WO1997033979A9

WO1997033979A9 - Transgenic model of hepatitis c virus infection

Info

Publication number: WO1997033979A9
Application number: PCT/US1997/003939
Authority: WO
Filing date: 1997-03-13
Publication date: 1997-11-27

Abstract

A transgenic mammal whose hepatocytes express one or more of the HCV proteins C, E1, or E2.

Description

TRANSGENIC MODEL OF HEPATITIS C VIRDS INFECTION

Statement as to Federally Sponsored Research This invention was made with Government support under R01CA63117 and R01CA54524 awarded by the National Institutes of Health. The Government has certain rights in the invention.

Background of the Invention Primary hepatocellular carcinoma (PHC) is one of the leading causes of death from cancer worldwide. In Southeast Asia and Sub-Saharan Africa, chronic hepatitis B virus (HBV) infection is by far the single most important factor associated with the development of hepatocellular carcinoma. In contrast, the prevalence of PHC, parallel to that of HBV infection, is relatively low in North America and Northern Europe. The recent cloning of hepatitis C virus (HCV) has led to an intense epidemiological investigation of the causative role of HCV in various liver diseases, including PHC. The results of these preliminary studies have clearly linked HCV infection to the subsequent development of PHC in Japan, Africa, and Europe. Reports regarding the prevalence of anti-HCV antibodies in the United States varied between 10 and 40%. The cloning of hepatitis C virus (HCV) has provided an important insight into the viral pathogenesis of hepatocellular carcinoma. The virus is a positive- stranded RNA virus containing a genome of approximately 9,400 nucleotides. This sequence codes for a single, large polyprotein of approximately 3,010 amino acids.

The structural organization and computer sequence comparison indicate that HCV is a distant relative of the flaviviruses and pestiviruses. The 5' untranslated region of the HCV virus shares significant homology (45- 49%) with the equivalent region of pestiviruses and is highly conserved (>95%) among all the HCV isolates. Like the flaviviruses and pestiviruses, HCV encodes a large polyprotein precursor through which individual viral proteins are processed co-translationally and/or post- translationally through the combined actions of host- and viral-encoded proteases.

The viral structural proteins are encoded by the N-terminal portion of the genome, followed by the nonstructural genes (NS2-NS5) . Computer sequence comparison and comparative hydropathy plot indicates that the structural proteins are divided into the core capsid protein (C) , the envelope protein (EN1) , and a third protein probably representing a second envelope protein, designated as EN2/NS1 (Houghton et al., 1991, Hepatology 14:381-388; Hijikata et al., 1991, 8£.5547-5551; Weiner et al., 1991, Virology 18.0:842-848).

Using various protein expression systems, the C, EN1 and EN2/NS1 glycoproteins have been shown to have molecular weights of 19 Kd, 33 Kd, and 72 Kd, respectively (Harada et al., 1991, J. Virol. 65:3015- 3021) The functions of NS2 and NS4 proteins are unknown at present. Their predicted structures are highly hydrophobic and probably represent membrane-associated proteins. The NS3 protein is probably a helicase/protease, and the NS5 gene, which encodes a large polypeptide, may function as an RNA-dependent RNA polymerase during viral replication.

A considerable amount of information regarding the HCV nucleotide sequences has accumulated. Sequence comparisons among various HCV isolates reveal significant sequence heterogeneity. Based on this sequence information, three basic groups of HCV strains have been classified (Houghton et al., 1991, Hepatology 14:381- 388) . Several regions of variable and hypervariable sequences have been identified in the putative envelop proteins EN1 and EN2/NS1 (Weiner et al., 1991, Virology 180:842-848; Hijikata et al., 1991, Biochem. Biophys. Res. Comm. 175:220-228) . The hypervariable region in the envelope protein may be a consequence of a strong selection pressure on a protective epitope(s) of either B or T cells. This is reminiscent of the hypervariable neutralizing domain in the V3 loop of the envelop protein of human immunodeficiency viruε-1 (HIV-1) . This speculation, if confirmed, will provide important information regarding virus-receptor interaction, host- immunological responses and most importantly, the future development of a vaccine against HCV.

In contrast to retrovirus infection, the HCV virus replicates through an RNA intermediate (Choo et al., 1989, Science 244:359-362; Houghton et al., 1991, Hepatology 14.:381-388) . There is no evidence at present that the HCV genome can integrate into the host genome through a reverse transcription mechanism. The observed chronicity associated with HCV infection cannot be explained by latency through the above mechanism. Similar to flaviviruses and pestiviruses, HCV may infect mononuclear cells or other cell types. These cells may be important as a source of viral transmission or as a reservoir for latent infection. In addition, lymphoid cells infected by HCV may modulate the host immune response to HCV. Since HCV is an RNA virus and replicates without proofreading capacity, significant sequence diversity can result. This virus is therefore capable of mutating rapidly to evade host immune surveillance leading to chronic and persistent infection.

Approximately 50% of patients infected with HCV progress to a chronic form (Dienstag et al., 1986, Sem. Liv. Dis. e>:67-81). Chronic infection with HCV is often associated with severe liver disease, cirrhosis and eventually hepatocellular carcinoma. The presence of cirrhosis in most cases of HCV-associated PHC suggests that HCV probably plays an indirect role in hepatocarcinogenesis. The chronic injury model is likely the key step in the pathogenetic mechanism.

Summary of the Invention We developed transgenic mice expressing all three structural components, i.e., C, El, and E2 proteins, because we believed that together these proteins mediate the damage caused either directly by HCV infection, or indirectly by the action of cytotoxic T cells. The genetic constructs encoding these foreign proteins are not infectious in people.

The model of the invention is less expensive than larger primate models, and offers the key advantage of a carefully studied immunologic and genetic animal system. Transgenic expression is a cheap and safe method for testing therapeutic compounds, and provides a unique opportunity to assess individual genetic strategies for treatment.

Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

Detailed Description The drawings will first be described. Drawings

Fig. 1 is a schematic representation of the genomic organization of the hepatitis C virus. The encoded structural proteins are represented by "C", "El", and "E2"; non-structural proteins are represented by "NS2" through "NS5".

Fig. 2 is a map of plasmid pALBHCV Fig. 3 is a map of plasmid pMUPHCV. Fig. 4 is a set of three sequences showing the PCR primers used to amplify HCV genes. Fig. 5 is a Western blot analysis of HCV core protein expression in liver.

Fig. 6A and Fig. 6B are pairs of photographs of ethidium bromide-stained gels, and illustrate RT-PCR analysis of HCV mRNA expression in liver. Total RNA was harvested from livers of Tg.MC4 at indicated weeks of age (Fig. 6A) and Tg.AC lines (Fig. 6B) . PCR for HCV (upper photograph in Fig. 6A and Fig. 6B) and /3-actin (lower photograph in Fig. 6A and Fig. 6B) were performed after first strand synthesis using an oligo(dT) primer. Fragments were separated on a 1.2% agarose gel and stained by ethidium bromide. Ten, one, and 0.1 pg of the plasmid pMUCHCV were used as template for positive controls. Duplicate mice are shown at each time point in the Tg.MC4 mice. Control reactions lacking reverse transcriptase were negative in all circumstances (not shown) .

Fig. 7 is a Northern blot analysis of HCV mRNA expression in liver. PolyA RNA of liver in Tg.AC lines was separated on a 1% agarose gel containing formaldehyde and transferred to nylon membrane. It was probed by a ³ P-labelled HCV probe. The same membrane was then re- probed by a ³²P-labelled GAPDII probe to control for gel loading. Fig. 8 is a Western blot analysis of HCV core protein expression in liver. Liver extracts of Tg.MC4 and Tg.AC lines were separated on an 15% SDS- polyacrylamide gel and transferred to PVDF membrane. It was probed using an anti-HCV core monoclonal antibody. It was then re-probed using an anti-actin antibody to control for gel loading. Controls included lysates of insect cells expressing HCV proteins (Bac+) , and insect cells expressing 0-glucuronidase (Bac-) .

Fig. 9 is a series of twelve photomicrographs illustrating immunohistochemical analysis of HCV proteins in liver. The livers of Tg.MC4 (A, D) , Tg. AC1 (B, E, F, H-L) and wild type (C, G) were fixed and sections were stained using rabbit anti-core antibody (C-F) , rabbit anti-E2 antibody (G-L) , and rabbit preimmune serum (A, B) . Photomicrographs were obtained with magnification of 200X (A-D, G) , lOOx (E, F, H-K) and 800X (L) . Core antibody staining was predominantly cytoplasmic although the arrows in 9F indicate positive core antibody staining in occasional nuclei compared with negative core staining in adjacent nuclei.

Cloning and Sequencing of the HCV Structural Genes The genomic organization of HCV is shown in Fig. l. The structural genes cluster around the 5' region of the viral genome. Comparative hydropathy plots and computer prediction of secondary structure of these gene products have defined the junctions and domains of these proteins (Kato et al., 1991, Proc. Natl. Acad. Sci., USA 87:9524-9528; Takamizawa et al., 1991, J. Virol. .65:1105- 1113) . At least three structural proteins can be predicted from the sequences: core capsid (C) , ENl, and EN2/NS1.

An HCV cDNA clone, type lb from a Japanese patient with chronic hepatitis, was generously provided by Dr. Kunitada Shimotohno, National Cancer Center Research Institute, 5-1-1, Tsukiji, Chuo-ku, Tokyo 104, Japan. We cloned a fragment containing the entire core, El and E2 sequences (Stul-Tthllll fragment - GenEMBL HPCJCG nucleotides 280 to 2834) into the EcoR I and Stu I sites of pFASTBac 1 (Gibco-BRL Gaithersburg, MD) . HCV cDNA clone can be obtained from any of a number of available strains, using standard methods.

This baculovirus construct was used for high level expression HCV structural proteins; infected SF9 cells were shown to express HCV proteins of appropriate size. This cell lysate was used as protein control in subsequent studies.

Preparation of Constructs for Generating Transgenic Mice We placed the portion of the HCV genome that encodes the structural proteins into the multiple cloning site of a pCDl expression vector, which is a derivative of CDM8 (Aruffo et al., 1987, Proc. Natl. Acad. Sci. USA,

M:8573-8577) . we ligated the vector and the HCV fragment via an Accl restriction site (at nt 308) 5' to the AUG initiation codon for the HCV polyprotein and a Hindlll site in the linker sequence 3' to the putative COOH-terminus of processed EN2/NS1. The HCV fragment spans 2210 bp. The individual structural proteins can be separately cloned into the expression vector using PCR to introduce start and stop codons at the presumed boundaries of each protein. The boundaries of each of the structural proteins, the genes that encode them, and the PCR primers designed to amplify each gene are shown in Fig. 4. The N-terminal primers are designed with sequences conforming to Kozak's rules (Kozak, 1984, Nucleic Acids Res. 12.:857-872) , and an in-frame AUG codon (Primers 1, 3, and 5). The C-terminal primers (primers 2, 4, and 6) encode two consecutive TAG stop codons. In order to prepare DNA constructs that can be used to generate transgenic animals, the HCV cDNA was placed under the control of the murine albumin enhancer/promoter (Fig. 2; plasmid pALBHCV) and under the control of regulatory sequences of the Mouse Urinary Protein (MUP) promoter region (Fig. 3; plasmid pMUPHCV) . The DNA fragments used in the transgenic experiments were obtained by digesting pALBHCV with Nhel and Nsil to remove the pGEM7 sequences, and by digesting pMUPHCV with Aatll to remove the pSVSport vector sequences. The rationale for selecting these regulatory elements, particularly the MUP enhancer/promoter, is discussed below.

Alternative means of directing expression of HCV proteins in the liver could include, but are not limited to, placing the HCV sequences under the control of liver- specific promoters such as those associated with the genes for fatty acid synthase, glutamine synthetase, apoVLDL-II, apolipoprotein E, human apolipoprotein E/C-I, apolipoprotein C-III, metallothionein-I, transthyretin, apolipoprotein B, and alpha-1-antitrypsin. HCV sequences could also be placed under the control of inducible promoters.

Our initial transgenic studies investigated mice bearing all three of the structural components of the HCV genome, but it would be feasible to omit one or more of these components in order to study individual genes. Similarly, any or all of the components could be expressed as mutants that contain a substitution or deletion of one or more amino acids. Generation of Transgenic Mice

Transgenic mice were made using the Aatll-Asel fragment of pMUPHCV and the Pvul-Nsil fragment of pAlbHCV. Fertilized oocytes from FVB inbred mice (Taconic; Germantown, NY) were microinjected at the single-cell stage and candidate transgenic offspring were obtained using standard methods. Transgenic mice containing the HCV transgenic constructs were identified by Southern blots using DNA isolated from individual mouse tails. Transgenic lines were produced by breeding heterozygous transgenic mice to inbred FVB mice.

The pups born to foster mothers were screened at three weeks of age by genomic Southern hybridization and PCR amplification in order to identify founder animals. Four MUP-HCV founders and eleven Alb-HCV founders were identified. RNA from the transgenic construct encoding HCV was identified in the first generation of mice. MUPHCV mice and AlbHCV mice with the highest levels of viral RNA were studied for expression of HCV core protein to demonstrate that the viral structural proteins were present (Fig. 5) . This immunoblot provided clear evidence that the AlbHCV transgenic mice were making HCV core protein in their livers.

Analysis of gene expression Total RNA was prepared in cesium chloride gradients from liver homogenized in guanidine thiocyanate. Five μg of total RNA was treated with DNase I (Gibco-BRL) and reverse-transcribed with Superscript (Gibco-BRL) using oligo(dT) primers. One μl (1/20 volume of reaction mixture) of the first-strand cDNA was amplified with PCR using HCV core primers(330-351; 5'- ATGAGCACAAATCCTAAACCTC-3' , 726-747;

5'-CAAGCGGAATGTACCCCATGAG-3') . PCR was performed for 35 cycles (94°C for 30 sec, 60°C for 30 sec, 72°C for 1 min) . A resulting 418 bp fragment was visualised on 1.2% agarose gels. PCR for /?-actin was also performed for 25 cycles using the same conditions with commercial sense and anti-sense primers (Stratagene; LaJolla, CA) .

PolyA RNA was purified from the cesium-purified RNAs using oligo(dT) magnetic particles (Promega;

Madison, WI) . Two μg of polyA RNA was separated in 1% agarose gel containing formaldehyde and transferred to nylon membrane (Bio-Rad; Hercules, CA) . Hybridizations were performed with ³²P-labeled HCV probe. The same membrane was re-probed with a ³²P-labelled glyceraldehyde 3-phosphate dehydrogenase (GAPDH) probe to control for sample loading. Analysis of protein expression

Liver samples were homogenized in 3 volumes of RIPA buffer + 1% sodium dodecyl sulphate (SDS) (0.15 M NaCl, 50 mM Tris pH 8.0, 0.5% sodium deoxycholate, 1% SDS, and 1% NP40) . After centrifugation to clear cell debris, the protein lysates were electrophoresed in 15% SDS-polyacrylamide gel and transferred to PVDF membrane (Millipore; Bedford, MA) by electroblotting. The blot was probed with anti-core monoclonal antibody (C7-50A7; a generous gift of Dr. Jack Wands) and developed with horseradish peroxidase-conjugated anti-mouse immunoglobulin antibody (Amersham) using an enhanced chemiluminesence kit (DuPont NEN; Boston, MA) . The membrane was re-probed with anti-actin monoclonal antibody (Amersham; Arlington Heights, II) to control for protein loading.

Immunostaining

Livers from mice were fixed in 4% paraformaldehyde in phosphate buffered saline, sectioned and subjected to immunohistochemistry using a Vectastain ABC kit (Vector Laboratories, Burlingame, CA) . Rabbit polyclonal antibodies against HCV core and E2 (obtained from Dr. Richard Lesniewski of the Abbott Laboratories) were used as primary antibodies in dilution of 1:100. As negative controls, nonimmune rabbit serum or BSA were used. For secondary antibody, biotinylated goat anti- rabbit IgG was used at a dilution of 1:500. The slides were counterstained with hematoxylin. Transgenic lines Potential founder mice were analyzed for the

HCV transgene by Southern blot and PCR analyses of tail DNA. Four founders containing MUPHCV sequences and eleven founders containing AlbHCV were identified and designated as Tg.MCl-4 and Tg.ACl-11. Transgenes were not successfully transmitted in Tg.MC3, AC2, AC4, and AC7 lines. The Tg.ACl line contained two transgene insertion sites. On subsequent breeding, two sublines, each containing one of the transgenes, were identified by

Southern analysis (data not shown) and maintained as two sublines (Tg.ACl-l,ACl-2) . Expression of the transgene was analyzed in three Tg.MC lines and ten Tg.AC lines

(including the two sublines) .

Expression of HCV mRNA in livers of transgenic mice We analyzed liver-specific transgenic expression of HCV mRNA. Northern blot studies of the Tg.MC lines were initially negative. Using highly sensitive HCV-specific RT-PCR, we detected HCV mRNA in two of three Tg.MC lines (Tg.MC2 and MC4) . Since Tg.MC4 expressed higher levels of HCV mRNA than MC2, we analyzed Tg.MC4 in detail. Although the MUP promoter was thought to be developmentally regulated in previous studies, several time points analyzing expression of the MUPHCV sequences revealed no regulation in our transgenic lines after birth (Fig. 6A) . However since RT-PCR is only semi-quantitative, levels of TG.MC4 transcripts may be somewhat more variable than demonstrated.

In order to study the pathogenic effects of HCV viral, we also developed albumin-driven HCV constructs to obtain high levels of expression in mice that should be immunologically tolerant. HCV mRNA was detected in seven of ten Tg.AC lines (Fig. 6B) . The expression levels were generally higher in the Tg.AC lines than seen in Tg. MC4. The highest level of expression was detected in Tg.AC3. Northern blots confirmed the result of RT-PCR (Fig. 7) .

A strong signal was detected in Tg.AC3 and weaker expression was seen in Tg.ACl-0, Tg.ACl-l and Tg.ACl-2 ines.

Expression of HCV proteins in livers of transgenic Mice Expression of HCV core protein was confirmed on Western blots of liver lysates extracts using a monoclonal anti-core antibody (Fig. 8) . Lysates of insect cells expressing the HCV structural proteins revealed a prominent 22 kd band. The same band was clearly visualized in Tg.AC3, AC1-0, AC1-1 and AC1-2. Expression levels of protein correlated well with mRNA levels. A prominent 50 kD band was also detected in Tg.AC3 and may represent unprocessed polyprotein. This band was also evident in the lysate of insect cells.

Immunohistochemical staining confirmed the HCV core expression and revealed expression of HCV E2 proteins (Fig. 9) . We used Tg.MC4 and AC1-0 for histologic studies because the highest expressing line (Tg.AC3) was lost. Hepatocytes of HCV transgenic mice were positively stained by both anti-core and anti-E2 antibodies (Fig. 9D, E, F) , while the wild type negative control mice showed no staining (Fig. 9 H, I, J, K, L) . The expression of HCV proteins was rather heterogeneous, with the highest level of expression in the perivenular hepatocytes (Fig. 9 E, F, J, K) . Patchy staining of hepatocytes was also seen scattered in various regions of the hepatic lobules (Fig. 9D) . E2 staining was visualised on the plasma membrane as well as in the cytoplasm of hepatocytes (Fig. 91, J, K, L) . It is interesting to note that the cytoplasmic staining of E2 appeared to be distributed in a reticular and punctated pattern suggestive of distribution in the endoplasmic reticulum. Similar, though to a lesser extent, cytoplasmic staining was also visualised for the core protein (Fig. 9E) . Overall core staining was present predominantly in the cytoplasm with occasional nuclear staining (Fig. 9F, arrows) . These findings are consistent with the biosynthetic pathway of HCV structural proteins. We did not see any histologic abnormalities associated with expression of HCV structural proteins in these mice up to six months of age.

Advantages One of the inherent shortcomings of some transgenic models is that the product of the transgene can be expressed as a self antigen, to which the host animal is tolerant. As a consequence, immunologically mediated responses which are important in a viral infectious process cannot be studied. We addressed this problem by utilizing an adult-specific promoter. Alternatively, but less desirably, an inducible promoter could have been used. Most of the inducible promoters permit low but significant expression of the protein product of the transgene, which is probably sufficient to be recognized as a self antigen.

The mouse major urinary proteins (MUP) are a family of proteins synthesized at high levels in liver and sweat glands, and excreted in urine by kidney or in sweat by the sweat glands after the rodent reaches puberty (Hastie et al., 1979, Cell 17:449-457). There are several types of MUPs, some of which are exclusively produced by hepatocytes. These genes are silent in utero when the predominate intrathymic education of the lymphoid precursors takes place. The enhancer and promoter elements of these liver-specific MUPs represent ideal regulatory sequences for the expression of a transgene in an adult-specific manner. This promoter has been successfully used to express the SV40 T antigen in transgenic mice for such a purpose, and the expression was indeed adult- and liver-specific (Held et al., 1989, EMBO J. 8:183-191) .

The mice of the invention can be used to assess the immunobiology and pathogenesis of hepatocellular injury associated with HCV infection. For example, spleen cells primed with recombinant vaccinia virus containing HCV structural genes can be transferred into the transgenic animals to elucidate this complex process. Thus, the mice can be used to study the question of whether immune mechanisms can contribute to cellular damage caused by viral infections, and whether either antibodies or cytotoxic lymphocytes may play an adverse role in the pathogensis of HCV infection. HCV-specific T helper and cytotoxic lymphocytes have been demonstrated in the liver and peripheral blood of chronically infected individuals, and these cellular components of immune responses are also directed against multiple viral proteins. These effector cells presumably participate in the lysis of virus-infected hepatocytes and clearance of virus. However, if this response is incomplete, then the virus will persist and cytokine-mediated inflammatory responses may contribute to hepatic damage.

Whether expression of viral proteins leads to hepatocellular injury or other sequelae of HCV infection such as HCC development indepently of the host immune response can also be studied. Although nearly all cases of HCC associated with chronic HCV infection appear in the setting of preexisting cirrhoses, case reports of HCC in the setting of non-cirrhotic or near-normal liver suggest that HCV may be able to cause HCC without invoking the theory of chronic injury and regeneration. The mice of the invention can be used to address this issue. Analysis of HCV sequences fails to reveal proteins that are similar to known classical oncogenes. The N-terminal portion of the NS3 protein which encodes a serine protease has been shown to be capable of transforming fibroblasts in tissue culture. The HCV core protein appears to itself suppress transcription in some transfection studies. Similar studies directed at other viral gene products may shed additional light on directly-acting viral pathogenic activities.

Animal models using transgenic mice have been successfully used to address similar issues in other viral infections, such as hepatitis B virus infections. Since transgenic mice expressing an antigen in utero become tolerant to the antigen, pathogenic effects of cloned viral products can be examined in the absence of an immune response. The apparent co-localization of core and E2 in the mice of the invention may suggest that the structural proteins can form complexes with the endoplasmic reticular membrane as part of an early step in virion assembly. These findings are consistent with observations made in various tissue culture systems. In contrast to one study (Dubuisson et al. (1994) , J. Virol.) in which E2 was not detected at the cell surface, we observed strong hepatocellularsurface localization of the E2 protein in our transgenic mice. Our findings in an in vivo system may more accurately reflect the natural biosynthetic pathway of HCV structural proteins. Despite high level expression of these proteins, the transgenic livers remain histologically normal. This observation suggest that a direct cytopathic effect of HCV structural proteins is unlikely. However, we cannot exclude subtle effects of viral proteins on hepatocyte function. We have not as yet determined the potential long-term effects of expression of HCV structural proteins on hepatocytes. Complete phenotypic analysis of these mice will likely provide additional insights into the pathogeneic mechanisms of HCV. Furthermore, these transgenic mice will provide a useful tool to study the immune response of HCV infection. Adoptive transfer experiments using well characterized HBV transgenic mice and cytotoxic lymphocyte clones have provided important information on the immunobiology of HBV infection.

Other embodiments are within the following claims. For example, transgenic mice of the invention can be made using not just microinjection, but any suitable method, such as homologous recombination into embryonic stem cells.

Any liver-specific promoter/enhancer sequences can be used, e.g., metallothioneien, serum amyloid P component (SAP) promoter (Zhao et al., J.Biochem., 111:736-738. 1992), or αl-antitrypsin promoter (Sifers et al., Nucleic acid Research, .15:1459-1475, 1987. Also, other adult-specific promoter/enhancer sequences can be used, e.g. , the α2-u globulin promoter (Da Costa Soares et al., Mol.Cell.Biol., 7:3749-3758, 1987).

Regarding the HCV DNA, the example described herein employed type lb HCV cDNA from a particular patient; it is of course contemplated that the HCV DNA can be obtained from any suitable source, e.g., other patients, and that any HCV strain can be used as the source of DNA. These various strains will be expected to exhibit fairly wide sequence variability, as HCV is a virus which exhibits a high mutation rate. It is also contemplated that mutations can be intentionally introduced into the HCV sequences used to make the transgenic mice of the invention. What is claimed is:

Claims

1. A transgenic mammal whose hepatic cells express one or more of the HCV proteins C, El, or E2.

2. The mammal of claim 1, wherein the hepatic cells express the HCV proteins C, El, and E2.

3. The mammal of claim 1 wherein said hepatic cells are hepatocytes.

4. The mammal of claim 1 wherein the gene encoding the HCV protein is under the transcriptional control of a liver-specific regulatory sequence.

5. The mammal of claim 1, wherein said regulatory sequence is adult-specific, whereby immune tolerance of the HCV protein is stimulated.

6. The mammal of claim 5, wherein the mammal is a mouse and the gene encoding the HCV protein is under the transcriptional control of a murine regulatory nucleotide sequence.

7. The mouse of claim 6, wherein said murine nucleotide sequences include the Mouse Urinary Protein enhancer/promoter region.

8. The mouse of claim 7 wherein said murine nucleotide sequences include the Mouse Albumin enhancer/promoter region.