20. New Technologies for Genome Sequencing and Expression Analysis

Wayne P. Rindone, John Aach, Martha Bulyk, George M. Church, Jason Hughes, Abby Mcguire, Pam Ralston, Martin Steffen, and Saeed Tavazoie

Department of Genetics, Harvard Medical School, Boston, MA 02115

wrindone@arep.med.harvard.edu

We pursue new, basic technologies for collecting low-cost, quantitative molecular data, useful for a variety of sequencing and/or functional genomics goals. We have developed a set of new methods for producing microarrays of Acrydite-immobilized PCR colonies and subsequent Fluorescent In Situ Sequencing (FISSEQ) on the slides. We have integrated mRNA expression data from SAGE, Chip, and microarray experiments into the ExpressDB database and developed a flexible Web-based search engine for exploring the data for clustering and DNA-motif analysis. The process also lead to insights into methodological improvements that would make such information more amenable to comparison among other laboratories.

Nucleic Acids Res 27(24):e34;pp. 1-6

Nature Genetics 22:281-5

http://arep.med.harvard.edu/

21. High-Performance DNA Sequencing and Analysis

Richard A. Mathies

Department of Chemistry, University of California, Berkeley, CA 94720

rich@zinc.cchem.berkeley.edu

Capillary array electrophoresis (CAE) systems¹ coupled with high sensitivity detection provided by energy-transfer labeling reagents² is now the accepted standard for high-throughput DNA sequencing facilities. Further advances are focused on the development of capillary array systems capable of running more than 96 capillaries as illustrated by Scherer et al.³ and the development of microfabricated CAE systems that provide higher throughput as well as the important ability to integrate microfluidic chemistries. Toward this end we have recently shown that microfabricated CE channels only 7-cm long can produce >500 bp 4-color sequencing reads in under 30 minutes⁴.

Microfabrication also permits the production of very high density electrophoretic analysis devices that provide unprecedented analysis throughput. Radial microplates coupled with a novel rotary confocal scanning system have been developed that can rapidly analyze 96 genotyping samples in parallel in seconds on 5-cm long channels⁵. To integrate arrays of 10-15 cm long channels on a radial microplate, it is necessary to devise ways to fold channels without reducing resolution. We have determined that "pinched turns", where the channel width is reduced before the turn and widened after the turn, enable folded designs that minimize turn-induced broadening while maintaining facile matrix introduction⁶.

Sequencing results on radial microplates with 96 15-cm long channels fabricated on a 6"-diameter wafer will be presented. We have also developed microfabrication methods for the integration of nanoliter volume PCR sample preparation directly with microfabricated CE analysis systems⁷.

Genotyping results will be presented along with plans for integrated thermal cycling devices. The implementation of microfabricated separation systems with integrated chemistries will be the next paradigm shift in DNA sequencing and genomic analysis.

• Kheterpal, I. and Mathies, R. A. Capillary Array Electrophoresis DNA Sequencing, Analytical Chemistry, 71, 31A-37A (1999).
• Xie, J., Hung, S.-C., Glazer, A. N. and Mathies, R. A. Energy Transfer Fluorescent Labels for DNA Sequencing and Analysis, in Topics in Fluorescence Spectroscopy, Volume 7: ANA Technology, in press (1999).
• Scherer, J. R., Kheterpal, I., Radhakrishnan, A., Ja, W. W. and Mathies, R. A. Ultra-High Throughput Rotary Capillary Array Electrophoresis Scanner for Fluorescent DNA Sequencing and Analysis, Electrophoresis 20, 1508-1517 (1999).
• Liu, S., Shi, Y., Ja, W. W. and Mathies, R. A. Optimization of High-Speed DNA Sequencing on Microfabricated Capillary Electrophoresis Channels, Anal. Chem. 71, 566-573 (1999).
• Shi, Y., Simpson, P., Scherer, J. R., Wexler, D., Skibola, C., Smith, M. T. and Mathies, R. A. Radial Capillary Array Electrophoresis Microplate and Scanner for High-Performance Nucleic Acid Analysis, Anal. Chem. 71, 5354-5361 (1999).
• Paegel, B. M., Hutt, L. D., Simpson, P. C. and Mathies, R. A. Turn Geometries for Minimizing Band Broadening in Microfabricated Capillary Electrophoresis Channels, Analytical Chemistry, submitted.
• Lagally, E., Simpson, P. C. and Mathies, R. A. Monolithic Integrated Microfluidic DNA Amplification and Capillary Electrophoresis System, Sensors and Actuators B, submitted (1999).

22. Radial Capillary Array Electrophoresis Microplate and Scanner for High-Performance DNA Sequencing and Analysis

Yining Shi¹, Brian M. Paegel¹, James R. Scherer¹, Peter C. Simpson¹, David Wexler¹, Christine Skibola², Martyn T. Smith², and Richard A. Mathies¹

¹Department of Chemistry and ²School of Public Health, University of California, Berkeley, CA 94720

yining@zinc.cchem.berkeley.edu

The design, fabrication and operation of a radial capillary array electrophoresis (CAE) microplate and scanner for high-throughput DNA analysis are presented¹. The microplate consists of a central common anode reservoir coupled to 96 microfabricated separation channels connected to sample injectors on the perimeter of the wafer. Detection is accomplished by a laser-excited rotary confocal scanner with four-color detection. Loading of 96 samples in parallel is achieved using a pressurized capillary array system. High-quality separations of 96 pBR322 restriction digest samples are achieved in <120 s using a 4"-diameter microplate. The practical utility and multicolor detection capability of this system is demonstrated by analyzing 96 methylenetetrahydrofolate reductase (MTHFR) alleles in parallel using a non-covalent 2-color staining method. This work establishes the feasibility of high-performance genotyping with capillary array electrophoresis microplates.

To explore the capabilities of our radial CAE microplate and scanner for high-speed and high-throughput DNA sequencing, we have designed and fabricated a CAE microplate containing 96 folded 12-cm-long separation channels on a 6"-diameter wafer. While high-quality four-color sequencing separations can be achieved on 7-cm-long straight microchannels², integration of an array of such straight channels into a 6"-diameter wafer in the radial format is difficult due to the dimensional restrictions of the wafer. To address this issue, Paegel et al. introduced an optimized channel design which allows the fabrication of 96 folded separation channels on a 6"-diameter wafer. Each of the 96 folded 12-cm-long channels has two complementary tapered turns that minimize turn-induced band broadening during electrophoresis separations^3,4. Four-color sequencing separations and automatic base-calling analyses of 96 single stranded M13mp18 DNA sequencing samples with our 6"-diameter radial CAE microplate and scanner will be presented.

• Y. Shi, P. C. Simpson, J. R. Scherer, D. Wexler, C. Skibola, M. T. Smith and R. A. Mathies. Anal. Chem. 1999, 71, 5354-5361.
• S. Liu, Y. Shi, W. W. Ja and R. A. Mathies. Anal. Chem. 1999, 71, 566-573.
• B. M. Paegel, L. D. Hutt, P. C. Simpson and R. A. Mathies. Anal. Chem. (submitted).
• See abstract "Turn Geometries for Minimizing Band Broadening in Microfabricated Capillary Electrophoresis Channels," by B. M. Paegel, L. D. Hutt, P. C. Simpson, and R. A. Mathies.

23. Turn Geometries for Minimizing Band Broadening in Microfabricated Capillary Electrophoresis Channels

Brian M. Paegel, Lester D. Hutt, Peter C. Simpson, and Richard A. Mathies

Department of Chemistry, University of California, Berkeley, CA 94720

brian@zinc.cchem.berkeley.edu

Microfabricated capillary electrophoresis (CE) devices have dramatically increased the speed and performance of chemical and biochemical analyses¹. As larger numbers of microfabricated structures are placed on a wafer to form capillary array electrophoresis microplates², or as one attempts to fabricate longer channels to enhance resolution in sequencing applications³, it becomes necessary to fold channels. Folding CE channels gives rise to turn-induced band broadening due to dispersion forces in the turn. To circumvent this limitation, microfabricated channels were constructed with a variety of turn geometries for the purpose of minimizing turn-induced band broadening. Column efficiencies of channels with a variety of turn designs were determined by quantitating the resolution of separations of HaeIII digests of phiX174 bacteriophage DNA. Most advantageously, tapered turns were created by narrowing the channel width at the start of a turn to reduce the channel width, followed by widening the channel at the end of the turn. The radius of curvature of the turn, the length of the tapered region, and the degree of tapering were explored. These experiments were performed using our novel rotary scanner⁴, which permits the simultaneous interrogation of a separation at three or more points along a serpentine channel. Serpentine channels were monitored before the first turn, after one turn, and after two turns. Turns with a minimum radius of curvature (250 µm), a minimum length of taper (55 µm), and a maximum tapering ratio (4:1 separation channel width to turn channel width) were found to provide the highest number of theoretical plates for the 271 and 281 base pair fragments of the phiX174 HaeIII ladder. The optimal turn configuration was then used to perform M13 DNA sequencing separations with an effective separation length of 15 cm. High-quality separations to 800 bp were observed in only 35 minutes. These extended length channel designs have been incorporated in high-throughput 96-channel microplates for DNA sequencing⁵.

• Simpson, P. C., Woolley, A. T. and Mathies, R. A. J. Biomedical Microdevices 1998, 1, 7-26.
• Simpson, P. C., Roach, D., Woolley, A. T., Thorsen, T., Johnston, R., Sensabaugh, G. F. and Mathies, R. A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 2256-2261.
• Liu, S. R., Shi, Y., Ja, W. W. and Mathies, R. A. Anal. Chem. 1999, 71, 566-573.
• Shi, Y., Simpson, P. C., Scherer, J. R., Wexler, D., Skibola, C., Smith, M.T. and Mathies, R.A. Anal. Chem. 1999, 71, 5354-5361.
• See abstract entitled "Radial Capillary Electrophoresis Microplate and Scanner for High-Performance DNA Sequencing and Analysis," Y. Shi, B. M. Paegel, J. R. Scherer, D. Wexler, C. Skibola, M. T. Smith and R. A. Mathies.

24. Integrated Microfluidic DNA Amplification and Analysis Systems

Eric T. Lagally¹, Daojing Wang², Charles Emrich², and Richard A. Mathies²

¹UCB/UCSF Joint Graduate Group in Bioengineering and ²Department of Chemistry, University of California, Berkeley, CA 94720

lagally@zinc.cchem.berkeley.edu

Microfabrication technology is an effective method for creating integrated devices for chemical and biochemical analysis^1-3. Our early work in the development of integrated devices included the manufacture of a hybrid Si polymerase chain reaction (PCR) reactor mated with a glass capillary electrophoresis (CE) device⁴ and the development of a CE device with an integrated electrochemical detector⁵. In more recent work, we have developed a fully integrated DNA analysis system microfabricated in glass consisting of controlled fluid delivery using active and passive elements, PCR amplification, and direct coupling to a capillary electrophoretic separation⁶. Samples are introduced at a common sample bus and loaded precisely into a 280-nanoliter volume PCR reactor using valves and hydrophobic vents. The sample is cycled between three temperatures using a resistive heater mounted on the bottom of the chip, and the amplification products are then directly injected and separated on a capillary electrophoresis channel. The device takes only 33 seconds/cycle, representing a vast improvement over conventional thermal cycling systems, which can take up to 5 minutes/cycle. Amplicons from the M13/pUC19 plasmid have been produced from only 20 starting copies/µL or 5 copies in the reactor. This amplification is among the most sensitive compared both to previous static systems, which require ~6,000 starting copies⁷, and to continuous-flow geometries which require as many as ~108 starting copies⁸. The high sensitivity of this device allows studies at the single molecule level.

We have also developed a microfluidic DNA capture chamber for sequencing sample clean-up and concentration. This device uses microfluidic elements to flow raw sequencing samples through a filter chamber filled with oligonucleotide-labeled capture beads. The extension products of interest are selectively captured on the beads and subsequently released using formamide and heat. The capture chamber is directly connected to a capillary electrophoresis channel for immediate sequencing. These results demonstrate a key link in the development of an integrated microfluidic system that performs complete genetic analyses at sub-microliter volumes.

• Simpson, P. C., Roach, D., Woolley, A. T., Thorsen, T., Johnston, R., Sensabaugh, G. F. and Mathies, R. A. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 2256-2261.
• Simpson, P. C., Woolley, A. T., Mathies, R. A. Journal of Biomedical Microdevices 1998, 1, 7-26.
• Liu, S. R., Shi, Y., Ja, W. W. and Mathies, R. A. Anal. Chem. 1999, 71, 566-573.
• Woolley, A. T., Hadley, D., Landre, P., deMello, A. J., Mathies, R. A., et al. Anal. Chem. 1996, 68, 4081-4086.
• Woolley, A. T., Lao, K. Q., Glazer, A. N., Mathies, R. A. Anal. Chem. 1998, 70, 684-688.
• Lagally, E., Simpson, P. C. and Mathies, R. A. Sensors and Actuators B, in press (2000).
• Cheng, J., Shoffner, M. A., Hvichia, G. E., Kricka, L. J. and Wilding, P. Nucleic Acids Res. 1996, 24, 380-385.
• Kopp, M. U., de Mello, A. J. and Manz, A. Science 1998, 280, 1046-1048.

25. High-Speed High-Throughput Mutation Detection

Qiufeng Gao, Ho-Ming Pang, and Edward S. Yeung

Ames Laboratory, Iowa State University, Ames, IA 50011

yeung@ameslab.gov

Single-nucleotide polymorphism (SNP) detection has been the focus of much attention recently. Although many methods have been reported, low-cost, high-throughput and high-detection-rate methods are still in demand. We present a fast and reliable mutation detection scheme based on temperature-gradient capillary electrophoresis. A large temperature gradient (10 °C) was applied with a precision of 0.02 °C and a temperature ramp of 0.7 °C/min. Multiple unlabeled samples from PCR reaction were injected and analyzed. Ethidium bromide was used as the intercalating dye for laser-induced fluorescence detection. The mutations were identified by comparing the electrophoretic patterns of the heteroduplex with that of a homoduplex reference without prior knowledge of the DNA sequence. Mutations in all five test samples were successfully detected with high confidence. This scheme is demonstrated in 96-capillary array electrophoresis for screening single-point polymorphism in large numbers of samples.

26. Micro-Fabricated Devices for Concentrating DNA by Induced-Dipole Trapping

Charles Asbury and Ger Van Den Engh

Department of Molecular Biotechnology, University of Washington, Seattle WA 98195-7730

engh@biotech.washington.edu

DNA molecules placed in a divergent electrical field experience an attractive force towards regions of higher field strength. This force is a result of a charge-dipole that is induced along the molecule's axis. The dipole interacts with the field gradient. Because induced dipoles always oppose the field, the attractive force is independent of the fields polarity. Migration due to dipole forces can be observed with both AC and DC fields. In contrast, electrophoretic forces, which are due to the native charge of DNA molecules, always move the molecules towards the positive electrode. Homogeneous oscillating fields with a 50% duty cycle do not cause a net displacement of DNA. In oscillating fields with steep gradients the molecules move towards the field's origin.

We are developing small chambers for manipulating DNA that allow independent application of both electrophoretic and induced-dipole forces. The devices consist of thin metal layers on a quartz substrate. By combining the two types of forces, cohorts of DNA molecules can be concentrated and moved with high precision. Dipole traps concentrate DNA out of a dilute solution. Electrophoretic forces can then be employed to move DNA cohorts between traps.

We are seeking the optimal conditions for dipole trapping of DNA. The ionic composition of the medium and the frequency, strength, and gradient, of the field are important. We current use a salt concentration below 10 mM and use fields oscillating between 30-100 Hz. By comparing the rate of Brownian movement and diffusion the trapping forces can be quantitated. We are using this information to develop devices in which DNA can be separated by size without the use of a sieving medium. We will present a gold-on-quartz device that consists of a capillary lined with dipole traps. Such capillaries can be combined with other modules to perform complex operation of small cohorts of DNA molecules.

27. Fully Automated Multiplexed Capillary Systems for DNA Sample Analysis

Qingbo Li, Thomas E. Kane, Changsheng Liu, Harry Zhao, Gary W. Loge, John Kernan, Songsan Zhou, Kevin Levan, Heidi Monroe, and David Fisk

SpectruMedix Corporation, 2124 Old Gatesburg Road, State College, PA 16803

qbli@spectrumedix.com

SpectruMedix has developed a commercial 96-capillary electrophoresis instrument for DNA analysis. All operation steps are automated, including capillary conditioning, gel filling, sample introduction, electrophoresis, and data acquisition. It can perform seven consecutive runs without human intervention. Simple yet highly efficient optical design renders an extremely robust detection system that shows excellent stability. The instrument uses a CCD detector to simultaneously record fluorescence signal from all 96 capillaries with on-column laser excitation. Multi-wavelength detection is implemented with a miniaturized spectrometer utilizing a transmission grating. A replaceable linear-polymer matrix provides high-performance separation for DNA sequencing and genotyping. Currently, the instrument is capable of routinely separating 500+ bases with 98% basecalling accuracy in a 2-hr run. By using a gel matrix that is optimized for longer read, up to 770 bp separation with 98% basecalling accuracy has been achieved in a 3-hr run including capillary conditioning, gel filling, sample introduction, and electrophoresis. The instrument includes an electronic unit that allows monitoring the electrophoresis current for each of the 96 capillaries. Further, a recently developed algorithm allows automated color deconvolution matrix file construction, avoiding the need for a calibration run. This process further enhances the robustness of the instrument. A comparison of the SCE9600 performance to an ABI377 using the same sample has been performed.

A prototype 384-capillary array electrophoresis instrument has been developed for higher throughput analysis. The 384-capillary instrument design is based on the SCE9600 platform, so the 96-capillary instrument can be readily upgraded to obtain 4X higher throughput. The injection end of the 384-capillary array is configured in 16x24 format so that it is compatible with commercial 384-well microtiter tray technology. The instrument is capable of performing one genotyping or sequencing run within 2 hours. In fully automated mode, the instrument will analyze 4,608 DNA samples in a 24-hour day.

The instrument includes an electronic unit that allows monitoring the electrophoresis current for each of the 96 capillaries. This has proved to be a valuable technique for protocol diagnosis and development. Automated basecalling software analyzes a set of 96-capillary data within minutes. Further, a recently developed algorithm allows automated color deconvolution matrix file construction, avoiding the need for a calibration run. This process further enhances the robustness of the instrument since instrumental drift, if there is any, is automatically corrected by the algorithm. A comparison of the sequencing performance of a SCE9600 to the performance of an ABI377 using the same sample has been performed.

A prototype 384-capillary array electrophoresis instrument has been developed for higher throughput analysis. The 384-capillary instrument design is based on the SCE9600 platform, so the 96-capillary instrument can be readily upgraded to obtain 4X higher throughput. The injection end of the 384-capillary array is configured in 16x24 format so that it is compatible with commercial 384-well microtiter tray technology. The instrument is capable of performing one genotyping or sequencing run within 2 hours. In fully automated mode, the instrument will analyze 4,608 DNA samples in a 24-hour day.

28. Development and Evaluation of a PCR-Based Sequencing Routine for Use on the ABI 3700 Capillary Machine

Lynne Goodwin, Owatha Tatum, Olga Chertkov, Judith Cohn, and P. Scott White

Bioscience Division and DOE Joint Genome Institute, Los Alamos National Laboratory, Los Alamos, NM 87545

swhite@telomere.lanl.gov

The rapid throughput of the new generation of capillary electrophoresis instruments for automated DNA sequencing presents unique problems for high throughput applications. Increased demands for sequencing substrate of high quality place strains on template preparation routines and equipment, and alternative strategies that scale well and cost less are needed. We have implemented a PCR-based strategy for creating sequencing template that requires only a few, easily automated steps, and costs a fraction of commercial robotic plasmid prep protocols. The template production process begins with overnight culture of subclones, followed by pin-stamp inoculation of PCR plates (96 or 384-well), and finishes with an enzymatic treatment of the PCR products to provide suitable sequencing template. Forward and reverse sequencing reactions are then performed directly on these templates using vector primers, followed by a single isopropanol precipi-tation, and resuspension in 1/10 TE. The 1/10 TE resuspension buffer allows samples to remain on the capillary instrument at ambient temperature for extended periods of time, which is necessary to obtain essentially hands-off automated sequencing of 6 to 8 96-well plates in a 24 hr period.

Concerns about the quality of sequence derived from PCR templates generally focus on homopolymer and simple repeat stretches that are difficult to accurately reproduce using Taq DNA polymerase. In addition, the number of failed reactions and the read length of PCR template-generated sequence is thought to be less than sequence generated from highly purified plasmid DNA. If the costs are favorable and the data of sufficient quality, then such a template production strategy is attractive; otherwise, the higher up front cost of plasmid template preparation is money well spent, as long as throughput is not limited. We have examined the quality of data obtained from an ABI 3700 using this process, and directly compared subclone sequence generated from plasmid templates on ABI 377 instruments. The results will be presented, as will a discussion of the implementation of this process.

29. Rapid and Accurate Detection of Human Functional SNPs Using a Base Stacking Microelectronic DNA Chip

Glen Evans¹, David Canter², Purita Ramos¹, Ray Radtkey², Ron Sosnowski², Gene Tu², James O'Connell², and Michael Nerenberg²

¹Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, TX and ²Department of Molecular Biology, Nanogen, Inc., San Diego, CA

gaevans@home.com

Large scale genomic sequencing is revealing thousands of useful functional and non-functional single nucleotide polymorphisms (SNPs). Technology for the accurate, rapid and expandable assessment of large numbers of human SNPs in parallel is needed for research and medical applications. We describe a novel technology for SNP assessment on microelectronic silicon-based DNA chips that utilizes short fluorescently-labeled oligonucleotide reporters and targets of amplified source DNA. This assay takes advantage of base stacking energies in the design of probes and has the ability of accommodate different sized amplicons in parallel. This assay has been utilized to assay functional SNPs in parallel controlle by electronic fields induced on the chip surface and two-color fluorescence detection. A panel of model functional SNPs has been developed to evaluate the utilize of this method. These markers include polymorphic sites in genes for HH (hemochromatosis), Factor V, EH1 (epoxide hydrolase 1), EPHX2 (epoxide hydrolase 2) CYP19 (cytochrome P450), DTD (diastrophic dysplasia sulfate transporter), GSTA1 (alpha glutathione S transferase), GSTA12 (microsomal glutathione S transferase), NAT1 (N-acetyl transferase 1), NAT2 (N-acetyl transferase 2), ColA1 (type IV collagen), ApoCIII (apolipoprotein cIII), MGC24 (PNA-binding glycoprotein), PPP2R1B (lung cancer susceptibility polymorphism) and others. SNP detection from amplified genomic target DNA can be carried out on 100 SNPs on a single microelectronic chip with accuracy exceeding that of DNA sequencing. We have utilized this system for the systematic genotyping of more than 200 individuals for 15 SNPs, also determined by DNA sequencing, with virtually 100% accuracy. This system is imminently suited to point-of-care genetic diagnosis, forensic applications, medical diagnostics as well as large scale human genotyping for pharmacogenomics applications.

30. DNA Sequencing by Single Molecule Detection

Peter M. Goodwin, Hong Cai, James H. Werner, James H. Jett, and Richard A. Keller

Bioscience Division, Los Alamos National Laboratory, M888, Los Alamos, NM 87545 USA

pmg@lanl.gov

We are developing a method, based upon single fluorescent molecule detection, to sequence individual DNA strands. The method consists of: (1) polymerase incorporation of fluorophore-labeled nucleotides into a strand of DNA complementary to the target sequence; (2) anchoring a single fragment of fluorescently-labeled DNA, in flow, upstream of the detection volume of an ultrasensitive fluorescence flow cytometer; (3) exonuclease digestion of the free end of the anchored DNA strand to sequentially release single,fluorophore-labeled nucleotides into the flow stream; and (4) detection and identification of the individual, released fluorophore-labeled nucleotides in the order of exonuclease cleavage. We have made considerable progress towards a demonstration of single molecule DNA sequencing. Up to three of the nucleotide types, labeled with fluorophores, have been incorporated into strands of DNA 2-7 kilobases in length. Multiple strands of fluorescently-labeled DNA have been attached to microspheres and individual microspheres have been anchored in flow upstream of the detection volume. Exonuclease cleavage of fluorescently-labeled DNA on individual microspheres anchored in flow has been observed. We have detected and identified single, tetramethylrhodamine-labeled dUMPs and Rhodamine-6G-labeled dCMPs enzymatically released from DNA strands containing both types of labeled nucleotides. The two fluorescent species were identified by correlated measurements of single molecule fluorescence burst intensity and intra-burst fluorescence lifetime. We present preliminary data demonstrating the detection of single, labeled nucleotides released by the processive exonuclease digestion of a single DNA strand.

This work was supported by the US Department of Energy, Office of Biological and Environmental Research.

31. New Optical Methods for Sequencing Individual Molecules of DNA

Jonas Korlach, Michael Levene, Stephen W. Turner, Mathieu Foquet, Harold G. Craighead, and Watt W. Webb

Applied & Engineering Physics, Clark Hall, Cornell University, Ithaca, NY 14853

jk109@cornell.edu

A new method for determining the base pair sequence of a single molecule of DNA by following the dynamical stepwise activity of DNA polymerase synthesizing the complementary strand of a given template strand is under development. The technical challenges consist in the development of suitable enzymatic systems and in the recognition of individual sequential base additions. Replacing spatial resolution of bases in the DNA by temporal resolution of sequential nucleotide additions is made possible by using near-field and multiphoton laser optics for chromophore processing, and time-resolved photon counting for detection. Confinement of the excitation volume far below the diffraction limit by nanostructured devices permits an increase of substrate concentrations by about three orders of magnitude above the nanomolar range (required for the enzymatic systems under study), but still allowing sequential single molecule recordings and analysis.

The approach should enable the creation of a very fast sequencing protocol with long read lengths, and potentially highly parallel, integrated systems with large throughput. Each development step towards the sequencing goal appears fertile for the generation and improvement of analytic research systems capable of following biochemical and molecular biological processes (e.g., enzymatic activities) at the single molecule level. The optical tools will enable a characterization of these processes previously unattainable by conventional biochemical analysis. Specifically in respect to the sequencing proposal, this amounts to new knowledge of the photophysical and dynamical behavior of single DNA molecules, the generation and use of new fluorescent labels that can be incorporated into DNA in high densities, and the study of enzymes acting on DNA at the level of individual bases.

32. High Throughput Multiplexed mtDNA SNP Scoring Using Microsphere-Based Flow Cytometry

P. Scott White, Alina Deshpande, Lance Green, Yolanda Valdez, David C. Torney, and John P. Nolan

Bioscience Division and DOE Joint Genome Institute, Los Alamos National Laboratory, Los Alamos, NM 87545

white_paul_scott@lanl.gov

We have developed a flow cytometry-based minisequencing platform capable of extremely high-throughput, low cost assays. These are no-wash assays analyzed at less than 1 minute/sample, with superior sensitivities. Furthermore, with commercially available multiplexed microspheres we can score dozens to hundreds of SNPs simultaneously. Multiplexing, coupled with high throughput rates, makes it possible to score several million SNPs/day at costs that are a fraction of competing technologies. In addition, these assays can easily be integrated into conventional liquid handling automation systems, and require no unique instrumentation for setup and analysis.

Multiplexing is enhanced by universal capture tags consisting of carefully designed, unique DNA tails incorporated into each minisequencing primer. These are complementary to address tags attached to discrete populations of microspheres in multiplexed sets. This enables simultaneous minisequencing of many SNPs in solution, followed by capture onto the appropriate microspheres for multiplexed analysis by flow cytometry. High signal-to-noise ratios, ease of setup, flexibility in format and scale, and low cost of these assays make them versatile and valuable tools for studies at a wide range of scales where SNP scoring is needed.

A typing method that could rapidly and inexpensively score 100 or more SNPs located in the mitochondrial genome would be extremely valuable in population and evolutionary biological studies, as well as a powerful forensics tool. We present results from multiplexed analyses of mtDNA and HLA SNPs, performed on a few large PCR amplicons, each containing numerous SNPs that have been scored simultaneously.

33. Mass Spectrometric Analysis of Genetic Variations

Lloyd M. Smith

Department of Chemistry, University of Wisconsin-Madison, Madison, WI

smith@chem.wisc.edu

In the last decade two powerful new tools for the mass spectrometric analysis of biomolecules have been developed, Matrix-Assisted Laser Desorption Mass Spectrometry (MALDI-MS), and Electrospray Ionization Mass Spectrometry (ESI-MS). The power of these methods lies in their ability to produce and mass analyze intact gas phase ions from very large molecules such as proteins and nucleic acids. The speed, accuracy, and sensitivity of the technologies make them well-suited to address a number of problems in genetic analysis, including the analysis of DNA sequence, genetic variations, and gene expression. Results in these areas will be presented, including recent work in which single nucleotide polymorphisms (SNPs) in genomic DNA may be analyzed without need for a prior PCR amplification step.

34. Affinity Capture and Mass Spectrometry of Targeted Proteins in Mice

Stephen J. Kennel¹, Gregory B. Hurst², Linda J. Foote¹, and Michelle V. Buchanan^1,2

¹Life Sciences Division and ²Chemical and Analytical Sciences Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6124

buchananmv@ornl.gov

We are developing new mass spectrometry-based methods for large-scale screening of targeted proteins in mice, taking advantage of the ability of mass spectrometry to detect compounds sensitively and to identify both normal and modified proteins unambiguously. In this initial study, we are working with the Mammalian Genetics and Development Section at ORNL to develop a method for large-scale screening of cytokine levels in mouse serum samples as a means to detect subtle abnormalities leading to chronic inflammatory diseases. Modified levels of cytokines in serum are indicative of inflammation, a condition associated with a wide variety of disease states. The new methodology involves capture of targeted cytokines from mouse serum onto antibody-derivatized aminopolystyrene beads, washing, elution, and analysis by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). The selectivity of the affinity separation is complemented by the m/z measurement capability of mass spectrometry, providing speed and specificity advantages over conventional ELISA techniques. Tumor necrosis factor a (TNF-a), a 17 kDa proinflammatory cytokine that has a relatively high serum concentration in acute reactions and seems to be an early effector in inflammatory cytokine cascades, has been used as a model analyte (Hurst, G.B.; Kennel, S.J.; Foote L.J.; Buchanan, M.V. Anal. Chem. 1999, 71, 4727-4733). In parallel with MALDI-MS, experiments using 125I-labeled TNF-a and gamma detection allow independent optimization of the affinity capture, and indicate that the capture methodology is viable from <100 pg/mL to >50 ng/mL. MALDI-MS currently allows reliable detection down to 1 ng TNF-a in an initial 100 mL sample volume (mouse limited), and we are working to improve this figure. To demonstrate selectivity, TNF-a spiked into mouse serum can be concentrated onto the beads and detected by MALDI-MS with little interference from the many other components present in serum. Control experiments indicate that non-specific binding is minor. Preliminary MALDI-MS results on other cytokines (IL-6, IL-1b, IL-2, and IFN-g) indicate that the MALDI matrix conditions must be carefully optimized for each cytokine to allow sensitive detection.

Research sponsored by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by Lockheed Martin Energy Research Corp. for the U. S. Department of Energy under Contract No. DE-AC05-96OR22464.

35. Rapid Quantitative Measurements of Proteomes

Richard D. Smith, Ljiljana Pasa Tolic, Mary S. Lipton, Pamela K. Jensen, Gordon A. Anderson, and Timothy D. Veenstra

Pacific Northwest National Laboratory, Richland, WA 99352

rd_smith@pnl.gov

The patterns of gene expression, protein post-translational modifications, covalent and non-covalent associations, and how these may be affected by changes in the environment, cannot be accurately predicted from DNA sequences. In addition, direct protein measurements now constitute the most effective method for determining open reading frames for small proteins. Therefore, proteome characterization is increasingly viewed as a necessary complement to complete sequencing of the genome. Approaches for proteome characterization are increasingly based upon mass spectrometric analysis of in-gel digested electrophoretically separated proteins, allowing relatively rapid protein identification compared to conventional approaches. However, this technique remains constrained by the speed of the 2-D gel separations, the sensitivity needed for protein visualization, the speed and sensitivity of subsequent mass spectrometric analyses for identification, and the limitations of spot visualization for quantitation.

Our objective is to circumvent the limitations of this approach by directly characterizing the cell's polypeptide constituents by combining fast separations and the mass accuracy and sensitivty obtainable with Fourier transform ion cyclotron resonance (FTICR) mass spectrometry. Several approaches are presently being pursued; one based upon the analysis of intact proteins and the second upon global approaches for protein digestion and accurate peptide mass analysis (i.e. the use of "accurate nmass tags"). A key attraction of FTICR is the enhanced facility for protein identification based upon the use of genome sequence data. Alternative versions of proteomes using stable isotope labeling are applied for the purposes of accurate quantitation. We describe the status of our efforts towards the development of a high throughput proteomics capability.

We thank the Office of Biological and Environmental Research, U. S. Department of Energy, for support of this research under contract DE-AC06-76RLO 1830.

36. DNA Characterization by Electrospray Ionization-FTICR Mass Spectrometry

David S. Wunschel, Bingbing Feng, Ljiljana Pasa Tolic, Mary S. Lipton, and Richard D. Smith

Pacific Northwest National Laboratory, Richland, WA 99352

rd_smith@pnl.gov

Mass spectrometry offers the potential for high speed DNA sequencing and ultra-sensitive characterization. Ongoing work in the laboratory is exploring approaches based upon electrospray ionization (ESI) and/or Fourier transform ion cyclotron resonance (FTICR) mass spectrometry. These efforts have included advanced methods for the characterization of polymerase chain reaction (PCR) products¹, enzymatically produced oligonucleotide mixtures, modified DNA and the development of methods for the analysis of DNA large fragments. High mass accuracy measurements for PCR products allowing a single base substitutions to be detected at >250 bp level with de novo identification of an unreported base substitution. This capability also allows the identification of small differences in mass such as those arising from methylation². Study of DNA damage/modifications in their sequence context will likely have to occur from within multi-component mixtures. The capability for this has been demonstrated using a multi-component reaction where a base pair deletion was identified with the putative identification of inter-operon variability within a single bacterial strain. These efforts are also being extended to exploit the non-destructive nature of FTICR for recovery (i.e., "soft-landing") of mass-selected modified DNA segments, following high resolution FTICR analysis and separation (i.e., high resolution sorting), for subsequent cloning or PCR³. This capability allows for the direct selection and analysis of individual components from within mixtures. Alternatively, DNA species that cannot be identified through traditional sequencing methodologies, those containing base modifications, can be isolated and the nature and position of the modification identified. Most importantly this provides an approach for identification of low abundance modifications where few if any alternatives for their detection exist.

• D. S. Wunschel, D. C. Muddiman and R. D. Smith, Advances in Mass Spectrometry, Volume 14, E.J. Karjalainen, A.E. Hesso, J.E. Jalonen, U.P. Karjalainen, Eds., Elsevier Science Publishers B.V., Amsterdam, 377-406 (1998).
• D. S. Wunschel, L. Pasa Tolic, B. Feng and R. D. Smith, J. Amer. Soc. Mass Spectrom., in press.
• B. Feng, D. S. Wunschel, C. D. Masselon, L. Pasa-Tolic and R. D. Smith, J. Amer. Chem Soc., 121, 8961-8962 (1999).

37. DNA Sequencing via Electrospray and Ion/Ion Chemistry in an Electrodynamic Ion Trap

Scott A. McLuckey¹, James L. Stephenson, Jr.², and Gregory B. Hurst²

¹Department of Chemistry, Purdue University, West Lafayette, IN 47907-1393 and ²Chemical and Analytical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831

mcluckey@purdue.edu

We are pursuing a methodology for high speed DNA sequencing based on electrospray ionization mass spectrometry employing gas-phase ion/ion chemistry in a quadrupole ion trap. DNA sequencing via mass spectrometry has been pursued by a number of groups in recent years due to its promise for the obviation of time-consuming electrophoresis-based separations required with established sequencing strategies. By far, most effort has been directed toward matrix-assisted laser desorption ionization (MALDI) combined with time-of-flight mass spectrometry. While a MALDI-based approach may yet fulfill its promise, limitations encountered in ionizing relatively large DNA oligomers have proved to be difficult to overcome. In contrast, ionization of large DNA oligomers is not a limitation for electrospray ionization. However, electrospray-based approaches for high speed DNA sequencing have not been extensively pursued due to spectral congestion associated with the multiple charging phenomenon that is characteristic of electrospray. The formation of multiple charge states from a single oligomer severely limits the mixture complexity amenable to direct analysis via electrospray. For this reason, electrospray usually follows a separation method, such as liquid chromatography or capillary electrophoresis, when applied to mixtures. We have recently shown that gas-phase ion/ion chemistry involving oppositely charged ions within a quadrupole ion trap greatly expands the mixture analysis capability of electrospray. In this work, the idea is to subject Sanger mixtures to electrospray and ion/ion chemistry as a core element in a strategy for high speed DNA sequencing. This talk describes the methodology and progress to date.

38. DNA and Protein Analyses on Microfabricated Devices

R. S. Foote, Y. Khandurina, I. M. Lazar, Y. Liu, T. McKnight, L. C. Waters, S. C. Jacobson, R. S. Ramsey, and J. M. Ramsey

Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6142

footers@ornl.gov

Microfabricated, microfluidic devices are being developed for both nucleic acid and protein analyses. An integrated system for rapid PCR-based analysis on a microchip has been demonstrated. The system couples a compact thermal cycling assembly based on dual Peltier thermoelectric elements with a microchip gel electrophoresis platform. This configuration allows fast (~ 1 min/cycle) and efficient DNA amplification on-chip followed by electro-phoretic sizing and detection on the same chip. On-chip DNA concentration has been incorporated into the system to further reduce analysis time by decreasing the number of thermal cycles required. The concentration-injection scheme enables detection of PCR products after performing as few as 10 thermal cycles with a total analysis time under 20 min, with a starting template copy number of fewer than 15 molecules per injection volume.

Electrophoretic separations of proteins have been carried out on microchips with on-chip, post-column labeling for detection by laser-induced fluorescence. The post-column labeling format avoids peak broadening and loss of resolution due to heterogeneous product formation in prelabeling reactions. Two-dimensional separations of tryptic peptides were demonstrated on microchips that combine micellar electrokinetic chromatography (MEKC) and high-speed capillary electrophoresis (CE). Effluent from the first dimension is sampled onto the second dimension every few seconds and the entire analysis is completed within 10 minutes. Structures that incorporate an electrospray element have also been devised and sub-attomole sensitivity demonstrated for peptide samples on a time-of-flight (TOF) mass analyzer. Proteolytic digestions with trypsin can be performed directly on the chip and the peptide fragments analyzed by TOFMS for protein identification. Tryptic peptides could be generated in less than 10 min for analysis of femtomol or subfemtomol amounts of protein.

39. Stable Isotope Assisted Mass Spectrometry Allows Accurate Determination of Nucleotide Compositions of PCR Products

Xian Chen¹, Zhengdong Fei², Lloyd M. Smith², E. Morton Bradbury^3,4, and Vahid Majidi¹

¹CST-9, Chemical Science and Technology Division and ³B-3, MS M888, Biological Division, Los Alamos National Laboratory, Los Alamos, NM 87544; ²Department of Chemistry, University of Wisconsin-Madison, 1101 University Avenue, Madison, WI 53706-1396; and ⁴Department of Biological Chemistry, School of Medicine, University of California at Davis, Davis, CA 95616

xchen@telomere.lanl.gov

In parallel with the large-scale sequencing effort, the human genome project will need the next generation tools for accurate and efficient analyses of the enormous pool of DNA sequences. Such analyses are required for; (a) validation of DNA sequences; (b) comparison of a parent (known) sequence with a related (unknown) sequence, and (c) characterization of sequence polymorphisms in various genes especially those associated with genetically inherited human diseases. Here, we report a novel method that combines stable isotope 13C/15N-labeling of PCR products of the target sequences with analysis of the mass shifts by mass spectrometry (MS). The mass-shift due to the labeling of a single type of nucleotide (i.e., A, T, G, or C) will reveal the number of that type of nucleotide in a given DNA fragment. Using this technique, we have accurately determined nucleotide compositions of DNA fragments. The method has also been applied to score a known single nucleotide polymorphism. The comparisons of nucleotide compositions determined by our method among homologous sequences are useful in sequence validation, sequence comparison, and characterizations of sequence polymorphisms.

40. Hybridization Detection

Tom J. Whitaker and Kenneth F. Willey

Atom Sciences, Inc., Oak Ridge, TN 37830

whitaker@atom-sci.com

Two projects aimed at developing new techniques that measure DNA hybridization to oligonucleotide (ODN) probes on DNA chips will be described. Although these techniques have similar goals, they differ widely in cost, complexity, sensitivity, and application. One method uses laser-based mass spectrometry analysis of stable isotopes of Sn atoms sputtered from Sn-labeled DNA target molecules. A 10-kV ion beam is used to sputter the sample from the surface and a wavelength-tunable laser efficiently and selectively ionizes the neutral Sn atoms for time-of-flight mass spectral analysis. The extreme sensitivity of this laser ionization technique has previously enabled high spatial resolution measurements of trace impurities in a variety of substrates. The technique also facilitates quantitative measurements by avoiding the effects of variation in secondary ion yield, a problem that plagues SIMS (secondary ion mass spectrometry) analyses. We are currently working with samples supplied by Affymetrix, Inc. to explore application of the technique for quality control of in-situ formation of ODN probes and DNA hybridization.

The second technique uses an inexpensive electronic detection method to determine if hybridization has occurred at a specific probe site. The low cost and small dimensions of this device make it ideal for point-of-care applications. The method takes advantage of the fact that an ODN probe can form an insulating self-assembled monolayer between a gold surface (on which the dielectric is attached) and a conducting liquid. Because of the extremely thin dielectric, the resulting capacitor has a very high specific capacitance. Calculations indicate that the change in effective dielectric thickness caused by the binding of a fraction of a monolayer of DNA to the ODN probe layer will produce a significant, and easily measurable, change in capacitance. Results of initial experiments aimed at verifying this concept will be provided and future experiments to enhance the capacitance change will be described.

41. Automation Using Packard Multiprobe Robots for Finishing

Christine Munk, Judy Buckingham, Marie Krawczyk, Elizabeth Saunders, David Bruce, and Mark Mundt

Bioscience Division and DOE Joint Genome Institute, Los Alamos National Laboratory, Los Alamos, NM 87545

cmunk@lanl.gov

We have recently adopted an automated approach for cherry-picking of subclone DNA to streamline our finishing process. The input for this process is a list of subclones for finishing reactions that is generated using the Phrap .ace file. A script sorts the resequencing reactions by single strand gap size and calculates source and destination for each subclone DNA. The script outputs two text files, one of which is formatted to be imported into an MP Table sample transfer program and contains source and destination locations for the DNA transfer. The second output file is used to make the sample sheet, which is imported into the data collection software on the ABI 377.

For cherry-picking subclones, we initially tried a Packard Multiprobe I equipped with disposable tips to transfer DNA from deepwell cluster tubes. We missed ~20% of the DNAs because the edge near the top of the tips got hung up on the top of the DNA tube, preventing the tip from reaching the DNA at the bottom of the tube. To address this problem, we moved to a fixed-tip robot, and the DNA prep procedure was changed to collect DNA into microtiter plates. The small diameter of the fixed tips allows them to reach the bottom of both the deepwell and microtiter plates. In our current process, we use a fixed-tip Multiprobe I robot with MP Table software to transfer subclone DNAs from round-bottom microtiter plates and to rearray these into cycleplates for sequence reaction setup. We have not experienced problems with DNA cross-contamination using a simple water rinse of the tips between samples. The cost of disposable tips has been eliminated. Use of microtiter plates instead of deepwells has greatly reduced the amount of storage space needed.

42. Automated, Low Cost Isolation of Blood or Bacterial Genomic DNA

Brian Bauman, Tuyen Nguyen, Zuxu Yao, Tony Zucca, Dan Langhoff, and William MacConnell

MacConnell Research Corporation, San Diego, CA 92121

macres@macconnell.com

The isolation of the genomic DNA from blood, bacteria, and virus is a necessary starting point for molecular diagnosis of infection, genetic disease, inherited traits and identity determination, and other research applications. The ability to rapidly and reproducibly isolate DNA from blood and other bodily samples will be continually required to identify, characterize and treat factors involved in human disease and disorders. This process needs to be automated by means that are affordable to clinical or research labs. In Phase II we are further developing prototypes for a fully automatic high-throughput blood or bacteria genomic DNA isolation instrument and disposable processing cassettes. This instrument uses a derivative of electrophoretic separation technology that we developed for purification of plasmid DNA. Proof-of-concept data has been obtained during Phase I. The instrument gives high yields of pure DNA over a wide range of sample concentrations.

The separation technology makes use of programmed electrophoresis of the lysed sample which is placed in between boundaries of agarose separation material. The process can be accomplished using a microprocessor-controlled programmable power supply in conjunction with a multi-sample disposable processing cassette, electrophoresis rig and fluid handling components.

Thus far, our prototype instrument and cassettes allows automatic purification of human blood or bacterial DNA in as little as 24 minutes. The resulting DNA is pure enough for use in restriction digests, and could be used as a template for PCR, PCR sequencing and RFLP analysis. In addition, we found that the process successfully purifies genomic DNA from a wide range of sample cell numbers, such as 104 to 109 bacterial cells or 0.2 to 500 microliters of human blood. Interestingly, the DNA yield from the above trials was greater than 75% of the theoretical amount of genomic DNA present in these samples.

In Phase II we are completing trials to: (1) optimize the electrophoretic purification technique, (2) determine the full range of sample volumes and cell numbers that can be processed, (3) test the activity of purified genomic DNA in a variety of molecular biology procedures, (4) test for cross-contamination between samples purified in the same run, (5) construct a 24 well and 96 well prototype instruments that automate the overall method, (6) write control software and (7) contract for the construction of an injection mold for the production of the processing cassettes. The method is being expanded to process DNA from a variety of sample types including viruses.

The products developed in this work will be commercialized by MacConnell Research in the form of instruments and supplies sold for genomic DNA isolation.

This research is being supported by DOE SBIR Phase II grant number DE-FG03-98ER82612.

43. The Use of Electrode Arrays for the Synthesis of Biomolecular Affinity Probes

Francis Rossi, Christopher Ashfield, Karl Maurer, and Donald Montgomery

CombiMatrix Corporation, 887 Mitten Road, Suite 200, Burlingame, CA 94010

frossi@combimatrix.com

We have developed an active semiconductor chip composed of over 1000 individually addressable electrodes that is used to synthesize microarrays of biomolecular affinity probes. Arrays are prepared by coating the semiconductor chip with a porous polymer support in which synthesis occurs. The underlying electrodes are used to electrochemically generate reagents from inert precursors. By switching on individual electrodes, or patterns of electrodes, reactions can be conducted at defined locations of the chip.

The technology has been successfully used to prepare DNA oligonucleotide probe arrays from commercially available reagents. To synthesize an array, electrodes are biased as anodes at defined locations. This generates acid, which removes the DMT protecting group from the 5'-hydroxyl group of nascent oligonucleotides. Extension of the deprotected hydroxyl group using standard DNA phosphoramidite reagents adds the next base of the oligonucleotide.

We are now extending this technology to the synthesis of peptide probe arrays for the identification and analysis of gene products. Peptides were prepared by first immobilizing an Fmoc-protected amino linker to the porous polymer support. The amino groups were deprotected at the desired locations with an electrochemically generated base. The resulting free amines were reacted with an activated Fmoc-protected amino ester using conventional techniques to give peptides.

44. Development of a High Throughput Peptide Nucleic Acid Synthesizer

J. Shawn Roach¹, Simon Rayner¹, Lynn Mayfield², David R. Corey², and Harold "Skip" Garner¹

¹Center for Biomedical Inventions, Department of Internal Medicine and ²Howard Hughes Medical Institute, Department of Pharmacology and Department of Biochemistry, University of Texas Southwestern Medical Center at Dallas, Dallas, TX

roach@ryburn.swmed.edu

Peptide Nucleic Acids (PNAs) are synthetic analogs of DNA in which the phosphodiester backbone has been replaced with 2-aminoethyl glycine linkages, but maintaining the four natural nucleobases. A PNA strand will bind to DNA with the same sequence complementarity as standard DNA/DNA base paring, but PNA/DNA binding occurs more rapidly and more tightly than DNA/DNA binding. Much research has gone into the potential applications of PNAs as antisense and diagnostic agents.¹ However, a major obstacle in PNA research becoming more widespread has been the high cost of the PNAs. A high throughput PNA synthesizer will afford an economy of scale and reduce the synthetic cost of PNAs. We report the development of a high throughput PNA synthesizer capable of producing up to 192 different PNAs in one synthesis run. The synthesizer is based on high throughput DNA synthesis technology developed at the University of Texas Southwestern Medical Center at Dallas to support the Human Genome Project.² The synthesizer consists of an XY table, a series of valves plumbed to an injection head for reagent delivery, two vacuum chucks for reagent removal and a computer that controls the synthesis procedure. Synthesis is conducted in 96-well fritted filter plates, using standard solid phase Fmoc-PNA synthesis chemistry. The quality of the PNAs produced from the synthesizer is assessed using RP-HPLC and MALDI MS.

• Nielsen, P.E. Curr. Op. Biotech. 1999, 10:71-75.
• Rayner, S., Brignac, S., Bumeister, R., Belosludtsev, Y., Ward, T., Grant, O., O'Brien, K., Evans, G. and Garner, H. Genome Res. 1998, 8, 741-747.

45. MicroArray of Gel Immobilized Compounds on Chip

V. Vasiliskov, A. Stomakhin, B. Strizhkov, S. Tillib, V. Mikhailovich, A. Sobolev, A. Kuhktin, and A. Mirzabekov

Joint Human Genome Program: Biochip Technology Center, Argonne National Laboratory, Argonne, IL 60439 and Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow 117984

Micro Array of Gel Immobilized Compounds on a Chip (MAGIChip^TM) are produced by immobilizing oligonucleotides¹, DNA, enzymes, antibodies, and other proteins² on a photopolymerized micromatrix of polyacrylamide gel pads. Recently, a photoco-polymerization technique was introduced for more rapid and inexpensive manufacturing of the microchips containing gel pads from 10´10´5 m to 100´100´20 m and larger in size³.

MAGIChips are efficient for carrying out direct hybridization tests, as well as for oligonucleotide ligation, single-base extension⁴, and PCR amplification of DNA. The fluorescence microscope has been devised for quantitative and real-time monitoring of hybridization, measuring the thermodynamic parameters of DNA duplexes, and measuring kinetics of enzymatic reaction on MAGIChips. On-chip MALDI-TOF mass spec-trometry was successfully tested for polymorphism analysis of DNA⁵. These technologies are demonstrated for identification of microorganisms, detection of their genes, and screening for mutations. Simple equipment and procedures have been developed for monitoring the hybridization of amplified DNA with oligonucleotide microchips and on-chip amplification. This allows us to carry out fast and inexpensive screening of Mycobacterium tuberculosis drug resistant mutations.

Work supported by the Department of Energy, Office of Health and Environmental Research under Contract No. W-31-109-ENG-38; Cooperative Research and Development Agreement No. 970192 between Argonne National Laboratory, Motorola, and Packard Instruments; Defense Advanced Research Project Agency under Interagency Agreement No. AO-E428; and the Russian Foundation of Fundamental Research under Grant 96-04-49858.

References

1. Yershov, G., Barsky, V., Belgovskiy, A., Kirillov, Eu., Kreindlin, E., Ivanov, I., Parinov, S., Guschin, D., Drobishev, Dubiley, S. & Mirzabekov, A. (1996) Proc. Natl. Acad. Sci. USA 93, 4913-4918.

2. Arenkov, P., Kuhktin, A., Gemmell, A., Chupeeva, V., & Mirzabekov, A. (2000) Anal. Biochem. (in press).

3. A. Vasiliskov, V., Timofeev E., Surzhikov S., Drobyshev, A., Shick, V., & Mirzabekov, A. BioTechnique 27, 592-606

4. Stomakhin, A, Vasiliskov, V., Timofeev, E., Schulga, D., Cotter, R., & Mirzabekov, A. (2000) Nucleic Acids Res. (in press).

147. Molecular Gates for Improved DNA Cleanup and Handling in Microfabricated Devices

Paul W. Bohn, T.C. Kuo, Wenju Feng, Lisa Sloan, and Jonathan V. Sweedler

Department of Chemistry and the Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL 61801

bohn@scs.uiuc.edu

The development of integrated systems capable of automated accurate sequence generation from sample introduction to sequence output is an important goal of the DOE Human Genome Project. Microfabricated DNA analyzers (such as microfabricated PCR systems with integrated CE systems) have been developed that offer a number of important advantages compared to the traditional large scale methods. However, the actual interface between these microfabricated subassemblies can be problematic. For example, to obtain the highest quality sequencing results, sample cleanup is required after PCR reactions but before introducing the sample to a CE separation. We are developing unique integration technology based on "molecular gates." Molecular gates can be thought of as intelligent (externally controllable) adsorption membranes. The successful molecular gate has the ability to "capture" a preselected DNA band on-device after PCR analysis to allow sample cleanup, release the analyte for electrophoretic sequencing and even to capture a particular DNA band eluting from the separation channel for further characterization. This allows easier interfacing between the separate components of a total "lab-on-a-chip" sequencer. In the first year of this project, we have optimized molecular gate technology and are developing the protocols for high efficiency and fast sample capture and release. The ability to have the molecular gate discriminate against small molecules is important, and we are currently able to drive specific analyte classes into and out of the gate structure. The overall device will offer the ability to desalt and cleanup the DNA between the various subassemblies of an integrated DNA system. Current work involves integrating these devices into microfabricated electrophoresis system for improved sample cleanup capabilities.

150. A modular integrated system for high throughput DNA sequencing

Voula Kodoyianni, Ying Ge, Geoffrey K. Krummel, Jessica M. Severin, Michael S. Westphall, Michael T. Borchardt, Laura Grable, Anne S. Olsen² and Lloyd M. Smith

Department of Chemistry, University of Wisconsin-Madison, Madison, WI and ²DOE Joint Genome Institute, Walnut Creek, CA

vkodoyia@facstaff.wisc.edu

We have developed an integrated modular system for high throughput DNA sequencing. Our system has four major components: a) a robotic template preparation module which converts M13 containing bacterial cultures from random subclone libraries to ready-to-sequence DNA b) a robotic module for sequencing either M13 DNA using extension dye primer chemistry or plasmid DNA using Big Dye Terminator chemistry c) a custom made electrophoresis module and d) a software package (GelImager and Basefinder) that converts raw gel images to ready-to-assemble sequence data.

This system is being "hardened", meaning thoroughly debugged in production operation, in the context of a collaborative project with the DOE Joint Genome Institute (JGI) to sequence minimally overlapping, mapped BAC and cosmid clones that cover 1 Mb of H19q13.2 and its syntenic region in mouse. These DNA regions encode a cluster of zinc finger proteins. A detailed comparison of the human gene sequences to their murine orthologs will allow better understanding of the regulation, evolution and functional diversification of these proteins.

Our sequencing strategy can be divided into five stages: a) clone validation; to ensure no deletions have occurred; b) random shotgun library construction; two libraries are made for each BAC/cosmid, a single stranded library in M13 (1.2-2 kb inserts) and a plasmid library in pBC SK+ (2.5-5 kb inserts); c) shotgun sequencing; a 96-well format is used beginning with DNA purification, followed by sequencing of DNA template with dye primer (M13) or dye terminator (plasmids) chemistries,and gel electrophoresis; d) gel data processing (using GelImager and Basefinder) and sequence assembly using phrap; e) finishing the assembled reads to 99.99% accuracy using consed and autofinish and f) annotation before submission to Genbank.

We are currently sequencing 14 human clones (8 cosmids and six BACs) and 4 murine clones ( 4 BACs). Our progress in analyzing these clones will be presented.

DOE - DE-FG02-97ER62386 and NIH - R01HG01886.

153. Implementation and Evaluation of Large Scale, Low Cost Sequencing Technologies at Stanford

Michael J. Proctor, Stanford University

The Stanford DNA Sequencing and Technology (SDSTC) was funded in Jan 1999 to make its high throughput DNA sequencing technology available to the DOE Joint Genome Institute (JGI) in Walnut Creek. The SDSTC has developed a suite of custom robotic instruments for the automation of shotgun sequencing of genomic DNA. The primary focus of this development has been the reduction of the cost of sequencing genomic DNA through the elimination of labor and through increases in the efficiency fo the use of reagents and disposables. The system should be completed by the end of this year. The system will have a cost per sequencing lane of $0.30 and have a throughput of 7000 samples/day. All components of the system will be made available to the JGI. In order to reduce the cost of the instrumentation delivered to JGI, we have joined with an industrial partner (GeneMachines Inc., San Carlos). All manufacturing and technical support of the instrumentation delivered to JGI is handled by GeneMachines. Over the past year, JGI has successfully imported and implemented 3 of our sequencing modules: the point sink shearing device; the plaque/colony picker; and the enhanced oxygen atmosphere shaker. At the SDSTC, we are continuing to implement improvements to these instruments as well as to the M13 prep station and plasmid prep station. These latter two instruments should be ready for export with expanded capabilities during the next year of this grant. The final instrument, the small volume thermal cycler, will be in production implementation at the SDSTC this fall. We will work with the JGI to make this technology available to them over the next year.