Detecting single base substitutions, mismatches and bulges in DNA by temperature gradient gel electrophoresis and related methods

Application of Temperature-Gradient Gel Electrophoresis for Detection of Prion Protein Gene Polymorphisms in Polish Świniarka Sheep

This study presents preliminary data on the polymorphism in the prion protein gene of Świniarka sheep using temperature gradient gel electrophoresis (TGGE). Available data indicate that sensitivity to scrapie is associated with polymorphisms in three codons of prion protein gene: 136,154, and 171. The TGGE method was used to detect point mutations in these codons responsible for sensitivity or resistance to scrapie. This study revealed presence of an allele encoding valine (V) in codon 136, which is associated with high sensitivity to scrapie and occurred in the form of heterozygous allele together with alanine (AV). The highest variability was observed in codon 171, with presence of arginine (R) and glutamine (Q) in the homozygous (RR or QQ) as well as the heterozygous form (RQ). The results of examination of fifty sheep DNA samples with mutations in codons 136, 154, and 171 demonstrated that TGGE can be used as a simple and rapid method to detect mutations in the PrP gene of sheep. Several samples can be run at the same time, making TGGE ideal for the screening of large numbers of samples.

Heat-transfer-based detection of SNPs in the PAH gene of PKU patients

Conventional neonatal diagnosis of phenylketonuria is based on the presence of abnormal levels of phenylalanine in the blood. However, for carrier detection and prenatal diagnosis, direct detection of disease-correlated mutations is needed. To speed up and simplify mutation screening in genes, new technologies are developed. In this study, a heat-transfer method is evaluated as a mutation-detection technology in entire exons of the phenylalanine hydroxylase (PAH) gene. This method is based on the change in heat-transfer resistance (R_th) upon thermal denaturation of dsDNA (double-stranded DNA) on nanocrystalline diamond. First, ssDNA (single-stranded DNA) fragments that span the size range of the PAH exons were successfully immobilized on nanocrystalline diamond. Next, it was studied whether an R_th change could be observed during the thermal denaturation of these DNA fragments after hybridization to their complementary counterpart. A clear R_th shift during the denaturation of exon 5, exon 9, and exon 12 dsDNA was observed, corresponding to lengths of up to 123 bp. Finally, R_th was shown to detect prevalent single-nucleotide polymorphisms, c.473G>A (R158Q), c.932T>C (p.L311P), and c.1222C>T (R408W), correlated with phenylketonuria, displaying an effect related to the different melting temperatures of homoduplexes and heteroduplexes.

Genome profiling (GP) method based classification of insects: congruence with that of classical phenotype-based one

BACKGROUND

Ribosomal RNAs have been widely used for identification and classification of species, and have produced data giving new insights into phylogenetic relationships. Recently, multilocus genotyping and even whole genome sequencing-based technologies have been adopted in ambitious comparative biology studies. However, such technologies are still far from routine-use in species classification studies due to their high costs in terms of labor, equipment and consumables.

METHODOLOGY/PRINCIPAL FINDINGS

Here, we describe a simple and powerful approach for species classification called genome profiling (GP). The GP method composed of random PCR, temperature gradient gel electrophoresis (TGGE) and computer-aided gel image processing is highly informative and less laborious. For demonstration, we classified 26 species of insects using GP and 18S rDNA-sequencing approaches. The GP method was found to give a better correspondence to the classical phenotype-based approach than did 18S rDNA sequencing employing a congruence value. To our surprise, use of a single probe in GP was sufficient to identify the relationships between the insect species, making this approach more straightforward.

CONCLUSION/SIGNIFICANCE

The data gathered here, together with those of previous studies show that GP is a simple and powerful method that can be applied for actually universally identifying and classifying species. The current success supported our previous proposal that GP-based web database can be constructible and effective for the global identification/classification of species.

Modeling of 2-D DNA display

2-D display is a fast and economical way of visualizing polymorphism and comparing genomes, which is based on the separation of DNA fragments in two steps, first according to their size and then to their sequence composition. In this article, we present an exhaustive study of the numerical issues associated with a model aimed at predicting the final absolute locations of DNA fragments in 2-D display experiments. We show that simple expressions for the mobility of DNA fragments in both dimensions allow one to reproduce experimental final absolute locations better than experimental uncertainties. However, our simulations also point out that the results of 2-D display experiments are not sufficient to determine the best set of parameters for the modeling of fragments separation in the second dimension and that additional detailed measurements of the mobility of a few sequences are necessary to achieve this goal. We hope that this work will help in establishing simulations as a powerful tool to optimize experimental conditions without having to perform a large number of preliminary experiments and to estimate whether 2-D DNA display is suited to identify a mutation or a genetic difference that is expected to exist between the genomes of closely related organisms.

Validation of dye-binding/high-resolution thermal denaturation for the identification of mutations in the SLC22A5 gene

Primary carnitine deficiency is an autosomal recessive disorder of fatty acid oxidation resulting from defective carnitine transport. This disease is caused by mutations in the OCTN2 carnitine transporter encoded by the SLC22A5 gene. Here we validate dye-binding/high-resolution thermal denaturation as a screening procedure to identify novel mutations in this gene. This procedure is based on the amplification of DNA by PCR in capillaries with the dsDNA binding dye LCGreen I. The PCR reaction is then analyzed in the same capillary by high-resolution thermal denaturation. Samples with abnormal melting profiles are sequenced. This technique correctly identified all known patients who were compound heterozygotes for different mutations in the carnitine transporter gene and about 30% of homozygous patients. The remaining 70% of homozygous patients were identified by a second amplification, in which the patient's DNA was mixed with the DNA of a normal control. This screening system correctly identified eight novel mutations and both abnormal alleles in six new families with primary carnitine deficiency. The causative role of the missense mutations identified (c.3G>T/p.M1I, c.695C>T/p.T232M, and c.1403 C>G/p.T468R) was confirmed by expression in Chinese hamster ovary (CHO) cells. These results expand the mutational spectrum in primary carnitine deficiency and indicate dye-binding/high-resolution thermal denaturation as an ideal system to screen for mutations in diseases with no prevalent molecular alteration.

Sequence variability and functional analysis of MutS of hypermutable Pseudomonas aeruginosa cystic fibrosis isolates

In this study, we investigated the variability of MutS among Pseudomonas aeruginosa recovered from cystic fibrosis (CF) patients. Sequencing of the mutS gene of 15 hypermutable P. aeruginosa isolates obtained from different patients revealed high rates of nucleotide substitutions as compared to that of strain PAO1. Significantly more synonymous than non-synonymous nucleotide substitutions have been found, indicating that generally MutS is highly conserved. The functional analysis of MutS variants by complementation of a PAO1 mutS mutant revealed 5 isolates with a defective MutS due to frameshift mutations or amino acid substitutions. This work supports the hypothesis that the respiratory tract of CF patients represents an environment that favors the selection of highly adaptive mutator phenotypes.

Change of the Protein p53 Electrochemical Signal According to its Structural Form – Quick and Sensitive Distinguishing of Native, Denatured, and Aggregated Form of the “Guardian of the Genome”

Presence of mutated and/or structurally modified (e.g., denatured, aggregated) protein p53 form is associated with several disorders such as Alzheimer's disease, Parkinson's disease, prion diseases, and many types of tumours. The aim of this work was to distinguish native, denatured and aggregated form of full-length p53 by flow injection analysis coupled with electrochemical detector (FIA-ED). Firstly FIA-ED method used for protein native form determination was optimized (detection limit 45.8 amol per 5 mul injection; 3 x S/N). In addition the technique was applied to identify p53 structural forms (denatured and aggregated). It was found out that denatured form provides about three times higher electrochemical response (protein structure unfolding, approach of more electroactive centers - aminoacid residues - towards electrode surface) in comparison with native form. On the other hand, aggregated form offers lower response (steric eclipse of electroactive protein parts) when compared with the signal of native form. The obtained data show that we are not only able to sensitively determine native, denatured, and aggregated structural forms of p53 protein but also to distinguish them.

Detection of single nucleotide polymorphism in the FUT2 gene by temperature gradient gel electrophoresis

A simple and rapid analysis system for single nucleotide polymorphisms (SNPs) was investigated for the FUT2 gene using the temperature gradient gel electrophoresis (TGGE) method. The 426-bp or 259-bp FUT2 fragments were amplified from heterozygous samples using primers, and the heteroduplex and homoduplex bands were detected by TGGE. The FUT2 fragments amplified from homozygous samples were denatured and re-annealed with a known sequence fragment, forming heteroduplex bands which were analyzed by TGGE. The fragment patterns of homoduplex and heteroduplex bands in TGGE were specific to the genotypes, and more specifically, the number of heteroduplex bands and the migration of the homoduplex bands corresponded to the number of nucleotide substitutions and the change in GC content due to the substitutions, respectively. Thus, the TGGE technique can act as a high-throughput method for the detection and the presumption of sequences of known and unknown SNPs in the FUT2 gene.

Interstrand cross-links: a new type of gamma-ray damage in bromodeoxyuridine-substituted DNA

Interstrand cross-links (ICL) represent one of the most toxic types of DNA damage for dividing cells. They are induced both by natural products (e.g., psoralens + UVA) and by several chemical agents, some of which are used in chemotherapy (e.g., carboplatin and mitomycin C). Although repair mechanisms exist for interstrand cross-links, these lesions can induce mutations, chromosomal rearrangements, and cell death. Here, we report, for the first time, the formation of ICL by gamma-rays in brominated DNA. It is well established that the radiosensitization properties of bromodeoxyuridine (BrdUrd) result primarily from the electrophilic nature of the bromine, making it a good leaving group and leading to the irreversible formation of a uridinyl radical (dUrd(*)) or uridinyl anion (dUrd-) upon addition of an electron. We observe that the radiolytic loss of the bromine atom is greatly suppressed in double-stranded compared to single-stranded DNA. We have used a model DNA containing a bulge, formed by five mismatched bases, and have observed a linear dose-response for the formation of strand breaks on the single-stranded regions of both the brominated strand and the opposite nonbrominated strand. Surprisingly, we have observed the formation of interstrand cross-links exclusively in the mismatched region. Thus, we propose that the radiosensitization effects of bromodeoxyuridine in vivo will almost certainly be limited to single strand regions such as found in transcription bubbles, replication forks, mismatched DNA, and possibly the loop region of telomeres. Our results suggest that interstrand cross-links may contribute to the radiosensitization effects of BrdUrd. These findings may have profound implications for the clinical use of bromodeoxyuridine as a radiosensitizer, as well as for the development of targeted radiosensitizers.

Single-strand-specific radiosensitization of DNA by bromodeoxyuridine

The effects of bromodeoxyuridine (BrdUrd) substitution for thymidine on gamma-ray-induced strand breakage were determined in single- and double-stranded oligonucleotides and double-stranded oligonucleotides containing a mismatched bubble region. BrdUrd does not sensitize complementary double-stranded DNA to gamma-ray-induced strand breakage, but it greatly sensitizes single-stranded DNA. However, when the BrdUrd is present in a single-stranded bubble of a double-stranded oligonucleotide, the non-base-paired nucleotides adjacent to the BrdUrd as well as several unpaired sites on the opposite unsubstituted strand are strongly sensitized. The radiosensitization properties of BrdUrd result primarily from the electrophilic nature of the bromine, making it a good leaving group and leading to the irreversible formation of the uridine-yl radical (dUrd(.)) or the uridine-yl anion (dUrd(-)) upon addition of an electron. The radiolytic loss of the bromine atom is greatly suppressed in double-stranded compared to single-stranded DNA. Thus we propose that the radiosensitization effects of bromouracil in vivo will likely be limited to single-strand regions such as found in transcription bubbles, replication forks, DNA bulges and the loop region of telomeres. Our results may have profound implications for the clinical use of bromodeoxyuridine (BrdUrd) as a radiosensitizer as well as for the development of targeted radiosensitizers.

Segregation of partly melted molecules: isolation of CpG islands by polyacrylamide gel electrophoresis

The technique of segregation of partly melted molecules (SPM) is a convenient and efficient method to isolate DNA fragments associated with CpG islands. The approach is conceptually simple and uses denaturant gradient gel electrophoresis to separate DNA molecules digested with restriction endonucleases. The SPM methodology has successfully been applied to the identification of genes from anonymous, unsequenced DNA fragments and CpG islands methylated in human cancer. In this article the theoretical background and practical application of the SPM method is reviewed.

Rapid, comprehensive screening of the human medium chain acyl-CoA dehydrogenase gene

Newborn screening by tandem mass spectrometry (MS/MS) identifies patients with medium chain acyl-CoA dehydrogenase (MCAD) deficiency the most frequently observed disorder of fatty acid oxidation. Molecular genetic analysis is becoming a common tool to confirm those identified as affected by prospective screening and for carrier detection in family studies. The A985G (K304E) mutation accounts for approximately 80% of mutant alleles in MCAD deficient patients, presenting symptomatically, while greater variability of mutant alleles is observed among cases identified through prospective screening. Aside from A985G, the mutation spectrum in MCAD deficient patients is heterogeneous such that comprehensive gene analysis is required. Traditionally the MCAD gene is assayed by sequencing the entire coding region. Although effective and definitive, this approach is expensive, turn around time is slow, and is poorly amenable to a clinical service molecular genetics laboratory. Dye-binding/high-resolution thermal denaturation is a rapid and homogeneous method by which to scan a PCR product for evidence of sequence aberration. PCR is performed in capillaries in the presence of the dsDNA-binding dye LCGreen I and subsequently the DNA/dye complexes are analyzed by high-resolution thermal denaturation. DNA sequencing was limited to fragments displaying abnormal melting profiles. Of 18 specimens analyzed, 11 have a genotype consistent with MCAD deficiency and seven have a genotype consistent with carrier status. Clinical and biochemical data corroborate that the genotype results identified the affected patients and differentiates them from carriers. The entire process is homogeneous requiring no post-PCR manipulation and is completed in under 3 h.

Amplicon melting analysis with labeled primers: a closed-tube method for differentiating homozygotes and heterozygotes

BACKGROUND

Common methods for identification of DNA sequence variants use gel electrophoresis or column separation after PCR.

METHODS

We developed a method for sequence variant analysis requiring only PCR and amplicon melting analysis. One of the PCR primers was fluorescently labeled. After PCR, the melting transition of the amplicon was monitored by high-resolution melting analysis. Different homozygotes were distinguished by amplicon melting temperature (T(m)). Heterozygotes were identified by low-temperature melting of heteroduplexes, which broadened the overall melting transition. In both cases, melting analysis required approximately 1 min and no sample processing was needed after PCR.

RESULTS

Polymorphisms in the HTR2A (T102C), beta-globin [hemoglobin (Hb) S, C, and E], and cystic fibrosis (F508del, F508C, I507del, I506V) genes were analyzed. Heteroduplexes produced by amplification of heterozygous DNA were best detected by rapid cooling (>2 degrees C/s) of denatured products, followed by rapid heating during melting analysis (0.2-0.4 degrees C/s). Heterozygotes were distinguished from homozygotes by a broader melting transition, and each heterozygote had a uniquely shaped fluorescent melting curve. All homozygotes tested were distinguished from each other, including Hb AA and Hb SS, which differed in T(m) by <0.2 degrees C. The amplicons varied in length from 44 to 304 bp. In place of one labeled and one unlabeled primer, a generic fluorescent oligonucleotide could be used if a 5' tail of identical sequence was added to one of the two unlabeled primers.

CONCLUSION

High-resolution melting analysis of PCR products amplified with labeled primers can identify both heterozygous and homozygous sequence variants.

Commonly conserved genetic fragments revealed by genome profiling can serve as tracers of evolution

We developed a method to produce, identify and analyze DNA fragments for the purpose of taxonomic classification. Genome profiling (GP) is a strategy that identifies genomic DNA fragments common to closely related species without prior knowledge of the DNA sequence. Random PCR, one of the key technologies of GP, is used to produce fragments and may be used even when there are mutations at the priming site. These fragments can then be distinguished based on the information of mobility and melting pattern when subjected to temperature gradient gel electrophoresis (TGGE). Corresponding fragments among several species, designated as commonly conserved genetic fragments (CCGFs), likely have the same genetic origin or correspond to the same gene. The criteria for identification of CCGFs has been defined and presented here. To assess this prediction, some of the fragments were sequenced and were confirmed to be CCGFs. We show that genome profiles bearing evolutionarily conserved CCGFs can be used to classify organisms and trace evolutionary pathways, among other profound applications.

A database for the provisional identification of species using only genotypes: web-based genome profiling

BACKGROUND

For a long time one could not imagine being able to identify species on the basis of genotype only as there were no technological means to do so. But conventional phenotype-based identification requires much effort and a high level of skill, making it almost impossible to analyze a huge number of organisms, as, for example, in microbe-related biological disciplines. Comparative analysis of 16S rRNA has been changing the situation, however. We report here an approach that will allow rapid and accurate phylogenetic comparison of any unknown strain to all known type strains, enabling tentative assignments of strains to species. The approach is based on two main technologies: genome profiling and Internet-based databases.

RESULTS

A complete procedure for provisional identification of species using only their genomes is presented, using random polymerase chain reaction, temperature-gradient gel electrophoresis, image processing to generate 'species-identification dots' (spiddos) and data processing. A database website for this purpose was also constructed and operated successfully. The protocol was standardized to make the system reproducible and reliable. The overall methodology thus established has remarkable aspects in that it enables non-experts to obtain an initial species identification without a lot of effort and is self-developing; that is, species can be determined more definitively as the database is used more and accumulates more genome profiles.

CONCLUSIONS

We have devised a methodology that enables provisional identification of species on the basis of their genotypes only. It is most useful for microbe-related disciplines as they face the most serious difficulties in species identification.

Spatial temperature gradient capillary electrophoresis for DNA mutation detection

A continuous spatial temperature gradient was established in capillary electrophoresis by using a simple temperature control device. The temperature profile along the capillary was predicted by theoretical calculations. A nearly linear spatial temperature gradient was established and applied to DNA mutation detection. By spanning a wide temperature range, it was possible to perform simultaneous heteroduplex analysis for various mutation types that have different melting temperatures.

Solution-based scanning for single-base alterations using a double-stranded DNA binding dye and fluorescence-melting profiles

DNA molecules differing by as little as a single-base substitution have traditionally been distinguished by gel electrophoresis-based methodologies that exploit differences in the sequence-specific properties of double-stranded DNA (dsDNA) such as melting temperature and secondary conformational configuration. By comparison, solution-based fluorescence methods using sequence-specific probes are limited to detecting mutations restricted to very short segments of DNA ( approximately 20 bp). We describe a solution-based fluorescence method that discriminates between wild-type and mutant sequences using a dsDNA binding dye, and interrogates a region of >200 nucleotides. This method is based on melting theory and entails fluorescence monitoring of the melting temperatures of GC-clamped amplicons subjected to gradual and progressive thermal denaturation in the presence of a constant concentration of urea. Heterozygous samples are easily identified by the lower melting temperatures of the less thermodynamically stable heteroduplex mismatches from the wild-type:mutant DNA hybrids as compared to the more stable wild-type Watson-Crick duplexes. All of the four possible sets of mismatches (A.G/T.C, T.G/A.C, G.G/C.C, and T.T/A.A) represented in 17 heterozygous mutations distributed throughout the length of 20 different amplicons (104 to 212 bp), were distinguished from the wild-type by their altered melting profiles. This methodology is advantageous in that it obviates gel electrophoresis or labeled oligonucleotide probes. Significantly, it expands the region of interrogation for detection of single-base changes using fluorescence-based methods in solution, and is amenable for automation and adaptation to high-throughput systems.

A commented dictionary of techniques for genotyping

Several tools, differing in their technical and practical parameters, are available for the detection of point mutations as well as small deletions and insertions. In this article, a dictionary featuring over fifty methods for detection of mutation is presented. The distinguishing principle for each method is briefly explained. Sorting of and discussion on the methods give the reader a brief introduction to the field of genotyping.

Temporal temperature gradient gel electrophoresis (TTGE) as a tool for identification of Lactobacillus casei, Lactobacillus paracasei, Lactobacillus zeae and Lactobacillus rhamnosus

AIMS

To develop a tool for rapid and inexpensive identification of the Lactobacillus casei complex.

METHODS AND RESULTS

Lactobacillus casei, Lactobacillus paracasei, Lactobacillus zeae and Lactobacillus rhamnosus were identified by PCR-amplification of the segment between the U1 and U2 regions of 16S rDNA (position 8-357, Escherichia coli numbering) and temporal temperature gradient gel electrophoresis (TTGE). Seven tested Lact. paracasei strains were divided into three TTGE-subgroups.

CONCLUSION

TTGE successfully distinguished between the closely-related target species. TTGE is also a powerful method for revealing sequence heterogeneities in the 16S rRNA genes.

SIGNIFICANCE AND IMPACT OF THE STUDY

Due to rapid and easy performance, TTGE of PCR-amplified 16S rDNA fragments will be useful for the identification of extended numbers of isolates.

Different characteristics distinguish early versus late arising adaptive mutations in Escherichia coli FC40

The Escherichia coli strain FC40 has frequently been employed to investigate the mechanism of adaptive mutations. The strain cannot utilize lactose due to a +1 frameshift mutation that reduces beta-galactosidase to about 1% of normal levels. Cells undergo a high rate of mutation from Lac- to Lac+ when cells are grown with lactose as the sole energy source. Almost all Lac+ colonies arising 3-6 days after plating result from a base pair deletion in runs of iterated base pairs within a 130-bp target region. In this study we characterized Lac+ colonies arising 3-10 days after plating. Temperature gradient gel electrophoresis (TGGE) was used to detect mutations in the target region as a function of the day a colony appears. TGGE results confirmed the occurrence of mutations within the target region in 36 of 37 FC40 Lac+ colonies arising on days 3-7. However, mutations in this region were not detected in 23 of 37 Lac+ colonies arising from days 8-10. Sequencing data verified the TGGE results. Half of the Lac+ mutants arising on days 8-10 with no base pair change in the target region were unstable and exhibited a Lac- phenotype after successive growth cycles in rich medium. The results suggest that amplification of the lac operon region is a common factor in late arising colonies, and that different characteristics distinguish early and late arising Lac+ colonies.

Heteroduplex mobility assay detects DNA mutations for differentiation of closely related phytoplasma strains

We present the first use of DNA heteroduplex mobility assay (HMA) to detect the point mutations including substitutions and deletions/insertions in 16S rDNA of aster yellows phytoplasma (AY27) and to differentiate phytoplasmas collected from field samples of clover proliferation (CP) and alfalfa witches'-broom (AWB). The phytoplasmal 16S rDNA fragment was amplified from AY27 by polymerase chain reaction (PCR) and cloned into a plasmid vector. The cloned DNA fragment was subjected to in vitro mutation to produce 1- to 4-base substitutions and 1- to 3-base deletions. The mutated 16S rDNA fragments were analyzed by HMA. The results showed that a single two-base substitution or a single-base deletion/insertion in the 529 bp DNA fragment was directly detected and that a DNA divergence at a level of as low as 0.2% was detectable by HMA. Heteroduplex mobilities were affected by the number and composition of the phytoplasma DNA bases in mismatches or gaps and were proportional to the degree of DNA divergences. Gaps caused greater retardation in heteroduplex mobility than mismatches did. HMA was highly sensitive in detecting the mixed infections of phytoplasmas. In analyses of CP and AWB field samples collected in Alberta, two CP and one AWB phytoplasma isolates were differentiated from others by HMA but not by restriction fragment length polymorphism (RFLP). Therefore, HMA provides a simple, rapid, highly sensitive and analytical method to detect and estimate the genetic divergence of phytoplasmas when other methods such as RFLP are not readily applicable.

Monitoring hybridization during polymerase chain reaction

The polymerase chain reaction (PCR) is usually analyzed by gel electrophoresis for size separation of PCR products. Additional separation techniques, such as single-stranded conformational polymorphism (SSCP), denaturing gradient gel electrophoresis (DGGE), temperature gradient gel electrophoresis (TGGE) and denaturing high-performance liquid chromatography (DHPLC), can also be used to scan for sequence alterations. These techniques are all based on the effect of PCR product hybridization on mobility. Hybridization can also be monitored with fluorescence during PCR without chromatographic or electrophoretic separation. Continuous monitoring of PCR allows the detection, quantification and sequence specificity of PCR products to be assessed, often without any need for further analysis. In such a closed system, PCR quantification with sensitivity to the single copy level can be achieved using either double-stranded DNA binding dyes or fluorescently labeled allele-specific oligonucleotide (ASO) probes. Melting curve analysis with ASO probes can be used to genotype various alleles, including single base alterations. The integration of rapid cycle PCR and ASO probes in an automated system greatly facilitates research and clinical applications of nucleic acid analysis in genetics, oncology, and infectious disease.

Electrophoretic separation of small DNA fragments in the presence of electroosmotic flow using poly(ethylene oxide) solutions

A new and simple method was demonstrated for separating phi X-174/Hae III DNA restriction fragments and DNA markers V and VI, respectively, without filling capillaries with polymer solutions prior to analysis. Using this novel method, poly(ethylene oxide) (PEO) solutions containing ethidium bromide migrated into capillaries by electroosmotic flow (EOF) during the separation. Two DNA fragments (123 and 124 bp) in markers V and VI were well-resolved. RSD values for the separation of phi X-174/Hae III DNA restriction fragments were less than 0.52% for 3 runs using a single 75-micron capillary and less than 3.96% using three different 75-micron capillaries. A highly viscous polymer solution prepared from 3% PEO was also used for separation of DNA markers V and VI. Theoretical plates up to 11.91 million/m and separation times of less than 7 min were achieved in the separation of phi X-174/Hae III DNA restriction fragments using a 10-micron capillary and a 2% PEO solution. Advantages of this method include simplicity, short separation times, the ability to use highly viscous polymer solutions for separating small DNA fragments, and the possibility of introducing several different polymer solutions into capillaries to extend the DNA separation range.