Printer Friendly

Capillary electrophoresis-based heteroduplex analysis with a universal heteroduplex generator for detection of point mutations associated with rifampin resistance in tuberculosis.

The Human Genome Project has provided more information about human disorders and diseases and presented new tools and opportunities for medical advances. As the number of genes linked to specific diseases grows, the development of technically simple and cost-effective methods to detect genetic mutations becomes increasingly important. Additionally, the Human Gene Mutation Database ( reports that >90% of disease-causing mutations in humans are attributable to microlesions, such as single-base substitutions and small deletions and/or insertions, rather than gross lesions (repeat expansions, complex gene rearrangements, gross insertions and/or deletions). Therefore, it is vital that mutation detection methods are sensitive to the content of nucleotide variations at the single-base level.

Although DNA sequencing is considered the "gold standard", this technology is still too cumbersome and costly to be effective for high-throughput screening of mutations (1). As a result, alternative mutation detection strategies have been developed and can be categorized as specific or scanning methods. Specific methods identify particular, characterized sequence variations, whereas scanning methods detect uncharacterized sequence variations. Heteroduplex analysis (HDA)[3] is a popular gel-based mutation scanning method that relies on the differential mobility of short DNA fragments with complete sequence complementarity vs those containing one or more mismatches. The traditional slab gel format has a parallel processing capacity that allows multiple analyses to be conducted simultaneously, but the approach is not amenable to automation and is quickly being adapted to a capillary format, which offers potential improvements in cost, resolution, speed, quantification, and automation (2-5). Additionally, slab gels are limited to agarose and polyacrylamide compositions, whereas several matrices are available for capillary electrophoresis (CE), such as hydrophilic polymers and cellulose derivatives that can improve the resolution in the heteroduplex assay. Replaceable linear polyacrylamides (LPAs) and other entangled polymer solutions, in particular, offer lower viscosity, achieve high-efficiency separations, and permit rapid replacement of the sieving matrix and, thus, recycling of the capillary (3, 4, 6-8).

An early report using HDA and CE described a method that utilized an entangled polymer matrix under nondenaturing conditions to resolve small conformational differences created by mismatches in duplexed DNA (9). In some cases, the method was unable to discriminate between the homo- and heteroduplexes, but in no case was it able to separate all generated fragments. Another report described the use of a temperature gradient during electrophoresis for mutation detection using HDA (10). Various base substitutions were screened in different length fragments by comparison to their respective reference samples. In this variation of HDA, the lower melting temperature of heteroduplex DNA molecules containing mismatches relative to the homoduplex molecules was exploited to partially denature the DNA during electrophoresis. The approach used a temperature gradient that was selected to cover the entire range of melting temperatures of duplexes containing various mismatches encountered in the samples for a universal screening method that was able to detect single-base substitutions in a 96-capillary array for high throughput.

A third report explored the use of various matrices, surface coatings, and silanizing reagents in HDA to discriminate between mutations in BRCA1, a breast cancer susceptibility gene with mutations ranging from a single-base substitution to a 40-base deletion (11). Using their optimized method, the authors achieved effective mutation discrimination in <10 min, and optimal performance included hydroxyethyl cellulose (HEC) as the polymer network with a poly(vinylpyrrolidone) coating covalently coupled to the capillary by chlorodimethyloctylsilane.

In some of the previous reports, the HDA technique was used to identify mutations in a sample by annealing the gene fragment (amplified by PCR) to a reference, which contained the wild-type sequence. In heterozygous individuals in whom genes possessing different allelic forms are present in a single copy within the genome, DNA heteroduplexes can be formed in trans without use of a reference (12). However, in either case, a single-base substitution in the sample generates a very slight conformational difference in the heteroduplex that is typically more difficult to resolve by conventional electrophoretic methods than an insertion or a deletion mutation, especially when using HDA (1). Fortunately, high-resolution HDA can be achieved by annealing the DNA fragment with a specially constructed third-party DNA, referred to as a universal heteroduplex generator (UHG) (12). This synthetic sequence mimics the genomic DNA sequence but contains controlled nucleotide substitutions, deletions, or insertions at nucleotide positions opposite to and contiguous with known mutation sites within the genomic DNA. This controlled variation in the UHG greatly enhances electrophoretic sorting of single-base substitutions in the samples and increases the resolution between homo- and heteroduplex DNA (see Fig. 1).

The HDA-UHG technique has been used in a slab gel format for drug susceptibility testing in tuberculosis (TB) (13). The disease, caused by Mycobacterium tuberculosis (Mtb), is responsible for 2.5 million deaths each year, is the leading cause of mortality in adults attributable to an infectious agent, and accounts for 26% of all preventable adult deaths globally (14,15). Additionally, the emergence of Mtb strains that are resistant to antimycobacterial drugs such as rifampin and isoniazid increases the cost and difficulty in treatment and may require a more toxic regimen (16).

In particular, resistance to rifampin is associated with missense, insertion, and deletion mutations in the rifampin resistance-determining region (RRDR) of rpoB, which encodes the [beta] subunit of the DNA-dependent RNA polymerase (17,18). The HDA-UHG method has been used in a slab gel format for simultaneous detection of Mtb with qualitative assessment of susceptibility to rifampin (13). However, this analysis requires >2 h of separation time and post staining with ethidium bromide.


Here we report a capillary-based HDA-UHG method that uses nondenaturing conditions for the detection of single-base nucleotide substitutions in the Rif region of rpoB that give rise to the rifampin-resistant (Rmp-r) phenotype of Mtb. This assay is a modification of the slab gel-based assay designed for detection of Mtb and rifampin susceptibility testing of Mtb directly from sputum specimens. The effects of intercalating dyes and ultraviolet (UV) absorbance vs laser-induced fluorescence (LIF) detection methods were investigated. In addition, various types of entangled polymer solutions appropriate for the high-resolution separation of duplexed DNA in capillary formats were investigated, including HEC, methyl cellulose (MC), and LPA in terms of their ability to allow the rapid separation of heteroduplexes resulting from single-base substitutions in this application. These characteristics can be important for any HDA method that examines point mutations enhanced by a generator. Optimized conditions were developed to detect single-base substitutions and achieve discrimination between all four fragments of the homo-/heteroduplex mixtures while determining rifampin susceptibility among TB samples.

Materials and Methods


N,N,N',N'-tetramethylethylenediamine (TEMED) and [gamma]-[(methoxyacryl)oxy]propyltrimethoxysilane ([gamma]-MAPS) were obtained from Amresco. [gamma]-MAPS was titrated to pH 3.5, using glacial acetic acid, before use. Stock solutions of 10X Tris-acetate-EDTA (TAE; 400 mmol/L Tris-acetate, 10 mmol/L EDTA) and Tris-borate-EDTA (TBE; 1.0 mol/L Tris, 0.9 mol/L boric acid, 0.01 mol/L EDTA) were purchased from Life Technologies. The fluorescent intercalating dyes YOPRO1 (MT 629) and TOPRO5 ([M.sub.r] 697) were purchased from Molecular Probes and stored at -20[degrees]C until used. MC (MT 88 000) and HEC ([M.sub.r] 250 000) were purchased from Sigma Chemical Co. High-molecular weight ([M.sub.r] 5 X [10.sup.6] to 6 X [10.sup.6)] LPA was purchased from Polysciences.


All samples and the UHG were obtained from the Laboratory Research Branch of the National Hansen s Disease Programs at the School of Veterinary Medicine, Louisiana State University (LSU). The samples consisted of pooled PCR products containing a 193-bp fragment from the RRDR of the rpoB gene of Mtb that were amplified from different Rmp-r strains of Mtb (minimum inhibitory concentration >2 mg/L rifampin). The PCR products used in this analysis were from sputum samples or from clinical strains of Mtb. The Rmp-r strains evaluated included YE14, YE67, and YE68, which contained single-base mutations that represent ~85% of all mutations giving rise to rifampin resistance in Mtb, and other strains giving less frequently observed mutations (see Table 1) (17). The PCR product PLN-2 contained a 193-bp fragment of the wildtype rifampin-susceptible (Rmp-s) rpoB gene from Mtb H37Rv (ATCC no. 25618; American Type Culture Collection) (19).

Initially, the UHG was synthetically prepared (GeneLab, LSU) and PCR-amplified using the rpo105 and rpo273 primer set to make a double-stranded 181-bp fragment (see Fig. 2). The fragment was cloned into PCR 2.1, a TA cloning vector (Invitrogen). The stock UHG was amplified from this plasmid by PCR, quantified spectrophotometrically at 260 nm, adjusted to 10 mg/L, and stored at -70[degrees]C until use. Compared with the wild-type rpoB sequence, the UHG contained four 3-bp deletions and four 2-bp substitutions, which were strategically placed to enhance electrophoretic discrimination of the common mutations encountered in Rmp-r Mtb clinical specimens.

Heteroduplex formation was performed as follows: 25 [micro]L of PCR product was mixed with an equal volume of UHG, heated to 94[degrees]C for 5 min, and allowed to cool slowly to room temperature over 30 min in a programmable Techne Genius thermocycler. The program was used to lower the temperature by 2[degrees]C every 30 s. The samples were then maintained at 25[degrees]C or at -20[degrees]C until electrokinetic loading onto the capillary.


Slab gel analysis was performed using Novex[TM] precast 10% TBE polyacrylamide gels from Invitrogen according to procedures published previously (20). The running buffer consisted of 0.6X TBE. DNA fragments were detected in ethidium bromide-stained gels (0.5 mg/L) by UV transillumination.


All capillaries for CE analysis were precoated with LPA and prepared in advance according to a modified Hjerten method (21). Briefly, 75-[micro]m (i.d.) capillaries were conditioned with 1 mol/L NaOH and 1 mol/L HCl and finally rinsed with copious amounts of deionized [H.sub.2]O. The capillaries were then filled with a 1:1 (by volume) solution of [gamma]-MAPS and methanol and allowed to stand overnight at 25[degrees]C. The following day, capillaries were rinsed with [H.sub.2]O and filled with a 4% acrylamide solution in 1 X TBE containing 1 mL/L of both TEMED and a 100 g/L ammonium persulfate solution. After being filled with the acrylamide solution, the capillaries were allowed to rest horizontally for 30 min at 25[degrees]C before being rinsed with [H.sub.2]O and dried for storage.


HEC solutions were prepared by dissolving the appropriate amount of polymer in 1 X TAE and applying heat and stirring overnight without boiling. The MC solution was prepared in 1X TBE as described elsewhere (8). Briefly, the polymer was added to distilled water, which was heated to 75[degrees]C. The solution was then placed on ice and stirred until clear (~15 min). The appropriate volume of 10 X TBE was added to make a final concentration of 1 X TBE. High-molecular weight LPA gels were prepared by dissolving the polymer in 1 X TBE with thorough mixing at 25[degrees]C for several hours. The staining dye was added to the entangled polymer solutions before the capillary was filled. Because of the high viscosity of high-molecular weight LPA, the capillary was filled with the gel under pressure (800 psi) in a pressure vessel that was designed in-house. The staining dye was added to the anode and cathode reservoir vials, which contained 1X TBE buffer.

Unless otherwise stated, the following conditions were used for electrophoresis. Separations in MC and HEC were performed in an LPA-coated capillary with an effective capillary length of 50 cm. Injection conditions were optimized to minimize the injection plug with respect to the length of the capillary to reduce zonal dispersion. Typically, injections were performed at -5 kV for 60 s, and the electric field strength used during the separation was 125 V/cm. When LPA was used as the matrix, the LPA-coated capillary length was 20 cm (effective length), and the electric field strength was 132 V/cm. Samples were injected for 1.5 min at -10 kV. In each case, the matrix was refilled into the capillary between each run. Electrophoresis was operated in the reverse-polarity mode (negative potential at the injection end of the capillary).

Electrophoretic separations with UV detection at 254 nm were performed on a commercially available capillary-based electrophoresis system, specifically, the Beckman P/ACE System 5510 (Beckman Instruments). Data acquisition was performed using the Beckman Gold Chromatographic software. Separations using LIF detection at 520 ran were performed using the Beckman P/ACE CE fitted with a LIF module consisting of an Argon ion laser ([[lambda].sub.ex] = 488 nm) and a 520 nm emission filter placed in front of the photomultiplier tube.

Results and Discussion


To maximize heteroduplex formation, the concentrations of PCR product and UHG must be optimized. Therefore, experiments using a titer or dilution series of PCR products and UHG were performed. Gel analysis of titer experiments was performed in which heteroduplex formation was carried out using 1:2 (one part sample in two parts total volume), 1:4, and 1:6 dilutions of both PCR product and UHG. It was determined that 1:2 dilutions of both the PCR product and UHG in [H.sub.2]O were optimal for heteroduplex formation as evidenced by clearly visible heteroduplex bands. Dilutions of 1:4 and 1:6 produced errant bands in addition to the primary heteroduplexes typically seen (data not shown).


The mutations shown in Table 1 were evaluated using slab gel electrophoresis (see Fig. 3). The utility of this assay is twofold: it simultaneously identifies Mtb apart from other mycobacterial and bacterial species and determines rifampin susceptibility (13). The presence of Mtb is definitively indicated by a 193-bp homoduplex in the electropherogram that results from highly specific PCR amplification. Both 181-bp UHG and 193-bp homoduplexes form at higher concentrations than the heteroduplexes because of increased thermodynamic stability induced by complete sequence complementarity in the homoduplexes.

Rifampin resistance was indicated by heteroduplex profiles, which are attributable to various base substitutions in the 193-bp PCR product and distinct from that observed for known Rmp-s strains (Fig. 3, lanes 2-6). The Rmp-s genotype (Fig. 3, lane 7) was not observed in the analysis of mutant strains because these samples were generated from Rmp-r strains of Mtb only. Acquired resistance typically occurs as a consequence of poor patient compliance to antidrug therapy. In this case, the resistant strain is selected for and the susceptible strain is eventually eliminated. Subsequent transmission of the resistant strain to another person leads to drug resistance from the outset and is known as primary resistance.

Similarities were seen between PLN-2 and YE67 heteroduplex profiles. The YE67 mutant, which contains the Asp516Va1 mutation, is generally difficult to discriminate from wild type by the HDA technique in the slab gel format. The reason for this appears to be that codon 516 is distant from any deletion or substitution in the generator; therefore, a mutation at this site causes a one-base mismatch that is not enhanced by the generator. Consequently, this mismatch makes little conformational difference in either heteroduplex compared with those generated by wild-type Mtb. Qualitative assessment of RR9 and O29 was also surprisingly difficult considering the close proximity of deletions or insertions in the generator. However, all mutants were distinguishable from the wild type on close inspection. For discussion purposes, the lower and upper heteroduplex bands have been designated as HD1 and HD2, respectively.



For the modification of HDA-UHG to the capillary format, we used 1.0% HEC, an entangled polymer commonly used to fractionate duplexed DNA (6, 9, 22, 23). We first investigated the addition of staining dyes (mono-intercalators) to the electrophoresis running buffer and their effects on the resolution of the single-base differences observed in the samples containing mutations in rpoB, using HDA and CE.

It has previously been demonstrated that the addition of mono-intercalators can improve the efficiency and resolution in CE compared with unstained DNAs (24), but addition of bis-intercalators can degrade separation efficiency because of various binding motifs (25, 26). In addition, intercalating dyes can cause slight unraveling and elongation of the duplex, potentially making HDA difficult when electrophoresis is used. Representative electropherograms are shown in Fig. 4 for the YE67 Rmp-r sample analyzed with unstained or with YOPRO1- or TOPRO5-stained double-stranded DNAs. Baseline separation of both the homoduplexes and heteroduplexes was achieved with the YOPRO1-stained DNA, whereas the unstained DNA produced baseline resolution of the homoduplexes but could not resolve the heteroduplexes. When the TOPRO5-stained sample was electrophoresed, the heteroduplexes comigrated and the homoduplexes did not achieve baseline resolution. DNA stained with YOPRO1 migrated more slowly than the unstained and TOPRO5-stained DNA. Because the addition of the staining dyes neutralizes the negative charge present in the duplexed DNA as well as slightly unraveling the duplex, the slower mobility of the YOPRO1-stained DNA was indicative of a higher loading of dye to the duplexed DNA and/or a higher binding affinity of the stain for the duplexed DNA (24).

Comparison of the selectivity values for the homoduplexes between the two staining dyes and the native DNA indicated no significant differences, which would be expected because the selectivity is determined primarily by the physical properties of the sieving matrix (see Table 2). However, the plate numbers were substantially higher for the YOPRO1-stained DNA, producing higher resolution factors for the YOPRO1-stained homoduplexes. In the case of the heteroduplexes, the selectivity values between the heteroduplexes were significantly larger for the YOPRO1-stained DNA, whereas for the unstained and TOPRO5-stained DNA, the heteroduplexes comigrated.


The apparent lower affinity of TOPRO5 for these mismatched duplexes is surprising because Scatchard analysis of the binding constant for fully matched duplexes using TAG (a structurally similar staining dye to TOPRO5) indicated an affinity comparable to that of YOPRO1 (24). The additional methine unit in the bridging methine chain could produce a lower binding affinity for these mismatched duplexes compared with the shorter dye, YOPRO1, because both dyes have the same net charge (+2). As such, it is clear that although staining dyes can improve the electrophoretic resolution in HDA for Mtb samples, judicious choice in the dye is necessary to augment discrimination of the heteroduplexed DNA.


Although UV detection at 254 nm can be implemented, the detection method suffers from poor sensitivity because of the short optical pathlength associated with CE. Consequently, LIF detection using intercalating dyes is the method of choice in applications requiring detection of low concentrations (25, 26). This is particularly important for HDA because the heteroduplexes form at lower concentrations and are typically less stable thermodynamically than the homoduplexes. As such, the population of heteroduplexes can be significantly smaller compared with the homoduplexes.

The YE67 mutation was analyzed with both UV detection at 254 nm and LIF detection at 520 nm using 0.3% MC as the sieving matrix and YOPRO1 as the staining dye. The resolution and selectivity between the UHG and HD1 (UHG/HD1) as well as between the heteroduplexes (HD1/HD2) were calculated in addition to the efficiency (plate numbers) of each fragment between the two detection strategies. Although the selectivity between fragments using either detection method remained the same (1.03 for UHG/HD1), efficiencies were slightly higher with LIF than with UV absorbance (2.23 and 1.15 X [10.sub.6] plates/m, respectively, for HD1). This produced a slight increase in the resolution between all fragments when LIF was used compared with UV absorbance (6.15 and 4.94, respectively, for UHG/HD1). An altered heteroduplex profile generally indicated the presence of mutant DNA; however, increased resolution between homoduplexes permitted increased certainty in the identification of Mtb, whereas enhanced resolution between heteroduplex fragments led to higher confidence in mutation discrimination.


We next evaluated LPA, HEC, and MC as potential separation matrices using LIF detection with YOPRO1 as the intercalating dye and YE67, the most difficult sample to discriminate. LPA has been used extensively as a separation matrix in CE because of the high resolutions achieved, whereas HEC and MC were chosen for their ease of use; their lower viscosity allows simple regeneration of the capillary between runs compared with LPAs (3). MC, in particular, was chosen for its ability to achieve high resolution at low concentrations of the entangled polymer (7).

For each matrix, the resolution and selectivity between UHG/HD1 and HD1/HD2 were calculated along with the efficiency of each fragment and are reported in Table 2. The UHG/HD1 resolution provides quantitative information about the separation of the homoduplexes from the heteroduplexes, with the UHG being chosen as an internal standard because of its invariable sequence between analyses. Alternatively, HD1/HD2 resolution values quantitatively describe the separation between the heteroduplexed peaks and the HDA profile for identifying rifampin resistance or susceptibility.

The matrix consisting of 3% LPA was able to exploit the conformational differences in the heteroduplexes compared with the homoduplexes as evidenced by the pronounced retardation of the heteroduplexes compared to the homoduplexes [resolution (R) >20 between UHG/ HD1]. However, this also caused decreased loading of the heteroduplexes onto the capillary because of biases induced by electrokinetic injection, producing lower signals for the heteroduplexes (data not shown). Significantly lower efficiencies were observed for the heteroduplexes; however, a poor signal-to-noise ratio added uncertainty to the determination of these values. Additionally, although use of LPA as the sieving matrix yielded baseline resolution between all fragments in less than 35 min, the high viscosity of the gel made on-line replacement of the matrix in the commercial instrument difficult because of the higher pressure required to replace the matrix.

With 1.0% HEC, all duplexes were baseline resolved and mutation discrimination was possible (Fig. 5), but a longer column length compared with the LPA matrix was required to achieve comparable resolution. The analysis required 35 min, but efficiencies were an order of magnitude lower than those achieved with 0.3% MC, as shown in Table 2. With MC, the optimized separation achieved efficiencies >1 X [10.sub.6] plates/m for all fragments in less than 30 min (Fig. 5). This matrix provided the best efficiency and resolution between heteroduplexes among those evaluated here, in agreement with previous reports that demonstrated high resolution at similar concentrations of this polymer (8,27). Therefore, MC was selected for use in further analyses.



Using the established conditions for base substitution detection with HDA-UHG in a capillary format, we analyzed all five mutations using 0.3% MC with YOPRO1 staining and LIF detection. PLN-2 was first analyzed in three trials to establish both reproducibility and wild-type conditions to be used for mutation discrimination. The analysis produced a SD of 0.10 in resolution between fragments with negligible SD in the selectivity between fragments (see Table 3). Additionally, baseline resolution (R >1.5) was obtained between all fragments, and efficiencies were >1 X [10.sub.6] plates/m. Using these data, we calculated 95% confidence limits for the resolution between the heteroduplexes and also between UHG/HD1 (see Table 3) for PLN-2 to allow comparison and identification of HDA mutant profiles.

Each Rmp-r sample containing a single-base substitution was then analyzed. Resolution between the heteroduplexes and UHG/HD1 in the Rmp-r samples was at least 1 SD above or below the 95% confidence intervals calculated and therefore could be classified as containing the Rmp-r genotype. Additionally, the mutation contained in YE67, which was the most difficult to qualitatively discriminate in the slab gel format, was easily identified. As in the slab gel format, similarities were observed in the HD1/HD2 resolution in PLN-2 (2.27 < R < 2.47) and RR9 (R = 2.49); however, comparison between UHG/HD1 resolutions established RR9 as an Rmp-r strain. Sample 029, which also showed similarities to PLN-2, was easily established as Rmp-r.

In conclusion, an efficient CE-based HDA-UHG method was developed for the detection of single-base pair substitutions in PCR-amplified gene fragments. By exploring the use of staining dyes, entangled polymer sieving matrices, and detection schemes, we could identify point mutations in the RRDR of rpoB of Mtb that give rise to the rifampin-resistant phenotype. Of the various methods evaluated, use of 0.3% MC as a polymer network and YOPRO1 staining with LIF detection appears to be optimal for this HDA application. The 0.3% MC entangled polymer solution yielded good separation and had low viscosity, which allowed it to be easily replaced in the capillary, thus permitting regeneration of the capillary and ease in operation. Additionally, the YOPRO1 staining dye gave the best resolution for the Rmp-s and Rmp-r heteroduplexes. In this application, the LPA coating appears to be suitable; however, the effects of the silanizing reagent used for LPA coating of the capillary were not investigated in this report, nor were alternative polymeric coatings, variables shown to have surprisingly profound effects on the CE separation in other HDA techniques (11). It has also been speculated that high salt concentrations in PCR products deteriorate LPA-coated capillaries. Diminished lifetimes of the LPA-coated capillary were not noted in this application, possibly because of the dilutions of the PCR products required during heteroduplex formation. The CE-based HDA-UHG substantially reduced the time required for analysis for Mtb and rifampin susceptibility using slab gel analysis, from 2.5 h to 30 min. It is highly amenable to automation and substantially reduces the amounts of reagents required; consequentially, it reduces the cost of the assay. Additionally, use of the UHG allowed improved discrimination of single-base substitutions compared with other methods reported. These improvements in the speed and throughput for HDA of genetic mutations may be of value for clinical applications as well as for further modification of this technique to the microchip format.

We would like to thank the National Institutes of Health for supporting this research (Grant R24 CA84625). G.A.T. acknowledges the Louisiana Board of Regents for a doctoral fellowship and Dr. Isiah M. Warner (LSU Department of Chemistry) for the gracious loan of an instrument. We also wish to thank Tana Pittman and Laynette Spring (Molecular Biology Research Department, Laboratory Research Branch, National Hansen s Disease Programs at the School of Veterinary Medicine, LSU) for excellent technical help.


(1.) Nataraj AJ, Olivos-Glander I, Kusukawa N, Highsmith WE Jr. Single-strand conformational polymorphism and heteroduplex analysis for gel-based mutation detection. Electrophoresis 1999; 20:1177-85.

(2.) Oda R, Clark R, Katzmann J, Landers J. Capillary electrophoresis as a clinical tool for the analysis of protein in serum and other body fluids [Review]. Electrophoresis 1997;18:1715-23.

(3.) Pariat YF, Berka J, Heiger DN, Schmitt T, Vilenchik M, Cohen AS, et al. Separation of DNA fragments by capillary electrophoresis using replaceable linear polyacrylamide matrices. J Chromatogr A 1993;652:57-66.

(4.) Heller C. Capillary electrophoresis of proteins and nucleic acids in gels and entangled polymer solutions. J Chromatogr A 1995;698: 19-31.

(5.) Siles B, Collier G, Reeder D, May W. The use of a new gel matrix for the separation of DNA fragments: a comparison study between slab gel electrophoresis and capillary electrophoresis. Appl Theor Electrophor 1996;6:15-22.

(6.) Bae Y, Soane D. Polymeric separation media for electrophoresis: cross-linked systems or entangled solutions. J Chromatogr A 1993;652:17-22.

(7.) Han F, Huynh BH, Ma Y, Lin B. High-efficiency DNA separation by capillary electrophoresis in a polymer solution with ultralow viscosity. Anal Chem 1999;71:2385-9.

(8.) Roed L, Arsky I, Lundanes E, Greibrokk T. Rapid and reproducible capillary electrophoretic separation of double-stranded DNA fragments in a simple methyl cellulosic sieving system. Chromatographia 1998;47:125-34.

(9.) Cheng J, Kasuga T, Mitchelson KR, Lightly ERT, Watson ND, Martin WJ, Atkinson D. Polymerase chain reaction heteroduplex polymorphism analysis by entangled solution capillary electrophoresis. J Chromatogr A 1994;677:169-77.

(10.) Gao Q, Yeung E. High-throughput detection of unknown mutations by using multiplexed capillary electrophoresis with poly(vinylpyrrolidone) solution. Anal Chem 2000;72:2499-506.

(11.) Tian H, Brody L, Mao D, Landers J. Effective capillary electrophoresis-based heteroduplex analysis through optimization of surface coating and polymer networks. Anal Chem 2000;72:5483-92.

(12.) Bidwell J, Wood N, Clay T, Pursall M, Culpan D, Evans J, et al. DNA heteroduplex technology. Adv Electrophoresis 1994;7:311-51.

(13.) Williams DL, Spring L, Gillis TP, Salfinger M, Persing DH. Evaluation of a polymerase chain reaction-based universal heteroduplex generator assay for direct detection of rifampin susceptibility of Mycobacterium tuberculosis from sputum specimens. Clin Infect Dis 1998;26:446-50.

(14.) Raviglione MD, Snider J, Kochi A. Global epidemiology of tuberculosis. Morbidity and mortality of a worldwide epidemic. JAMA 1995;273:220-6.

(15.) Bloom B, Murray C. Tuberculosis: commentary on a reemergent killer. Science 1992;257:1055-64.

(16.) World Health Organization. World Health Organization global tuberculosis programme. Geneva: WHO, 1997.

(17.) Ramaswamy S, Musser J. Molecular genetic basis of antimicrobial agent resistance in Mycobacterium tuberculosis: 1998 update. Tuber Lung Dis 1998;79:3-29.

(18.) Telenti A, Imboden P, Marchesi F, Lowrie D, Cole S, Colston MJ, et al. Detection of rifampicin-resistance mutations in Mycobacterium tuberculosis. Lancet 1993;341:647-50.

(19.) Miller L, Crawford J, Shinnick T. The rpoB gene of Mycobacterium tuberculosis. Antimicrob Agents Chemother 1994;38:805-11.

(20.) Williams DL, Limbers CW, Spring L, Jayachandra S, Gillis TP. PCR-heteroduplex detection of rifampin-resistant Mycobacterium Tuberculosis. In: Pershing D, ed. PCR protocols for emerging infectious diseases. Herndon, VA: ASM Press, 1996:122-9.

(21.) Hjerten SJ. High-performance electrophoresis: elimination of electroendosmosis and solute adsorption. J Chromatogr 1985;347: 191-8.

(22.) Grossman P, Soane D. Capillary electrophoresis of DNA in entangled polymer solutions. J Chromatogr 1991;559:257-66.

(23.) Quesada M. Replaceable polymers in DNA sequencing by capillary electrophoresis. Anal Biotechnol 1997;8:82-93.

(24.) Owens CV, Davidson YY, Kar S, Soper S. High resolution separation of DNA restriction fragments using capillary electrophoresis with near-IR, diode based, laser-induced fluorescence detection. Anal Chem 1997;69:1256-61.

(25.) Figeys D, Arriaga E, Renborg A, Dovichi N. Use of the fluorescent intercalating dyes POPO-3, YOYO-3, and YOYO-1 for ultrasensitive detection of double-stranded DNA separated by capillary electro phoresis with hydroxypropylmethyl cellulose and non-cross-linked polyacrylamide. J Chromatogr 1994;669:205-16.

(26.) Kim Y, Morris MD. Separation of nucleic acids by capillary electrophoresis in cellulose solution with mono- and bis-intercalating dyes. Anal Chem 1994;66:1168-74.

(27.) McGregor D, Yeung E. Optimization of capillary electrophoretic separation of DNA fragments based on polymer filled capillaries. J Chromatogr A 1993;652:67-73.


[1] Chemistry Department, Louisiana State University, Baton Rouge, LA 70803. [2] Molecular Biology Research Department, Laboratory Research Branch, National Hansen s Disease Programs at the School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA 70803.

[3] Nonstandard abbreviations: HDA, heteroduplex analysis; CE, capillary electrophoresis; LPA, linear polyacrylamide; HEC, hydroxyethyl cellulose; UHG, universal heteroduplex generator; TB, tuberculosis; Mtb, Mycobacterium tuberculosis; RRDR, rifampin resistance-determining region; LSU, Louisiana State University; Rmp-r and Rmp-s, rifampin-resistant and -susceptible; UV, ultraviolet; LIF, laser-induced fluorescence; MC, methyl cellulose; TEMED, N,N,N',N'-tetramethylethylenediamine; [gamma]-MAPS, [gamma]-[(methoxyacryl)oxy]propyltrimethoxysilane; TAE, Tris-acetate-EDTA; and TBE, Tris-borate-EDTA.

* Author for correspondence. Fax 225-388-3458; e-mail

Received February 20, 2001; accepted April 19, 2001.
Table 1. Samples investigated. (a)

Sample Base substitution Amino acid change

YE14 TCG[right arrow] or [vector]TTG Ser531Leu
YE67 GAC[right arrow] or [vector]GTC Asp516Val
YE68 CAC[right arrow] or [vector]TAC His526Tyr
RR9 TCG[right arrow] or [vector]TTG Ser522Leu
O29 CTG[right arrow] or [vector]CCG Leu533Pro
PLN-2 Wild type Wild type

(a) Each sample contains a 193-bp PCR-amplified fragment of the rpoB
gene associated with the development of rifampin resistance in Mtb and
a base substitution that produces the amino acid changes.

Table 2. Matrix comparison. (a)

 Selectivity Efficiency, [10.sup.5]

Matrix HD1/UHG HD2/HD1 UHG HD1 HD2

3.0% LPA 1.1176 1.0088 27.64 9.69 4.95
0.3% MC 1.026 1.0104 16.18 23.51 24.96
1.0% HEC 1.0434 1.0162 4.16 8.25 6.12


Matrix UHG/HD1 HD1/HD2

3.0% LPA 23.32 1.24
0.3% MC 6.15 2.76
1.0% HEC 5.54 2.33

(a) Comparison of the selectivity, efficiency, and resolution
obtained for sample YE67 using the matrices 3.0% LPA, 0.3% MC,
and 1.0% HEC. In each separation,
staining was performed using YOPRO1.

Table 3. Mutation discrimination. (a)

 Selectivity Efficiency, [10.sup.6]

PLN-2 Trials HD1/UHG HD2/HD1 UHG HD1 HD2

Mean 1.0241 1.0104 1.68 1.93 1.48
SD 0.0001 0.0000
YE14 1.0268 1.0097 2.25 2.33 2.42
YE67 1.026 1.0104 1.54 2.23 2.37
YE68 1.0251 1.0123 1.97 2.46 2.48
RR9 1.0256 1.0119 1.18 1.41 1.40
O29 1.0259 1.0111 2.33 2.31 2.41


PLN-2 Trials UHG/HD1 HD1/HD2

Mean 5.64 (5.49; 5.80) (b) 2.38 (2.28; 2.48) (b)
SD 0.14 0.09
YE14 7.08 2.63
YE67 6.15 2.76
YE68 6.49 3.39
RR9 5.07 2.49
O29 6.89 2.98

(a) PLN-2 (wild-type) was analyzed in three trials to obtain
reproducibility data and 95% confidence limits for the resolution
between homo- and heteroduplex as well detection. Each Rmp-r
as between heteroduplexes using 0.3% MC, 1 mmol/L YOPRO1, and LIF
sample was also analyzed, and corresponding data were included.

(b) Values in parentheses are upper and lower 95% confidence limits.
COPYRIGHT 2001 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Molecular Diagnostics and Genetics
Author:Thomas, Gloria A.; Williams, Diana L.; Soper, Steven A.
Publication:Clinical Chemistry
Date:Jul 1, 2001
Previous Article:Accumulation of free 3-hydroxy fatty acids in the culture media of fibroblasts from patients deficient in long-chain L-3-hydroxyacyl-CoA...
Next Article:Discontinuation of the bleeding time test without detectable adverse clinical impact.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters