1
Gene therapy
Gene therapy, the treatment or prevention of disease by gene transfer, is
regarded by many as a potential revolution in medicine. This is because gene therapies
are aimed at treating or eliminating the causes of disease, whereas most current drug
treat the symptoms. The treatment of human disease by gene transfer originally was
envisioned as a means to treat diseases arising from single-gene defects. Inherited
diseases encompass a wide range of disorders wherein a defective gene leads to failure
to synthesize a particular protein or to inadequately treat by conventional
pharmacological means. Therapy based on the replacement of the missing or defective
protein is available for only a few of these disorders. Furthermore, these therapies are
only partially effective in improving the manifestations of the disease and are
accompanied by significant complication. For most genetic disease, providing the
missing protein in a therapeutic fashion is not feasible due to the complex and fragile
nature of the protein and the need to deliver the protein to a specific subcellular location
(ie., cell surface expression, lysosomal localization, etc.).
Providing a normal copy of the defective gene to affect tissues would
circumvent the problem of delivering complex proteins, as the protein could be
synthesized within the cells using the normal cellular pathways. Although the defective
gene is present in all cells of an individual with an inherited disorder, only a few tissues
or organs actually express the gene and therefore are affected. Defects in genes that
function in all cells of the body (so-called housekeeping genes) usually result in such
severe abnormalities that embryonic development cannot occur. The limited number of
2
tissues affected by most inherited disorders greatly simplifies the requirements for
effective gene therapy, since a functional copy of the gene need be provided only to
those tissues that actually require it. The goal of gene therapy, therefore, is to
genetically correct the defect in only part of the body. Since this type of therapy is
designed not to alter the genetic structure of reproductive organs, it does not prevent the
genetic disorder from being passed on to subsequent generations.
The earliest human gene transfer experiments began in 1989 with
lymphocyte marking studies. While offering no therapeutic benefit, these initial studies
showed that gene transfer could be safely carried out and provided insight into many of
the technical difficulties of human gene transfer (Rosenberg et al., 1990). Lymphocytes
were likely targets for initial gene therapy attempts because they can be isolated easily
and manipulated ex vivo. Thus, tissue targeting can be effected by physical removal and
manipulation of the recipient cells, rather than by design of the gene delivery system,
which has so far proved difficult. Lymphocytes were also attractive because they are the
cellular locus of several inherited and acquired disorders (e.g., severe combined
immunodeficiency, HIV infection, graft versus host disease, and a variety of
malignancies). Furthermore, in addition to being readily isolated, lymphocytes may be
expected to be long-lived on return to the recipient and therefore can potentially provide
lasting benefits in chronic disorders. Thus, lymphocyte gene transfer provides an
important model for gene therapy and continues to be developed for many disorders. In
September 1990, the first human gene therapy trial with therapeutic potential began.
The ex vivo gene transfer of adenosine deaminase (ADA) gene into the lymphocytes of
3
a child with what is normally a lethal deficiency of this enzyme was carried out at the
National Institutes of Health (Anderson et al., 1990). In 1995,the journal Science
contains the first published summary of a landmark experiment in genetic therapy. The
report shows that two children with adenosine deaminase (ADA) deficiency who
received transplanted genes are healthy five years after beginning the treatment
.R.Michael Blaese of the National Center for Human Genome Research, a senior author
on the paper , was enthusiastic about the result, but critics point out that the girls’health
could also be attributed to the drug PEG-ADA from Enzon Inc. However, gene
therapists hope to apply similar strategies to diseases such as diabetes, cancer and AIDS
(Washington Post.,1995).
The majority of gene therapy trials under way are for the treatment of acquired
disorders such as AIDS, malignancies, and cardiovascular disease, rather than diseases
arising from single gene defects. The application of gene therapy to acquired disorders
has proceeded faster than applications for single-gene defects for several reasons.
Principle among these reasons is that the long-term gene expression (months to years)
that is likely necessary to treat genetic diseases has been difficult to achieve. The
availability of a large pool of candidate patients with severe and immediately lifethreatening acquired disorders (most notably cancer and AIDS) provides a clinical
setting to develop new strategies for DNA delivery that may be applied later to inherited
disorders. In contrast to the inherited diseases where a genetic defect has been well
characterized, in most applications of gene therapy to acquired illnesses, the molecular
basis of the disease is less well understood. Rather than correct a known underlying
4
defect, the approach has been to add new molecular functions that are capable of
altering the course of the disease, or to block an existing function, rather than correct an
underlying deficiency.
Gene therapy in AIDS
Know your enemy: the HIV life cycle and points of intervention
HIV is a member of the lentiviral class of retroviruses and, for obvious reasons,
is one of the best-studied viruses. A brief overview of its life cycle follows ( Fig. 1).
The infectious cycle begins with the binding of virus particles to target-cell membranes
via the virus envelope glycoprotein gp120/gp41. Initially, the virus binds to the CD4
protein, followed by attachment to an obligatory coreceptor, which has been identified
recently as one or more members of the chemokine receptor gene family. The main
target cells for HIV infection are T helper cells and macrophages. The viral core
complex then penetrates the cell where a multistep process of reverse transcription,
nuclear entry and integration into the host cell chromosome occurs. By integration into
the chromosome of the target cell, HIV now becomes a permanent part of that cell's
genome and can be considered an acquired genetic disease. In the CD4 + T cell, HIV
does not actively replicate unless the cell becomes activated and host cell transcription
factors help to jump-start the HIV promoter [the long terminal repeat (LTR)] to make a
series of early regulatory mRNAs that code for the Tat, Rev and Nef proteins.
5
Figure 1. A simplified model of the human immunodeficiency virus (HIV) life cycle and potential targets
for gene therapy. (a) Binding: HIV binds to CD4 and a coreceptor on the target cell. Soluble CD4 might
block this process. After (b) internalization and (c) uncoating, the viral genome, which is composed of
single-stranded RNA, is (d) reverse transcribed into DNA, a process that is catalyzed by reverse
transcriptase (RT). This step might be blocked by ribozymes that cleave the genomic RNA or by
intrabodies against RT. The genomic DNA is imported into the nucleus, becomes integrated into the
host-cell genome, and is transcribed and translated by the host-cell machinery. This process is
regulated by Tat and Rev, which feed back on the integrated HIV genome to (f) activate HIV RNA
synthesis and transport. Tat and Rev are therefore major targets for anti-HIV gene therapy, and might
be blocked by ribozymes, antisense RNA decoys and trans-dominant Rev. (g) The assembly process
6
itself might also be blocked using trans-dominant proteins or intrabodies. Finally, (h) the mature virus
buds from the cell surface. The infectivity of the mature particles might be blocked using Trojan horses
(Richard.,1999)
The Tat protein is the key player in the early stages of HIV
RNA synthesis, with more Tat leading to increasing amounts of HIV RNA. As the Rev
protein accumulates in the cells, this second key regulator causes a shift in the type of
HIV mRNAs produced to longer transcripts that code for accessory proteins and the
structural proteins that are necessary for new virus particle assembly 7. The HIV genes
encoding the Tat and Rev regulatory proteins are very attractive targets for anti-HIV
gene therapy. These two essential regulatory proteins mediate trans-activation of HIV
gene expression by binding on to the HIV mRNA. Tat binds to a region called TAR
(trans-activation-response element) near the 5'-end of the HIV RNA and Rev interacts
with sequences in the envelope gene termed RRE (Rev-responsive element). The
interaction between Tat and TAR can lead to a potent trans-activation of viral gene
expression by inducing transcriptional initiation and/or elongation. The Rev–RRE
interaction also strongly trans-activates HIV by facilitating the extranuclear transport of
unspliced or singly spliced mRNA molecules, which encode HIV structural proteins and
the accessory proteins Vpr, Vpu and Vif. Viral structural proteins associate near the
plasma membrane and these assembly complexes result in the budding of new virus
from the cell and a new infection cycle can begin.
7
Most studies on genetic inhibition of HIV have focused on nucleic acid-based
approaches. The main advantage of nucleic acid-based anti-HIV gene therapies is that
there is little chance of an immune response to the nucleic acid (in contrast to proteinbased approaches). Nucleic acid-based anti-HIV genetic inhibition strategies can be
divided into Ribozyme, Transdominant mutant, and for lack of a better qualifier,
“Trojan horses”.
Ribozyme
Ribozymes (Rzs) are small catalytic RNA molecules that possess sequencespecific RNA cleavage activity. They occur naturally, but can also be artificially created
to target specific sequences in cis or trans. Rzs have been successfully used to target
and destroy both viral and cellular RNAs, in cell culture systems and in animals.
Investigators in the HIV field were perhaps the earliest and most avid group to adopt
Rzs as gene therapeutics. Several features of Rzs make them attractive for HIV gene
therapy: simplicity of design; lack of immunogenicity of RNA; and the possibility of
designing multiple Rzs against conserved regions of the viral genome to overcome viral
resistance. Intracellular expression of anti-HIV Rzs in T cells or monocyte/macrophage
progeny derived from vector-transduced precursor cells have been shown to inhibit HIV
replication .
Successful use of Rzs to knockdown target gene expression is dependent on a
number of factors, including target site selection as well as Rz gene delivery,
expression, stability and intracellular localization. The two types of Rz that have been
8
used most extensively are hairpin and hammerhead Rzs. Each has its own minimal
target sequence requirements, the hairpin Rz requiring a GUC at the site of cleavage,
whereas the hammerhead Rz requires NUH (where N denotes any base and H denotes
A, C or U) (for review see. Not all target sites, however, are accessible for cleavage;
secondary structures, binding of proteins and nucleic acids, and additional esoteric
factors will influence in vivo Rz activity. Rzs can be introduced into cells as genes by
transfection or viral vector transduction, or as 'drugs' by using chemically synthesized
Rzs. Synthetic Rzs can be further stabilized with various base substitutions and 3' and 5'
modifications. Intracellular delivery, in this case, is typically achieved by lipid-mediated
transfection.
Depending on the application, either delivery method has its advantages and
disadvantages. Delivery of the Rz genes can provide stable intracellular expression and
promoter choice can allow cell or tissue-specific expression; however, dosing studies
and actual application in animals can be problematic. Delivery of synthetic Rzs allows
short term, high level availability, making it easier to use as a 'drug'; however, Rz
stability and pharmacokinetics may present significant challenges (Welch et al.,1998).
The ribozyme technology has emerged as a potentially powerful extension of the
antisense approach to gene inactivation. Intracellular expression of hammerhead
ribozymes and a hairpin ribozyme directed against HIV-1 RNA has been shown to
confer significant resistance to HIV-1 infection.
We investigated if a hairpin ribozyme targeted against HIV-1 would be effective
in virus inhibition. The first HIV-1 ribozyme we analyzed was engineered to cleave the
9
5'-leader sequence of HIV-1/HXB2 clone RNA at positions +111/112 from the cap
site. We observed that this ribozyme suppressed virus expression in Hela cells cotransfected with proviral DNA from diverse HIV-1 strains. Moreover, the antiviral
effect was primarily due to the catalytic rather than antisense property of the ribozyme.
We then "immunized" human CD4+ T cell lines (Jurkat and Molt 4/8) intracellularly
with the ribozyme gene delivered in murine retroviral vectors driven either by an
internal human tRNAval (pol III) promoter or directly by the retroviral LTR. There was
no apparent deleterious effect of ribozyme expression on cell proliferation or long-term
viability. Higher levels of ribozyme expression were consistently obtained with the pol
III promoter. Cells expressing ribozyme were resistant to challenge from diverse strains
of HIV-1, including an uncloned clinical isolate. No reverse transcriptase activity or
virus infectivity was detectable in the culture supernatants of Jurkat cells expressing the
ribozyme driven by tRNAval promoter up to 35 days after challenge with HIV-1/HXB2. Transduction of primary lymphocytes with the ribozyme vector also completely
blocked infection by HIV-1. In addition to inhibiting virus expression from integrated
proviral DNA, expression of the ribozyme also significantly decreased (by
approximately 100-fold) the efficiency of incoming virus to synthesize viral DNA.
These results indicate that transfer and expression of the ribozyme gene interfered with
both early and late events in the HIV-1 replication cycle and conferred long-term
resistance to HIV-1 infection. The extremely encouraging results obtained so far with
this ribozyme led us to expand our efforts to seriously develop ribozyme gene therapy
against HIV-1 infection. A phase I clinical protocol to test the safety and function of the
10
ribozyme in transduced human peripheral blood lymphocytes in vivo has been
developed and approved by the NIH Recombinant DNA Advisory Committee.
Although the loss of CD4+ cells is the major proximate cause of the immune
deficiency of AIDS, other cells such as macrophages, dendritic cells, and brain
microglial cells are also infected. Moreover, natural senescence of the transduced cells
would periodically require repeated, large-scale transfers of gene-altered T cells, a
process that is too logistically complex, expensive, and dependent on specialized
expertise to be practical. Although the initial efforts to examine transduction of
peripheral blood T cells will yield invaluable information for the feasibility of particular
gene therapies for HIV-1 or even confer limited therapeutic benefits on the patients,
gene transfer into pluripotent hematopoietic stem (progenitor) cells would be a preferred
strategy to achieving sustained immune reconstitution. Stem cells possess two important
characteristics for this purpose: the ability to give rise to all hematopoietic lineages and
the capacity for self-renewal. Stable transduction of stem cells might thus allow the
permanent repopulation of all hematopoietic cell lineages of the immune system with
intracellularly immunized cells.
We have used retroviruses carrying the ribozyme driven by two different pol III
promoters to transduce hematopoietic stem cells from different sources, including adult
bone marrow, fetal cord blood, and mobilized peripheral blood. High transduction
efficiency was obtained with immunoaffinity-enriched, cord blood-derived CD34 cells
prestimulated with a variety of cytokines and growth factors. Ribozyme expression in
the CFU as detected by RT-PCR assays was also highly efficient. Moreover, the
11
ribozyme was persistently expressed over fifty days (the experiment period).
Transduction and ribozyme expression had no apparent deleterious effect on phenotype
or proliferation, as determined by clonogenic assays and growth curve assessments. The
cultured stem cells differentiated into macrophages/monocytes that could be infected by
several HIV-1 strains. Challenge with a monocyte-tropic strain of HIV-1 showed that
transduction of their progenitor cells with the ribozyme gene conferred relative
resistance to HIV-1 infection. The inhibition of virus expression is supported by the
expression of the ribozyme gene in these cells. These results suggest the feasibility of
stem/progenitor cell gene therapy for HIV-1-infected patients(Wong-Staal.,1994).
Transdominant mutants
Site-directed mutagenesis approaches have identified specific amino acid
sequences of both the Tat and Rev proteins that, when altered, turn the mutant protein
into a dominant negative (or trans-dominant) protein. When these trans-dominant
proteins are synthesized in HIV-infected cells, the function of the wild-type HIV
proteins is efficiently blocked by the action of one defective monomer, disrupting the
functioning of the multimeric protein complex. The structural proteins Gag and Env are
also potential targets for the generation of trans-dominant proteins. These transdominant proteins interfere with their wild-type products at multiple stages of the viral
life cycle, from Gag polyprotein processing and core assembly, to interaction of core
proteins with the envelope proteins, to budding of infectious particles from the infected
cell. Some trans-dominant Gag proteins prevent the liberation of virions altogether,
12
whereas others lead to the release of defective particles that cannot infect new target
cells.
Tat and Rev both function by binding to specific sequences of HIV RNA (TAR
and RRE, respectively). This nucleic acid-binding property affords the potential that
decoy sequences could be synthesized in the cell to draw Tat and Rev away from their
natural targets and potentially decrease the effectiveness of these key proteins
14–16
.
Molecular decoys to both Tat and Rev are very effective in inhibiting HIV replication in
vitro and are currently in clinical trials.
We have previously reported that a mutant Rev protein (dRev) lacking its
nucleolar targeting signal remained out of nucleic in expressed cells and strongly
inhibited the function of Rev. To investigate the effect of dRev on HIV-1 replication,
we established several dRev-expressing human cell lines with two different vector
systems and examined virus production in these cells. An HIV-1-derived vector
containing drev cDNA was constructed and introduced into CD4-positive HeLa cells
and cells of the human T-cell line CCRE-CEM (CEM). In dRev-expressing HeLa cells,
virus replication, syncytium formation, and cell death caused by HIV-1 infection were
remarkably suppressed, and the same vector also conferred a resistant phenotype on
CEM cells. The production was also suppressed in CEM cells containing drev gene
driven by a cytomegalovirus promotor. In addition, we found that dRev gene did not
cause nucleolar dysfunction in a transient assay, in contrast to other transdominant
mutant and wild type Rev. Since dRev cannot migrate into nuclei, it is expected not to
interfere with nuclear/nucleolar function of the host cell. We conclude that drev is one
13
promising candidate as an antiviral molecule for gene therapy against AIDS (Furuta et
al.,1995).
It was previously shown that a tat mutant (tat22) where cysteine 22 is substituted
by glycine behaves as a transdominant negative mutant in Jurkat T cells lytically or
latently infected by HIV-1. In this study we demonstrate that tat22 controls HIV-1
replication in primary cells. This effect was observed both after in vitro infection of
peripheral blood mononuclear cells (PBMCs) from normal donors and after reactivation
of the latent infection in PBMCs from seropositive patients. The antiviral effect of tat22
was limited to conditions of low virus production. The use of tat22 may be promising
for a gene therapy approach to AIDS during the asymptomatic phase of the disease
allowing control of virus replication in infected cells and inhibition of virus spread to
uninfected cells (Rossi et al.,1997).
Trojan horse
The Trojan horse anti-HIV strategy brings together a variety of experimental
strategies that have a common goal: the co-packaging of the anti-HIV agent into the
newly formed HIV particle. During the final stages of HIV replication, HIV genomic
RNA is specifically incorporated into HIV particles based on recognition of an RNA
packaging signal. Thus, various investigators have linked HIV packaging signals (or
tRNA primers) to anti-HIV RNA, in the hope that these agents will associate with HIV
RNAs during assembly. When the particle is released, not only it will be biologically
14
attenuated by this molecular hitchhiker, but the anti-HIV agent will be delivered to
another cell.
Gene delivery
The ultimate goal in the management of inherited as well as acquired diseases is
a rational therapy with the aim to eliminate the underlying biochemical defects, rather
than a symptomatic treatment. Among other approaches somatic gene therapy is a
promising candidate to meet these objectives and appears to have the potential to
revolutionize modern medicine. To allow efficient transfer of the therapeutic genes, a
variety of gene delivery techniques have been developed based on viral and non-viral
vector systems. For the success of this technology it is vital to achieve regulated and
sustained expression of foreign genes in specific target tissues. This will be crucial for
the widespread application of somatic gene therapy. So far none for the gene delivery
systems is able to meet the requirements of safety, efficiency and specificity
demonstrating that vector research will be an important focus in the development of
optimized transfer methods.
To allow efficient transfer of the therapeutic genes, a variety of gene delivery
techniques have been developed based on viral and non-viral vector systems. For the
success of this technology it is vital to achieve regulated and sustained expression of
foreign genes in specific target tissues. This will be crucial for the widespread
application of somatic gene therapy. So far none for the gene delivery systems is able to
15
meet the requirements of safety, efficiency and specificity demonstrating that vector
research will be an important focus in the development of optimized transfer methods.
The ideal DNA delivery system would be one that could accommodate a broad
size range of inserted DNA, was available in a concentrated form, was easily produced,
could be targeted to specific types of cells, would not permit replication of the DNA,
could provide long-term gene expression, and was nontoxic and nonimmunogenic. Such
a DNA delivery system does not exist, and none of the available technologies for in vivo
gene transfer is perfect with respect to any one of these points. As of 1995, three gene
transfer systems (retroviral vectors, adenoviral vectors, and liposomes) had been used in
human gene therapy trials, with a total clinical experience of a few hundred patients
worldwide. Consequently, the following discussion will highlight conceptual strategies
and issues to be refined, rather than clinical experience.
T cells and macrophages are the main target cells for HIV infection, and both of
these cell types are derived from hemopoietic stem cells (HSCs). Although HSCs would
appear to be the most logical cells to target in HIV gene therapy, there are both
technical and theoretical reasons that might make this difficult. First, several clinical
trials have revealed that the current gene transfer vectors of choice, murine leukemia
virus (MLV)-based retroviral vectors, result in stable gene transfer into only a few
percent of mature progeny cells, even when cytoablation is used to make room for the
transferred cells (Chu et al.,1998). Second, successful engraftment of a small percentage
of gene-modified cells in adult patients is not likely to contribute greatly to the build-up
of sufficient numbers of protected mature lymphocytes. It is usually argued that these
16
cells could have a survival advantage in an HIV-infected host and would accumulate
over time, but there would still be a majority of unprotected cells that would serve as a
constant source of infectable cells. Another potential advantage of targeting genes to
HSCs is that, with the current MLV-based vectors, transducing HSCs is the only route
of introducing genes into the monocyte/macrophage lineage.
Mature T cells make up the bulk of HIV-infected cells in patients and there is
reasonably good evidence that these mature T cells are a very long-lived cell population
that largely maintains its numbers by division of mature cells and not by development
from primitive cells, at least in adults (Walker et.al .,1998). Again, although in vitro
gene-transfer efficiencies in T cells are significantly better than are those seen in HSCs,
engineering large numbers of cells is still technically very demanding. It might well be
that targeting both cell types will be the most effective strategy.
Prospects for gene therapy
The small number of published reports, along with meeting abstracts and
presentations, strongly suggest that Baltimore's proposal for intracellular immunity can
indeed prolong the survival of anti-HIV- engineered T cells. New anti-HIV approaches,
which continue to be published regularly, can provide significant in vitro protection to
engineered cells. There is an enormous gap between the laboratory investigation and
development of clinical anti-HIV gene therapy protocols. The current viral vector-based
gene-transfer methods are costly and would be difficult to apply on a large scale. If
genetic therapy is to become a practical option for HIV-infected individuals, then the
17
long-awaited paradigm shift from in vitro to in vivo gene transfer will have to occur.
Of the available gene-transfer vectors, only the lentiviral vectors appear to have this
potential (Naldini et al.,1996). The notion of using HIV-based vectors to fight HIV
seems appealing, but in reality, regulatory and technical concerns might outweigh the
potential utility of this system. These conceptual concerns must also face the reality that
for many infected individuals, the current highly active anti-viral therapy (HAART)based therapies are associated with significant clinical improvement, so why risk the
unknown of gene therapy? If the experience of HAART has taught us anything, it is that
combination anti-HIV therapies are superior to monotherapy. It is therefore logical to
consider that addition of genetic inhibition strategies to existing 'standard' anti-viral
therapy might be of clinical benefit to patients. It is conceivable that HAART could be
used to decrease viral load while genetic therapies could then be targeted at the latent
population of HIV-infected cells.
If the current handful of clinical trials is found to be associated with therapeutic
benefit and minimal risk, patients might feel more comfortable about entering clinical
gene therapy trials. Only time will tell, but it should be an exciting few years ahead.