Purification Process For some researchers, isolating and purifying DNA is a housekeeping task: a routine but necessary part of most experiments, and one often taken for granted-until, of course, a restriction enzyme fails to cut a plasmid or a PCR reaction fizzles. For others, including contributors to the Human Genome Project, DNA purification represents an important, rate-limiting step in a lengthy quest for information. Yet wherever scientists' interests lie along this spectrum and whatever the source of the DNA they're isolating, odds are that newly developed kits, devices, or technologies have streamlined their work in recent years.
FILTER UNIT: Spin columns with Millipore's Ultrafree-MC use a cellulose ultrafiltration membrane.
Most such suppliers offer their kits based on DNA-binding silica matrices such as glass, silica particles, or diatomaceous earth. Typically, bacterial cells from an overnight culture are pelleted and resuspended in a lysozyme-RNase solution. Then a chaotropic salt solution is added, which conditions nucleic acids to bind specifically to the matrix: a glass-fiber or silica-gel filter, or a silica resin-filled column. To speed filtration, these matrices are generally mounted in spin columns or on vacuum manifolds. The bound DNA is washed, then eluted with a low-salt buffer or with water. The resulting plasmid is usually ready for restriction digestion or in vitro transcription, among other procedures. Manufacturers of silica-based miniprep kits include Bio-Rad as well as Ambion Inc. of Austin, Texas; Indianapolis-based Boehringer Mannheim Corp.; Promega Corp. of Madison, Wis.; and Qiagen Inc., located in Chatsworth, Calif.
WIZARD PLUS: One of two versions of Promega Corp.'s Wizard purification system.
Promega and Qiagen, among other manufacturers, have also adapted their plasmid miniprep protocols for the purification of DNA from M13 and lambda bacteriophage, as well as for large-scale plasmid preps. "Most large-scale plasmid kits give us high-quality DNA that meets our needs for cloning or probes," maintains Bruce Eckloff, a research technician at the Mayo Foundation in Rochester, Minn. In the past, Eckloff says, he had relied on high-performance liquid chromatography (HPLC) methods to purify large amounts of high-quality plasmid, which he in turn preferred to the venerable technique of isolating plasmid on cesium chloride gradients by centrifugation (as described in D.B. Clewell, D.R. Helinski, Proceedings of the National Academy of Sciences, 62:1159-66, 1969).
Like Eckloff, many researchers have apparently abandoned this time-consuming procedure, which also involves the use of hazardous chemicals. "The standard cesium chloride-ethidium bromide centrifugation technique has now become the old grey mare of plasmid DNA isolation and purification," Paul Hengen, a postdoctoral researcher at the National Cancer Institute, wrote recently in A.M. Griffin and H.G. Griffin, eds., Molecular Biology: Current Innovations and Future Trends, Wymondham, U.K., Horizon Scientific Press, 1995, pp. 39-50. "Although the technique is still held as the golden shrine of DNA purifications due to the purity of recovered DNA, it relies on a 24 to 48 hour centrifugation step," he wrote.
As an alternative, "the use of high-end purification equipment such as a capillary electrophoresis system or HPLC for . . . [plasmids] is gaining popularity," Hengen noted in the book. However, he added that because of "the expense and time required for setting up and maintaining a system in a low-funded lab . . . these methods are not widely used in molecular biology laboratories."
That's not to say that molecular biologists have single-mindedly embraced kits, however. "Researchers sometimes feel a sense of loss of control over experimental procedures by using kits," Hengen wrote in an article that described "kit wars" among molecular biologists who subscribed to an Internet newsgroup (P. Hengen, Trends in Biological Sciences, 19:46-7, 1994). "Care has to be taken to avoid slipping into a quick and easy way of doing science under the guise of saving time, especially with the heavy pressure to produce results more quickly than the competitors." He concluded that "ultimately, the decision to use a kit becomes one of practicality, with cost weighed against the time and effort needed to make the solutions and the availability of materials."
Much as manufacturers responded to the challenge of providing plasmid DNA for automated fluorescent sequencing, they have also sought to meet the growing demand for DNA purification among users of the polymerase chain reaction (PCR). For these clients, suppliers have developed systems for the extraction and purification of genomic DNA for PCR templates and the purification of synthetic oligonucleotide primers. They have also developed cleanup protocols for PCR products prior to further manipulation. Using a kit such as the Puregene DNA Isolation Kit from Gentra Systems Inc. of Minneapolis, researchers can isolate genomic DNA from whole blood, cultured cells, bacteria, or yeast in about an hour via salt precipitation; isolation from plant and animal tissues requires an additional hour for cell lysis. DNAzol, a guanidine-detergent lysing solution available from Life Technologies Inc. in Gaithersburg, Md., also speeds genomic DNA precipitation, while the QUICK-Geno genomic isolation kit from Clontech Laboratories Inc. of Palo Alto, Calif., uses silica beads to separate genomic DNA from other components of lysed cells.
FROM HOEHRINGER MANNHEIM: The Plant DNA Isolation Kit extracts genomic DNA.
REAGENT: Gentra Systems' Generation DNA Purification System for rapid purification of genomic DNA from tissues or fluids.
"We offer four types of purification, depending on the scale of the synthesis and the intended use for oligonucleotide," Bouthillette explains. For most PCR primers, which are typically shorter than 30 bases in length, simple desalting through gel filtration columns like those sold by Pharmacia Biotech Inc. of Piscataway, N.J., is usually sufficient, he says. His company synthesizes some oligonucleotide primers, such as those destined for reverse-transcription or mutagenic PCR, with terminal trityl groups so that they can be purified away from any failure sequences by reverse-phase chromatography. For syntheses larger than 80 optical density units, researchers at his company perform this separation by HPLC. For smaller yields, they use gravity or vacuum-driven cartridges such as those supplied by Glen Research Corp. in Sterling, Va. Lastly, says Bouthillette, they generally resolve oligonucleotides longer than 50 bases by polyacrylamide gel electrophoresis, then excise the band representing full-length oligonucleotide prior to eluting the DNA.
Following PCR, amplified DNA must be purified away from unincorporated primers and other reaction components. Several manufacturers offer PCR cleanup kits based on silica matrices, similar to those used for plasmid minipreps, like the PCR Clean Up Kit from Boehringer Mannheim, Bio-Rad's Prep-A-Gene DNA Purification Kit, and the QIAquick PCR Purification Kit from Qiagen. Spin columns with Ultrafree-MC filter units from Millipore Corp. of Bedford, Mass., on the other hand, use a cellulose ultrafiltration membrane to reversibly bind double-stranded PCR products.
Isolating PCR products presents challenges similar to those encountered by researchers removing DNA fragments of comparable size from agarose gels, so it's not surprising that many suppliers of PCR cleanup kits have also developed products for gel extraction. The gel matrix is typically dissolved with sodium perchlorate (or, in the case of low melting-point agarose, with heat); after that, the protocols proceed much like those for recovering PCR products. Millipore's spin-column methods for purifying DNA from agarose gels and PCR differ significantly, however. In the company's gel-extraction protocol, centrifugal force drives the DNA and buffer from the gel and across a low-binding filter, which holds back the agarose. The DNA must then be precipitated from the effluent using standard methods, such as high salt and ethanol.
While great gains have been made in the efficient purification of DNAs the size of bacterial plasmids and smaller, and also of genomic DNA, the "middle ground" between them remains relatively fallow, according to Richard Guilfoyle, a research scientist in the laboratory of Lloyd Smith in the chemistry department at the University of Wisconsin, Madison. Scientists in that laboratory devise mapping and sequencing strategies for complex genomes, such as the human genome. In one of their protocols, they chemically cleave large fragments of DNA directly from the genome. "For these approaches to be truly effective, improved DNA methods have to be developed for isolating large pieces of genomic DNA, up to several megabases in length," Guilfoyle explains.
He and his coworkers have developed one such method, which they call Triple-Helix-Mediated Affinity Capture. In one application of this technique, they flanked the cloning site in a cosmid vector with homopurine-homopyrimidine triplex-forming sequences, then synthesized biotinylated oligonucleotides bearing the complementary homopyrimidine triplex to bind to the modified cosmid insert (H. Ji et al., Genetic Analysis: Tools and Applications, 11:43-7, 1994). Following digestion of the cosmid to release the insert, they captured the biotinylated oligonucleotide-insert triplex on streptavidin-coated magnetic particles, available from suppliers including Bangs Laboratories of Carmel, Ind.; Dynal Inc. of Lake Success, N.Y.; and PerSeptive Biosystems Inc. of Framingham, Mass. The DNA-bound magnetic particles were separated from solution by application of a magnet to the outside of the tube, and the insert DNA eluted from the complex under alkaline conditions.
"Using this method, we recover greater than 95 percent of the cosmid insert, and it's very clean," Guilfoyle avers. And because this method doesn't require centrifugation, he says, it not only is gentle-and thus potentially adaptable to capture even larger inserts-but also could be automated.
FULLY AUTOMATED: The Autogen 740 DNA purification system from Integrated Separation Systems processes samples.
Meanwhile, Trevor Hawkins and coworkers at the Whitehead Institute/Massachusetts Institute of Technology Center for Genomic Research in Cambridge, Mass., have developed another high-throughput DNA purification method based on magnetic beads. Termed SPRI, for solid-phase reversible immobilization, the technique hinges on the researchers' observation that many forms of DNA bind reversibly to carboxyl-coated magnetic particles (from the same suppliers as the aforementioned streptavidin-coated magnetic particles). Lloyd Smith, Richard Guilfoyle, and coworkers at the University of Wisconsin are developing protocols based on the same principle.
"SPRI uses the inexpensive carboxyl-coated magnetic particles that comprise the base material for most magnetic particle manufacture," Hawkins explains, in an introduction to the technique which appears on his Web page (http://www-genome.wi.mit.edu/~tlh/). Also featured on his page are SPRI protocols for M13 and PCR product isolation. "Under conditions of high polyethylene glycol and salt concentration, the surface of these particles binds both single- and double-stranded DNA, including sequencing reaction products, PCR products, M13 phage, lambda phage, plasmids, cosmids and BACs," he continues. Since their initial discovery of this affinity method in 1994, Hawkins and his coworkers have carried out tens of thousands of M13 and PCR product purifications. They have also combined SPRI with an automated sequencing device called the Sequatron to produce a robotic system that both purifies and sequences DNA without human intervention, he reports.
"I think the real value here is automation," Hawkins writes in response to an E-mail query from The Scientist. "Other nonmagnetic particle methods exist which can be used for M13 isolation which cost about the same in supplies, but require many people to perform the process. If we did not use SPRI to allow our automation of purification, sequencing, and DNA manipulation, we would need a further four or five people to produce 2,000 reactions per day."
Alison Mack is a freelance science writer based in Wilmington, Del.