Single-cell analysis avoids sample processing bias

Microscale separation tools such as capillary chromatography and capillary electrophoresis (CE) allow the study of metabolism in individual cells. In this work, we demonstrate that single-cell analysis describes metabolism more accurately than analysis of cellular extracts. We incubated HT29 cells (human colon adenocarcinoma) with a fluorescently labeled metabolic probe. This disaccharide, LacNAc, was labeled with a fluorescent dye, tetramethylrhodamine (TMR). The probe was taken up by the cells and metabolized to a number of products that retained the fluorescent label. We then split the cells into two batches. A cellular extract was prepared from one batch and analyzed by CE with laser-induced fluorescence (LIF) detection. The cells from the second batch were used for single-cell analysis by CE-LIF. Separation and detection conditions were identical for extract and single-cell analyses. We found that the electropherogram obtained by averaging the results from a number of single cells differed significantly from the cell extract electropherogram. Differences were due to sample processing during extract preparation. Disruption of the cells liberated enzymes that were compartmentalized within the cell, which allowed non-metabolic reactions to proceed. The accumulation of these non-metabolic products introduced a bias in the cell extract assay. During single-cell analysis, cells were lysed inside the capillary and the separation voltage was applied immediately to separate the enzymes from their substrates and prevent non-metabolic reactions. This paper is the first to report that CE analysis of single cells provides more accurate metabolic information than the CE analysis of a cellular extract.

Construction and evaluation of a capillary array DNA sequencer based on a micromachined sheath-flow cuvette

A capillary array electrophoresis DNA sequencer is reported based on a micromachined sheath-flow cuvette as the detection chamber. This cuvette is equipped with a set of micromachined features that hold the capillaries in precise registration to ensure uniform spacing between the capillaries, in order to generate uniform hydrodynamic flow in the cuvette. A laser beam excites all of the samples simultaneously, and a microscope objective images fluorescence onto a set of avalanche photodiodes, which operate in the analog mode. A high-gain transimpedance amplifier is used for each photodiode, providing high duty-cycle detection of fluorescence.

Single-cell analysis using capillary electrophoresis: influence of surface support properties on cell injection into the capillary

Capillary electrophoresis (CE) is an important tool of chemical cytometry. Whole-cell analysis using CE starts with cell injection into the capillary by either siphoning or electroosmosis. However, strong adherence of the cell to the support surface can prevent efficient cell injection and lead to irreproducible analysis. Here we evaluated several surfaces as potential cell supports for HT29 cells (human colon adenocarcinoma). These cells strongly adhered to the surface of untreated glass or polystyrene. Hydrophobic coating with dimethyldichlorosilane (DMS) or Sigmacote did not significantly reduce cell adhesion. In contrast, cell adhesion was reduced significantly when the surface was modified with hydrophilic polymers (hydrogels) such as poly(2-hydrohyethyl methacrylate) (PHEMA) and polyvinyl alcohol (PVA). In addition to their pronounced antiadhesive properties, PHEMA and PVA coatings were the most biocompatible (had highest survival of cells in contact with surface). Hydrogel-coated polystyrene plates were tested as a commercial alternative to hydrogel-coated glass slides. The cell adhesive properties of such plates were similar to those of PHEMA and PVA. However, the biocompatibility of the plates was lower than that of the other surfaces tested. Moreover, in contrast to PHEMA- and PVA-coated glass slides, the plates were sensitive to UV light and therefore should not be used when fluorescent image microscopy with UV excitation precedes CE. The analyses of the data obtained showed that PHEMA- and PVA-coated glass slides were the most suitable cell supports for cell injection into the capillary.

Instrumentation for chemical cytometry

Capillary electrophoresis is ideally suited to chemical analysis of individual cells. Small mammalian somatic cells (approximately 15 microns in diameter) can be analyzed by injecting the intact cell into a capillary, lysing the cell, separating and detecting the cellular components, and reconditioning the capillary prior to the next injection. In this paper, we report on technical improvements to single-cell analysis. We designed an inexpensive multipurpose single-cell injector that facilitates the following: (i) monitoring of injection, (ii) reproducible pressure- or electrokinetic-driven injection of the cell, (iii) complete cell lysis by SDS within 30 s of injection, and (iv) pressure-driven capillary reconditioning. Furthermore, we report on the analysis of glycosylation and glycolysis in single human carcinoma cells (HT29 cell line). The reliability and quality of the analysis is confirmed by comparing electropherograms from single cells and those from purified cell extracts.

Migration time correction for the analysis of derivatized amino acids and oligosaccharides by micellar capillary electrochromatography

Migration-time reproducibility is essential in the use of capillary electrophoresis to identify components in mixtures. Two methods based on the migration time of either one or two reference markers are proposed for improving migration time reproducibility. These methods were evaluated to determine the migration time reproducibility for phenylthiohydantoin-amino acids, fluorescein thiohydantoin-amino acids, and tetramethylrhodamine labeled oligosaccharides. In the best case, the relative standard deviation of the migration time was reduced from >3% without correction to <0.04% with the two-marker correction.

One-dimensional protein analysis of an HT29 human colon adenocarcinoma cell

A single HT29 human colon adenocarcinoma cell was introduced into a fused-silica capillary and lysed, and the protein content was fluorescently labeled with the fluorogenic reagent 3-(2-furoyl)quinoline-2-carboxaldehyde. The labeled proteins were separated by capillary electrophoresis in a submicellar buffer and detected by laser-induced fluorescence in a postcolumn sheath-flow cuvette. Several dozen components were resolved. A number of experiments were done to verify that these components were proteins. Most components of the single-cell electropherogram had the same mobility as components present in the 30-100 kDa fraction of a protein extract prepared from the cell culture. One component was identified as a approximately 100 kDa protein by co-injecting the sample with purified protein obtained from an SDS-PAGE gel. Protein expression varied significantly between cells, but the average expression was consistent with that observed from a protein extract prepared from 10(6) cells.

A multiple-capillary electrophoresis system for small-scale DNA sequencing and analysis

A five-capillary system has been developed for DNA sequencing and analysis. The post-column fluorescence detector is based on a sheath-flow cuvette. The instrument provides uniform and continuous illumination of the samples. The cuvette virtually eliminates cross-talk in the fluorescence signal between capillaries. Discrete single-photon counting avalanche photodiodes provide high efficiency light detection. The instrument has detection limits (3sigma) of 130 +/- 30 fluorescein molecules injected onto each capillary. Over 650 bases of sequence at 98.8% accuracy were generated in 100 min at 50 degrees C from M13mp18. Separation and detection of short tandem repeats proved efficient and accurate with the use of internal standards for direct comparison of migration times between capillaries.

Correlating cell cycle with metabolism in single cells: combination of image and metabolic cytometry

BACKGROUND

We coin two terms: First, chemical cytometry describes the use of high-sensitivity chemical analysis techniques to study single cells. Second, metabolic cytometry is a form of chemical cytometry that monitors a cascade of biosynthetic and biodegradation products generated in a single cell. In this paper, we describe the combination of metabolic cytometry with image cytometry to correlate oligosaccharide metabolic activity with cell cycle. We use this technique to measure DNA ploidy, the uptake of a fluorescent disaccharide, and the amount of metabolic products in a single cell.

METHODS

A colon adenocarcinoma cell line (HT29) was incubated with a fluorescent disaccharide, which was taken up by the cells and converted into a series of biosynthetic and biodegradation products. The cells were also treated with YOYO-3 and Hoechst 33342. The YOYO-3 signal was used as a live-dead assay, while the Hoechst 33342 signal was used to estimate the ploidy of live cells by fluorescence image cytometry. After ploidy analysis, a cell was injected into a fused-silica capillary, where the cell was lysed. Fluorescent metabolic products were then separated by capillary electrophoresis and detected by laser-induced fluorescence.

RESULTS

Substrate uptake measured with metabolic cytometry gave rise to results similar to those measured by use of laser scanning confocal microscopy. The DNA ploidy histogram obtained with our simple image cytometry technique was similar to that obtained using flow cytometry. The cells in the G(1) phase did not show any biosynthetic activity in respect to the substrate. Several groups of cells with unique biosynthetic patterns were distinguished within G(2)/M cells.

CONCLUSIONS

This is the first report that combined metabolic and image cytometry to correlate formation of metabolic products with cell cycle. A complete enzymatic cascade is monitored on a cell-by-cell basis and correlated with cell cycle.

Detection of duck hepatitis B virus DNA fragments using on-column intercalating dye labeling with capillary electrophoresis-laser-induced fluorescence

A rapid on-column DNA labeling technique is used to detect viral restriction DNA fragments by capillary electrophoresis-laser induced fluorescence detection. Intercalating dyes such as POPO3 or ethidium homodimer-2 are incorporated into the detection buffer. The cationic dyes migrate into the capillary during electrophoresis and bind to the oppositely migrating DNA fragments. A post-column sheath-flow fluorescence detector is used in the experiment. Excellent labeling efficiency is achieved at minimal background fluorescence by diluting the dyes to between 1 x 10(-7) M and 5 x 10(-7) M in a buffer with low ionic strength relative to the running buffer within the capillary. This dilute sheath-flow buffer allows stacking of dye molecules inside the capillary when an electric field is applied. Calibration curves using a series of DNA size markers (between 72 and 1353 base pairs) were linear over an order of magnitude in DNA concentration. Sensitivity also increased linearly with fragment length, and detection limits ranged from 4 x 10(-14) M to 5 x 10(-13) M for the size-standards. Analysis of cloned viral DNA using duck hepatitis B virus demonstrated a concentration detection limit of 3.9 x 10(-16) M. Last, the technique produced very high separation efficiency, 14 x 10(6) theoretical plates which is greater than 47 x 10(6) plates m-1, for the duck hepatitis B viral genome.

Labeling effects on the isoelectric point of green fluorescent protein

We studied the effects of fluorescent labeling on the isoelectric points (pI values) of proteins using capillary isoelectric focusing with laser-induced fluorescence detection (cIEF-LIF). Specifically, we labeled green fluorescent protein (GFP) from the jellyfish Aequorea victoria with the fluorogenic dye 3-(2-furoyl)quinoline-2-carboxaldehyde (FQ). cIEF-LIF was used to monitor the native fluorescence of GFP and showed pI changes in GFP's FQ-labeled products. Multiple labeling of GFP with FQ produced a series of products with pI values shifted towards a low pH. We verified cIEF-LIF results with traditional slab gel IEF. Our cIEF-LIF technique can routinely detect 10(-11) M of FQ-labeled protein, whereas traditional slab gel IEF with silver stain detection gives detection limits of 10(-7) M in the same samples.

Capillary electrochromatography with thermo-optical absorbance detection for the analysis of phenylthiohydantoin-amino acids

Capillary columns were packed with 3 microns C18 stationary phase, interfaced with an ultraviolet-laser based thermo-optical absorbance detector, and evaluated for separation of a mixture of phenylthiohydantoin-amino acids. These columns demonstrated consistent performance with a relative standard deviation (RSD) for migration time of less than 1.5% and a separation efficiency of 216,000 plates/m for the electroosmotic flow marker, thiourea. The thermo-optical absorbance detector was based on a 248 nm krypton-fluoride excimer laser. Detection limits (3 sigma) ranged from 1.6 to 4.8 x 10(-7) M phenylthiohydantoin (PTH)-amino acid injected onto the column, which is a factor of three superior to those obtained in micellar electrokinetic chromatographic analysis of these compounds. A mixture of 17 PTH amino acids was injected onto the capillary; 13 components were nearly baseline resolved in 14 min.

Species identification and phylogenetic relationships based on partial HSP60 gene sequences within the genus Staphylococcus

The phylogenetic relationships among 36 validly described species or subspecies within the genus Staphylococcus were investigated by cloning and sequencing their 60 kDa heat-shock protein (HSP60) genes using a set of universal degenerate HSP60 PCR primers. The cloned partial HSP60 DNA sequences from nine Staphylococcus aureus strains were highly conserved (97-100% DNA sequence similarity; mean 98%), indicating that the HSP60 gene of multiple isolates within the same species have little microheterogeneity. At the subspecies level, DNA sequence similarity among members of S. aureus, Staphylococcus schleiferi, Staphylococcus cohnii and Staphylococcus capitis ranged from 91 to 98%. At the interspecies level, sequence similarity among 23 distinct species of staphylococci ranged from 74 to 93% (mean 82%). By comparison, the highest sequence similarity of Bacillus subtilis and Escherichia coli with members within the genus Staphylococcus was only 70 and 59%, respectively. Importantly, phylogenetic analysis based on the neighbour-joining distance method revealed remarkable concordance between the tree derived from partial HSP60 gene sequences and that based on genomic DNA-DNA hybridization, while 16S rRNA gene sequences correlated less well. The results demonstrate that DNA sequences from the highly conserved and ubiquitous HSP60 gene offer a convenient and accurate tool for species-specific identification and phylogenetic analysis of staphylococci.

Single cell studies of enzymatic hydrolysis of a tetramethylrhodamine labeled triglucoside in yeast

Several hundred molecules of enzyme reaction products were detected in a single spheroplast from yeast cells incubated with a tetramethylrhodamine (TMR) labeled triglucoside, alpha-d-Glc(1-->2)alpha-d-Glc(1-->3)alpha-d-Glc-O(CH2)8CONHCH2- CH2NH- COTMR. Product detection was accomplished using capillary electrophoresis and laser induced fluorescence following the introduction of a single spheroplast into the separation capillary. The in vivo enzymatic hydrolysis of the TMR-trisaccharide involves at least two enzymes, limited by processing alpha-glucosidase I, producing TMR-disaccharide, TMR-monosaccharide, and the free TMR-linking arm. Hydrolysis was reduced by preincubation of the cells with the processing enzyme inhibitor castanospermine. Confocal laser scanning microscopy studies confirmed the uptake and internalization of fluorescent substrate. This single cell analysis methodology can be applied for the in vivo assay of any enzyme with a fluorescent substrate.

Nonaqueous capillary electrophoretic separation and thermo-optical absorbance detection of five tricyclic antidepressants and metabolism of amitriptyline by Cunninghamella elegans

We developed a technique based on nonaqueous capillary electrophoresis and laser-based thermo-optical absorbance detection to assay five antidepressants with similar structures and mass-to-charge ratios. A mixture of methanol and acetonitrile with ammonium acetate was essential to achieve baseline resolution of these compounds. We investigated the effects of ammonium acetate concentration, temperature, applied voltage, and capillary length on separation efficiency. The nonaqueous capillary electrophoresis and laser-based thermo-optical absorbance detection technique was used to study the metabolism of amitriptyline by Cunninghamella elegans. Sample preparation procedures were simplified for fast screening of the parent drug and its metabolites. Reproducible electropherograms were obtained from replicate cultures of C. elegans growing in the presence of amitriptyline.

On-line nonaqueous capillary electrophoresis and electrospray mass spectrometry of tricyclic antidepressants and metabolic profiling of amitriptyline by Cunninghamella elegans

An on-line nonaqueous capillary electrophoresis-electrospray mass spectrometry (ESI-MS) technique was developed using a commercial ion spray interface. The nonaqueous capillary electrophoresis ESI-MS system was used to profile tricyclic antidepressants of similar structures and mass-to-charge ratios. We found that pure methanol can be used as a sheath liquid to obtain stable ion spray from nonaqueous capillary electrophoresis. The flow rate of the coaxial nebulizing gas affected baseline signals, separation efficiency, and migration times. Other nonaqueous capillary electrophoresis operating conditions and electrospray parameters were optimized for enhanced baseline separation and high sensitivity detection. The effect of sample stacking on separation and detection was evaluated. The calculated detection limits were approximately 3 pg injected onto the capillary. ESI mass spectra of tricyclic antidepressants from a single quadrupole MS were obtained and elucidated. The information was used to propose fragmentation pathways of the tricyclic antidepressants. The method was also used to analyze the metabolites of amitriptyline produced by the fungus Cunninghamella elegans. Sixteen metabolites were detected and most of them were tentatively identified as demethylated and/or hydroxylated, and/or N-oxidized products.

Picomolar analysis of proteins using electrophoretically mediated microanalysis and capillary electrophoresis with laser-induced fluorescence detection

We report a method for the analysis of picomolar concentration proteins using electrophoretically mediated microanalysis (EMMA) to label proteins on-column with a fluorogenic reagent. Labeling is followed by capillary zone electrophoresis separation and postcolumn detection based on laser-induced fluorescence. The method provides a concentration detection limit (3 sigma) of 3 x 10(-13) M for conalbumin. The method provides separation efficiency of 300,000 theoretical plates. Protein extract from a human colon adenocarcinoma cell line generated a dozen major components and many minor components in a 12-min separation; the protein extract from 2.5 cells was used for this analysis. When compared to UV absorbance detection, the EMMA method provides 7,000,000-fold improvement in detection limit.

General protease assay method coupling solid-phase substrate extraction and capillary electrophoresis

Capillary electrophoresis with laser-induced fluorescence detection was used to develop a universal, highly specific protease assay. In this method, a peptide, biotinylated at the N-terminus, is labeled with fluorescein at a lysine residue near the C-terminus. Impurities are removed from the fluorescence labeling mixture by solid-phase extraction of the substrate on immobilized streptavidin, followed by extensive washing. The purified fluorescent substrate is dissociated from the streptavidin and incubated with the protease. The peptide sequence between the biotin and fluorescent label contains the cleavage sequence of the protease of interest. After cleavage, the fluorescent product does not contain a biotin group. A second solid-phase extraction is used to remove unreacted substrate to dramatically lower the background signal. The product is detected by capillary electrophoresis, which provides powerful discrimination against products generated by nonspecific proteases. With chymotrypsin as a test protease, product was detected with as little as 10 pg/mL (4.6 x 10(-13) M) chymotrypsin, or 5 amol of enzyme in the 10-microL sample volume.

Sodium dodecyl sulfate-capillary electrophoresis of proteins in a sieving matrix utilizing two-spectral channel laser-induced fluorescence detection

We report a method for protein labeling, separation by capillary electrophoresis in a polymer sieving matrix, and detection by laser-induced fluorescence. Different dyes are used to label standard and sample proteins. A two-spectral channel detector resolves fluorescence from the sample and standards. Comparison of the migration time of the sample and standards permits the precise determination of molecular weight, irrespective of variations in run-to-run migration times.

Cationic and anionic polymeric additives for wall deactivation and selectivity control in the capillary electrophoretic separation of proteins in food samples

Both cationic and anionic polymeric additives were used for the capillary electrophoretic separation of proteins in food samples. The cationic polyelectrolyte polydiallyldimethylammonium chloride was more effective in minimizing protein-wall interactions at pH 3 than at pH 7, presumably due to greater repulsion between the adsorbed polymer and proteins. Improved resolution was observed in the presence of the co-additive sodium octanesulphonate, presumably due to ion-pairing interactions with protein sample components. The anionic polymer dextran sulfate produced relatively high efficiencies, 120,000-180,000 theoretical plates, for protein separation, presumably because the polymer adsorbed to the capillary wall, rendering the surface more hydrophilic. In addition to reduced protein-wall interactions, improved resolution was observed, presumably due to analyte-polymer ion-exchange/ion-pairing interactions. When poly(vinyl sulphonic acid) was used instead of dextran sulfate, broader profiles were obtained and fewer components were resolved, presumably due to reduced wall deactivation that is related to the lower hydrophilicity of poly(vinyl sulphonic acid).

Surface modification based on Si-O and Si-C sublayers and a series of N-substituted acrylamide top-layers for capillary electrophoresis

Two approaches were used to prepare a series of surface-modified capillaries. In the first, a sublayer was formed by coupling gamma-methacryloxypropyltrimethoxysilane to the surface silanol groups forming an SI-O bond; a top layer was then formed by polymerizing acrylamide in the capillary, which reacted with the sublayer. In the second approach, a sublayer was formed by silanol chlorination, followed by Grignard coupling of vinylmagnesium bromide to form an Si-C bond at the surface; a top layer was formed by polymerizing either acrylamide (AA), dimethylacrylamide (DMA), N-acryloylaminoethoxyethanol (AAEE), or N-acryloylaminopropanol (AAP) onto the sublayer. The Si-Cpoly(AA) capillaries were more stable and produced an approximately 10-fold lower electroosmotic flow compared to the Si-O-poly(AA) capillaries. The Si-C sublayer was used to compare the performance of all four top layers. Electroosmotic flow decreased in the order: Si-O-poly(AA), Si-C-poly(AA), Si-Cpoly(AAEE), Si-C-poly(DMA), and Si-C-poly(AAP). Si-C-poly(AA) showed evidence of irreversible degradation at pH 9 already after 40-50 runs. Si-C-polyAAP-coated capillaries demonstrated superior efficiency and migration time reproducibility for a number of alkaline proteins and for fluorescently labeled ovalbumin. Excellent performance was maintained, in the case of poly(AAP), for a least 300 runs (of 30 min duration) at pH 9.0.

DNA sequencing by capillary electrophoresis

Capillary electrophoresis has been under development for DNA sequencing since 1990. This development has traveled down two parallel tracks. The first track studied the details of DNA separation by gel electrophoresis. Early work stressed rapid separations at high electric fields, which reached the extreme of a 3.5 min sequencing run at 1200 V/cm. While fast separations are useful in clinical resequencing applications for mutation detection, long read-length is important in genomic sequencing. Unfortunately, sequence read-length degrades as electric field and sequencing speed increases; this tradeoff between read-length and sequencing speed appears to be a fundamental result of the physics of DNA separations in a polymer. The longest sequence sequencing read-lengths have been obtained at modest electric fields, high temperature, and with low concentration noncrosslinked polymers. In parallel with our understanding of DNA separations, the second track of DNA sequencing development considered the design of large-scale capillary instruments, wherein hundreds of DNA samples can be sequenced in parallel. Real-world application of these very high throughput capillary electrophoresis systems will require significant investment in sample preparation technology.

Study of the enzymatic transformation of fluorescently labeled oligosaccharides in human epidermoid cells using capillary electrophoresis with laser-induced fluorescence detection

Isomeric oligosaccharides of both beta Gal(1-->3)beta GlcNAc (type I) series and beta Gal(1-->4)beta GlcNAc (type II) series were studied by using capillary electrophoresis (CE) with laser-induced fluorescence (LIF) detection. A mixture of phenylboronic acid and sodium tetraborate was used in the CE running buffers to improve the electrophoretic separation of the oligosaccharides. Both series of the tetramethylrhodamine (TMR)-labeled substrates [beta Gal(1-->3)beta GlcNAc-O-TMR and beta Gal(1-->4)beta GlcNAc-O-TMR) and their potential enzymatic products were baseline resolved using CE. The high resolution provided by CE and the excellent detection limit (8.10(-23) mol, or 50 molecules) by LIF allowed for the determination of minor enzyme products in the presence of excess unreacted substrate. The action of competing enzymes acting on the common type I sequence was monitored after the incubation of human epidermoid carcinoma cells (A431) with a fluorescent substrate (beta Gal(1-->3)beta GlcNAc-O-TMR). The CE-LIF analyses showed the formation of both synthetic and hydrolytic products, suggesting the actions of glycosyltransferases and glycosidases in the cells.

Detection of Aequorea victoria green fluorescent protein by capillary electrophoresis laser induced fluorescence detection

Aequorea victoria green fluorescent protein was assayed by capillary electrophoresis using post-capillary laser-induced fluorescence detection in a sheath flow cuvette. The limit of detection was 3.0 x 10(-12) M protein in an injection volume of 17 nL, corresponding to a mass of 3100 molecules.

Double coupling Edman chemistry for high-sensitivity automated protein sequencing

Capillary electrophoresis with laser-induced fluorescence detection (CE-LIF) is a promising new method for the analysis of protein sequencing products. It gives 10 zmol (1 zmol = 10(-21) mol) limits of detection (3 sigma) for fluorescein thiohydantoin (FTH) amino acids. We have developed a separation for the (FTH)-amino acid products generated from 18 of the 20 coded amino acids. The extremely low volume requirement associated with CE-LIF makes it incompatible with commercial sequencers. For this reason, we have also been developing a miniaturized sequencer that can be more easily coupled to our detection system. Both the CE-LIF system and the miniaturized sequencer are described.

Pseudo-coulometric loading in capillary electrophoresis DNA sequencing

While injection volumes in capillary electrophoresis are typically in the nanoliter range, it is difficult to physically prepare and manipulate samples much smaller than a microliter. As a result, only a small fraction of the analyte contained with the sample volume is transferred to the capillary. This problem is particularly acute in DNA sequencing applications, where on-column stacking is difficult and where the sequencing sample is relatively expensive to prepare. We report a method that transfers 75% of the DNA contained within a 3 microliters sample onto a capillary for DNA sequencing. This method relies on the use of very low ionic strength formamide to resuspend the DNA after an ethanol precipitation. The use of low ionic strength formamide achieves two tasks. First, it produces a very high resistance sample, which increases the voltage drop across the sample and decreases the field across the capillary. This electric field manipulation ensures that DNA fragments do not migrate down the capillary during the loading process, allowing long injection periods without excessive band-broadening. Second, the low ionic strength of the formamide increases the transference number of the DNA; more of the current passing through the injection tip of the capillary is carried by DNA fragments. In the limit of complete elimination of impurity ions from the loading solvent, current passing through the sample is carried only by DNA fragments and loading becomes a coulometric process.