418 Application of Therapeutic Artificial miRNAs in the CNS Non Allele Specific Silencing of Mutant and Wildtype Huntingtin Demonstrates Therapeutic Efficacy in Huntington''''s Disease Mice Molecular The[.]
Trang 1Molecular Therapy Volume 17, Supplement 1, May 2009
broader and more rapid use of the OPEN ZFN selection methodology
and provide additional support for the potential use of ZFNs as an
important approach for gene therapy
416 Treatment of Canine Leukocyte Adhesion
Defi ciency Using a SIN Lentiviral Vector and
Human CD18 Promoter Expressing Canine CD18
Michael J Hunter,1 Laura M Tuschong,1 Cedar J Fowler,2
Everette J R Nelson,1 Tanya H Burkholder,3 Thomas R Bauer,
Jr.,1 Dennis D Hickstein.1
1 Experimental Transplantation and Immunology, National
Cancer Institute, Bethesda, MD; 2 HHMI/NIH Research Scholars
Program, Howard Hughes Medical Institute, Chevy Chase, MD;
3 Department of Veterinary Resources, National Institutes of Health,
Bethesda, MD.
Hematopoietic stem cell gene therapy would be enhanced by the
development of vectors harboring tissue-and developmental
stage-specifi c cellular promoters to express the therapeutic transgene in the
target cell population This approach would mitigate the potential
adverse effects of inappropriate tissue expression and might be
expected to reduce insertional activation of nearby oncogenes that
lead to oligoclonal hematopoiesis and leukemia To develop and test
a modifi ed promoter cassette with the features described above for
our target disease canine leukocyte adhesion defi ciency (CLAD), we
cloned portions of the human CD18 promoter (1 Kb, 776 bp, and 306
bp) into a SIN lentiviral vector upstream of the canine CD18 cDNA,
and used this vector to transduce CLAD CD34+ cells In CLAD,
defects in the leukocyte integrin CD18 result in the inability to express
CD11/CD18 heterodimers on the leukocyte surface leading to
life-threatening bacterial infections Transduction of CLAD CD34+ cells
in vitro with the SIN lentiviral vector with the 1 kb human CD18
promoter resulted in the highest percentage of CD18+ cells; nearly
15% of the CLAD CD34+ cells were CD18+ when assessed 5 days
after an overnight transduction We used the SIN lenti vector with the 1
Kb human CD18 promoter to treat two dogs with CLAD Autologous,
CLAD CD34+ cells were transduced overnight and infused following
a single, non-myeloablative dose of 200cGy total body irradiation
(TBI) The percentage of CD18+ leukocyte compartments 8 weeks
following infusion of the two dogs were comparable: dog A1, CD18+/
PMN 0.3%, CD18+/CD3+ cells 0.4%, CD18+/B-cells 0.7%, CD18+/
CD14+ cells 0.9%; dog A2, CD18+/PMN 0.5%, CD18+/CD3+
cells 0.8%, CD18+/B-cells 1.2%, CD18+/CD14+ cells 1.4% Both
treated dogs are now 4 months of age and have had correction of
the CLAD phenotype In contrast, untreated CLAD dogs succumb
to overwhelming infection within a few months of life Reversal
of the CLAD phenotype with low numbers of CD18+ neutrophils
results from selective migration of CD18+ neutrophils from the blood
into the tissues Although both CLAD dogs in this study have had a
clinical response, additional regulatory elements of the human CD18
promoter/enhancer will be required to ensure long-term reversal of
the phenotype in this and other genetic leukocyte diseases These
studies represent requisite translational studies in the development
of new vector designs for the treatment of children with the human
counterpart of CLAD, namely LAD
417 Effi cient MGMTP140K-Mediated Selection
of Long Term Repopulating Cells in a Nonhuman Primate Model
Brian C Beard,1 Grant D Trobridge,1,2 Megan L Welsh,1 Hans-Peter Kiem.1,2
1 Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA; 2 Department of Medicine, University of Washington, Seattle, WA.
In vivo selection of genetically modified hematopoietic repopulating cells has many potential therapeutic applications For some applications which require relatively high levels of gene marking, such as hemoglobinopathies, in vivo selection may be required to increase initially low levels of gene-modifi ed cells Here we demonstrate effi cient post-transplantation selection of long-term hematopoietic repopulating cells using methylguanine methyltransferase (MGMTP140K) in a primate model In vivo selection was studied in macaques (M nemestrina) that received CD34-enriched cells transduced with VSVG-pseudotyped HIV-derived lentivirus vectors expressing MGMTP140K and a reporter gene green fl uorescent protein (GFP) or MGMTP140K only Two macaques were conditioned with a myeloablative dose of total body irradiation and a third macaque was conditioned with a nonmyeloablative dose of busulfan (4 mg/kg/day for 2 days) After stable engraftment monkeys were treated with O6-benzylguanine (O6BG) and BCNU Following myeloablative transplantation the monkeys transplanted with cells gene-modifi ed with a vector expressing MGMTP140K and GFP, in vivo selection was determined
by fl ow cytometry In one monkey (following 2 cycles of O6BG/ BCNU) the granulocytes rose from ∼20% to 70% and in the lymphocytes from ∼20% to 50% In the second monkey (following
a single cycle of O6BG/BCNU) the granulocytes rose from ∼25%
to 42% and in the lymphocytes from ∼12% to 22% Following nonmyeloablative transplantation and a single cycle of O6BG/BCNU the monkey transplanted with cells gene-modifi ed with the vector expressing only MGMTP140K increased overall gene marking, determined by real time (RT)-PCR, in total white blood cells rose from a provirus copy number of 0.04 (∼4% gene marking) to 0.16 (∼16% gene marking) Aside from transient elevated liver enzymes following O6BG/BCNU treatment no additional extra-hematopoietic toxicity has been observed Importantly, multilineage selection of hematopoietic cells was achieved and clonality studies are underway using a combination of LAM-PCR and a modifi ed whole genome pyrosequencing approach In summary, MGMT selection is effi cient and well tolerated in macaques and these large animal studies should
be highly predictive for clinical applications and will help to further improve HSC gene therapy
Oligonucleotide Therapies
418 Application of Therapeutic Artifi cial miRNAs in the CNS: Non-Allele-Specifi c Silencing
of Mutant and Wildtype Huntingtin Demonstrates Therapeutic Effi cacy in Huntington’s Disease Mice
Ryan L Boudreau,1 Jodi L McBride,1 Ines Martins,1 Shihao Shen,1
Yi Xing,1 Barrie J Carter,2 Beverly L Davidson.1
1 University of Iowa, Iowa City, IA; 2 Targeted Genetics, Seattle, WA.
RNA interference (RNAi) provides a promising therapeutic approach to treat several human diseases However, the safety of RNAi-based therapies remains a concern Previously, we compared the effi cacy and safety of short-hairpin RNA (shRNA) and artifi cial microRNA (miRNA) expression vectors in vitro and in vivo We found that shRNAs are more potent but induce toxicity in cell cultures and in mouse brain, whereas artifi cial miRNAs are expressed at lower levels and display better safety profi les We have since tested the
Trang 2Molecular Therapy Volume 17, Supplement 1, May 2009 Copyright © The American Society of Gene Therapy
S164
artifi cial miRNA strategy in a pre-clinical therapeutic trial aimed at
assessing the viability of a non-allele-specifi c silencing strategy for
Huntington’s disease (HD) HD is a dominant, fatal neurodegenerative
disease caused by mutant huntingtin (htt) protein which primarily
affects the striatum Currently, there are no effective treatments
for HD, thus warranting the investigation of novel therapeutic
strategies Recently, we and others demonstrated that silencing
mutant htt using RNAi provides therapeutic benefi t in HD mice
We have since found that silencing wildtype htt in adult mouse
striatum does not induce overt toxicity However, given the role of
htt in various cellular processes, it remains unknown whether
non-allele-specifi c silencing of both wildtype and mutant htt is a viable
therapeutic strategy for HD Here, we tested whether co-silencing
wildtype and mutant htt provides therapeutic benefi t and is tolerable
in HD mice Mice were injected intrastriatally with viral vectors
expressing htt-specifi c artifi cial miRNAs Following treatment, HD
mice showed signifi cant reductions in both wildtype and mutant
htt, and demonstrated improved motor coordination and survival
As a preliminary assessment of the safety of a non-allele-specifi c
silencing strategy for HD, we performed transcriptional profi ling to
evaluate the effects of reducing wildtype htt in adult mouse striatum
We identifi ed gene expression changes that are concordant with
previously described roles for htt in various cellular processes In
addition, several abnormally expressed transcripts associated with
early-stage HD were also differentially expressed in our studies, but
intriguingly, those involved in neuronal function changed in opposing
directions Whether this reversal of HD-related transcriptional defi cits
will occur and prove benefi cial in a therapeutic setting remains to
be determined Together, these encouraging and surprising fi ndings
support further testing of non-allele-specifi c RNAi therapeutics for
HD This work was funded by the Lori C Sasser Fellowship, National
Institute of Health and Hereditary Disease Foundation
419 Lethal Toxicity Caused by Striatal Delivery
of shRNAs in Mouse Models of DYT1 Dystonia:
Implications for Therapeutic Design
Janine N Martin,1 Nicolle Wolken,6 Timothy Brown,4 William T
Dauer,3 Michelle E Ehrlich,2 Pedro Gonzalez-Alegre.5
1 Genetics, U of Iowa, Iowa City, IA; 2 Neurology, Mt Sinai School
of Medicine, New York City, NY; 3 Neurology and Pharmacology,
Columbia U, New York City, NY; 4 Biochemistry and Molecular
Biology, Thomas Jefferson U, Philadelphia, PA; 5 Neurology, U of
Iowa, Iowa City, IA; 6 Neurology, U of Iowa, Iowa City, IA.
DYT1 is the most common inherited dystonia, a disabling
neurologically based movement disorder This incurable disease is
caused by the deletion of a glutamic acid residue in the protein torsinA
(torA(∆E)) A common, dominantly inherited mutation likely resulting
from a dominant negative effect of torA(∆E) over torA(WT) indicates
allele-specifi c silencing of torA(∆E) could be a potential therapeutic
strategy for DYT1 We have previously tested shRNAs that achieve
this goal in cultured neuronal cells without triggering infl ammatory
responses and we completed the following studies to determine the
effi cacy of those hairpins in vivo Two different mouse models of the
disease, DYT1 knockin (KI) mice (129/SvJ strain) and transgenic
(TG) mice overexpressing human torA(∆E) under the DARPP32
promoter (DARPP32-TorA(∆E)) (C57BL/6 background), along with
control littermates were used The study design included bilateral
striatal injections of rAAV2,1.CMV.GFP vectors encoding either
U6shTorA(∆E) (therapeutic vector), U6shMis (control mismatched
shRNA) or no shRNA at 2-6 months of age and behavioral testing
(open fi eld, rotarod and staircase reaching tests) at baseline and post
injection before sacrifi cing the mice for biochemical and histological
analysis Unexpectedly, a cohort of DYT1 KI and control mice (n:
10 per genotype/vector) displayed mortality rates of 70% and 53%
for those receiving the therapeutic and control hairpins respectively,
occurring 3-4 weeks post injection, while the GFP only control vector did not induce toxicity In parallel experiments completed in DARPP32-TorA(∆E) TG mice and control littermates, the mortality rate was 41% and 28% in those receiving the therapeutic vector and control shMis respectively, with no toxicity in the GFP only group
Mortality in the TG model was reduced when compared to the KI
model and occurred 6-8 weeks post injection, perhaps refl ecting the different genetic background This delay allowed us to complete behavioral analysis While baseline behavioral analysis did not display any difference, TG and control mice receiving the therapeutic vector and control shRNAs exhibited signifi cant hyperactivity in open fi eld behavior and did signifi cantly worse on rotarod testing 6 weeks post injection when compared to those injected with GFP only vectors, indicating striatal neuronal dysfunction In conclusion, our studies demonstrate that expression of U6-shRNA in the mammalian brain can lead to fatal toxicity, even when cell culture studies did not predict toxicity Furthermore, the genetic background of rodents modifi es their sensitivity to this form of toxicity, a factor that should be taken into consideration in the design of therapeutic RNAi trials Future studies will explore the mechanism of toxicity and modifying the therapeutic vector to abolish toxicity
420 Rational Design of MicroRNA-Based RNA Interference Leads to More Potent Targeting of
Hepatitis B Virus and Long Term Viral Inhibition In
Vivo
Kathy Keck,1 Ryan Spengler,1 Maria Scheel,1 Ye Ding,2 Anton P McCaffrey.1
1 Internal Medicine, University of Iowa, Iowa City, IA; 2 Wadsoorth Center, New York Department of Health, Albany, NY.
Hepatitis B virus (HBV) is a small DNA virus that chronically infects 400 million people Previously, we conducted proof-of-principle experiments showing that RNA interference (RNAi)
could degrade HBV RNAs in mice (McCaffrey et al 2003 Nature
Biotechnology 21, 639) Recently Grimm et al expressed the short
hairpin RNA (shRNA), HBVU6#2, described in our previous study, using self-complementary adeno-associated virus serotype 8
(scAAV8) in HBV transgenic mice (Grimm et al 2006 Nature 441,
537) While this resulted in substantial HBV knockdown in mice,
it also resulted in acute toxicity and death The authors concluded that the high levels of shRNA expression required to observe HBV knockdown oversaturated the RNAi machinery preventing endogenous microRNAs (miRNAs) expression We have utilized recent mechanistic insights to rationally design more potent HBV RNAi triggers We expressed our HBV RNAi triggers in the context
of the endogenous miRNA, miR30 All our rationally designed HBV RNAi triggers showed signifi cant silencing and eight were signifi cantly more potent than HBVU6#2 A two step model was used to model the hybridization of the guide strand with the target RNA A regression analysis identifi ed thermodynamic features that were highly correlated with RNAi activity Interestingly, miRNA-based RNAi triggers compete with endogenous miRNAs Pairs of miRNA-based RNAi triggers were expressed in HBV transgenic mice using scAAV8 HBV serum viral genomes were reduced by two logs over 50 days HBV surface antigen expression was similarly reduced Inhibition slowly diminished until remaining stable at ∼90% knockdown for over 120 days No fatality was observed Thus, we have demonstrated long term stable viral inhibition using scAAV8 delivery of miRNA-based RNAi These general approaches can be applied to inhibiting other viruses as well as treating dominant genetic disorders or cancer Supported by NIH 1R01AI068885-01A1