Gene therapy for human alpha1-antitrypsin deficiency in an animal model using SV40-derived vectors

GA

YD*�

C

BohbsdtacrtmmprhsRhmwvt�ghCtd

Tfpa(lbptb

GASTROENTEROLOGY 2004;127:1222–1232

ene Therapy for Human �1-Antitrypsin Deficiency in annimal Model Using SV40-Derived Vectors

U–YOU DUAN,* JIAN WU,* JIAN–LIANG ZHU,‡ SHU–LING LIU,‡ IWATA OZAKI,§

AVID S. STRAYER,� and MARK A. ZERN*Transplant Research Institute, University of California, Davis Medical Center, Sacramento, California; Departments of ‡Medicine and

Pathology, Anatomy and Cell Biology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, Pennsylvania; and §Saga Medical
ollege, Saga, Japan
lvcyg

dT�dhmtpfirRttabts(cmr

dOtt

z4s

ackground & Aims: In most genetic diseases, the goalf gene therapy is to deliver a particular transgene;owever, sometimes a deleterious gene product muste eliminated. Because of the promise of recombinantimian virus 40 (rSV40) vectors, we tested their ability toeliver a transgene and to target a transcript for destruc-ion by direct administration of the vectors to the liver ofn animal model for human �1-antitrypsin (�1-AT) defi-iency. Methods: Therapy of human �1-AT deficiencyequires stable transduction of resting hepatocytes, botho deliver wild-type �1-AT and to inhibit production ofutant �1-AT. Transgenic mice carrying the mutant hu-an �1-AT PiZ allele were treated through an indwellingortal vein catheter with a simian virus 40 (SV40)-de-ived vector carrying a ribozyme designed to target theuman transcript. Results: Treated transgenic micehowed marked decreases of human �1-AT messengerNA and the protein in the liver, and serum levels ofuman �1-AT were decreased to 50% � 5% of pretreat-ent values 3–16 weeks after transduction. Moreover,hen normal mice were treated with an SV40-derivedector containing a modified human �1-AT complemen-ary DNA engineered to be resistant to cleavage by the1-AT ribozyme, they expressed human �1-AT messen-er RNA and protein in their livers and serum levels ofuman �1-AT remained >1 �g/mL for 1 year.onclusions: These results represent the initial stepsoward a novel approach to the gene therapy of �1-ATeficiency.

o be effective, a viral gene delivery vehicle for liver-directed therapy of genetic diseases should (1) provide

or stable transgene expression sustained over a prolongederiod; (2) be relatively nonimmunogenic and nontoxic tovoid hepatocyte necrosis and allow repeat administration;3) be available at high titers to enable transduction of aarge percentage of hepatocytes; (4) transduce resting cells,ecause most hepatocytes are not cycling at any specificoint in time; and (5) be replication deficient. To satisfyhese requirements, we have established a vector systemased on simian virus 40 (SV40) for gene delivery to the

iver. Our previous studies indicated that SV40-derivedectors meet all of the criteria previously mentioned, in-luding stable expression of the transgene for more than 1ear. Thus, they are excellent candidates for in vivo hepaticene therapy.1–3

�1-Antitrypsin (�1-AT) deficiency is a genetic disor-er that leads to emphysema and chronic liver disease.he lung disease is believed to reflect insufficient normal

1-AT activity in the circulation,4 whereas the liverisease occurs because abnormal �1-AT accumulates inepatocytes.5 To address both the hepatic and pulmonaryanifestations of the disease, we developed a strategy

hat would both inhibit the synthesis of the abnormalrotein and lead to synthesis of the normal protein. Therst objective was accomplished by using hammerheadibozymes designed to cleave mutated �1-AT messengerNA (mRNA) at a specific site. We determined that

hese ribozymes inhibited �1-AT expression in a hepa-oma cell line, PLC/PRF/5.6,7 The second step in thispproach was to deliver functional �1-AT that would note recognized by the ribozymes used to target the mu-ated �1-AT mRNA. Therefore, we modified the baseequences of the human �1-AT complementary DNAcDNA) at the site targeted by the ribozyme, withouthanging the normal amino acid sequence, to yield anRNA product resistant to the site-specific hammerhead

ibozyme cleavage.7

In the present study, we report the use of the SV40-erived ribozyme and modified cDNA vectors in mice.ur results suggest that this novel approach for the gene

herapy of �1-AT deficiency is an effective strategy whenhe SV40-derived vectors are used as a delivery system.

Abbreviations used in this paper: �1-AT, �1-antitrypsin; ELISA, en-yme-linked immunosorbent assay; rSV40, recombinant simian virus0; RT-PCR, reverse-transcription polymerase chain reaction; SV40,imian virus 40.

© 2004 by the American Gastroenterological Association0016-5085/04/$30.00

doi:10.1053/j.gastro.2004.07.058

hgpsnppug(pcaglepparp

ysntmcrfinf(PrEafipfpce(v

�Bmat

rstb

F(tdwwSFatRppnwrbhP

October 2004 �1-ANTITRYPSIN DEFICIENCY AND GENE THERAPY 1223

Materials and Methods

Recombinant SV40 Vectors

Production of recombinant Tag-deleted SV40 virusesas been described elsewhere.1–3,7 Briefly, the Tag and tagenes were excised from the viral genome and replaced by aolylinker with bacterial phage SP6 and T7 promoters down-tream from 2 tandem SV40 early promoters. This viral ge-ome, cloned into pGEM13 as a NotI fragment,1–3 is calledSV5. The SP6 and T7 sequences were included in theolylinker to facilitate the sequencing of the cloned inserts bysing standard SP6 and T7 primers. The modified SV40 viralenome was recombined with a portion of pT7Blue plasmidNovagen, Madison, WI) at the PmeI site by removingGEM13 sequence from pSV5 with NotI to form a new SV40onstruct, pT7SVP. SV40 viral DNA in pT7SVP remainslmost the same as wild-type SV40 after removing Tag and tagenes. It contains intact late genes (3 capsid genes, one regu-atory gene), regulatory sequences (SV40 promoters, polyAs forarly and late genes, and so on), origin of SV40 replication, andacking signals. The pT7 plasmid DNA as a carrier inT7SVP contains the replication origin of the plasmid and anmpicillin-resistant gene. Recombinant SV40 (rSV40)-derivedeplication-deficient viruses could be produced by removingT7 plasmid carrier after gene cloning.A modified �1-AT cDNA resistant to an �1-AT ribozyme

et capable of producing a wild-type protein has been de-cribed in detail.7 The construct entails altering the thirducleotide of each amino acid codon in the region flanking thearget region of the AT589 ribozyme. As a template for theodified �1-AT cDNA, the full-length cDNA of �1-AT was

loned into pT7Blue-T vector to generate pT7 �1-AT byeverse-transcription polymerase chain reaction (RT-PCR)rom the RNA extracted from HepG2 hepatoma cells. Forntroduction of the mutations, 2 primers containing modifieducleotide sequences of �1-AT were synthesized. First, 2 PCRragments were amplified by using Vent DNA PolymeraseNew England Biolabs, Mississauga, Ontario, Canada). These 2CR products were mixed and annealed, and then an extensioneaction was performed with Taq DNA polymerase (Perkin-lmer, Branchburg, NJ). This mixture was used for PCRmplification as the template for the generation of the modi-ed �1-AT. The PCR products were then cloned into theT7Blue-T vector to generate pT7m �1-AT. A 1.3-kilobaseragment containing the modified �1-AT cDNA excised fromT7�-AT with XbaI and XhoI, blunted by Klenow, was thenloned into the blunted XhoI site of pT7SVP vector to bexpressed under the control of the 2 SV40 early promotersFigure 1). The orientation of modified �1-AT cDNA waserified by PCR using primers SP6 and T7.

The details of the construction and screening of functional

1-AT ribozymes have been reported by us previously.6

riefly, 11–15 bases of antisense sequence against �1-ATRNA were flanked on both sides of the hammerhead motif to

llow the ribozyme to associate with �1-AT mRNA throughheir complementary sequences. For the construction of the

ibozymes, 2 complementary oligonucleotides were synthe-ized on the DNA synthesizer (model 392; Applied Biosys-ems, Inc., Foster City, CA). The ribozymes were synthesizedy incubating 2 oligonucleotides to form a hemiduplex, and

igure 1. Schematic illustration of rSV40-derived vectors: (A) pSV-mAT) and (B) pSV(AT589T). (A) Tag and tag genes were excised fromhe wild-type SV40 genome, and a polylinker was replaced at theownstream site of the SV40 early promoter (SV40-P), which overlapsith the SV40 origin of replication. The modified SV40 viral genomeas recombined with a portion of pT7Blue plasmid to generate a newV40 construct, pT7SVP, as described in Materials and Methods.ull-length modified cDNA of human �1-AT was cloned into pT7SVP,nd the resulting vector is named as pSV(mAT), including an addi-ional copy of SV40-P head-to-tail with the existing viral promoter. (B)ibozyme 589T cassette was inserted into pT7SV(�) and resulted inSV(AT589T), in which ribozyme 589T is under the control of the tRNAromoter (tRNA-P) in this cassette containing PolIII termination sig-als. Both rSV40s were produced by removing pT7 plasmid carrierith PmeI, recirculating it, and transfecting it into COS-7 cells. The

esulting rSV40s packaged and amplified by COS-7 cells were thenand purified by ultracentrifugation. AT589T, human �1-AT ribozyme;m-alpha AT, the human modified �1-AT cDNA; SV40-PA, SV40 polyA;meI, restriction nuclease.

Ptptcp

ticeiSCsip

vaMsw1ct�totMqg�R

aSieitTtthTvpaec

je

pettiSssnvsotcwpCGlgbp(TTrCsto

mbufyp

aaidupaT�pm

1224 DUAN ET AL. GASTROENTEROLOGY Vol. 127, No. 4

CR amplifications were performed. The PCR products werehen cloned directly into the pT7Blue-T vector to generate alasmid pT7ATRzs. The ribozyme construct, containing aransfer RNA promoter,6 was then cloned into the multipleloning site of pT7SV(�) vector that was derived fromBSV-1, containing the modified SV40 genome (Figure 1).2

In all cases, recombinant viruses were produced by removinghe carrier plasmid from rSV40 vectors with PmeI, recircular-zing it, and transfecting it into a packaging cell line (COS-7ells). Tag is required for virus replication8,9; however, Tagxpressed by a packaging cell line can support viral replicationn trans.10 The resulting modified SV40 viruses were calledV(mAT) and SV(AT589T). Crude viruses from cell lysates ofOS-7 cells were band purified by discontinuous sucrose den-

ity gradient ultracentrifugation. Virus stocks were titrated byn situ PCR using a method we developed specifically for thisurpose.11

In Vitro Transduction

The human hepatoma cell lines HLE, HLF (kindly pro-ided by Health Science Research Resources Bank, Tokyo, Japan),nd PLC/PRF/5 (from American Type Culture Collection,anassas, VA) were grown in Dulbecco’s modified Eagle medium

upplemented with 10% fetal bovine serum. HLE and HLF cells,ith a very low level of �1-AT mRNA, were seeded at 5 �05 cells/60-mm dish and transduced the next day with SV(mAT)ontaining the modified �1-AT cDNA (100 �L of rSV40 at aiter of 1.2 � 1012 IU/mL). PLC/PRF/5 cells with a high level of

1-AT mRNA were also seeded at 5 � 105 cells/60-mm dish andransduced with SV(AT589T) containing the ribozyme (100 �Lf rSV40 at a titer of 1.2 � 1012 IU/mL). Total RNA fromransduced or untransduced cells was extracted using RNeasy

ini Kit (Qiagen, Valencia, CA) at 72 hours after transduction,uantitated, and kept at �80°C until further use. cDNA wasenerated from the RNA template and used for the evaluation of

1-AT mRNA levels in the transduced cells by quantitativeT-PCR as described in the following text.

Animal Experiments

Transgenic mice expressing mutant human �1-AT (PiZllele) were a kind gift from Stratagene, Inc. (La Jolla, CA).urgical procedures for the placement of an indwelling catheternto the portal vein and subsequent injections through the cath-ter were performed according to Vrancken et al.12 and describedn detail13 and approved by the Animal Care and Use Adminis-rative Advisory Committee of the University of California, Davis.he animals were anesthetized with pentobarbital (60 mg/kg) for

he surgical procedures and recovered 2–4 hours after the opera-ion. The transgenic mice, which were confirmed to possess theuman �1-AT PiZ gene by PCR, were used for the experiments.ransgenic mice or ICR mice were injected with the recombinantiruses multiple times through the indwelling catheter in theortal vein while under ether anesthesia. The injections followeddesigned protocol, as detailed in Results. The total volume of

ach injection was approximately 0.3 mL. Blood samples wereollected by tail incision at various intervals following viral in-

ection, and the animals were killed to obtain liver tissue at thend of the observation period.

In Situ RT-PCR

This method for evaluating transduced viral gene ex-ression, developed by Bagasra et al., has been describedlsewhere.14,15 Briefly, paraffin-fixed sections of the livers wereransferred to specially designed, siliconized slides. Slides werehen fixed and treated with proteinase K and hydrogen perox-de. To perform the amplification of RNA sequences for theV40 viral vector, reverse transcription was performed for thepecific sequences using antisense primers. For this purpose,pecimens were first treated with ribonuclease-free deoxyribo-uclease in a humidified chamber. Reverse transcription of theiral transcripts to cDNA was performed using the down-tream oligonucleotide (oligo[dT]20; Invitrogen Life Technol-gies, Carlsbad, CA) as a primer. A slide well without reverseranscriptase was used as a negative control to test for DNAontamination. A DNA/in situ/PCR amplification procedureas then performed using a viral gene-specific primerair (forward primer, 5=AAACTGTGACTGGTGTGAGCG-TG3=; reverse primer, 5=ACCCCAATGTCTGGGGTCAA-ATA3=, with an amplicon of 220 base pairs from the over-

apping DNA fragment between the SV40 VP1 and VP2enes). These slides were placed in a specifically designed heatlock of a thermocycler for amplification. Hybridization waserformed with the biotinylated oligonucleotide probes5=CAGGAATGGCTGTAGATTTGTATAGGCCAGATGA-TACTATGATATTTTATTTCCTGGAG3=; Sigma Genosys,he Woodlands, TX). The tissue sections were washed to

emove unbound probes and incubated with avidin-peroxidase.olor was developed with 3-amino-9-ethylcarbazole. The tis-

ues were counterstained with Gill’s hematoxylin. An addi-ional control for the in situ PCR experiments included the usef nonsense primer sets for vector amplification.

RNA Analysis

Total RNA both from the livers of transgenic and ICRice and from PLC/PRF/5, HLE, and HLF cells was extracted

y a modification of the method of Chomczynski and Sacchi16

sing the RNeasy Mini Kit (Qiagen). Expression of mRNArom livers was detected by Northern blot hybridization anal-sis as previously described6,7 using a human �1-AT cDNArobe and using human transferrin and 28S as controls.Gene expression of human �1-AT, mouse �1-AT, mouse

lbumin, and mouse �-actin from cells and mouse livers waslso evaluated by real-time quantitative RT-PCR as describedn detail by us previously.17 Following the digestion witheoxyribonuclease I, total RNA was used to generate cDNA bysing ThermoScript RT-PCR Systems and oligo(dT)20 (Ap-lied Biosystems) to prime first-strand synthesis. The gener-ted cDNA was used for quantitative PCR amplification.aqMan PCR was used to quantitate expression of human

1-AT in comparison with human-specific glyceraldehyde-3-hosphate dehydrogenase (in human hepatoma cell lines) orouse-specific �-actin (in mouse tissue) as housekeeping con-

t��toPaacphcs

eehwaumgCaiCwbQ

d�

imCb�amgabSctamp�d

l�ec

tPt0w

T

P

P

a

c


rols. SYBR PCR was used to quantitate expression of mouse

1-AT and mouse albumin in comparison with mouse-specific-actin as housekeeping control. Following amplification with

he TaqMan Universal PCR Master Mix (Applied Biosystems)r SYBR PCR Master Mix (Applied Biosystems) in the ABIrism 7700 Thermal Cycler (Applied Biosystems), semilogmplification curves were determined by calculation of ��CT

nd expression levels were normalized to the housekeepingontrols. All primers and probes are listed in Table 1 (primers/robe set for human-specific glyceraldehyde-3-phosphate de-ydrogenase amplification; Applied Biosystems) and were spe-ifically designed based on human or murine target geneequences.

Protein Extraction and Western BlotAnalysis

Proteins from livers of transgenic and ICR mice werextracted as previously described18 and were detected by West-rn blot analysis as previously described.6,7 Primary goat anti-uman �1-AT antibodies (MP Biomedicals, Inc., Irvine, CA)ere used for detection of human �1-AT, and primary rabbit

nti-mouse albumin antibodies (MP Biomedicals, Inc.) weresed for detection of mouse albumin. Donkey anti-goat im-unoglobulin (Ig) G horseradish peroxidase conjugate and

oat anti-rabbit IgG horseradish peroxidase conjugate (Santaruz Biotechnology, Inc., Santa Cruz, CA) were used as secondntibodies for imaging by enhanced chemiluminescence imag-ng reagents (Amersham Bioscience, Buckinghamshire, UK).ruz Marker MW standard (Santa Cruz Biotechnology, Inc.)as loaded for reference. Densitometry analysis for Westernlots was performed with a Molecular Dynamics Image-uaNTTM program (Sunnyvale, CA).

Enzyme-Linked Immunosorbent Assay forHuman �1-AT in Mouse Serum

Mouse blood samples were collected by tail incision atifferent intervals after the injection of viral vectors. Human-AT levels in mouse serum were quantified by enzyme-linked

able 1. Primer/Probe Sets Used in This Study

Gene Primer/probe sets

rimers for TaqMan PCRHuman �1-AT F: TCGCTACAGCCTTTGCAAT

P: AGCCTTCATGGATCTGAGCR: TTGAGGGTACGGAGGAGT

Mouse �-actin F: ACGGCCAGGTCATCACTATP: CAACGAGCGGTTCCGATGR: ATACCCAAGAAGGAAGGC

rimers for SYBR PCRMouse �1-AT F: GGGTGACACTCACACGCAG

R: TGGATGTCAGCCTCCGATGMouse albumin F: GCAAGGCTGCTGACAAGG

R: GGCGTCTTTGCATCTAGTGMouse �-actin F: ACGGCCAGGTCATCACTAT

R: ATACCCAAGAAGGAAGGC

All hybridization probes were quenched with 6-carboxytetramethylrhoein; R, reverse primer.

1

mmunosorbent assay (ELISA) with mouse anti-human �1-ATonoclonal antibodies (Chemicon International, Inc., Temecula,A) as coating antibodies (in 35 mmol/L absorption carbonateuffer, pH 9.6, coating overnight, at 4°C) and rabbit anti-human

1-AT IgG (Boehringer Mannheim, Indianapolis, IN) as capturentibodies in blocking buffer (1% bovine serum albumin in 1.0mol/L phosphate-buffered saline, pH 7.4). Peroxidase-conju-

ated goat anti-rabbit IgG (Dako, Carpinteria, CA) was applied asdetection antibody in washing buffer (0.1% Tween 20 in

locking buffer), and o-phenylenediamine (Sigma Chemical Co.,t. Louis, MO) was used for color development in phosphate-itrate buffer (pH 5) containing 1.5 �L/mL of 30% H2O2 at roomemperature for 150 seconds. The reaction was stopped by theddition of 2.0 mol/L H2SO4 before spectrophotometric measure-ent of absorbency at 490 nm. The sensitivity of detection for

urified human �1-AT (Sigma Chemical Co.) and for human

1-AT in mouse serum was 0.1 ng/mL. This ELISA could notetect any human �1-AT in normal mouse serum.

Statistical Analysis

Data from in vitro transduction experiments were ana-yzed using unpaired Student t test. The serum levels of human

1-AT in transgenic mice before and after rSV40 infection werevaluated by Wilcoxon signed rank sum test. P � 0.05 wasonsidered statistically significant.

Results

Transduction of SV40-Derived Ribozymeand �1-AT–Modified cDNA Vectors inHepatoma Cell Lines

Three days after transduction with SV(mAT) con-aining the modified �1-AT cDNA, quantitative RT-CR showed that human �1-AT expression in 2 hepa-oma cell lines, HLE and HLF, was enhanced by 14- �.6-fold and 13.2- � 0.4-fold, respectively, comparedith untransduced cells (Figure 2A). Human � -AT

5=-Labelinga Concentration (nmol/L)

300GG FAM 300

50300

FAM 30050

30050

30050

30050

ine. F, forward primer; P, hybridization probe; FAM, 6-carboxyfluores-

GCTCCTCCTGCCCTGGA

A

AACATGTGGA

dam

1

mS3c

S�9(tlrw(

wt

ncahtps

h

Fvimssa(iaia

FlhSf�mga(Prdukbi


RNA levels in PLC/PRF/5 cells transduced withV(AT589T) containing the ribozyme were reduced to0% � 3.8% at 72 hours compared with untransducedells (Figure 2B).

In Vivo Transduction Efficiency

Two inoculations daily for 4 days with theV40-derived vector designed to deliver the modified1-AT cDNA, SV(mAT), transduced approximately0% of hepatocytes, as evaluated by in situ RT-PCRFigure 3A). This figure shows the presence of RNAranscribed from the modified virus. Control mouseiver was negative (Figure 3B). Pathologic immuneesponses, hepatocyte necrosis, and inflammationere not apparent despite high levels of infection

Figure 3C). Serum alanine aminotransferase levels

igure 2. Effects of SV40-derived vectors on human �1-AT mRNAevels in human hepatoma cells. (A) Quantitative RT-PCR analysis ofuman �1-AT expression in HLE and HLF cells after transduction withV(mAT), containing human �1-AT–modified cDNA. RNA was extracted

rom the transduced HLE or HLF cells 72 hours after transduction, and1-AT mRNA levels were evaluated by quantitative RT-PCR using hu-an glyceraldehyde-3-phosphate dehydrogenase as a housekeeping

ene. The data were summarized from 2 independent experimentsnd expressed as fold increase compared with untransduced cells.B) Quantitative RT-PCR analysis of human �1-AT mRNA levels inLC/PRF/5 cells after transduction with SV(AT589T), containing �1-ATibozyme. Human �1-AT mRNA levels in the transduced cells wereetermined 72 hours after the transduction by quantitative RT-PCRsing human glyceraldehyde-3-phosphate dehydrogenase as a house-eeping gene and expressed as relative levels of gene expressionased on untransduced controls. The data were summarized from 5

ndependent experiments. ***P � 0.001 compared with controls.

ere normal when the mice were tested 2 days afterhe last injection (data not shown).

Long-term Human �1-AT Production in ICRMice After the Inoculation of an SV40-Derived Vector Containing Modified �1-ATcDNA

When RNA was extracted from the livers oformal mice that had been treated with an rSV40 vectorontaining the modified �1-AT cDNA, Northern blotnalysis showed successful transduction of the modifieduman �1-AT in the mouse liver. No human �1-ATranscripts were found in the untreated mice, even withrolonged exposure. A representative Northern blot ishown in Figure 4A.

In additional experiments in different mice, the serumuman �1-AT levels in treated mice were monitored by

igure 3. In situ RT-PCR of a liver section after injection with SV(mAT)ia the portal vein. Mice were injected 8 times over 4 days through anndwelling catheter in the portal vein. Each injection consisted of 0.3L of high-titer virus (1 � 1012 IU/mL). For the amplification of RNAequences for SV40, reverse transcription was performed for thepecific sequences using antisense primers as described in Materialsnd Methods. Brown color is indicative of SV40 RNA within the cells.A) Injected mouse liver. The mouse was killed 2 days after the lastnjection. (B) Untransduced mouse liver. (C) Histology of the liver from

mouse after 2 inoculations a day for 4 days with SV(mAT), contain-ng modified �1-AT cDNA (H&E staining). The mouse was killed 4 daysfter the last injection.

Emetcsbit

db

i�ba�rbsptlie

cevrtP(mobkrwiai�3

FmoiiInmawd(hmolpa

Fdci4��Nt


LISA, following the injection of SV(mAT) containing theodified human �1-AT cDNA. Serum human �1-AT lev-

ls remained 1 �g/mL for 50 weeks (Figure 5) in theransduced mice and were undetectable in untransducedontrol mice. Western blot analysis indicated that a sub-tantial amount of human �1-AT protein was synthesizedy the transduced livers of 2 mice killed 65 weeks afternjection (Figure 4B). No human �1-AT protein was de-ectable in the untreated ICR mice at all time points.

Inhibition of Human �1-AT Production inTransgenic Mice After the Inoculation of anSV40-Derived Vector With a Ribozyme

The same injection schedule of 8 injections over 4ays was used with SV(AT589T) containing the ri-ozyme in an attempt to inhibit �1-AT gene expression

igure 4. Northern blot and Western blot analyses of human �1-ATRNA and protein in ICR mice. (A) Northern blot hybridization analysisf human �1-AT mRNA transcripts in mouse liver after intravenous

njection of SV(mAT), containing modified �1-AT cDNA. Mice werenjected 8 times over 4 days with 0.3 mL of high-titer virus (1 � 1012

U/mL). Fourteen days after beginning inoculations with the recombi-ant virus, the mice were killed. RNA was extracted from 2 individualice (1 and 2) that were infected with SV(mAT). (B) Western blotnalysis of human �1-AT protein in the mouse livers. The liver proteinas extracted 65 weeks after administration of 8 injections over 4ays with 0.3 mL of high-titer virus (3.5 � 1012 IU/mL). (Upper panel)for detection of human �1-AT): lane 1, transgenic mice carryinguman PiZ allele, 9.0 �g; lane 2, mouse B in Figure 5, 60 �g; lane 3,ouse C in Figure 5, 60 �g; lane 4, control ICR mouse. The dilutionf the primary antibody was 1:100, and detection antibody was di-

uted at 1:1000. (Lower panel) for detection of mouse albumin, 15 ngrotein was loaded in each lane; the primary antibodies were dilutedt 1:500 and the second antibodies at 1:2000.

n transgenic mice that chronically produce the human1-AT protein in their livers and release it into theloodstream. As shown in Figure 6A by Northern blotnalysis, ribozyme treatment selectively inhibited human1-AT expression in these mice. The transcript was

educed as much as 85% when assayed 14 days aftereginning the inoculations. Western blot analysishowed a virtual absence of detectable human �1-ATrotein in the livers of transgenic mice, corresponding tohe observed large decrease in the human �1-AT mRNAevels (Figure 6B). Coomassie blue staining of proteinndicated that equal amounts of protein were added toach lane (data not shown).

The effectiveness of transduction with rSV40 ribozymeonstructs over a longer period was also tested in differentxperiments. Mice transgenic for human PiZ were treatedia the portal vein route with the rSV40 carrying theibozyme against the PiZ transcript, and their livers wereested by Western blot analysis 4 weeks later. The targetediZ protein was reduced 4 weeks after the transduction

data not shown). In another experiment, using differentice, quantitative RT-PCR analysis showed that the levels

f human �1-AT mRNA in the mouse livers were reducedy as little as 21%, to essentially 100% in 4 mice that wereilled between 6 and 16 weeks after transduction with theibozyme construct. The average reduction in these 4 miceas 57%. No change in mouse albumin mRNA was found

n these livers. The changes of human �1-AT in serum werelso monitored by ELISA in these mice (Figure 7). Admin-stration of the ribozyme decreased serum levels of human1-AT to 50% � 5% of pretreatment values (P � 0.01)–16 weeks after transduction (Figure 7A and B), whereas

igure 5. Human �1-AT levels in 3 ICR mice transduced with SV40-erived modified human �1-AT cDNA vector. Blood samples wereollected by tail incision after the injection of the construct through anndwelling catheter in the portal vein. Mice were injected 8 times over

days of 0.3 mL of high-titer virus (3.5 � 1012 IU/mL). Human serum1-AT levels were determined by ELISA using specific anti-human1-AT antibodies, and the value for each mouse is shown separately.o human �1-AT protein was detectable in untreated ICR mice at all

ime points.

sctm�(tmPfPattn

StPmcHIopphrtimtpbe

FdscM(ieLtddt1Pls1

Fimlhshttm


erum human �1-AT levels in 7 control mice were un-hanged. Western blot analysis of liver tissues from 4 ofhese 5 transgenic mice shows a marked reduction in hu-an �1-AT protein synthesis (Figure 8A). Serum human

1-AT level in mouse D was reduced by 99% at 6 weeksFigure 7B); human �1-AT protein in its liver correspondedo its serum �1-AT level (Figure 8A, lane 4; Figure 8B,ouse D) without any change in mouse albumin synthesis.iZ transcripts were undetectable by quantitative RT-PCR

rom the liver of that mouse. Moreover, quantitative RT-CR showed that the expression levels of mouse �1-AT,lbumin, and �-actin mRNA from that mouse remainedhe same as in control transgenic mice despite the essen-ially complete loss of the human �1-AT transcripts (dataot shown).

igure 6. Inhibition of human �1-AT in transgenic mice by SV40-erived ribozyme construct. (A) Northern blot analysis of PiZ tran-cripts in livers of transgenic mice that were treated with SV(AT589T),ontaining a specific hammerhead ribozyme against human �1-AT.ice were injected 8 times over 4 days with 0.3 mL of high-titer virus

1.3 � 1012 IU/mL). Fourteen days after the injection through anndwelling catheter in the portal vein, the mice were killed. RNA wasxtracted from the liver, and Northern blot analysis was performed.ane 1, Hep G2 cells; lane 2, ICR mouse; lane 3, untransduced mouseransgenic for PiZ allele; lane 4, transgenic mouse inoculated once aay for 4 days; lane 5, transgenic mouse inoculated twice a day for 4ays. (B) Western blot analysis of the same experiment shown in A. Aransgenic mouse was injected twice a day for 4 days and then killed4 days after infection (the same mouse as shown in lane 5 in A).roteins from HepG2 cells and untransduced transgenic (Tg) mouse

iver were used as positive controls. HepG2 cells, 5 �g; mouse liveramples (20 �g each sample). The primary antibodies were diluted at:400 and the second antibodies at 1:2000.

Discussion

In the present study, we have shown that anV40-derived construct containing a ribozyme is effec-ive in decreasing human �1-AT mRNA levels in PLC/RF/5 cells, a line expressing high levels of �1-ATRNA, whereas a modified cDNA construct is effica-

ious in enhancing �1-AT mRNA levels in HLE andLF cells, 2 cell lines with a low basal �1-AT secretion.

n animal experiments, we have shown that the injectionf rSV40 viruses through an indwelling catheter in theortal vein led to the successful transduction of a highercentage of hepatocytes in vivo, without precipitatingepatocyte necrosis or an obvious pathologic immuneesponse. Approximately 30% of cells expressed theransgene protein after only one inoculation. When 8noculations of an rSV40 were delivered over 4 days,ore than 90% of hepatocytes appeared to express the

ransgene as assessed by in situ RT-PCR. This injectionrotocol provided enough expression to markedly reduceoth serum protein levels and liver human �1-AT genexpression for up to 4 months (the length of the exper-

igure 7. Serum human �1-AT levels in transgenic mice after thenjection of SV40-derived ribozyme construct. Ribozyme-inoculatedice were assayed at intervals, shown in the graph, after the inocu-

ations. Mice were injected 8 times over 4 days with 0.3 mL ofigh-titer virus (4.4 � 1012 IU/mL). Serum �1-AT levels were mea-ured by ELISA as described in Materials and Methods. (A) Three micead relatively high basal levels of human �1-AT. (B) Two additionalransgenic mice with relatively low basal levels of human �1-AT beforehe inoculation. Serum human �1-AT levels in 7 control transgenicice were unchanged during the experiments.

it

leaiPrb

owhmroitpptpctr

IcsmnhfipgsvsscdpbttgSdssgtleorobmtnh

FobwjIthlTlamb1llnlaapaacs


ment to date) when a hammerhead ribozyme was used inransgenic mice expressing the human PiZ allele.

As shown in Figure 6, one of the mice had extremelyow levels of human �1-AT, as demonstrated by North-rn and Western blot analysis. In another experiment,nother transgenic mouse had almost no human �1-ATn its liver as measured by real-time quantitative RT-CR and Western blot and by ELISA of its serum. Theseesults would indicate that nearly all hepatocytes areeing effectively transduced because, contrary to gain-

igure 8. Sustained inhibition of human �1-AT synthesis in the liversf transgenic mice by SV40-derived ribozyme construct. (A) Westernlot analysis of human �1-AT protein in livers of transgenic mice thatere injected with SV(AT589T)-containing ribozyme. Mice were in-

ected 8 times over 4 days with 0.3 mL of high-titer virus (4.4 � 1012

U/mL). (Upper panel) Western blot analysis of human �1-AT protein inhe transgenic mice. Lane 1, transgenic (Tg) control mouse 1 carryinguman PiZ allele without ribozyme treatment (serum human �1-ATevel, 854 �g/mL); lane 2, the molecular weight standard; lane 3, theg control mouse 2 without ribozyme treatment (serum human �1-ATevel, 747 �g/mL); lane 4, mouse D in Figure 7 was killed 6 weeksfter injection (�1-AT levels before ribozyme, 175 �g/mL); lane 5,ouse A in Figure 7 was killed 12 weeks after injection (�1-AT levelefore ribozyme, 1830 �g/mL); lane 6, mouse B in Figure 7 was killed2 weeks after injection (�1-AT level before ribozyme, 1306 �g/mL);

ane 7, mouse C in Figure 7 was killed 16 weeks after injection (�1-ATevel before ribozyme, 2439 �g/mL); lane 8, control ICR mouse (CM,ot transgenic). Twenty micrograms of protein was loaded in each

ane; the primary antibodies were diluted at 1:400 and the secondntibodies at 1:2000. (Lower panel) Western blot analysis of mouselbumin levels in the corresponding mouse liver. Fifteen nanograms ofrotein was loaded in each lane; the primary antibodies were dilutedt 1:400 and the second antibodies at 1:2000. (B) Quantitativenalysis for Western blots by densitometry. Relative density wasalculated based on transgenic control mouse 2. The labeling is theame as in A.

f-function studies, transduction of only 50% of cellsith the ribozyme construct would leave levels of theuman �1-AT at only 50% of baseline. Not all of theice that were treated with the ribozyme had similar

esults, but RT-PCR and ELISA data showed an averagef 50% or more reduction in levels of the human proteinn the 5 mice tested. Although unproven, it would seemhat a 50% reduction in the production of the mutatedrotein might yield a significant improvement in theathophysiology of the disease process in humans. Fur-hermore, these are the initial studies, the proof-of-rinciple experiments, so more consistent robust resultsould certainly be expected with further optimization ofhe experimental system and repeated injection of theecombinant viruses.

The fact that serum human �1-AT levels in healthyCR mice after the delivery of the modified �1-ATDNA were sustained for more than 1 year confirms thetability of rSV40 gene delivery to the liver.3 Further-ore, it appears that the �1-AT transgene expression was

ot diminished in regenerating liver from a mouse thatad been assayed by partial hepatectomy earlier.19 Thesendings provide strong evidence that suggest, but do notrove, that vector integration into the hepatocellularenome was the mechanism by which rSV40 deliversuch sustained transgene expression. Further evidence forector integration of rSV40 viruses has recently beenhown in several experimental conditions.19,20 It has beenhown that rSV40 vectors integrate randomly into theellular DNA in both resting and dividing cells withinays of transduction and that this integration may ex-lain long-term transgene expression.3,19 The mechanismy which rSV40 vectors integrate is unclear; however,here is a report describing a possible association betweenhe SV40 ori and nuclear matrix.21 Activation of cellularenes at integration sites has not been associated withV40.22 On the other hand, expression of transgeneselivered by rSV40-derived vectors is not observed to beilenced over time, as has sometimes been the case witheveral other vector systems.23,24 Once a level of trans-ene expression is established in a population of cellsransduced by rSV40, that level persists indefinitely. It isikely that host proteins or protein complexes silencexpression from other vectors. This involves recognitionf their consistent integration patterns (i.e., terminalepeats/unique sequence/terminal repeats) and inhibitionf expression of the genes between the terminal repeatsy methylation,23 histone deacetylation,24 or othereans. The SV40 genome is circular and thus lacks

erminal repeats. Accordingly, such a silencing mecha-ism seems to be avoided in rSV40 vectors. The levels ofuman � -AT in the serum of the mice that were trans-
1

d�tbd�apriyt

ihTgflcsmTaTtn

dtsg�anooh�ar�fiurmta

sc

hpabrapscvamtavd

spsavyhmegtaUentdbrvtvibtc

msrrytsu


uced with the modified cDNA in our experiments were5% as high as would be deemed adequate for the

reatment of lung disease in �1-AT deficiency. Thus,efore human gene therapy can be entertained for lungisease using the modified cDNA, improvement in1-AT gene therapy must be achieved with the use ofdditional constructs that include different enhancers/romoters or other interventions. The present resultsepresent a proof-of-principle for human gene therapy,ncluding the stable expression of the transgene for 1ear, and show levels of ribozyme expression that appearo be physiologically relevant.

�1-AT deficiency, which was used as a model systemn these studies, is one of the more common lethalereditary disorders in white people of European descent.he deficiency state is caused by mutations of the �1-ATene, which produces a 52-kilodalton glycoprotein thatunctions as an antiprotease.25 Various combinations of ateast 17 different mutations of the �1-AT gene are asso-iated with a decreased serum level of �1-AT and aignificant risk for developing emphysema.26 A subset ofutations is associated with hepatitis and cirrhosis.5

hese latter mutations all involve the production ofbnormal proteins; they do not include null mutations.he pathogenesis of the liver disease is believed to be due

o the hepatotoxic effect of the accumulation of an ab-ormal �1-AT protein in hepatocytes.Although several studies (including the intravenous

elivery of adenoviral or adeno-associated viral vectors orhe topical administration of plasmid �1-AT in lipo-omes) have focused on the delivery of the normal �1-ATene into hepatocytes or airway cells to restore normal1-AT production and to protect lung tissue,27–31 thesepproaches do not affect the liver disease. In addition, aon–gene therapy approach has used chemical chaper-nes to reverse the cellular mislocalization or misfoldingf mutated �1-AT in hepatocytes. However, the agentas only transient effects in enhancing serum levels of1-AT in transgenic mice.32 It would seem that the bestpproach to treat �1-AT deficiency disease is to botheduce the production of the endogenous mutant form of1-AT protein and increase the synthesis of the normal

unctional protein. The novel approach used in this studynvolved inhibiting abnormal �1-AT protein productionsing a gene-specific ribozyme and the synthesis of aibozyme-resistant wild-type protein by engineering aodified �1-AT cDNA whose product was not suscep-

ible to ribozyme cleavage. To our knowledge, no similarpproach has been reported previously.

The potential of ribozymes to target particular tran-cripts has been studied and reported extensively inell-free systems and in cultured cells. Several reports

ave described ribozyme activity when ribozyme-ex-ressing cells transfected ex vivo are reimplanted intonimals in vivo33–35 or in animals transgenic for ri-ozymes.36 Ribozyme delivery in vivo has also beeneported, showing at least transient effects37,38 and ther-peutic potentials.33 Those studies reinforce the greatotential of ribozymes in vivo if an adequate deliveryystem were available. We show here that rSV40 vectorsan deliver durable ribozyme activity to animal livers inivo, with significant attenuation of serum �1-AT levelsnd liver �1-AT gene expression over approximately 4onths. This stability in expression and high transduc-

ion efficiency are characteristic of SV40 gene deliverynd offer the promise that these vectors may be usefulehicles for correcting genetic defects, such as �1-ATeficiency or abnormalities in bilirubin metabolism.20

Another major advantage of the SV40-derived vectorystem seems to be a higher level of transduction com-ared with other viral vector systems or nonviral vectorystems, such as plasmid DNA or liposomes. First, thebility of SV40 to infect almost all nucleated cell types ofirtually all mammals has been documented for manyears.39 The cellular receptor for SV40 is the majoristocompatibility complex class I; therefore, antibody toajor histocompatibility complex class I blocks SV40

ntry.40 In the case of wild-type SV40 infection, the earlyene proteins (Tag and tag) are made and presented tohe immune system; as a consequence, infected cellsctivate both T-cell responses and antibody responses.nlike wild-type SV40 infection, those antigens that

licit both cell-mediated immune responses that elimi-ate infected cells and neutralizing antibody responseshat impair repeat infection are not made by cells trans-uced by rSV40 vectors that lack early genes. Therefore,ecause rSV40 can evade the immune system, multipleeadministrations are well tolerated with the originalector and/or another rSV40,20 and this enhances theransduction capabilities. Finally, the ability of theseiral vectors to be lyophilized without significant loss ofnfectivity means that high concentrations of viruses cane delivered in small volumes, approaching more closelyhe titers needed to deliver genes to a high percentage ofells in vitro and in vivo.

Safety is a major concern for any vector used in hu-ans. None of the SV40-derivative viruses thus far con-

tructed lacking large T antigen have been shown toeplicate in cells lacking Tag. Replication-competentevertants have not been detected even 11 passages be-ond the initial generation of the virus. Because of theheoretical possibility of such an occurrence, it is reas-uring that wild-type SV40 seems to be relatively innoc-ous in humans. DNA sequences resembling those found

iassrtttp

pocSocnpdpthwSmtmrhpg

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2


n the SV40 genome have been described in severalnecdotal case reports of rare human tumors,8,9 but theource of these sequences and their significance are ob-cure. Moreover, epidemiologic studies of Salk vaccineecipients have shown no significant harmful sequelae inhe millions of people who inadvertently received wild-ype SV40 virus.10,11 Thus, harmful side effects, even inhe unlikely event of revertant contamination of viralreparations, would seem to be extremely low.The results of the present study and of previous re-

orts1–3 indicate that the SV40 vector system has manyf the properties necessary for the development of alinically valuable gene therapy agent. Tag-deletedV40-derivative viruses do not cause hepatocyte necrosisr pathologic immune responses, are stably expressed,an be concentrated to high titers, and can transduceoncycling cells effectively and stably. Moreover, multi-le injections are well tolerated and enhance the trans-uction capabilities; therefore, it is reasonable to antici-ate that essentially all hepatocytes can be readilyransduced. The ability to deliver transcripts to 90% ofepatocytes in mice using rSV40-derived vectors bodesell for this approach in human gene therapy. Thus, theV40-based vector system may be applied to the treat-ent of a variety of disorders requiring that most hepa-

ocytes be infected, such as hepatitis B or C disease, oretabolic abnormalities such as �1-AT deficiency, bili-

ubin metabolism abnormalities, Wilson’s disease, oremochromatosis. Finally, our results represent a proof ofrinciple in the development of a novel approach to theene therapy of �1-AT deficiency.

References1. Strayer DS, Kondo R, Milano J, Duan L-X. Use of SV40-based

vectors to transduce foreign genes to normal human peripheralblood mononuclear cells. Gene Ther 1997;4:219–225.

2. Strayer DS, Zern MA. Gene delivery to the liver using simian virus40-derived vectors. Semin Liver Dis 1999;19:71–81.

3. Strayer DS, Zern MA, Chowdhury JR. What can SV40-derivedvectors do for gene therapy. Curr Opin Mol Ther 2002;4:313–323.

4. Hubbard R, Crystal R. �1-antitrypsin augmentation therapy for�1-antitrypsin deficiency. Am J Med 1988;84:52–62.

5. Errickson S, Velez R. Risk of cirrhosis and primary liver cancer in�1-antitrypsin deficiency. N Engl J Med 1986;314:736–739.

6. Ozaki I, Zern MA, Liu SL, Wei DL, Pomerantz RJ, Duan LX. Ri-bozyme-mediated specific gene replacement of �1-antitrypsingene in human hepatoma cells. J Hepatol 1999;31:53–60.

7. Zern MA, Ozaki I, Duan L-X, Pomerantz R, Liu S-L, Strayer DS. Anovel SV40-based vector successfully transduces and expressesan �1-antitrypsin ribozyme in a human hepatoma-derived cell line.Gene Ther 1999;6:114–120.

8. Myers RM, Tjian R. Construction and analysis of simian virus 40origins defective in tumor antigen binding and DNA replication.Proc Natl Acad Sci U S A 1980;77:6491–6495.

9. Tornow J, Polvino-Bodnar M, Santangelo G, Cole CN. Two sepa-rable functional domains of simian virus 40 large T antigen:

carboxyl-terminal region of simian virus 40 large T antigen isrequired for efficient capsid protein synthesis. J Virol1985;53:415–424.

0. Lehn H. Human papillomavirus DNA replication mediated by sim-ian virus 40 T antigen in trans. J Gen Virol 1986;67:1581–1589.

1. Strayer DS, Duan L, Ozaki I, Milano J, Bobraski LE, Bagastra O.Titering replication-defective virus for use in gene transfer. Bio-techniques 1997;22:447–450.

2. Vrancken Peeters MJTFD, Lieber A, Perkins J, Kay MA. Method formultiple portal vein infusions in mice: quantitation of adenovirus-mediated hepatic gene transfer. Biotechniques 1996;20:278–285.

3. Liu L, Zern MA, Lizarzburu ME, Nantz MH, Wu J. Poly(cationiclipid)-mediated in vivo gene delivery to mouse liver. Gene Ther2003;10:180–187.

4. Bagasra O, Hauptman SP, Lischner HW, Sachs M, Pomerantz RJ.Detection of human immunodeficiency virus type 1 in mononu-clear cells by in situ polymerase chain reaction. N Engl J Med1992;326:385–391.

5. Bagasra O, Seshamna T, Hansen J, Pomerantz RJ. In situpolymerase chain reaction and hybridization to detect low abun-dance nucleic acid targets. In: Ausubel FM, et al., eds. Currentprotocols in molecular biology. New York, NY: Wiley, 1995:14.8.1–14.8.23.

6. Chomczynski P, Sacchi N. Single-step method of RNA extractionby acid guanidinium thiocyanate-phenol-chloroform extraction.Anal Biochem 1987;162:156–159.

7. Wege H, Le HT, Chui MS, Liu L, Wu J, Giri R, Malhi H, Sappal BS,Kumaran V, Gupta S, Zern MA. Telomerase reconstitution immor-talizes human fetal hepatocytes without disrupting their differen-tiation potential. Gastroenterology 2003;124:432–444.

8. Frizell E, Liu SL, Abraham A, Ozaki I, Eghbali M, Sage EH, Zern MA.Expression of SPARC in normal and fibrotic livers. Hepatology 1995;21:847–854.

9. Strayer DS, Branco F, Zern MA, Yam P, Calarota SA, Nichols CN,Zaia JA, Rossi J, Li H, Parasher B, Ghosh S, Chowdhury JR. Durabilityof transgene expression and vector integration: recombinant SV40-derived gene therapy vectors. Mol Ther 2002;6:27–237.

0. Sauter BV, Parasher B, Chowdhury NR, Kadakol A, Iiran Y, Singh H,Milano J, Strayer DS, Chowdhury JR. A replication-deficient rSV40mediates liver directed gene transfer and a long-term ameliorationof jaundice in Gunn rats. Gastroenterology 2000;119:1348–1357.

1. Baiker A, Maercker C, Piechaczek C, Schmidt SB, Bode J,Benham C, Lipps HJ. Mitotic stability of an episomal vector contain-ing a human scaffold/matrix-attached region is provided by associ-ation with nuclear matrix. Nat Cell Biol 2000;2:182–184.

2. Strayer DS, Milano J. SV40 mediate stable gene transfer in vivo.Gene Ther 1996;3:581–587.

3. Yamano T, Ura K, Morishita R, Nakajima H, Monden M, Kaneda Y.Amplification of transgene expression in vitro and in vivo using anovel inhibitor of histone deacetylase. Mol Ther 2000;1:574–580.

4. Chen WY, Bailey EC, McCune SL, Dong JY, Townes TM. Reacti-vation of silenced, virally transduced gene by inhibitors of histonedeacetylase. Proc Natl Acad Sci U S A 1997;94:5798–5803.

5. Crystal R, Brantly M, Hubbard R, Cruiel D, States D, Holmes M.The �1-antitrypsin gene and its mutations: clinical consequencesand strategies for therapy. Chest 1989;95:196–208.

6. Crystal RG. The �1-antitrypsin gene and its deficiency states.Trends Genet 1989;5:411–417.

7. Dasi F, Benet M, Crespo J, Crespo A, Alino SF. Asialofetuin lipo-some-mediated human �1-antitrypsin gene transfer in vivo results instationary long-term gene expression. J Mol Med 2001;79:205–212.

8. Jaffe H, Danel C, Longenecker G, Metzger M, Setoguchi Y,Rosenfeld M. Adenovirus-mediated in vivo gene transfer andexpression in normal rat liver. Nat Genet 1992;1:372–378.

2

3

3

3

3

3

3

3

3

3

3

4

R2m

AAiF

A2T


9. Song S-H, Morgan M, Ellis T, Poirier A, Chesnut K, Wang J-M,Brantly M, Muzyczka N, Byrne BJ, Atkinson M, Flotte TR. Sus-tained secretion of human �1-antitrypsin from murine muscletransduced with adeno-associated virus vectors. Proc Natl AcadSci U S A 1998;95:14384–14388.

0. Kay M, Baley P, Rothenberg S, Leland F, Ponder KP, Woo S.Expression of human �1-antitrypsin in dogs after autologoustransplantation of retroviral transduced hepatocytes. Proc NatlAcad Sci U S A 1992;89:372–378.

1. Stecenko AA, Brigham KL. Gene therapy progress and prospect:alpha-1 antitrypsin. Gene Ther 2003;10:95–99.

2. Burrows JAJ, Willis LK, Perlmutter DH. Chemical chaperones medi-ate increased secretion of mutant �1-antitrypsin (�1-AT) Z: a poten-tial pharmacological strategy for prevention of liver injury and em-physema in �1-AT deficiency. Proc Natl Acad Sci U S A 2000;97:1796–1801.

3. Bai J, Banda J, Lee NS, Rossi J, Akkina R. RNA-based anti-HIV-1gene therapeutic constructs in SCID-hu mouse model. Mol Ther2002;6:770–782.

4. Larsson S, Hotchkiss G, Andang M, Nyholm T, Inzunza J, JanssonI, Ahrlund-Richter L. Reduced beta 2-macroglobulin mRNA levelsin transgenic mice expressing a designed hammerhead ri-bozyme. Nucleic Acids Res 1994;22:2242–2248.

5. Lieber A, Kay MA. Adenovirus-mediated expression of ribozymesin mice. J Virol 1996;70:3153–3158.

6. Lednicky JA, Garcea RL, Bergsagel DJ, Butel JS. Natural simianvirus 40 strains are present in human choroid plexus and ependy-mona tumors. Virology 1995;212:710–717.

7. Carbone M, Pass HI, Rizzo P, Marinetti M, DiMuzio M, Mew DJ,Levine AS, Procopio A. Simian virus 40-like DNA sequences inhuman pleural mesotheliomas. Oncogene 1994;9:1781–1790.

8. Mortimer EA, Lepow ML, Gold E, Robbins C, Burton GJ, Frau-meni JF. Long-term follow-up of persons inadvertently inocu-lated with SV40 as neonates. N Engl J Med 1981;305:1517–1518.

9. Fried M, Prives C. The biology of simian virus 40 and polyomavi-rus. Cancer Cells 1986;4:1–16.

0. Breau WC, Atwood WJ, Norkin LC. Class I major histocompatibilityproteins are an essential component of the simian virus 40receptor. J Virol 1992;66:2037–2045.

Received May 19, 2003. Accepted July 8, 2004.Address requests for reprints to: Mark A. Zern, M.D., Transplantesearch Institute, University of California, Davis Medical Center, 4635nd Avenue, Suite 1001, Sacramento, California 95817. e-mail:[email protected]; fax: (916) 734-8097.Supported in part by National Institutes of Health grantsA06386 (to M.A.Z.) and CA44800 and AI41399 (to D.S.S.) and themerican Liver Foundation and Alpha-1 Foundation (to M.A.Z.). J.W.s a recipient of the Liver Scholar Award by the American Liveroundation.Presented in part at the 54th annual meeting of the Americanssociation for the Study of Liver Diseases (Hepatology 2003;38:29A) and the sixth annual meeting of the American Society of Geneherapy (Mol Ther 2003;7:S86–S87).

Documents

Gene therapy for human alpha1-antitrypsin deficiency in an animal model using SV40-derived vectors