Title:  Anti-herpesviral agent

United States Patent:  6,337,074

Assignee:  Medical Research Council (London, GB)

Filed:  January 25, 1999

PCT NO:  PCT/GB97/02025

102(e) Date:  January 25, 1999

PCT PUB. Date:  February 5, 1998

Foreign Application Priority Data:  Jul 26, 1996[GB] (9615730)

An antiviral agent capable of disrupting the association of two viral proteins required for DNA replication in herpesviruses. The agents disrupt the association of UL8 and POL in HSV-1 or the association of equivalent homologues of these proteins in other herpesviruses (for example UL 102 and UL54 in HCMV). Suitable agents are peptides which mimic the C-terminal or C-proximal portion of UL8 (or its homologues) for example the peptide IELVFTGVLAGVWGEGGKFV. Peptidomimetic compounds of such peptides are also suitable anti-viral agents. An assay to test for agents capable of disrupting association of POL and UL8 (or homologues thereof) is also described.

Description of the Invention

The present invention relates to an anti-viral agent effective against herpesviruses and to an assay for screening for other suitable anti-viral agents.

Herpesviruses include Herpes Simplex Virus types 1 and 2 (HSV-1 and HSV-2), Human Cytomegalovirus (HCMV), Epstein-Barr Virus (EBV) and Equine herpesviruses 1 and 4 (EHV-1 and EHV-4). The term "Herpesvirus" is used herein to refer to any virus of the herpesvirus family, including viruses in the .alpha. group (e.g. HSV 1 & 2, EHV 1 & 4), the .beta. group (e.g. HCMV) and in the .gamma. group (e.g. EBV).

Infections due to HSV have been successfully treated for many years through use of the drug acyclovir, a nucleoside analogue. Acyclovir is relatively non-toxic to the human host since it does not adversely affect the activity of the mammalian homologue of the targeted viral protein. However, similar low toxicity regimes for treating all herpesviruses have not yet been found. Whilst HCMV is treatable via use of the drug gancyclovir (Coen, 1992) the application of this drug is limited by its toxicity, poor bioavailability and the emergence of drug-resistant variants (reviewed by Coen 1992; Haffey & Field 1995; Filley et al 1995). A low-toxicity treatment for HCMV is particularly of interest as infection by this virus can cause congenital abnormalities in the newborn exposed to the virus by maternal transmission, and is also extremely problematic to immunocompromised patients, for example patients suffering from AIDS, or those on immunosuppressive therapy for cancer or following organ transplant.

The genome of herpes simplex virus type 1 (HSV-1) encodes seven proteins essential for origin dependent viral DNA synthesis (Wu et al., 1988). The genes encoding these proteins, and their protein products, are known in the art as UL5, UL8, UL9, UL29, UL30, UL42 and UL52. (McGeoch et al., 1988). Frequently the names of the genes are italicized, eg UL5, to avoid possible ambiguities. The UL30 protein, the catalytic subunit of the heterodimeric HSV-1 DNA polymerase, is also known as POL. Homologues of all seven genes have been identified in other alphaherpesviruses and human herpesviruses 6 and 7 (HHV-6 and HHV-7). Other beta- and gammaherpesviruses encode homologues of all these proteins except UL9. For convenience the terminology of the HSV-1 proteins will be used to refer not only to that particular protein but also its equivalent in other herpesviruses. Thus, as used herein the term "UL8" refers not only to UL8 of HSV-1 itself, but also to the HCMV homolgue UL102 and to equivalent homologues in other herpesviruses. Similarly, as used herein the term "POL" (or "UL30") refers not only to POL of HSV-1 itself, but also to the HCMV homolgue UL54 and to equivalent homolgues in other herpesviruses.

The functions of these proteins and their interactions may be summarised as follows. The UL9 product is an origin-binding protein (OBP) and the UL29 product (ICP8) a single-stranded DNA binding protein. These two proteins can interact via the C-terminus of UL9 (Boehmer and Lehman, 1993; Boehmer et al., 1994). The UL30 protein (POL) and UL42 proteins comprise the catalytic and accessory components, respectively, of a dimeric DNA polymerase (reviewed by Challberg, 1991; Weller, 1991) and interact via residues at or near the C-terminus of POL (Digard & Coen, 1990; Digard et al., 1993, 1995; Marsden at al., 1994; Stow et al., 1993; Tenney et al., 1993). The UL5, UL8 and UL52 proteins form a trimeric complex that exhibits both DNA helicase and DNA primase activities (Dodson et al., 1989; Crute et al., 1989). The UL5 protein is largely responsible for DNA helicase activity (Gorbalenya et al., 1989; Zhu & Weller, 1992), and the UL52 protein contributes an essential role in DNA priming (Klinedinst & Challberg, 1994; Dracheva et al., 1995) and these two proteins can form a stable subassembly that retains both functions (Calder & Stow, 1990; Dodson & Lehman, 1991; Crute et al., 1991). The UL8 component has auxiliary effects on the DNA primase activity, stimulating primer synthesis and/or utilization on a natural-sequence single-stranded DNA template (Sherman et al., 1992; Tenney et al., 1994), and is also required for efficient nuclear entry of the trimeric complex. (Calder et al., 1992; Marsden et al., 1996). UL8 is capable of binding separately to the UL5 and UL52 proteins and can also interact specifically with UL9 (McLean et al., 1994). The latter interaction with OBP may serve to recruit the helicase-primase into an initiation complex at the viral origins.

Further evidence for the occurrence of multiple interactions between DNA replication proteins has been provided by immunofluorescence experiments. In cells infected with HSV-1 in the presence of inhibitors of viral DNA synthesis UL29 (ICP8) localises to punctate structures within the nucleus termed "pre-replicative sites" (Quinlan et al., 1984). The requirement for each of the DNA replication proteins in the formation of these sites has been studied by the use of viral mutants with defects in individual replication proteins (Liptak et al., 1996; Lukonis et al., 1996). It was observed that proteins UL5, UL8, UL9 and UL52 are all necessary for the localisation of UL29 (ICP8) into pre-replicative sites and that mutants with defects in any of the other six DNA replication genes are affected in the ability of POL to localize to these sites. Although these data suggest that the DNA polymerase holoenzyme is the last component to be recruited (Liptak et al., 1996) they do not identify the specific interactions involved in this event.

It has now been found that the protein UL8 interacts with POL. Further, it has been found that disruption of the POL/UL8 interaction is possible. Examples of molecules, monoclonal antibodies and peptides that specifically disrupt the interaction have been identified.

The present invention provides an anti-viral agent capable of combatting replication of a herpesvirus by interfering with the association of UL8 and POL (as defined above).

Both the UL8 and POL proteins of HSV-1 have been previously described in the literature (e.g. Parry et al., 1993; Gottleib et al., 1990).

Furthermore the amino acid/DNA sequences of UL8 and POL from HSV-1 are available from publically accessible Genbank and EMBL databases under Nos. P10192/M19120 and P04293/M12356 (and several other entries), respectively.

The UL8/POL association is an association between two viral proteins, that are significantly different from any protein in the mammalian host organism (for HSV-1, the host is humans). Although homologues of POL are present in mammalian cells they are considerably diverged. No cellular homologue of UL8 is known. For the virus to overcome disruption of such a viral protein: viral protein interaction a double mutation, i.e. a mutation in each of the viral proteins involved, may be required. Alternatively the range of single mutations that overcome disruption, yet allow the two proteins to interact normally may be severely restricted. The probability of such reversion occurring is thus relatively low rendering this type of interaction attractive as a potential target for therapeutic agents. Additionally, as neither UL8 nor POL have close homologues in mammalian cell metabolism, the toxicity of an agent which specifically interacts with these proteins will be low.

The anti-viral agent may be a peptide or more preferably a non-peptidal compound having peptidomimetic properties. Such a non-peptidal compound will be preferred since it will be resistant to enzymic breakdown by peptidases. Suitable anti-viral compounds may include peptides having an amino acid sequence derived from the C-terminal or C-proximal region of UL8, a functional equivalent of such a peptide, or a peptidomimetic compound therefor.

The computer program "Predict-Protein" (EMBL-Heidelberg) makes a strong prediction of the presence of an alpha-helical region near the C-terminus of HSV-1 UL8 (amino acids 709-728) with the very C-terminus (residues 729-750) predicted to be in looped or extended structures (perhaps as a "tail"). The secondary structure predictions for the C-terminal regions of the UL8 homologues of bovine herpesvirus 1 (BHV-1), human cytomegalovirus (HCMV, betaherpesvirus) and Epstein-Barr virus (EBV, gammaherpesvirus) are all similar in that an alpha-helical region of approximately 20 amino acids is strongly predicted to occur within 10-26 amino acids of the C-terminus. The most inhibitory HSV-1 peptide we have identified (peptide 7, amino acids 719-738) is derived from across the junction of the predicted alpha-helix and "tail" portions at the C-terminus of UL8 and is 20 amino acids in length. We consider it likely that the predicted conserved structures in the C-terminal regions of the other herpesvirus UL8 homologues discussed above are similarly involved in interactions with the POL homologues and peptides representing similar regions might be able to disrupt the POL/UL8 interactions in these viruses. Thus the peptide is preferably derived from the free "tail" portion and/or the .alpha.-helix portion of the C-terminus of UL8. Optionally the peptide is as small as possible, eg less than 6 amino acids, but can be eg 10,14 or more amino acids in length, particularly where the peptide is derived wholly or partially from the .alpha.-helical region of the C-terminus of UL8.

The anti-viral agent is preferably effective against a herpesvirus selected from HSV, HCMV, Human herpesvirus 8 (HHV8), EBV and EHV 1 & 4. HCMV is of particular interest. The antiviral agent is preferably also effective against proteins homologous to UL8 and POL (eg UL102 and UL54 respectively for HCMV). Generally the anti-viral agent will be selected to mimic at least a portion at or near the C-terminus of the UL8 homologue of the specific target virus.

In a further aspect, the present invention provides an assay to determine the ability of a test substance to interfere with the association of UL8 and POL. The assay comprises the following steps:

i) providing a first viral component;

ii) exposing said first viral component to a test substance followed by a second viral component, or exposing said first viral component to a second viral component followed by a test substance;

iii) washing to remove any second viral component and/or test substance not associated with the first viral component; and

iv) detecting the presence, and optionally determining the amount, of second viral component associated with said first viral component.

The first or second viral components may be localised on a surface, such as a blotting membrane, or an assay plate for ELISA etc. Preferably the first component is immobilised in such a manner, although the invention contemplates the possibility of the assay being carried out in solution.

The first viral component may be POL or UL8. Where the first viral component is POL, the second viral component will be UL8. Where the first viral component is UL8, the second viral component will be POL.

If the assay is to test the ability of the test substance to interfer with UL54/UL102 association, the first viral component may be UL54 or UL102. Where the first viral component is UL54, the second viral component will be UL102. Where the first viral component is UL102, the second viral component will be UL54.

Detection of the presence and/or amount of second viral component associated with the first viral component may be conducted by any convenient means. Generally detection may be via a monoclonal antibody, the presence of which is established by exposure to a second labelled monoclonal antibody in a typical ELISA-style assay. Alternatively, the second viral component may be labelled (eg radioactively) to determine its binding to the first viral component.

Suitable monoclonal antibodies (Mabs) for use in the assay of the present invention have been produced (see Examples 2 and 3) and form a further aspect of the present invention. In particular the POL-specific Mab 13185 is suitable for use in the assay of the present invention where POL of HSV-1 is the second viral component. Mabs 804 and 805 are UL8-specific Mabs and are suitable for use in the present invention where UL8 of HSV-1 is the second viral component. Hybridoma cell-lines have been deposited for Mabs 13185 and 805 at the European collection of animal cell cultures at ECACC, Porton Down, Wiltshire on Jul. 26, 1996 under Accession Nos 96072640 and 96072639 respectively.

Identification of MAb 814 as an antibody that inhibits the POL/UL8 interaction and the mapping of its epitope to between amino acids 470 and 671 suggests that the C-terminus may not be the only region of UL8 to contribute to POL binding. For the POL/UL42 interaction the C-terminus was found to contribute 75% of the binding energy (Marsden et al 1994). The relative contribution of different regions of UL8 to POL binding remains to be determined.

By analogy with other DNA replication systems it is considered likely that initiation of HSV-1 DNA synthesis involves the formation of an initiation complex at one or more of the replication origins. The first stage in this process would be the binding of UL9 to its recognition sequence. The interaction of UL9 with UL8 might then serve to recruit the viral helicase-primase complex (UL5, UL8 and UL52) (McLean et al., 1994). In addition, ICP8 both interacts physically with UL9 and can stimulate its helicase activity (Boehmer & Lehman, 1993; Boehmer et al., 1994). These five proteins together therefore have the potential to open up the duplex DNA in the origin region and synthesize RNA primers. The interaction between POL and UL8 which we have now identified may play an important role in bringing the viral DNA polymerase (POL/UL42 heterodimer) into the complex to initiate DNA synthesis. In addition a direct physical interaction between the polymerase and helicase-primase complexes may be important in co-ordinating the unwinding of the duplex and the synthesis of RNA primers on the lagging strand at the advancing replication fork. This model, summarized in FIG. 10, is entirely compatible with that proposed by Liptak et al. (1996) in which UL5, UL8, UL9, UL29(ICP8) and UL52 are assembled at prereplicative sites followed by recruitment of POL, which is facilitated by UL42. Our finding provides the basis for the recruitment of the POL/UL42 complex. Amongst the many questions that remain to be answered is whether the affinities of the different proteins for each other is influenced by the presence of other proteins in the complex. It is possible, for example, that binding of POL to UL8 reduces the affinity of UL8 for UL9 allowing the helicase-primase-polymerase complex to migrate away from the origin to the replication forks.

The interaction of POL with UL8 may represent a possible new target for the action of an antiviral agent. A UL8 protein lacking the C-terminal 34 amino acids is unable to support viral DNA synthesis in a transient transfection assay indicating that this region of the UL8 protein performs an essential replicative function. Although this provides evidence consistent with a key role for the UL8/POL interaction, it should be noted that we cannot exclude the possibility that this region of the protein is also necessary for some other essential function.

Our identification of peptides that block this interaction should also encourage further studies of this region and the search for more potent inhibitors. In the case of the HSV ribonucleotide reducase, following the initial discovery that peptides corresponding to the C-terminus of the small subunit inhibited enzyme activity (Cohen et al., 1986; Dutia et al., 1986), it proved possible to identify more active peptidomimetic compounds that could function intracellularly (Luizzi et al., 1994; Moss et al., 1995). The POL/UL8 interaction may be an especially attractive new target for two reasons. First, both proteins are present in infected cells in low amount in contrast to POL/UL42 and R1/R2 where one or both of the interacting proteins are abundant viral products. Second, the POL/UL8 interaction appears to be relatively weak as suggested by the observation that in contrast to POL/UL42 and R1/R2 they do not co-purify from infected cells and also by the ability of peptide 7 to block the interaction equally effectively when pre-incubation with POL was omitted. Such a weak interaction may be more readily blocked by an antiviral compound than a strong interaction.

Mabs 817, 818 and 819 all recognised peptide 5, that corresponds to residues 722 to 750 of UL8, and to a lesser extent peptide 3 (amino acids 726-750). However the Mabs do not recognise peptide 2 (amino acids 728-750) or peptide 7 (amino acids 719-738). It is therefore probable that all three MAbs recognize the same epitope located within the C-terminal 29 amino acids of UL8 and minimally involving the region spanning amino acids 727-739.

The present invention also provides a method of combatting replication of a herpesvirus, said method comprising providing an agent capable of disrupting the association UL8 and POL.

Further, the present invention provides a method of combatting an infection caused by a herpesvirus, said method comprising administering an antiviral agent as described above.

Additionally the present invention provides the use of an agent capable of interfering with association of POL/UL8 for combatting herpesvirus replication or infection.

Claim 1 of 3 Claims

What is claimed is:

1. An antiviral agent which prevents or hinders replication of a herpesvirus in vitro by specifically binding to POL or UL8, thus inhibiting the association between UL8 and POL, wherein "UL8" is defined as UL8 of HSV-1 or the homologues thereof in other herpesviruses and "POL" is defined as POL of HSV-1 or homologues thereof in other herpesviruses, wherein said agent is a peptide selected from the group of peptides:

a) VFTGVLAGVWGEGGKFVYPFDDKMSFLFA (SEQ ID NO: 5);

b) IELVFTGVLAGVWGEGGKFV (SEQ ID NO: 7);

c) DEWVRSLAVDAQHASKRVASEGLRFFRLNA (SEQ ID NO: 11) and;

TWLEERDEWVRSLAVDAQHAARRVAS (SEQ ID NO: 12).

____________________________________________
If you want to learn more about this patent, please go directly to the U.S. Patent and Trademark Office Web site to access the full patent.