| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
CONCISE REVIEW |
a University of Pittsburgh School of Medicine, Department of Molecular Genetics and Biochemistry, Pittsburgh, Pennsylvania;
b Department of Chemical Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania;
c Department of Neurology and GRECC, Pittsburgh VA Healthcare System, Pittsburgh, Pennsylvania, USA
Key Words. Herpes simplex virus • Gene therapy • Nervous system • Glioma • Neuropathy • Pain • Arthritis • Muscular dystrophy
Joseph C. Glorioso, Ph.D., University of Pittsburgh School of Medicine, Department of Molecular Genetics and Biochemistry, E1240 Biomedical Sciences Tower, Pittsburgh, Pennsylvania 15261, USA. Telephone: 412-648-8106; Fax: 412-624-8997; e-mail: glorioso{at}pitt.edu
 |
ABSTRACT |
|---|
 Top Abstract IntroductionBasic Biology of HSV-1Using HSV-1 to Make...Applications of HSV VectorsConclusionsReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionBasic Biology of HSV-1Using HSV-1 to Make...Applications of HSV VectorsConclusionsReferences |
|---|
Delivery of transgenes to cells may be achieved by various means. The most potent gene delivery reagents occur naturally. Viruses function to proficiently transport their genetic material to the interior of host cells. Following this, advantage is taken of cellular metabolic machinery to effect viral replication, usually to the detriment of the host. Harnessing the ability of viruses to enter cells and deliver their genetic payload, while deleting viral functions responsible for pathogenic effects, has allowed the construction of several series of gene transfer vectors based on different virus families. Viruses have evolved specific features to enable their successful propagation in vivo, using a variety of strategies such as integration, latency and immune evasion to optimize interaction with the host organism. Teleologically, it might be expected that exploitation of viral functions that are preserved in gene therapy vectors and are consistent with efficient gene delivery in the absence of host metabolic derangement would enhance the utility of viral vectors.
Our laboratory has developed a vector system based on disabled herpes simplex virus (HSV). Replication-defective HSV vectors are nonpathogenic but retain many advantageous features of the wild-type virus, which may be exploited to enhance the delivery and expression of therapeutic and experimental transgenes. The vectors have found numerous applications. In this review, we illustrate ways in which different aspects of the complexities of HSV biology have been exploited for specific gene delivery applications.
|
BASIC BIOLOGY OF HSV-1 |
|---|
|
TopAbstractIntroductionBasic Biology of HSV-1Using HSV-1 to Make...Applications of HSV VectorsConclusionsReferences |
|---|
; [2]). The mature virion consists of the following components:
|
Virtually all of the proteins and glycoproteins of the mature virion are encoded by viral genes. The HSV genome consists of 152 kb of dsDNA, arranged as long and short unique segments (UL and US) flanked by repeated sequences [14-17]. Eighty-four viral genes are encoded, and these may be divided into essential and non-essential genes, according to whether their expression is necessary for viral replication in a permissive tissue culture environment (Fig. 1B
). Nonessential genes often encode functions that are important for specific virus-host interactions in vivo, for example immune evasion, replication in non-dividing cells or shutdown of host protein synthesis; these genes may be deleted in the generation of gene therapy vectors, allowing the insertion of exogenous genetic material [18-19].
During lytic infection, viral genes are expressed in a tightly regulated, interdependent temporal sequence [2, 20, 21], reviewed in [2] (Fig. 3). Transcription of the five immediate-early (IE) genes, ICP0, ICP4, ICP22, ICP27, and ICP47 commences on viral DNA entry to the nucleus. Expression of these genes is regulated by promoters that are responsive to a viral structural protein, VP16, which is transported to the host cell nucleus with viral DNA. VP16 binds to cognate motifs within the IE promoter sequences and associates with cellular transcription factors to promote transcriptional activation of the IE genes [7, 22, 23]. Expression of IE genes initiates a cascade of viral gene expression (Fig. 2). Transcription of early (E) genes, which primarily encode enzymes involved in DNA replication, is followed by expression of late (L) genes mainly encoding structural components of the virion [2, 20, 21]. Of the IE gene products, only ICP4 and ICP27 are essential for expression of E and L genes, and hence viral replication [24-26].
|
|
. Following primary cutaneous or mucosal inoculation, the virus undergoes lytic replication in the infected epithelia. Viral particles are released at the site of the primary lesion; they may enter sensory neurons whose axon terminals innervate the affected area. The nucleocapsid and tegument are carried by retrograde axonal transport from the site of entry to the neuronal soma in the dorsal root ganglia or trigeminal ganglia, where the viral genome and VP16 enter the nucleus [27-29]. At this point, one of two chains of events may ensue. First, the lytic replicative cycle described above may take place. This pathway results in neuronal cell death and egress of infectious particles. Alternatively, the viral DNA can enter the latent state. During latency, the viral genome persists as a stable episomal element, sometimes for the lifetime of the host [30]. The DNA adopts a chromatin-like structure; it is probably not extensively methylated [31, 32]. No IE, E, or lytic L genes are expressed during latency, but a set of nontranslated RNA species, the latency-associated transcripts (LATs), is produced and detectable in the nuclei of latently infected neurons [33-37, and below]. At a time point that may be remote from the establishment of latency, alterations in the host-virus interaction may cause "reactivation" of the viral infection. IE genes are expressed and the lytic cascade of gene expression follows, resulting in the production of mature virions. The nucleocapsid and glycoproteins are transported by separate anterograde axonal transport pathways to the peripheral nerve terminals, where they are assembled and released [38-39]. The processes regulating the establishment of and reactivation from latency are not well understood. The LATs are a hallmark of HSV latency; the major 2.0 kb and 1.5 kb species are abundant, stable lariat introns that arise by splicing of a primary transcript [40-43]. The functions of the LATs remain unknown, although several putative roles have been suggested. These include: efficient establishment of latency [44, 45]; effective reactivation from latency [46-52]; antisense regulation of IE gene transcripts [53-55]; prevention of apoptosis in infected neurons [56]; expression of proteins [57] that may compensate for the absence of IE gene expression during latency [58], and functions relating to RNA-mediated catalysis [59]. However, it is clear that the LAT genes are not an absolute requirement for establishment, maintenance, or reactivation from latency [60-63]. This has important implications for vector construction, as it is possible to insert transgenes within the LAT loci, disrupting the LAT genes and using the LAT cis-acting regulatory sequences to drive transgene expression (see below).
|
USING HSV-1 TO MAKE GENE THERAPY VECTORS |
|---|
|
TopAbstractIntroductionBasic Biology of HSV-1Using HSV-1 to Make...Applications of HSV VectorsConclusionsReferences |
|---|
Wild-type HSV infection is toxic and invariably results in lysis of many cell types. Blocking viral replication after cellular entry arrests the HSV life cycle and prevents lytic infection. As E and L gene expression, and therefore replication, is fully dependent upon the expression of IE genes, generation of replication-incompetent vectors can be accomplished by disruption of one or another essential IE gene, ICP4 or ICP27 [24] (Fig. 3). An ICP4 null mutant, for example, is unable to replicate in noncomplementing cells in culture [24]. Unfortunately, this single manipulation is not adequate to completely prevent cytotoxicity, as ICP22, ICP27, and ICP0 are toxic to host cells in many situations [66-68]. IE genes are negatively regulated by ICP4, such that infection with an ICP4 null mutant results in their overexpression, resulting in extensive cell death in the absence of viral replication [18, 24, 69]. To prevent cytotoxicity, a series of vectors has been generated that are multiply deleted for IE genes [18, 68, 70] (Fig. 4). Characterization of these vectors in vitro shows that deletion of the genes encoding ICP4, 22, and 27 improves the vector cytotoxicity profile when compared with the ICP4 null parent vector [18]. In addition, toxicity associated with the triple (ICP4–:ICP22–:ICP27–) mutant is less than that of either corresponding double (ICP4–:ICP22– or ICP4–:ICP27–) mutant [18]. Quintuple IE mutants (ICP0–:ICP4–:ICP22–:ICP27–:ICP47–) have been produced, and are entirely nontoxic to cells. The genomes of these vectors are able to persist for long periods in cells [68]. However, vectors grow poorly in culture and express genes at very low levels in the absence of ICP0 [70, 71]. Retention of the trans-activator ICP0 allows efficient expression of viral genes and transgenes, and allows the virus to be prepared to high titer. Recent work has shown that the post-translational processing of ICP0 in neurons is different from that in glia [72]. It appears that, although ICP0 mRNA is efficiently expressed in both cell types, ICP0 undergoes proteolytic degradation in neurons. It might be predicted that a vector carrying an intact ICP0 gene would not be toxic to neurons and may even be advantageous for oncological applications, where ICP0-induced cell cycle arrest [73] may be desirable. The issue of whether ICP0 expression is deleterious in other situations is currently under evaluation. Deletion of ICP47 restores the antigenic loading of MHC class I molecules at the surface of infected cells [74-76]. This may potentially confer advantages in the gene therapy of malignancy, although the utility of this modification is unclear at present. For most other applications, where immune evasion and high-level transgene expression are desirable, triple IE mutants (ICP4–: ICP22–:ICP27–) have been used. These vectors show minimal cytotoxicity in vitro and in vivo, are efficient vehicles for transgene delivery, and can be grown efficiently in cells that complement the absence of ICP4 and ICP27 in trans [18, 67].
|
|
APPLICATIONS OF HSV VECTORS |
|---|
|
TopAbstractIntroductionBasic Biology of HSV-1Using HSV-1 to Make...Applications of HSV VectorsConclusionsReferences |
|---|
).
|
|
|
). Finally, sequences contained within LAP2 seem able to direct high-level and sustained latent-phase transcription from some heterologous promoter elements ([88, 91] and Goins, unpublished observations). As the LATs are not an absolute requirement for the establishment of latency, it has been possible to insert transgenes into the LAT loci to utilize all of the relevant cis-acting sequences [92, 93]. Studies aimed at identifying the best ways in which to use the latency promoter system for neurological gene therapy are ongoing.
PNS (Figs. 5, 6)
Of all the potential target organs of HSV vector-mediated gene therapy, the PNS seems likely to yield the most successful results. The latent life cycle of the wild-type virus occurs in the sensory ganglia of peripheral nerves. Natural viral functions are, therefore, already optimized for the delivery and expression of genetic material in this tissue; deleting the capacity of the virus to reactivate and cause pathological disturbances would be expected to yield a safe and efficient system for peripheral nerve gene delivery.
In vitro, a triple IE mutant virus, TOZ.1, was able to efficiently infect dorsal root ganglion cultures, and to drive sustained transgene expression [18]. There was little or no evidence of vector toxicity in these neurons, although supporting cells present in the cultures showed a more robust cytopathic response to infection with clear evidence of cell death and metabolic disturbance [18]. This may be at least partially attributable to the difference in post-translational processing of ICP0 between neurons and glia [72]. Other groups have examined the electrophysiological responses of primary sensory neurons in culture to infection with HSV vectors [94]. In contrast to wild-type virus, which abolishes transmembrane Na+ currents following infection, transduction with a replication-defective vector has no effect on the electrophysiological profile of these cells. In vivo, peripheral inoculation allows replication-defective vector to exploit the same retrograde axonal transport system used by wild-type virus to reach the sensory ganglia [64, 78, 91-93]. In contrast to the in vitro setting, glial toxicity is not seen, as the virus does not replicate within the ganglia and is unable to enter Schwann cells. Peripheral inoculation with replication-defective vectors apparently effects less efficient nerve transduction than with replication-competent virus [91]. This is probably a manifestation of the initial round of lytic replication in epithelia, which amplifies the virus dose delivered to nerves. Indeed, with an elevated dose of replication-deficient vector, it is possible to demonstrate efficient nerve transduction via peripheral cutaneous infection [78, 91]. In addition, motor neurons may be targeted by viral entry through their peripheral termini in muscle [92, 95, 96].
Long-term expression of transgenes has proved feasible in the peripheral nervous system. Indeed, we initially identified the LAP2 promoter by its ability to drive virus-delivered reporter gene activity in latent infection of trigeminal ganglia in vivo for up to 300 days [78]. A subsequent study demonstrated the viability of using the LAP2 promoter to drive therapeutic transgene expression in vivo [64]. A nerve growth factor (NGF) transgene was expressed in the context of an attenuated vector, KLN (Fig. 4), using the LAP2 promoter to drive transcription. Significant quantities of the transgene product were present within latently infected trigeminal ganglia at 28 days post-infection [64]. Others have corroborated these data, showing that insertion of reporter genes into the LAT transcript allows stable long-term expression in dorsal root ganglia and motor neurons [91]. In addition, LAT promoter-driven reporter gene activity was demonstrable in the brain stem motor nuclei of mice 10 months after acute ear infection with a replication-competent virus [92]. These studies provide hope that the LAT promoter system may be useful in motor as well as sensory neurons.
We have investigated several applications for HSV vectors in the peripheral nervous system (Figs. 5, 6). The first of these is the delivery of trophic factors to support neurons in models of neuropathic disease. In an initial series of experiments, vectors producing nerve growth factor were generated (KLN, SLN; Fig. 4) [64]. Biologically active NGF, capable of inducing differentiation of PC12 cells, was produced in cultured neurons, stably and in significant quantities, for several weeks. The NGF was sufficient to protect primary dorsal root ganglion neuron cultures from an oxidative insult with H2O2, probably by induction of antioxidative genes (Fig. 6). We have now extended these studies to the treatment of peripheral nerve disease in vivo, using animal models of toxic sensory neuropathy. It appears that prior expression of NGF by sensory nerves is able to protect them from a toxic insult in vivo, with clear implications for prophylactic therapy of chemotherapy-induced peripheral neuropathy (Goss, Goins, Lacomis et al., unpublished observations). In addition, we have addressed protection of the autonomic nervous system from the chronic metabolic stress of diabetes mellitus; the effects of chronic hyperglycemia are ameliorated by HSV-mediated NGF expression in a model of diabetic cystopathy [97]. For the therapy of polyneuropathy arising from systemic disease, it may be more appropriate to deliver trophic neuronal support into the circulation; a subsequent section addresses the issue of circulating protein delivery (Fig. 8).
|
). Opioid receptors that are activated by enkephalin are located on the presynaptic terminals of primary afferent fibers, and the post-synaptic terminals of second order neurons, which mediate pain transmission in the dorsal horn of the spinal cord [100, 101]. Activation of these receptors leads to inhibition of neurotransmitter release from the primary afferents, and inhibits activation of second-order neurons within the CNS both directly and indirectly [102, 103]. In initial studies, the footpads of mice were inoculated with a conditionally replicating virus expressing pre-proenkephalin [98] (KHPE; Fig. 4). Ectopic expression of the human form of the neurotransmitter within dorsal root ganglia was demonstrated following infection. Sensitization of unmyelinated or small myelinated nerve fibers with pharmacological agents was abolished by enkephalin expression within the nerve fibers, but the transmission of basal nociceptive responses was unaffected. The antihyperalgesic action of the enkephalin transgene was abolished by naloxone, an antagonist at enkephalin receptors. These findings indicate that the pre-proenkephalin transgene product is processed, packaged and released in a similar manner to the native protein, with which it shares identical receptors [98]. A replication-defective vector (SHPE; Fig. 4) is currently under evaluation in a formalin model of chronic pain in rats [104] (Fig. 6) and a nerve-ligation mechanical allodynia model of neuropathic pain in rats (Hao et al., unpublished observations). The results raise the exciting possibility of novel approaches to the treatment of chronic pain by intrasynaptic delivery of analgesic molecules that are associated with significant side effects when delivered systemically.
CNS In contrast to viral latency observed in the PNS, infection of the CNS with wild-type HSV-1 results in a rapidly fatal hemorrhagic encephalitis, which is fully dependent upon viral replication; replication-defective HSV vectors do not cause this dramatic effect. However, eliminating expression of multiple IE genes appears crucial in minimizing CNS neuron toxicity. Thus, a single IE mutant (ICP4–) virus was toxic to cultured cortical neurons [105], which showed minimal evidence of toxicity or metabolic disturbance when infected with a triple IE mutant vector (ICP4–:ICP22–:ICP27–) [18]. The same appears true in the brain in vivo. Thus, use of a single IE mutant (ICP27–) gave rise to cell death and an inflammatory response following intraparenchymal injection [94], whereas a triple IE mutant [18] caused a small degree of tissue damage that was similar to that seen with saline injection, and was presumably partially mechanical in origin [106]. Importantly, there is no evidence that direct introduction of disabled HSV into the cerebral parenchyma can effect reactivation of latent wild-type virus [107].
Following direct inoculation of the CNS parenchyma with replication-defective HSV-1, local transduction occurs only for a few millimeters around the needle track [94, 106, 108]. This is not enhanced significantly by increasing injection volume. The axonal transport of HSV in transduced neurons gives rise to transduction of neurons at locations remote from the injection site, for example within the substantia nigra following striatal injection [109]. Introduction of the virus into the cerebrospinal fluid by cisternal puncture [110-111] allows transduction of the pia and arachnoid mater overlying the brain; meningeal cells are transduced over a wide area after a single injection, but there is little expression in the underlying neuropil. Current delivery techniques are thus best suited to the introduction of transgenes within well-circumscribed anatomical areas, or systems with defined connectivity within the CNS, or using the meninges as a depot for therapeutic protein secretion into the cerebrospinal fluid (CSF).
Viral DNA persists long-term following intracerebral inoculation with a replication-defective HSV vector [108]. Use of viral promoters other than the latency promoters gives rise to short-term transgene expression in CNS neurons, as might be expected [94, 106, 108, 112]. Long-term expression using the latency promoter system has been demonstrated in the context of a replication-competent attenuated vector [113], and it is known that, following acute infection, replication-competent and neuro-attenuated vectors persist in CNS neurons where they transcribe the LAT genes [113-115]. Stable CNS gene expression using the latency promoters has not yet been reported from a replication-defective vector. Recently, we have shown that tightly regulated expression from a replication-defective virus can be achieved in the CNS short-term using an inducible promoter system [116].
Using replication-defective vectors, we have established the principle that neuroprotection may be effected in the CNS by transient HSV-mediated expression of appropriate transgenes [112]. The substantia nigra pars compacta (SNc) of the mesencephalon is the primary site of neuronal loss in Parkinson's disease. One animal model of SNc cell loss involves injecting the ipsilateral midbrain of rats with a neurotoxin, 6-hydroxydopamine (6-OHDA). This causes extensive apoptotic cell death among the SNc neuron pool. bcl-2 is an antiapoptotic protein that prevents opening of the permeability transition pore complex and subsequent release of cytochrome-C from mitochondria during initiation of the mitochondria-dependent programmed cell death cascade. Rats pretreated with a replication-defective HSV vector directing expression of bcl-2 (THZ/bcl; Fig. 4) in the substantia nigra showed much less SNc cell loss in response to 6-OHDA than controls. Markers of cell survival, including phenotypic and behavioral markers for SNc dysfunction were much improved in this paradigm by prior treatment with the bcl-2 expressing vector. Although the relevance of the rat 6-OHDA model to the pathogenesis of Parkinson's disease is unclear, this study shows that replication-defective HSV vectors may effect phenotypic improvements in pathological CNS processes by delivery of therapeutic genes.
Recently, the use of the meninges as a depot site for synthesis and secretion of anti-inflammatory cytokines into the CSF has been reported [110, 111]. Replication-defective HSV vectors expressing interferon-
were injected into the cisterna magna or cerebral ventricles of mice. The vectors established stable infection of the leptomeninges, ependyma, and choroid, and secreted the cytokine in detectable quantities into the CSF [117]. Subsequent experiments using an interleukin-4-expressing HSV vector showed that the presence of the vector-derived anti-inflammatory cytokine was sufficient to ameliorate the pathological phenotype arising from experimental allergic encephalitis, an animal model of autoimmune CNS inflammation [114].
Further developments in vector delivery and use of the latency promoter system in the CNS may allow more widespread and long-term pathologies to be addressed, e.g., Alzheimer's disease and Huntington's disease.
Malignant Glioma Malignant glioma is a common, fatal malignancy of the CNS. An invasive tumor margin and sensitive local environment preclude complete resection, explaining the inability to effect curative surgery. Glioma are attractive targets for delivery of therapeutic transgenes using HSV vectors; distant metastasis occurs only under unusual circumstances, so the tumors are highly localized [118, 119]. This enables direct inoculation of the tumour or postoperative tumor cavity with recombinant vector, utilizing the route of CNS delivery that is effective with current technology. In addition, transient high-level transgene expression may be desirable to eradicate tumour cells, obviating the need for long-term transcriptional elements. The following data might apply equally well to other localized solid tumors, although we have concentrated our studies on models of glioma.
A series of replication-defective HSV vectors has been produced that deliver anticancer transgenes to malignant glioma cells [19, 69, 120-122] (Figs. 4, 7). The crucial features of HSV biology being exploited in this application are: A) the large capacity for the insertion of foreign genetic material, allowing simultaneous delivery of multiple therapeutic transgenes; B) the high infectivity of HSV-1 allowing infection of glioma cells at low multiplicity of infection, and C) the ability to produce large amounts of pure vector and infect at high multiplicity. In a series of experiments, it was shown that a significant antitumor response is greatly augmented by simultaneous delivery of multiple genes designed to induce toxicity to tumor cells by both direct transduction and by lysis of surrounding cells.
The UL23 gene of HSV1 herpes simplex encodes a thymidine kinase (HSV-TK) that functions to phosphorylate deoxypyrimidines with broad substrate specificity. This property allows the conversion of a prodrug ganciclovir (GCV) into its active form by HSV-TK, but not by its cellular counterpart. The phosphorylated form of GCV acts as a defective nucleoside analogue that becomes incorporated into replicating DNA and causes premature strand termination. Activated GCV is therefore toxic only to cells undergoing DNA replication, such that toxicity towards actively dividing tumor cells is much greater than to neurons or quiescent glia. It is not necessary to transduce all tumor cells with the HSV-TK gene, as in many cases, surrounding nontransduced cells are killed following GCV administration. This phenomenon is referred to as "bystander lysis" [121, 123]. Studies show that HSV-TK can be effectively delivered to tumors using a replication-defective HSV vector (TOZ.1; Fig. 4). Tumor lysis greatly exceeds the degree of transduction, and the reduced direct toxicity of multiply IE-deleted vectors enhances destruction of tumor cells by allowing greater expression of the transgene product [69].
In vitro bystander lysis is largely attributable to uptake of activated GCV by HSV-TK negative cells [121, 124]. The mechanisms responsible for bystander lysis in vivo are complex. Passage of activated GCV from HSV-TK positive to HSV-TK negative cells plays a key role, in addition to effects attributable to necrosis-induced inflammation and disruption of vasculature [125-127]. Activated GCV may pass from cell to cell through gap junctions [126, 128, 129]. These are intercellular channels formed by a number of proteins including connexin-43 [130]. Gliomas are often defective in connexin expression [131] and intercellular gap junctions [132]. On this basis, connexin-43 was incorporated into the anti-tumor vector (TOCX; Fig. 4). Connexin alone had an in vivo effect on animal survival that was comparable with that of HSV-TK/GCV, whereas the combination of HSV-TK/GCV and connexin proved to be synergistic [121] (Fig. 7).
In parallel experiments, viruses were generated that deliver HSV-TK with a secreted factor designed to effect destruction of neighboring tumor tissue (TH:TNF; Fig. 4). Tumor necrosis factor alpha (TNF-
) is a potent anti-tumor cytokine that demonstrates a range of actions against malignant cells, including the induction of apoptosis via activation of TNF- receptors, enhancement of HLA antigen expression in tumors, and immunomodulatory effects such as induction of natural killer- and cytotoxic T lymphocyte-mediated tumor lysis [133-138]. The molecule is too toxic to deliver systemically [137, 138], but the ability of HSV vectors to accommodate multiple transgenes readily enables its incorporation into a locally administered suicide gene therapy paradigm. Analysis of these vectors showed that TNF- expression had a direct antitumor effect in vivo and in vitro [120]. In vivo, the effect of gamma irradiation was augmented greatly by the simultaneous expression of TNF-, and combination therapy with TNF-, gamma irradiation and HSV-TK/GCV led to the ablation of many of the tumors [122] (Fig. 7).
Finally, vectors were generated that united the connexin-enhanced suicide gene therapy strategy with the TNF-/radiotherapy combination (Nurel-C; Fig.4). These vectors are currently under evaluation and may enter clinical trials in the near future. Further refinements under investigation include the use of additional transgenes to radiosensitize tumor cells, and evaluation of other biochemical targets to correct the basic mechanisms of tumor genesis.
Skeletal Muscle
Gene therapy for treatment of hereditary muscular diseases presents a number of challenges. First, the tissue compartment is vast; a large tissue mass is distributed over an immense area of the body, necessitating the use of large quantities of vector in conjunction with some means of systemic delivery. Second, the tissue is postmitotic. The commonest hereditary muscular disease, Duchenne muscular dystrophy (DMD), presents additional problems for gene therapy. DMD is an X-linked recessive disorder affecting 1 in 3,500 live male births. The phenotype is a manifestation of mutations that prevent expression of a muscle cytoskeletal protein, dystrophin. The cDNA encoding dystrophin is 14 kb in length, which is too large for the limited cloning capacity of many vector systems [139, 140].
The properties of replication-defective HSV vectors that may be useful for this application are: A) HSV has a large transgene capacity, adequate to accommodate the entire dystrophin cDNA and appropriate regulatory elements; B) HSV is maintained episomally in nondividing cells including muscle, and C) it is possible to manufacture vector at high titer (up to 109 pfu/ml) without contamination from helper virus.
In vitro, transduction of myoblasts and myotubes has been demonstrated with a single IE mutant (ICP4–) HSV vector expressing a reporter gene (SHZ; Fig. 4), but the single mutant virus caused cytotoxicity in infected cells [141]. In vivo, transduction of many muscle fibers in newborn muscle was demonstrated, but the transduction efficiency was much lower in adult fibers [141]. Two problems have been identified. First, the maturation-dependent reduction in transduction efficiency [142] is partly attributable to the muscle basal lamina acting as a physical barrier to viral particle access [143]. Second, vector-related cytotoxicity from single IE mutant virus and an immune response directed against viral antigens resulted in short-lived transgene expression [144]. Improved transduction and reporter gene expression was demonstrated following construction of triple IE mutant vectors (ICP4:ICP22:ICP27) (THZ; Fig. 4) both in vitro and in vivo [145].
Vectors expressing full-length or truncated dystrophin (THD and SHD, respectively; Fig. 4) were generated [146]. In vitro, these directed dystrophin synthesis in dystrophin-null myotubes. As expected, the triple mutant was less toxic than the single mutant. Importantly, the cellular localization of dystrophin at the cytoplasmic surface of the sarcolemma was restored when the full-length construct was delivered; diffuse cytoplasmic staining was evident with expression of the truncated form [146]. In vivo, sub-sarcolemmal dystrophin staining was evident in dystrophin-null muscle, but only in a small area adjacent to the site of vector injection [146].
Cell-mediated gene delivery might allow solution of some of the problems pertaining to systemic gene delivery. Several cell populations have been described that are capable of systemic engraftment and fusion with myofibers [147, 148]. Although these populations engraft at levels below the therapeutic threshold for phenotypic correction of DMD using current protocols for isolation, expansion, and engraftment, it may be possible in the future to use these populations to achieve widespread genetic correction in muscle. Ex vivo transduction of these cells with HSV vectors encoding multiple transgenes could be utilized to introduce immunosuppressive, myogenic-inducing, and other therapeutic transgenes, simultaneously. Hybrid vectors are currently under development; these should allow the stable integration of HSV-delivered sequences into the genomes of cell populations that will subsequently undergo extensive cell division (see below).
Stem Cells
Various stem cell populations have shown potential as tissue repopulating and gene delivery vehicles [149, 150]. The ability to transfer genes into human stem cell populations may permit cell-mediated delivery of therapeutic transgenes or even allow the fate of the cells to be specified, which is an exciting prospect worthy of further study. Transduction of stem cells has been reported using several vector systems. HSV-1 possesses many features that suggest that it may be a useful gene delivery vector for stem cell applications: A) HSV vectors are capable of incorporating large or multiple transgenes; it may be possible to express the multiple genes necessary to induce differentiation to a specific cell lineage; B) HSV infects dividing and non-dividing cells, and C) the viral genome persists episomally without disturbing host cell metabolism, removing the possibility of insertional mutagenesis and implying that the pluripotent nature of the cells would be unaffected by transduction-related toxicity issues.
Infection of stem cell populations has been demonstrated using replication-defective HSV vectors. CD34+ human mobilized peripheral blood and monkey bone marrow cells were infected with a triple IE mutant (ICP4–:ICP22–:ICP27–) replication-defective HSV-1 vector (TOZ.1; Fig. 4, [18]) [151]. Twelve hours following infection, flow cytometry analysis showed that almost all of the cells expressed a reporter gene encoded by the virus. The infected culture showed similar cell viability to an uninfected sample at days 2, 4, and 7 post-infection, consistent with our previous experience using this vector in other cell populations [18]. Transduced monkey CD34+ cells were transplanted into monkeys with skin autografts, which were biopsied 5 days later. Vector transduced endothelial cells were detected in the walls of nascent vasculature within the skin graft, and mononuclear cells from the peripheral blood and bone marrow showed reporter gene expression for over 3 weeks following transplantation. GCV treatment induced progressive necrosis of the vasculature supplying the autograft, which detached after 5 days' treatment [151]. This indicates that a functional and potentially therapeutic gene product may be introduced into stem cells using an HSV vector; the possibility of allowing HSV-TK-expressing stem cells to contribute to tumor neovascularization followed by GCV ablation is exciting. We are currently analyzing the utility of HSV-1 vectors in transducing CD34+ cells, purified from human umbilical cord blood. HSV-1 cell entry receptors are abundantly expressed on human cord CD34+ cells allowing us to achieve transduction levels of greater than 85% with a replication-defective vector (Wechuck et al., unpublished observations).
Results using other types of HSV vectors have been variable. First, a glycoprotein H (gH) deleted, disabled infectious single-cycle (DISC) HSV-2 vector has been shown to infect human bone marrow CD34+ stem cells [152]. Nearly 100% of the cells expressed a reporter gene 24-48 hours following infection at a low multiplicity (MOI = 2.0), but expression declined thereafter; no cytotoxic effects were reported, although it seems unlikely that this minimally disabled HSV-2 mutant could effect transduction in the complete absence of toxicity. Second, human peripheral blood CD34+ cells have been transduced with a replication-competent attenuated HSV-1 vector (ICP34.5–) [153]. Twenty four hours after infection at MOI = 10, flow cytometry analysis demonstrated that only 10% of the cells expressed a reporter gene. Those cells expressing the reporter gene were collected, and it was shown that 100% of the sorted population expressed a second reporter gene encoded by the same vector. The poor level of transduction in this report may be attributable to differences in vector, cell population, or infection technique.
It is worth noting that the nonintegrating nature of HSV-1 vectors will preclude their use in unmodified form for delivering genes to cells whose final destinations are many cell divisions away from the transduction event. In some respects, this may be advantageous for cell fate specification, where transient expression of differentiation signals may be optimal. The development of integrating HSV vectors, however, may allow the stable incorporation of HSV delivered sequences into stem cells to effect genetic replacement, for example.
Arthritis and Secreted Proteins
Rheumatoid arthritis (RA) affects 1%-2% of the population worldwide, reduces life expectancy, and is associated with significant morbidity [154]. Current treatment options include symptomatic treatments and disease-modifying agents. Available drugs are frequently ineffective and may be toxic; in addition, most fail to exert chondoprotective effects required to maintain the integrity of an inflamed joint. Recent advances in understanding the cellular and molecular pathophysiology of arthritis permit the rational design of therapies aimed at intervening in the basic disease mechanisms. It has now become clear that, irrespective of etiology, much of the intra-articular pathology of arthritis is driven by cytokines [155-158], particularly interleukin-1 (IL-1) and TNF-. Administration of either Il-1 receptor antagonist (IL-1Ra) or a soluble form of the TNF- receptor (TNFSR) as purified proteins has been effective in ameliorating inflammation in animal models of RA [159-161].
The development of gene therapy vectors allowing long-term local modulation of inflammatory cytokines would represent a major advance in the treatment of RA. Both ex vivo and in vivo approaches have been employed to treat RA using both nonviral and viral delivery vehicles. Successful ex vivo introduction of IL-1Ra in a rabbit model of RA using retroviral vectors [162-164] has culminated in phase I [165] and II [166] human clinical trials. IL-1Ra or TNFSR delivery to the joints of animals with experimental models of RA have been achieved using several other methods [167-171]. Many of the published studies have demonstrated a diminution in the inflammatory response, or in some cases, protection of articular cartilage. However, some vehicles simply exacerbated the inflammatory reaction, and transgene expression was short-lived in the majority of instances. Initial results using replication defective HSV vectors expressing IL-1Ra or TNFSR (T/O-IL1 and T/O-TNFsr; Fig. 4) demonstrated transient local expression of these products in the synovium and beneficial phenotypic consequences following intra-articular inoculation [172].
We sought to exploit the latency behavior of HSV-1 vectors by infecting the nerve terminals innervating joints. It was hoped that the vector would become latent within the relevant dorsal root ganglion cells and effect chronic, sustained production and secretion of transgene product from sensory nerve terminals into the joint space. Interestingly, neurons innervating the joint space were inconsistently infected, and the vector did not persist in synovial cells. However, we found that intra-articular inoculation of NGF expression vectors (SHN, SLN; Fig. 5) resulted in significant increases in joint lavage NGF levels, of a similar magnitude to those previously observed with IL-1Ra and TNFSR. Surprisingly, NGF was detected in the blood plasma and was demonstrable at increased levels up to one year after the initial inoculation [65]. Subsequent analysis showed that the primary site of transduction was joint ligament tissue. Interestingly, both the human cytomegalovirus immediate-early promoter, which directs transient expression in many other circumstances, and the LAP2 promoter remained active in ligament cells for an extended period. We have recently started to evaluate the utilization of adipose tissue as a depot site for the secretion of therapeutic proteins into the joint spaces and circulation. Adipose tissue represents an attractive target for vector transduction and delivery of transgene products; fat is accessible, abundant, and well vascularized, enabling secreted factors to gain access to the intravascular compartment. It appears, therefore, that replication-defective HSV vectors are able to persist in non-neuronal "depot" tissues and express transgenes long term. We are currently investigating ways of exploiting this property for the development of novel therapeutic strategies to target several diseases. First, the sustained intra-articular delivery of soluble inhibitors of inflammation is likely to represent an important advance in chronic arthritis. Second, the expression of nerve growth factor in the circulation may ameliorate the chronic peripheral neuropathies associated with many systemic diseases. Finally, this technology may enable the delivery of circulating proteins to replace crucial factors missing in genetic diseases such as the hemophilias, 1-antitrypsin deficiency, and C1-esterase deficiency.
|
CONCLUSIONS |
|---|
|
TopAbstractIntroductionBasic Biology of HSV-1Using HSV-1 to Make...Applications of HSV VectorsConclusionsReferences |
|---|
We now have access to an unprecedented amount of information concerning genetic structure and function following a quarter of a century of identifying disease-determining genes and the recent publication of the draft human genome sequence. Our ability to use this information for clinical applications may be partially determined by the availability of means to effect genetic modifications to human cells. Ongoing investigation will determine if the technology described here eventually becomes established in clinical practice.
|
REFERENCES |
|---|
|
TopAbstractIntroductionBasic Biology of HSV-1Using HSV-1 to Make...Applications of HSV VectorsConclusionsReferences |
|---|
