402 High Efficiency Transduction of Primary Human Monocytes Derived Dendritic Cells (moDCs) by Recombinant Adeno Associated Virus Vectors Containing a Point Mutation in the Surface Exposed Serine Resi[.]
Trang 1Molecular Therapy Volume 19, Supplement 1, May 2011
Copyright © The American Society of Gene & Cell Therapy S155
AAV CAPSIDS & TRAFFICKING
400 Characterization of Chimpanzee-Derived
Novel Natural Variants of AAV5
Li Zhong,1,3 Mengxin Li,1,2,4 Jun Xie,1,4 Qin Su,1 Ran He,1 Yu
Zhang,1,4 Jason Goetzmann,5 Terry Flotte,3 Guangping Gao.1,4
1 Gene Therapy Center, University of Massachusetts Medical
School (UMMS), Worcester, MA; 2 Division of Hematology/
Oncology, Department of Medicine, University of Massachusetts
Medical School, Worcester, MA; 3 Department of Pediatrics,
University of Massachusetts Medical School, Worcester, MA;
4 Department of Microbiology and Physiology Systems, University
of Massachusetts Medical School, Worcester, MA; 5 New Iberia
Research Center, University of Louisiana at Lafayette, New Iberia,
LA.
Recent development in AAV vectorology resulted in a panel of
effi cient vehicles that were originated from natural serotypes/variants,
direct evolution, or rationale design AAV5-derived vector stands out
in transduction of some major organs, such as lung, joint and brain
in animal models In an attempt to isolate natural variants of AAV5,
we rescued 30 full length capsid sequences that were closely related
to AAV5 from chimpanzee tissues including liver (Lv), heart (Ht)
and spleen (Sp) A subset of these novel AAV5 variants that were
retrieved from either cellular DNA by PCR (CHt6.1, CHt6.6, CHt6.10,
CSp8.6, and CSp8.8) or RNA by RT-PCR (CLv-M2, CLv-M6,
CLv-M8, CHt-P6 and CHt-P8) was selected for vector development
and evaluation Differences in capsid sequences between AAV5 and
these selected novel variants ranged from 1 to 5 amino acids Three
vectors can be packaged effi ciently (CHt6.1, CHt6.10 and CHt-P6),
one can be packaged at lower titer (CSp8.8), and the other six cannot
be packaged These data indicated that several amino acids are critical
for AAV5 packaging Those four vectors were evaluated in 293 cells
in vitro and showed similar transduction effi ciency as AAV5, except
for CSp8.8, which barely transduced 293 cells These four vectors
along with AAV5 were further evaluated in C57BL/6 mice in vivo
for α1-anti-trypsin (AAT) gene transduction after intravenous (i.v.),
intramuscular (i.m.), and intra-nasal (i.n.) administration at dose of
1.5×1010 - 1.0×1011 GC/animal Our data demonstrated CHt-P6 topped
all other vectors and led AAV5 in AAT expression by 2-3 folds after
i.v., i.m and i.n delivery for 12 weeks Additionally, while CHt6.10
vector led to poor AAT expression, CHt6.1 and CSp8.8 accomplished
decent transduction in between In an attempt to compare tissue/
cell tropisms, nLacZ transduction by AAV5 and CHt-P6 at dose of
lung The data did not demonstrate signifi cant differences in liver
and muscle transduction between these two vectors after i.v and i.m
injection, but showing a slight lead by CHt-P6 in nLacZ transduction
in lung after i.n delivery In the lung, CHt-P6 targeted both alveoli
and airway epithelia as did AAV5 Also, CHt-P6 did not target other
organs such as lung, muscle, heart and kidney after i.v injection
This study provided useful information on development of novel
AAV5 variants vectors, which could be an alternative choice for
gene therapy in lungs Further studies are important to elucidate the
potential correlations among the capsid structure, viral infectivity, and
critical functional domains in the AAV5 capsid, which may contribute
to its interesting vector biology
401 Widespread Transduction of Brain
Parenchyma after Intraventricular Injection of
AAV9 Compared to AAV4 and AAV5
Annagiusi Gargiulo,1 Nicolina C Sorrentino,1 Carmine
Spampanato,1 Alessandro Fraldi,1 Enrico M Surace.1
1 TIGEM-Telethon Institute of Genetics and Medicine, Naples, Italy.
Overcoming the diffi culty of effi cient delivering and distribution of
therapeutic agents globally to the CNS represents a major challenge
for the treatment of most brain disorders Current protocols for viral
vector-mediated gene delivery have showed several limits Stereotaxic injection of the vector to the CNS parenchyma results only in localized brain transduction while intra-vascular administration may results
in extensive brain transduction but is hampered by the presence of the blood brain barrier (BBB) and by the systemic exposure of the therapeutic vector Instead, intra-ventricular administration leads to a wide transduction of several brain structures of cerebral parenchyma and has the advantage of a limited extra-CNS exposure In this study
we tested transduction characteristics of adeno-associated virus (AAV) vector serotype 9 upon intra-ventricular administration and compared
it to AAV4 and AAV5 vectors, currently considered the best vector serotype upon intra-ventricular delivery (Liu et al., 2005; Fraldi et al., 2007) To evaluate levels and pattern of viral transduction we injected intra-ventricularly P0/P1 wild-type mice with viral vectors containing both Luciferase and EGFP genes Mice were examined 3 weeks after vector delivery by Luciferace assay and EGFP staining To determine precisely vector transduction distribution, luciferase levels were evaluated in different brain slices over the anterior-posterior axes Intra-ventricular administration of AAV9 outperforms all serotypes tested, mainly AAV4 on an average of 8-fold Robust luciferase expression levels were found higher in the olfactory bulb and the striatum compared to hippocampus and cerebellum EGFP signal confi rms vector transduction in these brain sections To determine cerebral cell subtypes transduction we are currently performing co-localizations staining with neuronal, glial and endothelial markers The results here reported support the use of AAV9 as a promising vector for the treatment of brain disorders affecting the CNS globally such as the Multiple Sulfatase Defi ciency (MSD)
402 High-Effi ciency Transduction of Primary Human Monocytes-Derived Dendritic Cells (moDCs) by Recombinant Adeno-Associated Virus Vectors Containing a Point Mutation in the Surface-Exposed Serine Residue
George V Aslanidi,1 Chen Ling,1 Ashley T Martino,1 Lakshmanan
Arun Srivastava.1
1 Department of Pediatrics, University of Florida Coleege of Medicine, Gainesville, FL; 2 Department of Biochemistry and Molecular Biology, University of Florida College of Medicine, Gainesville, FL.
Gene-modifi ed dendritic cells (DCs) are able to modulate antigen-presenting cell functions and induce therapeutic immunity or tolerance
in an antigen-specifi c manner However, there are confl icting reports
on the transduction effi ciency of dendritic cells (DCs) by commonly used rAAV vectors We hypothesized that rAAV2 vectors fail to transduce DCs effectively, in part, because serine/threonine-specifi c protein kinases phosphorylate surface exposed serine and/or threonine residues followed by ubiquitination of AAV capsids, which leads to cellular proteasome-mediated degradation In the present studies,
we have observed that selective inhibitors of JNK and p38 MAPK serine/threonine kinases improve the transduction effi ciency of primary human monocytes-derived DCs (moDCs) by rAAV2 up to 4- and 6-fold, respectively (Fig 1A) We reasoned, therefore, that mutations of the surface-exposed serine residues might allow the vectors to evade phosphorylation and subsequent ubiquitination and, thus, prevent proteasome-mediated degradation Indeed, site-directed mutagenesis of a serine residue at position 662 to valine (S662V)
in moDCs (Fig 1A) We next evaluated the possibility whether S662V-AAV vectors could be used for generating antigen-specifi c cytotoxic T-cells (CTLs) To this end, S662V-AAV2 vectors encoding
a truncated human telomerase (hTERT) gene were generated, and used to stimulate CTLs against a human hepatocellular carcinoma cell line, Huh7 Our data show that S662V-AAV2-hTERT-transduced DCs
Trang 2Molecular Therapy Volume 19, Supplement 1, May 2011 Copyright © The American Society of Gene & Cell Therapy
S156
AAV CAPSIDS & TRAFFICKING
resulted in rapid (within one week), specifi c T-cell clone proliferation
and generation of robust CTLs An in vitro CTL assay was set up,
and specifi c cell lysis was determined by fl uorescence-activated cell
sorting (FACS) analysis of live/dead cell ratios (Fig 1B) These
results suggest that high-effi ciency transduction of moDCs by
serine-modifi ed AAV vectors is feasible, which supports the potential utility
of these vectors for future human DC vaccine studies
Fig 1 (A) Transduction effi ciency of scAAV2 vectors in primary
human moDCs following pre-treatment of cells with JNK and MAPK
inhibitors, or following site-directed mutagenesis of the
surface-exposed serine (S) to valine (V) at position 662 (S662V) (B) Killing
curve for specifi c lysis by S662V-AAV2-hTERT vector-transduced
moDCs against Huh7 cells
403 A Capsid Composition Variant Containing
four VP Proteins Caused by a Novel Start Codon at
Position 219 of the Nucleotide Sequence of AAV8
Cap
Olga Zelenaia,1 Federico Mingozzi,1 Bernd Hauck,1 Shangzhen
Zhou,1 Katherine A High,1,2 J Fraser Wright.1,3
1 Center for Cellular and Molecular Therapeutics, Children’s
Hospital of Philadelphia, Philadelphia, PA; 2 Howard Hughes
Medical Institute, Children’s Hospital of Philadelphia,
Philadelphia, PA; 3 Pathology and Laboratory Medicine, University
of Pennsylvania School of Medicine, Philadelphia, PA.
Recombinant adeno-associated virus serotype 8 vectors generated
by transient transfection of HEK293 cells and highly purifi ed by
combined column chromatography and gradient ultracentrifugation
steps were observed to contain a VP band in addition to the canonical
VP1, 2 and 3, when assessed by SDS-PAGE / silver staining The
additional band migrated at a position intermediate between the
VP1 and VP2 as determined by Western blotting using monoclonal
antibody B1, and was designated ‘VP1.5’ Sequencing of the
packaging plasmid used to generate this vector indicated a single
nucleotide difference from the expected AAV8 Cap sequence, with G
replacing C at position 219 This nucleotide exchange is not predicted
to change the amino acid sequence of the encoded proteins, but
results in a novel CTG start codon at a position intermediate between
the VP1 and VP2 start codons Estimated by band intensity using
silver staining, or Coomassie blue staining (with higher loading)
of SDS-PAGE gels, the anomalous VP1.5 band was present at a
comparable quantity to VP1 and VP2 Dynamic light scattering
analysis indicated that highly purifi ed AAV8 particles containing
VP1.5 were indistinguishable from normal recombinant AAV8,
both demonstrating a radius in the range 13-15nm Examination of
normal and VP1.5-containing AAV8 particles by negative staining
electron microscopy indicated similar morphology; however, a slight
loss of symmetry in the variant particles was noted In a side by
side experiment to characterize the functional activity of the VP1.5
variant, a human coagulation FIX expression cassette driven by the
liver-specifi c human Alpha 1 Anti-Trypsin (hAAT) promoter was packaged using normal or VP1.5 variant capsids and the vectors were purifi ed using an optimized double cesium gradient ultracentrifugation purifi cation method (Grimm et al, 2005; Ayuso et al, 2010) Based
on titering by two methods (protein based and qPCR) to ensure comparability, the respective vectors were injected via tail vein in B57/Bl6 mice at a dose of 2.5E10 vg/mouse (fi ve mice per dose) Both vectors resulted in indistinguishable, high level expression of hFIX at all time points tested (2, 4 and 6 weeks), with > 75ug/mL of circulating hFIX measured at week 6 Animals receiving the variant capsid showed a lower anti-capsid IgG response than AAV8-injected mice These data characterize a novel capsid composition observed in recombinant AAV8, caused by a novel translational start codon in Cap, that retains normal structure and functional activity Manipulation of capsid protein VP length and composition may be exploited to further optimize recombinant vectors
404 Systematical Evaluation of Tropism of Different AAV Serotype Vectors in Adult and Neonatal Mice Following I.V Injections
Li Zhong,1,3 Mengxin Li,1,2,4 Xin Mu,1,4 Christopher Boisvert,1,4
Dmitry Ratner,1,4 Qin Su,1 Ran He,1 Sheng Chen,1,2,4 Jun Xie,1,4
Yu Zhang,1,4 Hongwei Zhang,1,4 Daniel Lucking,5 Terry Flotte,3
1 Gene Therapy Center, University of Massachusetts Medical School, Worcester, MA; 2 Division of Hematology/Oncology, Department of Medicine, University of Massachusetts Medical School, Worcester, MA; 3 Department of Pediatrics, University
of Massachusetts Medical School, Worcester, MA; 4 Department
of Microbiology and Physiology Systems, University of Massachusetts Medical School, Worcester, MA; 5 Olympus America, Inc, Center Valley, PA.
Twelve adeno-associated virus (AAV) serotypes and more than
100 AAV variants have been characterized in human and non human primate (NHP) populations, which have gained much attention as potentially useful vector for gene transfer and gene therapy Due to their different capsid surface properties, these new vectors present novel tropism profi le in vivo In this study, we compared the transgene expression in different organs after intravenously injecting different self-complementary (sc)AAV serotypes and variants - EGFP vectors
at doses of 4x1012 GC/animal in 10-week old adult mice (AAV1, 6, 6.2, 7, 9, rh10 and rh39) and 4x1011 GC/animal in postnatal 1-day old newborn mice (AAV1, 2, 5, 6, 6.2, 7, 9, rh10, rh39 and rh43) The organs and tissues were collected, sectioned and examined under
fl uorescence microscopy after 3 weeks post-injection In adult mice,
we found that AAVrh10 led to dramatical GFP expression in adipose tissues, renal tubules in kidney, pancreas, keratinocytes in skin and stomachs, although it was the least effi cient vector to transduce liver and also was modest in transducing cardiac muscles The AAV mediated-GFP expression is low in the lung, and is high in the skeletal muscles without much difference among all the vectors In neonatal mice, we found that the transduction profi le was different The AAV-mediated GFP expression was rare in pancreas and gastric intestines The transduction effi ciency was low in liver and stomachs But all the vectors led to strong GFP expression in the skeletal muscles and cardiac muscles In the kidney, the AAV2 vector transduced renal tubules and glomeruli very effi ciently, whereas AAVrh10 led to highly-effi cient GFP expression in renal papilla In the skin, AAVrh39 and AAVrh10 led to high GFP expression in keratinocytes In the lung, AAVrh10 resulted in highly effi cient GFP expression, and both alveolar and airway epithelia cells were targeted AAV7 and AAV9 also transduced the lung well, however, primarily targeted alveolar cells Overall, our data not only provided the AAV tropism information after high dose I.V administration, but also indicated some AAV vectors have unique tropism, such as AAVrh10, which targeted kidney,