The Discovery of Tumor Antigens for Therapeutic Cancer Vaccines    abstract
Advisor:  Paul Harris

RNA: Method, Mechanism and Medicine   abstract
Advisor:  Ron Prywes

MALDI-TOF Mass Spectrometry: A Sustainable Platform for High-Level Multiplex SNP Detection   abstract
Advisor:  Jingyue Ju

Using Biotechnology to Discover New Antibiotics and Vaccines  abstract
Advisor:  Carol Lin

Cell-based and Tissue Engineering Approaches as Potential Solutions to Type I Diabetes   abstract
Advisor:  Clark Hung

Aging, Genetics and Environment  abstract
Advisor:  Carol Lin

Finishing Reactions for one BAC pair (1D17B11) of the Legionella Genome Project   abstract
Advisor:  James Russo

Genomics and Proteomics: Interwoven Fields at Work   abstract
Advisor:  Carol Lin

The Science and Policy of Stem Cells:  A Comparative Analysis of the United States and British Systems of Policy Formulation abstract
Advisor:  Michael Crow

Cell Replacement Strategies for Parkinson’s Disease  abstract
Advisor:  Carol Lin

Hepatoblastoma: A Scientific Investigation  abstract
Advisor:  Jessica Kandel

Pharmacogenomics: Opportunities and Challenges in Breast Cancer  abstract
Advisor:  Carol Lin

Genomic and Proteomic Applications in the Emerging Field of Pharmacogenomics  abstract
Advisor:  Daniel Kalderon

Pre-mRNA Splicing: Finding an Exon in a Haystack of Introns  abstract
Advisor:  Lawrence Chasin

Developing Non-invasive Drug Delivery Systems  abstract
Advisor:  Carol Lin

Madhurika Sankar, M.A.

Cancer Therapeutics and Research - Current Avenues, Recent Advances    abstract
Advisor:  Carol Lin

Viral Vectors for Gene Therapy   abstract
Advisor:  Carol Lin

Manali D. Talathi, M.A.

An Examination of the Three Most Researched Viruses and an Analysis of the Simian Virus-40 as Delivery Systems In Gene Therapy  abstract
Advisor:  Paul Fisher

Optimization of DNA Sequencing Using Matrix Assisted Laser Desorption Ionization Time of Flight Mass Spectrometry With Biotinylated ddNTPs  abstract
Advisor:  Jingyue Ju

Development of a Generic Polyketide Host from an Industrial Production Strain and New Genetic Tools for Combinatorial Biosynthesis   abstract
Advisor:  Carol Lin

Tumor Analysis Using DNA Microarrays: Bridging the Research Laboratory and the Clinic  abstract
Advisor:  Robert Pollack

The past decade marked the beginning of a cooperative effort to identify molecular markers that allow immune cells to attack and extinguish cancerous cells from the human body. Various methods have been developed to aid in this discovery effort. Beginning with the mastery of cultivating antigen specific T cells in vitro, biochemical and molecular genetic techniques such as CTL immunoscreening, SEREX, peptide elution, and representation difference analysis have been developed and/or converted into viable strategies for tumor antigen discovery. Each of these techniques has contributed significant knowledge regarding both the identity of these and antigens and the nature of their presentation to the immune system. Recently, bioinformatic and genomic methods have been utilized in this search, providing new leads in the vast hunt. Ultimately, medical professionals hope to use these antigens to target tumors in therapeutic vaccine therapy. Their ability to do so will greatly depend on the accurate identification of and the continued discovery of tumor antigens for an increasingly larger pool of studied tumor types. This paper discusses the merits and limits of these methods, and proposes a general approach to using these methods more effectively.

RNA interference or RNAi is a conserved gene-specific response to double-stranded RNA. It uses a post-transcriptional mechanism to interfere with the expression of protein-coding genes. It can be utilized as a powerful reverse genetics technique in many organisms including C. elegans, Drosophila, mice and even humans. While many of the important genes and enzymes involved are known, some investigation is still necessary in order to elucidate its mechanism. This knowledge will enable scientists to harness the power of RNAi and apply it to clinical medicine.

Single nucleotide polymorphism analysis is an increasingly important component of a wide range of biological investigation, ranging from biomedical research to ecology and evolutionary biology. This field has a host of applications including the characterization of population structure, the study of genetic diversity, and the identification of loci underlying the genetic component of complex phenotypic variation. However, in order to realize their utility, robust and efficient technologies are required for SNP discovery and genotyping. Currently, there exists a myriad of techniques for these purposes, yet no single method has emerged as a universal remedy for the range of problems that face individual investigators. What is clear is that cost, reliability, and precision are important criteria when selecting a particular technology. Towards these ends, matrix assisted laser desorption/ionization time of flight mass spectrometry coupled with streptavidin mediated solid phase capture represents a potential platform for the rapid identification of multiple single nucleotide polymorphisms in a highly effective manner.

To validate its utility, 20 sites were simultaneously assessed using two synthetic, single-stranded templates corresponding to regions of exon 7 (100 base pairs) and exon 8 (110 base pairs) of the human p53 gene. Conducting single base extension to incorporate biotinylated dideoxynucleotides in a single reaction permitted the purification of all 20 extension products away from unextended primer. Subsequent matrix assisted laser desorption/ionization time of flight analysis resolved 20 independent peaks at molecular masses equivalent to their anticipated values. Consequently, this approach could serve as an appropriate technique for the classification and functional analysis of single nucleotide polymorphisms.

Antibiotics and vaccines are the foundation behind the fight against pathogens. With the increase in antibiotic resistance as well as new and reemerging diseases the need for new antimicrobials is more urgent than ever. Also, the possible role of infectious agents in chronic diseases as well as organisms that have proven evasive targets for vaccines has only added to the growing dilemma physicians and scientists are beginning to face. The traditional method of random screening of compounds that kill microorganisms has only produced analogues of compounds that were discovered more than thirty years ago. Genomics is proving to be a valuable alternative to traditional drug discovery. The sequences of entire pathogen genomes are being completed at a rapid pace. These whole genome sequences are used as a starting point to find genes that might be essential to the organism's survival and/or necessary to induce infection. This data is than used to experimentally knock out these genes in order to validate the essentiality of the gene. Once an essential gene is found its protein product is than analyzed to see if it can be used as a therapeutic target for drug discovery. A more global approach is to use microarray technology to monitor gene expression of a pathogen and/or a host cell before and after infection. Proteomics is a relatively new field that promises to both identify all the proteins expressed by a pathogen and show how these proteins interact with each other and the host. Proteomics will therefore compliment genomic data and aid in rational drug design. Genomic and proteomic technology has been able to find many possible targets for antibiotics and vaccines. For example, pathogens have virulence genes that are not expressed by their nonpathogenic relatives. Therefore these genes can be the focus for antimicrobial genomics. Another example is the histidine kinase signal transduction system, which is a signaling mechanism that many bacteria use but is absent in mammals. In summary, molecular biology research might be a promising solution to the underestimated threat of infectious disease.

This paper focuses on the disease of diabetes, and more particularly, those individuals suffering from Type I diabetes. Current treatments for this disease are analyzed and include immunotherapy drugs, glucose monitoring and insulin delivery in both "open-loop" and "closed-loop" systems. Significant limitations with these current treatments are highlighted. Accordingly, a cell-based or tissue engineering approach is proposed as the most promising path to treatment. Many cell or tissue engineered therapies, at various stages of research and development, are analyzed across multiple criteria. In general, key factors and considerations are proposed for cell and tissue engineering therapies. These therapies include options such as human whole organs and islets, xenogenic approaches, and human embryonic and adult stem cells. Strengths and weaknesses of each approach are assessed to include analysis of the implant materials utilized. One hypothetical tissue engineering approach is offered as a possible ideal solution.

Aging is characterized by the accumulation of potentially harmful alterations in biomolecules such as proteins and DNA. Such alterations accompany advancing age, lead to gradual damages to diverse physiological functions and eventually result in increasing probability of disease and death. Oxidant-antioxidant redox homeostasis and the protein turnover play an important role in maintaining normal and functional life. It is believed that free radical damage to cell is the main mechanism for aging process. In this review, I summarize the effects of oxygen radical on the most vital molecules for life, DNA and protein, and the natural repair system against the damage. Based on our understanding of these aging mechanisms, it is possible to establish ways for prevention and future therapies.

Legionnaires' disease was first reported in 1976 at a hotel in Philadelphia and it was quickly revealed to be caused by a microorganism called Legionella pueumophila. Legionella pueumophila can cause this pneumonia or milder respiratory illness, especially in old or immunocompromised individuals. A group of research laboratories in Columbia is studying Legionella pueumophila and having the whole genome sequence of Legionella pueumophila will help them to discover the genes involved in the mechanism of pathogenesis. Therefore, a Legionella Genome Project sponsored by the National Institutes of Health is being carried out in the Genome Center at Columbia University. The goal of this project is to sequence the whole genome of Legionella pueumophila and share this information with the whole research community. Two approaches were used to sequence the whole genome: (1) whole genome shotgun sequencing and (2) clone-based (BAC) shotgun sequencing. My summer project focused on one BAC pair (1D17B11). First, Consed, a shared software package, was used to identify problematic bases within the sequence of 1D17B11 then primers were designed surrounding these bases in order to permit resequencing reaction to improve coverage. Second, at the time I initiated this work, the 1D17B11 region contained seven contigs with gaps between them, therefore primers were designed at the ends of each contig and PCR reactions were carried out using the designed primers to close these gaps. After resequencing reactions, most of the problem bases have reached the 3-fold quality. With regard to the PCR finishing reactions, several of them walked into the gaps between pairs of contigs; subsequent work resulted in joining of the seven contigs into two larger contigs.

With the publication of the rough draft of the human genome sequence, one may ponder the thought that the genomic era is near the end and the era of proteomics is in the forefront. However, when drawing on history, one realizes that genomics was born decades ago and proteomics also has its roots in the distant past. In fact, one can not make a finite distinction between the two fields; one realizes that the boundary between two are blurred as they both serve to enhance each other. Genomics can not exist without proteomics and vise versa. Only with these two fields working together can the drug discovery phase truly reach another level, one where our knowledge of the genes within our cells and the protein products it makes play a crucial role in developing uncharted drugs to tackle currently incurable diseases.

While research into stem cells has been heralded as one of the most promising fields in science, progress toward recognizing the intended therapeutic applications of these cells through regenerative medicine has historically been hindered in the United States as a result of a system of ineffectual policy formulation. This issue is examined herein.

As the basis of effective legislation lies in a thorough comprehension of the matter that is to be regulated, an extensive review of the scientific aspects of human embryonic, germ and adult stem cells is presented. Each type of stem cell has distinct biological characteristics, differing potential for medical therapy, and is steeped in varying levels of ethical controversy. In turn, the approach to governing the research and development of the different kinds of stem cells has been dissimilar. This is revealed through an examination of the policies established in the United States that have both directly and indirectly impacted stem cell technologies. The ineffectiveness of these regulations is highlighted by a parallel account of the history of British stem cell policy and an analysis of the implications resulting from the perceived distinctions.

With this understanding of stem cell policy in the United States, dating from the 1970's to present political decisions on the issue, in combination with the comparative analysis made in regard to British stem cell policy, a set of recommendations are set forward. In short, for the United States to retain its global leadership of science and technology it will be necessary for legislators to design policies in a timely fashion, which guide scientific research without restraining advancement.

Parkinson's disease is a neurodegenerative disorder characterized clinically by rest tremor, rigidity, and bradykinesia, and pathologically by a loss of midbrain dopamine neurons with a subsequent loss in striatal dopamine. Treatment with levodopa works initially, but reduced efficacy and development of motor complications has lead to alternative approaches to treating Parkinson's disease. Cell replacement strategies represent a new hope for patients suffering from Parkinson's. Fetal neural grafts have been investigated clinically and have shown some benefit to patients. Technical and ethical constraints have forced investigators to find alternative cell replacement therapies including xenografts, embryonic stem cell therapy, and adult stem cell therapy.

Hepatoblastoma, the most common primary liver tumor in children, accounts for over 2.5% of all primary tumors and nearly 50% of those that are malignant in children. In children, in the first 2 years of life, hepatoblastoma accounts for over 40% of the primary liver tumors. As with adults, however, the most frequently occurring malignancies in the liver of children are metastatic tumors, including neuroblastoma, lymphoma, and rhabdomyosarcoma. However, primary liver tumors account for only 0.5% to 2.0% of pediatric tumors, with hepatoblastoma thus accounting for 0.1% to 0.5% with the prevalence of 1 in 120,000. Hepatoblastoma is observed worldwide. Hepatoblastomas are not unique to humans and have been reported in animals including sheep, mice, cattle and horses. Here I propose to carry out a scientific investigation of the features of hepatoblastoma. I will review current research to outline the most recent status of this disease. I will review histology, epidemiology, the clinical and laboratory features of hepatoblastoma, its pathology, (including molecular markers), clinical management, including prognosis, current and potential future management options for hepatoblastoma.

Patients with identical clinical symptoms respond differently to the same drug therapy. Oftentimes, the suitability of a drug for an individual with breast cancer and other diseases is established through a trial and error process. Pharmacogenomics is the study of genetic factors that mediate a person's drug response and the science that may change the above trial and error process. Pharmacogenomic analysis promises to identify disease susceptibility genes thus revealing new drug targets. This will not only lead to a novel drug discovery process, but also to an individualized application for drug therapy and new insights into disease prevention. Current concepts in drug therapy focus on large patient populations that are gathered irrespective of potential individual genetic-based differences on drug response. In contrast, pharmacogenomics may offer a more effective therapy on smaller patient populations whose disease phenotype has been thoroughly characterized genetically. It still remains to be seen whether and to what extent this personalized, genetics-based approach to medicine will result in improved and economically feasible therapies.

The drug discovery industry has and will continue to benefit from the relatively new disciplines of genomics and proteomics. Advances made in the multidisciplinary field of genomics have lead to the near completion of the once unimaginable goal of sequencing the entire human genome. Once complete, the conversion of vast amounts of DNA sequencing data into the identification of novel genes and gene products will lead to a better understanding of normal physiology and disease states. Further, proteomics holds great potential as it represents a depiction of an organism at the phenotypic rather than the genotypic level. Information gained through proteomic studies includes the characterization of novel proteins and the identification of new protein-protein interactions and pathways. Eventually, genomic and proteomic studies will lead to a better understanding of the underlying mechanisms of disease. Subsequently, a better understanding of disease will lead to superior therapeutics with higher efficacies and reduced adverse effects. The emerging discipline of pharmacogenomics attempts to apply the innovative techniques of genomics and proteomics in order to better understand disease, drug response and eventually to produce more effective drugs with fewer side effects. The following is a review of genomic and proteomic techniques and some recent applications in pharmacogenomic studies.

Pre-mRNA splicing is a fundamental process of higher organisms that mediates the transfer of genetic information from DNA to protein. In humans, alternative splicing adds to the complexity of the proteome. The major challenge encountered when trying to understand the splicing process is splice site detection, which defines exon boundaries. Initial studies indicating that splice site consensus sequences were the determining factors for splice site choice have proven to be too simplistic when one considers that these consensus sequences are frequently located throughout the genome. Years of research has resulted in the identification of complex regulatory networks composed of exonic enhancer and silencer elements, intronic enhancer and silencer elements, and several trans-acting factors that work synergistically and antagonistically to ensure proper splice site choice. Additionally, secondary structures and the possibility of a nuclear open reading frame scanning mechanism have also been implicated to operate in the splicing network. Despite the progress that has been made, a lot more information will be needed before we can hope to make ab initio splice site predictions.

Developing drug delivery systems that can deliver drugs to its target sites at a desired rate and duration will create opportunities for novel drug therapies. By adjusting drug release rates at appropriate times and at specific destinations, the therapeutic effects of drugs will be maximized. Biotherapeutics have a substantial need for non-invasive drug delivery systems. Currently, potential biotherapeutics are largely dependent on injectable or parenteral delivery, an inconvenience to most patients. On the other hand, reformulations can significantly improve the therapeutic effects of existing pharmaceuticals. The overall objective in developing unique drug delivery systems lies in improving clinical outcomes. Decision maps or drug therapy profiles specific to the diseased condition must guide the development of drug delivery systems. Designing systems that maintain drug efficacy throughout its non-invasive routes of administration - oral, transdermal and pulmonary - must be achieved. Such delivery vehicles include natural or synthetic polymers, liposomes, or combinations of both. The most promising research in delivering biotherapeutics is pulmonary delivery. Even so, oral delivery is preferred and efforts to improve diffusion across the intestinal epithelium should be continued. Future design parameters will include responsive drug delivery systems to symptomatic triggers. Polymers can be designed to release drugs to changes in temperature and pH or can be designed to release multiple drugs at various rates.

By any measure, the amount of information gathered over the past two decades about the workings of cancer is without parallel in the history of biomedical research, fueled further by the mapping of the human genome. Some of this knowledge has already been put to good use, building molecular tools for detecting and determining the aggressiveness of certain cancer types. However, despite vast insight into cause, new curative therapies have remained relatively elusive, one reason is that the difference between tumor cells and healthy ones is minimal; tumor cells being mutated in a minute fraction of the genome. Nonetheless the course of the battle is changing. These subtle yet unique differences between normal and cancerous cells provide excellent targets for intervention by newly developed, targeted anti-cancer molecular therapeutics. This research endeavor will reward us in the coming years with cancer therapies that earlier generations could not have dreamed possible.

Gene therapy holds great promise and potential as a successful form of treatment for a multitude of diseases and disorders. Vectors are used in gene therapy to transfer a gene of choice to a host cell in order to alleviate the symptoms of disease. Viruses, by nature penetrate and infect host cells, which in turn takes over the cells machinery producing viral proteins. If this property is harnessed, the gene that is introduced in the virus can be expressed. By exploiting the infective nature of viruses as host delivery vehicles, functional genes can be delivered into desired cells with the option of long-term or short-term expression of this gene. The current limitations, however, of administering viral vectors as a form of therapy is their immunogenicity and potential for undesired dispersment to other cells. This thesis examines the advantages and limitations that viral vectors have as vehicles for gene delivery and why they have a great potential as a successful form of treatment.

The field of gene therapy is very young and experimental. Researchers have not been triumphant at developing successful gene therapy techniques because of many factors. The foremost impediment is the gene delivery vehicle. At present, the most common vectors are viruses. Scientists have tried to take advantage of the virus's ability to encapsulate and deliver their genes to human cells in a pathogenic manner by manipulating the virus's genome to remove the disease-causing genes and to insert therapeutic genes. However, the most commonly used viruses introduce problems to the body such as immune and inflammatory responses, gene control and targeting issues, and toxicity. The limitations with these viruses made scientists seek novel viruses for use in gene therapy. Three of the most extensively researched viruses are analyzed as well as the prospect of using a novel virus, Simian virus-40 as a gene delivery vehicle in this paper. SV40 is a small double stranded, circular DNA virus of rhesus monkey origin. Viral vectors based on SV40 are easily manipulated, produced at a high titer and provide high levels of transgene expression in almost any cell type. The limitations with this vector include packaging capacity and the lack of in vivo studies. Future studies with this novel vector will assess its true value as a delivery system. While SV40 does not provide the solution to the current problems faced by gene therapy, it does provide diversity to the available systems of gene delivery. The solution may lie in the development of systems that incorporate attributes of more than one kind of viral vector and thus should be vigorously investigated.

In the post-genomics era, there is a great demand for improved DNA sequencing technologies. Currently, the most commonly used technique for DNA sequencing is capillary array electrophoresis with laser induced fluorescence detection. However, in recent years, matrix assisted laser desorption ionization (MALDI) time of flight (TOF) mass spectrometry (MS) has emerged as a potential alternative due to its advantages of speed and accuracy. This study focuses on improving DNA sequencing read lengths using a synthetic DNA template mimicking a portion of an HIV protease gene and a MALDI-TOF mass spectrometer in combination with a solid phase capture technique, which was recently developed at Columbia University. The basic methodology of DNA sequencing with mass spectrometry is to generate a ladder of DNA fragments using the Sanger chain termination method followed by solid phase capture and subsequent analysis by mass spectrometry. The current factors limiting sequencing read lengths are DNA fragmentation and adduct contamination. However, this study attempts to compensate for these limitations by modifying the ddNTP/dNTP ratios used in Sanger sequencing reactions to shift the ladder distribution in favor of longer oligonucleotides. As a result, this study separately investigates the optimum ratios for all four biotinylated terminators. These results are then combined and fine-tuned to increase the sequencing read length to 34 base pairs, thus representing an improvement of 9 base pairs over previously published results. This study demonstrates that MALDI-TOF-MS has the potential to complement DNA sequencing by electrophoresis with laser induced fluorescence detection in the future, but further improvements are needed.

Polyketides from actinomycetes and other natural sources are well noted for their broad range of therapeutic activities. Combinatorial biosynthesis, a novel drug discovery strategy, exploits the modular nature of complex polyketide synthases and their related biosynthetic pathways to generate novel polyketides. The ability to manipulate the catalytic activities of these multifunctional enzymes has led to the design and production of many novel analogs for development of lead compounds. Construction of "unnatural" polyketide products libraries by genetic manipulation of polyketide synthases promises to be a valuable tool for drug discovery. However, one major challenge to the construction of very large compounds libraries is the decline in production levels associated with many genetically modified polyketide synthases, particularly those in which multiple domains have been modified. One of the strategies for higher levels of polyketide exploits the overproduction capabilities of industrial strains that normally make far greater quantities of polyketides. A problem with these strains is a lack of established vector/promoter systems and a apparent high barrier to traditional Streptomyces DNA transformation methods.

In this thesis I review the importance of polyketides for the treatment of diseases and the opportunities offered by the Combinatorial Biosynthesis approach in the modern drug discovery process. In addition, this thesis detailed the research I conducted to address two fundamental problems for the production of new polyketides: the heterologous expression for overproduction and the design of new genetic tools for combinatorial biosynthesis.

Recent research efforts have yielded cogent evidence for the potential of DNA microarray technology to assist in cancer classification, diagnosis, and identification of targets for therapeutic intervention. Still, as of yet, there is no clinical evidence indicating a statistical improvement in life expectancy of cancer patients as a direct result of microarray-based tumor analysis techniques. Several barriers are preventing the application of the research laboratory results in the clinic. These include the lack of adequately specific anti-cancer drugs, as well as the weakness of microarray technology due to current data analysis limitations. It is expected that with time, new advances will aid in eliminating these barriers and DNA microarray technology will greatly enhance the specificity and efficacy of cancer treatment.