Mathies People: Dave Duberow

Early detection and characterization of cancer is vital in the selection and execution of a successful treatment regimen. In recent years, promising advancements have been made in the molecular and genetic detection of human cancer, the genetic determination of cancer risk factors, and the use of high-throughput genetic analysis to stratify tumors for treatment. While much of this work has focused on proteomics, nuclear genetics, and expression analysis, much less attention has been placed on the role of mitochondrial genetic variation and mutation in cancer.

A compact, circular genome of 16,569 bp, the mitochondrial DNA (mtDNA) encodes several proteins vital in the oxidative phosphorylation pathway. This function, coupled with studies examining the suspected role of mitochondria in apoptosis, establishes the potential existence of carcinogenic mutant sites within the mtDNA, as tumor cells generally exhibit abnormally high metabolic activity and resistance to programmed cell death. Indeed, recent work on paired blood and tumor samples has revealed a high frequency of mtDNA variation in several forms of cancer including bladder, head and neck, lung, and prostate [1,2]. A thorough analysis of mtDNA sequence variation in a variety of tumors would be valuable in characterizing the relevance of this variation in cancer risk, onset, and treatment. Unfortunately, the high cost and lengthy analysis times associated with sequencing-based assays have been a major hurdle in the effort to implement such techniques clinically. In addition, conventional sequencing techniques fail to account for genetic heterogeneity, thereby rendering mutation detection within tumor margins impractical.

Recent advances in the Mathies Lab, beginning with the development of the 96-lane microfabricated capillary array electrophoresis bioprocessor (Figure 1), have laid the foundation for overcoming these obstacles. The device features 96 electrophoretic separation channels, each 200 mm wide, 30 mm deep, and 17 cm long, arranged radially on a 6-inch glass wafer and converging on a common anode. Sample injection is performed electrokinetically via 48 cross-injectors, with two adjacent lanes sharing common cathode and waste wells. In initial tests, the 96-lane chip was capable of producing 1.7 kbp/min of high-quality sequencing data, with a total separation time under 30 minutes [3].

A more recent innovation of the Mathies Lab is the comparative sequencing technique known as Polymorphism Ratio Sequencing (PRS), whereby two DNA samples are sequenced using a modified dideoxy-termination technique with four universal energy-transfer dye-labeled primers. Reaction mixtures are pooled to allow simultaneous sequencing of two bases from each DNA sample in a single capillary. Polymorphic sites are elucidated as peaks in a squared difference plot for each base sequenced (Figure 2). Correlation of all four traces provides a fingerprint by which the specific nature of the polymorphism can be identified. Performing sample and reference separations in the same lane yields an inherently controlled data set and allows for the quantitation of genetic variation within heterogeneous tissues, with a detection limit of 5% minor allele frequency [4].

My initial experiments have been directed toward the optimization and streamlining of PRS protocols on the 96-lane chip using standard mtDNA samples. A rigorous optimization of injection voltage conditions has yielded consistent, reliable, and high-quality electrophoretic separations across the entire device. Following the development of integrated, user-friendly data analysis software in collaboration with Jing Yi in Terry Speed’s statistics group, we are currently engaged in a pilot study of human lung cancer samples in which paired blood and tumor specimens will be analyzed in search of cancer-specific mutations. Once characteristic variants have been identified, work will commence on the design and validation of a more integrated system capable of performing on-chip DNA extraction, PCR amplification, Sanger sequencing, and capillary electrophoresis. Our final goal is to develop a portable instrument for clinical applications both in the diagnosis stage and in the operating room, where analysis of tumor margins could provide molecular pathology to monitor tumor excision in real time.

Figure 1: Layout of the 96-lane Microfabricated Capillary Array Electrophoresis Bioprocessor

Figure 2: Typical PRS “A” trace from Hypervariable Segment 2 of standard mtDNA samples. Verified polymorphic sites at base positions 185, 188, and 228 are clearly visible both in the overlaid traces (below) and as peaks in the squared difference plot (above).

<pre-	-next>	Dave Duberow
		Dave Duberow

		Ph.D. Student, UCB Chemistry B.S. Chemistry, Penn State Erie, 2004

Early detection and characterization of cancer is vital in the selection and execution of a successful treatment regimen. In recent years, promising advancements have been made in the molecular and genetic detection of human cancer, the genetic determination of cancer risk factors, and the use of high-throughput genetic analysis to stratify tumors for treatment. While much of this work has focused on proteomics, nuclear genetics, and expression analysis, much less attention has been placed on the role of mitochondrial genetic variation and mutation in cancer. A compact, circular genome of 16,569 bp, the mitochondrial DNA (mtDNA) encodes several proteins vital in the oxidative phosphorylation pathway. This function, coupled with studies examining the suspected role of mitochondria in apoptosis, establishes the potential existence of carcinogenic mutant sites within the mtDNA, as tumor cells generally exhibit abnormally high metabolic activity and resistance to programmed cell death. Indeed, recent work on paired blood and tumor samples has revealed a high frequency of mtDNA variation in several forms of cancer including bladder, head and neck, lung, and prostate [1,2]. A thorough analysis of mtDNA sequence variation in a variety of tumors would be valuable in characterizing the relevance of this variation in cancer risk, onset, and treatment. Unfortunately, the high cost and lengthy analysis times associated with sequencing-based assays have been a major hurdle in the effort to implement such techniques clinically. In addition, conventional sequencing techniques fail to account for genetic heterogeneity, thereby rendering mutation detection within tumor margins impractical. Recent advances in the Mathies Lab, beginning with the development of the 96-lane microfabricated capillary array electrophoresis bioprocessor (Figure 1), have laid the foundation for overcoming these obstacles. The device features 96 electrophoretic separation channels, each 200 mm wide, 30 mm deep, and 17 cm long, arranged radially on a 6-inch glass wafer and converging on a common anode. Sample injection is performed electrokinetically via 48 cross-injectors, with two adjacent lanes sharing common cathode and waste wells. In initial tests, the 96-lane chip was capable of producing 1.7 kbp/min of high-quality sequencing data, with a total separation time under 30 minutes [3]. A more recent innovation of the Mathies Lab is the comparative sequencing technique known as Polymorphism Ratio Sequencing (PRS), whereby two DNA samples are sequenced using a modified dideoxy-termination technique with four universal energy-transfer dye-labeled primers. Reaction mixtures are pooled to allow simultaneous sequencing of two bases from each DNA sample in a single capillary. Polymorphic sites are elucidated as peaks in a squared difference plot for each base sequenced (Figure 2). Correlation of all four traces provides a fingerprint by which the specific nature of the polymorphism can be identified. Performing sample and reference separations in the same lane yields an inherently controlled data set and allows for the quantitation of genetic variation within heterogeneous tissues, with a detection limit of 5% minor allele frequency [4]. My initial experiments have been directed toward the optimization and streamlining of PRS protocols on the 96-lane chip using standard mtDNA samples. A rigorous optimization of injection voltage conditions has yielded consistent, reliable, and high-quality electrophoretic separations across the entire device. Following the development of integrated, user-friendly data analysis software in collaboration with Jing Yi in Terry Speed’s statistics group, we are currently engaged in a pilot study of human lung cancer samples in which paired blood and tumor specimens will be analyzed in search of cancer-specific mutations. Once characteristic variants have been identified, work will commence on the design and validation of a more integrated system capable of performing on-chip DNA extraction, PCR amplification, Sanger sequencing, and capillary electrophoresis. Our final goal is to develop a portable instrument for clinical applications both in the diagnosis stage and in the operating room, where analysis of tumor margins could provide molecular pathology to monitor tumor excision in real time. . Figure 1: Layout of the 96-lane Microfabricated Capillary Array Electrophoresis Bioprocessor Figure 2: Typical PRS “A” trace from Hypervariable Segment 2 of standard mtDNA samples. Verified polymorphic sites at base positions 185, 188, and 228 are clearly visible both in the overlaid traces (below) and as peaks in the squared difference plot (above).

References and Publications Fliss, M.S.; Usadel, H.; Caballero, O.L.; Wu, L.; Buta, M.: Eleff, S.M.; Jen, J.; Sidransky, D. Science. 2000, 287, 2017-2019. Maitra, A.; Cohen, Y.; Gillespie, S.E.D.; Mambo, E.; Fukushima, N.; Hoque, M.O.; Shah, N.; Goggins, M; Califano, J.; Sidransky, D.; Chakravarti, A. Genome Research. 2004, 14, 812-819. Paegel, B. M.; Emrich, C. A.; Wedemayer, G. J.; Scherer, J. R.; Mathies, R. A. Proc. Natl. Acad. Sci. U. S. A.. 2002, 99, 574-579. Blazej R.G.; Paegel B.M.; Mathies R.A. Genome Research. 2003, 13, 287-293.