815 Development of Non Integrating Foamy Vectors for Gene Therapy Molecular Therapy Volume 17, Supplement 1, May 2009 Copyright © The American Society of Gene Therapy S312 RNA VIRUS VECTORS III the ge[.]
Trang 1Molecular Therapy Volume 17, Supplement 1, May 2009 Copyright © The American Society of Gene Therapy S312
RNA VIRUS VECTORS III
the gene transfer effi ciency to actively dividing cells, in particular
when vectors were applied, and 2) Overall gene delivery effi ciency
could be highly over-estimated, depending on post-transfection (or
transduction) time periods These are crucial questions needed to be
addressed for in vitro and in vivo gene delivery, but to quantitatively
understand and compare various vector systems cannot be achieved by
experiments alone In this study, cells were cultured to form colonies
where only the cells in periphery actively spread and proliferate A
computer-simulated stochastic cell growth model was utilized in order
to fi t to the gene expression rates collected at various post-tranfection
(or transduction) time points, and then the specifi c gene transfer rates
to actively dividing and non-dividing cells at the time of transfection
(or transduction) were determined The results indicated: 1) There
was a signifi cant difference in the apparent gene transfer effi ciency,
(e.g., 82, 56 and 40% using HEK 293 cells and VSV-G pseudotyped
MoMLV vectors) 24h after transduction, depending on the ratio of
actively-dividing vs non-dividing cells and time of transduction, 2)
The true values of viral transduction rates are relatively consistent
(e.g., approx 75 and 0% transduction rates to actively dividing and
non-dividing cells over the course of 2h exposure) This means that
gene transfer effi ciency based on the number of gene expressing cells
varies up to several folds with post-transfection (or transduction)
time as well as ratio of actively dividing vs non-dividing cells
More extended examples, including transfection rates to various
types of actively dividing and non-dividing cells by different kinds
of gene delivery vectors (e.g., lentiviral and non-viral gene delivery
to actively-dividing vs non-dividing cancer cells) will be presented
Implications of the fi ndings for in vivo gene delivery will also be
discussed
Therapy Vectors for the Treatment of Cystic
Fibrosis
Anna R Kwilas,1,3 Mark A Yednak,3 Liquin Zhang,4 Peter L
Collins,5 Raymond J Pickles,4 Mark E Peeples.2,3
1 Integrated Biomedical Science Graduate Program, The Ohio
State University, Columbus, OH; 2 Department of Pediatrics,
The Ohio State University, Columbus, OH; 3 Center for Vaccines
and Immunity, The Research Institute at Nationwide Children’s
Hospital, Columbus, OH; 4 Cystic Fibrosis/Pulmonary Research
and Treatment Center, University of North Carolina at Chapel Hill,
Chapel Hill, NC; 5 Laboratory of Infectious Diseases, NIAID, NIH,
Bethesda, MD.
Cystic fi brosis (CF) is the most common lethal recessive genetic
disease in the Caucasian population It is caused by any of a number
of mutations in the CF transmembrane conductance regulator (CFTR)
gene A mutation in the CFTR gene, leading to the production of a
partially or completely non-functional protein, results in dehydration
of mucosal surfaces and an inability to remove pathogens from
the respiratory tract leading to morbidity and mortality in the CF
population Since CFTR was fi rst cloned in 1989 a major goal of CF
research has been to develop an effective gene therapy for this disease
Respiratory syncytial virus (RSV) naturally infects the ciliated cells that line the respiratory tract, the same cells that express CFTR
In addition, although RSV does induce an immune response, this response does not prevent subsequent infection suggesting that an RSV-based vector might be effectively readministered, a requirement since the apical ciliated cells do not divide We have inserted the large (4.5 kb) CFTR gene into four sites in the recombinant, green
fl uorescent protein-expressing (rg)RSV genome to generate virus expressing increasing amounts of the CFTR protein Two of these rgRSV-CFTR viruses were capable of expressing CFTR The inclusion of the CFTR gene in the (rg)RSV genome did cause a delay
in the growth of the viruses, however this was overcome by 72 hours post infection The CFTR expressed by the rgRSV-CFTR viruses was functional in primary human airway epithelial cultures derived from
CF patients and infection with these viruses resulted in a minimum of 30% of the normal chloride channel function seen in non-CF primary human airway epithelial cultures Unfortunately, ciliated airway cells infected with RSV die within 5 days post infection To create an improved RSV-based vector we have generated an RSV “replicon”
by removing the three glycoprotein genes from a full-length RSV cDNA The replicon is able to replicate in cells without killing them
or producing infectious virus, allowing the isolation and expansion
of replicon-containing cells Providing the viral glycoproteins in trans enables the production of “one-step virus” (OSV) which is
capable of delivering a replicon to fresh cells where it produces its encoded proteins, but cannot spread to additional cells We inserted the CFTR gene into the RSV replicon genome in four positions, all four of which produced the CFTR protein When CFTR-expressing replicon OSV were used to infect primary human airway epithelial cultures, replicon containing cells remained within the cultures for 12-20 days, indicating that replicon containing cells have a survival advantage over those infected with the complete virus
Vectors for Gene Therapy
David R Deyle,1 Yi Li,1 David W Russell.1
1 Department of Medicine, Division of Hematology, University of Washington, Seattle, WA.
Foamy virus (FV) is a retrovirus that has been developed for gene therapy because it has not been associated with disease, is able to infect many different hosts and tissue types, and has the largest genome of all retroviruses After infection, FVs can stably integrate into the host DNA and facilitate effi cient transgene expression However, integration of exogenous DNA into the genome can cause insertional mutagenesis To avoid the issues of insertional mutagenesis, previous attempts have been made to create a functional non-integrating foamy virus (NIFV) vector similar to integrase-defi cient lentiviral vectors However, these efforts have been unsuccessful To further investigate this problem, we constructed NIFV vectors by introducing mutations in the highly conserved DD35E catalytic core motif of the
FV integrase gene Three independent vectors stocks containing the expression cassette MSCV-GFP were generated for two single mutant NIFV vectors packaged with point mutations D131V or D188A, a double mutant NIFV vector packaged with both D131V and D188A mutations, and an integrase-profi cient FV vector Genome titers were determined by Southern blot analysis and all titers ranged from 8 x 10^8 to 3 x 10^10 genomes/ml, showing that NIFV vector stocks were comparable to the integrase profi cient FV vector stocks All three NIFV vectors and the integrase-profi cient FV vector expressed GFP in human fi broblasts and HT1080 cells two days after infection Unlike integrase profi cient FV vectors that showed stable GFP expression,
fi broblasts and HT1080 cells transduced with NIFV vectors had a decreasing percent of GFP+ cells over a 10 day period, refl ecting
a lack of integration and dilution over time in the proliferating culture We also generated NIFV and integrase-profi cient FV vectors
Trang 2Molecular Therapy Volume 17, Supplement 1, May 2009
PRECLINICAL AND CLINICAL APPLICATIONS OF AAV
containing a CMV-nuclear localized lacZ-PGK-Neo-Ori (CnZPNO)
cassette and transduced fi broblasts Again, all three NIFV vectors and
the integrase-profi cient FV vector expressed the transgene, nuclear
localized lacZ, as evidenced by the presence of
beta-galactosidase-positive foci In addition, after selecting infected fi broblasts with
G418, we found that the NIFV vectors integrated at a frequency
several logs lower than integrase-profi cient FV vectors NIFV
vector-chromosome integration junctions from G418 resistant fi broblast
clones will be sequenced to determine the mechanism of
integrase-independent vector insertion Since non-integrating viral genomes
are known to exist in various forms, including linear, one-LTR or
two-LTR circles, we are currently utilizing Southern blot analysis to
identify the predominate DNA species Our NIFV vector would be
especially advantageous for those applications in which short-term
gene expression is required, such as expressing a factor to infl uence
development, inducing a cellular phenotype, or expressing a genomic
modifying protein (e.g Cre recombinase) Also, NIFV vectors may be
useful for targeted gene repair In summary, our novel NIFV system
expresses transgenes, infects different cell types, has applications
for gene therapy, and most importantly, does not integrate, greatly
reducing the risk of genotoxicity
Preclinical and Clinical Applications of AAV
Their Responsiveness to Hepatic AAV Gene
Transfer
Hua Li,1 Katherine High,2 Hildegund C J Ertl.1
1 Immunology Program, The Wistar Institute, Philadelphia, PA;
2 Children’s Hospital of Philadelphia, Philadelphia, PA.
Hepatic adeno-associated virus (AAV)-serotype 2-mediated
gene transfer results in sustained transgene product expression in
experimental animals but not in human subjects We hypothesized
that loss of transgene product expression in humans might be caused
by immune memory mechanisms that become reactivated upon
AAV vector transfer We tested the effect of hepatic AAV2-hF.IX or
AAV8-hF.IX gene transfer on AAV capsid-specifi c CD8+ T cells in
mice AAV capsid-specifi c CD8+ T cells were found to proliferate in
response to hepatic AAV gene transfer indicating that degradation of
the AAV particles triggered a recall response The kinetics of the recall
response analyzed by adoptive transfer of CD8+ T cells into mice that
had received AAV-2hF.IX or AAV-8hF.IX vectors previously showed
differences in the degradation rates of AAV-2 and AAV-8 vectors
Other functional properties of AAV capsid-specifi c CD8+ T cells
were analyzed and the most remarkable fi nding was that in mice AAV
capsid-specifi c CD8+ T cells require a rather high concentration of
their cognate antigen in order to exhibit lysis The implication of these
fi ndings on the potential role of AAV capsid-specifi c CD8+ T cells in
limiting hepatic AAV gene transfer in humans will be discussed
Mutant Alpha One Antitrypsin Using Recombinant
AAV Delivered shRNA
Christian Mueller,1 Qiushi Tang,1 Brian O’Sullivan Murphy,1 Sofi a
Braag,1 Terence R Flotte.1
1 UMass Gene Therapy Center, UMass Medical School, Worcester,
MA.
Alpha-1 antitrypsin (AAT) defi ciency is one of the most common
genetic diseases in North America, with a carrier frequency of
approximately 4% in the US population Homozygosity for the most
common mutation caused by a single base pair change (Glu342Lys,
PI*Z) leads to the synthesis of a mutant protein, which accumulates
and polymerizes within hepatocytes rather than being effi ciently
secreted This lack of secretion causes severe serum defi ciency
predisposing patients to chronic lung disease 12 to 15% of patients with PI*ZZ also develop liver disease, which can be severe, even in infancy This is thought to be due to toxic effects of the accumulated mutant Z-AAT within the hepatocyte Thus, an approach to reduce AAT-defi cient liver disease will likely require some mechanism to decrease the amount of Z-AAT within hepatocytes Here we describe studies of allele specifi c small interfering RNAs (siRNAs) designed to down regulate PI*Z specifi c AAT within hepatocytes Two different siRNA sequences were identifi ed and cloned into a recombinant adeno-associated virus (rAAV) backbone as U6 driven shRNA, one with the single base pair change at position 10 (p10) and one with the change at position 16 (p16) of the targeting siRNA molecule Each construct was able to reduce Z-AAT levels, but experiments
on wildtype PiM AAT expressing cell culture models showed the greatest PiZ specifi c reduction with p10, whereas the p16 construct also signifi cantly down regulated wildtype AAT The rAAV-U6p10 was then packaged into AAV8 capsids and used in vivo to transduce the livers of human Z-AAT over-expressing transgenic mice These studies show a decrease in total human AAT after 2 weeks, and a clearing of Z-AAT accumulation by immunohistochemistry in the livers The rAAV8-U6-p10 vector may hold promise as a potential therapy for patients with AAT liver disease
Delivery by Adeno-Associated Virus Vectors Carrying Large Genomes in Hemophilia Mouse and Dog Models
Junjiang Sun,1 Paul E Monahan,1 R Jude Samulski,1 Clinton D Lothrop, Jr 2
1 Gene Therapy Center, University of North Carolina at Chapel Hill, Chapel Hill, NC; 2 Biochemistry and Molecular Genetics, School of Medicine, University of Alabama, Birmingham, AL.
The use of adeno-associated virus (AAV) vectors for correction of many clinically relevant congenital defi ciencies has been hampered by the relatively small size of the AAV genome (4680 nt) and ineffi cient AAV vector delivery of genes that are larger than the wild type (wt) size We previously demonstrated in vitro that proteasome inhibitor (PI) treatment concurrent with AAV delivery improves transduction using larger than wt expression cassettes that exceed the size of the wtAAV genome We have extended these studies using AAV2 or 8 vectors carrying a 5.6 kb factor VIII cassette in hemophilia A mice treated with or without proteasome inhibitor PI enhancement of expression was observed that was increased on average 6-fold (AAV2) and 3-fold (AAV8); enhanced expression persisted throughout the
1 year of observation In addition, a limiting dose of 1 x 1013 vg/kg AAV8.canine FVIII was delivered via portal vein to two hemophilia
A dogs after receiving PI I.V.; two hemophilic dogs received vector without PI No toxicity was observed; specifi cally, there were no abnormalities of liver transaminases, blood platelet, white blood cell
or other cell counts, or development of FVIII inhibitory antibodies The Whole Blood Clotting Time (WBCT), which is normally prolonged to greater than 20 minutes in hemophilic dogs, corrected
to the normal range (6-10 minutes) by the fi rst timepoint at one week
in the dogs receiving PI Over one year after vector delivery dogs receiving AAV8cFVIII alone had a mean WBCT of 13.6 minutes, and experienced 9 bleeding episodes apiece, a bleeding rate not different from untreated hemophilic dogs followed in parallel Dogs receiving vector with PI had a mean WBCT of 9.0 minutes, and correction of the hemophilic phenotype was evidenced by zero bleeding episodes Because male gender may be associated with higher hepatic expression from AAV vectors, a second litter of dogs was subsequently treated with two females receiving AAV + PI and one male receiving vector alone After fi ve months follow-up, the PI-treated females show the same pattern of improved hemostatic correction The fi rst litter has now been followed for 27 months remaining asymptomatic with