Recently through the development of gene targeting, we have been able to introduce specific gene mutations into mammalian cells via homologous recombination. While this technology has practical applications in the field of gene therapy, we have used gene targeting in mouse embryonic stem cells to both directly and indirectly make animal models for human genetic disorders.
The most commonly used targeting vector (TV) is the replacement vector which contains a linear array of isogenic genomic DNA on both sides of the mutated sequence. Following introduction into cells, the TV is thought to align itself with the native chromosomal sequences via its shared homology, which facilitates homologous recombination, either via double-reciprocal exchange or gene conversion. Relative to random integration events, the occurrence of homologous recombination is rare in mammalian cells. In order to enrich for targeted cell. lines, we have devised a positive- negative selection protocol, whereby transformed cells possessing random integrations of the targeting vector are selectively killed. Using this strategy, a positive selectable marker, neo or lacZ:neo is inserted within the coding region of a 10 kb plasmid subclone that is flanked by negative selectable markers HSVtk1 and tk2 .
In order to obviate problems encountered with constructing gene TVs in plasmid, we have devised methods for generating replacement vectors and gene targeting using l. phage vectors . A phage vector, lSyrinx2TK, bearing the negative selectable marker genes HSV-tkl and tk2 adjacent to a Spi+ stuffer region is compatible for the construction of gene targeting vectors by conventional restriction-ligation or recombination-mediated methods. In vivo recombination between plasmids and phage allows the introduction of neo and lacZ-neo mutations into precise positions in phage vectors using methods which select for double-crossover recombinants. One method, which utilizes the l gam gene to select against single crossover recombinants enables us to introduce specific point mutations into phage TVs using transplacement mutagenesis.
Currently, TV construction is the rate determining step in any gene targeting protocol: isogenic DNA must first be isolated and subsequently modified to generate the TV. Initially, recombinogenąc phage vectors were developed to isolate specific sequences from genomic libraries containing millions of random clones. Using an ES cell DNA library constructed in lSyrinx2TK, it should be possible to both isolate isogenically- derived DNA and generate a gene targeting vector in a single step via plasmid-phage recombination screening. Work is in progress to construct a lSyrinx2TK genomic library and generic neo recombination plasmids in order to test this hypothesis.
Muscle cell presents a very useful system to produce and deliver recombinant proteins into the circulation since myoblasts can be readily isolated from muscle biopsies, expanded in culture, and genetically transfected in culture to synthesize large amount of recombinant proteins. The implanted myoblasts can fuse to each other to form new myofibers or fuse to existing myofibers that have natural access to the blood circulation and a long life span. Direct intramuscular injection of an expression plasmid vector has also been demonstrated to have an effective transfection rate of local muscle cells and a considerable expression level of the gene tranfected. Another potential is that myoblasts derived from a patient can be reimplanted after genetic modification. Although gene therapy for the correction of genetic defects in muscle itself has been limited in that only a small portion of the defective muscle fibers can be affected, gene therapy by muscle cells is an approach of clinical significance, since a limited population of the engineered muscle cells is desired here. Several recent studies have used infectious viral vectors or viral promoter-derived plasmid vectors to direct the expression of recombinant proteins in muscle cells to correct Parkinson's disease (tyrosine hydroxylase) and hemophilia B (factor IX), to produce human growth hormone, cytokines, erythropoietin, and kallikrein (a vasodilation factor). It has been shown that those muscle-expressed protein factors had significant systemic effects on the host animals. However, the introduction of viral gene elements into the host cells has a potential risk of causing new genetic disorders. A better approach, therefore, is to use an endogenous muscle-specific gene promoter to direct the expression of recombinant proteins, that will also require less stringency for tissue-specific transfection. Another advantage of using all endogenous elements for the construction of transgenes for the expression of recombinant therapeutic protein reagents is the minimization of possible host immunorejection of the transfected muscle cells.
Cardiac and skeletal muscle troponin T (TnT) are major proteins expressed in cardiac or skeletal muscle cells in a highly tissue-specific manner. hl-calponin is an abundant actin- associated protein specifically expressed in smooth muscle cells. In our studies of muscle-specific protein gene expression and structure-function relationships, we have cloned all the three gene promoters. We have initiated collaborative studies with Dr. D.R. Edwards (Cancer Biology Research Group at U of C) to develop an approach using these muscle-specific gene promoters to direct the expression of TIMP-2 (a metalloproteinase inhibitor to act on anti-tumor invasion and metastasis) in muscle cells aiming for application in cancer therapy. We are also in the progress of expressing in muscle cells: TNFa, a cytokine that stimu1ates T-cell and B-cell proliferation and has direct cytotoxic activity on many tumor cells; and G-CSF, an established therapeutic reagent used in cancer patients receiving chemotherapy. We are also interested in using the smooth muscle specific hl-calponin promoter to express antisense RNA to cell cycle regulatory factors for the prevention of post-coronary angioplasty restenosis.
Retrovrial vectors are used in more than 70% of all human gene therapy protocols since the retroviral plasmids and the packaging cell system are well-established and retroviral transduction allows for permanent integration into target cell chromosomes. The current retroviral vector and packaging system is derived from murine leukemia virus (MLV). To generate a gene therapy vector, the gene of interest is cloned into a replication-defective retroviral plasmid which contains two long terminal repeats (LTR), a primer binding site, a packaging signal, and a polypurine tract essential to reverse transcription and the integration functions of retrovirus after infection. To produce viral vector, the plasmid form of a vector is transfected into a packaging cell line which produces Gag, Pol and Env of the retroviral structural proteins required for particle assembly. A producer cell line is usually generated using a selective marker, often a G418 resistant gene carried by the retroviral vector. Up to 107 of infectious particles can be produced from producer cell lines.
Infection of human CD4+ lymphocytes by HIV-1 induces rapid cell death. The rapid spread of infectious HIV in vivo is the key to disease progress. We have designed and tested an improved retroviral vector which is expressed at high levels in a wide variety of cell lines and is further inducible by HIV Tat. The vector is comprised of a modified LTR, a modified MLV packaging signal, an intron for splicing, and the HIV packaging sequences. Human CD4+ lymphocytes transduced with this vector exhibited almost 100% protection from both HIV-1 and HIV-2 infection and were void of the ensuing cell killing effects for longer than 40 days. The use of this vector would be advantageous in HIV gene therapy protocols. We plan to propose a clinical trial using this vector within the next year.
Our gene therapy program also includes anti-cancer immunogene therapy research. A promising idea for cancer therapy is to enhance the tumor immunogenicity in patients. To test this, we have developed vectors expressing several immuno-therapeutic genes: the human T-cell costimulator B7-2, the apoptosis ligand FAS, and therapeutic cytokines IL-2, IFN-g, GM-CSF and IL-12. Retroviral vectors have been constructed to express multiple genes simultaneously. Our studies have demonstrated coexpression of B7-2 and GM-CSF in long term tumor cell lines, primary fibroblasts and fresh tumor cultures. In cancer gene therapy efficacy studies, we are using two assay systems: the in vitro cell-mediated immunity assay and the in vivo hu-PBL-SCID/beige mouse model. The preliminary results showed that a T-cell costimulator B7-2-transduced glioblastoma cell line, but not the untransduced tumors, was efficiently rejected by allogeneic human PBLs in reconstituted hu-PBL-SCID mice. In collaboration with Dr. Petruk's group, a clinical trial protocol will be proposed within the next few months.
Improving the safety and efficacy of vectors for gene therapy continues to be an important focus for gene therapy research. In recent months we have asked the question as to whether it is possible to target cationic liposomal vectors to specific cell lines by means of monoclonal antibodies (mAb) against cell-associated antigens, or via ligands against cell surface receptors, and if this would result in an increase in the level of transfection of the targeted cell population. Cationic liposomes (DOTAP/DOPE, 1 :1) were coupled to mAb or asialofetuin via a thioether bond, either at the liposome surface or at the terminus of a polyethylene glycol-lipid derivative inserted into the liposome bilayer. Liposomes were associated with a plasmid containing the reporter gene, ū-galactosidase. In vitro experiments were conducted at a variety of DNA:lipid ratios and lipid concentrations. The cell lines tested to date consisted of two adherent cell lines, human colon carcinoma and human hepatoma and a suspension culture, human T-cell Iymphoma. In each case the attachment of a targeting ligand at the surface of the cationic liposomes resulted in a 3 to 12-fold increase in the level of transfection of the targeted cell line. This ligand-mediated increase in DNA delivery was most evident a low lipid concentrations, which little or no significant transfection was observed with non-targeted liposomal vectors. This approach appears to be successful in increasing the delivery and expression of DNA in specific cell lines in vitro.
Purine nucleoside phosphorylase (PNP) and adenosine deaminase (ADA) deficiencies are associated with T cell immunodeficiency and severe combined immunodeficiency disease respectively.
As progress towards developing model systems in which to examine therapeutic strategies, we have recovered three independent mutations at the PNP locus in the mouse. The mutants excrete PNP substrates in proportion to the severity of the enzyme deficiency. Two of the more severe mutations show progressive attrition in the number of thymocytes after two months of age, and an accumulation of CD4-CD8- T cell precursors and corresponding reduction in CD4+CD8+ cells in the thymus. In mature mice there is a 50% reduction in the number of peripheral Thy-l' cells and apparently normal levels of B cells. Peripheral response to T cell mitogens and IL-2 is decreased by as much as 80%. The PNP deficient mouse exhibits parallel immunological deficits to the human disorder and is an ideal model system in which to examine novel therapeutic approaches including the efficacy of haematopoietic stem cell gene therapy.
We have also followed from birth a male with ADA deficiency. This child immediately received ADA enzyme replacement. At the same time cord blood stem cells were recovered, transfected with an ADA construct and reintroduced into the patient. We have monitored the ongoing effectiveness of PEG-ADA therapy and peripheral blood Iymphocytes for evidence of ADA expression.
Previous attempts to treat diabetes mellitus (DM) with somatic gene transfer has failed because expression of the insulin transgene is not regulated by glucose. To overcome this problem, we constructed a chimera, pS14-Ins comprised of a glucose sensitive promoter (-4316 to +18) from the rat liver S14 gene fused to the human preproinsulin cDNA. When pS14-Ins was transfected into human hepatoma cell lines, HuH17 or HepG2 we detected immunoreactive insulin (lRI) using a human C-peptide RIA. Unfortunately, IRI in the media derived from the cells was not regulated by glucose. To observe the expected S14-response to glucose, we measured CAT-activity in primary cultured hepatocytes transfected with pS14-CAT (gift from H. Towle, Univ. of MN). Whereas, CAT-activity was lowest in cells cultured in the presence of 2.5~5.0 mM glucose, a 4-fold rise was observed upon exposure of these cells to 30 mM glucose. A detailed analysis showed that CAT-activity rose in a sigmoidal fashion with the predominant change in this activity occurring within the glucose range of 5.0 to 11 mM. In an identical set of studies, we substituted pS14-Ins for pS14-CAT and found IRI to be much lower in the cytosol compared to the culture media of hepatocytes, thus showing clearly that IRI was secreted from the transfected cells. More importantly, the abundance of IRI in the media rose in response to increasing concentrations of glucose. Insertion of the above chimeras into liver would allow us to test the feasibility of somatic gene therapy for DM using the S14-promoter. Hence, we modified a reported technique for the direct injection of DNA and used it to test both pS14-CAT and pS14-Ins in rat liver. Results showed the expected induction of both CAT-activity and endogenous S14-expression in response to carbohydrate feeding, T3 and dexamethasone. The abundance of IRI detected in the serum from rats injected with pS14~Ins was also regulated by the same set of stimuli. In summary, our results show that the sensitivity of the S14-promoter to glucose is highest in the range of 5.0-11 mM. The direct injection of pS14-Ins into liver enabled this tissue to synthesize and secrete IRI in response to factors known to regulate the S1 4-gene. These results may be useful in developing gene therapy for DM.
Immunogene therapy refers to the introduction of new genes coding for immunoreactive proteins into human cancer cells for the purpose of stimulating host immune responses against the tumor. We have been investigating the use of GM-CSF (cytokine) and B7-2 (T-cell costimulatory molecule) as genes for Immunogene Therapy For Glioblastoma using a model in SCID-beige mice. These mice are deficient in mature T/B lymphocytes and in NK cells. However, they can be repopulated effectively with human lymphocytes via intraperitoneal injection. In a series of experiments, mice were injected subcutaneously with cells from the D54MG human glioblastoma cell line previously transduced with the gene for either GM-CSF or B7-2 via a retroviral vector. A group of these mice were then reconstituted with human peripheral blood lymphocytes from a single donor. In addition, some mice received untransduced wild type- D54MG cells and a portion of these received human peripheral blood lymphocytes.
The experiments have shown a significant inhibition of growth of subcutaneously implanted gliomas in reconstituted mice with D54MG cells transfected with GM-CSF. A greater inhibition of tumor growth in reconstituted mice transfected with B7-2 costimulatory gene was observed. Immunohistochemical staining of tumor specimens from reconstituted D54MG-B7-2 mice showed an increased infiltrating human lymphocytes in the tumor. Results are still pending for the reconstituted D54MG-GM-CSF mice. Although the results are preliminary, they suggest that an effective immune response can be generated against human glioblastoma multiforme tumors by transfecting the tumor cells with genes for GM-CSF and B7-2. Further work on this model will involved repeating these experiments with a dual-cistronic retroviral vector carrying both the GM-CSF and B7-2 genes and with a control retroviral vector that carries only a non-specific reporter gene. In addition, tumor infiltrating lymphocytes (TILs) which have been isolated from the tumors of the reconstituted mice are being grown in culture. These TILs will be used to reconstitute mice that will subsequently be challenges with wild type D54MG to determine whether increased inhibition of tumor growth compared to fresh human peripheral blood lymphocytes is achieved (i.e. transference of immunological memory). Finally, it is our endeavour to develop an autologous SCID-hu model (i.e. utilizing tumor and PBL from the same glioblastoma patient) to rule out nonspecific allogeneic immune responses.
Malignant mesothelioma presents a frustrating therapeutic challenge due to the universally poor outcome despite current treatment modalities. Survival from the time of diagnosis is usually 8 - 10 months. In many centers, management consists of palliative/supportive care only. In an attempt to impact on this grim prognosis, the potential of gene therapy in malignant mesothelioma is being explored. The specific strategies that are described are 1) the use of a virally derived suicide gene (Herpes Simplex thymidine kinase (HStk)), 2) the targeting of a tumor specific marker, and 3) the down regulation of an essential growth related gene.
Gene therapy makes use of recombinant DNA technology to transfer a desired gene into the target tumor cell. This can be achieved through the use of viral vectors either with direct viral infection, the introduction of actively secreting virus producer cells to the tumor cell milieu or the bombardment of gene carrying particles to tumor sites.
The research program proposed will investigate first the feasibility of HStk-gancyclovir gene therapy in mesothelioma using three different delivery methods (retroviral, adenoviral and gene gun transfection). The successful insertion and expression of the HStk gene in mesothelioma would enable precise targeting of the gene modified cells with gancyclovir. The in vitro transduction of mesothelioma will be evaluated in both human mesothelioma - nude mouse and hamster mesothelioma - hamster tumor models. Determination of GCV sensitivity in the gene modified cells will include optimization of GCV dosage, timing and duration of drug therapy. The dynamics of viral or gene gun transfection on in vivo transduction of tumor will be investigated. PCR analysis for the presence of HStk gene in normal organs and tissues will be used to evaluate the potential risk of gene transfer to nontumor tissues and subsequent potential for systemic toxicity with GCV therapy.
The second route of investigation will be to further characterize a unique association of SV40 like sequences and large T antigen expression in 60% of mesothelioma and see if this can be exploited to specifically differentiate normal from tumor tissue. Since up to 5% of the total large T antigen can be identified on the cell surface on SV40 transfected cells, the viral antigen may serve as a unique tumor marker for specifically targeting mesothelioma cells. Anti-large T antigen monoclonal antibody viral conjugates will be tested initially in vitro and then in vivo to determine their ability to target mesothelioma (large T positive and negative cell lines). Large T antigen effect on transformation and oncogenicity might be manipulated by interruption of large T synthesis with antisense oligonucleotides or plasmids. Development of antisense oligonucleotide sequences and antisense producing plasmids and evaluation of their in vitro effects on large T antigen positive and negative cell lines will be the initial direction of research. Large T expression will be detected with immunohistochemistry and immunoprecipitation. Changes in transformation will be evaluated with soft agar assays to determine anchorage independent growth, growth assays in 0%, 1%, and 10% serum conditions, and tumorigenicity in an in vivo model.
The third and final strategy in this proposal will be to define the mechanism behind the decreased tumorigenicity associated with antisense to IGFR1. Antisense to insulin - like growth factor receptor (IGFR1) mRNA has been demonstrated to decrease the proliferation of mesothelioma in vitro and decrease tumorigenicity in vivo. The approach used will be FACS analysis of transfected antisense/sense clones and wild type cells to determine any changes in cell cycle. Changes in the transformed phenotypes will be evaluated as discussed above. Animal models will be developed to evaluate the immune response (humoral and cell mediated) to the transfected clones. If an increased immune response is a significant component behind the decreased tumorigenicity of the transfected clones, this would present wide reaching therapeutic implications in affecting not only the gene modified tumor cells but also other sites of tumor (metastasis).
The ultimate goal of this program will be the development of therapeutic approaches that can be used in mesothelioma patients.
Treatment of Hemophilia A has been complicated by transmission of viral infections and continuing problems with the complications of hemophilia, particularly hemophilic arthropathy. High purity recombinant and human derived FVIII are safe and stable, although expensive and inconvenient. Transplantation of spleen and liver have been curative but are complicated and have significant morbidity and mortality. Gene therapy for hemophilia should be efficient, safe, and convenient.
Our hypothesis is that:
Work to date:
Our current aims are:
Biochemical, molecular genetic, and electron microscopic investigations have shown that the B domain of the F VIII protein, is structurally separate, functionally unnecessary, and decreases the efficiency of FVIII expression in tissue culture cells. DNA sequences within 600 bases of the boundaries of the B domain deletion, are dominant inhibitors of transcription of F VIII. In addition, expression of foreign genes using retroviral vectors is at relatively low levels and large DNA fragments are difficult to package in intact retrovirus. The experiments described will improve expression through manipulation of the FVIII cDNA and of the promoter in the gene therapy clone. Combining anti-HIV strategies and FVIII sequences in the same vector provides a unique opportunity to study gene therapy in HIV infected hemophilia A patients.
Chemotherapy of cancer has not advanced to any substantial degree over the past several decades. In general, chemotherapeutic anticancer drugs do not have molecular targets that are specific to cancer cells. The result is that these drugs (e.g. cis-platin; 5-FU) have high toxicities, very low therapeutic indicies and low efficacy once multi-drug resistance develops. Their clinical use is more-often-than-not associated with poor quality of life. In strong contrast, it has been possible to direct antiviral chemotherapy (e.g. ganciclovir) against very specifāc molecular targets within the virus-infected cell, thereby providing low general toxicity and a very high safety factor.
Gene therapy is one of the most exciting spin-offs of basic molecular biology.1 Several approaches to the application of gene therapy in oncology have been envisioned. The most clinically-advanced concepts involve the introduction, into the cancer cell, of genes that will encode for molecular targets not normally found in mammalian cells.2 Indeed, a number of clinical trials are based on the introduction of the Herpes simplex virus (HSV) gene that encodes for viral thymidylate kinase (TK+). Once the gene is in place in the target (cancer) cell, therapy can be effected simply by the administration of a highly molecularly-targetted and systemically non-toxic antiviral drug such as ganciclovir.3 Our group at the University of Albetra has a long-standing interest in the development of radiodiagnostic and therapeutic antiviral nucleosides, particularly those with high aflānity for virus-encoded enzymes, including HSV-thymidylate kinase (SV-TK+). We have developed metabolically stable and highly potent antivirals such as E-5-(2-iodovinyl)-1-(2-deoxy-2-fluororibofuranosyl)uracil (IVFRU) for therapy and when radiolabelled, for diagnostic imaging of focal HSV infections.4 Imaging Herpes simplex encephalitis (HSE) has been of particular interest to us, so we have focussed on the development of IVFRU-prodrugs that will penetrate the intact blood-brain-barrier (BBB).5 This research has direct application to monitoring gene transfection and transduction in gene therapy of a number of pathological conditions, including cancer. Detection of viral gene expression is vital to proper timing of the therapeutic intervention. It is also probably the only practical means by which to detect viral gene expression in non-target tissue, a side-effect that could result in catastrophic toxicity if the antiviral drug were administered to patients expressing the viral 8ene in a vital tissue.
We have developed and/or adapted gene-transduced cell lines for both in vitro and in vivo study of several radiolabelled antiviral nucleosides and selected chemical delivery systems. Our data show selective uptake, both in vitro and in vivo (murine models), by the transfected model, and provide a rational basis for additional study in both experimental and clinical studies.
Our studies on the genetic transformation of insects include a considerable component that relates to gene therapy objectives. To achieve our goals, we are exploring new methods for stable (embryo) transformation and in vivo expression of recombinant proteins. A focal point of our work is the design of vectors and delivery systems that permit efficient integration of transgenes into the host genomes, preferably in all the cells of the host, and efficient tissue-specific or constitutive ex transformed organisms.
From a vector point of view, we are developing integration vectors (plasmids) that may direct efficient chromosomal integration and transgene expression on the basis of (i) homologous recombination; (ii) expression of specific transposases; and (iii) gene conversion mediated by the action of intron-encoded endonucleases. Vectors in this category are delivered to the hosts by direct injection into eggs following fertilization.
For direct delivery into tissues, including gonads (for transmission to the progeny), we are exploring the engineering of insect viruses with the aim to inactivate functions that are crucial for the completion of the infectious cycle following DNA replication. Our goal is to produce defective viruses that can infect, propagate into but not kill the cells of their hosts.
With regard to delivery (mainly for the plasmids), we are trying to find ways to replace the rather cumbersome embryo injections with delivery by direct injection of DNA into the male and female gonads and in the abdominal cavity. The injected DNA is tested both in its naked form and following complexing with lipid particles. Recombinant viruses, on the other hand, are delivered in all tissues by a simple injection in the haemocoel of the animal.
Eukaryotic chromosomes contain a variety of distinct elements that are essential to ensure their accurate reproduction and maintenance within cells, including origins of replication, centromeres and telomeres. Unfortunately, the identification and functional characterization of these structures have been hampered by their large size, repetitive sequence elements, and difficulty in purification. Because loss, gain or duplication of function of these elements can lead to gross chromosomal abnormalities and a consequent disease state, it is appropriate to establish an experimental system that will enable these structures to be identified and their function or dysfunction assessed. We therefore propose to undertake the identification, assembly and test of the structural elements necessary for the proper function and stability of a mammalian artificial chromosome (MAC). To achieve this goal, we will create a chimaera between an existing Yeast Artificial Chromosome (YAC) and selected human and other mammalian genomic sequences. The stability and autonomous replication of the MAC in human cells will be evaluated by a variety of assays, both before and after the addition of candidate origins, telomeres and centromeres. Members of the group who are based at McGill University will contribute their expertise in the isolation and characterization of origins of replication, and the Calgary group will contribute primarily to the analysis of centromeric and kinetochore-associated DNA fragments, as well as their expertise in yeast genetics, microinjection and gene expression technologies. Both groups will participate in the optimization of chromosomal structure and testing of the compatibility of centromeric and replicative sequences, and will collaborate with scientists at Geron Corporation on the incorporation and testing of mammalian telomeric structures. The development of a functional MAC will capitalize on the advances arising in many different fields of cell and molecular biology and genetics and provide a technological resource that would be extremely useful to other groups investigating genomic structure and function. In addition, the availability of a MAC may permit the development and test of effective model systems for chromosomal disorders such as triploidy and aneuploidy, for the changes in chromosomal structure and stability that accompany aging and oncogenesis, and could ultimately facilitate the design of improved vectors for gene therapy and biotechnology.
The Elliott laboratory in interested in using gene therapy to modulate immune responses against transplanted islets. There are two aspects to the immune responses which occur against transplanted islets. First, there are the allo- or xenogeneic responses which occur against any tissue graft. Second, autoimmune diabetics are know to have a strong anti-islet beta cell autoimmune response, and since it will be these patient who will receive the islet grafts, this causes a second immunological barrier to successful engraftment. We have used two different methods to introduce immunosuppressive genes into islets--recombinant adenoviruses and biolistics. Thus far two recombinant adenoviruses have been constructed, expressing either mouse IL4 or mouse IL10 under control of the CMV immediate early promoter-enhancer. These are non-replicating, E1 deleted adenoviruses. The second method used to introduce genes into islets involves use of biolistic transformation, or the "gene gun". In this case DNA is coated onto microparticles of gold and these are blasted into islets using a hand-held, high pressure helium driven device. We have found in control experiments that firefly luciferase activities of 4% of those obtained with a recombinant luciferase adenovirus can be achieved by biolistic transformation. This is a much higher level of expression than can be achieved by any other classic method of transfection (e.g. lipofectamine). In terms of immunosuppressive molecules being tested, in addition to TH2 type cytokines such as IL4 and IL10, we are also using CTLA4Ig, and have work in progress using soluble Fas ligand.
The work in my lab is currently focused on exploring viral anti-inflammatory proteins and calreticulin as potential therapeutic agents for atherosclerosis. Several viral proteins SERP-1 (a serine proteinase inhibitor with anti-inflammatory properties), T-2 (a TNF receptor homologue), and T-7 (an interferon gamma receptor homologue) reduce plaque formation after primary and secondary balloon injury in rabbit and microswine models of atherosclerosis (collaboration with Dr. G. McFaddens lab). Calreticulin, specifically the C domain, is a naturally occurring calcium binding protein with anti-thrombotic that we have been demonstrated to reduce intimal hyperplasia in a rat model of balloon induced iliac arterial injury atherosclerosis (collaborative effort with Dr. M. Michalaks lab). The SERP-1 gene has been introduced into COS and CHO cells for expression. Introduction of these agents through a viral vector as potential genetic therapeuric agents for the treatment of atherosclerosis has been considered. Drs. G. McFadden and C. Bleakley are currently investigating other viral gene products that reduce antigenicity of the expressed proteins. Dr. M. Michalak is currently studying the distribution of calreticulin in transgenic mice. These agents that we are studying therefore have already proven benefit in animals models and may potentially be introduced as genetic therapy.
Malignant gliomas are among the most treatment refractory of all tumors. Although post-operative radiotherapy has been shown to confer a modest improvement in survival, most patients die of failure to control local disease. Therefore, therapeutic strategies which disrupt the cellular and molecular systems utilized by tumor cells to repair radiation-induced lesions may greatly improve local control. The DNA double strand break (dsb) is considered to be the lethal lesion induced by ionizing radiation. We have recently shown that DNA-dependent protein kinase, a member of the phosphatidylinositol 3 kinase (PI3K) superfamily, is required for the repair of dsb (Science 267, 1183, 1995). Wortmannin, a fungal metabolite, is a specific and irreversible inhibitor of PI3K activity (Cancer Res. 54, 2419, 1994). Preliminary results from our lab suggest that treatment of human malignant glioma cells with low concentrations of wortmannin produce minimal cytotoxicity. However, exposure of wortmannin treated cells to ionizing radiation results in an ~2-fold increase in radiosensitivity. We therefore propose a gene therapy strategy whereby wortmannin would be selectively delivered to tumor cells and then activated within the radiation field. Treatment of cells with ionizing radiaiton has been shown to result in transcriptional activation of several early response genes, including c-fos, c-jun, EGR-1 and NFkB. Constructs containing a radiation-inducible promoter upstream of cDNA encoding for wortminnin, or an equivalent PI3K inhibitor, would be utilized. An advantage of using a radiation-inducible promoter is that multiple fractions of radiation (25-30) are commonly used to treat glioma. Small incremental gains in radiosensitivity are magnified by an exponent equal to the number of frations (Int. J. Radiat. Oncol. Biol. Phys. 54, 4266, 1994). In addition, newly developed techniques of conformal radiotherapy allow for specific targeting of radiation delivery with maximal normal tissue sparing.
Cytotoxic T lymphocytes are important effectors of anti-viral immunity, and induce target cell death either by secretion of cytoplasmic granules containing perforin and granzymes, or via signalling through the Fas cell surface antigen. Although it is not known whether the granule-mediated and Fas-mediated cytolytic mechanisms share common components, proteinase activity has been implicated as an important feature of both pathways. The orthopoxviruses cowpox virus and rabbitpox virus each encode three members of the serpin family of proteinase inhibitors, designated SPI-1, SPI-2 and SPI-3. Of these, SPI-2 (also referred to as crmA in cowpox virus) has been shown to inhibit the proteolytic activity of both members of the interleukin-1b converting enzyme (ICE) family and granzyme B. Cells infected with cowpox or rabbitpox viruses exhibit resistance to cytolysis by either cytolytic mechanism. Whereas mutation of the crmA/SPI-2 gene was sufficient to relieve inhibition of Fas-mediated cytolysis, in some cell types mutation of SPI-1, in addition to crmA/SPI-2, was necessary to completely abrogate inhibition. In contrast, viral inhibition of granule-mediated killing was unaffected by mutation crmA/SPI-2 alone, and was relieved only when both the crmA/SPI-2 and SPI-1 genes were inactivated. These results suggest that an ICE-like enzymatic activity is involved in both killing mechanisms and indicate that two viral proteins, SPI-1 and crmA/SPI-2, are sufficient to inhibit both cytolysis pathways. These molecules may be useful in the design of gene therapy vectors as they would allow transduced cells to resist T cell mediated killing.
The incidence of Squamous cell cancer of the Head and Neck is approximately 42,800 cases per year in the U.S., with worldwide projections of more than 500,000 annually. The overall survival among these patients has remained unchanged at approximately 45% for nearly 30 years. Treatment failures amongst these patients are local and regional; only 10-15% of patients die of distant metastases.
We believe that promising interventions at the molecular level, e.g. gene therapy, will hold great promise in the future. We want to insert a cytokine gene into tumor cells in vitro and use them essentially as a tumor cell vaccine. Animal studies have demonstrated that cytokine secreting tumor cells can produce a significantly greater T cell response in host animals than non gene modified tumor cells. Various cytokine genes are currently under active investigations. We want to use GM-CSF as it has been shown to produce potent and long lasting antitumor activity in mouse model.
It is difficult to consistently and quickly generate Head and Neck Cancer tumor cell lines. Fibroblast cells are now being used in some current protocols as they can be quickly cultured, quickly expanded to large numbers and easily transduced. They can then be used as a vehicle to deliver high and constant levels of GM-CSF to the vaccine site.
We propose to carry out a phase I study of effect of genetically modified autologus cancer cells expressing CD86 (B7-2) costimulatory ligand and fibroblasts expressing GM-CSF in patients with advanced, recurrent, unresectable Head and Neck Squamous cell carcinomas. Alternatively, fibroblasts genetically modified to secrete IL-12 can also be explored.
The vector is a retroviral vector, and is derived from a murine leukemia virus similar to the vectors employed by a number of investigators in the field. The cytokine vector utilized in this investigation (GM-CSF) contains a recombinant human GM-CSF cDNA under control of the MLV promoter or other strong eukaryotic promoters.
Conversion of cholesterol into bile acids represents a major pathway for the removal of whole body cholesterol. This process occurs exclusively in the liver and the first and rate limiting step is catalyzed by cholesterol 7alpha-hydroxylase (cyp7). Comparison of species that have differential sensitivity to diet-induced hypercholesterolemia suggest that bile acid synthesis is more efficient in resistant species.
We are now in the process of investigating the feasibility of using targeted gene delivery as a way to manage hypercholesterolemia. Our initial application of this technology is to express a cyp7 transgene specifically in the liver. Ultimately, this therapy may be useful as a primary intervention but we envisage its use as an adjunct to existing pharmacologic therapies to manage extreme cases of hypercholesterolemia.
The current experimental paradigm involves the use of an inducible promoter to maximize the expression of the transgene and the use lysine conjugated asialoorosomucoid (poly-L-lysine ASOR) glycoprotein to facilitate the targeted delivery of the transgene to the liver. Introduction of the complex into mice shows very minimal spillover of transgene expression to other tissues. There is now preliminary evidence showing that increasing the activity of cyp7 in the liver can increase cholesterol catabolism. Further experiments are intended to determine the magnitude of cyp7 activity required to achieve a therapeutically usable effect.