45 Implementation of a Reproducible and Effective Process for the GMP Production of Lentiviral Vectors Molecular Therapy Volume 17, Supplement 1, May 2009 Copyright © The American Society of Gene Ther[.]
Trang 1Molecular Therapy Volume 17, Supplement 1, May 2009
Copyright © The American Society of Gene Therapy
S18
CELL PROCESSING AND VECTOR PRODUCTION
43 A Novel Inducible System for Highly
Effi cient Production of Recombinant
Adeno-Associated Virus (rAAV) Vectors in Insect Sf9 Cells
George Aslanidi,1 Kenneth Lamb,1 Sergei Zolotukhin.1
1 Pediatrics, University of Florida, Gainesville, Fl.
Production of clinical grade gene therapy vectors for human trials
remains a major hurdle in advancing cures for a number of otherwise
incurable diseases In the current study, a novel simple and effi cient
system of rAAV production in insect cells is described The system
takes advantage of DNA regulatory elements from two unrelated
viruses - Autographa californica multiple nuclear polyhedrosis virus
(AcMNPV) and AAV2 The endpoint design consists of only two
components: 1) stable Sf9-based cell line incorporating integrated
copies of rep and cap genes, and 2) Bac-GOI (gene of interest fl anked
by AAV inverted terminal repeats) Rep and cap genes are designed
to remain silent until the cell is infected with Bac-GOI helper which
provides both rAAV transgene cassette and immediate-early (IE-1)
transcriptional transregulator Infection with Bac-GOI initiates the
rescue/amplifi cation of the integrated AAV helper genes resulting
in dramatic induction of the expression and assembly of rAAV The
integration cassette incorporates an IE-1 binding target sequence from
wild type AcMNPV, a homologous region 2 (hr2) In addition, the
cassette includes Rep-binding element (RBE), a target site for AAV2
Rep78/68 proteins By binding to RBE, Rep proteins initiate a feed
forward loop mediating the rescue/amplifi cation of the integrated
cassette All four Rep proteins (78, 68, 52, and 40) are expressed
from one uninterrupted AAV2 rep sequence thus utilizing Sf9 cells
splicing machinery Furthermore, the system had been also utilized
to modulate the stoichiometry of VP1/VP2 proteins thus increasing
phospholipase A2 content of the particle and improving infectivity of
alternative serotypes such as AAV8 produced in insect Sf9 cells The
arrangement provides high levels of Rep and Cap proteins in every
cell thus improving rAAV yields by 10-fold The described vectors
are modular in design and could be utilized for the production of
other multiprotein complexes
44 Effi cient Construction of Producer Cell
Lines for SIN, Clinical Lentiviral Vectors by
Concatemer Array Transfection
Robert E Throm,1 Annastasia A Ouma,1 Anantharaman
Chandrasekaran,1 Timothy Lockey,1 Michael Greene,1 Sheng
Zhou,1 Derek A Persons,1 Harry Malech,2 Brian P Sorrentino,1
John T Gray.1
1 St Jude Children’s Research Hospital, Memphis, TN; 2 NIAID,
NIH, Bethesda, MD.
We describe a transfection and selection method for introducing
vector genome expression cassettes into packaging cells which
effi ciently generates high titer producer cell lines for vectors containing
self-inactivating (SIN) LTRs and chromatin insulators These vector
design features have been shown to reduce genotoxic side effects, and
yet clinical production of such vectors frequently utilizes cumbersome
transient transfection methodology, as stable producer cell derivation
is hampered by the fact that SIN vector genomes cannot be introduced
into packaging cells by viral transduction methods We optimized
transfection and selection methods of stable producer cell derivation
by constructing in vitro ligated concatemeric arrays containing a
25:1 ratio of a SIN, GFP vector genome expression cassette and an
antibiotic resistance cassette, which were subsequently transfected
into our newly developed doxycycline regulated HIV-VSVG stable
packaging cell line After selection with drug, individual clones were
isolated, expanded, and screened for productive titer When compared
to clones isolated after transfection of supercoiled plasmids mixed at
identical ratios, clones derived from concatemeric array transfection
yielded 10-fold higher titers (mean titer=1.76x107 tu/ml vs 1.8x106
tu/ml, p=0.0002) We additionally showed by Southern blot analysis
of producer cells that engineering of the concatemeric array to ensure the directional assembly of monomers facilitated the stable
maintenance of the array in vivo We have successfully applied this
new methodology to generate stable producer cell lines for 4 separate SIN, insulated clinical vectors, all of which produce titers comparable
to or exceeding those generated by transient transfection The furthest progressed vector expresses the human common γ-chain for SCID-X1 gene therapy, and our top clone reliably generates unconcentrated titers of 3-10 x 107 tu/ml on ED7R cells (a γ-chain defi cient human T-cell line) This producer line has been expanded under GMP and vialed as a Master Cell Bank for clinical production Cells derived from this bank were passaged without selection for one month prior
to seeding in a small scale WAVE bioreactor, which after induction generated high titer supernatants (>3x107 tu/ml) for 6 consecutive daily harvests Another line developed for SCID-X1 gene therapy generates the IL2RG-Revgen vector, which contains the entire genomic locus of the human common γ-chain gene in the reverse orientation Titers from this line also exceed 107 tu/ml The third line produces an EF1α driven gp91phox vector for treatment of chronic granulomatous disease, and the fourth contains a complex γ-globin vector for treatment of beta-thalassemia and sickle cell anemia Both of these lines generate titers greater than 106 tu/ml, and yet still await production optimization In summary, the concatemeric array transfection method allows reliable generation of high-titer stable producer cell lines for safety modifi ed, clinically relevant vectors
45 Implementation of a Reproducible and Effective Process for the GMP Production of Lentiviral Vectors
Patricia Noguiez-Hellin,1 Otto-Wilhelm Merten,1 Sylvain Fauchille,1 Nicolas Laroudie,1 Céline Dugué,1 Sabine Charrier,1 Maria-Antonietta Zanta-Boussif,1 Helene Chautard,1 Didier Caizergues,1 Marina Radrizzani,2 Alessandro Aiuti,3 Luigi Naldini,3 Anne Galy.1
1 Genethon, Evry, France; 2 Molmed, Milan, Italy; 3 Tiget, San Raffaele Hospital, Milan, Italy.
We have obtained encouraging preclinical data supporting the development of a lentiviral vector (LV) for clinical gene therapy application in Wiskott Aldrich Syndrome (WAS) To implement the highest standards of quality and safety, the clinical studies will be performed with highly-purifi ed vector particles For this purpose,
we developed a process for the production of LV in cGMP This process is based on the transient transfection of a selected subclone
of HEK293T cells followed by the downstream purifi cation and concentration of the VSVg pseudotyped viral particles The process has been carefully validated jointly both at Genethon and Molmed/ Tiget We report here the manufacture of clinical grade vectors which was successfully implemented in Genethon’s manufacturing facility Two batches of the WAS lentiviral vector were produced in GMP The manufacturing operations started by cultivating cells from the master cell bank in 24 ten-stack CF-10 (CellFactories) After amplifi cation, the cells were transfected with 4 GMP-grade plasmids bringing in the functions necessary for the production of the vector (rev, gag-pol, VSV-g, vector with the therapeutic transgene) The supernatant fl uid containing the LV was harvested, fi ltered, purifi ed and concentrated by
a multi-step downstream processing protocol including ion exchange chromatography, concentration/diafi ltration, and gel fi ltration Such vector preparation was purifi ed by a factor of about 1000 fold from protein and DNA contaminants and concentrated about 200 fold In the fi nal product, specifi c host cell contaminants such as SV40large
T antigen or E1A sequences were shown to be reduced Process by-products such as residual benzonase were no longer detectable High titers of at least 1x 109 i.g./ml were obtained reproducibly in both batches, yielding a total amount of at least 2.5 x 1011 i.g produced
Trang 2Molecular Therapy Volume 17, Supplement 1, May 2009
CELL PROCESSING AND VECTOR PRODUCTION
per batch The vector integrity was confi rmed by analysis of the
integrated proviral sequences and RCLs were not detectable The
vector was functional as demonstrated by gene transfer or transduction
of immortalized WASp-defi cient B cell lines or CD34+ cells, and
there was no evidence of cellular toxicity induced by the vector
These functionality assays revealed that the vector introduces about
1 vector copy per cell in hematopoietic progenitor cells when the
vector concentration was 5-10 x 107 ig/ml Under these conditions
the transduction effi ciency was about 40-50% The WAS LV batch
produced at Genethon was therefore conform and released for clinical
use In conclusion, the WAS project has permitted the implementation
of large-scale GMP production, purifi cation, and control of advanced
HIV1-derived lentiviral technology at Genethon which should be
useful for other applications
46 Production of Lentivirus by Transient
Transfection of HEK 293T Grown on Spherical,
Polystyrene Microcarriers
Scott R Witting,1 Aparna Jasti,1 Sarah Dolan,1 Kenneth Cornetta.1
1 Medical and Molecular Genetics, Indiana University School of
Medicine, Indianapolis, IN.
Lentiviral vectors have evolved into an effective means of obtaining
long-term transgene expression and a valuable tool for human
gene therapy Therefore, techniques to effi ciently produce large
amounts of clinical grade vector are needed Transient transfection
of mammalian cells with a plasmid cocktail containing the necessary
vector components and transgene is a proven method of obtaining
high titer lentiviral vector Advantages include time/cost savings by
eliminating the tedious process of developing packaging cell lines
and the ability to quickly utilize newly characterized pseudotypes
Because transient lentivirus production on a large scale requires the
cumbersome task of manipulating numerous cell culture vessels (i.e
cell factories), we explored methods to simplify the process One
possibility is the use of microcarriers as a cell growth substrate to
replace the surface area of multiple culture vessels and downsize to
a single vessel or bioreactor To investigate this possibility, Hillex II
microcarriers were added to small shaker fl asks such that 100 cm2 of
growth surface was present Initial cell seeding experiments of HEK
293T cells were performed to determine the cell count and seeding
time required to obtain 70-90% microcarrier surface coverage with
less than 10% unbound cells The optimum conditions were found
to be 6.25 x 106 cells in 5 mL total volume, a seeding time of three
hours, and gentle mixing every 45-60 minutes Cells were transfected
with a plasmid cocktail, including vector plasmids and a GFP
transgene plasmid, using jetPEI in concentrations consistent with
those successfully used in static culture Infectious titer, as measured
by GFP expression, averaged 5.7 x 106 IU/mL When scaled up to
a 500 mL wave bioreactor, Hillex II microcarriers tended to clump
without strong agitation - which in turn caused cell detachment We
next compared lentivirus production with Hillex II (specifi c gravity
= 1.11) to a lower density microcarrier, ProNectin F (specifi c gravity
= 1.02), in shaker fl asks Infectious titers were similar at 2.3 x 106
and 2.9 x 106 , respectively Additionally, the inclusion of 5% fi coll
in the media was evaluated to increase the suspension effi ciency of
ProNectin F microcarriers This would allow for less harsh agitation
conditions in wave bioreactor productions Infectious titers from cells
in 5% fi coll were comparable to those without fi coll (3.2 x 106 and 2.9
x 106 IU/mL, respectively) Current work involves scaling up harvest
volumes in a wave bioreactor using ProNectin F microcarriers and
media containing 5% fi coll
47 Rapid Generation of Genetically-Modifi ed Primary T Cells and Antigen-Specifi c CTL for Preclinical and Clinical Applications Using a Novel Cell Bioreactor
Juan F Vera,1 Lara Brenner,1 Ann M Leen,1 Minhtran C Ngo,1 John Wilson,2 Helen E Heslop,1 Gianpietro Dotti,1 Cliona M Rooney.1
1 Baylor College of Medicine, Texas Children’s Hospital, The Methodist Hospital, Houston; 2 Wilson Wolf Manufacturing, New Brighton, MN.
The generation of genetically-modifi ed primary T cells and antigen-specifi c cytotoxic T lymphocytes (CTL) for adoptive transfer is time-consuming, requiring 4-12wks to produce suffi cient cells for therapeutic purposes, and expensive (media + plastics + cytokines + man hours), impeding the broader clinical application of T cell therapy T cell growth is limited by gas exchange, nutrients, and waste accumulation Bioreactors developed to overcome these limitations tend to be complex, involving mechanical rocking or stirring and continuous perfusion, which increases the expense and limits the number of products to the number of mechanical devices that can
be housed and maintained We have now explored the use of a new static gas permeable (GP) Cell Bioreactor for T cell expansion This device is essentially a fl ask with a gas permeable membrane supported
by a plastic lattice as its base The O2/CO2 exchange from the base allows large volumes of media to be added thereby reducing nutrient limitations and waste build-up, and consequently the manipulation required to sustain cell expansion We tested 2 sizes, 10cm2 (GP40) and 100cm2 (GP2000) that hold a maximum of 40mL and 2000mL of media, respectively We fi rst evaluated the expansion of genetically-modifi ed OKT3-stimulated T cells engrafted with a chimeric T cell receptor targeting the Kappa light chain of the B cell The cultured cells required minimal manipulation, with a single media change per week, while those cultured in 24-well plates required manipulation every 2-3 days (media replenishment and reseeding) to sustain optimal expansion Using the GP bioreactor we observed a 54.6 fold T cell expansion within 7 days (range 37.8-64.6) vs 15.3 fold (range 11.8-29.2) Evaluation of the cells cultured using both systems revealed less apoptosis in the bioreactor-expanded cells, suggesting that the increased cell numbers resulted not from an increased rate
of proliferation but from a cumulative increase in viable cells Importantly T cell phenotype and function was preserved Cytolytic activity evaluated by Cr51 release showed strong killing of a Kappa+
B cell tumor (Daudi) (99±5%vs89±2%, Bioreactor vs plate) with low killing of an irrelevant target, K562 (15±15%vs10±6%) Using
a single GP2000 we could produce up to 8E+08 genetically-modifi ed
T cells, which would have required approximately 320 wells (>13 plates) We also evaluated the expansion of Epstein-Barr virus (EBV) CTL by coculturing APCs with established EBV-CTL at
an optimized cell density These culture conditions enhanced CTL expansion (42.5 fold ±14.8 vs 3.4 fold ±1.2 within 7 days) without requiring media change Their phenotype and antigen specifi city, as evaluated by tetramer and IFN-γ ELIspot, was similar to conventional CTLs In summary, we have successfully utilized GP Cell Bioreactor technology to induce optimal in vitro T cell expansion with minimal handling We have also demonstrated that this system is suited to the clinical grade expansion of other cell types