited in any of the control donors in the absence of HIV exposure. The generation in vitro of primary CTL able to kill virus-infected cells requires special conditions that were not present here, such as the use of specialised antigen-presenting cells³⁶"³⁸ or peptide modified to allow access to the class I MHC processing pathway³⁹. The similarity of the kinetics of the HIV- specific responses and those raised against influenza and EBV peptides strongly suggests that these are secondary, or memory, CTL responses primed by exposure to live virus.

The lack of detectable CTL activity in the other exposed seronegative donors may simply mean that, against a background of relatively low HIV prevalence in the general population, those women have managed to avoid or escape exposure to infectious HIV. Alternatively, it may reflect the limitations of our restimulation protocol, which would be expected to detect only HLA-B35 or B53-restricted CTL primed by virus very similar in sequence to that of the peptides used, and would not therefore detect CTL with other specificities. Also, the follow-up studies suggest that some HIV-specific CTL responses may be transient, perhaps in relation to the timing of exposure.

If the CTL have been primed by exposure to live HIV, it is surprising that these women remain persistently seronegative. These findings are consistent wi the hypothesis that exposure to low doses of virus primes predominantly cellular (or Th-1) immune responses at the expense of humoral (or Th-2) responses^4,40. In macaques, low dose simian immunodeficiency virus (SIV) primes Th-1 responses without antibody production (CTL were not measured) and these animals are then protected against a high dose challenge⁴¹. In murine leish aniasis, protection results from cell-mediated immunity without antibody production induced by low-dose immunisation⁴². The detection of responses which are cross-reactive between HIV- 1 and -2 may be relevant, since Gambian prostitutes are likely to have been first exposed to HIV-2 which appears to be less pathogenic⁴³, and has a lower transmissibility and virus load than HIV-1 infection⁴⁴-⁴⁵. Initial exposure to HIV-2 may have led to protective immunity against HIV-1 in women who make a CTL response that is cross-reactive between the two viruses. Cross-reactive neutralising antibody responses between HIV-1 and -2 are rare⁴⁶; cross-reactive T-helper⁴⁷ and CTL⁴⁸ responses can occur but are not always seen^29,49"51 so cross-protection may be dependent on the extent of conservation between HIV-1 and -2 in the epitopes selected by particular class 1 molecules. The strongest responses observed in our seronegative donors were to epitopes which are probably conserved for functional reasons^31,32. If an effective cross-reactive CTL response could clear infection with a second strain of HIV, this may explain why not all donors with dual sero-reactivity for HIV-1 and -2 are actually infected with both viruses⁵²"⁵⁴.

The question of whether these women are protected against HIV infection is crucial, and may depend on the persistence of their CTL. Although the CTL response to HIV can be extremely long-lasting at the clonal level in infected patients¹⁷ (P. Moss et al. , manuscript submitted), it is not known how long specific CTL persist in the absence of antigen. Some studies have demonstrated long-lasting CTL immunity without persisting antigen, although this remains controversial^55"57. The observation of transient HIV-specific CTL in some uninfected infants of infected mothers^11,12 and the diminution in detectable CTL after three months in these women may indicate that CTL levels wane over time. However, studies in a comparable group of women in Nairobi show that a minority (5 %) remain uninfected despite repeated exposure over long periods of time and appear to be genuinely resistant to infection (Plummer, F A et al, 1993, IXth International Conference on AIDS, and manuscript submitted).

These findings of HIV-specific CTL in a group of repeatedly exposed but uninfected women suggest that CTL generation may be a central component of protective immunity against HIV and underline the importance of CTL induction in HIV vaccine design. Methodology

HIV testing. Ethical approval was obtained from the The Gambia Government/MRC Ethical committee and informed consent prior to HIV testing was given by each donor in the study. Donors were tested for HIV-1 and -2 infection with two type-specific competitive enzyme-linked immunosorbent assays (Murex Diagnostics, Dartford, Kent, UK) and with type-specific peptide-strips (Pepti-Lav, Diagnostics Pasteur, Marnes-la- Cougette, France).

Generation of HIV-specific CTL from infected donors

PBMCs were separated from whole blood, and CTL cultures were established by taking one eighth of cells, stimulating them for 24 hours with PHA, washing once and adding them back to the remaining cells²¹. Cells were cultured in RPMI 1640 (Gibco) with 10% fetal calf serum (Gibco) (R/10) and antibiotics for one week, then 10% Lymphocult T (Biotest) was added for the second week. Standard ⁵¹-chromium release assays were performed after two weeks using HLA-B35-matched or control mismatched target B- lymphoblastoid cell-lines (B-LCLs) labelled with ⁵¹-chromium and pulsed with each of the HIV peptides or a control influenza peptide at a concentration of lOmM. Background chromium release was less than 20%. Per cent lysis was calculated from the formula 100 x (E-M/T-M), where E is the experimental release, M is the release in the presence of R/10 medium and T is release in the presence of 5% Triton X-100 detergent. Results were regarded as positive if recognition of the HIV peptide was greater than 10% above that of a control peptide on at least two occasions.

CTL studies in exposed seronegative donors. After testing serum for HIV antibodies, PBMCs were separated from 10-15mls of heparinised blood and 2 x 10⁶ cells were stored at -80°C, for future virological studies. Class I HLA typing was performed by serology, and peptide-stimulated cultures were set up from donors with HLA-B35 or B53. Their cells were pulsed as a pellet for one hour with 100 μM of each B35-restricted peptides from HIV-1 and -2, or the previously identified B53 -restricted peptide from HIV-2 (ref. 29), at a concentration of 100 μM, then diluted in R/10 to a final concentration of 10 μM and cultured at 2 x 10⁶ cells per well in a 24-well Costar plate. IL-2, in the form of Lymphocult-T, was added to a dilution of 10% on day three. Assays for specific CTL activity were carried out after 7-14 days in culture, using HLA-B35 or B53-matched target cells pulsed with the appropriate peptides as above. Some CTL cultures were also tested against matched target cells infected with recombinant vaccinia virus expressing full-length HIV-1 pol or an irrelevant protein, β-galactosidase (kind gifts of a B Moss).

Cells from the donors without HLA-B35 were stimulated with autologous PHA blasts infected with HIV-1 (IIIB) and a Gambian strain of HIV-2 (CBL-20) (ref. 30). These cultures were tested for HIV-specific CTL activity after 7-14 days using autologous or HLA-matched target cell-lines infected with recombinant vaccinia viruses expressing HIV gag, pol and nef genes.

Control studies. Peptide-stimulated cultures using HIV peptides were set up from known seropositive donors with HLA-B35 or B53 and a control group of volunteers at low-risk of HIV infection. (Low-risk volunteers were European laboratory workers, seronegative donors from a rural area of The Gambia with low HIV prevalence and seronegative donors from The Gambian Blood Bank). CTL cultures were generated from both HIV-exposed and unexposed donors, using the same protocol, to an HLA-B35-restricted influenza matrix peptide (Influenza A matrix protein 128-135) and an HLA-B35 -restricted EBV peptide (EBNA 3a 458-466).

Virological studies in the seronegative donors with HIV-specific CTL. Nested PCR was carried out on DNA extracted from PHA blasts on two occasions, three months apart, for donors OXll and OX12, and on one occasion for donor OX14, using gag primers for HIV-1 (ref. 58) and LTR primers for HIV-2 (ref. 59). Virus isolation was performed as described previously³⁰, on two samples from each donor taken three months apart. Briefly, PBMCs from the women were stimulated with PHA for three days, then cultured in the presence of IL-2 with fresh donor PBMCs added at weekly intervals over a 35-day period. Cultures were monitored for the appearance of syncitia or other cytopathic effects, and were co-cultivated with indicator cell-lines at regular intervals. Supernatants were collected from the cultures every three to four days and assayed for RT activity using a highly sensitive PCR assay (at least 100 times more sensitive than conventional assays)³³.

Since obtaining the results described above and in Table 2, further follow-up studies have been conducted. As a result, HIV-specific CTL have now been detected in blood samples from 5 out of the 6 repeatedly exposed prostitutes. Over a period of 2 years, the women have remained healthy, seronegative and HIV-negative (by both virus culture and PCR). The CTL generated from the women are shown to be cross-reactive between HIV-1 and HIV-2; that is, they kill target cells pulsed with either the HIV-1 or the HIV-2 version of a given epitope.

Table 1 : Identification of peptides presented by HLA-B35 to CTL from HIV-infected donors

Protein Amino acid HIV strain Amino acid consensus residues sequence

Nef 75-82 HIV-1 & -2 VPLRMPTY

Pol 329-337 HIV-1 HPDIVIYQY

HIV-2 NPDVILIQY p24 260-269 HIV-1 PPIPVGDIY

HIV-2 NPVPVGNIY pl7 130-138 HIV-1 NSSKVSQNY

HIV-2 PPSGKGGNY

Table 2 : Detection of HIV-specific CTL

Method of Risk Group CTL detected Stimulation

Virus stimulation High-risk seronegative 0/11 Known seropositive 3/4

Peptide stimulation High-risk seronegative 0/3 HLA-B53 donors Low-risk seronegative 0/2 Known seropositive (HIV-2) 2/2

Peptide stimulation High-risk seronegative 3/6 HLA-B35 donors Low-risk seronegative (Gambian) 0/12 Low-risk seronegative (European) 0/7 Known seropositive (HIV-1) 2/3 References

1. Ho, D.D. e l. , N. Engl. J. Med. 321, 1621-1625 (1989).

2. Schecter, M.T. et al.. Lancet 341, 658-659 (1993).

3. Clerici, M. , et al, J. infect Dis. 165, 1012-9 (1992).

4. Clerici, M. and G. Shearer, Immunol Today 14, 107-111 (1993).

5. Clerici, M. et al, AIDS 7, 1427-33 (1993).

6. Clerici, M. et al, J.A.M.A. 271, 42-6 (1994).

7. Levy, J.A. , aids 7, 1401-1410.

8. Miedema, F. et al, Science 262, 1074-1075 (1993).

9. Salk, J. et al., Science 262, 1075-1076 (1993).

10. Cheynter, R. et al, Bar. J. Immunol. 22, 2211-2217 (1992).

11. Rowland-Jones, S.L. et al, Lancet 341, 850-861 (1993).

12. Aldhous, M.C. et al, Clin. Exp. Immunol. 97, 61-7 (1994).

13. Langlade-Demoyen, P. et al, J. Clin. Invest. 93, 1293-1297 (1994).

14. Walker, B.D. et al, Nature 328, 345-8 (1987).

15. Plata, F. et al, Nature 328, 348-351 (1987).

16. Carmichael, A. et al, J. exp. Med. 177, 249-56 (1993).

17. Kalams, S.A. et al, J. exp. Med. 179, 1261-1271 (1994).

18. Mabey, D.C.W. et al, Br. Med. J., 83-86 (1988).

19. Wilkins, A. et al, AIDS 5, 1529-30 (1991).

20. Pickering, H.M. et al, AIDS 7, 271-3 (1993).

21. Nixon, D.Ε. et al, Nature 336, 484-487 (1988).

22. Elliott, T. et al, Chem. Opin. Immunol. 3, 854-866 (1993).

23. Allsopp, C.E. et al . Am. J. hum. Genet. 50, 411-21 (1992).

24. Martinon, F. et al , Eur. J. Immunol 21, 2171-2176 (1990).

25. Hill, A.V.S. et al, Nature 360, 434-439 (1992).

26. Kos, F.J. and A. Mullbacher, Eur. J. Immunol. 22, 3183-5 (1992).

27. Neefjes, J.J. et al. Ham. Immunol 16, 169-181 (1986).

28. Culmann, B. et al, J. Immunol. 146, 1560-1565 (1991).

29. Gotch, E. et al, J. Immunol. 151, 3361-9 (1993).

30. Schulz, T.F. et al, J. Virol 64, 5177-5182 (1990). 31. Kohlstaedt, L.A. et al, Science 356, 1783-1790 (1992).

32. Shugas, D.C. et al, J. Virol. 67, 4639-4650 (1993).

33. Silver, J.T. et al, N.A.R. 21, 3593-3594 (1993).

34. Safrit, J.T. et al, J. exp. Med. 179, 463-72 (1994).

35. Burrows, S.R. et al, J. exp. Med. 179, 1155-1161 (1994).

36. Carbone, F.R. et al , J. exp. Med. 167, 1767-1779 (1989).

37. Macatonia, S.E. et al, J. exp. Med. 1255-1264 (1989).

38. De Bruijn, M. et al, Eur. J. Immunol. 22, 3013-20 (1992).

39. Schild, H. et al, J. exp. Med. 174, 1665-8 (1991).

40. Salk, J. et al, Science 260, 1270-1272 (1993).

41. Clerici, M. et al , AIDS 8, 1391-1395 (1994).

42. Bretscher, P. A. et al, Science 257, 539-542 (1992).

43. Markovitz, D.M. , Ann. intern. Med. 118, 211-8 (1993).

44. Simon, P. et al, AIDS 7, 1411-1417 (1993).

45. De Cock, K.M. et al, J.A.M.A. 270, 2083-6 (1993).

46. Robert-Guroff, M. et al, J. Virol. 66, 3602-8 (1992).

47. Pinto, L.A. et al , AIDS 7, 1389-1391 (1993).

48. Nixon, D.F. et al, AIDS 4, 841-5 (1990).

49. Phillips, R.E. et al, Nature 354, 453-9 (1991).

50. Johnson, R.P. et al, J. exp. Med. 175, 961-71 (1992).

51. Johnson, R.P. et al, J. Virol. 67, 438-45 (1993).

52. George, J.R. et al, Lancet 340, 337-339 (1992).

53. Peeters, M. et al, Lancet 340, 339-340 (1992).

54. Leonard, G. et al, AIDS 7, 1185-1189 (1993).

55. Mullbacher, A. , J. exp. Med. 179, 317-321 (1994).

56. Loos, L., et al, Nature 369, 648-52 (1994).

57. Gray, D. and P. Matzinger, J. exp. med. 174, 969-974 (1991).

58. Simmonds, P. et al, J. Virol. 64, 864-72 (1990).

59. Berry, N.J. , et al, AIDS Research and Human Retroviruses 8, 1031-1037 (1994). Figures

Fig. 1. Cross-reactivity between HIV-1 and HIV-2 in the recognition of peptides presented by HLA-B35 by CTL from HIV-infected donors. CTL assays were performed using two week-old bulk CTL cultures from donors with HLA-B35 known to be infected with either HIV-1 or HIV-2, using B35-matched targeted cell-lines pulsed with each of the peptides and a control peptide at a concentration of 10 μM. a) CTL bulk cultures from donors infected with either HIV-1 (AN) or HIV-2 (FS, KS) show specific recognition of HIV nef peptide^75"82, b) Cultures from HIV-2-infected donor 90.17 show specific recognition of HLA-B35 target cell lines pulsed with either the HIV-1 or HIV-2 versions of the HLA-B35-restricted pl7 and p24 peptides, c) CTL grown from a British HIV-1 infected haemophiliac donor are tested on HLA-B35-matched target cells pulsed with increasing concentrations of either the HIV-1 or HIV-2 versions of the peptide pol 329- 337.

Fig. 2. HLA-B35 -restricted CTL using peptide-stimulated cultures from high-risk seronegative donors OXll, 12, 14 and 17. The CTL assays shown were performed on day 12-14 in culture, using a target cell-line matched only at HLA-B35, pulsed for one hour with each of the peptides and a control influenza peptide at 10 μM. The Effector: Target ratio (E:T) varied according to the number of cells that could be harvested from each culture, but was between 12-50.J . The lysis in the presence of a control influenza peptide is shown in the dark columns and that with the specific HIV peptide is shown in the dotted columns. Results were regarded as positive if recognition of the HIV peptide was more than 10% above that of a control peptide on at least two occasions. Donor OX17 was regarded as showing no specific CTL activity and no other positive results were obtained in cultures from the remaining exposed seronegative donors with HLA-B35 or HLA-B53.

Fig. 3. Peptide-stimulated cultures from the second bleed of high-risk seronegative donors OXl l and OX12, a) Recognition of virus-infected cells. Peptide-stimulated cultures were established using the HIV-1 pol peptide as described, and used on day 15 in culture. Target cells were B-LCL from donor FS, matched with the CTL only at HLA- B35, infected with recombinant vaccinia virus expressing full-length HIV-1 pol or an irrelevant protein, 0-galactosidase. E:T ratios were 16: 1 (dark columns) and 4J (striped columns), b) Peptide titration for the HIV-1 pol peptide. Peptide-stimulated cultures from donor OX12 were used on day 22; at an E:T ratio of 6: 1, in a CTL assay using HLA-B35 matched target cells with the HIV-1 pol peptide diluted in the assay to the concentrations shown, c) Time course of CTL activity from the peptide-stimulated cultures. CTL assays using cultures stimulated with either the HIV-1 pol peptide in the case of donor OX12 or the influenza matrix peptide in the case of donor OX11 were performed using similar E:T ratios, over a 24 day period. The figures shown are specific lysis, i.e. with the lysis in the presence of a control peptide substracted from that of the peptide of interest.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT:

(A) NAME: Medical Research Council

(B) STREET: 20 Park Crescent

(C) CITY: London

(E) COUNTRY: United Kingdom

(F) POSTAL CODE (ZIP): WIN 4AL

(G) TELEPHONE: (0171) 6365422 (H) TELEFAX: (0171) 3231331

(ii) TITLE OF INVENTION: Improvements in or Relating to Vaccines

(iii) NUMBER OF SEQUENCES: 7

(iv) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentin Release #1.0. Version #1.30 (EPO)

(2) INFORMATION FOR SEQ ID NO: 1:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

His aa Asp He Val He Tyr Gin Xaa 1 5

(2) INFORMATION FOR SEQ ID NO: 2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 8 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

Val Xaa Leu Arg Met Pro Thr Xaa 1 5 (2) INFORMATION FOR SEQ ID NO 3

O) SEQUENCE CHARACTERISTICS

(A) LENGTH 9 ammo acids

(B) TYPE amino acid

(C) STRANDEDNESS

(D) TOPOLOGY linear

(π) MOLECULE TYPE peptide

(xi) SEQUENCE DESCRIPTION SEQ ID NO 3

Asn Xaa Ser Lys Val Ser Gin Asn Xaa 1 5

(2) INFORMATION FOR SEQ ID NO 4

(l) SEQUENCE CHARACTERISTICS

(A) LENGTH 9 amino acids

(B) TYPE amino acid

(C) STRANDEDNESS

(D) TOPOLOGY linear

(n) MOLECULE TYPE peptide

(xi) SEQUENCE DESCRIPTION SEQ ID NO 4

Pro Xaa He Pro Val Gly Asp He Xaa 1 5

(2) INFORMATION FOR SEQ ID NO 5

( ) SEQUENCE CHARACTERISTICS

(A) LENGTH 9 amino acids

(B) TYPE amino acid

(C) STRANDEDNESS

(D) TOPOLOGY linear

(n) MOLECULE TYPE peptide

(xi) SEQUENCE DESCRIPTION SEQ ID NO 5

Pro Xaa Ser Gly Lys Gly Gly Asn Xaa 1 5

(2) INFORMATION FOR SEQ ID NO 6

(i) SEQUENCE CHARACTERISTICS

(A) LENGTH 9 amino acids

(B) TYPE amino acid

(C) STRANDEDNESS

(D) TOPOLOGY linear (ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

Asn Xaa Val Pro Val Gly Asn He Xaa 1 5

(2) INFORMATION FOR SEQ ID NO: 7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS:

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:

Asn Xaa Asp Val He Leu He Gin Xaa 1 5

Claims

1. A vaccine against AIDS, comprising a peptide sequence of HIV capable of generating a protective cytotoxic T lymphocyte (CTL) response in humans.

2. A vaccine according to claim 1, comprising a peptide sequence of HIV-1 pol protein.

3. A vaccine according to claim 1 or 2, comprising a peptide having a sequence in accordance with any one of the following consensus sequences: HXDIVIYQZ; VXLRMPTZ; NXSKVSQNZ; PXIPVGDIZ; PXSGKGGNZ; NXVPVGNIZ; and NXDVILIQX; where X and Z may be any amino acid residue, and where X may be the same as Z.

4. A vaccine according to claim 3, where X is proline and Z is tyrosine.

5. A vaccine according to any one of the preceding claims, comprising a peptide sequence of HIV capable of being recognised by HLA-B35-restricted CTLs.

6. A vaccine according to any one of the preceding claims, further comprising one or more other HIV-encoded components.

7. A vaccine according to any one of the preceding claims, further comprising one or more peptide sequences of HIV capable of being recognised by CTLs restricted by a human leukocyte antigen other than B35.

8. A vaccine according to any one of the preceding claims, wherein one or more peptide sequences of HIV is present as a fusion polypeptide.

9. A vaccine according to any one of the preceding claims, comprising a bacterial or viral vector, inert carrier or adjuvant.

10. A method of preventing or treating AIDS in a human subject, comprising administering to the individual an effective amount of a vaccine according to any one of the preceding claims.

11. A method of preventing or treating AIDS in a human subject, comprising administering autologous CTL which have been primed with a peptide sequence present in the vaccine of any one of claims 1-5.

12. A method of making a vaccine against AIDS capable of generating a protective CTL response to HIV, comprising mixing with a physiologically acceptable carrier substance an effective amount of a peptide having a sequence in accordance with any one of the following consensus sequences: HXDIVIYQZ; VXLRMPTZ; NXSKVSQNZ; PXIPVGDIZ; PXSGKGGNZ; NXVPVGNIZ; and NXDVILIQX; where X and Z may be any amino acid residue, and where X may be the same as Z.

13. A method according to claim 12, wherein X is proline and Z is tyrosine.

14. A method according to claim 12 or 13, further comprising mixing one or more other peptide epitopes of HIV with the carrier substance.

15. Use of a peptide in making a vaccine against AIDS capable of generating a protective CTL response against HIV in humans, wherein the peptide has a sequence in accordance with any one of the following consensus sequences: HXDIVIYQZ; VXLRMPTZ; NXSKVSQNZ; PXIPVGDIZ; PXSGKGGNZ; NXVPVGNIZ; and NXDVILIQX; where X and Z may be any amino acid residue, and where X may be the same as Z.