Download Gene therapy delivery tools poised for success in ocular

Retina
Gene therapy delivery tools poised for
success in ocular gene therapy
Gearoid Tuohy PhD
in Monte Carlo
A NUMBER of years ago at an international
gene therapy conference a leading academic
researcher commented that the three great
challenges for making gene therapy a
realistic treatment option were “delivery,
delivery, and delivery”. This unfortunately
was not an over-statement. However, in the
past 10 years significant strides have been
made in developing technologies that are
now capable of highly specific reproducible
delivery of exogenous genes into a broad
range of tissues and cells.
At this year's EURETINA conference,
two leading gene therapy researchers in
ocular disease described the current stateof-the-art regarding gene delivery to the
retina. These technologies recently
culminated in the initiation of the first
human gene therapy trial for the treatment
of Leber congenital amaurosis (LCA), at
London's Moorfields Eye Hospital.
The medical and scientific community are
now identifying therapeutic opportunities
to deliver specific gene transcripts in an
effort to correct the course of the
underlying pathology. One of the keys to
achieving such ambitious goals is “vector
technology” – the technology used for
getting foreign genes into cells and by far
one of the most popular tools for gene
delivery are virus vectors.
Different vectors are now available for
transfecting a wide range of tissues and
cells. Each specific vector has advantages
and disadvantages that chiefly depend on
the gene that needs to be delivered and the
target cell or tissue for delivery.
Adenovirus has been the most used gene
therapy vector in clinical trials to date.
However, a large volume of research into
applying gene therapy in ocular disease has
focused on the use of adeno-associated
vectors (AAV).
Alberto Auricchio MD, based at the
Telethon Institute of Genetics and Medicine
(TIGEM) and Department of Paediatrics,
“Federico II” University, Napoli, Italy, told
delegates of recent progress made on
adeno-associated virus (AAV) – a gene
therapy vector which possesses several
features which make it attractive for gene
delivery to a range of organs in vivo,
including the retina.
AAV contains single stranded DNA and
is non pathogenic to humans. The virus is
able to transduce both dividing and nondividing cells such as those in the retina
over the long term following a single subretinal administration. One of the most
attractive features of AAV is that it exists in
over 100 different “flavours”, called
serotypes, which have been isolated in the
last few years. Such a portfolio of serotypes
is extremely valuable as it permits
48
researchers to take each serotype and
convert it into a gene transfer vector
capable of infecting specific cell types in a
way that is both efficient and simple.
The AAV particle is made of a genome
encased in a protein from which biologists
can derive gene therapy vectors by
exchanging parts of the capsid protein. This
is useful because each of the surface
proteins on the capsid is the main
determinant of a vector's ability to infect
certain cell types and drive expression of
the particular gene of interest.
The ability of AAV vectors with different
external surface proteins to transduce
different cell types has been demonstrated
by Dr Aurrichio's group and others using
the retina as the target tissue.The broad
range of different serotypes, some of which
have been converted into gene therapy
vectors, are currently being tested for their
ability to transduce the retina.
Turning genes on only where they are
required
Choosing a suitable capsid type represents
just one way to direct or target gene
expression in specific cell types. Another
way to direct gene expression following
subretinal administration or vitreal injection
is to engineer the gene of interest with cell
type specific promoters – sequences of
DNA upstream of the gene that regulate
gene expression. For example, expression
of a given transgene may be restricted to a
specific cell type such as the retinal pigment
epithelia (RPE) or ganglion cells by using
gene promoters specific to transcripts
expressed in those cell types.
Alternatively, using the rhodopsin or
rhodopsin kinase promoter limits
expression of a transgene to rod
photoreceptors.This capability provides
additional comfort in terms of regulatory
safety concerns to insure that a given
transgene is expressed only in the target
cell.
Dr Aurrichio showed examples of how
AAV can be used to correct defects in
some animal models including severe retinal
diseases where most of the genes are
expressed in photoreceptors and require
gene replacement as a therapeutic strategy.
For example in recessive RP, brought about
by a homozygous mutation of the ≤
subunits of the phosphodiesterase (≤PDE)
gene, delivery of a normal ≤PDE gene has
brought about functional rescue from the
disease state.
In other ocular applications, for example
in dominant disease, RNAi can be used to
specifically knock down the mutated allele
leaving the normal allele untouched.This
approach has been tested in an ADRP
(autosomal dominant retinitis pigmentosa)
model carrying a Pro23His rhodopsin
mutation whereby in vitro and in vivo
Table of vectors used
expression of RNAi can knock down the
mutant rhodopsin by 70 per cent leaving
the wild type rhodopsin gene untouched.
Gene therapy may also be applied to
animal models of ocular neovascularisation
such as retinopathy of prematurity (ROP)
and AMD. One model of ROP that induces
neovascularisation exposes newborn mice
to high levels of oxygen and then returns
animals to normal oxygen levels at which
point an AAV carrying an angiogenic gene
can be introduced to block
neovascularisation.
Lentiviral vectors provide an
alternative delivery tool
A very different viral vector used for gene
therapy applications was presented by Dr
Kam Balaggan MRCOphth, specialist
registrar at Moorfields Eye Hospital and
research fellow in the Division of Molecular
Therapy, Institute of Ophthalmology,
London headed by Prof Robin Ali. Lentiviral
vectors, unlike adenovirus and AAV, are
RNA viruses with a capacity to efficiently
transduce both dividing and non-dividing
cells and are associated with minimal
intraocular immune responses.
The principal advantages of lentiviral
vectors are that they accommodate a large
carrying capacity and mediate high levels of
transgene expression. For certain
applications they can deliver genes too
large for AAV to carry while also having
predictable kinetics of expression, typically
expressing at a very early stage following
administration. Nearly all the different types
of lentiviral vectors described to date can
efficiently target ocular RPE cells with some
possessing the capacity to also transduce
photoreceptors and other neurosensory
retinal cells with varying efficiency after
conventional subretinal delivery.
Lentiviral vectors may be derived from
primate sources such as HIV and the simian
equivalent (simian immunodeficiency virus)
or from non-primate sources such as horse
(equine infectious anaemia virus-EIAV), cat
(feline immunodeficiency virus-FIV) and
cow (bovine immunodeficiency virus-BIV).
The principal advantage of such derivatives
is that their use further minimises any
potentially deleterious consequences given
that their wild type forms are known to be
non-pathogenic in humans. Lentiviral
vectors integrate efficiently into the host
cell genome and so the transgene may
propagate to all daughter cells following cell
division.
The wild type HIV genome, typically
around 9,700 base pairs in size, can be
engineered into a gene therapy vector
because most of the pathogenic
components of these viruses can be safely
removed without causing any diminution in
either viral titre or activity.To create vector
tools the viral genome may be split across
three separate plasmids, coding for all the
essential sequences necessary for
transcription and integration of the viral
genome, insertion of the gene of interest
and a back bone with as little as 550 base
pairs (down from the original 9,700 base
pairs) so very little of the original
pathogenic vector remains.
As two of the plasmids do not have any
encapsidation sequence they do not get
incorporated into the viral particle but
function purely to provide the structural
proteins and enzymes necessary for correct
functioning of the lentiviral particle.The
natural tropism for wild type HIV-1 is the
CD4+ lymphocyte. However, it is possible
to expand this affinity beyond CD4+ cells
by removing the original protein encoding
the wild type viral envelope and replacing
with a gene encoding the envelope protein
from a different virus.
Dr Balaggan and colleagues have tested a
number of re-engineered vectors for
delivery of genes to the retina in animal
Retina
models, using green fluorescent protein as a
marker gene to track transduction and
expression efficiencies. Using the EIAV
vector Dr Balaggan's research group have
stably and efficiently transduced retinal
pigment epithelial cells in animal models
resulting in high levels of GFP expression
within just three days of subretinal
administration and persisting for at least 16
months.
The implications of maintaining transgene
expression for 16 months following a single
injection in itself illustrates the
attractiveness of gene therapy because such
an approach could be used for the longterm delivery of therapeutic proteins. For
example in exudative AMD, genes encoding
anti-VEGF proteins could be delivered
resulting in sustained VEGF targeting after
just a single injection, in contrast to the
repeated delivery that is required of
current pharmacological anti-VEGF agents
including Lucentis and Avastin. As it is likely
that life-long administration of
pharmacological agents will be required to
sustain therapeutic efficacy, the advantages
of a gene-based approach for patients with
exudative AMD could be immense.
Clinical evaluation of lentiviral vector
tools
Use of viral vectors in clinical gene therapy
has generally yielded few notable successes
aside from two trials based on providing
gene based therapies for SCID (severe
combined immune deficiency), a disease of
both cellular and humoral immunity, a form
of which may be caused by a mutation in
the IL2RG gene which encodes the 'alpha'
chain of the interleukin-2 receptor.
Research teams in France, and later in
the UK, successfully used an integrating
retroviral vector to deliver a correct
functional copy of the IL2RG gene to
autologous bone marrow cells which were
then transplanted back into patients
resulting in a dramatic recovery. Following
treatment, patients produced normal
numbers of T cells, did not need to live in
specialised environments and responded
well to childhood immunisation.This was a
significant milestone in the history of gene
therapy.
However, even in advance of the SCID
trials, there were concerns that apparent
random integration of exogenous DNA
into host cells could, in theory, activate a
proto-oncogene or deactivate a tumour
suppressor gene.This concern became a
reality when four patients from one of the
trials developed T cell leukaemia thought in
at least two patients to arise from the
vector genome inserting near to a protooncogene (LMO2) which is known to
induce T cell leukaemia formation.
To address the issue of insertional
mutagenesis, a new generation of non-
integrating lentiviral vectors has been
developed and are currently undergoing
testing.These “integration-deficient”
lentiviral vectors are expected to
substantially reduce the risk of insertional
mutagenesis in addition to reducing the risk
of inadvertent germ line transmission, for
applications involving post-mitotic or
essentially post-mitotic cells, such as those
in the eye.
Non-integrating lentivirus show
equivalent efficiencies
Dr Balaggan's group in collaboration with
colleagues from Prof Adrian Thrasher's
group at the Institute of Child Health,
London have developed and tested selfinactivating advanced vectors carrying a
single substitution mutation in one position
of the gene encoding the integrase enzyme
of HIV which catalyses the integration of
the double-stranded DNA into the host
cell chromosome.The result of this
alteration reduces the integration frequency
of HIV vector by 10,000 fold in vitro.
“The implications of
maintaining transgene
expression for 16 months
following a single injection
in itself illustrates the
attractiveness of gene
therapy because such an
approach could be used
for the long-term delivery
of therapeutic proteins”
To test the feasibility of a gene therapy
approach to choroidal neovascularisation
Dr Balaggan and colleagues have used EIAV
vectors to deliver angiostatic genes
(endostatin and angiostatin) originally
characterised for their tumour suppression
characteristics now known to have multiple
mechanisms of action including the
inhibition of VEGF, and also imparting a
positive effect on endothelial cell apoptosis.
Animal studies showed that
administration of the therapeutic genes
inhibited angiogenesis in the order of 50-60
per cent. An additional efficacy measure of
permeability was recorded by calculating
the proportional degree of increased
leakage from early phases to latter phases
resulting in a reduction of approximately 25
per cent. In the therapeutic groups the size
of the lesion is substantially smaller than
that in the control groups as is the degree
of leakage between early and late phases.
Dr Balaggan's group also evaluated the
efficacy of the newer non-integrating HIV1
vectors, in models of retinitis pigmentosa
secondary to RPE defects. Analysis of two
models of early onset retinal dystrophy, the
RPE65 mutation which accounts for about
10 per cent of LCA cases and the MERTK
deficiency modelled in the RCS rat,
substantial rescue was observed in function
in the RPE65 deficient mouse in the order
of 300-500 per cent increase in 'b' wave
amplitudes over fellow un-injected eyes.
Similarly, in the MERTK deficient rat,
preservation of function at six to eight
weeks was recorded compared with uninjected eyes where no rescue was seen
with control vectors.These technologies
and pre-clinical results provide added
confidence for the use of lentiviral vectors
in gene therapy applications for a range of
ocular diseases.
[email protected]
[email protected]
Studies have demonstrated efficient and
sustained transgene expression in vivo in
both rodent ocular and brain tissue. In
addition, Dr Balaggan and colleagues have
shown substantial rescue of clinically
relevant rodent models of retinal
degeneration with such vectors. Dr
Balaggan believes that “this shows that the
high efficiency of gene transfer and
expression mediated by lentiviruses can be
harnessed in vivo without a requirement
for vector integration”.
A major advantage of gene therapy is its
potential for localised sustained delivery of
a therapeutic molecule following a single
injection.This contrasts significantly with
anti-VEGF treatments, which may require
repeated intraocular injections, presenting a
cumulative risk of ocular complications.
49