JP2002523053A - Targeted alphaviruses and alphavirus vectors - Google Patents

Abstract

  (57) [Summary] The present invention provides a recombinant alphavirus comprising at least one epiviral non-alphavirus sequence. This sequence is (i) present in the E1, E2 or E3 protein either as an insert or as a replacement of one or more residues. This sequence then (ii) binds directly and selectively to a region on the target cell-specific surface receptor. The present invention also provides nucleic acids associated with the alphavirus, and methods of treating cancer in mammals using the alphavirus and related nucleic acids.

Description

DETAILED DESCRIPTION OF THE INVENTION

(Government Assistance) The present invention provides, in part, a Breast Cancer Research Prog.
This was done with funding from the Department of Defense according to ram Idea Award No. BC9617725.

This application claims priority from provisional application number 60 / 095,138, filed July 30, 1998, under 35 USC 119. The contents of this provisional application are incorporated herein by reference.

TECHNICAL FIELD [0003] The present invention relates to a recombinant alphavirus comprising at least one non-alphavirus sequence, wherein the non-alphavirus sequence results in this vector targeting to a particular target cell. The present invention also relates to a method of treating cancer using the targeted vector of the present invention.

Background Art Alphaviruses are members of the Togaviridae family. Alphaviruses are + strand RNA viruses that are replicated and packaged in the cytoplasm of infected cells. There are many known strains of human tropics. These include: eastern equine encephalomyelitis virus, western equine encephalomyelitis virus, Venezuelan equine encephalomyelitis virus, Sindbis virus (including mutant Ockelbo virus and Babanki virus), chikungunya virus, Onion virus, Igbo Ora virus, Ross River virus, Mayalovirus, and Barmah forest virus. Examples of alphaviruses that are usually trophic for non-human animals include the following: eastern equine encephalomyelitis virus, western equine encephalomyelitis virus, Venezuelan equine encephalomyelitis virus, and getavirus. Another example of an alphavirus is Interna
Tional Committee of Taxonomy of Viru
ses, "The Classification and Nome"
nclature of Viruses "in that publication. This sixth report was published in 1995. Other examples are available on the Internet at http: // www. ncbi. nih. gov / htbin-po
st / Taxonomy / wgctorg? id = 11019 & 1v1 = 3.

[0005] Many viruses that make up the genus Alphavirus are closely related in molecular characteristics and structure. A common antigenic determinant is that immunodiagnostic techniques (eg, fluorescent antibodies, EL
(ISA, RIA and complement fixation).

[0006] Sindbis virus is a representative prototype alphavirus and has been extensively studied. Sindbis virus is distributed over vast areas of Europe, Africa, Asia and Australia. The virus then infects humans but usually does not have a recognized clinical disease. In certain defined parts of the world, Sindbis virus is an important cause of fever, arthritis and rash. The symptoms of infection with Sindbis virus are minor and transient and usually do not require medical attention. Assuming that the Sindbis virus used in the context of the present invention is targeted, this virus is attenuated and therefore unlikely to cause disease.

[0007] Sindbis virus comprises a nucleocapsid enveloped in a lipoprotein envelope with T = 4 icosahedral symmetry. Sindbis is a single-stranded + chain RN of + polarity associated with 240 copies of the 29 kd capsid protein (C).
Contains the A genome (11,703 bases). The lipid bilayer of this lipoprotein envelope is derived from the host cell plasma membrane during budding. The general structure of this genome is shown in FIG.

[0008] The complete genomic sequence of Sindbis virus is described in Status, E .; G. FIG. Vir
<RTIgt; org. </ RTI> (1984) 133: 92-110, which is hereby incorporated by reference. As shown in FIG. 1, the genome is capped at the 5 ′ end; about two-thirds of the genome encodes a nonstructural protein, and copies the + strand to the − complement. Is enough to bring. 3 'of this genome
One-third contains a subgenomic promoter that controls the transcription of the portion encoding the structural gene, including the capsid and envelope proteins described above. Immediately upstream of this promoter is the viral binding region required for replication. 3 '
There is a poly A chain at the end. The portion of the -strand corresponding to the region encoding this structural protein is transcribed and capped as a subgenomic + strand RNA, and
The genomic RNA is then packaged and translated into structural proteins that complete the virions.

[0009] Alphaviruses, especially Sindbis, have been used as vectors for the expression of heterologous genes for many years. U.S. Pat. No. 5,091,309 relates to a Sindbis virus D1 RNA vector, in particular a vector comprising nucleotides 1-241 and 1928-2314 of Sindbis virus D1 RNA, between nucleotides 1-241 and 1928-2314. Alternatively, an RNA sequence complementary to a restriction endonuclease site for further insertion of a heterologous sequence can be inserted. This vector also contains poly A adjacent to nucleotide 2314. These vectors have been described as being useful for expressing heterologous sequences in cells. Similarly, U.S. Patent No. 5,217,879 relates to infectious Sindbis viruses, including: (i) heterologous, upstream from the structural coding region of the virus and at least 5 nucleotides downstream from the start of the subgenomic RNA of the virus. Code sequence, and (
ii) a viral binding region located upstream from the viral structural coding region. This vector is described as being useful for expressing a heterologous sequence in a cell. Schles
inger, TIBTECH (January 1993) 11: 18-22 disclose recombinant Sindbis viruses expressing heterologous genes, as well as alphavirus vectors expressing heterologous epitopes that can function as immunogens.

[0010] The use of alphaviruses as carriers for heterologous polypeptides by incorporation of heterologous sequences into permissive sites in the virion envelope protein,
Bredenbeck et al., Seminars in Virology (199
2) 3: 297-310. Mutagenesis by random insertion was used to identify insertion sites in the envelope proteins of Sindbis virus (eg, E1 and E2) that allow recovery of chimeric viruses with growth characteristics similar to the parent Sindbis virus. . Such chimeric viruses can be used to develop live attenuated virus vaccines, and the insertion of a peptide ligand or larger binding domain into the virus surface protein will ultimately result in recombination into a particular cell type It may allow for delivery of the RNA expression construct.

[0011] Numerous changes in E1 and E2 affecting Sindbis neurotoxicity are:
Disclosed (Lustig et al., J Virology 62 (7): 232).
9-2336 (1998)). More specifically, it changes the neurotoxicity of Sindbis,
A single amino acid change at position 172 of the E2 glycoprotein has been disclosed (Tucke).
r et al., J Virology 65: 1551-1557 (1991). Neutralizing epitopes derived from the external G2 glycoprotein of Rift Valley fever virus
Infectious Sindbis virus containing glycoprotein and E3 glycoprotein is Lo
Ndon et al., Proc Natl Acad Sci USA (1992) 89
: 207-211. Such chimeric viruses have been described as potentially useful for developing live attenuated virus vaccines. Vaccines based on alphaviruses are also described in PCT application WO 99/11808, and alphavirus vectors with reduced cytotoxic effect are described in PCT application WO 99/18226. Specific expression of alphavirus vectors in bone marrow cells is described in WO 98/36779 and US Pat.
No. 407.

[0012] Frolov, et al., Proc Natl Acad Sci USA (199)
6) 93: 11371-11377 provides an overview of the use of alphaviruses as recombinant vectors at that time. This article, incorporated herein by reference, provides a useful description of the alphavirus itself and its life cycle. It has also been suggested in the context of gene therapy that targeting alphaviruses to specific cell types is useful.

PCT publications WO 95/07994 and WO 96/17072 describe various Sindbis expression vectors. Some of these have been targeted to desired cell types. The highlighted approach to targeting involves tissue-specific expression, which depends on the presence in the targeted cell of RNA sequences that interact with the vector to effect expression of the heterologous gene. It has also been suggested that the selectivity of the vector can be manipulated by making chimeric envelope proteins from various strains of the alphavirus. Upon passage, human IL-2 and the envelope protein are combined to target the vector to cells bearing the IL-2 receptor, or to include the neutralizing domain of gp120 from HIV to provide normal E2 tropism. It has also been briefly described that it may also be possible to disrupt and provide CD4 cell targeting. Further details are given, but only on the construction of hybrid envelopes from various strains of alphavirus.

More recently, attempts to target an alphavirus-based vector to selected cells by providing a gene encoding a chimeric envelope protein comprising a portion of the IgG binding domain of Protein A Has been made. This allows the virus to bind to cells bound by IgG specific for the surface molecule. As described in PCT Publication WO 98/44132, this system provides for the linking of the virus to target cells selected through an intermediate IgG specific for its surface molecule characteristic of its target (linkag).
e) is required. There is no direct binding of the virus to the cell surface. This work has been previously described by Ohno, K. et al. Et al., Nature Biotech (
1997) 15: 763-767 by the same group.

[0015] Vectors other than alphavirus vectors have been modified to target them to specific cellular targets in the context of gene therapy. Mille
r et al., FASEB J (1995) 9: 190-199. Vectors, including retroviral vectors, adenoviral vectors, liposome vectors, and molecular conjugates, are described as being useful as well as transcription targeting in the context of cell surface targeted gene therapy. Retroviruses displaying single chain antibodies to cell surface proteins expressed on various human carcinoma cell lines have been disclosed (Chu et al.,
J Virol 69 (4): 2659-2663 (1995)). The avian retrovirus envelope protein has been modified to include an RGD-containing peptide that binds to integrins (Valsesia-Wittmann et al., J. Am.
Virol 68 (7): 4609-4619 (1994)). Similar modifications have been suggested for other retroviruses (eg, murine leukemia virus). Moloney murine leukemia virus (MoMLV) has been modified to include an erythropoietin sequence as an insert into the viral envelope, thereby allowing targeting to the erythropoietin receptor (Kasahara et al.,
Science 266: 1373-1376 (1994)). Adenovirus fibers have been modified to include ligands specific for cell surface receptors (US Pat. No. 5,543,328). MoMLV containing a fusion protein of an antibody fragment and the viral envelope Pr80 ^env has been disclosed for targeted binding to mouse fibroblasts (Russell et al., Nucleic Acids).
Research (1993) 21 (5): 1081-1085). MoMLV vectors targeted to human low-density lipoproteins via chimeric envelope proteins containing single-chain variable fragments derived from mAbs have also been disclosed (Somia et al., Proc Natl Acad Sci USA (1995)).
92: 7570-7574). Han et al., Proc Natl Acad Sc.
i USA (1995) 92: 9747-9751 discloses the insertion of a sequence encoding human heregulin into the MoMLV envelope to target the virus to human breast cancer cells. Adenoviruses expressing single-chain antibodies have also been described (WO 94/10323). In addition to modifying adenovirus, WO
94/10323 discloses the modification of influenza and vaccinia viruses and virus-like particles so that they bind to target cells. The modifications listed include mAbs, ScFvs, dAbs and minimal recognition units.

[0016] The above vectors have disadvantages. For example, retroviral vectors are disadvantageous due to low transduction and infection efficiency. In addition, retroviral vectors carry the risk of insertional mutagenesis and the generation of replication-competent virus upon integration of the proviral DNA into the host genome. In addition, retroviral vectors do not infect non-proliferating target cells. Because they cannot integrate their proviral DNA into the genome of non-proliferating cells. Targeted retroviral vectors also require the presence of wild-type and modified envelope glycoproteins in mature virions to establish infection. Adenovirus vectors are produced at high titers and can transfer genes efficiently to replicating and non-replicating cells, but adenoviral vectors are disadvantageous. This is because, due to the complexity of adenovirus vectors, adenovirus vectors are difficult to modify in the receptor binding domain. Furthermore, infection with adenovirus vectors results in non-specific inflammation and anti-vector cell immunity, and as a result, only transient gene expression.

As described further below, the use of alphavirus as a vector specifically targeted to the cells to be treated has advantages over previously constructed targeted vectors. In a preferred embodiment, the targeted cell is a cancer cell, and the alphavirus produces a negative effect either by the infection itself or by an additional heterologous sequence. The vectors of the present invention offer advantages over those available in the prior art. In vitro, alphaviruses facilitate the construction of recombinant alphaviruses due to their simple genomic structure, and are highly viable from RNA transfected cells without the need for cumbersome amplification or selection procedures. Enables the production of multivalent recombinant virus stocks.
In vivo, the alphavirus has no risk of insertional mutagenesis, the ability to infect proliferating and quiescent cells, does not require wild-type envelope glycoproteins for the production of infectious mature virions, and It offers the advantage of infection of the targeted cells equivalent to or better than infection by (ie, unmodified) alphavirus. In addition, high levels of heterologous gene expression can be achieved. This is because viral RNA replication and translation signals are used. In addition, production of infectious RNA transcripts does not involve DNA intermediates and replication is in the cytoplasm, thus it does not involve the integration of the viral genome into the host genome. In addition, alphaviruses have the ability to kill targeted cells by apoptosis or by eliciting a host immune response against viral structural proteins expressed on the surface of targeted cancer cells. The immune response to cells infected by the alphavirus assists in eliminating cells killed by apoptosis.

DISCLOSURE OF THE INVENTION The present invention provides a recombinant alphavirus comprising an epiviral non-alphavirus sequence designed to target the virus to a particular cell type. Preferably, the epiviral non-alphavirus sequence comprises an E1 glycoprotein, E2
Present as an insert or a substitution of one or more amino acid residues in a glycoprotein and / or E3 glycoprotein. In a preferred embodiment, the epiviral sequence is directed to a receptor on the cancer cell.

The at least one epiviral non-alphavirus sequence can be, for example, a single chain antibody (scAb) to a particular cell surface receptor or a ligand to a particular cell surface receptor, or a binding domain to a particular cell surface receptor. . A binding domain for a specific cell surface receptor may include a single epitope or more than one epitope. Additional epiviral non-alphavirus sequences, which enhance the vector's ability to target the desired cells, may also be included. Thus, additional amino acid sequences, which also bind to receptors characteristic of the desired cell, can be included. Preferred target cells are
It is a cancer cell or, more usually, a cell characteristic of cancer growth. Cells characteristic of cancer growth can be the cancer cells themselves or cells that support metastasis or cancer growth (eg, neovascular cells required for cancer to survive). Preferably, the epiviral non-alphavirus sequence binds to erbB-2. More preferably, the epiviral non-alphavirus sequence binds to erbB-2. At this time, it heterodimers with erbB-3 or erbB-4. A preferred sequence that binds to heterodimerized erbB-2 is the EGF domain of α-heregulin.

[0020] The recombinant alphavirus may further contain and express a heterologous sequence, preferably under the control of a viral subgenomic promoter. Heterologous sequences can be full length or partial coding sequences for proteins or factors that directly or indirectly induce cell death. For example, the heterologous sequence is the full-length or partial coding sequence of Bak or Bax, and encodes a ribosome that binds, cleaves, and destroys Bcl-2 mRNA. Alternatively, the heterologous sequence induces cell death in the presence of a cytotoxic drug (eg, the heterologous sequence encodes a thymidine kinase and the cytotoxic drug is ganciclovir). Depending on the intended use, the heterologous sequence may encode any desired protein, including markers for infected cells.

[0021] Transcription of the genome of the recombinant alphavirus can be under the control of a non-alphavirus promoter. In that case, the promoter is preferably a cancer-specific promoter or an inducible promoter.

More specifically, the present invention is a recombinant Sindbis virus as described above.

The present invention further provides a recombinant DNA vector into which the RNA genome of the above-mentioned recombinant alphavirus can be transcribed. A substantially pure RNA genome of the above-described recombinant alphavirus is also provided.

[0024] In addition to the above, the present invention provides a method of treating cancer in a mammal. If desired, the cancer to be treated is of epithelial cell origin (eg, breast or ovarian cancer). One method involves administering to the mammal a cancer therapeutically effective amount of a recombinant alphavirus as described above. I do. Recombinant alphaviruses treat cancer by binding to cancer cells and subsequently infecting and lysing. Another way is
Administering a therapeutically effective amount of the above-described recombinant vector or RNA genome to a mammal. The recombinant vector or RNA genome treats cancer by entering cancer cells. Here, it produces an RNA genome, which is then packaged in a viral particle, which lyses the cancer cell and selectively binds and infects another cancer cell, And dissolve.

A method for inserting a heterologous sequence into the E2 glycoprotein coding region of the Sindbis virus genome is also provided by the present invention. Similarly, methods are provided for replacing a portion of the E2 glycoprotein coding region of the Sindbis virus genome with a heterologous sequence.

A method for detecting a cancer cell is also provided. The method comprises the steps of contacting a cell with a recombinant alphavirus, wherein the alphavirus is desirably non-cytolytic and replication-deficient, and comprises a marker gene (eg, including a green fluorescent protein) and its marker gene. Detecting the expression of the marker gene, wherein detecting the expression of the marker gene indicates the presence of a cancer cell.

Another method provided by the present invention is a method for determining new cancer-specific cell surface molecules or receptors. This method combines sequences from a library of binding domains or random sequences with the recombinant alphavirus glycoprotein E1.
Introducing one or more amino acid residues of E2 or E3 as an insertion or substitution, contacting a cancer cell with the recombinant alphavirus, selecting a recombinant alphavirus that binds to the cancer cell, and Identifying a cancer-specific cell surface molecule or receptor bound by the recombinant alphavirus using the recombinant alphavirus that binds to the cancer cells.

DETAILED DESCRIPTION OF THE INVENTION Background to Alphaviruses Alphavirus infections in humans can range from asymptomatic seroconversion to disease, but most alphaviruses have clinical Does not result in disease or results in transient, transient, debilitating fever (eg, Semliki Forest virus and Venezuelan equine encephalomyelitis virus (VEE))
). Some alphaviruses invade the central nervous system and cause encephalitis in a small percentage of infections (eg, eastern equine encephalomyelitis (enc)
epalitis) virus (EEE), western equine encephalomyelitis (encepha)
litis) virus (WEE) and Highland J
Virus). Inactivated whole virus vaccines are available in EEE, WEE and VE
E is available as a new investigational drug for infection in humans. In this regard, live attenuated VEE vaccines have been used in researchers without adverse effects and have been used in Texas to prevent the epidemic of VEE in epidemics. Other alphaviruses can cause acute arthropathy with permanent joint sequelae (eg, Chikunganya, Mayaro, Onion, Ig
bo Ora, Ross River, Sindbis and Barmah forests); however, live attenuated vaccines have been developed, and the use of the virus according to the invention has been targeted, and thus, as an attenuated vaccine Has been attenuated. Yet another alphavirus (eg, aura virus, beval virus,
(Beggy Creek, Cabassou, Fort Morgan, Getavirus, Kyzylagach, Middelburg virus, Nudum virus, Pixnavirus, Sagiyama virus, Unavirus and Watallowa virus) are not known to cause recognized human diseases. . In view of the above, it is desirable to use an alphavirus that is either associated with a disease or with a non-fatal sudden death disease. Sindbis virus is preferred.

[0029] The alphavirus of the invention, in a preferred embodiment, comprises an epiviral non-alphavirus amino acid sequence responsible for targeting the desired cells embedded in the envelope. This target sequence is usually found there and is provided as a fusion protein with an envelope protein that is responsible for the tropism of the virus in its natural state. Therefore, an explanation of the nature of this envelope is relevant.

[0030] Two viral glycoproteins, E1 and E2, are embedded in the lipoprotein envelope. E1 and E2 partition as 80 trimers on the outer surface of the viral envelope. Each trimer has three E1-
Consists of an E2 heterodimer, which protrudes from the membrane in a triangular shape. At this protrusion, the three heterodimers wrap around each other to form the axis of the protrusion, and then form a three-part head. A one-to-one interaction exists between the glycoprotein heterodimer and the capsid protein subunit.
The C-terminal region of E2 is responsible for binding to its capsid. Through crosslinking research,
It has been suggested that these trimers are maintained by non-covalent E1-E1 interactions. Upon binding of the viral envelope to the cell membrane during infection,
E1-E2 heterodimers disassociate and form E2 and E1 homotrimers. E1 has been demonstrated to play a role in the fusion of the cell membrane with the viral envelope.

[0031] E1 and E2 each have a molecular weight of about 50 kilodaltons and are anchored to the viral envelope by transmembrane anchors in the C-terminal region.
Sindbis virus E2 is 432 amino acids long and has a cytoplasmic domain of 33 residues. Sindbis virus E1 is 439 amino acids long and has a two residue cytoplasmic domain. Both E1 and E2 are glycosylated, where E2 has two or three sugar chains and E1 has two sugar chains. The number and position of this sugar chain is not absolutely conserved among alphaviruses. E of Sindbis virus and Semliki forest virus
1 and E2 are known to have palmitic acid residues attached at or near their transmembrane anchors. The major neutralizing epitope present in the region of E2 amino acids 170-220 is sequence-variable among Sindbis virus strains and Ch
poorly constructed as determined by computer analysis of the putative secondary structure of ou-Fasman and Robson-Garnier, highly charged (i.e., hydrophilic), and containing glycosylation sites (this Suggests that this epitope is epiviral on the surface of the virion).

The viral structural protein is derived from the 26S subgenomic mRNA by using the polyprotein H _Two Translated as NC-E3-E2-6K-E1-COOH. Capsid (C
) Is a polypeptide chain having serine protease activity and from which it is synthesized
From itself. E2 is derived from the precursor PE2, which is
Cleavage into E2 and small glycoprotein E3 in the lansgolgi network
Is done. E3 does not maintain an association with mature Sindbis virions, but
Maintain meetings with Ki Forest Virions. 6K is a small hydrophobic peptide,
It serves as a linker between E1 and E2 in the polyprotein,
It has been found to associate with mature virions in sub-concentration amounts. The capsid
After the protein has been cleaved from the polypeptide chain, the N-terminal signal sequence is
Into the endoplasmic reticulum, after which the polyprotein
It is cleaved into individual proteins by nalase. Sugar chain is the protein
It is added when passing through the secretory pathway. PE2 is a trans-Golgi network
Is cleaved to E2 by a calcium-dependent serine protease.

[0033] The envelope glycoprotein has domains involved in neutralization, blood aggregation, early events in cell infection and virulence. Three neutralization sites have been identified in E2 and one neutralization site has been identified in E1. E2
One of the neutralization sites (ie, Ec) in amino acids 62, 96 and 152
Is a conformational epitope in The remaining two neutralization sites at E2 (ie, Ea and Eb) are at amino acids 196-216;
Together, they make up the major neutralizing domain of Sindbis. Similar regions exist in other alphaviruses. For example, the major neutralizing domain in Venezuelan equine encephalomyelitis virus is located at amino acids 180-210 at E2 and has a sequence just at the N-terminus, while the major neutralizing domain in Ross River virus is E2 amino acid 216 , 234 and 246-251. E2a and E2b are physically very close. This is because these two are independently mutable, but compete strongly for E2a and E2b mAbs in ELISA. Traditional E1 and E2 epitopes become epiviral during the early stages of infection due to conformational changes. These and other potential epitopes become epiviral during treatment at high temperatures, reducing agents or low pH. Some of these changes may involve the reduction of key disulfide bonds in glycoproteins,
For example, disassociation can disrupt the association between proteins.

[0034] Antibodies directed against the E2 protein primarily neutralize viral infectivity. Therefore, the E2 protein is thought to be important for binding the virus on the cell surface. Ionic interactions between Sindbis virus and the host cell surface are also important for attachment and initiation of infection.

As noted above, genomic RNA acts as mRNA for translation of polyproteins, which are cleaved at or after translation to produce the nonstructural proteins nsP1, nsP2, nsP3 and nsP4. give. Genomic RNA also acts as a template for the synthesis of the complementary negative strand, which
It then provides a template for the synthesis of plus-strand genomic RNA and a transcript of the 26S subgenomic RNA, which is translated into the capsid, PE2,6K and E1 proteins.

Construction of the Alphavirus of the Invention The recombinant alphavirus may be replication deficient, but is preferably replication compatible and contains epiviral non-alphavirus sequences. The epiviral non-alphavirus sequence is preferably present as an insertion or substitution of one or more amino acid residues in the E1 and / or E2 glycoprotein or the E2 and / or E3 glycoprotein. It then binds directly and selectively to at least one region of the target cell-specific surface receptor. "Non-alphavirus (sex)" means that the sequence is not naturally present in any alphavirus. "Epivirus (sex)" refers to a sequence whose sequence is any part of the viral envelope or part of the viral protein coating (except for the internal surfaces of the viral envelope or virion protein coating, El, E2 and possibly E3 ( )) When E3 is part of a mature virion. For example, E1 glycoprotein,
When present on the E2 or E3 glycoprotein, or in the region that penetrates E2 and E3, the epiviral sequence is either present on the outer surface of the viral protein coat, or (first after the establishment of the viral cell complex). As a result of the conformational changes that occur during the early stages of infection (as indicated by the transient epitope exposure of E1 and E2 when virions infect the plasma membrane within 30 minutes of The outer surface of the

[0037] Epiviral sequences are designed to target the virus to a selected population of cells that have characteristic cell surface receptors. "Cell surface receptor" means any molecule present on the cell surface that can be recognized by any type of peptide. Thus, a “receptor” may or may not have any transduction capacity when bound to a suitable ligand or other binding agent. What is required for a "receptor" is that the receptor have sufficient amounts on its desired target cell to successfully interact with the epiviral sequence in the recombinant alphavirus to mediate infection by the virus. And the host cell desired in the environment in which the host cell is targeted such that its receptor causes selective infection of the desired target cell as compared to the cell in the environment being treated. It is all about being sufficiently distinctive and selective.

There may be various reasons for selecting a particular cell population to be targeted. For example, selective gene therapy that is appropriate for a particular type of differentiated or undifferentiated cell requires such selectivity. A particularly preferred embodiment discussed below is the targeting of cancer cells or other unwanted cells whose destruction is the ultimate goal of infection. However, vectors are not limited to this particular purpose. The nature of the cell targeted depends on the purpose of the infection.

In general, the cell surface receptor characteristics of a target cell can be recognized by a single-chain antibody directed to antigen-specific interaction with the receptor. However, a particular receptor, which may normally be present on a particular type of cell, may itself have a ligand specific for the receptor which may then be used as an epiviral amino acid sequence. If the ligand is non-peptidic, a peptide that mimics the conformation and change pattern of the actual ligand can be used.

The present invention is illustrated herein using representative examples of Sindbis virus prototypes of the Alphaviridae family. Suitable positions for insertion or replacement of a sequence by an epiviral sequence may be described with respect to the position in the Sindbis virus genome and the protein it encodes. Other members of the Alphaviridae family have permissible positions for insertions or substitutions similar to those of Sindbis when substituted for Sindbis virus. Thus, the Sindbis virus-specific positions described in the following paragraphs and elsewhere of this application can be predicted by selecting corresponding positions in these strains for other Alphaviridae members.

In general, epiviral non-alphavirus sequences are present in any region of the E2 glycoprotein. More preferably, the epiviral non-alphavirus sequence is in the region of about 60 to about 243 E2 amino acids of Sindbis. Most preferably, the epiviral non-alphavirus sequence is present in a region of about 60 to about 114, about 114 to 175, about 60 to about 175, or about 175 to about 243 E2 amino acids. Alternatively, the epiviral non-alphavirus sequence is
It is present in the N-terminal half of E2, preferably E2 amino acids 1 to 17
2, most preferably from E2 amino acid 62 to about E2 amino acid 172. Another possible site for insertion or replacement by a non-alphavirus sequence (so that the non-alphavirus sequence becomes epiviral in the E2 glycoprotein) is an anti-E2 mAb that counteracts viral infection. It can be mapped by generating idiotype antibodies. In this regard, the replaced envelope glycoprotein domain is preferably poorly conserved between certain alphavirus strains and poorly constructed to avoid alteration or loss of important secondary structure. And is easily accessible (ie hydrophilic)
.

When the non-alphavirus sequence is present as an insert, the insert contains about 3 to about 423 amino acids, more preferably about 6 to about 200 amino acids, and most preferably about 9 to about 68 amino acids. . If the non-alphavirus sequence is present as a replacement, preferably the replacement comprises about 3 to about 423 amino acids, more preferably about 6 to about 200 amino acids, and most preferably about 9 to about 68 amino acids. .

Such a recombinant alphavirus may further comprise at least one further epiviral non-alphaviral sequence as described above, which together with the epiviral alphaviral sequence as described above Or separately, from about 60 to about 2 of the E2 amino acid residues of Sindbis in the E2 glycoprotein.
43 or about one or more of about E2 amino acid residues 60 to about E2 amino acid residues 243, or about E2 amino acid residues 1 to about E2 amino acid residues 172 of the E2 glycoprotein. Exists as a substitution of the amino acid residue of

Accordingly, the present invention also provides a method for inserting a heterologous sequence into the E2 glycoprotein coding region of the alphavirus genome. This method involves introducing a restriction enzyme site at a nucleotide site corresponding to E2 amino acid residue 60, 114 or 175 of a Sindbis virus into a vector that can be transcribed into the RNA genome of an alphavirus, and Inserting a heterologous sequence into the DNA. The present invention also provides a method for replacing a part of the E2 glycoprotein coding region of the Sindbis virus genome with a heterologous sequence. This method provides for the Sindbis E2 amino acid residue 60, 114 or 1 into a vector that can be transcribed into the RNA genome.
And introducing a restriction enzyme site at the nucleotide site corresponding to 75. This restriction enzyme site is then combined with either another restriction enzyme site at one of the two remaining amino acid residues or a BclI restriction enzyme site naturally occurring at a nucleotide site corresponding to E2 amino acid residue 243 of Sindbis. E2 amino acids 60-114, 60-175, 114-175, 11
The nucleotide corresponding to 4-243 or 175-243 is deleted. The deleted nucleotide is then replaced with a nucleic acid sequence encoding a heterologous sequence.

A given site in a viral structural protein is permissive for an insertion or substitution (ie, a substitution or insertion at that site may result in structural protein synthesis, virus production, and production in a cancer cell line or non-cancerous cell line). ), Which may result in the ability to establish the perfection of a subject, can be determined according to any one of a number of known methods in the art. For example, anti-Sindbis virus antibodies can be used in immunoprecipitations or Western blots to investigate viral protein synthesis in cells, and to examine the presence of viral proteins in supernatant fluid or purified virions. Similarly, the presence of one or more non-alphavirus sequences can also be determined according to methods known in the art. For example, antibodies to the non-alphavirus sequence can be used to investigate the presence of the non-alphavirus sequence in virions purified by immunoprecipitation or Western blot.
Alternatively, the recombinant alphavirus may be assayed for its ability to kill cancer-derived or non-cancer-derived cell lines.

[0046] Whether a given recombinant alphavirus has sufficient activity for use in the present invention can be determined according to methods known in the art. For example,
The degree of cell killing of the cancer cell line after infection with the targeted alphavirus can be determined by trypan blue exclusion or propidium iodide assay,
It can then be measured by FACS analysis. Alternatively, the ability of the recombinant alphavirus to kill cancer cells can be measured by using an animal model, such as a nude mouse model. Preferably, the recombinant alphavirus is about 20%
It kills up to about 50% of cancer cells, more preferably about 50% to about 80%, most preferably about 80% to about 100%. However, if the recombinant alphavirus spreads throughout the tumor without causing tumor cell death,
It can be determined that it has sufficient activity. The expression of viral structural proteins in tumor cells acts as immunotherapy and results in the destruction of tumor cells by the host's immune system.

An alteration in the putative receptor binding domain of an envelope glycoprotein (eg, E1, E2, or E3) prevents misfolding, altering processing, or altering processing to an extent that prevents the formation of virus particles from infected cells. Care must be taken to ensure that the transport of glycoproteins to the cell surface does not fail (eg, Doms et al., Virology (199).
3) 193: 545-562). In this regard, the E2 glycoprotein has two N-linked glycoprotein sites, one of which is amino acid residue 1
At 96 is located within the neutralizing epitope E2ab. Substitution of this region, such as by a ligand or scAb, can result in misfolding of the glycoprotein. This is because sugar chains are required for proper folding. The neutralizing epitope E2c includes a region that has been found to be important for virus penetration. Replacement of this region with non-alphavirus sequences can interfere with this function. The alphavirus fusion function resides in the E1 glycoprotein. In this regard, E1
Glycoproteins and E2 glycoproteins heterodimerize, and thus their proper folding, and therefore their function, depend on each other.

Care was taken to ensure that the modification of the receptor binding domain by insertion could easily be removed from the virus by RNA recombination, thereby not producing a wild-type alphavirus that is no longer targeted. There must be. However, the only danger associated with the production of most wild-type alphaviruses is
An asymptomatic infection that leads to polyarthritis and rash, which alleviates and resolves when the virus is swept away by the immune system as described above.

If no virus is produced after RNA transfection of the cells, pulse-chase analysis and immunoprecipitation of metabolically labeled infected cell lysates is performed,
It can be determined if the modified glycoprotein is not misfolded and transported to the cell surface. For example, if the glycoprotein is misfolded, it will be shorter,
Less important parts of the glycoprotein should be replaced by non-alphavirus sequences, shorter non-alphavirus sequences should be used, or non-alphavirus sequences should be inserted instead of replacements Should be introduced into the alphavirus as a product.

In general, chimeric glycoproteins are synthesized using specific oligonucleotide primers that include selected non-alphavirus sequences (“inserts”) and homologous alphavirus sequences that include unique restriction endonuclease sites. It can be produced in recombinant alphaviruses by amplifying the insert for insertion into all or part of the alphavirus structural protein or for their replacement. Alternatively, a restricted Bal-31 digestion from convenient restriction enzyme sites is performed,
The digestion may be re-digested to an acceptable insertion site, followed by blunt end ligation of the insert.

[0051] The recombinant alphavirus may further comprise at least another epiviral non-alphavirus sequence. The binding of additional sequences may improve the targeting specificity of the recombinant alphavirus. This at least another episomal non-viral sequence is also preferably, together with or separately from the episomal non-viral sequence described above, as an insertion or substitution of one or more amino acid residues, preferably E1, E2, or Present in the E3 glycoprotein. And this can be a scAb, ligand, or binding domain that binds directly and selectively to a target-specific cell surface receptor, or is an immunogen. However, if the at least another episomal non-viral sequence binds to the same receptor as the first episomal non-alpha virus peptide, then this is preferably the region bound by the first episomal non-alpha virus peptide. Bind to a different region. For example, a bi-targeted virus can be made with the non-alphavirus sequences of E2 and E3, or the non-alphavirus sequence is a cleavage site between E2 and E3 that is not part of the mature virion May be replaced. In some cases, incorporation of PE2 into mature virions results in a non-infectious virus. This “lethality” can be suppressed by mutation at the second site of E3 or E2.

If the additional epiviral non-alphavirus sequence is an immunogen, the immunogen may elicit an immune response against cells to which the recombinant alphavirus binds directly and selectively, and / or It can promote cytotoxicity to cells or produce cytokines that promote immunity. One example of a surface antigen is B.A. 7 Histocompatibility antigens; an example of a cytokine is GM-CSF.

A further improvement of recombinant alphavirus that specifically targets cancer cells is to optimize the recombinant alphavirus by passage of the recombinant alphavirus through multiple rounds of infection in the target cancer cells. Can be achieved by choice. This allows the engineered targeting epitope to vary and create the most compatible epitope. This selection process can also be used to determine new epitopes on the surface of specific types of cancer cells.

Epiviral Sequence As discussed above, the choice of epiviral sequence depends on the nature of the target cell. As used herein, “epiviral sequence” refers to an amino acid sequence fused to the envelope protein (s) of the virus. It also refers to the nucleotide sequence encoding the amino acid sequence displayed on the virus surface. Thus, "sequence" refers to either the peptide sequence level or the nucleotide sequence level, and it is clear from the context which is intended.

For the treatment of cancer, the epiviral sequence is preferably a scAb, a ligand, or, generally, a binding domain for a cancer-specific cell surface. The binding domain for a cancer-specific cell surface receptor is a single epitope, or 2
It may contain more than one epitope.

The choice of non-alphavirus sequences used to treat cancer depends on a number of factors, including the type of cancer being treated, the direct and selection between recombinant alphavirus and a given cancer-specific cell surface receptor. And the ability to achieve specific infectivity of target cancer cells. Thus, the ability to achieve direct and selective binding between a recombinant alphavirus and a given cancer-specific cell surface receptor is also due to a number of other factors, including the non-alphavirus sequences that alter the alphavirus E1 or E1
The ability to incorporate into two glycoproteins, such that the non-alphavirus sequences become epiviral and the packaging of the virus is not adversely affected. The ability of target cancer cells to achieve specific infectivity also depends on a number of other factors: binding to cancer-specific cell surface receptors results in internalization of the alphavirus RNA genome, viral replication, and packaging And ultimately, whether it affects cell lysis).

Examples of cancer-specific cell surface receptors include: placental alkaline phosphatase (testicular and ovarian cancer), pancarcinoma (pan carcinoma)
) (Small cell lung cancer), polymorphic epithelial mucin (ovarian cancer), prostate specific membrane antigen, α-fetoprotein, B lymphocyte surface antigen (B cell lymphoma), truncated EGFR (glioma), idiotype (B Cell lymphoma), gp95 / gp97 (melanoma),
N-CAM (small cell lung carcinoma), cluster w4 (small cell lung carcinoma), cluster 5
A (small cell carcinoma), cluster 6 (small cell lung carcinoma), PLAP (seminoma, ovarian and non-small cell lung cancer), CA-125 (lung and ovarian cancer), ESA (carcinoma), CD19, 22, or 37 (B cell lymphoma), 250 kD proteoglycan (melanoma), P55 (breast cancer), TCR-IgH fusion (childhood T cell leukemia), A blood group antigen in blood type B or O individuals (stomach and colon) Tumor).

Other examples of cancer-specific cell surface receptors include: erbB
-2, erbB-3, erbB-4, IL-2 (lymphoma and leukemia), IL
-4 (lymphoma and leukemia), IL-6 (lymphoma and leukemia), MSH (
Melanoma), transferrin (glioma), tumor vascular integrin, etc. Preferred cancer-specific cell surface receptors include erbB-2 and tumor vascular integrins (eg, CD11a, CD11b, CD11c, CD18, CD29,
CD51, CD61, CD66d, CD66e, CD106, and CDw14
5).

[0059] There are many known antibodies against cancer-specific cell surface receptors. C46 Ab against carcinoembryonic antigen (Amersham) and 85A12 Ab (Un
ipath), H17E2 Ab against placental alkaline phosphatase (ICR)
F), NR-LU-10 Ab against all carcinomas (NeoRx Corp.), HMFC1 Ab against polymorphic epithelial mucin (ICRF), W14 Ab against B-human chorionic gonadotropin, RFB4 Ab against B lymphocyte surface antigen (R
Oyal Free Hospital), A33 Ab against human colon carcinoma
(Genex), TA-99 Ab against human melanoma (Genex), c-e
Antibodies to rbB2 (JP 7309780, JP 8176200, and JP 7059588) and the like. scAbs can be developed based on such antibodies using techniques known in the art (eg, Bind et al., Science (
1988) 242: 423-426, and Whitlow et al., Methods.
(1991) 2 (2): 97-105).

[0060] Ligands can be identified by using a radioreceptor assay to screen conditioned media from many cells. Thus, the ligand can be isolated from the cell using known techniques. For example, protein purification techniques such as extraction, sedimentation, ion exchange chromatography, affinity chromatography, gel filtration and the like can be used. Any of a wide variety of procedures can then be used to clone the gene encoding the ligand. For example, a shuttle vector library of DNA inserts from cells expressing the ligand can be analyzed for the presence of the gene. Preferably, the amino acid sequence of the ligand is determined in part, for example, using an automated sequencer or fragmentation with cyanogen bromide or a protease (eg, papain, chymotrypsin, or trypsin). An oligonucleotide having a nucleotide sequence complementary to the oligonucleotide sequence capable of encoding the sequenced amino acid fragment is then identified, synthesized, and extracted from cells cultured under conditions that induce ligand synthesis. Hybridized to a cDNA library generated from the RNA. Alternatively, a library of expression vectors can be prepared by cloning DNA or cDNA from ligand-expressing cells into an expression vector. Expression vector libraries can be screened for anti-ligand molecules. The DNA molecule comprising the ligand gene or a part thereof further comprises:
DNA that can be amplified, such as by PCR, and then encodes an alphavirus
The vector can be introduced to produce a recombinant alphavirus that contains an epiviral ligand for binding to a cancer-specific cell surface receptor.

[0061] Nucleic acids encoding many binding domains are known and can be readily found by reference to publicly available nucleotide sequence databases such as EMBL and GenBank. This DNA can then be amplified, such as using the polymerase chain reaction, and the DNA encoding the alphavirus
The recombinant alphavirus may be incorporated into an NA vector and contain an epiviral binding domain for binding to a cancer-specific cell surface molecule or receptor, as demonstrated in the Examples.

Examples of binding domains include: EGF domain of α-heregulin, α-integrin domain, peptide motif of tumor vasculature, and ht
tp: // ampere. doe-mbi. ucla. edu: 8801 / da
t / dip. dat database and http: // bones. bioch
em. ualberta. ca / pedro / rt-1. What is described in the html database.

In general, there are many databases for ligands, binding domains, and cell surface molecules. For example, ftp: // kegg. genome. ad. j
p, http: // browweb. pasteur. fr / docs / vers
ions, http: // ampere. doe-mbi. ucla. edu:
8801 / dat / dip. dat, or http: // bones. bio
chem. ualberta. ca / pedro / rt-1. html1. checking ...

The epiviral non-alphavirus sequence can be one of a library of binding domains or random sequences. Such sequences can be used to find specific epitopes on the surface of cancerous cells. The resulting recombinant alphavirus is selected by passage and during this selection process, the “right” type is amplified. The resulting recombinant alphavirus can be sequenced in regions of the epiviral non-alphavirus sequence to determine specific sequences that bind to the target cell population. This information can then be used to manipulate more specifically targeted alphaviruses.

The erbB-2 receptor has been found in breast, ovarian, gastric, salivary and adenocarcinoma, and non-small cell carcinoma of the lung. Overexpression of erbB-2 in such cancers was found to correlate with poor prognosis. In vitro studies have e
Overexpression of rbB-2 strongly suggests that it may play an important role in tumor progression.

An example of a scAb binds c-erbB-2 (WO 93/161).
85). For sequences of antibodies that can be used to construct scAbs, see WO9.
3/21232 and http: // www. antibodyy resou
rce. See also com.

An example of a ligand that binds to erbB-2 is α-heregulin. α-Heregulin is obtained by converting erbB-2 into erbB-3 or erbB-.
4 binds to erbB-2 when heterodimerized by WO92 / 12
174). A preferred binding domain is the EGF domain of α-heregulin.
α-Heregulin is a human epidermal growth factor receptor, erbB-2, erbB
-3 and erbB-4 are ligands having affinity for breast cancer cells. Heregulin interacts indirectly with erbB-2 through heterodimerization with erbB-3 or erbB-4.

The specificity of binding can be determined, such as by a virus binding assay, using a radiolabeled virus in the presence and absence of a competitive monoclonal antibody (mAb). The degree of specific infectivity can be determined by measuring infectivity in the presence or absence of the mAb, and the reporter gene (
For example, assaying for activity of chloramphenicol acetyltransferase (CAT), or green fluorescent protein (GFP), measuring the number of infected cells by indirect immunofluorescence, and cell death by trypan blue exclusion By assaying for If only partial specificity of binding is shown, by serially passage the recombinant alphavirus in a culture of cells expressing the targeted receptor using positive and negative selection procedures, Vectors with higher specificity of binding may be selected. For example, the modified alphavirus can be incubated with baby hamster kidney (BHK) cells for 1 hour at 4 ° C. to remove cells that can bind to BHK cells in the population. After this negative selection, the supernatant can be used to infect breast cancer cells that overexpress erbB-2, or erbB-3 and erbB-4 (ie, positive selection). Other suitable methods are known in the art.

Additional Heterologous Proteins Recombinant alphaviruses may further contain and express heterologous sequences under the control of a viral subgenomic promoter. The nature of the heterologous sequence depends on the nature of the target cell and the purpose to be achieved. For example, if the cell is targeted for gene therapy, the heterologous sequence is a sequence that encodes the desired protein. If the cell population selected is the population to be labeled, the heterologous sequence encodes a reporter gene (eg, a gene encoding green fluorescent protein (GFP)). If the target cell is to be destroyed, the heterologous sequence is designed to produce a product that assists in this disruption.

In this latter case, preferably, the heterologous sequence directly or indirectly induces cell death and allows the full-length sequence or portion of the protein or factor to enable apoptosis induced by the recombinant alphavirus. It is a code sequence.
Examples of preferred coding sequences for proteins or factors that induce cell death include Bak
, Bax, Bid, Bad, Bik , Bif-2; Bcl-2 cell death-inducing coding sequence (e.g., including an N-terminal deletion), Bcl-x _L of cell death-inducing coding sequence (
For example, including N-terminal deletions), IAP-1, IAP-2, caspases, and kinases (eg, protein kinase Cθ, protein kinase Cδ, Akt /
PI (3) -kinase, DNA-PK, PITSLRE, DAP kinase, RI
P, JNK / SAPK, Daxx, Raf-1, Pim-1, NIK, MEKK
1, ASK1, PKR and the like. More preferably, the heterologous sequence is Bak
Or the full-length or partial coding sequence of Bax. For example, a cell death-inducing sequence (e.g., a full-length or partial coding sequence of Bak or Bax) desirably, as indicated above, may interact with erbB-2, for example, when indirectly interacting with heregulin. Used in recombinant alphaviruses targeted to erbB-2 on cancer cells that do not express erbB-3 and / or erbB-4 to heterodimerize.

In some cases described herein, the heterologous sequence comprises or encodes a ribozyme or antisense molecule against a gene having anti-apoptotic activity, to induce cell death, To facilitate the spread of the alphavirus, it is desirable to express a coding sequence with anti-apoptotic activity and create permanently infected cells. Examples of such genes include Bcl-
_{2, Bcl-x L, Bcl} -w, Bag-1, Mcl-1, cIap family, viral IAp of a gene, CrmA, P35, and FADD, A
Dominant negative forms of cell death-inducing genes such as SK, JUN / SAPK and PKC. Preferably, the heterologous sequence comprises or encodes a ribozyme that binds, cleaves and destroys Bcl-2 mRNA.

Alternatively, the recombinant alphavirus may further comprise and express, as a heterologous sequence, a coding sequence that promotes cell death in the presence of a cytotoxic agent. An example of such a coding sequence is the sequence of a thymidine kinase gene (eg, from a herpes virus). Expression of the thymidine kinase gene in cells treated with ganciclovir results in their death. Other such prodrug strategies can be used.

If the recombinant alphavirus contains and expresses heterologous sequences, one of two types of vectors may be used (eg, Huang et al., Virus Ge
nes (1989) 3 (1): 85-91; Bredenbeek et al. (199
Schlesinger et al. (1993), supra; or Fluoro.
v. et al. (1996), supra). One type of vector contains two copies of the viral subgenomic promoter. One promoter controls the synthesis of viral structural proteins and the other controls the expression of the heterologous sequence.
Vectors are self-replicating, produce infectious viral particles, and can spread from cell to cell.

[0074] The other type of vector is a vector in which a heterologous sequence replaces one or more structural protein genes. This type of vector requires helper RNA for packaging of the recombinant RNA into viral particles. Vectors do not produce infectious viral particles and therefore cannot spread from cell to cell.

Other Modifications If desired, the recombinant alphavirus may have a genomic transcriptionally non-alphaviral promoter (eg, a cancer-specific or inducible promoter).
Can be further modified to be under the control of Preferably, the cancer-specific promoter is a promoter that is only activated in cancer cells that are directly and selectively bound by the recombinant alphavirus. The use of a cancer-specific promoter provides another layer of protection and ensures that the alphavirus replicates only in cancer cells. An example of a cancer-specific promoter is CEA. Other promoters are
On the Internet, http: // www. genome. ad. jp / db
get-bin / www_bFind? can be found in the eukaryotic eukaryotic promoter database. Alternatively, and preferably, the promoter can be a cell-specific promoter, from which the target cancer cell or target supporting cell actively transcribes the mRNA.

In addition to the above, the present invention also provides a recombinant DNA vector into which the RNA genome of the above-described recombinant alphavirus (eg, a recombinant Sindbis virus) can be transcribed.

The present invention also provides a substantially pure RNA genome of the above-described recombinant alphavirus (including a recombinant Sindbis virus). "Substantially pure" is free of recombinant DNA vector and cellular components (if the RNA genome is produced from a recombinant DNA vector) or contains viral coat proteins and other viral components. No (if the RNA genome is purified and isolated from alphavirus particles (eg, Sindbis virus particles)).

One Use of Targeted Alphavirus-Cancer Treatment In view of the above, the present invention provides a method of treating cancer in a mammal. "Cancer" according to the present invention includes cancers characterized by abnormal cell growth and absence of contact inhibition, which may be evidence of tumorigenesis. The term includes cancers that are localized to the tumor as well as cancers that are not localized to the tumor (eg, cancers that spread locally from the tumor by invasion or systemically by metastasis). In theory, any type of cancer can be targeted for treatment according to the present invention. However, preferably, the cancer is of epithelial origin (eg, breast or ovarian cancer).

One method involves administering to a mammal in need of treatment for cancer a cancer treating effective amount of a recombinant alphavirus (eg, a recombinant Sindbis virus) as described above, where it is to be treated. Upon binding to cancer cells, the recombinant alphavirus infects the cells and lyses the cells, thereby treating the cancer. Treatment of cancer can be assessed, for example, by monitoring tumor growth decay and / or tumor regression, where "tumor growth" includes an increase in tumor size and / or number of tumors, and "tumor regression" Includes a decrease in tumor mass. Where the recombinant alphavirus comprises at least one other epiviral non-alphavirus sequence and the sequence is an immunogen, preferably the immunogen elicits an immune response against the cancer to be treated. Desirably, the cancer is a cancer of epithelial origin (eg, breast or ovarian cancer).

Another method of treating cancer in a mammal comprises administering to a mammal in need of cancer treatment an effective amount of a recombinant DNA vector or RNA genome as described above for treating cancer. Upon entry of the recombinant vector or RNA genome into the cancer cell, the recombinant DNA vector or RNA genome produces an RNA genome, which is then packaged into a viral particle, which lyses the cancer cell, and Selectively binds and infects other cancer cells, which are eventually lysed, thereby treating the cancer. If the recombinant DNA vector or RNA genome encodes at least another epiviral non-alphavirus sequence, and the sequence is an immunogen, preferably the immunogen will generate an immune response against the cancer to be treated. Provoke. Preferably,
The cancer is a cancer of epithelial origin (eg, breast or ovarian cancer).

In this regard, the method of treating cancer according to the present invention further comprises a non-cancerous cell (eg,
Pre-cancerous cells potentially infected with virus) or cells supporting tumor growth (eg,
Tumor angiogenesis endothelial cells). Targeting tumor angiogenic endothelial cells includes targeting integrins, for example, with a recombinant alphavirus that includes the RGD peptide as an epiviral non-alphavirus sequence.

[0082] The methods of the invention for treating cancer in a mammal can be used alone or in combination with radiation, chemotherapy and / or surgery. For example, such combination treatments can be used early or late in the progression of the cancer, including the metastatic stage. The recombinant alphavirus according to the invention can be introduced at the site of a mastectomy or oophorectomy to infect, for example, the remaining tumor cells after surgery. Recombinant alphavirus can also be introduced into the mammary gland by conduit cannulation.

According to the present invention, a recombinant alphavirus, preferably a recombinant Sindbis virus, a recombinant DNA vector into which the RNA genome of the recombinant alphavirus can be transcribed, or a RNA genome of the recombinant alphavirus requires Administered to a mammal. The means of administration can be by any suitable means. This may be due, in part, to a recombinant alphavirus or recombinant DNA vector or RN.
Determined by which of the A genomes is being administered. Suitable routes of administration include peritumoral, intratumoral, intravenous, intramuscular, intraperitoneal, subcutaneous, oral, rectal, ocular, intranasal, and the like. Peri- and intratumoral routes of administration (eg, by injection) are preferred. Administration by lipofection, direct DNA injection, microprojectile bombardment, liposomes, molecular conjugates (eg, for recombinant DNA vectors and RNA genomes) can also be achieved. As an alternative to in vivo administration, in ex vivo technology, cells can be removed, contacted with a recombinant alphavirus, recombinant DNA vector or RNA genome, and returned to the mammal. However, the method does not depend on any particular means of administration and should not be construed as such. Means of administration are well-known to those of skill in the art and are also exemplified herein.

For example, a targeted alphavirus (TA) can be delivered as a virus after being packaged in a suitable packaging cell line. Since TA can only invade targeted cells (eg, breast cancer cells), the packaging cell line must be made from a cell line of the same type as the target cell line. Thus, for a TA targeting breast cancer, the packaging cell line must be the same type of breast cancer cell for which the TA is to be targeted. Specifically, erb-B2-targeted TA is packaged in an erb-B2-overexpressing breast or ovarian cancer cell line cultured in vitro. The virus then
It can be concentrated (eg, by filtration or ultracentrifugation) and administered directly (as an aid to surgical treatment) at the site of the tumor or at the site of tumor resection.

[0085] The targeted alphavirus vector (TAV) can be delivered directly to the tumor site or surrounding site, where the TAV can produce TA that can spread through the tumor. Alternatively, the TAV may be delivered first to the packaging cell line or to the patient's own irradiated tumor cells, which act as packaging cells, which may then be delivered back to the tumor site. The number of cells can be introduced back to the patient is about 10 ³ to about 10 ¹⁴ cells per inoculation. Multiple inoculations may be required at the same site or at different sites.

When TAV is delivered as a nucleic acid, it can be delivered as either RNA or DNA. When TAV is delivered as RNA, it must first be transcribed from a DNA template to make the RNA. This RNA can be made by using standard in vitro transcription techniques known in the art. This RNA can then be complexed with liposomal or non-liposomal (collectively referred to as "lipoid") delivery systems to efficiently transfect cells with TAV RNA. Once the RNA has entered the cytoplasm of a cell, which may be a target cell or a non-target cell, the RNA
Can amplify itself to produce TA particles. From this first round of replication, TA
Can be produced, which then specifically infects and spreads through the target population of cancer cells. Cells producing the first round of virus die due to productive viral infection, but this is not critical. This is because the number of cells transfected is relatively small compared to the subsequent replication and expansion of TA through the target cell population (eg, tumor). If the vector is to be introduced intravenously, TAV can also be added to a liposome or non-liposome delivery system.
It can be targeted to tumors by targeting lipoid-RNA and incorporating molecules that more specifically complex with tumor vasculature. Tumor vasculature has specific surface molecules (i.e., integrins), and by incorporating appropriate ligands that target these integrins to lipoid-RNA complexes, the first round of TA leads to tumor vasculature. And then the TA particles can be seeded into the tumor. TA can then grow and spread through the tumor. In the case of tumor-integrin-targeted TAV, incorporation of the ligand into the lipoid-DNA complex further acts to facilitate targeting of integrin-tropic TAV to tumor vasculature. TAV lipoid-RNA can specifically infect tumor vasculature endothelial cells, and TA particles can then spread through adjacent tumor endothelial cells and kill them and cut off the tumor's blood supply.

[0087] TAV can also be delivered as a DNA-lipoid complex. To enable this, the promoter must be placed in close proximity to the virus to ensure the first round of expression of TAV. This is because alphaviruses are RNA viruses without DNA intermediates. One way to do this is to transfect the cells directly with plasmid DNA TAV without a eukaryotic promoter system. Even if the TAV plasmid DNA does not contain a eukaryotic promoter (but does contain potential sequences for eukaryotic transcription of the vector), the virus is transfected with the eukaryotic promoter-deficient plasmid DNA encoding the alphavirus. Can be produced from the isolated cells. This first round of virus can then be amplified by passage in a suitable packaging cell line or in vivo.

Another method of producing the first round of virus from DNA-lipoid TAV is
The co-expression of the bacteriophage polymerase gene in cells containing the AV plasmid DNA, which contains a promoter sequence for polymerase binding. An example of this is the co-expression of bacteriophage SP6 polymerase in cells containing TAV along with the SP6 polymerase promoter sequence. The SP6 polymerase can then bind to the SP6 promoter sequence,
The first round of TAV RNA can then be transcribed. The TAV RNA can then amplify, infect the target cell population tumor cells, and spread through it.
TAV targeting can be made more specific by linking the SP6 polymerase gene to a promoter that is only expressed in target cells. For example, SP6 polymerase can be linked to a tumor-specific promoter, such that the first round of virus is only produced from tumor cells and not other cell types.
Such a multi-layer approach assists in targeting TAVs and reduces their toxicity (non-cancer / non-tumor cell type infection).

Another method of generating the first virus from DNA-lipoid TAV is by directly ligating the TAV plasmid DNA sequence to a eukaryotic promoter. TCV
Can be made more specific for tumor cells by using a tumor-specific promoter. The concept of using a multi-layered strategy to promote a gradual increase in specificity (lipoid targeting + promoter targeting + virus targeting) is particularly important for clinical applications.

A preferred means for initial virus production is to use a carrier vector (eg, an adenovirus vector, or preferably a crHIV vector (Dropulic et al.,
WO 97/20060)) to replace the internal TAV sequence. Initial delivery can be achieved by delivery of the carrier vector to target cells, which in turn produces primary TAV-targeted alphavirus particles, which in turn infect the target cell population specifically.

A more preferred method of TAV delivery is inside a crHIV vector. crH
Since the size of the IV genome is smaller than the alphavirus genome, TAV is
It is separated into two components containing sequences that allow its recombination. Introduction of two heterologous crHIV TAV components into a packaging cell line is achieved by introducing two heterologous TAV components RN
Allows simultaneous packaging of A into one virion (because two crs
HIV RNA is co-packaged into one virion) and cr
Upon infection of target cells with an HIV vector, reverse transcription of crHIV RNA results in recombination and production of the entire provirus of the infectious targeted alphavirus. The targeted alphavirus provirus can be driven by a eukaryotic promoter and expressed in cells to produce TA particles. Alternatively, the provirus of the two targeted alphaviruses expresses its component RNA, which in turn produces the entire TA RNA genome for its amplification and production of target alphavirus particles.

The most preferred delivery method for TAV is to overlay the two components crHIV-TCA / AV inside the vector-helper crHIV direct delivery system. c
Upon infection of a cell with the rHIV vector-helper DNA-lipoid complex, a two-part crHIV-TA particle is initially generated for infection of the cell. Cells infected with two factors, crHIV-TA, generate cells containing the targeted alphavirus provirus. The cells produce alphavirus particles targeted for specific infection and killing of the target cell population. An array of promoter / vector permutations can be used to generate TA from specific cell types. This will assist in the development of highly specific TAs.

[0093] Recombinant alphavirus for use in the methods of the present invention may be stored in either crude or purified form. To produce the virus in crude form, the virus-producing cells can be cultured in a bioreactor, and the virus particles
Released from the cells into the culture medium. The virus can then be stored in crude form by adding a sufficient amount of formulation buffer to the culture medium containing the recombinant alphavirus to form an aqueous suspension. The virus may be stored in purified form by purifying the crude recombinant virus by passing through a filter, and then concentrating the virus, for example, by cross-flow concentration. DNase is added to digest exogenous DNA, and the digest can be diafiltered to remove excess media components and establish the virus in a more desired buffered solution. The diafiltrate is then passed through a column,
Then, the purified recombinant alphavirus is eluted. A sufficient amount of formulation buffer can then be added to the eluate so as to reach the desired final concentration of displacement and to minimize dilution of the recombinant virus. The aqueous suspension can then be stored (e.g., -70 <0> C) or immediately dried. Crude recombinant alphavirus can also be purified by ion exchange column chromatography. Aqueous suspensions can be dried by lyophilization or evaporation at ambient temperature. in this regard,
Such aqueous suspensions will contain components to preserve the activity of the recombinant alphavirus upon freezing and lyophilization or drying through evaporation. For example, the suspension may include buffers, high molecular weight structural additives, saccharides and water (see, eg, WO 96/17072). The lyophilized or dehydrated recombinant alphavirus can then be reconstituted (reconstituted) using, for example, water or dilute salt solutions. Other additives include components that enhance the transduction efficiency of the reconstituted virus.

[0094] Preferably, the recombinant alphavirus is administered to a mammal in the form of a pharmaceutically acceptable composition. The composition must not impair the ability of the recombinant alphavirus to bind directly and specifically to cancer-specific cell surface molecules, or cancer-specific cell surface receptors, on the cancer to be treated. .

[0095] Preferably, the recombinant DNA vector or RNA genome comprises a cationic lipid,
For example, it is administered by liposome means. Such liposomes are commercially available (eg, Life Technologies, Gibco BRL).
Lipofectin®, Lipofectamine ^™ , etc. supplied by Gaithersburg, MD). In addition, liposomes with increased metastatic capacity and / or reduced toxicity in vivo (eg, as reviewed in PCT Patent Application No. WO 95/21259) can be used in the present invention. For liposome administration, the recommendations identified in WO 93/23569 may be followed. Similarly, other delivery vehicles include hydrogels and controlled release polymers. If desired, liposomal formulations and the like can be targeted to cancer cells by making the liposomes present a mAb, scAb, ligand or binding domain (eg, for a cancer-specific cell surface molecule or receptor). (See, eg, Miller et al. (1995), supra).

Alternatively, recombinant DNA vectors are targeted to cells by linking the DNA to a mAB, scAb, ligand or binding domain, for example, by covalent attachment of a polycation (eg, polylysine) to the ligand. obtain. Then
Polycations bind to plasmid DNA via electrostatic interactions and condense, leaving ligand on the virus (see, eg, Miller et al. (1995) (
See above)).

Prior to administration to mammals, the genome of the recombinant DNA vector or RNA of the present invention can be formulated into various compositions by combining with a suitable pharmaceutically acceptable carrier or excipient. And can be formulated to be suitable for either human or veterinary applications.

Thus, a composition for use in the method of the present invention combines the genome of one or more of the foregoing recombinant alphaviruses, recombinant DNA vectors or RNA, preferably with a pharmaceutically acceptable carrier. And can include. Pharmaceutically acceptable carriers, as well as suitable modes of administration, are well-known to those of skill in the art. The choice of carrier will be determined in part by whether the recombinant alphavirus or the genome of the recombinant DNA vector or RNA is to be administered, and by the particular method used to administer the composition. . One skilled in the art will also recognize that various routes of administering the composition are available, and that more than one route may be used for administration,
It is understood that certain routes may provide a faster and more effective response than another route. Accordingly, there is a wide variety of suitable formulations of the compositions that can be used in the methods of the present invention.

The recombinant alphavirus, recombinant DNA vector, RNA genome, or composition comprising such an alphavirus, recombinant DNA vector, or RN A genome, alone or with one or more active agents Can be made into formulations suitable for parenteral administration, preferably intraperitoneal administration. Such formulations may include aqueous and non-aqueous solutions, isotonic injection solutions. This includes antioxidants, buffers, bacteriostats and solutes, and suspending, solubilizing, thickening, stabilizing, and preserving agents that render the formulation isotonic with the intended recipient's blood. It may include aqueous and non-aqueous sterile suspensions which may contain the agents. The formulation may be in unit or multi-dose closed containers, such as ampules and vials, and lyophilized requiring only the addition of a sterile liquid carrier (eg, water for injection) immediately before use. It can be stored in a stored state. Extemporaneous injection solutions and suspensions can be prepared from sterile powders, granules and tablets, as described herein.

Formulations suitable for oral administration include liquid solutions such as an effective amount of a compound dissolved in a diluent (eg, water, saline or fruit juice); capsules, cachets (
sachets or tablets (each containing a predetermined amount of active ingredient as a solid or granule); solutions or suspensions in aqueous liquids; and oil-in-water or water-in-oil emulsions. Tablet form comprises one or more lactose, mannitol, corn starch, potato starch, microcrystalline cellulose,
Gum arabic, gelatin, colloidal silicon dioxide, croscarmellose sodium, talc, magnesium stearate, stearic acid and other excipients, coloring agents, diluents, buffers, wetting agents, preservatives, flavoring agents, As well as a pharmaceutically acceptable carrier.

[0101] Similarly, formulations suitable for oral administration may include lozenges. This is the active ingredient of the flavor, usually sucrose and acacia or tragacanth; fragrances containing the active ingredients in inert bases, for example, gelatin and glycerin, or sucrose and acacia; and the active ingredients in suitable liquid carriers. Mouthwashes; and creams, emulsions, gels, and the like (containing, in addition to the active ingredient, carriers known in the art).

Aerosol formulations suitable for administration via inhalation can also be made. The aerosol formulation can be in a pressurized acceptable propellant, for example, dichlorodifluoromethane, propane, nitrogen, and the like.

[0103] Formulations suitable for topical application may be in the form of creams, ointments or lotions.

Formulations for rectal administration may be presented as a suppository with a suitable base comprising, for example, cocoa butter or a salicylate. Formulations suitable for vaginal administration comprise, in addition to the active ingredient, a carrier as is known to those skilled in the art to be suitable.
It may be presented as a pessary, tampon, cream, gel, paste, foam, or spray formulation.

In the context of the present invention, the dose administered to a mammal, especially a human, is sufficient to affect a therapeutic response in an infected individual over a reasonable time range.
The dose will be determined by the genomic titer of the particular recombinant alphavirus, recombinant DNA vector or RNA used for treatment, the severity of the cancer, and the weight and age of the infected individual. The size of this dose will also be determined by the presence of any adverse side effects that may accompany the use of the particular recombinant alphavirus, recombinant DNA vector or RNA genome used. It is always desirable whenever possible to keep harmful side effects to a minimum.

This dosage may be in unit dosage form such as a tablet or capsule. As used herein, the term "unit dose form" refers to a physically discrete unit suitable as a single dosage for human and animal subjects. Each unit contains a predetermined amount of the vector, alone or in combination with other anti-cancer agents. this is,
The amount has been calculated to be sufficient to produce the desired effect in combination with the pharmaceutically acceptable diluent, carrier, or vehicle. The details for the unit dosage form of the invention depend on the particular embodiment used and the effect to be achieved, as well as the pharmacokinetics associated with each compound in the host. The dose administered is the "effective amount for treating cancer" or the amount necessary to achieve an "effective level" in an individual patient.

Since “effective level” is used as a preferred endpoint for dose determination, actual dosages and schedules may vary depending on inter-individual differences in pharmacokinetics, drug distribution and metabolism. "Effective level" can be defined, for example, as the blood and tissue level desired in a patient. This is one according to the present invention.
It corresponds to the concentration of one or more recombinant alphaviruses or recombinant DNA vectors, which lyse the targeted cancer cells in assays that predict clinical anticancer activity.
The “effective level” for the genome of a recombinant alphavirus, recombinant DNA vector, or RNA of the present invention may also vary when the composition of the present invention is used in combination with other known anti-cancer agents.

Those of skill in the art can readily determine the appropriate dose, schedule and method of administration for precise formulation of the composition being used to achieve the desired “effective level” in the individual patient. obtain. Those skilled in the art will also appreciate direct analysis (eg, tumor biopsy or x-ray images of the tumor) or appropriate patient samples (eg, blood and / or tissue).
Suitable indices of the "effective level" of the compounds of the present invention can be readily determined and used by indirect analysis (e.g., PSA levels in blood).

Further, with respect to determining an effective level in a patient for the treatment of cancer,
Suitable animal models are available and have been extensively implemented to assess the in vivo efficacy of recombinant DNA protocols against cancer (eg, PCR
checking). These models include nude and SCID mice. Such models can also be used to evaluate the in vivo efficacy of the RNA genome.

In general, the genome of a recombinant alphavirus, recombinant DNA vector, or RNA in an amount sufficient to obtain a tissue concentration of about 10 ² to about 10 ¹² particles / ml is:
Preferably, especially about 10 ⁶ to about 10 ¹⁰ particles per ml are preferred. For certain applications, multiple doses per day are preferred. In addition, the number of doses depends on the means of delivery,
And depending on the particular recombinant alphavirus, recombinant DNA vector or RNA genome to be administered.

Pharmaceutical compositions, when used to therapeutically treat cancer, may include other formulations in combination with the recombinant alphavirus, recombinant DNA vector, or RNA genome of the present invention. In particular, an anticancer agent, eg, preferably a recombinant virus, nucleic acid / liposome formulation (or other nucleic acid delivery formulation) or another vector system (either as a virus particle or a nucleic acid / liposome formulation) For example, retroviruses or adenoviruses) are intended to be used. In addition, representative examples of these additional formulations that may be used, in addition to those previously described, include chemotherapeutic agents, immunostimulants, antiviral compounds, and other agents that may be used in treating cancer. As well as treatment regimens, including those recognized as alternative medicines. Anti-cancer compounds include, but are not limited to, angiostatin, endostatin, anti-HER-2 / neu antibody and tamoxifen. Immunomodulators and immunostimulants include, but are not limited to, various interleukins, cytokines, antibody preparations and interferons.

The method of treating cancer in a mammal can be adapted for use in vitro, for example, to determine cancer-specific expression of a protein. For example, non-alphavirus sequences can be introduced into a recombinant alphavirus or, more specifically, a recombinant Sindbis virus. And cancer-specific expression of proteins to which non-alphavirus sequences can bind can be assessed by lysis of the cancer cells with the recombinant alphavirus.

(Other Applications) In yet another aspect, the present invention provides a method for detecting a cancer cell. This method involves contacting a cell with a recombinant alphavirus of the invention, wherein the alphavirus is optionally non-cytolytic and replication-deficient and comprises a marker gene such as a green fluorescent protein. And detecting the expression of the marker gene, wherein detecting the expression of the marker gene includes the step of indicating the presence of a cancer cell.

Another method provided by the present invention is a method for determining new cancer cell-specific cell surface molecules or receptors. The method comprises the steps of introducing a sequence from a library of binding domains or a random sequence as an insertion or substitution of one or more amino acid residues of glycoprotein E1, E2, or E3 of the recombinant alphavirus. Contacting the virus with a cancer cell, selecting a recombinant alphavirus that binds to the cancer cell to identify a cancer-specific cell surface molecule or receptor to which the recombinant alphavirus has bound.

EXAMPLES The present invention is further described in the context of the following examples. These examples serve to further illustrate the invention and are not intended to limit the scope of the invention.

Example 1 Sindbis virus E2 glycoprotein with unique restriction endonuclease site
Introduction into the cytoplasmic region) Each of the three unique restriction endonuclease sites (ie, Mlu I, S
ph I, and Bst EII) were converted to two separate sites by site-directed mutagenesis.
Using PCR in the reaction of Sindbis virus clone TE (Dian)
e Dr. Griffith, The Johns Hopkins Unive
rity School of Hygiene and Public H
E2 glycoprotein) (available from E.H.). In one reaction
Using 5 'flanking primers and 3' mutagenic primers, and
Used a 5 'mutagenic primer and a 3' flanking primer. E2
5 'flanking primer for introduction of a Mlu I site at amino acid residue 175 (5
'SV-Stu I) and 3' mutagenesis primers were 5'-CT
 CGA CAT CCT TGA AGA G-3 '(SEQ ID NO: 1) and 5
'-ATG TATACG CGT GCG GTC TCG GC-3 '(
SEQ ID NO: 2), while the 5 'mutagenesis primer for the introduction of a Mlu I site.
Mer and the 3 'flanking primer (3' SV-Bcl I) were each 5'-
CGCACG CGT ATA CAT CCT ACC TG-3 '(sequence
No. 3) and 5'-ATT TCC CTT GGG CCG TGT-3 '
(SEQ ID NO: 4). Introduction of the Sph I site at E2 amino acid residue 60
5 'flanking primer (5' SV-Stu I) and 3 'mutagenic primer
Are 5'-CT CGA CAT CCT TGA AGA G-
3 '(SEQ ID NO: 5) and 5'-TTT GCG CAT GCT GCT C
CG CT-3 '(SEQ ID NO: 6), while the 5' mutagenesis primer and
The 3 'flanking primers (3' SV-Mlu I) were each 5'-AGCA GC ATG C GC AAA CAA GTA CC-3 '(SEQ ID NO: 7)
And 5'-ATG TATACG CGT GCG GTC TCG GC-
3 '(SEQ ID NO: 8). Bst EII site at E2 amino acid residue 114
Flanking primer (5 'SV-Sph I) and 3' mutagenesis for the introduction of
The developing primers were each 5'-AGC AGC ATG CGC AAA
CAA GTA CC-3 '(SEQ ID NO: 9) and 5'-CACTATGG
TAA CCG TTA CGC TGT-3 '(SEQ ID NO: 10).
5 'mutagenesis primer for introduction of Bst EII site and 3' flanking primer
The immers (3'SV-MluI) were each 5'-ACG GTT ACC
 ATA GTG AGT AGC AA-3 '(SEQ ID NO: 11) and 5'-
ATG TATACG CGT GCG GTC TCG GC-3 '(sequence
No. 12). The PCR products from the two reactions are each a new restriction
Contains a nuclease site at one end and therefore hybridizes to each other.
It was able to soy and was combined with two adjacent primers and amplified. Final P
The CR product is gel purified and digested with enzymes immediately upstream or downstream of the adjacent primer.
And run on an agarose gel to ensure that the fragments are the correct size
It was confirmed. Then, from Sindbis virus clone TE ("parent virus")
1200 bp Stu I-BSH II fragment (this is
Containing three introduced restriction sites in the protein), Stu I-BS
Subclones in SH II digested dual subgenomic vector TotoCat5
And thereby place the corresponding TotoCat5 E2 sequence (shown in FIG. 1).
Replace. FIG. 1 shows that the alphavirus expression vector TotoCat5 duplex
Map of the subgenomic vector (top) and the single subgenomic vector (bottom)
is there. Precise assembly of all constructs and new restriction sites
Analysis and dideoxy-mediated chain termination sequencing (Sequenase, Amers)
ham, Arlington Heights, Illinois).
I accepted. Two major neutralizing epitopes in E2 glycoprotein and unique restriction enzymes
The elementary sites are shown in the schematic diagram of FIG.

Example 2 (Construction of Recombinant Sindbis Virus Expression Vector ^Containing Non- ^Alphavirus Sequence α-Heregulin) The sequence of α-heregulin was obtained from p-GEX- ^HRGa177-244 (Wallashch et al., EMBO J ( 1995) 17: 4267-4275) (this is Axel
Ullrich (Department of Molecular Biol
ogy, Max Planck Institute for Biochem
IE, Martinsried, Germany). Α-Heregulin sequence cloned into SV vector (EGF domain)
Is 200 bp long. This sequence was amplified using PCR and the primers shown in Table 1 (underlined indicates restriction sites) and in-frame To
Subcloning into toCat5-E2ab or E2c.

(Table 1) (Primers for PCR amplification of heregulin sequence)

[0119]

[Table 1] A control vector containing an irrelevant scAb (eg, an anti-gp120 scAb) or an irrelevant ligand (eg, a gp120-derived CD-4 binding ligand) is also generated.

Example 3 (Construction of a recombinant Sindbis virus clone TE containing the non-alphavirus sequence α-heregulin as a substitution of E2 amino acids 175-243 (TEMluBc
l)) The 200 bp sequence for α-heregulin was obtained from p-GEX-HRG ^a177-244 and PCR and restriction endonuclease sites Mlu I and / or Bcl I at the 5 ′ end as shown in Table 1 Amplification was performed using encoding primers. Heregulin that was PCR amplified was subcloned into the E2 glycoprotein sequence of the recombinant Sindbis virus clone TE by replacing E2 amino acids 175-243 to generate TEMluBcl. Correct construction of the modified virus was confirmed by restriction enzyme analysis and DNA sequencing.

Example 4 (Recombinant Sindbis virus clone TE (TEBcl) containing the non-alphavirus sequence α-heregulin as an insertion at the Bcl site at E2 amino acid 243)
Construction of the 200 bp sequence for α-heregulin was obtained from p-GEX-HRG ^a177-244 and primers encoding the restriction endonuclease site Bcl I at the 5 ′ end as shown in PCR and Table 1 (5 'BclI and 3'Bcl
Amplified using I). The PCR amplified heregulin was subcloned into TEBcl by inserting a 200 bp sequence into the E2 glycoprotein sequence of the recombinant Sindbis virus clone TE at the BclI site of E2 amino acid 243.
Generated. The correct construction of the modified virus was confirmed by restriction enzyme analysis and DNA sequencing.

Example 5 Construction of Recombinant Sindbis Virus Expression Vector Containing Non-Alphavirus Sequence Anti-erbB-2 scAb The sequence of erb-2 scA was replaced with pFRP5 (Nance Hynes
(Fredrich Miescher Institute, Basel, S
(available from Witzerland). The erbB-2 sequence is
It is 711 bp long. The sequence is amplified using PCR and the TotoC
Subclones into at5-E2ab or TotoCat5-E2c or Sindbis virus clone TE. Anti-erbB-2 scA, TotoCa
For cloning of t5 into the E2ab site, the 5'Mlu I and 3 'Bcl I primers in Table 2 (underlined indicate restriction sites). For cloning of the anti-erbβ-2 scA into the E2c site of TotoCat5, the 5 ′ Sph I and 3 ′ Mlu I primers of Table 2 are used.

A control vector containing the anti-gp120 scAb is also generated.

Table 2 Primers for PCR amplification of anti-erbB-2 sequence

[0125]

[Table 2] Example 6 (at one of two different sites in E2 glycoprotein, or E2
Construction of a Recombinant Sindbis Virus Clone TE Containing a Non-Alphavirus Sequence NGR Binding Domain Specific for Binding to Angiogenic Endothelial Cells as a Substitution at a Site Across the E3 Glycoprotein) , Science (1998) 279: 3.
77-380 and amplified using PCR and the primers shown in Table 3. PCR amplified NGR domain was cloned into recombinant Sindbis virus clone T
Subcloning was performed by substituting E2 amino acids 61-73 or 190-202, or E3 amino acids 58-E2 amino acids 6, into the E2 or E2 / E3 glycoprotein sequence of E. The correct construction of the modified virus was confirmed by restriction enzyme analysis and DNA sequencing.

Table 3 Primers for PCR amplification of NGR sequences

[0127]

[Table 3] Example 7 Production of Recombinant Alphavirus Stock Infectious recombinant alphavirus RNA is transfected into BHK cells, and the supernatant is collected 24 hours after transfection. The supernatant after transfection contains the recombinant virus. The amount of virus present was titrated on BHK cells and measured in plaque-forming units per ml (pf
u) is determined by determining the number.

[0128] The amount of infectious units of virus per 1ml of supernatant also in the presence of actinomycin D, and ³ H-uridine to inhibit cellular RNA polymerase, propagating virus (which, ³ H Uridine into viral RNA). ^The supernatant containing the ³ H uridine-labeled viral RNA is TCA precipitated, counted in a liquid scintillation counter, and labeled wild-type with a known titer to determine the number of infectious units / ml. Compare to type culture. Alternatively, the viral stock supernatant can be immunoblotted to determine the relative amount of viral protein present in the supernatant.

[0129] Wild-type and heregulin-containing virus stocks were 10 ⁸ to 10 ⁹ pfu / ml.
Is expected to be contained. Plaque titration on BHK cells does not work. Because the tropism of the virus changes, the virus is unable to bind and the spread virus can be quantified using a stably transfected BHK cell line expressing the wild-type E2 glycoprotein. Because. The wild-type protein may complement non-alphavirus sequences containing E2 to allow spread of virus and plaque titers.

Recombinant alphavirus stocks can also be used for (i) co-transfection with helper RNA constructs, (ii) transfection of RNA into packaging cell lines, (iii) transfection of DNA, and (iv) SP
6 can be produced by co-transfection of DNA.

Example 8 Synthesis of Recombinant Virus Structural Proteins by Recombinant Alphavirus Vector DNA (2-5 μg) was linearized with SstI and SP6 DN
A-dependent RNA polymerase (RiboMAX; Promega, Madiso)
n, Wisconsin) at 5 mM rATP, rCTP, and rUTP, respectively.
, 0.6 mM rGTP, 3 mM cap analog (7 ^m G5'ppp5'G)
Transcribe in vitro using The reaction is incubated for 2 hours at 37 ° C.
RNA is phenol / chloroform extracted, precipitated with isopropanol, and suspended in RNase-free Tris-Cl-EDTA. RNA is purified from Lipofectin (GibcoBRL, Gaithersburg, Marylan).
d) or transfect into BHK21 (ATCC) cells using electroporation. The virus-containing supernatant was used for 24 hours of transfection.
Collect at time, clarify to remove cell debris, and store at -80 ° C. E2
To determine whether the α-heregulin or anti-erbB-2 scA glycoprotein-containing is incorporated into the virus, the proteins from the virus-containing supernatant were either radioimmunoprecipitated or not deleted from the recombinant virus. Immunoblot using anti-Sindbis virus rabbit serum that recognizes an epitope in the E2 glycoprotein. In addition, a protein immunoblot using anti-scA and anti-heregulin antibodies is performed to demonstrate that the scA or heregulin sequences are present within the E2 glycoprotein sequence.

Example 9 Production of Recombinant Viral Structural Proteins by Recombinant Alphavirus Generated by Plasmids TEMluBcl and TEBcl The plasmids of Examples 4 and 5 were linearized and recombinant Sindbis virus RNA was converted to 5 mM rATP, rCTP and rUTP, 0.6 mM r
SP6 with GTP, 3 mM cap analog (7 ^m G5'ppp5'G)
In vitro transcription was performed using DNA-dependent RNA polymerase (RiboMAX). The reaction was incubated at 37 ° C. for 2 hours. The RNA was phenol / chloroform extracted, precipitated with isopropanol, and resuspended in RNase-free Tris-Cl-EDTA. This RNA was converted to BHK21 (A
(TCC) cells were transfected using Lipofectin (GibcoBRL) or electroporation. The supernatant containing the virus was collected 24 hours after transfection, clarified to remove cell debris,
Then, it was stored at -80 ° C. E2 containing α-heregulin
To determine whether glycoproteins were incorporated into the virus, supernatant containing virus (50 μl from lysate of infected cells and 1 μl from stock supernatant).
00 μl) were immunoimmunoprecipitated or immunoblotted using anti-Sindbis virus rabbit serum that recognizes an epitope in the E2 glycoprotein that was not detected from the recombinant virus. In addition, a protein immunoblot using anti-scA and anti-heregulin antibodies is performed to demonstrate that the scA or heregulin sequences are present within the E2 glycoprotein sequence.

The E1, E2 and capsid proteins were converted to TE, TEMluBcl, TEB
Detection was performed in cell lysates infected with cl. Cell lysates from TEBcl infected cells displayed a Western blot band of the same size as wild type E2.
This indicated that the heregulin inserted at the BclI site might have been deleted from the virus by RNA recombination. The slowest migrating band in the lane of the TEBcl infection may represent a heregulin-containing E2 glycoprotein that is predicted to be about 7,300 daltons greater than wild-type E2, or this band is a precursor of E2 May represent PE2. Reverse transcription, followed by PCR using wild-type and heregulin-containing virus-specific primers, is performed to demonstrate the presence of heregulin in E2 (Santa Cruz Biot).
technology, Inc. , Santa Cruz, California
). Capsid and E2 glycoprotein were easily identified in the supernatant. E1
Glycoprotein detection was more difficult, but was present, as confirmed by immunoblot using anti-E1 mAB. The epitope presented by E1 in mature virions was not easily detected by anti-Sindbis virus rabbit serum.

The heregulin-containing viruses TEMluBcl and TEBcl were also
Early after transfection of NA, radioimmunoprecipitation (RIP) assays
Was analyzed. 2.5 μg of virus mixed with 10 μl of lipofectin
Eight hours after transfection of BHK cells with RNA, the cells are
Depleted for 40 minutes in medium lacking stain and methionine, and ³⁵ label (ICN, Irvine, California)
Pulsed with 0 μCi / ml medium for 1 hour. Next, the supernatant is collected and
The cells are washed three times with ice-cold PBS and RIPA buffer (150 mM Na
Cl, 10 mM EDTA, 1% Triton X-100, 0.5% Na ⁺ Deoxycholate, 0.1% SDS, 50 mM Tris (pH 8.0)
). Cell lysates were purified with anti-Sindbis virus rabbit antibody (1: 8
0 dilution) and immunocomplexes are isolated using Protein A Sepharose.
(Pharmacia). After washing the immune complex, the sump
And boil the protein and convert the protein to SDS-15% polyacrylamide (PA
GE) Separation by electrophoresis through a gel.

RIP assays of transfected cell lysates revealed similar findings to protein immunoblot of cell lysates infected with stock virus. E2 at the size of the wild-type protein was detected in TEMluBcl transfected cells. A prominent band just below 68 kd was detected in the TEBcl transfected cells, along with the wild type size E2 band. The wild type E2 band was as expected for TEMluBcl transfected cells. This is because the heregulin sequence is approximately the same size as the removed Sindbis virus sequence. PE2 in TEMluBcl transfected cells appears to be completely processed. As in the immunoblot above, the high molecular weight band (68k) in TEBcl transfected cells
D) can correspond to a PE2 or E2 glycoprotein containing a heregulin insert. Again, a band corresponding to wild-type E2 was found in TEBcl transfected cells. This suggests the presence of mixed infection with the wild-type virus, or removal of the heregulin sequence by RNA recombination. In wild-type infected cells, p110, a precursor protein composed of PE2 and E1,
Was larger in TEBcl transfected cells. This was consistent with the presence of heregulin in the larger precursor protein. T
TEMluBcl compared to PE2 in EMluBcl transfected cells
PE2 in transfected cells did not appear to be completely processed. This was most likely the result of a heregulin insert.

[0136] Together with protein immunoblots, these data indicate that alphaviruses can be made despite the removal of the 68 amino acid domain and replacement with heterologous sequences. This data also shows that alphaviruses with insertion of a larger heterologous sequence (about 200 bp) into the E2 glycoprotein can be created.

Example 10: Infection of BHK cells and MDA-MB-231 cells and breast cancer cells with wild-type Sindbis virus or heregulin-containing Sindbis virus BHK cells and MDA-MB-231 cells (low level or Expressing undetectable levels of erbB-2 and erbB-4 receptors) with wild-type Sindbis virus or heregulin-containing Sindbis virus (ie, TEM)
Infections were made with luBcl (heregulin replaces E2 amino acid residues 175-243) or TEBcl (heregulin sequence at amino acid 243) at a multiplicity of infection of 5. At various times post-infection, the amount of cell death as a result of viral infection was measured by trypan blue exclusion. For BHK cells, separate infections with wild-type virus and two heregulin-containing viruses resulted in 40-45% cell death 24 hours post-infection. 48 hours after infection, all BHK cells were killed. At 43 and 68 hours post-infection of MDA-MB-231 cells, approximately three times as many cells were killed by the heregulin-containing virus as compared to the wild-type virus. If MDA-MB-231 cells have low or undetectable levels of erbB-
The fact that heregulin-containing viruses can kill MDA-MB-231 cells more effectively than wild-type viruses if they express the 2 receptor and the erbB-4 receptor is surprising and unexpected. Was.

Example 11 Transfection of BHK Cells and SKBR3 Breast Cancer Cells Using Sindbis Wild-Type and Sindbis-Modified Virus RNA Genomes Infectious RNA was cloned into clone 12 (TE SphI BstEII heregulin).
And clone 15 (TE BstEII MluI heregulin), transcribed in vitro and used Lipofectin to BHK cells and SK
BR3 breast cancer cells (American Type Culture Collec)
Tion (ATCC), Rockville, Maryland; erbB-2
Is overexpressed). The number of dead cells was determined 48 hours after infection by trypan blue exclusion assay (Current Protocols i).
n Molecular Biology, edited by Ausubel et al., John W.
iley & Sons, Inc. , 1996, Addendum 18, Unit 11.5.
1. Cell Viability Test by Trypan Blue
Exclusion).

SphI at amino acid residue 60 of E2, BstEI at amino acid residue 114 of E2 derived from viral TE cDNA and generated in a two-step PCR reaction
The parent virus, which contains three unique restriction enzyme sites for MluI at amino acid residue 175 of I, E2, was less efficient in SKBR3 cells but was able to replicate in both cell lines. 48 hours after transfection with the parent virus, BHK
80% of the cells were killed in the culture, whereas 20% of the cells were killed in the SKBR3 culture. Heregulin-containing virus, clone 12 (E
Heregulin replaces two amino acid residues 60-114) and clone 15 (E
Two amino acid residues 114 to 175 are replaced by heregulin), mock (mock)
) SKBR3 when compared to transfected cells (Lipofectin only) and cells transfected with the Sindbis virus replicon expressing β-galactosidase gene (expression vector without structural protein).
Only cells could be killed (almost the same level as TE), but BHK cells could not be killed. Heregulin-containing virus showed cytotoxic effects only in SKBR3 cells.

Example 12 Clone 12, TE BstEII MluI Heregulin (
Evaluation of intracellular synthesis of alphavirus structural proteins by clone 15) and LKD102.10 (from parent virus TE) Clone 12, clone 15 and parent virus were pulse-labeled and transfected cell lysates were immunoprecipitated. . The presence of the capsid protein
It was easily detected in all samples, but decreased as compared to the wild-type virus TE. The band with a molecular weight slightly larger than that of wild-type E2 was clone 1
2 for transfected cells. This band, slightly larger than wild-type PE2, was most likely to represent PE2 due to the presence of heregulin. The putative PE2 for clone 15 was hardly detectable. This precursor appeared to be slightly smaller than wild-type PE2, despite the fact that an extra 8 amino acids were introduced into heregulin. Presumably, the faster migration was also due to folding / processing differences when compared to wild-type PE2. E1 was difficult to evaluate. Because it is a mock
) Because it overlaps the band present in the transfected cells. Wild type E2
Size E2 was transfected with parental RNA, and clone 1
Detected in cells transfected with 2 or clone 15. An increase in the size of E2 of 1500 daltons for clone 12 and 900 daltons for clone 15, as a result of the presence of heregulin, was this 15%
It was not degraded by separation of the protein in the gel.

[0141] These data demonstrate the synthesis of viral structural proteins in cells transfected with the recombinant alphavirus. The amount of structural protein produced was lower compared to wild-type virus TE. This indicates a defect in the step (s) of the viral replication cycle (assuming similar transfection efficiencies).

Example 13 Selectivity of Clone 12 Clone 12 RNA was replaced with a defective helper RNA, DH26S (Sondra).
(Obtained from Schlesinger) and BHK cells at a ratio of 1: 1 in co-electroporate. Approximately 5 × 10 ⁶ BHK cells were electroporated and seeded on 10 cm tissue culture plates in 10 ml of medium. The post-transfection supernatant was harvested 32 hours after electroporation, clarified to remove cell debris, and 200 μl was placed on a BHK cell monolayer in a 35 mm well at 37 ° C. for 1 hour. After infection, 1.5 ml of medium was added to the culture and the cells were incubated at 37 ° C for 26 hours. About 26-70% of the cells showed cytotoxic effects (CPE), 26 hours after infection, the supernatant was collected,
It was clarified and cell debris was removed. Cells were lysed using RIPA buffer. BHK cells or SKBR3 cells were infected with 200 μl of these supernatants at 37 ° C. for 1 hour. Twenty-seven hours after infection, supernatant and cell lysate from the second passage were collected as described above. CPE was observed during the second passage of clone 12 (complemented with helper virus RNA before the first passage). 27 hours after infection with supernatant from BHK cells infected with parental viral RNA, B
Approximately 80% of HK cells show CPE, whereas 90% of SKBR3 cells
Showed CPE. First of complemented clone 12 virus in BHK cells
Infection of cells with the supernatant from passage 2 resulted in 10% of BHK cells and 70% of SKBR3 cells displaying CPE. These results suggest the presence of clone 12, a heregulin-containing virus that infects erbB-2 overexpressing breast cancer cell lines. Western blots of cell lysates from two passages in BHK cells revealed the presence of heregulin-containing E2 in the BHK cell lysates of the first and second passages. The intensity of the lane showing BHK cells infected with the supernatant from the first passage of clone 12 complemented in BHK cells (second passage) is the clone 12 virus complemented with wild-type E2 (clone 12 virus). The intensity of the same band in the lane representing BHK cells infected with helper RNA (result of co-transfection of clone 12) (first passage). This is probably due to the presence of wild-type E2 in the complemented virus, resulting in more effective infection of BHK cells during the first passage. Also, the extent of CPE was less than at the time of BHK cell harvest at the second passage. Furthermore, the actual multiplicity of infection of the first and second passages is unknown. However, it is clear that the presence of heregulin-containing E2 in the cell lysate of the second passage indicates that the heregulin-containing virions are made and have the ability to infect and kill cells.

Example 14: Specificity of Targeted Recombinant Sindbis Virus Binding Encoding CAT and anti-erbB-2 scA, α-heregulin,
In vitro transcribed RNA from Sindbis virus containing unmodified, irrelevant scA or irrelevant ligand is separately infected into BHK-21 cells using Lipofectin, and progeny virions are converted to ³⁵ S-methionine. And ³⁵ S-cysteine in a medium (Tran ³⁵ S label, IC
N, Irvine, California). Radiolabeled offspring virions,
Precipitate from the supernatant 24 hours after infection using 10% polyethylene glycol 8000 in 0.5 M NaCl and purify by passing through a 15-40% potassium tartrate gradient. Purified radiolabeled virus was combined with binding medium (RPM without NaHCO ₃ , 0.2% BSA, 20 mM HEPE).
S, pH 7.4) and stored at -80 ° C. The number of infectious units per ml was determined in the presence of actinomycin D (a sufficient dose to inhibit cellular RNA polymerase) and ³ H-uridine (resulting in incorporation of ³ H-uridine into viral RNA). As measured by virus growth in a permissive breast cancer cell line. ^The supernatant containing the ³ H-uridine-labeled viral RNA was TCA precipitated,
Count in a liquid scintillation counter and calculate the percentage of ³ H-uridine incorporated into the viral RNA and compare to a labeled wild-type virus culture with a known titer.

A predetermined number of infectious units of test and control vectors are separately added to cell monolayers in the absence or presence of appropriate increasing dilutions of anti-erbB-2 or GST α-heregulin fusion protein. And bring the cell monolayer to 4 ° C.
(Specific virus binding to the cells but not into the cells) to determine the specific binding of the virus. erbB-2 or e
Specific binding of virus overexpressing rbB-2 and erbB-4 and SKBR
3 breast cancer cells, MDA-MB-453 breast cancer cells (ATCC), MDA-MB-36
1 and BT474 breast cancer cells (ATCC), erbB-2 non-overexpressing breast cancer cell lines MCF-7 (ATCC) and MDA-MB-231, cell lines known to be permissive for Sindbis virus infection (I.e., BHK-21 cells and N18 neuroblastoma cells), and cell lines known to be not permissive for Sindbis virus infection (i.e., IB4 (EBV-transformed B lymphocyte cell line)). use.

[0145] Supernatant was prepared with the appropriate virus, virus and mAb, or virus and α
-After the addition of heregulin, harvest at 30 minute intervals, count in a liquid scintillation counter and determine the amount of unbound virus. After washing. 1% cells
Lyse with SDS and count in a liquid scintillation counter to determine the amount of virus bound.

Example 15: Specific infectivity of targeted recombinant Sindbis virus SKBR3 cells, MDA-MD-453 cells, MDA-MD-361 cells (A
TCC), BT474 cells (ATCC), MCF-7 cells, MDA-MB-23
CAT or Green Fluorescent Protein (GFP, Clonetech, Palo Alto, C
alifonia) and anti-erbB-2 scA or α-
Specific infectivity by Sindbis virus containing heregulin, unmodified, irrelevant scA or irrelevant ligand, where appropriate anti-erbB-2 or GST
Determined by assaying for CAT activity in antibody dilutions that maximally inhibited virus binding in the binding assay in the presence or absence of the α-heregulin fusion protein. The cells are incubated with the vector and mAb for 1 hour at 37 ° C., the vector or vector and antibody combination is removed, the cells are washed twice with phosphate buffered saline, and fresh growth media is added. Six hours after infection, the cells are harvested and assayed for CAT activity using standard techniques (using a fluorescence microscope if GFP is used,
Detect its expression). The cells are lysed by three consecutive freeze-thaw cycles. The lysate is centrifuged and the supernatant is assayed for CAT activity. Since high levels of CAT expression were achieved with TotoCat5, the cell extract was:
Usually, a dilution of 1,000 or more is required to be within the linear range of the assay. [ ¹⁴ C] chloramphenicol (0.4 μCi), acetyl-CoA (0.5 m
M) and TrisHCl, pH 7.5 (0.25M) are added to the appropriate amount of cell lysate and incubated at 37 ° C for 1 hour. Chloramphenicol and acetylated chloramphenicol are extracted with ethyl acetate. The ethyl acetate is dried in a SpeedVac evaporator and the sample is resuspended in 30 μl of ethyl acetate. This sample is subjected to thin layer chromatography (TLC)
Spot on a sheet and develop it in a bath containing 19: 1 chloroform / methanol. Next, the TLC sheet is air-dried. The amount of monoacetylated chloramphenicol was determined using a phosphorimager (Molecular Dy).
quants).

In addition, the specific infectivity is determined by an indirect fluorescence assay. Test and control cells are grown on glass coverslips to 70% confluency and counted on a hemocytometer to determine the number of cells plated. Then
The cells are incubated with a predetermined number of infectious units of the targeted or non-targeted vector for 3 hours.
Infect for 1 hour at 7 ° C. Six hours after infection, the coverslips are washed twice with PBS (without calcium and magnesium) and then fixed with methanol at -20 ° C for 5 minutes. The methanol is removed and the cells are washed three times with PBS. The cells were then combined with blocking buffer (0.5% w /
v Gelatin / 0.2% w / v BSA) for 30 minutes at room temperature. The blocking buffer was removed and the primary antibody (ie, anti-Sindbis virus rabbit serum, anti-E2 mAb or anti-CAT rabbit serum (5′3
', Inc. , Boulder, Colorado)) at room temperature over 30 minutes. After washing three times with PBS, a secondary antibody (ie, fluorescein-conjugated anti-mouse or anti-rabbit IgG) is added at room temperature for 30 minutes. The coverslips are washed three times with PBS, air dried, and 2.5% DABC
O (1,4-diazobicyclo- [2,2,2] -octane (Sigma Che
m. Co. , St. Louis, MO)) containing Moviol 4-88 (
Mount on glass slides using Calbiochem) solution. Immunofluorescent staining is performed and cells expressing CAT or Sindbis antigen are counted.
Calculate the percentage of infected cells.

To determine the number of cells killed due to alphavirus infection, erb
B-2 expressing breast cancer cells and control cells are also infected with a predetermined number of infectious units of test and control vectors. Twelve hours after infection, the cells are harvested by incubating for 1-2 minutes in trypsin. After washing with PBS, an aliquot of the cells is diluted 5-fold in 0.2% trypan blue. The number of dead (ie blue) and viable cells is counted using a hemocytometer and the percentage of dead cells is calculated.

Example 16 Targeting Recombinant Sindbis Virus to Vascular Cells The vascular cell line SLK is a human Kaposi's sarcoma cell line that expresses the NGR receptor characteristic of cells in the vascular system. Arap, W et al., Science (1998)
279: 377-380 suggested targeting the tumor vasculature with the "NGR" peptide. Many of these peptides are disclosed in this publication, which is incorporated herein by reference. Exemplary peptides described below are:
For illustrative purposes only, of course, other related peptides may also be used.

Nucleotide sequence 5′-TGC AAC GGT CGC TGT GTA
TCG GGA TGT GCC GGT CGA TGT-3 ′ encodes the peptide sequence CNGGRCCVSSGCCAGCRC. This nucleotide sequence is inserted between amino acid 58 of E3 and amino acid 6 of E2 (NGR1
) Or between amino acid 61 and 73 of E2 (NGR2). Sindbis viruses with these modifications in the envelope were generated. The resulting virus was then tested for cytopathic effects on either the SLK cell line, which expresses the NGR receptor on its surface, or the BHK cell line, which is normally infected with wild-type Sindbis virus. As shown in the table below, the NGR1 and NGR2 strains are specifically targeted to SLK cell lines and are not effective against BHK. As expected, wild-type Sindbis virus only infects BHK.

[0151]

[Table 4] Example 17 In Vivo Targeting of Recombinant Sindbis Virus to be Targeted SKBR3 cells (1 × 10 ⁷ ) were multiply infected with 5 targeted alphavirus or control virus vectors. Infect at 37 ° C for 1 hour in vitro. After infection, the cells were transferred to 6 week old Swiss athymic nude mice (Jackson Laboratories, Bar Harbor,
Maine) is injected into the mammary fat pad. Over 4-6 weeks, the tumors were observed for rate changes and for the extent of growth compared to tumors formed from controls (non-infected SKBR3 cells).

In a separate set of experiments, 1 × 10 ⁷ SKBR3 cells are injected subcutaneously into the mammary fat pad of 6-week-old Swiss athymic nude mice. One week after injection, after the growing tumor became visible (tumor size about 5 mm in diameter), 0.1 ml of the appropriate volume (based on in vitro studies) (targeted virus and The viral RNA (of the non-targeted virus) is injected into the center of the tumor. The effect on tumor growth is measured with a vernier caliper for 4-6 weeks and the tumor volume (tumor volume = (width) ² × length / 2) is calculated. At the end of 4-6 weeks, the tumors are analyzed histologically and immunohistochemically.

In addition, using another set of mice, 10 days after injection of the viral vector, tumors were excised, formalin fixed, embedded in paraffin, sectioned, and stained with hematoxylin and eosin, Look for tumor necrosis. A terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay is used to detect tumor cells that have undergone apoptosis as a result of targeted alphavirus infection. The TUNEL assay detects endonucleolytically cleaved DNA by the addition of labeled dUTP with a terminal transferase at the end of the DNA. Tissue sections were deparaffinized, rehydrated, and dried at 10 / ml.
μg of proteinase K (Boehringer Mannheim, India)
(Anapolis, Indiana) by incubating for 15 minutes at room temperature and washing twice in PBS for 5 minutes. This tissue section is
40 μl of the labeling mixture (25 mM Tris-HCl, pH 7.2, 0.2 M
Potassium cacodylate, 5 mM cobalt chloride, 30 U of terminal deoxynucleotidyl transferase, 0.6 nmol of digoxigenin-11-dUTP,
0.15 mM dATP) and incubate in a humid chamber at 37 ° C. for 60 minutes. The reaction is stopped by incubating in 2 × SSC at room temperature for 15 minutes. Probes are detected using an alkaline phosphatase conjugated anti-digoxigenin Fab fragment using the Genius 3 non-radioactive nucleic acid detection kit from Boehringer Mannheim. Blocking buffer (Buffer 1 (Boehringer Mannheim) and 0.1% Triton X-100 and 2% normal lamb serum (NLS,
GibcoBRL, Gaithersburg, Maryland)) is added to the tissue section and the section is incubated for 30 minutes. 100% of 2%
1:50 in buffer 1 containing NLS and 0.1% Triton X-100
A 0-diluted anti-digoxigenin antibody is added to the tissue section, and the section is incubated for 30 minutes at room temperature in a humid chamber. The BM Genius nucleic acid detection system for blue detection of alkaline phosphatase activity with 4-nitroblue tetrazolium chloride and X-phosphate is incubated for 10-30 minutes according to the supplier's protocol. The slides were washed, dehydrated, and mounted in Permount (Sigma Chemical Co., St. Louis).
s, Missouri). TUNEL assay controls are:
Includes omission of terminal transferase or labeled nucleotide.

[0154] Immunoperoxidase staining using standard techniques is also performed to detect the presence of the viral antigen in the excised tumor and to determine its distribution. Tissue sections are immunohistochemically stained with anti-alphavirus rabbit serum using the avidin-biotin peroxidase conjugate method. Sections are first blocked with 2% normal rabbit serum in PBS and then incubated with anti-alphavirus rabbit serum at 1/100 stock dilution for approximately 45 minutes. Put this slide in PBS
Rinsing in biotinylated secondary antibody (1:10 in 2% normal rabbit serum)
(0 dilution). The slide is then washed and treated with 1% hydrogen peroxide in methanol to block endogenous peroxidase. Then, avidin DH-biotin complex (Vector La) in PBS
borateries, Nurlingame, CA) followed by 0.01%
Diaminobenzidine (0.5 μg / ml, Sigma) in PBS containing hydrogen peroxide
a Chem. Co. , St. (Louis, MO). The stain is darkened with CuSO ₄ and 0.15 M NaCl. Sections are counterstained with hematoxylin and eosin and photographed with a light microscope.

Example 18 Method of Delivering a Targeted Alphavirus (TA) or Alphavirus Vector (TAV) to a Mammal According to the Methods of the Invention TAV is in a suitable packaging cell line It can be delivered as a virus after packaging. An example of a packaging cell line for the production of TAV used in the treatment of breast cancer is a breast cancer cell line that expresses a target protein or factor for TAV infection. However, any cell line that can produce high titers of virus can be used. The TA is then concentrated (eg, by filtration, ultracentrifugation or another method) and administered directly (eg, as a surgical aid), eg, at the site of a tumor or at the site of tumor resection.

TAV construct DNA or R transcribed in vitro into packaging cell lines
The result of NA transfection is TA particle generation. Use standard transfection reagents (eg, Lipofectin (for RNA), lipofectamine, lipofectamine plus, e
ffectene, and clofectin (for DNA)). If the packaging cell line contains a target protein or factor for TA infection, the TA particles generated from the transfected cells will infect neighboring cells, resulting in amplification of the TA particles. This virus is used for transfection
-96 hours, more preferably 12-72 hours after transfection (this will vary depending on the packaging cell line used),
The supernatant is clarified by centrifugation to remove cell debris.

The TA particles are then concentrated (eg, preferably by filtration or ultracentrifugation). If filtration is used, the virus is concentrated from non-viral proteins and factors present in the supernatant using a Centricon 100-500 concentrator. The use of Centricon concentrators is known to users in the art,
And can be found in any Centricon user manual. If ultracentrifugation is used, the virus is 39,000 using a SW rotor (Beckman).
Centrifuge at rpm for 30 minutes. The concentrated virus is then titrated against the indicator cell line. For example, a similar erb-B2 TAV expressing green fluorescent protein (GFP) was titered against a breast cancer cell line overexpressing erb-B2 and the virus titer was determined by fluorescence microscopy for GFP positive cells. Can be measured and determined.

Placing TA on breast cancer cells is another method that can be utilized to determine viral titers. A limiting dilution of TCA is infected into a monolayer of breast cancer cells and its titer is determined by testing the monolayer for viral plaques 48-72 hours later. The titer of the virus, preferably a TA particles / ml to about 10 ² to 10 ¹³ For clinical application.

In treating breast cancer, the concentrated TA particles can be resuspended in a suitable solution (eg, saline or tris buffered saline) and then applied to or at the tumor site. Or can be injected. In one situation, TA particles may be applied to a surgical tumor resection site as an adjunct to surgical therapy. TA particles not only infect and kill any of the remaining cells at the surgical site, but are also taken up by local lymph nodes and reduce the residual tumor cells present in the lymph nodes that escaped surgery. Infect either. Alternatively, the TA suspension can be injected directly into the tumor mass without the need for surgery. Multiple injections are made periodically (preferably once a week or once a month) into the tumor to reduce tumor mass and prevent metastasis.

When the target for TA is not the tumor cells themselves, but are feeder cells (eg, tumor angiogenic endothelial cells), the TA suspension may be, for example, at a concentration of greater than about 10 ¹⁰ TA particles / ml. Such concentrations can be injected intravenously. The intravenous virus then binds to the angiogenic endothelial cells, replicates in the cells, and kills the cells. The virus produced from the first round of endothelial cell infection then spreads to adjacent tumor angiogenic endothelial cells and kills them as well, thus killing the cells lining the blood vessels that feed the tumor. I do.

The format for TAV delivery can be one of two types— (a) direct nucleic acid delivery to or at the tumor site, where the nucleic acid produces TA that propagates through the tumor Obtain) or (b) delivery of the nucleic acid to a packaging cell line (or to act as the patient's own irradiated tumor cell-packaging system), followed by delivery of cells containing TA to the patient. The number of cells introduced into the patient is preferably
About 10 ³ to about 10 ¹⁴ cells / inoculum. Multiple inoculum may be required and may be administered to the same or different sites.

When TAV is delivered as a nucleic acid, it can be delivered as either RNA or DNA. When delivered as RNA, it must first be transcribed from a DNA template to make the RNA. Such R
In vitro transcription of NA is known in the art. The TAV plasmid construct is:
Reaction buffer (eg, 0.5 μg DNA, 20 U RNAsin, 2.5 mM d) containing a polymerase (eg, SP6 polymerase) to drive transcription
NTP, 100 mM DTT, 40 mM Tris-HCl, pH 7.5, 25
Prior to in vitro transcription in mM MgCl ₂ , 2 mM spermidine, 10 mM NaCl ₂ ), it is linearized with appropriate restriction enzymes. This RNA is then converted to TAV R
For efficient transfection of cells with NA, they are complexed with liposome or non-liposome (collectively referred to herein as "lipoids") delivery systems. Such a lipoid delivery system is, for example, lipofectin complexed with other compounds or other proteins (such as polybrene or polyamine) to increase the efficiency and targeting of TAV transfection into cells. (Life technologies), Clonfectin (Qiagen)
), Superfect and Qiagen.

[0163] The TAV RNA enters the cytoplasm of a cell, which may be a target cell or a cell that is not a target cell. An example of a target cell is an erb-B2 overexpressing breast cancer cell to be killed by this TA / TAV. Examples of non-target cells are packaging cell lines, cells in tissue surrounding a resected tumor, lymph node cells, or endothelial cells. Once this RNA enters the cell's cytoplasm, it amplifies itself and translates its genome, producing TA particles. From this first round of replication, TA particles are produced that specifically infect, kill, and spread through the population of tumor cells. Cells transfected with TAV RNA
It is important to note that virus production can kill. Killing occurs when the lytic production of the virus kills that particular cell type, or when the cell is destroyed by the host immune system as a result of the expression of viral proteins on the cell surface.

In certain cell types, the virus can be produced chronically, so that TA / TAV
Production occurs for an extended period of time (Griffin and Hardwick, Annual
Review of Microbiology (1997) 51: 565-9
2 and Dryga et al., Virology (1997) 228 (1): 74-8.
3). Thus, TA / TAV produces cytolytic infections to infect certain cell types, such as non-target cells, to produce virus persistently from non-target cells, and at the same time in target cancer cells. Can be modified for

In an alternative approach, TA / TAV can be modified to persistently infect cancer cells and not induce lytic cytopathology as a result of viral replication. Cell death can then be induced by expression of the virus or a heterologous protein or genetic element. The protein can be a cell death protein or a ribozyme that destroys anti-apoptotic protein mRNA. Alternatively, the heterologous protein may be expressed on the surface of a cancer cell, which may then be targeted for destruction by the host's immune system.

The heterologous protein may be encoded by a gene that induces or prevents cell death, or its cognate antisense or ribozyme molecule, or a non-host antigen,
It can be an immunogen (widely defined as an antigen, cytokine or antibody or part thereof) that promotes target tumor cell destruction by the immune system, such as a cytokine or scAb. An example of a non-host antigen is the 4-1BB ligand (Melero et al., European Journal of Immunology (1998).
) 28 (3): 1116-21). Examples of cytokines include GM-CSF (Dun
ussi-Joannopoulos et al., Blood (1998) 91 (1):
222-30). An example of a single chain antibody heterologously expressed in cells is erb-B2 (
Thirion et al., European Journal of Cancer P
revention (1996) 5 (6): 507-11; and Zbar et al.
British Journal of Cancer (1998) 77 (5):
683-93).

The heterologous gene can be inserted into an alphavirus vector that contains a second subgenomic promoter for expression of the heterologous gene, such as a Sindbis virus vector. Thus, the heterologous gene inhibits tumor growth and / or
Or it can function to increase TA / TAV production. Optionally, the vector kills tumor cells in the host. Non-target cells producing the first round of virus can die from productive viral infection, but this is not critical. Because the number of transfected non-target cells is higher than the subsequent replication and amplification of TA / TAV (approaching 100%) through the target cell population, eg, tumor.
It is relatively small (1-20%), preferably (5-10%).

When the vector is introduced intravenously, the TA / TAV becomes lipoid-R
The NA complex can be further targeted to the tumor by being incorporated into liposomal or non-liposomal delivery system molecules, which more specifically target the tumor vasculature. For example, a peptide expressing an RGD domain that targets tumor vasculature is mixed with a lipoid-RNA complex to target TAV RNA to cells of tumor vasculature integrin molecules, specifically tumor angiogenic endothelial cells. (Ruos
lahti, Annual Review of Cell & Development
ental Biology (1996) 12: 697-715). Then, T
A / TAV infected tumor vasculature produces a first round of virus, which can inoculate TA particles from non-luminal endothelial surfaces into the cell mass of tumor cells. This tumor-specific TA then spreads from the tumor cells to the tumor cells, thus destroying the tumor.
Importantly, necrosis resulting from the destruction of tumor cells in proximity to the blood supply can produce tumor necrosis that prevents other tumor cells from receiving the necessary nutrients from the blood for survival.

The TAV RNA can be targeted to tumor vasculature by incorporating a peptide RGD ligand into a lipoid-RNA complex, where the TAV itself is the RGD domain of a tumor vasculature integrin molecule. Targeted. This translates from about 1 pM to about 100 mM, preferably about 1 nM, of the RGD peptide ligand.
This can be done by adding to the lipoid TAV RNA complex at a concentration of 約 1 μm. Therefore, this then becomes a two-layer system where the Lipoid R
Ligands in the NA complex initially promote one level of targeting to the tumor vasculature, and then a specific RGD domain on the virus is targeted by spreading and amplifying the virus through tumor vascular endothelial cells. Promote the second level.

An additional layer of specific targeting may be added to promote tumor specificity and low cytotoxicity. For example, this TAV is a DNA rather than an RNA-lipoid complex.
-Can be delivered as a lipoid complex. To enable this, a promoter must be placed in close proximity to the virus to ensure the first round expression of TA. This is because alphaviruses are RNA viruses without a DNA intermediate. One method is to transfect plasmid DNA directly into cells without the eukaryotic promoter system. Even if this plasmid DNA does not contain a eukaryotic promoter (but does contain the eukaryotic transcriptional potential sequence of the vector, such as the SP6 bacteriophage promoter), the virus will still be able to encode the alphavirus encoding true virus. It can be produced from cells transfected with a plasmid DNA deficient in a nuclear organism. The SP6 promoter TAV DNA can be transfected into cells using the lipoid transfection reagent described above at a concentration known in the art. For example, 5 μg of TAV is added to 20 μl for transfection into packaging cells.
Clonfectin (Quiagen), or can be mixed directly into the breast cancer surgical site. The transfected cells can produce a first round of virus, which is then amplified and spread into other tumor cells in a cell-specific manner.

One preferred method of producing a first round of virus from a DNA-lipoid complex is to use a TA comprising an SP promoter sequence proximal to the alphavirus genome.
Simultaneous expression of the SP6 bacteriophage polymerase gene in cells containing V plasmid DNA. This SP6 polymerase is expressed from a helper plasmid. The SP6 gene was derived from the pSR3 plasmid (Moss et al., Bi).
otechniques (1993) 14: 222-223), cloned into the pREP10 plasmid (Invitrogen, CA) or expressed from a recombinant vaccinia virus. This SP6 plasmid can then be expressed in eukaryotic cells where the SP polymerase can bind to the SP6 promoter sequence located on TAV. TAV RNA expression may be the result of translation of the alphavirus NS RNA-dependent RNA polymerase gene.
The polymerase can then bind to the TAV RNA promoter sequence and amplify the TAV genome in the cell and produce TA particles. These particles then infect and spread through the target cell population--tumor cells or cells that support their growth.

The system can be further improved by driving SP6 expression from a tumor-specific promoter, or a tumor capillary endothelial cell promoter. The latter promoter is preferred. Because it can inhibit expression in cells exposed to blood circulation other than endothelial cells. A promoter that is more specifically expressed in endothelial cells is the factor VIII promoter. Another promoter is the gamma glutamyl transpeptidase promoter (Dropulic et al., In V.
Intro Cellular & Developmental Biology (
1987) 23 (11): 775-81).

Another method of producing a first round of virus from DNA-lipoid TAV is as follows:
By linking the TAV plasmid DNA sequence directly to a eukaryotic promoter (Dubensky et al., Journal of Virology (1996)
) 70 (1): 508-19). For example, the alphavirus genome is cut from the plasmid vector and the pcDNA3.1 Amp expression vector (Inv
itrogen), where the alphavirus genome is
Located distal to the cytomegalovirus promoter (CMV). First T
AV RNA is transcribed from CMV-driven plasmid DNA and then TAV
RNA-dependent RNA polymerase binds to the transcription initiation site on TAV RNA and amplifies the amount of TAV RNA in the cell. To ensure that the correct 5 'TAV RNA terminus, which is important for RNA polymerase binding and transcription, is generated, a ribozyme can be inserted into the 3' region of the viral genome and C
Cleave the 5 'leader sequence between the MV transcription start site and the true 5' end of the TAV RNA. This ribozyme is linked to the 5 'end and
It is constructed to cut exactly at the first nucleotide at the NA 5 'end. TAV
Can be made more specific for tumor cells by using tumor-specific or vascular-specific promoters.

A preferred method of producing a first round of virus is to use a carrier vector (eg, a vaccinia vector, an adenovirus vector, or preferably a Dropul vector) driven by a suitable promoter for expression of TAV RNA in cells.
ic et al., crHIV vector as described in WO 97/20060).
By arranging the AV sequence. The first round of delivery can be achieved by delivery of the carrier vector to the target site, which can then produce the first round of targeted alphavirus particles, which can then specifically infect the target cell population .

A preferred method for delivery of TCA / AV is in a crHIV vector. c
Since the size of the rHIV genome is smaller than the alphavirus genome, this TC
A / AV can be separated into two components containing sequences that allow their recombination. The non-structural (NS) region, the J (joining) region, and a limited number of downstream nucleotides (if necessary, non-structural TAV RNA and structural (S) TAV
(Which allows efficient recombination between RNA) can be amplified by PCR using appropriate primers containing restriction sites for cloning into a crHIV vector. This PCR product was prepared using Vent polymerase (New Engl).
and Biolabs) to prevent mutations from being inserted into the PCR product. The PCR product is then cut with restriction enzymes, run on a gel, and the appropriate bands are purified from the gel using a Quiagen gel extraction kit (Qiagen). The cleaved PCR product is then ligated into a crHIV vector within the polylinker region of the vector. At the same time, the structural region of the TAV genome containing the modified envelope protein targeted to cancer cells, the J region, and a limited number of upstream sequences (if necessary, between the nonstructural and structural TAV RNAs). (Which allows for efficient recombination) can also be amplified by PCR and cloned into a crHIV vector. TA in two crHIV vectors
The V sequence is a eukaryotic promoter (eg, CMV, a tumor-specific promoter,
Or a vascular-specific promoter) or can be expressed directly from the crHIV long terminal repeat (LTR), which acts as a promoter in this vector system. The two crHIV-TAV (NS and S) constructs can be co-transfected with a helper construct into a packaging cell system, allowing production of recombinant crHIV-TAV. Since HIV encapsulates two genomic RNAs in its viral particles, some particles may contain crHIV-TAV.
It contains both the NS and S genomes, which can then infect cells.
During the reverse transcription of crHIV RNA, the NScrHIV-TAV genome and ScrH
Recombination between the IV-TAV genome can occur, resulting in the entire TAV genome being integrated into the host cell's genome. This TAV can then be expressed off the eukaryotic promoter, producing the initial TAV RNA, which is then amplified using Sindbis virus nonstructural RNA-dependent RNA polymerase. The targeted alphavirus particles then specifically infect tumor cells and spread by receptor-specific mechanisms.

The most preferred method for TAV delivery is to overlay the two component crHIV-TAV in a vector-helper crHIV direct delivery system. crHIV
Upon infection of the vector-helper DNA-lipoid complex, a first round of binary crHIV-TAV particles can be produced for infection of the cells. Two-component crH
Cells infected with IV-TAV can produce cells containing the targeted alphavirus provirus, which can continue to produce targeted alphavirus particles for specific infection and killing of the target cell population. Promoter / vector permutation (perm
arrays can be used to specifically produce TAV from specific cell types. This may assist in the generation of highly specific TA / TAV.

[0177] All references cited herein, including patents, patent applications, and publications, are incorporated herein by reference in their entirety.

Although the invention has been described with emphasis on the preferred embodiments, it is understood that modifications in this preferred embodiment may be prepared and used, and that the invention be described in detail herein. It will be apparent to those skilled in the art that other implementations are possible. The present invention is intended to include such modifications and alternative implementations. Accordingly, the present invention includes all modifications that fall within the spirit and scope of the present invention.

[Brief description of the drawings]

FIG. 1 is a diagram of the Sindbis RNA genome.

FIG. 2 is a map of the alphavirus expression vector TotoCat5 double subgenomic vector (top) and a single subgenomic vector (bottom).

FIG. 3 is a schematic diagram of portions of two major neutralizing epitope / receptor binding domains and three introduced restriction enzyme sites.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) C12N 7/00 G01N 33/48 M 4C087 C12Q 1/02 (C12N 7/00 G01N 33/48 C12R 1:92 ) // (C12N 7/00 (C12Q 1/02 C12R 1:92) C12R 1:92) (C12Q 1/02 C12N 15/00 ZNAA C12R 1:92) A61K 37/02 (81) Designated country EP (AT) , BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ, CF, CG, CI, CM) , GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, SD, SL, SZ, UG, ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, CU, CZ, DE, DK, EE, ES, FI, GB, GE, GH, GM, HR, HU, ID, IL, IS, JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS, LT , LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL, TJ, TM, TR, TT, UA, UG, US, UZ, VN, YU, ZW (72) Inventor Dropuric, Lisha United States 21042, Ellicott City, Golden Oak Drive 12637 (72) Inventor Hardwick, J. Marie United States Maryland 21212, Baltimore, Midhurst Road 2000 F-term (reference) 2G045 AA26 CB02 FB03 FB07 FB12 GC15 4B024 AA01 BA32 BA45 CA04 CA07 CA11 DA03 DA20 EA02 EA04 HA01 4B063 QA01 QA19 QQ01 QQ08 QR33 QR35 QR59 QR79 QR80 QS05 QS33 QX02 QX07 4B065 AA90X AA95X AA95Y CA24 CA25 CA44 CA46 4C084 AA02 AA13 CA01 CA53 DC50 NA14 ZB261 4C087 AA01 AA02 BB65 BC83 CA12 NA14 ZB26

Claims (21)

[Claims] 1. A recombinant alphavirus comprising an epiviral non-alphavirus peptide embedded in its envelope as a fusion protein with an envelope protein, wherein said peptide is located on a surface on a target cell. An alphavirus that specifically binds to its receptor. 2. The alphavirus according to claim 1, wherein the fusion protein is at least a part of an E2 glycoprotein and / or an E3 glycoprotein, or at least a part of an E1 glycoprotein and / or an E2 glycoprotein. Alphavirus fused to 3. The alphavirus according to claim 2, wherein the fusion protein substitutes at least some of the amino acids corresponding to amino acids 60 to 243 of E2 of Sindbis virus, or Sindbis. An alphavirus that is constructed by inserting the peptide between amino acids corresponding to amino acids 60-243 of E2 of a bisvirus. 4. The alphavirus of claim 2, wherein the fusion protein replaces at least some of the amino acids corresponding to amino acids 58 to 6 of E3 of Sindbis virus. Or an alphavirus constructed by substituting at least some of the amino acids corresponding to amino acids 73 of E2 to amino acids 73 of E2 of Sindbis virus. 5. The alphavirus of claim 1, further comprising an expressible nucleotide sequence encoding a further heterologous protein. 6. The alphavirus of claim 5, wherein the nucleotide sequence is under the control of a subgenomic promoter of the alphavirus. 7. The alphavirus according to claim 6, wherein the heterologous protein is a toxin protein or a reporter protein. 8. The alphavirus according to claim 1, which is a Sindbis virus. 9. The alphavirus according to any one of claims 1 to 8, wherein said epiviral non-alphavirus peptide specifically binds to a cell receptor characteristic of cancer progression. . 10. The alphavirus according to claim 9, wherein said receptor is an erbB-2 receptor or an NGR receptor. 11. The alphavirus according to claim 10, wherein said receptor is erbB-2 and said non-alphavirus peptide is an EGF domain of α-heregulin, or a simple binding to erbB-2. Containing a long chain monoclonal antibody, or wherein said receptor is an NGR receptor and said peptide is NG
Alphavirus which is an R peptide. 12. A recombinant alphavirus nucleic acid molecule comprising a non-alphavirus nucleic acid sequence in a nucleotide sequence encoding an E1 glycoprotein and / or an E2 glycoprotein or an E2 glycoprotein and / or an E3 glycoprotein. A nucleic acid molecule comprising a non-alphavirus nucleotide sequence that encodes a peptide that binds directly and selectively to a cell-specific surface receptor on a target cell. 13. The nucleic acid molecule of claim 12, wherein the alphavirus is a Sindbis virus. 14. The nucleic acid molecule according to claim 12 or 13, wherein
Recombinant DN capable of transcribing the alphavirus RNA genome or a part thereof
A nucleic acid molecule that is an A vector. 15. The nucleic acid molecule according to claim 12 or 13, wherein
A nucleic acid molecule that is an isolated alphavirus RNA or a portion thereof. 16. A nucleic acid molecule according to any of claims 12 to 15, wherein said peptide binds to a cancer cell receptor. 17. A method for treating cancer in a mammal, the method comprising administering to the mammal an effective amount of the recombinant alphavirus of claim 9, wherein the method comprises administering to the mammal an effective amount of the recombinant alphavirus of claim 9. When binding to cell surface receptors that exhibit the properties of progression of
Infecting cells presenting said receptor with said recombinant alphavirus and lysing said cells, thereby treating said cancer. 18. A method for treating cancer in a mammal, comprising administering an effective amount of the nucleic acid molecule of claim 16 to the mammal, the method comprising: The recombinant alphavirus infects and lyses the cells presenting the receptor upon binding to a cell surface receptor that exhibits the properties of
Thereby treating said cancer. 19. A method for detecting a target cell, the method comprising: contacting a composition suspected of containing the target cell with the recombinant alphavirus of claim 7; Detecting production, wherein detection of the reporter protein indicates the presence of the target cell. 20. The method according to claim 19, wherein said reporter protein is a green fluorescent protein. 21. A method for identifying a target cell-specific cell surface receptor, comprising: combining a composition comprising said target cell with a number of different epiviral non-alphavirus peptides displayed in its envelope region. Contacting with a library of alphavirus vectors comprising; selecting a recombinant alphavirus of the library that binds to the target cells; characterizing target cells that bind to members of the library; and thereby targeting Identifying a cell-specific cell surface receptor.