Michael Mathis, Ph.D.
Dept. of Cellular Biology and Anatomy
B. Jill Williams, Ph.D.
Dept. of Urology
Arrigo De Benedetti, Ph.D.
Dept. of Biochemistry
Ronald Klein, Ph.D.
Dept. of Pharmacology
Cherie-Ann Nathan, M.D.
Dept. of Otolaryngology
Kate Ryman, Ph.D.
Dept. of Microbiology and Immunology
Francesco Turturro, M.D.
Dept. of Medicine
Wei-Ming Duan, M.D., Ph.D.Dept. of Cellular Biology and Anatomy
Benjamin Li, M.D.
Dept. of Surgery
Qian-Jin Zhang, Ph.D.
Dept. of Cellular Biology & Anatomy
Plasmid DNA - liposome
Adult stem cells
Cancer (breast, head & neck, ovarian, prostate hematopoietic)
- Exploiting the translation initiation factor eIF4e (gene targeting)
- Cancer gene therapy paradigms (mechanisms)
- Novel viral vectors for cancer gene therapy (gene delivery)
- Gene therapy of neurodegenerative diseases
INNATE IMMUNITY IN ADENOVIRUS
GENE THERAPY OF CANCER
The following hypothesis will be tested in this project: immunological mechanisms involving natural killer (NK) cells contribute to the efficacy of adenovirus-mediated gene therapy in a mouse model for cancer, and these mechanisms can be enhanced to improve the efficacy of therapy. Gene therapies delivered by adenovirus vectors have been quite successful in rodent models, and complete elimination of tumor cells is often achieved, even with poor transduction efficience in vivo. These results suggest the mechanism of killing by the adenovirus involves a "bystander effect", in which uninfected neighboring cells are killed by a factor or signal from adenovirus-infected cells. We hypothesize that populations of NK cells in mice treated with adenoviral vectors are involved in the therapeutic response to cancer. In Specific Aim 1, we will determine the role of NK cell activation and recruitment in cancer gene therapy using the adenovirus vector. The role of NK cells in adenovirus-mediated gene therapy will be evaluated using syngeneic ovarian and breast tumor cells in immunocompetent mouse strains. The pattern of activation and tumoricidal activity of NK cell and cytokine expression will be determined after administration of adenoviral vectors in normal and tumor-bearing mice. Survival studies will be performed in which mice are depleted of NK cells. In Specific Aim 2, we will determine the mechanism of NK activation and the role of macrophages in the efficacy of NK cell-mediated tumor regression with adenovirus gene therapy. We will determine the mechanism of NK cell activation by adenoviruses through the regulation of the cytolytic mediator perforin. We will also measure expression of the apoptosis-inducing molecules, TNF-a and Fas ligand, produced by activated NK cells. The pattern of cytokine and chemokine expression will be investigated in NK cells after Ad-null administration. Changes in number, activation and tumoricidal activity of macrophages will be analyzed and survival studies will be performed in normal mice and in mice in which macrophages have been depleted. In Specific Aim 3, we will determine whether the activation of macrophages and/or NK cells can be enhanced in vivo to improve the efficacy of gene therapy. Initially, cytokine gene therapy treatments using adenoviruses expressing IL-2 (to activate NK cells) or IFN-g (to activate macrophages), or a combination of these will determine whether boosting NK and/or macrophage activity can increase efficacy of gene therapy. Allogeneic populations of lymphokine activated killer (LAK) cells will be generated in vitro and infected with Ad-IL-2 to test the hypothesis that these cells can remain cytolytic to tumor cells and deliver IL-2 in vivo. Understanding the mechanisms of the innate immune response to adenoviruses is important to identify strategies that enhance their effectiveness.
TRANSLATIONAL REGULATION OF HSV-TK IN CANCER GENE THERAPY
The purpose of cancer gene therapy is the elimination
of tumor cells while sparing non-tumor cells of the cytotoxic effects of the
cancer treatment. To accomplish this goal three separate objectives must be
achieved. First, the gene must be delivered to tumor cells at a sufficient
level to affect treatment. Secondly, once delivered the gene must be expressed
by the transduced cells at a frequency and duration sufficient to affect viability
of the cells. Thirdly, the gene therapy must be able to arrest tumor growth
or reduce the tumor mass at the site of occurrence to be considered therapeutic.
Over the past decade numerous animal studies have clearly shown that delivery
of the suicide gene Herpes Simplex Virus thymidine kinase (HSV-Tk) by an adenovirus
vector in conjunction with administration of the prodrug ganciclovir (GCV)
is a viable strategy for cancer gene therapy. However, phase I clinical trials
have only produced limited therapeutic responses. Poor gene delivery and limited
distribution within the tumor are likely limitations to the current methods.
Additionally, lack of tissue specificity in delivery and expression has also
limited the efficacy of the therapy by the necessity of lower GCV levels to
reduce toxicity in non-tumor tissues. In order to advance cancer gene therapy,
we propose to utilize an adenovirus to deliver a modified suicide gene to
tumor cells in a mouse model for breast cancer. The adenovirus vector will
allow gene delivery to a broad range of cells. Expression of the suicide gene,
however, will be regulated by a modification using a 5' upstream-untranslated
region (5'-UTR) such that translation of the modified suicide gene (Ad-UTk)
is restricted to cancer cells expressing high levels of the translation initiation
factor/proto-oncogene, eIF4E. Previous studies have shown that eIE4F is expressed
at higher levels than in normal surrounding tissues in a broad spectrum of
spontaneous human tumors. These finding allow a broad application for this
approach in human cancer treatment.
The hypothesis for the proposed studies states that expression of the translationally regulated suicide gene, Ad-UTk, will limit the cytotoxic effects of GCV to tumor cells expressing high levels of eIF4E. Preliminary data with the translationally restricted suicide gene (UTk) delivered by liposomes significantly reduced the non-tumor tissue cytotoxicity with Ad-UTk/GCV treatment compared with the unmodified suicide gene (Ad-HTk/GCV treatment in a mouse model. In order to proceed with this approach to Phase 1 human trials we have incorporated the translationally restricted UTk into an adenovirus and propose to demonstrate its efficacy in a mouse model for breast cancer by the following Specific Aims: Specific Aim I: To validate the delivery of a modified virus suicide gene using an adenovirus vector (Ad-UTk) to tumor tissues in a syngeneic mouse model for breast cancer. PCR and immunohistochemistry on tissue sections will determine suicide gene delivery. These studies will determine tissue specificity, efficacy of the delivery method, and duration of expression of Ad-UTk in a syngeneic BALB/c mouse model for breast cancer. In addition, cytotoxic effects of the treatment will also be determined at the cellular level in both tumor and non-tumor tissues by histological analysis. Specific Aim II: To validate the efficacy of Ad-UTk in a syngeneic mouse model of cancer treatment. Survival studies and PET scans will allow the determination of efficacy of the cancer treatment in vivo in the mouse model. PET analysis after gene delivery will allow determination of the tissue specificity of suicide gene expression.
DENDRITIC CELL-BASED GENETIC
FOR OVARIAN CANCER USING A SYNGENEIC MOUSE MODEL
In the following project, we propose to evaluate a dendritic
cell-targeted adenovirus vaccine expressing the simian virus 40 (SV40) large
T antigen (T-Ag) in a mouse model of ovarian cancer. We hypothesize that immunization
of dendritic cells with the SV40 T-Ag will be effective in inducing antigen-specific
cytotoxic T-lymphocyte (CTL) responses, and suppress the growth of ovarian
tumor cells expressing the SV40 T-Ag. Dendritic cells (DCs) are professional
antigen presenting cells (APCs) that direct the cellular immune response through
antigen presentation in the presence of appropriate co-stimulation. Multiple
techniques of antigen priming have been advocated to create a tumor-specific
DC vaccine. The expression of whole tumor antigens within the host antigen
presenting cells following vaccination, has been demonstrated to result in
the presentation of multiple tumor-associated epitopes in the context of MHC
class I and/or class II molecules. Although the ex vivo genetic manipulation
approach in priming DCs has been useful in establishing proof-of-principle
in an experimental setting, it is not appropriate for the development of practical
therapeutic anti-tumor regimens. If transduction of the DCs could be accomplished
In situ, the broad use of DC-targeted vaccines as anti-tumor agents
would be more feasible. However, a critical component of In situ transduction
is efficient targeting of the vector to DCs without perturbation of DC function.
To this end, we have generated an adenovirus (Ad) vector system that specifically
targets human and mouse DCs via the CD40 receptor via CAR-CD40 bispecific
adapter molecules and have shown that the CD40-targeted Ads efficiently transduce
DCs in vitro without interfering with DC function. The availability
of a DC-targeted Ad system makes possible the next step in establishing an
anti-tumor vaccination model system.
In the following experiments, our goal is to utilize a model of ovarian cancer in mice with an intact immune system. Although the immunodeficient mouse is an important animal model for the preclinical development of new cancer therapies, the major disadvantage of this model is the absence of T cells that may modulate therapeutic effects. The use of mouse ovarian cancer cell lines to form tumors in normal immune-intact mice will provide a model in which immune interactions in the establishment, progression, and treatment of ovarian cancer can be investigated. In addition, the use of an immune-intact mouse model will provide a more realistic representation of human cancer and the potential efficacy of gene manipulation strategies. There are only a few syngeneic models available for intraperitoneal ovarian cancer in mice. However, Connolly et al. [Cancer Res. 2003 63:1389-97] recently reported that female transgenic mice expressing the SV40 T-Ag under control of the Mullerian inhibitory substance type II receptor gene promoter, develop bilateral ovarian tumors in approximately 50% of cases. Histologically, these tumors are poorly differentiated carcinomas with occasional cysts and papillary structures present at the surface of the ovary. These tumors disseminate intraperitoneally, invade omentum, and form ascites as do human ovarian carcinomas. Cell lines derived from the ascites (MOVCAR cells) exhibit the properties of epithelial ovarian cancer, such as anchorage-independent growth, tumorigenicity in SCID immunocompromised mice, expression of epithelial cell markers, and organotropic implantation. The SV40 T-Ag is highly immunogenic, inducing both antibody and cytotoxic T lymphocyte (CTL) responses. Since this antigen is synthesized in MOVCAR cells, the SV40 T-Ag is an attractive candidate as a model system for the development of a DC-targeted cancer vaccine. Recently, we have demonstrated that MOVCAR cells express the SV40 T-Ag and form tumors in syngeneic immunocompetent B6C3F1 mice. Therefore we have all of the reagents necessary to directly evaluate the efficacy of a dendritic cell-targeted adenovirus vaccine, using a mouse model of ovarian cancer.
Specific Aim I: Transduction of DCs in vitro with a CD40-targeted Ad vector expressing the SV40 T-Ag (Ad5sv40-CFm40L) will result in a high level of transgene expression, and be effective in inducing an antigen-specific CTL response in vivo. We will assess the in vitro effects within isolated dendritic cells and in vivo in CTLs after infection with Ad5sv40-CFm40L. Specific Aim II: Dendritic cells infected ex vivo with Ad5sv40-CFm40L will generate protective tumor immunity in vivo against MOVCAR tumor cells expressing the SV40 T-Ag. We will establish a syngeneic immunotherapy model of ovarian cancer using MOVCAR tumor cells. Specific Aim III: In situ administration of Ad5sv40-CFm40L will generate tumor immunity against MOVCAR cells tumor cells expressing the SV40 T-Ag. We will determine whether systemic administration of mice with a single dose of Ad5sv40-CFm40L will result in protection against a challenge from a lethal dose of MOVCAR tumor cells.