WO1997030167A1

WO1997030167A1 - Method of treating liver disorders

Info

Publication number: WO1997030167A1
Application number: PCT/US1997/001564
Authority: WO
Inventors: James M. Wilson
Original assignee: The Trustees Of The University Of Pennsylvania
Priority date: 1996-02-13
Filing date: 1997-02-13
Publication date: 1997-08-21
Also published as: AU2251897A

Abstract

A method for enhancing ureagenesis in a subject, preferably a mammal, in need thereof is provided. The method involves administering to the mammal an effective amount of a recombinant virus which expresses in vivo at least one urea cycle enzyme. The method also may involve administration of additional recombinant adenoviruses expressing other of the urea cycle enzymes or a single recombinant adenovirus expressing more than one enzyme. The method also includes a step of administering an immune modulator with the recombinant virus.

Description

METHOD OF TREATING LIVER DISORDERS This invention was supported by National Institutes of Health Grant Nos. DK47757, DK09031, HD32649 and HD26979. The United States government has certain rights in this invention.

Field of the Invention

The present invention relates generally to the treatment of inherited or acquired liver failure due to compromised ureagenesis; and specifically, to the treatment of the liver with recombinant viruses.

Background of the Invention

Ureagenesis is the process by which ammonia (produced in the metabolism and from deamination of amino acids in the liver and kidney) is removed from the body by conversion to urea. Five enzymes, primarily expressed in the liver, are responsible for the urea cycle which results in ureagenesis. The urea cycle involves the combination of ammonia, carbon dioxide and ATP, which in the presence of the enzyme carbamyl phosphate synthetase forms carbamyl phosphate. This compound reacts with ornithine in the presence of the mitochondrial enzyme ornithine transcarbamylase (OTC) to form citrulline. Arginine is produced from citrulline by a two-step process involving aspartic acid, ATP, and the enzymes argininosuccinate synthetase and arginino-succinate lyase . Finally arginine is hydrolyzed in the presence of the enzyme arginase to yield urea and ornithine. Urea is transported to the kidneys and excreted in the urine and ornithine is recycled back into the urea cycle. Specific inborn genetic errors of enzymes of the urea cycle have been observed in humans. Such defects lead to a syndrome of early onset hyperammonemia, which can cause severe disability or death. For example, a deficiency of the hepatic enzyme OTC is associated with compromised ureagenesis and derangements in nitrogen metabolism leading to intolerance to protein. As liver function deteriorates so does the capability of the patient to breakdown protein, leading to accumulations of ammonia in the blood, which is a marker for metabolic consequences of life-threatening hyperammonemic encephalopathy in humans. The development of encephalopathy can lead to coma and eventually death. This X-linked recessive disorder is the most common inborn error of urea synthesis, with an estimated prevalence of 1:40,000 to 1:80,000 births [Nagata, N. et al, Amer. J. Med. Genet.. 39: 228-229 (1991)], i.e., affecting about 850 newborns per year. See, also, Batshaw, M. L. et al, New Engl. J. Med.. 306: 1387-1392

(1982)]. Survivors of the neonatal crisis and hemizygous males with partial OTC activity often experience recurrent episodes of potentially life threatening hyperammonemia precipitated by excess protein intake or catabolic stress later in life [Brusilow, S. W. , and

Horwich, A. L. , Urea Cycle Enzymes in "The Metabolic and Molecular Bases of Inherited Disease" (Scriver, C. R. , Beaudet, A. L. , Sly, W. S., and Valle, D. , eds) Vol. I, pp. 1187-1232, McGraw-Hill, Inc., New York (1995)]. A similar clinical syndrome is seen in 10-20% of heterozygous females who have <30% residual OTC activity in liver caused by disproportionate lyonization of the normal OTC allele.

Urea cycle disorders are extreme examples of the results of nonspecific liver damage and compromise of ureagenesis due to fallout of hepatocytes. Acquired diseases of the liver, such as cirrhosis or cancers, can cause inefficient urea cycling, due to a deficiency in the number of properly functioning hepatocytes. Such inefficiency in the urea cycle can result in intolerance to protein ingestion, mental deficiency, retarded development and function of the nervous system and excessive amounts of free ammonia in the blood.

The development of alternate pathway metabolic therapy has improved prognosis of urea cycle disorders, such as OTC deficiency. Metabolic derangements in ureagenesis were corrected in several severely afflicted hemizygotes following orthotopic liver transplantation [Largilliere, C. et al, J. Pediat.. 115: 415-417 (1989) cited above; Broelsch, C.E. et al, Ann. Surg.. 212: 368-377 (1990); and Todo, S. et al, Heoatology. 15: 419-422 (1992)]. A therapeutic strategy such as organ replacement, however, has problems due to the intrusiveness of the method, as well as the critical lack of organ donors. Thus, liver enzyme deficiencies, as well as other urea cycle disorders, continue to be devastating illnesses.

Liver-directed gene therapy for a variety of metabolic disorders, including OTC deficiency, has been reported. For example, a minigene expressing rat OTC cDNA from an SV40 promoter introduced into the germline of spf¹"* mice resulted in a phenotype conversion and increase in OTC activity to 80-90% of control in both liver and intestine [Cavard, C. et al, Nucleic Acids Res.. 16: 2099-2110 (1988)]. A similar approach was taken in spf mice with a construct carrying human OTC CDNA under control of the mouse OTC promoter [Jones, S. N. et al, J. Biol. Chem.. 265: 14684-14690 (1990)].

Recombinant adenoviruses have been evaluated as vectors for liver-directed gene therapy [Herz, J. , and Gerard, R. D., Proc. Natl. Acad. Sci., U.S.A.. 90: 2812-2816 (1993); Kozarsky, K. et al, J. Biol. Chem.. 269: 13695-13702 (1994); Stratford-Perricaudet, L. D. et al, Hum. Gene Ther.. 1 : 241-256 (1990); Morsy, M. et al, J. Clin. Invest.. 92: 1580-1586 (1993)]. Adenovirus is rendered defective for use as a vector by deleting the immediate early genes Ela and Elb and incorporating a minigene expressing the therapeutic protein. Adenovirus is efficiently targeted to hepatocytes in vivo following intravenous infusion. High level transgene expression can be achieved in virtually 100% of hepatocytes, most of which are fully differentiated and not dividing. The first use of El deleted viruses for gene therapy was in newborn spf**¹¹ mice [Stratford-Perricaudet, cited above]. Infusion of a vector containing a rat OTC CDNA into the newborn animals led to an increase in hepatic OTC actively in 4/15 mice which persisted for 1-2 months and was associated with decreased urinary orotic acid excretion. Experiments of adenovirus-mediated gene transfer to liver have not been as encouraging when performed in other species as well as in adult mice. Gene transfer to liver is similarly efficient in these experimental models; however, transgene expression is transient, often lasting less than 14-21 days, and associated with substantial hepatitis [Kozarsky et al, cited above; Ishibashi, S. et al, J. Clin. Invest. , 92: 883-893 (1993); Li, Q. et al, Hum. Gene Thera.. 4_.: 403-409 (1993); Yang, Y. et al, Proc. Natl. Acad. Sci. USA. 91: 4407-4411 (1994) (Yang I)].

There remains a need in the art for methods and compositions useful in improving the efficiency of ureagenesis in a liver damaged by acquired or inherited disease, thereby improving the prognosis and survival of such patients.

Summary of the Invention

In one aspect, the present invention provides a method of enhancing ureagenesis in a subject in need thereof comprising administering to said subject a recombinant virus capable of delivering a therapeutic transgene which expresses at least one urea cycle enzyme. The recombinant virus can express more than a single urea cycle enzyme. In another aspect, the invention provides a method of enhancing ureagenesis in a subject in need thereof comprising administering to said subject a combination of recombinant viruses, each virus capable of delivering a therapeutic transgene which expresses a different urea cycle enzyme.

In yet another aspect, the invention provides a method of preventing or treating hepatic encephalopathy due to hyperammonemia by enhancing ureagenesis by the methods of the invention. In a further aspect, the methods described above include a step of administering to said subject an effective amount of an immune modulator, said modulator substantially inhibiting the formation of neutralizing antibodies directed against the virus or of substantially reducing CTL elimination of virally-infected cells.

In still another aspect, there is provided a method for determining the rate limiting urea cycle enzyme necessary in a patient for the treatment of compromised ureagenesis. In yet a further aspect, the invention provides a recombinant adenovirus capable of delivering a therapeutic transgene which expresses at least one urea cycle enzyme.

Other aspects and advantages of the present invention are described further in the following detailed description of the preferred embodiments thereof.

Brief Description of the Drawings

Fig. 1 is a diagrammatic map of recombinant adenoviruses. See, Example 2. Fig. 2A is a graph illustrating the evaluation by Knodell score of periportal and bridging necrosis vs days post infusion of the recipient mouse liver receiving infusion of 5xlθ¹⁰ particles/mouse of H5.OlOCBhOTC (1st gen. hOTC) or H5.110CBhOTC (2nd gen. hOTC) recombinant adenovirus.

Fig. 2B is a graph illustrating the pathological response of Fig. 2A in mice receiving infusion of 1X10¹¹ particles of H5.OlOCMVmOTC (1st gen. mOTC) or H5.110CMVmOTC (2nd gen. OTC) recombinant adenovirus.

Fig. 2C is a graph illustrating the evaluation of intraglobular degeneration and focal necrosis in livers of mice treated as in Fig. 2A. Fig. 2D is a graph illustrating the evaluation of intraglobular degeneration and focal necrosis in livers of mice treated as in Fig. 2B.

Fig. 2E is a graph illustrating the evaluation of portal inflammation response of the recipient mouse liver treated as in Fig. 2A.

Fig. 2F is a graph illustrating the evaluation of portal inflammation response of the recipient mouse liver treated as in Fig. 2B.

Fig. 3A is a graph plotting urinary orotate excretion vs. days post infusion for spf/Y mice infused with 5X10¹¹ particles of H5.OlOCBhOTC or H5.010CMVlacZ.

Fig. 3B is a graph plotting urinary orotate excretion vs. days post infusion for spf/Y mice infused with SxlO¹¹ particles of Hδ.llOCBhOTC or H5.HOCBlacZ. Fig. 3C is a graph plotting plasma glutamine levels vs. days post infusion for spf/Y mice infused with δxlO¹¹ particles of H5.110CBOTC or H5.HOCBlacZ.

Fig. 4 is a bar graph illustrating the liver OTC activity in spf mice infused with recombinant adenoviruses: H5.OlOCBhOTC, H5.OlOCMVhOTC, H5.OlOCMVmOTC, or H5.OlOCMVlacZ. Untreated spf mice and untreated C3H mice are controls. Data are presented as OTC activity (μmol citrulline/mg protein/hr) .

Fig. 5A is a graph plotting urinary orotate excretion level vs days post infusion in spf mice infused with 2x10" particles of H5.OlOCMVmOTC, H5.llOCMVmOTC or H5.110CMVlacZ. Urinary orotic acid levels were measured in duplicate for each sample. Urinary orotate/mg creatinine are presented as a % of pretreatment levels and are the mean + SEM of at least 4 determinations.

Fig. 5B is a graph plotting plasma glutamine levels vs days post infusion in spf mice infused with the recombinant adenoviruses as described in Fig. 5A. Plasma glutamine are presented as a % of pretreatment levels and are the mean ± SEM of between 4 to 10 determinations.

Fig. 5C is a graph plotting urinary orotate excretion level vs days post infusion in spf*^Λ mice infused with 2x10" particles of H5.llOCMVmOTC. Urinary orotic acid levels were measured as described in Fig. 5A. Fig. 5D is a graph plotting plasma glutamine levels vs days post infusion in spf"¹- mice infused with 2xlOⁿ particles of H5.llOCMVmOTC. Plasma glutamine levels are presented as described in Fig. 5C.

Fig. 6 is a graph illustrating the dose response to nitrogen challenge (NH₄C1 in mmol/kg) of spf mice (closed circle) or C3H mice (open circle) . Average scores for each group were indicated. Scores for spf mice are noted by asterisk.

Fig. 7 is a graph illustrating the clinical response to nitrogen challenge of spf mice treated with gene therapy measured in days post viral infusion. Untreated control C3H mice (open triangles) ; untreated control spf mice (closed triangles) ; spf mice treated with OTC virus (open circles) ; and spf mice treated with lacZ virus (closed circles) . Average scores for each group are indicated. Scores for C3H control mice and spf mice treated with OTC virus were noted by asterisk.

Detailed Description of the Invention

The present invention provides a novel method for treating or preventing hepatic encephalopathy caused by compromised ureagenesis in mammals, regardless of the cause of improper functioning of the urea cycle. This method is useful in treating inefficient ureagenesis due to inherited genetic defects in urea cycle enzymes. This method is equally useful in treating compromised ureagenesis which results from non-specific acquired liver failure.

According to this invention, a urea cycle enzyme gene is delivered to the liver cells by administering a recombinant virus, preferably an adenovirus, capable of expressing the gene in vivo . Expression of a deficient urea cycle enzyme in vivo improves the efficiency of the processing of ammonia to urea in the subject's liver. This improvement can be obtained even in a liver in which the deficiency in the enzyme function is due to malfunctioning liver cells or an abnormally low number of properly functioning liver cells, e.g., caused by liver damage due to cirrhosis, rather than to a particular genetic deficiency. One or more of the missing or deficiently-expressed urea cycle enzymes may be supplied by one or more recombinant viral vectors, administered together or sequentially.

It is envisioned that employing somatic gene transfer to enhance ureagenesis can prevent the development of encephalopathy due to hyperammonemia. In most metabolic pathways there is one rate-limiting step. The rate-limiting step of an enzymatic pathway can change when the concentration of reactants, intermediates, and/or products is altered (by enzyme or gene deficiencies and tissue disorders) . Thus, it is anticipated that a vector carrying a gene capable of expressing a single urea cycle enzyme will be sufficient to improve the condition of hepatic encephalopathy in the context of liver failure. However, the methods of this invention also encompass the delivery of more than one gene therapy vector, each of which may express at least one of the five above-mentioned urea cycle genes, or a single vector capable of expressing more than one urea cycle enzyme.

Since the enzymatic reactions of the urea cycle are principally localized to the liver, the methods of this invention preferably target gene therapy vectors to hepatocytes to correct underlying metabolic derangements. The present invention thus corrects urea cycle metabolic derangements by hepatocyte-directed gene transfer rather than by complete organ replacement.

The methods of this invention preferably use recombinant adenoviruses carrying a urea cycle enzyme gene for the treatment of liver metabolic disease, i.e., hepatic encephalopathy. Hemizygous male or homozygous female OTC-deficient mice are recognized as authentic animal models of OTC deficiency, useful for the study of human hyperammonemia (see Example 1) . These OTC- deficient mice, develop hyperammonemia, seizures and coma when challenged with ammonia, thus simulating episodes that characterize the human disease. These life threatening episodes of hyperammonemia are used to demonstrate the efficacy of recombinant adenoviruses for correcting the metabolic defect in liver. I. Recombinant Viruses

Recombinant viruses, that are capable of delivering and stably integrating a functional, normal urea cycle enzyme gene to hepatocytes are used in this method. For convenience hereafter, the urea cycle enzyme gene will be referred to as the OTC gene. However, any of the other four liver enzyme genes, carbamyl phosphate synthetase (CPS) , arginino-succinate lysase (AL) , arginase (ARG) , and argininosuccinate synthetase (AS) , may be used according to the same techniques.

As described in more detail below, recombinant viruses for use in the present invention are desirably deleted in one or more viral genes, and contain a "minigene" containing the liver enzyme gene (e.g., OTC) under the control of regulatory sequences. Optional helper viruses and/or packaging cell lines supply to the recombinant viruses any gene products necessary for the deleted viral genes to ultimately be replicated and/or expressed. Suitable viruses useful in gene therapy are well known, including retroviruses, vaccinia viruses, poxviruses, adenoviruses and adeno-associated viruses, among others. While the methods of this invention are anticipated to be useful with all viral gene therapy vectors, the preferred viruses for use in the methods of the invention are adenoviruses [see, e.g., M. S. Horwitz et al, "Adenoviridae and Their Replication", Virology. second edition, pp. 1712, ed. B. N. Fields et al, Raven Press Ltd., New York (1990); M. Rosenfeld et al, Cell. 6_8:143-155 (1992); J. F. Engelhardt et al, Human Genet. Ther.. 1:759-769 (1993); Yang IV; J. Wilson, Nature. 3_65:691-692 (Oct. 1993); B. J. Carter, in "Handbook of Parvoviruses", ed. P. Tijsser, CRC Press, pp. 155-168 (1990) . Adenoviruses can be purified in large quantities and highly concentrated, and the virus can transduce genes into non-dividing cells.

The adenovirus sequences may be obtained from any known adenovirus type, including the presently identified 41 human types [Horwitz et al, Virology. 2d ed., B. N. Fields, Raven Press, Ltd., New York (1990)]. The DNA sequences of a number of adenovirus types are available from Genbank, including type Ad5 [Genbank Accession No. M73260]. A variety of adenovirus strains are available from the American Type Culture Collection, Rockville, Maryland, or available by request from a variety of commercial and institutional sources. Particularly desirable are human type C adenoviruses (Ad), including serotypes Ad2 and Ad5, which are not associated with human malignancies. For gene therapy, the adenovirus is preferably rendered replication defective by deleting the early gene locus that encodes Ela and Elb [K. F. Kozarsky and J. M. Wilson, Curr. Opin. Genet. Dev.. 2:499-503 (1993)]. Recombinant, defective adenoviruses optionally bearing other mutations, e.g., temperature sensitive mutations, deletions and hybrid vectors formed by adenovirus/adeno- associated virus sequences may also be used in this invention [see, for example, the viruses described in Kozarsky, cited above, and other references cited herein, which are incorporated by reference] .

The methods employed for the selection of viral sequences useful in a recombinant virus, the cloning and construction of the liver enzyme "minigene" and its insertion into a desired virus and the production of a infectious recombinant virus by use of helper viruses and the like are within the skill in the art given the teachings provided herein.

Useful recombinant adenoviruses for delivery of the liver enzyme gene (e.g., OTC) to the liver can contain adenovirus nucleic acid sequences ranging from a minimum sequence amount (a virus containing only the adenovirus cis-elements necessary for replication and virion encapsidation, but otherwise deleted of all adenovirus genes) to viruses characterized by deletions of only selected adenovirus genes. In either case, deleted gene products can be supplied in the recombinant virus production process by a packaging or helper cell line. Desirable "minimal" recombinant adenoviruses (Ad) vectors useful in the present invention are described in detail in co-owned International Patent Application

WO 96/13597, published May 9, 1996, which is incorporated by reference herein for the purpose of describing these vectors.

Recombinant, replication-deficient adenoviruses useful for the methods of this invention alternatively contain more than the minimal adenovirus sequences. These other Ad vectors may be characterized by deletions of various portions of gene regions of the virus, and infectious virus particles formed by the optional use of helper viruses and/or packaging cell lines. See, e.g.,

Kozarsky and Wilson, Curr. Qpin. Genet. Devel.. 3_:499-503 (1993) and references cited therein.

As one example, "first generation" recombinant viruses are formed by deleting all or a sufficient portion of the adenoviral early immediate early gene Ela (which spans mu 1.3 to 4.5) and delayed early gene Elb (which spans mu 4.6 to 11.2) so as to eliminate their normal biological functions. These replication-defective El-deleted viruses are capable of replicating and producing infectious virus when grown on an adenovirus- transformed, complementation human embryonic kidney cell line containing functional adenovirus Ela and Elb genes which provide the corresponding gene products in trans, the 293 cell [ATCC CRL1573]. The resulting virus is capable of infecting many cell types and can express a transgene but cannot replicate in most cells that do not carry the El region DNA unless the cell is infected at a very high multiplicity of infection. Another recombinant adenovirus, a "second generation" virus is characterized by the above described deletion in the Ela and Elb genes, as well as additional deletions and mutations of the adenovirus genome in genes E2a and E3 as follows. Second generation recombinant adenoviruses contain a mutation which produces temperature-sensitive (ts) virus, i.e., incorporation of the missense temperature-sensitive mutation in the DNA binding protein (DBP) E2a region found in the Ad5 H5tsl25 strain [P. Vander Vliet et al, J. Virol.. 15:348-354 (1975)] at 62.5 mu. A single amino acid substitution (62.5 mu) at the carboxy end of the 72 kd protein produced from the E2a gene in this strain produces a protein product which is a single-stranded DNA binding protein and is involved in the replication of adenoviral genomic DNA. At permissive temperatures (approximately 32°C) the ts strain is capable of full life cycle growth on HeLa cells, while at non-permissive temperatures (approximately 38°C) no replication of adenoviral DNA is seen. In addition, at non-permissive temperatures, decreased immunoreactive 72 kd protein is seen in HeLa cells.

Additionally, a "second generation" adenovirus is characterized by deletion of all or a portion of the adenovirus delayed early gene E3 (which spans mu 76.6 to 86.2) . The function of E3 is irrelevant to the function and production of the recombinant virus particle. For example, recombinant viruses are constructed with a therapeutic minigene inserted into the El-deleted region of the known mutant Ad5 sub360 backbone [J. Logan et al, Proc. Natl. Acad. Sci. USA. 8_1:3655-3659 (1984)]; or the Ad5 mutant dl7001 backbone [Dr. William Wold, Washington University, St. Louis; see, e.g., J. E. Engelhardt et al, Nat. Genet.. 4_:27-34 (1993)]. Both mutant viruses also contain a deletion in the E3 region of the adenoviral genome; in sub360, at 78.5 to 84.3 mu, and in dl7001, at 78.4 to 86 mu. The life cycle of both sub360 and dl7001 display wild type characteristics.

Recombinant adenoviruses useful in this invention may also be constructed having a deletion of the El gene, at least a portion of the E3 region, and an additional deletion or mutation within adenovirus genes other than El and E3 to accommodate the liver enzyme minigene and/or other mutations which result in reduced expression of adenoviral protein and/or reduced viral replication. For example, all or a portion of the adenovirus delayed early genes E2a (which spans mu 67.9 to 61.5), E2b (which spans mu 29 to 14.2) and E4 (which spans mu 96.8 to 91.3) may be eliminated or mutated in the recombinant adenovirus. Other recombinant adenoviruses are described in the literature [J. F. Engelhardt et al, Proc. Natl. Acad. Sci. USA. 9.1:6196- 6200 (June 1994); Y. Yang et al. Nature Genet.. 7_: 362- 369 (July, 1994) and references cited therein] . The liver enzyme minigene may be inserted into any deleted region of the selected Ad virus. By "minigene" is meant the combination of the liver enzyme gene and the other regulatory elements described below, which are necessary to transcribe the gene and express the gene product in vivo . The therapeutic gene contained within the recombinant virus is one or more urea cycle enzymes, as described above and known in the art. The human gene sequence for OTC is known [see, e.g., Jones, S. N. et al, J. Biol. Chem.. 265: 14684-14690 (1990) and references cited therein], as are the sequences for the other liver enzymes. For the sequences of CPS, see Y. Haraguchi et al, Gene. 107:335-340 (1991); for the sequences of AL, see W. E. O'Brien et al, Proc. Natl. Acad. Sci. USA. 8_3_:7211-7215 (1986); for the sequences of ARG, see Y. Haraguchi et al, Proc. Natl. Acad. Sci. USA_f 84_.:412-415 (1987); for the sequences of AS, see, G. E. Herman et al, Somatic Cell Mol. Genet.. 15(4) :289-296 (1989) and F. M. Boyce et al, Somatic Cell Mol. Genet.. 15(2) :123-130 (1989). These genes may be obtained from academic or commercial sources. Generally, the entire coding region of the liver enzyme sequence is used in the minigene. Urea cycle enzyme genes which are of human as well as other mammalian origins, e.g., rabbit, monkey, etc., may also be useful in this invention. The liver enzyme gene is operatively linked to regulatory components in a manner which permits its transcription. Such components include conventional regulatory elements necessary to drive expression of the liver enzyme transgene in a cell transfected with the recombinant virus. Thus the minigene also contains a selected promoter which is linked to the transgene and located, with other regulatory elements, within the selected viral sequences of the recombinant virus.

Selection of the promoter is a routine matter within the skill of the art and is not a limitation of this invention. Useful promoters may be constitutive promoters or regulated (inducible) promoters, which will enable control of the amount of the transgene to be expressed. For example, a desirable promoter is that of the cytomegalovirus (CMV) immediate early promoter/enhancer [see, e.g., Boshart et al, Cell. 41:521-530 (1985)]. Another desirable promoter/enhancer sequence is the chicken cytoplasmic β-actin promoter [T. A. Kost et al, Nucl. Acids Res.. ϋ(23):8287 (1983)]. The minigene may also desirably contain nucleic acid sequences heterologous to the virus sequences including sequences providing signals required for efficient polyadenylation of the transcript (poly-A or Pa) and introns with functional splice donor and acceptor sites. A common poly-A sequence which is employed in the exemplary vectors of this invention is that derived from the papovavirus SV-40. The poly-A sequence generally is inserted in the minigene following the transgene sequences and before the viral sequences. A common intron sequence is also derived from SV-40, and is referred to as the SV-40 T intron sequence. A minigene of the present invention may also contain such an intron, desirably located between the promoter/enhancer sequence and the transgene. Selection of these and other common vector elements (e.g., reporter genes, and sequences allowing autonomous replication) are conventional [see, e.g., Sambrook et al, "Molecular Cloning. A Laboratory Manual.", 2d edit., Cold Spring Harbor Laboratory, New York (1989) and references cited therein] and many such sequences are available from commercial and industrial sources as well as from Genbank.

Exemplary Ad viruses containing an OTC minigene used to demonstrate this invention are described in detail in Example 1 below. Other exemplary Ad viruses, including Ad viruses containing an AS minigene (Example 6) , Ad viruses containing an ARG minigene (Example 7) Ad viruses containing a CPS minigene (Example 8) , and Ad viruses containing an AL minigene (Example 9) , are also described in detail. Assembly of the selected DNA sequences of the adenovirus, the reporter genes or therapeutic genes and other elements into the recombinant adenovirus and the use of the helper viruses to produce an infectious virus is performed using conventional techniques. Such techniques include conventional cloning techniques of cDNA such as those described in texts [Sambrook et al, cited above], use of overlapping oligonucleotide sequences, polymerase chain reaction, and any suitable method which provides the desired nudeotide sequence. Standard transfection and co-transfection techniques are employed, e.g., CaP0₄ transfection techniques using the human embryonic kidney 293 cell line, which provides Ela function to El-deleted adenovirus. Other conventional methods employed include homologous recombination of the viral genomes, plaguing of viruses in agar overlay, methods of measuring signals generated by the reporter gene (e.g., enzymatic, colorimetic, etc.) and the like.

For example, following the construction and assembly of the desired minigene-containing plasmid vector, the vector is infected in vitro in the presence of an optional helper virus and/or a packaging cell line. Homologous recombination occurs between the helper and the vector, which permits the adenovirus-transgene sequences in the vector to be replicated and packaged into virion capsids, resulting in recombinant viral particles. The current method for producing such virus particles is transfection-based. Briefly, helper virus is used to infect cells, which are then subsequently transfected with an adenovirus plasmid vector containing an OTC (or other liver enzyme) transgene by conventional methods. About 30 or more hours post-transfection, the cells are harvested, an extract prepared and the recombinant virus vector containing the OTC (or other liver enzyme) transgene is purified by buoyant density ultracentrifugation in a CsCl gradient.

II. The Method of this Invention

The resulting recombinant adenovirus containing the liver enzyme minigene provides an efficient gene transfer vehicle for delivery of the liver enzyme gene to a patient in vivo or ex vivo and provide for introduction of the gene into a liver cell, alternatively, for the production of a medicament useful in treatment of a non- inherited liver disorder. These recombinant viruses can also be employed to produce the selected liver enzyme in vitro , if desired. The recombinant virus may be administered to a patient, preferably suspended in a biologically compatible solution or pharmaceutically acceptable delivery vehicle. A suitable vehicle includes sterile saline. Other aqueous and non-aqueous isotonic sterile injection solutions and aqueous and non-aqueous sterile suspensions known to be pharmaceutically acceptable carriers and well known to those of skill in the art may be employed for this purpose. The medicament or virus is administered in sufficient amounts to transfect the hepatocytes and provide sufficient levels of transduction and expression of the liver enzyme gene to improve the functioning of the urea cycle, without undue adverse effects or with medically acceptable physiological effects which can be determined by those skilled in the medical arts. Conventional and pharmaceutically acceptable routes of administration include direct delivery to the liver, intranasal, intravenous, intramuscular, subcutaneous, intradermal, oral and other parental routes of administration. Routes of administration may be combined, if desired.

Dosages of the viral vector will depend primarily on factors such as the level of compromised ureagenesis, the cause of such condition (i.e., genetic defect or acquired illness) , the selected urea cycle enzyme gene(s) and number of such genes to be delivered, and factors such as the age, weight and health of the patient. Thus dosages may thus vary among patients. For example, a therapeutically effective human dosage of the recombinant viruses is generally in the range of from about 0.1 to about 100 ml, and more preferably from about 0.1 to about 20 ml, of saline solution containing concentrations of from about 1 x IO⁷ to 5 x lθⁿ pfu/ l viruses, and preferably about 1 x 10" pfu/ml. A preferred adult human dosage is about 1 ml saline solution at the above concentrations. The dosage will be adjusted to balance the therapeutic benefit against any adverse effects. The levels of expression of the gene encoding the selected urea cycle enzyme can be monitored to determine the selection, adjustment or frequency of dosage administration.

Where it is desired to treat the compromised ureagenesis with more than a single urea cycle enzyme gene, it is anticipated that each enzyme gene will be present in a separate recombinant adenovirus, and that such viruses will be administered at approximately the same time and in approximately equivalent dosages. However, it should be understood that two or more urea cycle enzyme genes may be expressed from the presence of two minigenes in the same recombinant virus.

III. Co-administration of Virus with Immune Modulator

Current limitations of adenovirus- ediated gene therapy include the potential of cytotoxic T lymphocyte (CTL) responses to residually-expressed viral protein and possibly to OTC (or other liver enyzme) in patients with null mutations. In addition, previous studies have described the development of neutralizing antibodies to input viral proteins that block gene transfer upon second administration [Kozarsky et al, cited above] . Improved vectors deleted for adenoviral genes may diminish CTL responses to viral protein, but will have no effect on neutralizing antibody to viral capsid protein or CTL responses to OTC. Activation of CD4 T helper cells to input viral proteins is necessary for both B cell and CTL activation. Transient blockade of the initial CD4 T cell activation at the time of virus instillation has been shown to both prolong transgene expression and prevent formation of neutralizing antibody [Yang, Y. et al, Nature Med.. 1, 890-893 (1995) (Yang VI) ] .

Thus in one preferred embodiment, the method according to this invention for preventing or treating hepatic encephalopathy due to hyperammonemia and/or compromised ureagenesis in inherited or acquired liver disorders involves the coadministration of an immune modulator together with the recombinant adenovirus carrying an OTC gene. Other preferred embodiments involve preventing or treating hepatic encephalopathy by coadministration of an immune modulator together with a recombinant adenovirus carrying an ARG, CPS, AS, or AL genes. Alternatively, combinations of these recombinant adenoviruses may be utilized. A suitable amount of a preferably short-acting, immune modulator may be administered either concurrently with, or before or after administration of the recombinant adenovirus. The selected immune modulator is defined herein as an agent capable of inhibiting the formation of neutralizing antibodies directed against the recombinant virus of this invention or capable of inhibiting cytolytic T lymphocyte (CTL) elimination of the virus. The immune modulator may interfere with the interactions between the T helper subsets (T_H1 or T_m) and B cells to inhibit neutralizing antibody formation. Alternatively, the immune modulator may inhibit the interaction between T_H1 cells and CTLs to reduce the occurrence of CTL elimination of the virus. More specifically, the immune modulator transiently interferes with, or blocks, the function of the CD4 T cells.

The method of selection of immune modulators is disclosed in detail in co-pending United States patent application No. 08/394,032, which is incorporated herein by reference. In brief, selection of the immune modulator may be based upon the mechanism sought to be interrupted or blocked. See, e.g., Y. Yang et al, Nat. Medic.. l(9):890-893 (1995), incorporated by reference herein. Preferably, the method of inhibiting an adverse immune response to the gene therapy vector involves nonspecific inactivation of CD4 cells with monoclonal antibody, preferably, "humanized" antibodies which prevent the recipient from mounting an immune response to the blocking antibody. Such "humanization" may be accomplished by methods known to the art. See, for example, G.E. Mark and E. A. Padlan, "Chap . 4 . Humanization of Monoclonal Antibodies" , The Handbook of Experimental Pharmacology, vol. 113, Springer-Verlag, New York (1994), pp. 105-133, which is incorporated by reference herein.

In one embodiment, the immune modulator can be a cytokine, such as interleukin-12 [see, e.g., Y. Yang et al, Nat. Med.. 1(9):1-4 (Sept., 1995)] or gamma interferon [S. C. Morris et al, J. Immunol.. 152:1047-

1056 (1994); F. P. Heinzel et al, J. EXP. Med.. 177:1505 (1993)]. Another desirable immune modulator for use in this method which selectively inhibits the CD4+ T cell subset T_H1 function at the time of primary administration of the viral vector includes interleukin-4 [see, e.g., Yokota et al, Proc. Natl. Acad. Sci.. USA. 83_:5894-5898 (1986); United States Patent No. 5,017,691]. Such cytokines for use in this method are preferably in protein form, e.g., recombinantly produced using known techniques, or obtained commercially. Alternatively, a cytokine gene may be engineered into a recombinant virus and expressed in a target cell in vivo or ex vivo . It is also anticipated that active peptides, fragments, subunits or analogs of these cytokines which share the function of these proteins, will also be useful in this method.

To permit an effective second administration of virus in liver-directed gene therapy, the method preferably involves administration of more than one cytokine, specific dosing regimens and/or co¬ administration of an additional immune regulator, such as an antibody. However, the use of cytokines is advantageous because cytokines are natural products, and thus not likely to generate any adverse immune responses in the patient to which they are administered.

Still other immune modulators useful in this method are agents that specifically inhibit or deplete CD4+ cells, for example, by antibody to the CD4 protein. Among such agents include anti-T cell antibodies, such as anti-OKT 3+ [see, e.g., US Patent No. 4,658,019; European Patent Application No. 501,233, published September 2, 1992, among others; and commercially available antibody GK1.5 (ATCC Accession No. TIB207) ] . Depletion of CD4+ cells is shown to inhibit the CTL elimination of the virus.

Alternatively, any agent that interferes with, or blocks the interactions necessary for, the activation of B cells by T_H cells, and thus the production of neutralizing antibodies is useful in the methods of this invention. For example, an antibody to CD40 ligand (anti-CD40L) [available from Bristol-Myers Squibb Co; see, e.g., European patent application 555,880, published August 18, 1993] or a soluble CD40 molecule can be a selected immune modulator in this method. A soluble form of B7 or an antibody to CD28 or CTLA4, e.g., CTLA4-Ig [available from Bristol-Myers Squibb Co; see, e.g., European patent application 606,217, published July 20, 1994] can also be the selected immune modulator in this method.

Other immune modulators or agents that non- specifically inhibit immune function, i.e., cyclosporin A or cyclophosphamide, may also be useful in this method. Depending upon the cause of the patient's liver dysfunction, the subject may be undergoing long term treatment with a non-specific immunosuppressant. If not, however, a short course of treatment with cyclosporin A or cyclophosphamide is preferred where these agents are utilized.

A suitable amount or dosage of the immune modulator will depend primarily on the identity of the modulator, the amount of the recombinant virus bearing the OTC transgene which is initially administered to the patient, and the method of delivery of the virus. Other secondary factors such as those mentioned above, may also be considered by a physician in determining the dosage of immune modulator to be delivered to the patient. For example, a therapeutically effective human dosage of a protein immune modulator, e.g., IL-12 or γ-IFN, is generally in the range of from about 0.5 μg to about 5 mg per about 1 x 10⁷ pfu/ml recombinant adenovirus. Various dosages may be determined by one of skill in the art to balance the therapeutic benefit against any side effects. According to the method of this invention, the co-administration of recombinant virus and immune modulator occur within a close time proximity to each other. According to this aspect of the method, the immune modulator is administered simultaneously with the recombinant virus expressing the urea cycle enzyme gene. Alternatively, the immune modulator is administered prior to or subsequent to administration of the virus. It is presently preferred to administer the modulator concurrently with or no longer than one to three days prior to the administration of the virus. The immune modulator may be administered separately from the recombinant virus, or, if desired, it may be administered in admixture with the recombinant virus. The immune modulator may be administered in a pharmaceutically acceptable carrier or diluent, such as saline. For example, when formulated separately from the virus, the immune modulator is desirably suspended in saline solution. Such a solution may contain conventional components, e.g. pH adjusters, preservatives and the like. Such components are known and may be readily selected by one of skill in the art.

Alternatively, the immune modulator may be itself administered as DNA, either separately from the vector or admixed with the recombinant virus bearing the transgene. Methods exist in the art for the pharmaceutical preparation of the modulator as protein or as DNA [See, e.g., J. Cohen, Science. 259:1691-1692 (1993) regarding DNA vaccines] . Desirably the immune modulator is administered by the same route as the recombinant virus.

The immune modulator may be formulated directly into the composition containing the virus administered to the patient. Alternatively, the immune modulator may be administered separately, preferably shortly before or after administration of the virus. In another alternative, a composition containing one immune modulator may be administered separately from a composition containing a second immune modulator, and so on, depending on the number of immune modulators administered. These administrations may independently be before, simultaneously with, or after administration of the recombinant virus.

The administration of the selected immune modulator may be repeated during the treatment with the recombinant adenovirus carrying the transgene, during the period of time that the transgene is expressed, as monitored by assays suitable to evaluating the functioning of the urea cycle or with every booster of the recombinant virus. Alternatively, each reinjection of the same recombinant virus may employ a different immune modulator.

IV. Determination of Rate-Limiting Urea Cycle Enzyme

Another aspect of the present invention includes the determination of which of the five urea cycle enzymes is the rate-limiting enzyme involved in any particular liver dysfunction characterized by compromised ureagenesis.

In one assay embodiment, ureagenesis is compromised by partial liver resection in an accepted animal model of the indicated liver dysfunction. Recombinant viruses containing different urea cycle enzyme genes of this invention are administered singularly. After each virus and individual urea cycle enzyme gene is administered, ureagenesis is evaluated by conventional methods known to the art. Based on the observation of the functioning of ureagenesis following administration of each vector, one compares the results of administration of each enzyme gene and thereby determines which enzyme is necessary to improve ureagenesis. Alternatively, a combination of urea cycle enzymes in one or more recombinant adenoviruses are administered and their effects studied in the same manner to determine what combination of enzymes is necessary to improve ureagenesis.

In another assay embodiment, a toxic drug is administered to bring the animal subject to the point of hyperammonemia. Then the recombinant viruses are administered singularly or in combination, as described above, to determine the best treatment for the indicated liver dysfunction.

V. Examples of the Method

The following examples illustrate the methods for preparing suitable recombinant viruses useful in the therapeutic methods of this invention, and experiments designed to demonstrate the usefulness of the method in treating an animal model of a urea cycle enzyme deficiency. These examples are illustrative only and do not limit the scope of the invention.

Examples 1 and 2 discuss the animal model and OTC-expressing recombinant adenoviruses used in these experiments. Example 3 discusses the assay methods used to assess efficacy of treatment of an OTC deficiency. Example 4 demonstrates the utility of adenovirus-mediated gene therapy in preventing the biochemical and clinical sequelae of nitrogen challenge in OTC deficiency. Briefly described, adenovirus vectors of the present invention that express a normal murine OTC gene are infused into the blood of OTC deficient animals. Adenoviral vectors administered by this route target to liver and high level OTC expression is detected in virtually all hepatocytes. The animals are subsequently challenged with ammonia and evaluated for metabolic and clinical consequences. Animals treated with gene therapy vectors expressing an OTC gene fail to accumulate significant nitrogen metabolites and tolerate the ammonia challenge without clinical sequelae. This contrasts with mock-treated OTC deficient mice which accumulate large quantities of nitrogen containing metabolites in serum and uniformly develop morbid, if not mortal, consequences of the ammonia challenge.

Example 5 sets out the experimental protocols and results of treating the OTC-deficient mouse models with first and second generation adenoviruses carrying minigenes which differed in promoters and origin of the OTC gene. The experimental protocols systematically evaluated the impact of both the transgene and virus on the safety and efficiency of adenovirus-mediated gene transfer to liver. Recombinant "first generation" adenoviruses deleted in El contain either human OTC CDNA driven by a CMV-enhanced, β-actin promoter (H5.OlOCBhOTC) or mouse OTC CDNA driven by a strong CMV promoter (H5.OlOCMVmOTC) . Both viruses achieved high level gene transfer in vivo. However, expression of functional enzyme from the first generation virus carrying the human OTC was low and transient, and associated with substantial pathology with no observable impact on the underlying metabolic abnormalities (i.e., elevated urinary orotate and serum glutamine) . Infusion of first generation virus containing the mouse OTC in the mouse model was associated with a complete normalization of liver OTC enzyme activity, as measured by lysate and histochemical assays, and partial correction of the metabolic abnormalities. Expression of murine OTC CDNA and the associated metabolic improvements were transient, lasting less than 28 days.

Experiments were repeated with the second generation viruses in which the tsl25 mutation in the adenovirus gene, E2a, was introduced into the El-deleted vectors containing hOTC or mOTC minigenes (i.e. , H5.110CBhOTC and H5.llOCMVmOTC) . The benefit of incorporating the tsl25 E2a mutation into the recombinant adenovirus carrying the human OTC minigene was minimal in this construct. However, metabolic correction is complete and prolonged in spf mice treated with the second generation adenovirus containing the mouse OTC/CMV promoter. With the second generation recombinant adenovirus H5.llOCMVmOTC, expression of viral genes was reduced, inflammation was diminished, MOTC expression was prolonged for at least two months (normalized hepatic activity) and associated with complete metabolic correction of abnormalities in urinary orotate and serum glutamine. Thus, use of a sufficiently strong promoter with a species-homologous OTC cDNA was useful in achieving curative therapeutic gene expression in this model.

Data in these authentic animal models of OTC deficiency support the utility of recombinant adenoviruses for treating liver metabolic diseases. Complete and prolonged correction (up to 3 months) of the metabolic defect has been demonstrated in the animal model following a single infusion of purified virus. Infusion of recombinant adenovirus leads to reconstitution of OTC enzyme activity in liver that peaks between days 2 and 14 and begins to return to baseline by day 28. Sufficient OTC enzyme is expressed within 24 hours to both diminish nitrogen induced accumulations of plasma amino acids and improve the resulting neurological abnormalities.

This finding in the OTC-deficient mouse, an accepted animal model, supports the claimed method for using adenovirus-mediated gene therapy in the treatment of an acute episode of hyperammonemia in humans with a human OTC gene, where the goal is to reverse underlying metabolic derangements within 72 hours (See Example 8) . The use of such recombinant adenoviruses to stably correct the genetic defect in OTC deficiency and prevent the onset of hyperammonemia may be accomplished as described above by the co-administration of immune regulators with the viruses. Other modifications in the recombinant adenovirus that diminish its immunogenicity in combination with regimens of immune blockade are anticipated to further prolong the period of transgene expression and allow effective repeated administration of virus.

Example 6 describes construction of exemplary adenoviruses carrying AS. Example 7 describes construction of exemplary adenoviruses carrying ARG. Example 8 describes construction of exemplary adenoviruses carrying CPS. Example 9 describes construction of an exemplary adenovirus carrying AL. Example 10 describes use of the exemplary adenoviruses of the invention in the treatment of hyperammonemia.

The following examples provide preferred viral vectors and exemplary methods of the invention. The examples do not limit the invention.

Example 1 - Murine Models The best characterized murine model of OTC deficiency is the sparse fur (spf) mouse, in which a missense mutation in codon 117 of the OTC gene leads to a catalytically defective enzyme with hepatic OTC activity reduced to approximately 5-20% of wild-type levels at physiologic Ph [Veres, G. et al. Science. 237: 415-417 (1987) and Qureshi, I. A. et al, Pediat. Pes.. 13: 807-811 (1979)]. The 85% reduction in liver OTC activity in this mouse leads to an activation of de novo pyrimidine synthesis as evidenced by a 13-fold increase in excretion of orotic acid in the urine.

The other accepted murine model is the spf*¹* (abnormal skin and hair) mutant mouse, in which a point mutation in the final base pair of exon 4 of the OTC gene leads to aberrant splicing with markedly reduced levels of OTC mRNA and only 5% of normal OTC activity

[Doolittle, D. P. et al, J. Hered.. 65: 194-195 (1974) and Hodges, P. E. , and Rosenberg, L. E. , Proc. Natl. Acad. Sci.. U.S.A.. 86: 4142-4146 (1989)]. Both models are characterized by dysfunctional or reduced (10-30%) residual OTC enzyme activity in liver, resulting in metabolic and clinical abnormalities consistent with a partial OTC deficiency in humans. Hemizygous male and homozygous female pups have a mild phenotype that resembles human males and symptomatic females with partial OTC activity (i.e., late onset hemizygotes and symptomatic heterozygotes). For example, these mice are runted and have wrinkled skin with little to no fur early in development. The mice are asymptomatic on a normal diet with baseline abnormalities in plasma glutamine, citrulline, and arginine.

Injection of excess nitrogen into adult spf or spf*'^h hemizygotes precipitates a life threatening episode of nitrogen accumulation. Nitrogen accumulates in these animals, as evidenced by a 60% increase in serum glutamine. Conversion of ^1SNH₃ to urea is decreased to 50% of control and the level of serum citrulline, a urea cycle intermediate, is decreased to 25% of control. This episode is accompanied by symptomatic hyperammonemia, glutaminemia, severe orotic aciduria and neurological abnormalities similar to what is seen in OTC deficient patients exposed to excess nitrogen because of increased protein intake or enhanced catabolism of protein. The mouse strains spf, sρf"^h and C3HeB/J used in the Examples below were purchased from Jackson Laboratory (Bar Harbor, Maine) and maintained in the Wistar animal facility. Female heterozygous spf/X or spfVx mice were bred with normal male (X/Y) C3HeB/J mice to generate experimental spf/Y or spf"^h/Y mice. All animals used in the experiments described below were between 6 to 10 weeks of age. Example 2 - First and the Second Generation Recombinant Adenoviruses

A summary of recombinant adenoviruses used in this study is provided in Fig. 1. Nomenclature for the vectors was described in J. Engelhardt et al, Proc. Natl. Acad. Sci. USA. £1:6196-6200 (1994) [Engelhardt I] .

H5. OlOCBhOTC: This El-deleted first generation recombinant adenovirus was prepared as follows. The coding sequence for human OTC was removed from plasmid pHO-731 on a 1.0 kb Hinfl fragment [Horwich, A. L. et al, Science. 224: 1068-1074 (1984)], blunted with Klenow, ligated with Bell linkers, and cloned in direct orientation into the BamHI site of the retroviral vector pgagBA [Grossman, M. et al, Hum. Gene Ther. _f 3_: 501-510 (1992)].

A Xho I to Nhe I restriction fragment from this retroviral vector containing the chicken cytoplasmic β-actin promoter [T. A. Kost et al, cited above], the human OTC CDNA, and 3 ^• untranslated sequences including 130 bp of retroviral sequence was cloned in place of lacZ in the adenoviral vector pAd.CMVlacZ [previously described in detail in Kozarsky, K. et al, J. Biol. Chem.. 269: 13695-13702 (1994) ; see also, Boucher et al, Hum. Gene Thera.. 5:615-639 (1994)]. The resulting plasmid, called pAd.CBhOTC, contains adenovirus type 5 map units 0-1, a minigene capable of directing the expression of human OTC cDNA from a CMV-enhanced, chicken β-actin promoter, a polyadenylation signal, and adenovirus map units 9.7-16 in a plasmid backbone. This plasmid is deleted of adenoviral sequences spanning 1 to 9.6 map units. In preparation for production of virus, pAd.CBhOTC was linearized with Nhe I and transfected into 293 cells [ATCC CRL1573] with Clal/Xbal restricted sub360 genomic adenovirus type 5 DNA, which contains a small deletion in the E3 gene between m.u. 78.5-84.3 [Logan, J. , and Shenk, T., Proc. Natl. Acad. Sci.. U.S.A.. 81: 3655-3659 (1984)]. The resulting recombinant virus, designated H5.OlOCBhOTC was grown and purified through three rounds of plaque isolations.

H5.110CBhOTC: This second generation recombinant virus differs from H5.OlOCBhOTC only by a single base pair substitution in the E2a gene which generates a temperature sensitive viral DNA binding protein (DBP) capable of growth at 32°C but not 39°C. This virus was deleted in El (i.e., no Ad m.u. 1-9.2), defective in E2a due to the tsl25 mutation, and contains the above-described human OTC cDNA minigene.

The E2a ts mutation contained within this recombinant adenovirus was generated from the wild type Ad5 mutant strain H5.tsl25 [Ensinger, M. J. , and Ginsberg, H. S., J. Virol.. 10: 328-339 (1972)]. This temperature sensitive mutation inactivates essential gene product of E2a. This second generation virus has shown improved efficacy in gene transfer to mouse liver, and mouse, rat, and primate lung, using lacZ containing constructs in place of the human OTC gene [Yang II; Engelhardt I; Engelhardt, J. et al, Hum. Gene Thera. _f 5_: 1217-1229 (1994) (Engelhardt II); Goldman, M. J. et al, Hum. Gene Thera.. 6_: 839-85 (1995)]. In each case, expression of the transgene is prolonged for a variable period of time and inflammation associated with administration of first generation viruses is diminished. H5 . OlOCMVmOTC or H5. llOCMVmOTC: These first and second generation recombinant viruses differ from the above-described viruses in that they contain a strong constitutive CMV viral enhancer/promoter and carry mouse OTC cDNA rather than human. Thus, these adenoviral vectors deliver a species-homologous OTC gene to the experimental animal. It is possible that the murine OTC transgene product interacts with regulatory factors in the cell involved in ammonia homeostasis. Mouse OTC cDNA was generated by RT-PCR, cloned into pGEM-T vector (Promega, Madison, Wl) and restricted with Spe I and Sac II. A 1.5kb fragment containing mouse cDNA was isolated, blunted and cloned into EcoRV site of an adenoviral vector pAd.CMV-link (a plasmid containing the adenoviral sequences 0 to 16 map units deleted of Ela and Elb as described in the other adenovirus vectors into which a CMV promoter-polylinker cassette was cloned) .

The new plasmid, designated pAd.CMVmOTC, was linearized with EcoRI and cotransfected into 293 cells with Clal/Xbal restricted Ad5 viral DNA containing the sub360 mutation in E3 as described above to generate the first generation recombinant virus, H5.OlOCMVmOTC.

To generate the second generation recombinant adenovirus, pAd.CMVmOTC was linearized with EcoRI and cotransfected into 293 cells with Clal/Xbal restricted Ad5 viral DNA containing the tsl25 mutation in E2a as described above and the sub360 mutation in E3, resulting in H5.llOCMVmOTC. (For nomenclature of recombinant adenoviruses see Engelhardt II) .

Both viruses were purified through three rounds of plaque isolation. The mouse cDNA part of the recombinant viruses was sequenced in both directions.

H5. llOCMVlacZ: This second generation lacZ control virus was constructed as described in Engelhardt I, and purified through three rounds of plaque isolation. H5. OlOCMVhOTC: For use in studying the influence of the promoter, this first generation recombinant virus was prepared similarly to H5.OlOCMVmOTC, but the human OTC gene above was used to replace the mouse OTC gene described above.

Suitable adenoviral vectors capable of expressing the other liver enzyme genes may be readily prepared using methods similar to those described for the above vectors, and substituting the desired liver enzyme gene.

Example 3 - Assay Methods A . OTC Lysate Assay

Cells were harvested by scraping into mitochondria lysis buffer (0.5% Triton, 10 mM Hepes, pH 7.4, 2 mM DTT) , and total protein was extracted by three freeze-thaw cycles. Liver tissue was homogenized in mitochondria lysis buffer with a Polytron homogenizer. The homogenate was centrifuged in a microfuge at the maximum speed for 5 minutes; the supernatant was transferred to a new tube; and OTC enzyme activity was measured according to the assay described by Lee, J. T., and Nussbaum, R. L. , J. Clin. Invest., 8_4, 1762-1766 (1989) , with the following modifications.

Briefly, 2-10 μg of total cellular protein was added to 700 μL of reaction mixture (5 mM ornithine, 15 mM carbamyl phosphate, and 270 mM triethanolamine, pH 7.7) which was incubated at 37^βC for 30 minutes.

Reactions were stopped by adding 250 μL of 3:1 phosphoric acid/sulfuric acid (by volume) . Citrulline production was then determined by adding 50 μL of 3% 2,3-butane- dionemonoxime, incubating at 95-100°C in the dark for 15 minutes, and measuring absorbance at 490 nm.

JB. OTC Histochemistry

Slides of liver tissue, less than 3 mm thick, were fixed in 2% paraformaldehyde in PBS for 4-6 hours at room temperature and subsequently washed with PBS containing 10% sucrose for 2 hours, PBS containing

20% sucrose for 2 hours, and finally PBS containing 30% sucrose overnight. The liver slides were then embedded in OCT and sectioned for histochemical staining as described by Mizutani, A., J. Histochem. Cvtochem.. 16: 172-180 (1968) .

Briefly, a reaction medium was prepared first, which contained 12 mg carbamyl phosphate, Li salt; 20 mg L-ornithine dihydrochloride, 3.2 g sucrose; 16 ml 0.05M triethanolamine buffer, pH 7.2; 20 ml distilled water, and 4 ml 1% lead nitrate. The lead nitrate solution was added dropwise with continuous stirring and the solution was readjusted to pH 7.2 with 1 N NaOH. The slightly turbid substrate mixture was filtered and used immediately.

Sections were incubated for 30 minutes in reaction medium at room temperature, washed with distilled water three times, immersed in 0.37% ammonium sulfide for 1 minute, rinsed with distilled water again and mounted for light microscopic observations. Dark brown deposits of lead sulfide indicated the sites of OTC activity.

C. Determination of Urinary Orotate Mouse urine was collected at day -3, day

-1, and at different time points after virus infusion by leaving the animals in a metabolic cage overnight. Urinary orotic acid levels were measured in duplicate for each sample as described by Brusilow, S. W. , and Hauser, E. , J. Chromatograph.. 493: 388-391 (1989).

D. Determination of Plasma Amino Acids Plasma amino acids were analyzed by precolumn derivitization with o-phthaldialdehyde as previously described by Robinson et al, J. Neurochemi.. £1: 2099 2103 (1993)]. After centrifugation of heparinized blood, an aliquot of plasma was immediately precipitated with an equal volume of 0.8 N perchloric acid which contained the internal standards L-a-aminoadipate and L-a-amino-n- butyric acid. After centrifugation, an aliquot of the supematants was neutralized with 2 M KHC0₃. These samples were derivitized using an autosampler. External standards were injected after every fifth specimen.

E. Method of RNA hybridization analysis Total cellular RNA was isolated, fractionated on formaldehyde gel, and transferred onto Hybond-N nylon filters (Amersham, Arlington Heights, IL) . DNA fragments used as probes in RNA hybridizations were gel-purified and labeled with [α-³²P]dCTP by random priming.

F . Immunocytochemical Analysis

Immunofluorescence staining of adenoviral late gene products was performed as described by Kozarsky et al., cited above. The primary antibody was a polyclonal rabbit antibody specific to Ad5 late gene products (produced in Dr. Wilson's laboratory). The secondary antibody was a fluorescein isothiocyanate (FITC)-labeled goat anti-rabbit IgG (Chemicon, Temecula, CA) .

Example 4 - Nitrogen Challenges in Mice

The clinical and metabolic consequences of acute hyperammonemia that occur in OTC deficient humans, were simulated in the spf mouse to study the impact of in vivo gene therapy using vectors described herein. Protocol A: Individual cohorts of male C3HeB/J and spf/Y mice (6-10 weeks old) were weighed and injected intraperitoneally with a single dose (ranging from 4 to 10 mmol/kg) of a 0.64M solution of NH₄C1 (i.e., nitrogen challenge) . At 15-20 minutes after injection, blood was collected from the retro-orbital plexus of the mice and analyzed via HPLC for plasma amino acids and the mice were evaluated and scored by two observers using the scoring system described below. Protocol B : Blood samples were collected by retro-orbital bleeding three days before the experiment (day -3). On day 0, Spf/Y and C3H male mice at 6-10 weeks of age were injected with lxlO¹¹ particles of recombinant second generation adenovirus, either

H5.llOCMVmOTC or H5.HOCMVlacZ, in 0.1 ml of phosphate buffered saline (PBS) via the tail vein. Control mice were injected with PBS only. The animals were then injected with NH_tCl (10 mmol/kg) on days 1, 2, 7, 14, or 28 post virus infusion. Blood was collected from the retro-orbital plexus of the mice and analyzed via HPLC for plasma amino acids and the mice were evaluated and scored by two observers using the scoring system described below. Clinical Scoring System

The clinical scoring system was developed to quantify the clinical manifestations of hyperammonemia. After nitrogen challenge, clinical manifestations of hyperammonemia were scored using a system based on the appearance of ataxia (A) , seizures (S) and response to sound (R) . Ataxia is determined by pulling on the mouse's tail and observing gait. The response is scored as follows: 2, if the mouse is able to ambulate normally; 1, if the mouse staggers away; 0, inability to walk. Seizure intensities are scored as follows: 2, no seizure; 1, spontaneous myoclonus with spontaneous jerking movements and 0, tonic-clonic seizures characterized by rigid extension of all limbs. Finally, hyper- responsiveness to sound is determined by ringing a bell with sound frequency of 100 db 5 to 6 times in series. A normal mouse exhibiting no response to sound scores 3; twitching of extremities, 2; jumping, 1 and moribund, unable to right itself, 0.

Under this scoring system, a normal mouse is expected to receive an additive score of seven (A2S2R3) . Severely affected mice would receive a score of one (A0S1R0) . Any mice that die during the challenge would automatically receive a score of 0. Mice that exhibited tonic-clonic seizures always died after the seizures. The observers did not know the identity of animals in terms of treatment group during the period of observation.

Statistical Analysis

Statistical comparison of plasma amino acid values in mice before and after the nitrogen challenge were performed by the unpaired-sample t test. Plasma amino acid data and the clinical scores of responsiveness to nitrogen challenge from mice that received different treatments, e.g., untreated, lacZ treated, or OTC treated, were analyzed by the single factor analysis of variance (ANOVA) . Results

This study presents evidence of definitive clinical benefit of in vivo gene therapy that targets an OTC gene to liver cells in an authentic animal model for a lethal metabolic disease, as observed in Fig. 6 and Table I.

Fig. 6 presents the relationship between dose of NH₄C1 and clinical seguelae in the animals treated under Protocol A. Acute hyperammonemia, similar to that observed in patients, was simulated in spf mice following i.p. injection of NH₄C1 (i.e., nitrogen challenge). There was a direct correlation in spf animals between dose of NH₄C1 and severity of symptoms ranging from essentially no symptoms at 4 mmol/kg to virtually complete mortality at 10 mmol/kg. Animals developed ataxia, seizures, and frequently coma and death within 20 minutes of nitrogen challenge concurrent with the accumulation of plasma aspartate, alanine, and gluta ate. Under the same conditions, congenic normal C3H mice challenged with identical doses of NH₄C1 remained asymptomatic at all doses except 10 mmol/kg, where a few animals demonstrated nonlethal findings, without changes in plasma amino acids. A direct comparison of C3H to spf animals challenged with nitrogen revealed significant differences in clinical scores at all doses [i.e., 6 and 8 mmol/kg, (p<0.001 ) and 10 mmol/kg (p<0.05)] except at 4 mmol/kg (p>0.05).

Injection of OTC containing adenovirus into spf mice led to normalization of enzyme activity in liver which prevented the accumulation of plasma amino acids and spared the animals clinical sequelae following nitrogen challenge. Biochemical and clinical benefits of gene therapy were evident in animals challenged with nitrogen 24 hours after gene therapy with peak efficacy observed at 7 days, returning to pre-therapy conditions by 1 month.

Statistical comparison (single factor ANOVA) among the four groups of mice (C3H, spf/Y, spf/Y-OTC, spf/Y-lacZ) in Protocol B revealed significant differences between spf/Y and spf/Y-OTC mice at all time points (P<0.001 at day 1, 2, 7, and 14 p.i.; P<0.05 at day 28 p.i.). There was no significant difference between C3H and spf-OTC mice up to day 14 p.i., but the two groups of animals showed a significant difference at day 28 p.i. (P<0.001). There was no difference between spf and spf-lacZ mice at all time points except day 2 p.i. (P<0.05) . Table IA

Plasma Amino Acid Levels in Mice After NH4CI Challenge

Plasma Amino Acid fμM)

NH4CI

Row Mice Virus Challenge Ala Glu

1 none no 650+20 43.0+2.3 2 none yes 570+30 28.0+2.0-5

C3H

3 lacZ day 2 p.i. 634±31 36.0+5.7 4 lacZ day 7 p.i. 953+125# 40.7+4.0 5 lacZ day 14 p.i, 934+284 46.5+13.5

6 none no 619+47 45.8+6.3 7 none yes 1857+134§ 100.0±10.4§

8 lacZ day 1 p.i. 2020+300 150.2±31.4*

9 lacZ day 2 p.i. 1550±275 184.5±25.3***

10 spf lacZ day 7 p.i. 1368+190** 87.3+16.4

11 lacZ day 14 p.i 1844+127 143.8±17.5**

12 lacZ day 28 p.i 1560+187 76.0+13.8

13 OTC day 1 p.i. 1314±227* 72.6+19.9 14 OTC day 2 p.i. 640+51*** 48.0+6.4*** 15 OTC day 7 p.i. 983+83*** 52.0+3.7*** 16 OTC day 14 p.i 1031+154*** 54.5+10.0** 17 OTC day 28 p.i 2297+85** 83.9+5.1

I Statistically different from the corresponding values in Row 1 by the Unpaired-Sample I Test, P < 0.005.

S Statistically different from the corresponding values in Row 6 by the Unpaired-Sample I Test, P < 0.005.

# Statistically different from the corresponding values in Row 2 by the single factor ANOVA, P < 0.05.

## Statistically different from the corresponding values in Row 2 by the single factor ANOVA, P < 0.005.

* Statistically different from the corresponding values in Row 7 by the single factor ANOVA, P < 0.05.

** Statistically different from the corresponding values in Row 7 by the single factor ANOVA, P < 0.01.

*** Statistically different from the corresponding values in Row 7 by the single factor ANOVA, P < 0.005. Table IB

Plasma Amino Acid Levels in Mice After NH4CI Challenge

Plasma Amino Acid (uK)

NH4CI

Row Mice Virus Challenge Asp Gin

1 none no 13.3±1.2 431+15 2 none yes 10.1+1.2 465+20

C3H

3 lacZ day 2 p.i. 5.5±0.4 468+18 4 lacZ day 7 p.i. 14.8+3.3 727+54## 5 lacZ day 14 p.i. 16.5±6.5 663+52*

6 none no 15.6+0.9 745+305

7 none yes 127.9+40.3§ 670+44

8 lacZ day 1 p.i. 131.7+40.7 571+22*

9 lacZ day 2 p.i. 86.8+27.1 599+28

10 spf lacZ day 7 p.i. 71.8+18.7 588+46

11 lacZ day 14 p.i. 103.9±16.8 587+28

12 lacZ day 28 p.i. 135.3+33.9 939+28***

13 OTC day 1 p.i. 38.8+13.8 556+35* 14 OTC day 2 p.i. 6.5±0.7*** 573+40 15 OTC day 7 p.i. 24.3±2.3*** 712+23 16 OTC day 14 p.i. 15.5+2.3*** 614+42 17 OTC day 28 p.i. 178.7+17.7 1004+20***

1 Statistically different from the corresponding values in Row 1 by the Unpaired-Sample I Test, P < 0.005.

*** Statistically different from the corresponding values in Row 7 by the single factor ANOVA, P < 0.005. Table IC

Plasma Amino Acid Levels in Mice After NH₄C1 Challenge

Plasma Amino Acid fuM)^~

NH₄C1

Row Mice Virus Challenge Cit Arg

1 none no 55.8+2.8 97.5±9.3 2 none yes 86.0±5.5f 99.1+15.1

C3H

3 lacZ day 2 p.i. 71.5±8.8 99.0±23.3 4 lacZ day 7 p.i. 73.5+6.0 33.8+12.7## 5 lacZ day 14 p.i. n/a n/a

6 none no 19.7±1.0f 65.1±2.95 7 none yes 24.2+2.9 52.3+8.0

8 lacZ day 1 p.i. 27.5+2.8 68.2+5.6

9 lacZ day 2 p.i. 20.0+0.7 33.5+3.2

10 spf lacZ day 7 p.i. 34.8+2.8 27.0+9.7

11 lacZ day 14 p.i. n/a n/a

12 lacZ day 28 p.i. 16.8+1.2 22.8+1.6

13 OTC day 1 p.i. 28.2±3.3 77.6+4.4 14 OTC day 2 p.i. 20.0±1.7 57.3+8.6 15 OTC day 7 p. i. 35.3+2.2 39.0+16.5 16 OTC day 14 p.i. n/a n/a 17 OTC day 28 p.i. 21.0+1.1 31.6+2.3

*** Statistically different from the corresponding values in Row 7 by the single factor ANOVA, P < 0.005. Table ID

Plasma Amino Acid Levels in Mice After NH C1 Challenge

Plasma Amino Acid (uM)

NH₄C1

Row Mice Virus Challenge Lys

1 none no 179+14 2 none yes 204+15

C3H

3 lacZ day 2 p.i. 159+4 4 lacZ day 7 p.i. 281+31

5 lacZ day 14 p.i. 302+22

6 none no 201+9

7 none yes 288±11§

8 lacZ day 1 p.i. 273±28

9 lacZ day 2 p.i. 157+9***

10 spf lacZ day 7 p.i. 239+42

11 lacZ day 14 p.i. 243+22

12 lacZ day 28 p.i. 208+18***

13 OTC day 1 p.i. 274+30 14 OTC day 2 p.i. 177+11* 15 OTC day 7 p.i. 259+30 16 OTC day 14 p.i. 221+17 17 OTC day 28 p.i. 231+14

*** Statistically different from the corresponding values in Row 7 by the single factor ANOVA, P < 0.005. Table 1 summarizes the impact of OTC gene delivery using adenoviral gene therapy vectors on plasma amino acids following nitrogen challenge when evaluated 1, 2, 7, 14, and 28 days after gene therapy. Rows 1 and 6 of Table 1 represent a baseline of plasma amino acids in both C3H and asymptomatic spf/Y animals maintained on a normal chow diet (see, rows 1 and 6 of Table 1) .

The urea cycle intermediates citrulline and arginine were decreased in spf animals to 25% (p<0.005) and 70% (p<0.005) of levels founds in C3H mice providing evidence for defective ureagenesis. Plasma glutamine and urine orotate were elevated 2-fold (p<0.005) and 10-fold (data not shown) , respectively, in spf mice relative to C3H animals consistent with systemic accumulation of nitrogen in the setting of partial OTC deficiency.

Amino acids were measured as described in Example 3 in blood harvested prior to, and 20 minutes following, challenge with NH₄C1. No significant differences in plasma amino acids consistent with systemic nitrogen accumulation were noted in C3H animals following nitrogen challenge. This was not the case in spf animals who realized 3-fold increase in alanine (p<0.005), 8-fold increase in aspartate (p<0.005), and 2-fold increase in glutamate (p<0.005) after nitrogen challenge.

These results demonstrate that serum amino acids do not accumulate in nitrogen-challenged spf mice treated with OTC gene therapy. Gene therapy prevented the accumulation of alanine, aspartate, and glutamate in animals challenged with NH₄C1 2, 7, and 14 days after infusion of vector. Accumulation of these metabolites in spf animals at these time points was significantly less than what was observed in spf animals not pre-treated with vector (p<0.005). Partial correction of amino acid accumulation that was not statistically significant was observed at day 1 with values returning to baseline by day 28.

Analysis of liver tissue from selected animals demonstrated the following levels of OTC enzyme activity relative to that found in untreated C3H animals: day 0 (5%), day 1 (40%), day 2 (109%), day 7 (86%), day 14 (67%) , and day 28 (3%) .

The C3H and spf animals were treated with equivalent doses of lacZ virus to evaluate nonspecific effects of gene therapy on nitrogen metabolism and baseline ureagenesis. This is of concern because adenoviral vectors cause some degree of liver inflammation in mice at the doses used in this experiment (see below) . The results demonstrated that serum glutamine (p<0.005) and alanine (p<0.005) were mildly and transiently increased in C3H mice following gene transfer. Analysis of nitrogen-challenged spf/Y animals subsequent to infusion with lacZ virus demonstrated findings similar to that observed in untreated spf/Y animals: there was no statistically significant difference in elevated serum alanine, aspartate, and glutamate that occurred in nonvector treated versus lacZ vector treated spf mice (p<0.05 for most groups). This demonstrates that transgene derived OTC is responsible for preventing elevations in plasma amino acids following nitrogen challenge and confirms that virus-induced hepatotoxicity does not compromise residual ureagenesis in spf/Y animals.

Previous studies have demonstrated vector-induced, dose-dependent liver inflammation that peaks at days 7 - 10 and begins to resolve by day 289. Inflammation in this setting is probably multifactorial and includes antigen specific T cell responses to the genetically corrected hepatocytes, as well as direct toxicity of the virus. Little, if any, compromise in ureagenesis was detected in C3H and spf animals following the nonspecific effects of the virus.

Spf/Y animals were evaluated for NH₄C1 induced neurological abnormalities at different intervals following gene therapy. These data are presented in Fig.

7. Clinical manifestations of nitrogen challenge were prevented in spf animals treated with OTC vector. A statistically significant improvement in clinical outcome was detected as little as 24 hours after gene therapy that persisted through day 14 (p<0.001 for days 1, 2, 7, and 14) with a gradual decline to a nonvector treated phenotype by day 28 (p=0.05).

The clinical score was essentially unchanged in lacZ vector treated C3H mice. Spf mice treated with lacZ virus demonstrated clinical abnormalities following nitrogen challenge that were not significantly different than those observed in untreated spf animals after nitrogen challenge (p>0.05).

These studies demonstrate clinical efficacy of liver-directed gene therapy in an authentic animal model of OTC deficiency.

Example 5 - In vivo Delivery of Recombinant Adenoviruses to Mouse Liver

A . Experimental Protocol Spf/Y, spf„_h/Y and male C3HeB/J mice at

6-10 weeks of age were used in these experiments. Blood samples were collected by retro-orbital bleeding the day before the experiment (day -1) . Urine samples were collected at day -3 and day -1. Urine and plasma samples were collected at weekly intervals after viral infusion.

One group of Spf/Y mice were infused via the tail vein on day 0 with 5x10" particles of first generation H5.OlOCBhOTC or second generation H5.110CBhOTC human OTC-based recombinant adenovirus suspended in 0.1 ml of phosphate-buffered saline (PBS) . Animals were sacrificed at day 4, 7, and 14 post infusion.

Another group of Spf/Y mice were infused with 2xlOⁿ particles of first generation H5.OlOCMVmOTC or second generation H5.llOCMVmOTC mouse OTC-based recombinant adenovirus following the same protocol and these animals were sacrificed at day 7, 14 and 28.

Uninfected C3HeB/J, heterozygous spf/+, and hemizygous spf/Y mice were used as controls. Liver tissues from each animal were prepared for histochemical, biochemical, and molecular biological analysis.

B . Experimental Protocol

One group of Spf/Y mice were infused via the tail vein on day 0 with 5xl0¹⁰ particles/mouse of first generation H5.OlOCBhOTC or second generation H5.110CBhOTC human OTC-based recombinant adenovirus suspended in 0.1 ml of phosphate-buffered saline (PBS). Animals were sacrificed at day 7, 14, 21 and 28 post infusion.

A second group of Spf/Y mice were infused with lxlO¹¹ particles of first generation H5.OlOCMVmOTC or second generation H5.llOCMVmOTC mouse OTC-based recombinant adenovirus following the same protocol and these animals were sacrificed at day 7, 14, 21 and 28.

Liver tissues were harvested and evaluated for evidence of histopathology.

C. Experimental Protocol

Spf/Y mice at 6-8 weeks of age were infused with 5xlO^u particles of first generation viruses (H5.OlOCBhOTC or H5.010CMVlacZ) or second generation viruses (Hδ.llOCBhOTC or H5.HOCBlacZ) through tail vein. Urine and plasma samples were collected the day before the virus infusion, and at day 4, 7, and 14 post-infusion. D. Experimental Protocol

Spf/Y mice were infused with 2xlθⁿ particles of the first generation (H5.OlOCBhOTC or H5.OlOCMVmOTC) or second generation (H5.HOCBhOTC or H5.llOCMVmOTC) recombinant adenovirus. At day 4 post infusion, total liver RNA (10 μg) was isolated, fractionated in denaturing formaldehyde-agarose gels, transferred to nylon filter and hybridized with probes specific to the later viral gene product hexon or DBP. E. Experimental Protocol

Liver tissues of spf/Y mice infused with 2xlO^π particles of recombinant virus as described in Protocol D were harvested 4 days later. Fresh frozen sections (6 μm) were fixed in 100% methanol for 10 minutes, and analyzed for adenoviral late gene expression by immunofluorescence using an antibody specific to hexon.

F. Experimental Protocol

Spf/Y mice were infused with 2xlO^π particles of one of the following recombinant adenoviruses: H5.OlOCBhOTC, H5.OlOCMVhOTC, H5.OlOCMVmOTC, or H5.010CMVlacZ. Untreated spf mice and untreated C3H mice were used as controls. Liver tissue was harvested 3 days post infusion and analyzed for OTC activity by lysate enzyme analysis.

G. Experimental Protocol

Spf/Y mice at 6-8 weeks of age were infused with 2xl0¹¹ particle of first generation (H5.OlOCMVmOTC) or second generation virus (H5.llOCMVmOTC or Hδ.HOCMVlacZ) through tail vein. Spf^/Y mice of similar age were infused with 2xl0ⁿ particle of the second generation virus (H5.llOCMVmOTC) . Urine and plasma samples were collected at day -3, day -l, and at weekly intervals after virus infusion. H. Results of Assays Performed on Tissues The above experiments and assays demonstrated that the maximally-tolerated dose of recombinant first generation virus containing human OTC (H5.OlOCBhOTC) necessary to significantly increase OTC enzyme activity in liver in spf mice (i.e., 5 x 10^u particles) resulted in only a modest increase in OTC activity over baseline. This represents 5-fold more virus than is necessary to transduce >80% of hepatocytes based on experiments with similar vectors expressing a variety of reporter genes [see, e.g., Kozarsky et al, cited above; Morsy et al, cited above and Engelhardt I, cited above] .

Histochemical analyses of OTC activity in liver tissues from the spf mice infused with recombinant adenoviruses (magnification 100X) in Experimental Protocol A characterized the distribution and level of OTC expression. Analyses of C3H animals showed a dark brown reaction product (lead sulfide) in 100% of cells. This reaction product was absent in spf hemizygotes.

Heterozygotes showed two populations of OTC expressing cells consistent with lyonization of the X chromosome.

Histochemical analysis of spf liver removed 4 days after infusion of 5 x lθ" particles of first generation human OTC vector as described in

Experimental Protocol A revealed low level expression in most cells, diminishing to baseline by days 14 - 28, concurrent with the development of substantial but self-limited hepatitis (Fig. 2A) . Liver tissues were harvested from the animals of Experimental Protocols A and B at indicated time points in Figs. 2A through 2F following infusion and evaluated for evidence of histopathology by light microscopic inspection of paraffin sections stained with hematoxylin and erosin. Histopathology was evaluated using the criteria developed by Knodell R. et al, Hepatol.. 1: 431-435 (1981)]. The three criteria scored were: I, periportal & bridging necrosis; II, intralobular degeneration & focal necrosis; and III, portal inflammation. Analyses were performed on three to four animals per time point. The histogram shown in Figs. 2A through 2F is the average of at least three independent observations with SEM shown as error bars.

In contrast, histochemical analysis demonstrated OTC activity in the majority of hepatocytes of spf animals treated with first generation mouse OTC virus (Experimental Protocols A and B) . Transgene expression diminished with time, although OTC activity was detected in at least one-third of hepatocytes at day 28 of spf animals treated with first generation mouse OTC virus.

Expression of OTC enzyme as measured by the histochemical stain was higher and prolonged in the second generation mOTC virus than in the first generation viruses or the hOTC containing viruses. Inflammation was reduced in the second generation mOTC virus than in the first generation viruses or the hOTC containing viruses (Fig. 2B) .

Figs. 3A and 3B illustrate urinary orotate excretion in spf mice infused with recombinant adenoviruses carrying human OTC cDNA, as described in Experimental Protocol C. Urinary orotate/mg creatinine are presented as a % of pretreatment levels and are the mean + SEM of at least 6 determinations. There was no significant change in urinary orotate (Fig. 3A) in spf mice after infusion of 5 x 10^π particles of first generation human OTC vector H5.OlOCBhOTC, when compared to animals that received identical doses of H5.010CMVlacZ virus. Urinary orotate nonspecifically decreased to approximately 50% of pretreatment levels with both viruses H5.OlOCBhOTC and H5.010CMVlacZ possibly due to the associated hepatitis.

Fig. 3C illustrates plasma glutamine levels in spf mice of Experimental Protocol C infused with second generation recombinant adenoviruses carrying human OTC CDNA. The levels are presented as a % of pretreatment levels and are the mean + SEM of between 4 to 10 determinations. There was no significant change in serum glutamine in spf mice after infusion of 5 x 10^U particles of first generation human OTC vector

H5.OlOCBhOTC, when compared to animals that received identical doses of H5.OlOCMVlacZ virus.

An RNA blot analysis of liver tissues from spf mice infused with recombinant adenoviruses according to Experimental Protocol D demonstrated a large blot for the hexon RNA in liver RNA from spf/Y mouse that received first generation human OTC virus and first generation mouse OTC virus. Liver RNA from spf/Y mouse that received first and second generation human OTC virus and first and second generation mouse OTC virus revealed large blots for DBP. RNA from untreated spf liver showed neither the hexon nor DBP blots. The intensity of ribosomal RNA (18s and 28s) was similar in each lane, indicating equivalent quantities of electrophoresed RNA. Expression of late viral genes at the level of RNA was reduced in the second generation MOTC virus than in the first generation viruses or the HOTC containing viruses.

Evaluation of adenoviral late gene expression in liver tissue from these spf mice infused with recombinant adenoviruses as described in Protocol E demonstrated that first generation CMV/β-actin driven human OTC CDNA constructs (H5.OlOCBhOTC) and first generation CMV driven mouse OTC CDNA constructs (H5.OlOCMVmOTC) expressed detectable levels of the adenoviral hexon gene product. Using the corresponding second generation CDNA constructs, i.e., H5.110CBhOTC and H5.llOCMVmOTC, expression of late viral genes at the level of protein was reduced. This reduction in expression was more marked with use of the second generation MOTC virus than with use of the first generation murine OTC virus or the HOTC-containing viruses.

The OTC lysate assay performed on liver tissue from Experimental Protocol F produced the results illustrated in Fig. 4. Consistent with the above results, the highest OTC activity was detected in the spf mice treated with H5.OlOCMVmOTC. The OTC activity detected in the spf mice treated with H5.OlOCMVhOTC was equivalent to the OTC activity in the control C3H mice. The original vector expressing human OTC from the

CMV/β-actin promoter produced little enzyme activity above background in mouse spf liver. A three-fold increase in activity was achieved when the CMV promoter/enhancer was used to express the human OTC CDNA. An additional two to three-fold increase was achieved when the human OTC CDNA was replaced with the murine homolog in the CMV based vector. The results from Protocol F demonstrated that normal OTC enzymatic activity was achieved with 2 to 5-fold less mouse OTC virus than the maximally tolerated dose of human OTC virus, which only partially corrected mouse OTC deficiency. This further supports the observation that the human OTC CDNA product functions inefficiently in restoring OTC levels in this mouse model. Urinary orotic acid levels and plasma glutamate levels were measured as described above in Example 3 for the animals of Protocol G. The results are shown in Figs. 5A through 5D. Urinary orotate of animals treated with first generation mouse OTC virus decreased to 10% of pretreatment levels by day 7 and gradually returned to baseline by day 42 (Fig. 5A) . Correction of urinary orotate was significantly greater in animals treated with first generation MOTC virus than the nonspecific reduction seen with the lacZ virus, although it was not complete.

Abnormalities in urine orotate were completely normalized within 2-3 weeks of gene transfer in animals treated with the second generation MOTC virus. This metabolic parameter gradually returned to baseline levels but remained significantly improved in animals treated with the second generation MOTC virus as compared to animals infused with lacZ virus for 2-3 months. A similar correction of urine orotate was achieved in the other murine model of OTC deficiency, the spf*"¹ mouse, following infusion of second generation vector containing mouse OTC CDNA (Fig. 5C) . These experiments demonstrate complete correction of orotic acid overproduction following hepatic reconstitution of OTC in both the spf and spf**¹¹ mouse. With regard to the plasma amino acids, the impact of gene therapy was assessed and showed a normalization of serum glutamine reflecting a decrease in nitrogen stores. A 30% decline in serum glutamine was measured in animals treated with first generation MOTC. Correction of serum glutamine in this case lagged behind biochemical correction of enzyme activity and the peak reduction in urine orotic acid. The reason for this is unclear. However, it may be due to the combined effects of increased OTC enzyme expression with the superimposed but transient decline in liver function that occurs because of adenovirus induced hepatitis.

Animals treated with first generation murine OTC showed the most substantial improvement, realized 20 days after gene transfer, with serum glutamine approaching normal levels, and thereafter returning to baseline within 42 days of gene transfer (Fig. 5B) . In contrast, abnormalities in glutamine were completely normalized within 2-3 weeks of gene transfer in animals treated with the second generation MOTC virus. This metabolic parameter gradually returned to baseline levels but remained significantly improved in animals treated with the second generation MOTC virus as compared to animals infused with lacZ virus for 2-3 months. A similar correction of serum glutamine was achieved in the other murine model of OTC deficiency, the spf**¹¹ mouse, following infusion of second generation vector containing mouse OTC CDNA (Fig. 5D) .

However, adenovirus-mediated OTC gene transfer did not significantly impact on depleted serum citrulline. This is consistent with the experience of orthotopic liver transplantation in human carbamyl phosphate synthetase and OTC deficiencies, where all serum amino acids are corrected except citrulline, which remains low [Largilliere et al, cited above; Tuchman, M. , New Engl. J. Med.. 320: 1498-1499 (1989)].

The suboptimal performance of the first generation El deleted vector carrying either MOTC or HOTC is attributed, in part, to the inherent immunogenicity of first generation constructs. El deleted viruses express viral genes whose proteins are targets for destructive cellular immune responses. In contrast, the levels of viral late gene RNA and protein in the second generation viruses are diminished over those observed with the first generation viruses; associated hepatitis is also decreased.

The paucity of OTC expression achieved with the human OTC-based vectors in the mouse models, and the lack of significantly enhanced performance in second generation vectors containing the tsl25 mutation suggests problems with the human OTC CDNA functioning in the mouse model. Expression of the human OTC CDNA is still low, but slightly more stable in the second generation virus from that in the first generation virus.

One explanation is that the human OTC protein is not properly processed in a mouse cell to form catalytically-active, mitochondrial-localized enzyme. Direct sequence analysis of viral DNA confirmed the structure of the human OTC minigene ruling out the trivial explanation of mutation or rearrangement. Analysis of RNA from liver of treated animals revealed recombinant derived human OTC MRNA in excess of the level of endogenous OTC transcript found in normal human liver suggests possible post-transcriptional inefficiencies. Differences in amino acid sequence between the OTC murine and human homologs may affect the ability of the human enzyme to fold, oligomerize, and/or traffic to the mitochondria where it can function. In fact, 10 of the 26 amino acid differences present between mouse and human OTC are located in the leader sequence. Another explanation for the apparent dysfunction of the human OTC CDNA in mouse is that the human protein is viewed as a neoantigen in the spf mouse eliciting confounding and destructive immune responses in the mouse. Analysis of mice infected with human OTC adenovirus detected cytotoxic T cells but not antibodies to human OTC protein. While primary antigen specific cellular immune responses may decrease stability of transgene expression at day 7 and beyond, they cannot limit efficiency at early time points such as day 3. Previous studies have clearly implicated cellular immunity in the loss of transgene expression and associated inflammation that has characterized El deleted adenoviruses [Yang I, II, and III, cited above; Yang, Y. et al, Proc. Natl. Acad. Sci.. U.S.A.. 92: 7257-726 (1995) (Yang IV)]. It has been difficult quantifying the relative contribution of viral versus transgene protein to activation of destructive cellular immune responses. Most experiments have used transgenes whose protein products are easily distinguished from endogenous protein but which run the risk of being neoantigens (e.g. , lacZ luciferase, chloramphenicol acetyltransferase, etc.)

Previous experiments with El deleted viruses containing lacZ demonstrated CTLs to both viral proteins and E. coli β-galactosidase. However, adoptive transfer experiments and experiments in knock-out mice or those whose immune functions where depleted with antibody indicated immune responses to viral antigens are sufficient to ablate transgene expression [Yang, Y. et al, Gene Therapy, In press (1995) (Yang V) ] . Characterization of the performance of adenoviral vectors containing normal mouse OTC CDNA provides an opportunity to directly evaluate the relative contribution of viral protein versus transgene product in eliciting destabilizing cellular immunity. This is the first example in which adenovirus-mediated gene transfer has been performed with a transgene whose product should not be viewed as a neoantigen. Transgene-derived MOTC differs by only one amino acid from the spf protein and is identical to the product of the spf^Mh allele. Transgene expression with the mouse OTC CDNA vectors persisted longer than what has been consistently observed with vectors expressing non-self transgenes such as β-galactosidase. Incorporating the tsl25 mutation into this vector diminished late viral gene expression and further prolonged transgene expression. The expression of transgene eventually diminished to undetectable levels within 3-4 months of gene transfer. Example 6 - Recombinant Adenovirus Containing Human Argininosuccinate Synthetase (AS) CDNA

A plasmid pAS4/l/9 containing human AS cDNA was obtained from the ATCC [catalog #57074], An 1.5 kb PstI fragment containing AS cDNA was isolated, blunted and cloned into EcoRV site of an adenoviral vector pAd.CMV- linkl [X. Ye et al, J. Biol. Chem.. 221:3639-3646 (1996)]. The new plasmid, designated pAd.CMVhAS, was linearized with Nhel and cotransfected into 293 cells with Clal/Xbal restricted sub360 [J. Logan et al, Proc. Natl. Acad. Sci. USA. 81:3655-3659 (1984)] or sub360/tsl25 [M. Ensinger et al, J. Virol.. 10:328-339 (1972)] DNA. The resulting recombinant viruses (H5.010CMV2.AS and H5.HOCMVhAS) are purified by multiple- rounds of plaque purification and can be tested for functional AS activity by ¹ C-aspartate conversion [T. Su et al, Biochemistrv. £0:2956-2960 (1981)] and lysate [S. Ratner, in Methods in Enzymology II, pp. pp. 356-367, eds. S. Colowick and M. Kaplan, Academic, New York (1955) assays.

This vector is useful in liver-directed gene therapy as described above with respect to the OTC- containing vectors.

Example 7 - Recombinant Adenovirus Containing Human Arginase (ARG) cDNA

Human ARG cDNA was obtained from human liver tissue by RT-PCR using primers generated using the published sequence [Y. Haraguchi, Proc. Natl. Acad. Sci. USA, 84.:412-415 (1987)]. The primer sequences are: 5^» primer: 5'-AGCTCAAGTGCAGCAAAGAG-3 • [SEQ ID NO:l] 3' primer: 5'-TGACATGGACACATAGTACCT-3' [SEQ ID NO:2]. The ARG cDNA generated was cloned into pGEM-T vector (Promega, Madison, Wl) and restricted with NotI and Sphl. An 1.45 kb fragment containing human cDNA was isolated, blunted and cloned into EcoRV site of an adenoviral vector pAd.CMV-linkl [X. Ye et al, J. Biol. Chem.. 271:3639-3646 (1996)]. The new plasmid, designated pAd.CMVhARG, was linearized with Nhel and cotransfected into 293 cells with Clai restricted sub360/tsl25 [M. Ensinger et al, J. Virol.. 10:328-339 (1972) ] DNA. The resulting recombinant virus (H5.110CMVhARG) expressed functional human arginase in

293 cells as tested by ARG lysate assay [D. Greenberg, in Methods in Enzymology, II, pp. 368-374, eds. S. Colowick and M. Kaplan, Academic, New York (1955)].

A second arginase-expressing vector was constructed using the methods (including the same restriction enzyme site) described above for H5.110CMVhArg, except that the viral backbone utilized was dl7001. dl7001 is a type 5 adenovirus which contains a deletion in the E3 region between map units 78.4 through 86 and, more particularly, at nucleotides 594 and 3662 where nudeotide 1 corresponds to the EcoRI site of Ad5 at 27,331 bp. All E3 region open reading frames are deleted. The resulting vector is termed H5.020hArg. Hδ.HOCMVhArg and H5.020hArg were purified through three rounds of plaque isolation and have been found to express arginase activity in 293 cells. These vectors are useful in liver-directed gene therapy as described above with respect to the OTC-containing vectors.

Example 8 - Recombinant Adenoviruses Containing Human Carbamyl phosphate synthetase (CPS) cDNA

Human CPS cDNA was obtained from human liver tissue by RT-PCR using primers generated using the published sequence [Y. Haraguchi et al, Gene. 107:335-340 (1991)]. The primer sequences are as follows. 5• primer: 5'-AAGTCTTATCACACAATCTCATTAA-3' [SEQ ID NO:3] 3' primer: 5'-GCCCTGTTAAAGTGTCCTGAG-3' [SEQ ID NO:4]. To construct an adenovirus containing human CPS under control of a human cyto egalovirus promoter, the CPS cDNA was cloned into pGEM-T vector (Promega, Madison, Wl) to generate pGEM-CPS. pGEM-CPS is then restricted with Sail and SphI, and cloned into pAd.link (a plasmid containing the human Ad5 sequences, map units 0 to 16, which is deleted of Ela and Elb as described in X. Ye et al, J. Biol. Chem.. 271:3639-3634 (1996)). The resulting plasmid, pAd.CMVhCPS is linearized and co-transfected into 293 cells with Clal-digested dl327 virus DNA to generate H5.030CMVhCPS.

Similar methods may be utilized to construct an adenovirus containing human CPS under control of another promoter, e.g., a mouse albumin promoter. Suitable plasmids can be readily obtained or constructed by one of skill in the art. For example, pGEM-CPS can be restricted with Sail and SphI as described above and cloned into pAd.albBcl2 backbone [kindly provided by Shujen Chen] in which Bcl2 cDNA has been removed by BamHI digestion. The new plasmid, pAd.albhCPS, is linearized by Nhil and cotransfected with Clal-digested dl327 virus DNA into 293 cells to generate H5.030albhCPS. To construct an adenovirus containing human CPS under control of a metallothionien promoter, the human CPS cDNA generated is removed from pGEM-CPS as described above and inserted into the Nhrul site of pMT-LCR expression vector [kindly provided by R. Palmiter, University of Washington] . The hCPS mini-gene which includes MT promoter, hCPS cDNA and human growth hormone (hGH) polyadenylation sequence is removed by EcoRI and inserted into pAd.link, as described above. The resulting plasmid, pAd.MThCPS is linearized by Nhel and co¬ transfected into 293 cells with Clal-digested dl327 virus DNA to produce H5.030MThCPS.

The resulting hCPS vectors are assayed for activity in 293 cells as described in T. Nuzum and P.J. Snodgrass, The Urea Cycle (ed. S. Grisolia et al.), pp. 325-355, John Wiley and Sons, New York. These vectors are useful in liver-directed gene therapy as described above with respect to the OTC-containing vectors.

Example 9 - Recombinant Adenoviruses Containing Human Argininosuccinate lyase (AL) cDNA

Human AL cDNA is obtained from human liver tissue by RT-PCR using primers generated using the published sequence [W. E. O'Brien et al, Proc. Natl. Acad. Sci. USA. 8.3:7211-7215 (1986)]. The human AL cDNA is then cloned into pGEM-T vector (Promega, Madison, Wl) essentially as described in the examples above. The hAL cDNA is then inserted into pAd.CMVlink-1 [X. Ye et al, J. Biol. Chem.. 121:3639-3646 (1996)]. The resulting plasmid, pAd.CMVhAL, is linearized with Nhel and co¬ transfected into 293 cells with Clai digested H5.110 or H5.020 backbone.

The resulting hAL vectors are assayed for activity in 293 cells as described in T. Nuzum and P.J. Snodgrass, The Urea Cycle (ed. S. Grisolia et al.), pp. 325-355, John Wiley and Sons, New York. These vectors are useful in liver-directed gene therapy as described above with respect to the OTC- containing vectors.

Example 10 - Treatment of Hyperammonemia in Humans

The experiment illustrates the method of this invention for using gene therapy in the treatment of an hyperammonemia in a human patient with a defect in the argininosuccinate synthetase (AS) gene of the urea cycle. The recombinant virus H5.H0CMVhAS of Example 6 is administered intravenously to the adult human patient experiencing an acute episode of hyperammonemia due to the above-noted defect in a 1 ml dosage, i.e. a recombinant virus concentration of about 1 x IO¹¹ pfu/ml of saline solution. At the same time, a dosage of 1 μg/ml saline solution of anti-CD40L (Bristol-Myers Squibb Co.] is administered iv to the patient to suppress unwanted immune responses to the recombinant virus.

The levels of expression of the encoded urea cycle AS enzyme in serum are monitored periodically by conventional methods to determine the selection, adjustment or frequency of dosage administration and to assess the goal of reversing underlying metabolic derangements within 72 hours.

Using similar methods, the H5.HOCMVhArgr and H5.020hArg vectors of example 7 may be utilized in the treatment of a human patient with a defect in the arginase (ARG) gene of the urea cycle; the H5.030CMVhCPS, H5.030albhCPS, or H5.030MThCPS vectors of example 8 may be utilized in the treatment of a human patient with a defect in the carbamyl phosphate synthetase (CPS) gene of the urea cycle; and the H5.110hAL or H5.020hAL vector of example 9 may be utilized in the treatment of a human patient with a defect in the argininosuccinate lyase (AL) gene of the urea cycle. All articles identified herein are incorporated by reference. Numerous modifications and variations of the present invention are included in the above- identified specification and are expected to be obvious to one of skill in the art. Such modifications and alterations to the compositions and processes of the present invention are believed to be encompassed in the scope of the claims appended hereto.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: Trustees of the University of

Pennsylvania Wilson, James M.

(ii) TITLE OF INVENTION: Method of Treating Liver

Disorders

(iii) NUMBER OF SEQUENCES: 4

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Howson and Howson

(B) STREET: Spring House Corporate Cntr,

PO Box 457

(C) CITY: Spring House

(D) STATE: Pennsylvania

(E) COUNTRY: USA

(F) ZIP: 19477

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

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

(D) SOFTWARE: Patentin Release #1.0, Version #1.30

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: WO

(B) FILING DATE:

(C) CLASSIFICATION:

(vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 60/011,613

(B) FILING DATE: 13-FEB-1996

(VU) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 60/025,883

(B) FILING DATE: 06-SEP-1996

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Kodroff, Cathy A.

(B) REGISTRATION NUMBER: 33,980

(C) REFERENCE/DOCKET NUMBER: GNVPN.016CIP2PCT

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: 215-540-9200

(B) TELEFAX: 215-540-5818 (2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: other nucleic acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO:l: AGCTCAAGTG CAGCAAAGAG 20

(2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: other nucleic acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: TGACATGGAC ACATAGTACC T 21

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: other nucleic acid ( i) SEQUENCE DESCRIPTION: SEQ ID NO:3: AAGTCTTATC ACACAATCTC ATTAA 25 (2) INFORMATION FOR SEQ ID NOM:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: other nucleic acid (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: GCCCTGTTAA AGTGTCCTGA G 21

Claims

WHAT IS CLAIMED IS:

1. A method for enhancing ureagenesis in a subject in need thereof comprising administering to said subject an effective amount of a recombinant virus which expresses in vivo at least one urea cycle enzyme.

2. The method according to claim l, wherein said virus is a replication defective adenovirus comprising i. the DNA of, or corresponding to, a portion of the genome of an adenovirus which comprises adenovirus 5' and 3 • cis-elements necessary for replication and virion encapsidation in the absence of sequence encoding viral genes; and ii. a selected gene encoding a urea cycle enzyme operatively linked to regulatory sequences directing its expression, said gene linked to the DNA of (a) and capable of expression in a target cell in vivo or in vitro ; said DNA and gene encapsidated in a viral capsid.

3. The method according to claim l comprising administering more than one recombinant virus, each said virus expressing a different urea cycle enzyme.

4. The method according to claim 1 wherein said urea cycle enzyme is selected from the group consisting of carbamyl phosphate synthetase, ornithine transcarbamylase, argininosuccinate synthetase, argininosuccinate lyase, and arginase.

5. The method according to claim 1 wherein said subject is suffering from a genetic urea cycle enzyme deficiency.

6. The method according to claim 1 wherein said subject is suffering from nonspecific liver failure characterized by compromised ureagenesis.

7. The method according to claim 1 further comprising administering to said subject an effective amount of an immune modulator, said modulator substantially inhibiting the formation of neutralizing antibodies directed against said virus or of substantially reducing CTL elimination of said virus, whereby said virus may be readministered without immediate elimination by an immune response.

8. A method for treating a non-inherited disorder of the liver comprising administering a replication defective, recombinant adenovirus which expresses a urea cycle enzyme.

9. Use of a recombinant virus which expresses at least one urea cycle enzyme in a medicament for enhancing ureagenesis in a subject in need thereof.

10. Use of a replication defective, recombinant adenovirus which expresses a urea cycle enzyme in preparation of a medicament for treating a non- inherited disorder of the liver.