Capillary electrophoresis procedures for serum protein analysis: comparison with established techniques

Methods using automated capillary electrophoresis (CE) instrumentation are available for serum protein electrophoresis with monoclonal band quantitation, isoelectric focusing and sodium dodecyl sulphate-polyacrylamide gel electrophoresis separations. The advantages of CE over previous gel methods relate to the time and labour saved by the automated instrumentation. High pI monoclonal bands and cryoglobulin specimens can be successfully analysed by CE. However, if the CE application uses a standard company supplied kit, then the cost savings are often negated by the high cost of the kit. Improvements such as the inclusion of both a UV-Vis as well as a fluorescence detector as standard within the one commercial instrument, the production of coated IEF capillaries with a useful life of at least 100 samples, and the introduction of a capillary array into all commercial instrumentation would ensure greater use of CE within both the clinical and other protein laboratories.     

Assessing the efficiency of free light chain assay in monitoring patients with multiple myeloma before and after autologous stem cell transplantation along with serum protein electrophoresis and serum protein immunofixation

Monoclonal gammopathies are a group of disorders, referred to as paraproteinaemias, dysproteinaemias or immunoglobulinopathies, associated with monoclonal proliferation of plasma cells. Monoclonal immunoglobulin secreted by these cells is an indicator of clonal proliferation. The aim of this study is to analyze the efficiency of three methods: serum protein electrophoresis (SPE), serum protein immunofixation (IFE) and FLC (free light chain) assay for the diagnosis and monitoring of the tumor burden in multiple myeloma. In this study we have presented the dynamic evolution of 7 patients with intact immunoglobulin multiple myeloma (IIMM) (2 IgG, kapa; 3 IgG, lambda; 1 IgA, kappa; 1 IgA, lambda) and 2 patients with light chain multiple myeloma before and after autologous peripheral blood stem cell transplantation (PBSCT). All 7 patients fulfilled the four criteria for the diagnosis of IIMM: bone marrow plasma cells exceeding 20%, lytic bone lesions, identification and quantification of M protein by scanning densitometry of electrophoresis gels, IFE (immunofixation protein electrophoresis) confirmed and typed the M protein. All patients had been given cytotoxic chemotherapy (VAD or VELCADE) before autologous (PBSCT). In two of the patients with IIMM both SPE and kappa/lambda ratio fell towards normal range after autologous PBSC and both reported a relapse of the disease after 23 months and 19 months respectively. SPE could not normalize after chemotherapy and transplantation in three patients with IIMM, the kappa/lambda ratio being the only marker used to monitor the tumor kill. In one patient the kappa/lambda ratio could not normalize even after PBSCT still indicating the presence of plasma cell disorder at the time when IFE was still negative. 16 months after PBSCT both SPE and FLC indicated a relapse of the disease. Classical SPE failed to demonstrate the presence of M-protein in light chain multiple myeloma, the diagnosis being established by using IFE and the FLC assay. Because IFE is a qualitative method and its interpretation may be sometimes subjective, FLC was the only method used to follow the disease course. The measurement of kappa/lambda ratio proved to be more sensitive than SPE, IFE and the levels of free light chains kappa or lambda individually indicating whether the treatment is effective or not.     

Determination of nicotine and its metabolites in urine by solid-phase extraction and sample stacking capillary electrophoresis-mass spectrometry

The combination of capillary electrophoresis (CE) and mass spectrometry (MS) with solid-phase extraction (SPE) has been used for the identification of nicotine and eight of its metabolites in urine. The recovery of cotinine from cotinine-spiked urine, by C18 SPE, was found to be 98%. Smokers urine (200 ml) was preconcentrated 200-fold via SPE prior to analysis. The sample stacking mode of CE, when compared to capillary zone electrophoresis, was shown to improve peak efficiency by 132-fold. The combination of hydrodynamic and electrokinetic injection was studied with sample stacking/CE/MS. The on-column limits of detection (LOD) of nicotine and cotinine, by this technique, were found to be 0.11 and 2.25 microg/ml, respectively. Hence, LODs of nicotine and cotinine in urine after 200-fold preconcentration were 0.55 and 11.25 ng/ml, respectively.     

Three methods of capillary electrophoresis compared with high-resolution agarose gel electrophoresis for serum protein electrophoresis

We assessed the BioFocus 2000 capillary electrophoresis instrument for use in a routine clinical laboratory. We examined 210 serum samples received for serum protein electrophoresis by four methods: (1) The Bio-Rad HR015EC high-resolution serum protein kit on the BioFocus; (2) the Jenkins–Guerin (JG) method on the Applied Biosystems 270A HT Capillary Electrophoresis System (JG-ABI); (3) the Jenkins–Guerin method using the BioFocus (JG-BF); and (4) the quantitation of monoclonal bands found in 76 of the 210 samples was assayed by Helena Titan Hi-Res agarose gel electrophoresis (HRAGE). The correlation coefficient between the three sets of capillary electrophoresis monoclonal band results and the Helena quantitation was 0.92 or better. Although the quantitative comparison of monoclonal bands by HR015EC was very good, the lack of sharpness of monoclonal bands using the HR015EC kit meant our preference was to use the JG method on either the ABI or on the Biofocus.     

An unusual case of Waldenstr?m macroglobulinemia with half molecules of IgG in serum and urine

Immunochemical studies are described in an unusual case of Waldenstr?m's macroglobulinemia. Two monoclonal Igs (whole IgG1/kappa and IgG1/kappa half molecules) occurred in the serum in addition to the IgM monoclonal protein. Protein electrophoresis of the serum showed a monoclonal component in the gamma region, and the immunoelectrophoresis allowed detection of a monoclonal IgM/kappa and another abnormality represented by a double precipitin line in serum and urine, observed when antiserum anti IgG was used. The abnormal proteins were purified and further analyzed. The IgG-related proteins were whole four chains IgG monoclonal molecules, 1/2 IgG monoclonal molecules, composed of one heavy and one light chain, and residual polyclonal IgG. The half molecules were antigenically deficient with respect to normal IgG. The idiotypic analysis showed that the three monoclonal proteins shared idiotypic determinants. This patient had clinical and morphological findings of Waldenstr?m's macroglobulinemia and, as observed in other cases, the formation of half molecules was not associated with a distinct clinical syndrome.     

Electrophoretic characteristics of monoclonal immunoglobulin G of different subclasses

Monoclonal IgG are commonly observed in various B cell disorders, of which multiple myeloma is the most clinically relevant. In a series of serum samples, we identified by immunofixation 73 monoclonal IgG, including 63 IgG(1), 4 IgG(2), 5 IgG(3), and 1 IgG(4). The light chains were of kappa type in 45 cases, and of lambda type in 28 cases. These monoclonal IgG were further characterized by high resolution two-dimensional polyacrylamide gel electrophoresis (2-DE) in various isoelectric focusing conditions, as well as by 3-DE (2-DE of the proteins extracted from agarose after serum protein agarose electrophoresis). After 2-DE, 38 out of 73 monoclonal gamma chains (52%) were visualized using immobilized pH 3-10 gradients for isoelectric focusing. In 6 cases (8%), gamma chains were only detected using alkaline immobilized pH 6-11 gradients. In 3 cases (4%), 3-DE revealed monoclonal gamma chains hidden by polyclonal gamma chains. Finally, in 26 cases (36%), no monoclonal gamma chains were clearly visualized. Sixty-one monoclonal light chains (84%) were detected using immobilized pH 3-10 gradients, whereas 12 (16%) were not. Monoclonal gamma chains and light chains were highly heterogeneous in terms of pI and M(r). However, a statistically significant correlation (P<0.05) was observed between the position of the monoclonal IgG in agarose gel and the pI of their heavy and light chains (R=0.733, multiple linear regression). Because of the extreme diversity of their heavy and light chains, it appears that a classification of monoclonal IgG based only on their electrophoretic properties is not possible.     

Preparation and characterization of polyclonal and monoclonal antibodies against human aromatase cytochrome P-450 (P-450AROM), and their use in its purification

Aromatase cytochrome P-450 (P-450AROM) was partially purified from human placental microsomes by hydrophobic affinity chromatography using Phenyl-Sepharose and ion-exchange chromatography on DEAE-cellulose. The resulting preparation had a specific activity of 2 nmol/mg protein with respect to cytochrome P-450 content and displayed a type I difference spectrum upon addition of the substrate androstenedione. When the cytochrome P-450-enriched fractions were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and stained with Coomassie blue, there was an enrichment of two proteins having apparent molecular weights of 50,000 and 55,000. The bands containing these proteins were removed from unstained polyacrylamide gels and injected separately or together into three rabbits. An aliquot of the serum or an immunoglobulin (IgG) fraction prepared from the serum of the rabbit injected with the 55-kDa band or with both the 50- and 55-kDa bands inhibited aromatase activity of human placental microsomes by 80%; this IgG had no effect on 17 alpha-hydroxylase or 21-hydroxylase activities of human fetal adrenal microsomes. In contrast, the serum of the rabbit injected with the 50-kDa band had little capacity to inhibit placental aromatase activity. By immunoblot analysis, it was found that the IgG from the serum of the rabbit immunized with the 55-kDa protein bound specifically to a protein of 55 kDa in human placental microsomes. Monoclonal antibodies were prepared from a hybridoma cell line derived from the spleen cells of mice immunized against the 55-kDa protein. The monoclonal IgG was covalently linked to a Sepharose 4B column and was used for immunoaffinity chromatography of cytochrome P-450AROM. The finding that cytochrome P-450 and the 55-kDa protein were selectively retained by the affinity column and eluted with NaCl (2 M) and glycine (0.2 M, pH 3.0) and that this fraction contained aromatase activity upon reconstitution with purified NADPH-cytochrome P-450 reductase and phospholipid, is indicative that the 55-kDa protein is indeed cytochrome P-450AROM. These findings are also indicative that both the monoclonal and polyclonal IgGs are specific for human cytochrome P-450AROM.     

On-line coupling of size exclusion chromatography and capillary electrophoresis via solid-phase extraction and a Tee-split interface

An on-line size exclusion chromatography (SEC)-solid-phase extraction (SPE)-capillary electrophoresis (CE) system using a Tee-split interface has been developed for the analysis of peptides in biological fluids. The SEC column fractionates the sample by molecular size and the low-molecular-weight fraction, which contains the peptides, is directed to a C(18) SPE microcolumn, where the peptides are trapped and concentrated. The SPE column is desorbed with 425 nL acetonitrile and the effluent is sent to the Tee-split interface, which hydrodynamically splits (1:40) the flow and, thus, allows appropriate injection of analytes into the CE system. The performance of the system is investigated by the analysis of enkephalins in cerebrospinal fluid (CSF). It is demonstrated that the SEC step efficiently removes potentially interfering proteins, permitting reproducible SPE and CE. The total system provides efficient separations of the enkephalins with plate numbers up to 100,000. Concentration limits of detection (S/N = 3) for the peptides are about 100 ng/mL for injection of 20 microL spiked CSF samples. Plots of enkephalin peak areas versus concentration showed good linearity over the 0.25-10 microg/mL range (R2 > or = 0.985). Repeatability of migration time and peak area was within 2% and 10% R.S.D., respectively.     

A monoclonal IgG4 (lambda) with factor V inhibitory activity.

A monoclonal IgG4 (lambda) with inhibitory activity to human coagulation factor V was isolated from the serum of a patient with a fatal hemorrhagic diathesis by using a combination of DE-52 ion exchange chromatography and isoelectric focusing techniques. Using the criteria for defining a monoclonal immunoglobulin of restricted mobility on protein electrophoresis, immunoelectrophoresis, and isoelectric focusing, as well as neutralization with class, subclass, and light chain type antisera, we are the first to demonstrate a factor V inhibitor as a monoclonal IgG4 (lambda) detectable in serum or plasma.     

Capillary electrophoresis for clinical problem solving: analysis of urinary diagnostic metabolites and serum proteins

Many clinical laboratories employ gas chromatography-mass spectrometry (GC-MS) and high-performance liquid chromatography (HPLC) to detect abnormal compounds occurinng in urine and serum due to disease. The methods, particularly GC-MS, often require laborious sample pre-treatment, and separation times may exceed an hour. We describe the use of capillary electrophoresis (CE) equipped a with a diode-array detector in an attempt to improve the efficiency of an analytical system routinely used for diagnosis of human metabolic disease. It was found that urine samples could be injected directly onto the CE instrument without any pre-treatment, and over 50 metabolites were separated in 15 min. Identification of abnormal metabolites was based on migration times and characteristic diode-array spectra. The method readily diagnosed adenolysuccinase deficiency, 5-oxoprolinuria, propionic acidemia and disorders having orotic acid as diagnostic metabolite (e.g. the HHH-syndrome). The results show that CE may become a useful additional tool for diagnosis of metabolic disease. In a different project CE was used to study sera from the Janus-bank. This large serum bank comprises samples collected at intervals from nearly 300 000 blood donors. As the sera are stored at −25°C and not at a lower temperature, a major concern has been the stability of the specimens. GC-MS, 2D-protein electrophoresis, certain immunological assays and enzyme measurements have previously been used to evaluate the stability of the sera. We can now also show that the protein profile, as determined by CE, is remarkably stable even after 22 years of storage. The results moreover confirmed that the CE-method and traditional gel electrophoresis gave almost identical results, except for small amounts of fibrinogen which did not show up on the CE-pattern.     

Evaluation by ELISA of anisakis simplex larval antigen purified by affinity chromatography

In order to improve the specificity and sensitivity of the techniques for the human anisakidosis diagnosis, a method of affinity chromatography for the purification of species-specific antigens from Anisakis simplex third-stage larvae (L3) has been developed. New Zealand rabbits were immunized with A. simplex or Ascaris suum antigens or inoculated with Toxocara canis embryonated eggs. The IgG specific antibodies were isolated by means of protein A-Sepharose CL-4B beads columns. IgG anti-A. simplex and -A. suum were coupled to CNBr-activated Sepharose 4B. For the purification of the larval A. simplex antigens, these were loaded into the anti-A. simplex column and bound antigens eluted. For the elimination of the epitopes responsible for the cross-reactions, the A. simplex specific proteins were loaded into the anti-A. suum column. To prove the specificity of the isolated proteins, immunochemical analyses by polyacrylamide gel electrophoresis were carried out. Further, we studied the different responses by ELISA to the different antigenic preparations of A. simplex used, observing their capability of discriminating among the different antisera raised in rabbits (anti-A. simplex, anti-A. suum, anti-T. canis). The discriminatory capability with the anti-T. canis antisera was good using the larval A. simplex crude extract (CE) antigen. When larval A. simplex CE antigen was loaded into a CNBr-activated Sepharose 4B coupled to IgG from rabbits immunized with A. simplex CE antigen, its capability for discriminate between A. simplex and A. suum was improved, increasing in the case of T. canis. The best results were obtained using larval A. simplex CE antigen loaded into a CNBr-activated Sepharose 4B coupled to IgG from rabbits immunized with adult A. suum CE antigen. When we compared the different serum dilution and antigenic concentration, we selected the working serum dilution of (1/4)00 and 1 microg/ml of antigenic concentration.     

Determination of kanamycin in serum by solid-phase extraction,pre-capillary derivatization and capillary electrophoresis

An effective method based on solid-phase extraction (SPE) and capillary electrophoresis (CE) for the determination of kanamycin in human serum was developed and validated. Off-line SPE was employed for the isolation of kanamycin from serum on a carboxypropyl-bonded phase (CBA) weak cation-exchange cartridge. A mixture of 0.2 M borate (pH 10.5)-methanol (50:50, v/v) was used as analyte eluting solvent. After pre-capillary derivatization with o-phthalaldehyde/mercaptoacetic acid reagent, the sample was analyzed by CE with a separation buffer of 30 mM borax, pH 10.0, containing 16% (v/v) methanol. A linear response over the concentration range 5-40 microgram/ml was obtained with a detection limit of 2 microgram/ml. Intra-day and inter-day precision were 6.2 and 10.3% RSD, respectively. Recoveries of approximately 90% were found. For the determination of lower levels of kanamycin (<5 microgram/ml), NH(4)OH (25%, w/v)-methanol (30:70, v/v) was used for analyte elution. After evaporation, reconstitution and derivatization, the sample was analyzed by on-line field-amplified sample stacking (FASS) CE. Good linearity in the concentration range 0.4-5 microgram/ml was obtained with a detection limit of 0.1 microgram/ml. Intra-day and inter-day RSD were 3.4 and 11.2%, respectively. Recoveries of approximately 60% were found. The method was successfully applied to the analysis of kanamycin in sera of tuberculosis patients at peak level and trough level concentrations.     

Characterization of a solid-phase extraction device for discontinuous on-line preconcentration in capillary electrophoresis-based peptide mapping

Peptide mapping by capillary electrophoresis (CE) with UV detection is problematic for the characterization of proteins that can only be obtained at low micromolar concentrations. Dilution of peptide fragments during digestion of the protein can further reduce the detection sensitivity in peptide mapping to the point where analysis at sub-micromolar concentrations is not possible. A remedy to this problem is preconcentration (sample enrichment) of the proteolytic digest by solid-phase extraction (SPE). To minimize non-specific adsorptive losses during sample handling, on-line SPE–CE is preferred. However, packed-inlet SPE–CE is not always feasible due to either instrument or sample limitations. We describe here a simple method of preconcentration by discontinuous on-line SPE–CE, specifically applied to peptide mapping in low-pH separation buffer after protein digestion in a solid-phase enzyme microreactor. The SPE–CE system does not require application of a low pressure during electrophoretic separation to overcome reversed electroosmotic flow because the preconcentrator device is disconnected from the separation capillary before the electric field is applied. Up to a 500-fold preconcentration factor can be achieved with this device, which can be reused for many samples. Parameters such as the volume of desorption solution, the adsorption/desorption (chromatographic) process, reproducibility of packing the SPE preconcentrator and effects of sample concentration on the peptide map are investigated.     

Quantification of penicillin G during labor and delivery by capillary electrophoresis

In this study, a capillary electrophoresis (CE) method was developed as a means to measure levels of penicillin G (PCN G) in Group B Streptococcus (GBS) positive pregnant women during labor and delivery. Volunteers for this developmental study were administered five million units of PCN G at the onset of labor. Urine, blood, and amniotic fluid samples were collected during labor and post delivery. Samples were semi-purified by solid-phase extraction (SPE) using Waters tC18 SepPak 3 cc cartridges with a sodium phosphate/methanol step gradient for elution. Capillary electrophoresis or reversed-phase high-performance liquid chromatography (RP-HPLC) with diode-array absorbance detection were used to separate the samples in less than 30 min. Quantification was accomplished by establishing a calibration curve with a linear dynamic range. The tC18 SPE methodology provided substantial sample clean-up with high recovery yields of PCN G ( 90%). It was found that SPE was critical for maintaining the integrity of the separation column when using RP-HPLC, but was not necessary for sample analysis by CE where no stationary phase is present. Quantification results ranged from millimolar concentrations of PCN G in maternal urine to micromolar concentrations in amniotic fluid. Serum and cord blood levels of PCN G were below quantification limits, which is likely due to the prolonged delay in sample collection after antibiotic administration. These results show that CE can serve as a simple and effective means to characterize the pharmacokinetic distribution of PCN G from mother to unborn fetus during labor and delivery. It is anticipated that similar methodologies have the potential to provide a quick, simple, and cost-effective means of monitoring the clinical efficacy of PCN G and other drugs during pregnancy.     

Bioanalysis of tobramycin for therapeutic drug monitoring by solid-phase extraction and capillary zone electrophoresis

A method based on solid-phase extraction (SPE) and capillary zone electrophoresis (CZE) for the analysis of tobramycin in human serum is presented. An off-line SPE employing a carboxypropyl bonded phase (CBA) cartridge was used for the extraction of tobramycin from human serum. Adsorbed tobramycin was eluted from the CBA cartridge using a mixture of NH(3) (25%, w/v)-methanol (30:70, v/v). After evaporation, the analyte was reconstituted and derivatized with o-phthaldialdehyde (OPA)/3-mercaptopropionic acid (MPA). The resulting tobramycin-OPA/MPA derivative was purified, and then identified by mass spectrometry. The tobramycin-OPA/MPA derivative was then analysed by CZE with a background electrolyte (BGE) comprising of 30 mM sodium tetraborate pH 10.0-acetonitrile (ACN) (80:20, v/v) with ultraviolet detection at 230 nm. A linear response was observed in the range of 0.3-30 microg/ml with r(2) = 0.992. The sensitivity of the method was determined by its limit of quantitation (LOQ) and limit of detection (LOD) of 0.3 microg/ml and 0.1 microg/ml, respectively. SPE recovery ranged from 68 to 79% at the trough levels to 98% at the peak levels found in serum. Furosemide has been added as internal standard (IS) to improve precision. For the therapeutic range of tobramycin in serum (2-10 microg/ml) the relative standard deviation (R.S.D.) was less than 11% for the entire SPE/CE process. The method demonstrated excellent selectivity as shown by the lack of interference from a total of 20 drugs investigated. The method was then used in therapeutic drug monitoring of patients receiving the drug.     

Salvia plebeia extract inhibits the inflammatory response in human rheumatoid synovial fibroblasts and a murine model of arthritis

Salvia plebeia R. Br. has been used to treat a variety of inflammatory diseases and as an antioxidant in many countries, including Korea and China. In this study, we investigated the effects of S. plebeia extract (SPE) on inflammatory arthritis and the underlying mechanisms of action. We used a collagen-induced arthritis (CIA) mouse model. TNF-α-stimulated rheumatoid arthritis (RA) synovial fibroblasts were used to elucidate the underlying mechanisms of action. Oral administration of SPE improved the clinical arthritis score, footpad thickness, and histologic changes, as well as serum IgG1 and IgG2a levels. SPE administration inhibited Th1/Th2/Th17 phenotype CD4⁺ T lymphocyte expansion in inguinal lymph node and expression of inflammatory mediators such as cytokines, MMP-1, and MMP-3 in the ankle joint tissue. SPE significantly suppressed the expression of cytokines and MMP-1 by down-regulating NF-κB, Akt, and mitogen-activated protein kinases in RA synovial fibroblasts. Taken together, these results indicate that SPE is therapeutically efficacious against chronic inflammatory arthritis, suggesting that SPE is a candidate for treating RA.     

Chromatography-based on- and in-line pre-concentration methods in capillary electrophoresis

Capillary electrophoresis (CE) poses unique challenges in many different analytical applications, mainly to biological and complex samples and when only small amounts of sample are available, due to its low sample consumption. As a consequence, poor limits of detection are usually observed with this technique, especially with UV photodetectors. Minimal or no sample treatment is desirable in any analytical method to avoid external sources of contamination or errors and to provide a high throughput. On- and in-capillary sample pre-concentration strategies, based on solid-phase extraction (SPE) technology can take advantage of both techniques (SPE and CE), while avoiding sample contamination and tedious manipulations when the sample amount is an issue. Moreover, the combination can provide two-dimensional separations. This review collects the most recent strategies that merge SPE technology built on- and in-capillary pre-concentration for increasing sensitivity and/or selectivity.     

Chromatography-based on- and in-line pre-concentration methods in capillary electrophoresis

Capillary electrophoresis (CE) poses unique challenges in many different analytical applications, mainly to biological and complex samples and when only small amounts of sample are available, due to its low sample consumption. As a consequence, poor limits of detection are usually observed with this technique, especially with UV photodetectors. Minimal or no sample treatment is desirable in any analytical method to avoid external sources of contamination or errors and to provide a high throughput. On- and in-capillary sample pre-concentration strategies, based on solid-phase extraction (SPE) technology can take advantage of both techniques (SPE and CE), while avoiding sample contamination and tedious manipulations when the sample amount is an issue. Moreover, the combination can provide two-dimensional separations. This review collects the most recent strategies that merge SPE technology built on- and in-capillary pre-concentration for increasing sensitivity and/or selectivity.     

Protein analysis by membrane preconcentration–capillary electrophoresis: systematic evaluation of parameters affecting preconcentration and separation

Fast and efficient analysis of proteins in physiological fluids is of great interest to researchers and clinicians alike. Capillary electrophoresis (CE) has proven to be a potentially valuable tool for the separation of proteins in specimens. However, a generally acknowledged drawback of this technique is the limited sample volumes which can be loaded onto the CE capillary which results in a poor concentration limit of detection. In addition, matrix components in samples may also interfere with separation and detection of analytes. Membrane preconcentration–CE (mPC–CE) has proved to be effective in overcoming these problems. In this report, we describe the systematic evaluation of parameters affecting on-line preconcentration/clean-up and separation of protein mixtures by mPC–CE. Method development was carried out with a standard mixture of proteins (lysozyme, myoglobin, carbonic anhydrase, and human serum albumin). First, using MALDI-TOF-MS, membrane materials with cation-exchange (R-SO₃H) or hydrophobic (C₂, C₈, C₁₈, SDB) characteristics were evaluated for their potential to retain proteins in mPC cartridges. Hydrophobic membranes were found most suitable for this application. Next, all mPC–CE analysis of protein samples were performed in polybrene coated capillaries and parameters affecting sample loading, washing and elution, such as the composition and volume of the elution solvent were investigated. Furthermore, to achieve optimal mPC–CE performance for the separation of protein mixtures parameters affecting postelution focusing and electrophoresis, including the composition of the background electrolyte and a trailing stacking buffer were varied. Optimal conditions for mPC–CE analysis of proteins using a C₂ impregnated membrane preconcentration (mPC) cartridge were achieved with a background electrolyte of 5% acetic acid and 2 mM ammonium acetate, 60 nl of 80% acetonitrile in H₂O as an elution solvent, and 60 nl of 0.5% ammonium hydroxide as a trailing stacking buffer. The developed method was used successfully to separate proteins in aqueous humor, which contains numerous proteins in a complex matrix of salts.     

Capillary electrophoresis as a clinical tool

Clinical laboratories must produce accurate results for patients with a minimum turn-around time. Automated commercial capillary electrophoresis instrumentation has been available to the clinical laboratory for the past five years. Our laboratory has utilised capillary electrophoresis (CE) to automate serum protein electrophoresis. We have used the technique of CE to produce clinical results for nearly two years. CE methods are also available for the quantitation of haemoglobin variants, by both isoelectric focusing and free solution techniques. Micellar electrokinetic separations by CE have been developed for some specialised drug assays and for B-group vitamin analysis, while gel-filled capillaries have the capability to separate DNA fragments, such as PCR products. Isoenzyme analysis has shown possibilities by CE, but quantitative results are needed to be clinically useful. Analysis of amino acids for newborn screening programs and as an arterial clotting indicator are being developed. The next five years should see a proliferation of clinical laboratory methods using automated CE.