Sponsored by the U.S. Department of Energy Human Genome Program
Human Genome News Archive Edition
Human Genome News, September 1991; 3(3)
Joe Gray, Director, Division of Molecular Cytometry, University of California, San Francisco
Two key Human Genome Project goals-creation of a physical map of the genome and determination of genomic DNA sequence-require significant advances in biophysical instrumentation. To meet this need, the DOE Human Genome Program has initiated several instrumentation development efforts, some to support ongoing mapping and sequencing systems and others to explore completely new technologies. Several approaches are summarized below to provide an overview of current initiatives in this rapidly advancing field.
New instruments are beginning to contribute significantly to several important procedures used in physical map assembly:
Identifying contiguous cloned DNA fragments by hybridization usually requires the fragments to be deposited on a membrane to support the hybridization process-typically a rate-limiting step. Other rate-limiting aspects include preparation and replication of ordered clone collections (usually in microtiter trays) and detection of the labeled probe hybridizing to the DNA samples on the membrane.
Several laboratories are working to automate this process. Scientists Joseph Jaklevic and colleagues at Lawrence Berkeley Laboratory (LBL) have developed a robotic system that orders individual colonies into arrays to identify yeast or bacterial colonies carrying cloned DNA sequences; the system then transfers DNA from each arrayed colony onto a hybridization membrane. The group also is developing a system to record and store autoradiographic images showing the hybridized probe locations. A key component of this system is a photostimulable phosphorimaging plate that allows image generation without a film intermediate.
Tony Beugelsdijk and coinvestigators at Los Alamos National Laboratory (LANL) have automated the production of high-density DNA grids on nylon membranes, starting with cloned sequences arrayed in microtiter dishes. The number of separately resolvable DNA samples per unit area is increased 16-fold during transfer of the clones to the membranes. The group has also developed a system for replication of libraries stored in microtiter trays.
This procedure identifies overlapping cloned sequences by analyzing the restriction pattern produced by digestion with one or more restriction endonucleases. Overlapping clones will have several fragments of the same size. This process is limited by the electrophoresis time required for DNA fragment separation.
Anthony Carrano and coworkers at Lawrence Livermore National Laboratory (LLNL) use an Applied Biosystems, Inc. (ABI) electrophoresis system to automate and multiplex the process. In this approach fluorescent labels are ligated (attached) to restriction enzyme-digested cosmids. The fragments are separated electrophoretically and detected using a laser scanner. Multiplex operation is achieved by ligating fluorochromes (that fluoresce at different wavelengths) to separate cosmid digests, mixing these differentially labeled digests together with a standard, and running several in each gel lane.
Barry Karger and coinvestigators at Northeastern University are increasing separation speed by applying high-voltage capillary electrophoresis; they use linear polyacrylamide as the electrophoretic separation medium to achieve reproducible separation of DNA fragments ranging from 50 bp to several kb in 2 to 10 minutes. Norman Dovichi and colleagues at the University of Alberta, Edmonton, Canada, are developing a parallel capillary system that will support electrophoresis of 32 capillaries simultaneously.
The contig (contiguous set of clones) assembly methods described above are limited in that they do not always provide information on contig orientation and location on chromosomes. FISH allows localization of contig elements to metaphase chromosomes and establishment of their orientation.
Barbara Trask and coworkers at LLNL have developed dual-color FISH techniques applicable to metaphase chromosomes and interphase nuclei. Metaphase mapping localizes contig elements to within ~2 Mb along the genome, while interphase mapping allows contig element ordering in the 100-kb to 2-Mb range. Brigitte Brandriff and others at LLNL are developing hybridization to germ cell chromatin for ordering cosmids separated by < 100 kb.
Joe Gray (University of California, San Francisco; formerly at LLNL) and colleagues have developed a semiautomated computer-assisted microscope to facilitate mapping by FISH. This instrument allows simultaneous determination of the chromosomal locations of several differentially labeled clones. The number of simultaneously mapped probes is increased by identifying clones according to fluorescence intensity ratios.
Longer-range projects also are under way to facilitate DNA fragment analysis. Leonard Lerman's group at the Massachusetts Institute of Technology is evaluating the usefulness of thermal stability mapping, which is independent of restriction sites and does not require cloning. This approach, based on hybridization-pattern analysis to DNA fragments separated by two-dimensional denaturing gradient gel electrophoresis, is being tested using human genomic DNA and bacteriophage lambda DNA.
Sequencing the human genome and genomes of important model organisms is likely to require sequence information for 10(10)to 10(12) bp of DNA. Although substantial sequencing advances have occurred in recent years, current approaches are still not sufficiently powerful for large-scale projects. Improvements are needed to facilitate DNA preparation for sequencing and to increase sequencing rate and accuracy.
Routine sequencing procedures entail separation and size measurement of DNA fragments resulting from the four sequencing reactions needed for each sequence analysis. Separation procedures require several hours and can reliably separate fragments with single-base resolution to only a few hundred base pairs.
Investigators at Leroy Hood's California Institute of Technology laboratory played an important early role in alleviating this bottleneck by automating the separation, replacing autoradiographic fragment-size analysis with fluorometric detection, and developing a procedure known as multiplexing for running a few differentially labeled DNA fragment collections in each gel lane; the ABI line of commercial sequencers uses this approach.
George Church and colleagues at Harvard University have extended the multiplex approach by increasing to over 40 the differentially detectable DNA fragment collections that can be run in each lane and to over 700 the bases that can be resolved in each lane. A multiplex sequencing effort using Sanger biochemistry also is proceeding under Raymond Gesteland and Robert Weiss at the University of Utah. Both the Hood and Church laboratories are also working to increase sequencing efficiency through the automation of other steps, such as DNA extraction, amplification, fragment generation, and labeling.
The multiplex approach is made increasingly powerful by adding to the number of samples that can be analyzed in each lane. Bruce Jacobson and colleagues at Oak Ridge National Laboratory (ORNL) are using resonance ionization mass spectrometry to resolve fragments labeled with stable tin and iron isotopes.
Several laboratories are working on other approaches to improve the DNA fragment separation rate and extend the fragment size that can be separated per run. Capillary electrophoresis is useful in mapping and sequencing because it allows rapid fragment separation (see Karger). Future goals include developing fully automated instrumentation.
Investigators at Lloyd Smith's University of Wisconsin laboratory and Jacobson and his colleagues are using ultrathin 50- to 75- µm gels to increase the fragment separation rate (standard gel thickness is about 400 µm). In addition, these groups, as well as those of Peter Williams at Arizona State University and Winston Chen at ORNL, are applying mass spectrometry to high-speed fragment sizing. In this approach, individual DNA molecules are volatilized (e.g., by laser ablation), ionized, and sized by high-resolution mass spectrometry. Work is now concentrated on the volatilization and ionization steps of this process, which is attractive because of its speed and because of the potential for analysis of longer DNA fragments.
Several new strategies are being explored that have the potential to improve sequencing rate and cost by several orders of magnitude.
Flow cytometry. An approach under development by Richard Keller and colleagues at LANL uses flow cytometry to sequence one DNA molecule at a time by
This approach has the theoretical advantages of (1) requiring only a single molecule, (2) generating sequence information at a high rate, and (3) allowing application to long DNA fragments. Single-molecule detection has been demonstrated, and work is now under way to develop single-molecule manipulation techniques and base labeling and cleaving procedures. LANL and Life Technologies, Inc. (GIBCO BRL) have entered into a cooperative agreement to collaborate on the development of this technology [see HGN 3(1), 5-7 (May 1991)].
Scanning Tunneling and Atomic Force Microscopies. Several laboratories are exploring scanning tunneling (STM) and atomic force microscopies for DNA sequence analysis. These approaches, based on the direct visualization of individual bases in DNA molecules attached to atomically flat substrates, have the theoretical advantages of requiring only a few molecules and generating sequence data at a high rate. Near-atomic-resolution images of DNA molecules have been produced by several laboratories, but sample and substrate preparation and base recognition must be substantially improved before this approach is reliable.
Wigbert Siekhaus and colleagues at LLNL are developing improved methods for deposition and bonding onto substrates, such as pyrolytic graphite for STM DNA analysis. So far their system can resolve individual adenine and thymine molecules deposited on these substrates. Troy Wilson and others at Miguel Salmeron's LBL laboratory are developing STM with a decreased scanning force to reduce scanning damage and to explore new procedures for attaching DNA molecules to the scanning substrate. Raoul Kopelman and coinvestigators at the University of Michigan are developing molecular exciton microscopy, based on the interaction of mercury-labeled bases with a laser-illuminated crystalline scanning tip, to distinguish specific bases during scanning without application of any mechanical force. Bruce Warmack, Tom Ferrell, and ORNL colleagues have demonstrated that STM can routinely image DNA prepared by electrospray and have also shown that covalent attachment of DNA to a substrate surface offers advantages for STM imaging.
X-Ray Diffraction. Gray, Jim Trebes, and LLNL colleagues have described an approach to DNA sequence analysis based on X-ray diffraction from arrays of DNA molecules. Four samples, one for each base labeled with an efficient X-ray scatterer such as mercury, are prepared for each sample. The distances between labels (and hence the sequence) are determined by analysis of the X-ray diffraction pattern from an ordered array of labeled molecules. Theoretical studies indicate that illumination with a bright X-ray source, such as a synchrotron, might generate an interpretable diffraction pattern from a DNA sample in a few seconds. The feasibility of this approach has been demonstrated theoretically, and recent experiments suggest that the preparation of properly ordered arrays of molecules from nanogram amounts of DNA might be the most challenging aspect of this approach.
Sequencing by Hybridization (SBH). Another novel approach to sequencing comes from Radoje Drmanac, Radomir Crkvenjakov, and colleagues at Argonne National Laboratory, who are pursuing SBH of multiple, short oligonucleotide probes to DNA samples. Sequence information is determined by mathematical analysis of the probe hybridization pattern. The advantage of this technique is that it does not require preparation of labeled DNA samples or fragment sizing; however, it does require a large number of hybridization reactions (e.g., 500 to 3000 hybridizations of 6 to 8 mers for sequencing 1- to 10-kb DNA fragments). Miniaturization of SBH by forming arrays of either DNA or oligonucleotides attached to microbeads is being developed to facilitate this process.
Robert Foote and coworkers at ORNL are developing the synthesis of microscale arrays of defined oligonucleotide sequences on planar supports to simplify SBH. The synthesis of 65,536 8-mer oligonucleotides in only 32 chemical reactions can occur by methods under development. A practical way to implement this approach would be to hybridize fluorescently labeled DNA samples against the array and read the pattern instrumentally.
The electronic form of the newsletter may be cited in the following style:
Human Genome Program, U.S. Department of Energy, Human Genome News (v3n3).
The Human Genome Project (HGP) was an international 13-year effort, 1990 to 2003. Primary goals were to discover the complete set of human genes and make them accessible for further biological study, and determine the complete sequence of DNA bases in the human genome. See Timeline for more HGP history.
Published from 1989 until 2002, this newsletter facilitated HGP communication, helped prevent duplication of research effort, and informed persons interested in genome research.