WO2000015663A1

WO2000015663A1 - MIP-1α ISOFORM LD78β AND USES THEREOF

Info

Publication number: WO2000015663A1
Application number: PCT/GB1999/003059
Authority: WO
Inventors: Gerard John Graham; Robert John Benjamin Nibbs
Original assignee: Cancer Research Ventures Limited
Priority date: 1998-09-14
Filing date: 1999-09-14
Publication date: 2000-03-23
Also published as: GB9820014D0; AU5875799A

Abstract

The present invention relates to the finding that the isoform of the chemokine MIP-1α, referred to as LD78β or MIP-1αP, ha s high suppressive activity against the CCR5 co-receptor, implicated in the entry of HIV into cells. The invention provides methods to block entry of HIV into cells using MIP-1αP, the use of MIP-1αP in the treatment of HIV and a novel form of this chemokine, amino-oxypentane MIP-1αP, which has particularly advantageous properties.

Description

MlP-l ISOFQRM LD78B AND USES THEREOF

The present invention relates to an isoform of the chemokine MlP-lα and its use as a receptor antagonist.

Background to the Invention.

The mammalian immune system is controlled by a complex series of interactions between signalling molecules. One group of such molecules is the β-chemokines, which are a family of proteins of a molecular mass of about 8,000 kD. Chemokines play diverse roles in both inflammatory and non inflammatory situations¹ via activation of a family of heptahelical G protein coupled receptors². In addition, they have been implicated in the pathogenesis of HIV (ref . 3) . Chemokines are active on lymphocytes and monocytes by means of these G protein coupled receptors. Of this G-protein family, the receptor CCR5 is of particular significance in medicine. This receptor is implicated in the transmission of HIV-1 since this virus requires CCR5 as a co-receptor (along with CD4) for entry into T-cells. A truncated allele of CCR5 (which causes the loss of the three C-terminal transmembrane domains) results in resistance to HIV-1 infection or a reduction in the severity of disease progression (Samson et al , Nature 382; 722-725, (1996)).

It has been reported that the β-chemokines RANTES, MlP-lα and MlP-lβ are HIV-suppressive factors produced by CD8⁺ T-cells

(Ref. 4). Specifically, the chemokines MlP-lα, MlP-lβ and RANTES have been shown to be major HIV1 suppressive factors produced by T cells⁴. Subsequently, the chemokine receptor, CCR5, which recognises these ligands, was shown to serve as a critical cofactor mediating the entry of M-tropic HIV1 strains^5"11. Current data suggests RANTES is the most effective natural antagonist of CCR5-dependent HIV1 entry, whilst MlP-lα is somewhat less efficient⁵'^9,12.

It is believed that the binding of these factors to the CCR5 receptor is responsible at least in part for the suppressive effects of these β-chemokines. Current data suggest that RANTES is the most effective natural antagonist of CCR5- dependent HIV-1 entry into CD4 cells.

The human MlP-lα chemokine has been found to exist in two non-allelic isoforms (Refs 13, 14). The term MlP-lα, when used without qualification, is generally used to refer to the more widely used isoform, also called LD78α, the predominant experimentally used isoform. The other isoform, LD78β, differs in its mature sequence by three amino acids (see

Figure 1) . Of these, one is the substitution of serine at position 2 of the LD78α sequence by proline.

No functional significance attributed to this difference has been described, and indeed the widespread experimental use of the LD78α isoform suggests that those of skill in the art do not attach any major significance to this difference. This may be because it has been reported that the MlP-lα expressed in some cells is a truncated version of the LD78 sequence shown herein in Figure 1, in which the first 4 residues of the sequence are missing (e.g. Cocchi et al , Ref. 4). This truncation removes this region of difference between the isoforms .

Disclosure of the invention.

Recently, we have cloned a chemokine receptor which we have termed D6 (refs. 15-16) . Curiously, whilst murine and human D6 bind murine MlP-lα with high affinity, the closely related human MlP-lα interacts poorly with these receptors. Similarly, murine CCR5 binds murine MlP-lα with high affinity but does not recognise the putative human homologue¹⁵. The isoform of human MlP-lα that we have used in these studies is a commercially available form that has the amino acids ASLA at the amino terminus, and has previously been referred to as LD78α to distinguish it from a naturally occurring non- allelic variant, LD78β (refs. 13-14). The close sequence similarity between LD78α and murine MlP-lα led us to examine the possible amino acid residues that may be responsible for the discriminatory binding of these ligands. As shown in Figure 1, the only consistent feature present in the mCCR5 and hD6 ligands but absent in human LD78 is the proline residue at position 2 of the mature protein. LD78β, however, does have a proline residue in position 2, with two reciprocal serine/glycine swaps in the region between cysteines 3 and 4 being the only other differences between the mature LD78α and β peptides (Figure 1) . Also, consideration of other differences in the putative signal sequences of the two isoforms¹⁴ by predictive algorithms (not shown) , suggested there may be variation in the site of signal peptidase cleavage between the two isoforms during secretion, and that LD78 may not be produced with the ASLA amino terminal amino acids. Indeed, human MlPl has been shown to be produced naturally by CD8+ T cells⁴ as a -4 variant with the amino terminal sequence ADTPT. We have worked extensively with the LD78β isoform¹⁷ and numerous amino terminal sequencing exercises have consistently revealed a 'full length' amino terminus APLAADTPT, suggesting that the - 4 isoform identified in T cells is likely to be a naturally generated form of LD78 (henceforth referred to as LD78α-4) perhaps indicating alternative signal peptidase cleavage site usage by the two LD78 isoforms.

As a result of our observations, we have surprisingly found that LD78β, unlike the more commonly used a isoform, has high affinity for the promiscuous β-chemokine receptor D6, is the most potent natural CCR5 agonist and displays an HIV1 suppressive activity that is markedly higher than any other known natural CCR5 ligands. A proline residue at position 2 of LD78β is implicated in its enhanced activity and we propose that LD78 and β be renamed MlP-l S and MlP-lαP respectively to reflect their functional differences. In addition, our studies suggest that MlP-lαP more accurately represents the human functional homologue of murine MlP-lα.

The present invention thus provides a method of inhibiting the entry of an HIV-1 virus into a CD4⁺ cell which method comprises : bringing said cell and said virus into contact in the presence of a polypeptide comprising the LD78β isoform of IP-lα under conditions wherein said polypeptide binds to a CCR5 receptor on the surface of said cell and inhibits entry of said virus into the cell.

The invention further comprises a diagnostic method for determining the likely effectiveness of D78β therapy in a patient infected with, or at risk of infection from, HIV, which method comprises: isolating a sample containing CD4⁺ T-cells from said patient; bringing said T-cells into contact with a polypeptide comprising the LD78β isoform; and determining the extent to which said polypeptide can bind to said cells.

In the foregoing method, the sample may be a human body sample, such as blood or serum.

The present invention further provides the use of the polypeptide comprising the LD78β isoform of MlP-lα for the manufacture of a medicament for the treatment of HIV infection.

The polypeptide comprising the LD78β isoform of MlP-lα may be in the form of the wild-type protein or may comprise one or more side-chain, N-terminal or C-terminal modifications. For example, the polypeptide may be produced by expression by recombinant means known per se in the form of a fusion protein, wherein said LD78β isoform is fused to a second polypeptide sequence, for example a signal sequence which directs the fusion protein out of the host cell .

The LD78β isoform may also be chemically linked to groups, such as N-terminal protecting groups. For example, Simmons et al (Science 276, 276-279, 1997) describe modification of

RANTES by N-terminal protection with amino-oxy pentane (AOP) , Such a group, or another amino-oxy alkane (e.g. C2-10 alkane) may be linked to the N-terminal of LD78β using analogous methodology. W096/41813 provides methods for the transamination of the N-terminal of polypeptides (citing Dixon & Fields, Meth. Enzymol . 25, 409-419, 1979) and the subsequent protection of the N-terminal of such modified polypeptides with polyethylene glycol and derivatives thereof .

These and other methods known in the art may be used to modify LD78β to provide polypeptides useful in the present invention, for example in the diagnostic and other methods described herein.

The copy number ofLD78β varies between individuals, with up to six copies being present in some humans . The present findings indicate that the progreesion of HIV infection in an individual may be related, at least in part, to the copy number of LD78β. Thus in a further aspect, the invention provides a method for the diagnosis of an individual ' s susceptibility to the development or progression of HIV infection, which method comprises measuring the copy number of LD78β in an individual's genome, or measuring the level of expression of LD78β in the individual. Where increased copy numbers (i.e. more than one per haploid genome) are found, then the individual may be more resistant to the effects of HIV and to the progression of the infection into AIDS. The individual may be someone with or without HIV infection, or with unknown HIV status.

Description of the Drawings.

Figure 1 shows an alignment of the predicted protein sequence of some mature β-chemokines. Human and murine sequences are prefixed with ^vh' or ^xm' respectively. The two human MlP-lα isoforms (h D78 and β) are indicated with '{^,. The proline residue at position 2 (serine in hLD78α) , and the reciprocal serine/glycine changes between LD78α and β, are in bold and underlined. Binding to human D6 and murine CCR5 is indicated to the right of the alignment, and is derived from references 16 and 15 respectively.

Figure 2 shows the displacement of I¹²⁵-m IP-lo: from CHO cells expressing human MlP-lα receptors (A, Human D6 ; B, Human CCR5; C, Human CCR1) by human MlP-lα / LD78 isoforms.

Figure 3 shows dose-response curves for ligand-induced calcium ion fluxes in HEK293 cells expressing human MlP-lα receptors (A, CCR5-293 cells; B, CCR1-293 cells) .

Figure 4 shows that LD78β is a more potent inhibitor of HIV entry through CCR5 than other MlP-lα variants and RANTES (A, macrophage tropic strain JR.FL; B, strain ADA; C & D, SF162) . Figure 5 shows that AOP-LD78β has a higher binding affinity for CCR5 than unmodified LD78β.

Figure 6 shows that AOP-LD78β exerts its improved effects through down-regulation of cell surface receptors.

Detailed Description of the Invention.

Reference herein to the "LD78β isoform" means the protein of the sequence shown herein in Figure 1, or a variant thereof which retains the CCR5 binding ability of LD78β but which differs by from 1 to 10, such as 2 , 3, 4, 5 or 6 amino acids. A difference is one substitution, deletion or insertion. Where the difference is a substitution, the substitution may be conservative or non-conservative. Conservative substitutions may be made for example according to the following table, where amino acids on the same block in the second column and preferably in the same line in the third column may be substituted for each other:

Alternatively, any amino acid may be replaced by a small aliphatic amino acid, preferably glycine or alanine.

Where the LD78β is a variant of the sequence shown in Figure 1, the difference will be other than at position 2, where the proline residue will be retained. Preferably, the difference is also other than in the N-terminal 12 amino acids, up to and including the cys-cys pair, and also other than at any other cysteine residue.

A particularly preferred class of deletions or substitutions is the deletion or substitution of a charged amino acid, particularly one in the C-terminal region of the LD78β. We have previously found that such changes result in a protein which does not aggregate to the same extent as the corresponding wild type protein, thus making it more effective as an antagonist.

Substitutions of a charged amino acid which result in a loss of overall charge are preferred. Charged amino acids in the LD78β sequence of Figure 1 are D6, R18, D27, E30, K37, K45, R46, R48, D53, E56, E57, K61, D65 and E67. Preferably, E67 is substituted, either alone or in combination with one or more of D53, E56, E57, K61 and D65.

Particularly preferred specific substitutions may be lys (K) to asn (N) , asp (D) to asn (N) , or glu (E) to gin (Q) .

Thus for example a specific LD78β isoform which may be used in the invention is that shown in Figure 1 apart from the substitutions E67Q, or E67Q + D65N, or E65Q + D65N + E57Q. These and other isoforms which disaggregate may be made by methods analogous to those described in, for example, W094/24285.

Where a polypeptide comprising the LD78β isoform is to be used in a method of inhibiting the entry of an HIV-1 virus into a cell, the cell will be a CD4⁺ cell which is also expressing a CCR5 receptor capable of acting as a co-receptor for the HIV-1 virus. The cell may be any mammalian cell type, which either naturally expresses both CD4 and CCR5, or which has been genetically engineered to express these receptors. Examples of the latter are illustrated in the accompanying examples, and also widely available in the art, given the widespread interest in developing AIDS therapies which target the CCR5 receptor. Suitable cells include cultures of PBMCs, COS cells, CHO cells, HEK 293 cells and Hos (human osteosarcoma) cells .

The HIV-1 virus used in the method may be any nonsyncytium- inducing strain which uses CCR5 as a co-receptor, or an enveloped pesudotyped virus which uses this receptor, such as the JR.FL virus or ADA pseudotyped virus (for both see Deng et al, Nature 381, 661-666) .

In the method of the invention, the amount of the LD78β isoform which is present may vary according to the needs of those of skill in the art. For example, the method may be practised in vi tro for the assay of compounds capable of binding to and/or antagonising the CCR5 receptor. Generally, the concentration of LD78β in such assays will range from 100 p to 100 nM, preferably from 100 pM to 10 nM. This range of concentrations will be suitable for other in vi tro or in vivo purposes as well.

Similarly, where the invention is in predicting the likely effectiveness of LD78β therapy in an individual, the concentration of polypeptide may be in a similar range.

LD78β polypeptides may be formulated into pharmaceutical compositions. The compositions comprise the polypeptide together with a pharmaceutically acceptable carrier or diluent. Pharmaceutically acceptable carriers or diluents include those used in formulations suitable for oral, topical, or parenteral (e.g. intramuscular or intravenous) administration. The formulations may conveniently be presented in unit dosage form and may be prepared by any of the methods well known in the art of pharmacy. Such methods include the step of bringing into association the active ingredient with the carrier which constitutes one or more accessory ingredients. In general the formulations are prepared by uniformly and intimately bringing into association the active ingredient with liquid carriers or finely divided solid carriers or both, and then, if necessary, shaping the product.

For example, formulations suitable for parenteral administration include aqueous and non-aqueous sterile injection solutions which may contain anti-oxidants, buffers, bacteriostats and solutes which render the formulation isotonic with the blood of the intended recipient; and aqueous and non-aqueous sterile suspensions which may include suspending agents and thickening agents, and liposomes or other microparticulate systems which are designed to target the polypeptide to blood components or one or more organs.

Suitable liposomes include, for example, those comprising the positively charged lipid (N[l- (2 , 3-dioleyloxy) propyl] -N,N,N- triethylammonium (DOTMA) , those comprising dioleoyl- phosphatidylethanolamine (DOPE), and those comprising 3β[N- (n¹ ,N' -dimethylaminoethane) -carbamoyl] cholesterol (DC-Choi) .

Compositions may comprise any desired amount of the LD78β polypeptide. In part this will depend upon the intended formulation and its intended use. By way of general guidance the composition may comprise from about 1% to about 99%, for example from 10% to 90% of the LD78β polypeptide. The composition may comprise a mixture of more than one, for example two or three, LD78β polypeptides.

Where in vi tro use is contemplated, this will include ex- vivo, e.g. in the treatment of bone marrow other tissue from a subject which may be reimplanted into the subject after treatment .

In general, the methods of treatment will involve administering to a patient in need of treatment an effective amount of an LD78β polypeptide (or composition thereof) . Suitable routes of administration of compounds of the invention include oral or parenteral, and will depend in part upon the intended use and the discretion of the physician.

Thus the present invention also provides a method of treating an individual infected with HIV with an effective amount of an LD78β of the invention. In particular, the treatment may provide the use of an N-terminal protected LD78β.

At present, there is no established cure for HIV infection. Thus "treatment" includes the alleviation of symptoms, including the delay in the progression of the disease, the stimulation of the immune system to resist other illnesses, or the like, rather than necessarily cure.

Treatment of HIV infection may be by administering an LD78β by any suitable route, including for example intravenous or subcutaneous injection.

The amount of an LD78β polypeptide administered to a patient is ultimately at the discretion of the physician, taking account of the condition of the patient and the condition to be treated. Typical amounts of polypeptide administered may be in the region of lμg/kg body weight to 10 mg/kg body weight and preferably between lμg and lmg /kg body weight . The amount, formulation and route of administration should be chosen to achieve an effective level of the polypeptide of around 50pg/ml to lOμg/ml circulating in blood or other target tissue.

The dose for administration to a patient desirably will be selected to balance between the need for an effective amount of the polypeptide comprising the LD78β isoform on the one hand, and on the other, the need to avoid undesirable side- effects such as inflammation, nausea, etc. The present findings show that the LD78β polypeptide is advantageous over cytokines, including RANTES, which have been proposed for therapeutic purposes, particularly in view of its high affinity for the CCR5 receptor. For example, we have found that in the HIV-1 entry inhibition assay, performed with the JR.FL pseudotyped virus according to the protocol of Deng et al (Ref. 5), which is incorporated herein by reference, LD78β provides 100% inhibition of entry at a concentration of lOnM, compared to only 50% inhibition of entry with lOOnM of LD78 . Similar results were also achieved with the ADA pseudotyped virus. Thus LD78β achieves a more complete inhibition of entry of HIV-1 into a cell via CCR5 at a 10-fold lower concentration. Even compared to RANTES, we have found a more complete inhibition of entry. Using SF162 virus we have shown that at 10 nM RANTES achieves about 75 % inhibition of entry using the assay of Connor et al , Virology, 206, 935- 944, 1995, which is incorporated herein by reference, whereas LD78β achieves effectively 100% at the same concentration. Thus it is about 1 log more active.

Accordingly, the invention allows the use of an LD78β polypeptide for in vitro or in vivo uses to achieve a desired effect (particularly antagonism of HIV-1 entry to a cell) at a concentration of 1 log less than the concentration of either or both of RANTES or LD78 would be required to achieve the same effect .

Doses may be administered continuously, e.g in the form of a drip, or at discrete intervals, e.g twice daily, daily, weekly or monthly.

The LD78β may also be administered by way of a gene construct for delivery by gene therapy. For example, naked DNA constructs are known to be taken up by cells in muscle tissue. Alternatively, gene delivery vector systems such as those based upon viral vectors, e.g. adenoviral or retroviral vectors may be used to deliver a construct comprising nucleic acid encoding LD78β operably linked to a suitable promoter.

The invention is illustrated by the accompanying examples

Example 1: Binding of LD78 isoforms to murine MlP-lα receptors .

We have therefore tested the ability of the various LD78 isoforms to interact with known murine MlP-lα receptors stably expressed on CHO cells (Table 1) . In contrast to LD78α-4, LD78β binds with high affinity to both murine CCR5 and D6 , whilst LD78α-4 is a significantly better ligand than LD78β for murine CCR1. To assess the importance of the proline residue at position 2 of LD78β we have tested two other proteins. One, LD78β-4, has the first four amino acids (APLA) of LD78β removed, and a second, LD78α, has the four amino acids ASIA reintroduced onto LD78 -4. Both of these proteins behaved like LD78 -4 on the murine receptors, though there was a slight decrease in binding to CCRl by LD78α. This is shown in Table 1. Table 1

Table 1: Dissociation constants (in nM) for displacement of I¹²⁵-mMIP-lα from CHO cells expressing murine MlP-lα receptors, by different forms of human MlP-lα. Displacement curves were generated like those shown in Figure 2. Dissociation constants were calculated from these data using LIGAND software²⁵. * indicates data taken from reference 15. Chemokine sources: LD78α is 'human MIP-l ' purchased from Peprotech, London, UK; LD78α-4 is 'human MIP-lα' purchased from R&D Systems, Abingdon, UK; LD78β was prepared as described in reference 17 and LD78β-4 prepared by similar methods .

Example 2: Binding of LD78 isoforms to human receptors.

CHO cells expressing human receptors were also tested and showed similar results. Experiments were performed as described in refs 15 and 16, using I¹²⁵-labelled murine MlP-lα at a constant concentration of 600pM (for D6) , 3nM (for CCR1) or 9nM (for CCR5) , whilst increasing the concentration of unlabelled competitor LD78 protein. Remaining radioactivity bound after 90min, and three ice-cold PBS washes, was determined. Each point was done in triplicate, the average taken, and converted into a percentage of radioactivity bound in the absence of any unlabelled competitor chemokine.

Thus, with human D6, whilst the LD78β isoform binds with high affinity (Kd 5.5nM) , the -4 variants of the two isoforms bind only weakly (Kd 77nM for LD78α-4 and Kd 124nM for LD78β-4) and LD78 , which has the addition of the four amino acids ASLA to LD78α-4, does not exhibit an enhanced binding affinity (Figure 2A, □ = LD78β; 0= LD78β-4; o = LD78α;

. With human CCR5 , there is an ~6-fold higher binding affinity of LD78β (Kd 6.2nM) than is seen for the -4 LD78 proteins (Kds 35nM) , and the addition of the ASLA amino acids to LD78α-4 results in a further reduction in binding affinity (Figure 2A, symbols as Fig. 2A) . These results show that the proline residue at position 2 of LD78β is responsible for the enhanced binding of this isoform to murine and human D6 and CCR5. Conversely, human CCR1 binds the -4 variants of the LD78 isoforms consistently better than the 'full-length' proteins (Figure 2C, symbols as Fig. 2A) .

Example 3: Signalling through human CCR5 and CCR1 receptors.

These isoform binding variations are also reflected in the dose response for signalling through human CCR5 and CCR1 , as assayed by Ca²⁺ flux into HEK293 cells stably transfected with these genes.

Experiments were performed according to Nibbs et al¹⁵. ~6xl0⁶ cells loaded with Fura-2-AM were incubated at 37°C in a continuously stirred cuvette in a Perkin-Elmer LS50 Spectrometer (340nm (l_ex) ; 500nm (l_em) ) and fluorescence emission recorded every 100ms. After 2min, ligand was added to a defined concentration and fluorescence recorded every 100ms for a further 2min. To allow to control for variations in Fura-2 loading and cell number, cells were lysed with detergent and fluorescence emission recorded. Also, to control for day-to-day experimental variation, a full dose- response curve for murine MlP-lα was performed each time a different ligand was tested (data not shown) . The peak of ligand-induced fluorescence increase (after alterations to account for these variations) is presented as a percentage of the maximum achievable peak with each particlar transfected cell line.

With 'full-length' proteins on CCR5, half maximal signalling potency is seen at approximately lOnM for LD78 and at approximately 750pM for LD78β (Figure 3A; □ = LD78β; 0= LD78β-4;

X = hRANTES ; + = hMIP-lβ). LD78α was unable to induce a Ca²⁺ flux of the same magnitude as those seen with LD78β, even when it is added at very high concentrations (>lmM) (not shown) . The LD78α-4 variant shows a slight increase in signalling potency through CCR5 compared to LD78α, whilst removal of the terminal 4 amino acids from the β isoform reduces its activity approximately 10-fold. The potency of signalling with LD78β is 5-10 times that seen with RANTES and -100-fold that seen with MlP-lβ, and is therefore the most active CCR5 agonist described to date. We have detected Ca²⁺ fluxes with MCP2 , which has recently been identified as a CCR5 ligand¹⁸, but this ligand is not as potent as MlP-lβ in this assay. Signalling through human CCR1 reflects the results from the binding studies and suggests little difference in signalling potency between the 'full-length' and β isoforms of LD78 (Figure 3B, symbols as Fig. 3A) . The -4 variants however have a slightly higher signalling potency than the full length peptides. We have so far been unable to demonstrate a signalling role for human D6 (ref. 16) and so the functional consequences of the high affinity binding of LD78β to this receptor remain to be determined.

Example 4: LD78β as an HIV supressive agent.

Given the very high affinity binding of LD78β to CCR5, we investigated the potency of this chemokine as an HIV suppressive agent. Studies were carried out to investigate the potency of 'full length' and -4 variants of both isoforms of LD78 in suppressing entry of JR.FL envelope pseudotyped virus into CEMxl74-CCR5 cells. 5xl0⁴ CEMxl74-CCR5 cells, pretreated with chemokine for 30min, were incubated for 4 hrs with luciferase virus (5ng p24) , pseudotyped by either JR.FL (Fig. 4A; π = LD78β; 0= LD78β-4; O = LD78α_;

+ = AOP-RANTES) or ADA (Fig. 4B, symbols as Fig. 4A) according to protocol previously described⁵. Medium was then changed and the luciferase activity measured 3 to 5 days post-infection. The extent of inhibition of HIV entry was determined by comparing luciferase activity of chemokine-treated cells, with untreated cells⁵.

These studies demonstrated LD78β to be a substantially more potent suppressor of HIV entry that either LD78 or the -4 variants (Figure 4A) . Greater than half maximal inhibition of JR.FL pseudotyped viral entry was achieved with concentrations of LD78β as low as 5ng/ml whilst 10-20 fold higher concentrations of the other LD78 variants were required before 50% inhibition of viral entry was achieved. Similar results were seen with virus pseudotyped with the envelope of the ADA HIV1 strain, although all the variants were less able to inhibit entry (Figure 4B) .

Further investigations using replication-competent virus entry into CEMxl74-CCR5 cells and peripheral blood mononuclear cells (PBMC) were performed. The CEMxl74-CCR5 cells were infected with replication competent SF162 virus after pretreatment with chemokine. p24 concentration was determined after 5 days using ELISA (ref. 26) . These results were compared to the [p24] released from cells that had not been exposed to any chemokine. The extent of reduction by chemokine treatment determined the percentage of inhibition of HIV entry. Similarly, 5xl0⁵ PBMC were infected with replication-competent SF162 virus. PBMC were purified from HIV1 seronegative donors, PHA/IL-2 activated, and infected with virus as previously described²⁶. Five days post- infec ion [p24] was determined and percentage inhibition of HIV entry calculated relative to chemokine-untreated control as above .

The results are shown in Figures 4C and 4D, respectively (□ = LD78β;

X = hRANTES; + = AOP-RNATES) . These studies suggest that in addition to being by far the most effective LD78 variant in suppressing HIV entry into target cells, LD78β is also more potent than RANTES, although not quite as effective as the N-terminally modified form of this chemokine, AOP RANTES (ref. 19) . Note that RANTES and LD78 -4 can enhance HIV entry into PBMCs at the lower concentrations tested (Figure 4D) . These results present LD78β as the most potent naturally-occurring inhibitor of HIV1 entry through CCR5. Variants of this protein, such as AOP-linked LD78β, may exhibit enhanced HIV1 entry inhibition and/or receptor antagonism and have potential as HIV1 therapeutics.

LD78 -4 and LD78β-4 behave in a similar fashion in all assays we have performed, and we therefore believe that it is the proline residue at position 2 of LD78β that is crucial for its strong interaction with D6 and for its increased binding to, and signalling through, CCR5. It is likely that proline 2 in other β-chemokines is necessary for high affinity binding to D6. This, however, is not sufficient for D6 interaction, as SDF1 also contains a proline residue at position 2 yet shows no potential to bind to D6 (data not shown) . Interestingly, this residue in SDF1 is crucial for activation of its receptor CXCR4 , which acts as an entry cofactor for T-tropic HIV1 strains^20"23, and mutation of this residue to glycine generates a high affinity CXCR4 antagonist²³. Thus, proline 2 must be presented in the context of a β-chemokine to permit high affinity D6 interaction, whilst with CCR5 other domains are likely important in restricting the ligands for this receptor to MlP-lα, lβ, RANTES and MCP2. It is intriguing that both major HIV entry coreceptors have a preference for a proline residue at position 2 during receptor activation.

Murine MlP-lα consistently behaves in a similar fashion to LD78β, by binding avidly to murine and human CCR5 and D6, and slightly less well to CCR1 (ref. 15 and data not shown) and by being generated consistently in a 'full length' form. LD78α, on the other hand, either as a full length or -4 variant protein, only binds well to CCR1. We conclude therefore that LD78β is the functional human homologue of murine MIP-l : LD78 should be considered to be a related, but functionally distinct chemokine. To simplify the nomenclature of these two non-allelic variants and emphasise the importance of proline 2, we propose that the LD78β gene be renamed MlP-lαP and that LD78α be called MlP-lαS, indicating the change in predicted amino acid sequence to serine at this position.

The LD78β/MIP-lαP copy number varies between individuals, and this gene can in fact be absent from some individuals^13"14. β- chemokine production has been reported to be associated with a number of inflammatory and autoimmune diseases¹, and deletion of the MlP-lα gene in mice dramatically alters responses to several infectious agents²⁴. LD78β/MIP-lαP gene dosage may alter an individuals response in these pathological situations. Moreover, in light of the potent HIV entry inhibition by LD78β/MIP-l P described here, it is of considerable interest to test whether gene copy number affects the rate of progression to AIDS in HIV-infected individuals.

Example 5: Receptor binding studies with AOP-MIP-lαP. An AOP derivative of MlP-lαP was synthesised using solid phase synthesis (by Gryphon Sciences, CA, USA) in which an AOP-proline group was attached to the N-terminal of LD78β in place of proline.

(An alternative is to use similar techniques to those of Simmonds et al (Ref. 19), by generating an LD78β variant with serine in place of alanine at position 1, and modifying the serine to generate an aldehyde group which is then reacted to the hydroxylamine derivative of aminooxypentane to provide the AOP-LD78β molecule.)

The AOP-LD78β molecule was tested for binding to the human CCR5 receptor expressed in CHO cells using the methods outlined above in Example 2, but with labelled AOP-MIP-lαP being used at 20 nM.

The displacement of I¹²⁵-mMIP-lα (❖, □) or AOP-MIP-l P (Δ) is shown in Figure 5. As indicated, the results demonstrate that AOP-MIP-lαP is a more potent blocker of murine MlP-lα binding to CCR5 than MlP-lαP itself. The basis for this enhanced blocking activity appears not to be due to an enhanced affinity of AOP-MIP-lαP for CCR5 as AOP-MIP-lαP displacement of AOP-MIP-lαP from this receptor shows it to have a binding affinity that is broadly similar to that seen with the unmodified MlP-lαP. It is likely that the enhanced blocking seen with AOP-MIP-lαP is a reflection of the enhanced ability of this ligand to down-regulate cell surface receptor levels thus effectively denuding the cell surface of receptors. Essentially this is the same mechanism that is used by AOP-RANTES in blocking CCR5 function.

The fact that MlP-lαP is similar in potency to AOP-RANTES in anti HIV assays and that the AOP-MIP-lαP variant has an approximately 5-10 fold enhanced ability to block CCR5 binding of MlP-laP indicates that it is likely to be a more potent anti -HIV therapeutic than AOP-RANTES. This means that clinically relevant concentrations of AOP-MIP-lαP will be more easily achieved than with AOP-RANTES indicating that AOP-MIP-l P may a more attractive clinical agent.

Example 6: Receptor signalling studies with AOP-MIP-lαP.

The studies of Example 3 were repeated using the AOP-MIP-l P. It was found that AOP-MIP-lαP is similar in potency to its unmodified counterpart in signalling assays on either CCRs 1 or 5. This again supports the conclusion of Example 5 that the major enhancement of blocking potency seen with AOP-MIP- lαP is a consequence of more efficient receptor down regulation.

Example 7: Receptor down-regulation studies.

CHO cells expressing human CCR5 receptor as described in example 2 were incubated with 100 nM of AOP-MIP-lαP or MIP- l P for 1 hour, then washed and incubated with medium for the specified time periods indicated on Figure 6. An azide containing binding assay (see Refs. 15 & 16) was performed to measure the downregulation of receptors. As indicated, a 1 hour incubation with AOP-MIP-l P induces an 80% downregulation of cell surface receptor levels compared with the 50% seen with MlP-lαP. Even more significantly is the observation that the AOP-MIP-l P downregulation is much more prolonged than that seen with the unmodified MlP-lαP. In fact, over the 60 minute recovery period studied in the experiment, the CCR5 receptor levels in the AOP-MIP-lαP treated cells barely recovered from their 20% level whilst those in the MlP-lαP treated cells recovered essentially completely within 40 minutes. Ref erences .

1. Rollins, B.J. Blood 90, 909-928 (1997).

2. Murphy, P.M. Cytokine and Growth Factor Reviews 7, 47- 64 (1996) . 3. Clapham, P.R. Trends Cell Biol . 7, 264-268 (1997).

4. Cocchi, F. eϋ al . Science 270, 1811-1815 (1995).

5. Deng, H. et al . Nature 381, 661-666 (1996).

6. Alkhatib, G. et al . Science 272, 1955-1958 (1996).

7. Doranz, B. J. et al . Cell 85, 1149-1158 (1996). 8. Choe, H. et al . Cell 85, 1135-1148 (1996).

9. Dragic, T. et al . Nature 381, 667-673 (1996) .

10. Simmons, G. et al . J^". Virol . 70, 8355-8360 (1996).

11. Zhang, L., et al Nature 383, 768 (1996).

12. Trkola, A. et al . J. Virol. 72, 396-404 (1998). 13. Irving, S. G. et al . Nuc . Acids Res . 18, 3261-327

(1990) .

14. Nakao, M. , et al . Mol . Cell . Biol . 10, 3646-3658 (1990).

15. Nibbs, R. J. B., et al . J. Biol . Chem . 272, 12495-12504

(1997) . 16. Nibbs, R. J. B., et al . J. Biol . Chem. 272, 32078-32083 (1997) .

17. Graham, G. J. , et al . Growth Factors 7, 151-160 (1992).

18. Gong, W. et al . J. Biol . Chem . 273, 4289-4292 (1998).

19. Simmons, G. et al . Science 276, 276-279 (1997). 20. Feng, Y., et al . Science 272, 872-877 (1996).

21. Oberlin, E. et al . Nature 382, 833-835 (1996) .

22. Bleul, C. C. et al . Nature 382, 829-833.

23. Crump, M. P. et al . EMBO J. 16, 6996-7007 (1997).

24. Cook, D. N. et al . Science 269, 1583-1585 (1995). 25. Munson, P. J. & Robard, D. Anal . Biochem . 107, 220-239 (1980) . 26. Connor, R. I., et al . Virology 206 , 935-944 (1995).

Claims

1. A method of inhibiting the entry of an HIV-1 virus into a CD4⁺ cell which method comprises : bringing said cell and said virus into contact in the presence of a polypeptide comprising the LD78╬▓ isoform of MlP-l╬▒ (MlP-l╬▒P) under conditions wherein said polypeptide binds to a CCR5 receptor on the surface of said cell and inhibits entry of said virus into the cell .

2. A diagnostic method for determining the likely effectiveness of LD78╬▓ (MlP-l╬▒P) therapy in a patient infected with, or at risk of infection from, HIV, which method comprises: isolating a sample containing CD4⁺ T-cells from said patient ; bringing said T-cells into contact with a polypeptide comprising the LD78╬▓ isoform; and determining the extent to which said polypeptide can bind to said cells.

3. A method according to claim 2 wherein said sample is a human body sample selected from blood or serum.

4. A method according to any one of the preceding claims wherein said LD78╬▓ isoform comprises an amino-oxypentane moiety at its N-terminus.

5. The use of a polypeptide comprising the LD78╬▓ isoform of MlP-l╬▒ for the manufacture of a medicament for the treatment of HIV infection.

6. Amino-oxypentane LD78╬▓ (AOP-MIP-l╬▒P) .

7. AOP-MIP-l╬▒P for use in a method of treatment of the human or animal body.

8. A method of treating HIV infection which method comprises administering to a patient infected with HIV an effective amount of a chemokine selected from the group MIP- l╬▒P and AOP-MIP-l╬▒P.

9. A method of diagnosis or prognosis of the course of HIV infection in an individual which method comprises determining the copy number in the genome of the gene for MlP-l╬▒P or its copy number in a body sample .