Capillary Electrophoresis
– Techniques and Methods
Preconcentration
Injected sample amounts in CE are inherently small. Moreover, when using optical detectors, the optical pathway is limited. Consequently, the concentration sensitivity of CE methods can be poor. Improvement of concentration detection limits by sample preconcentration, either online or offline, remains a topic of research activity.
Electrophoretic Preconcentration
Electrophoretic sample preconcentration in CE often relies on an abrupt and temporary reduction of the migration velocity of analytes. This is clear in commonly applied methods like field-amplified sample stacking (FASS), large-volume sample stacking/injection (LVSS/LVSI), field-enhanced sample injection (FESI), sweeping, and electrokinetic supercharging (EKS). The fundamentals of electrokinetic processes occurring during FASS were studied experimentally and with computer simulation by Sestak and Thormann. The authors investigated the effect of injected plug lengths, buffer concentration, sample composition, and linear velocity for the analysis of cationic compounds.
Another study described the use of a free liquid membrane (FLM) to further enhance stacking efficiency of a EKS-CE-UV method. The FLM presents a water-immiscible organic solvent interface facilitating the electrically induced transfer of charged analytes, such as paraquat and diquat. The sensitivity gain was almost 2000-fold. Similar improvements in sensitivity were achieved by Cheng et al., who combined LVSI, anion selective exhaustive injection, and sweeping for the online preconcentration of tetrahydrocannabinol and metabolites. The resulting CE method allowed direct detection of target analytes in urine.
Multiple isotachophoresis (M-ITP) injections were explored to enhance sensitivity. In M-ITP, the ITP sample preconcentration procedure is repeated several times allowing injection of up to 300 times the normal volume. With 6 M-ITP cycles, quantification of the Aβ 1-40 amyloid peptide down to 50 nM was achieved using UV detection.
Chromatographic Preconcentration
Solid-phase extraction (SPE) remains a popular chromatographic preconcentration technique for CE. Zhao et al. demonstrated the effectiveness of offline C5 reversed-phase liquid chromatography (RPLC) prior to top-down capillary zone electrophoresis coupled to electrospray ionization mass spectrometry (CZE-ESI-MS) analysis of a yeast proteome. In total, 580 proteoforms and 180 protein groups were identified from 23 proteome fractions analyzed. Another study used C8 SPE cartridges for reduction of sample complexity and preconcentration. Melatonin and indole compounds in plant extracts could be detected down to low ppb levels. Rodriguez et al. performed SPE based on synthesized Fe3O4–fullerene–activated carbon magnetic adsorbents for analysis of azo dyes in wastewater in the low mg/L range.
In online SPE-CE a small trapping column is introduced just before or in the initial part of the separation capillary. Tascon et al. proposed an online SPE-CE-MS method for sensitive alkaloid analysis. A micro-C18-column ensured sample cleanup and simultaneous preconcentration, providing limits of detection (LODs) in the 2–77 pg/mL range for algal extracts. Espina-Benitez et al. used a short segment of silica-based monolith with a locally functionalized acrylamide derivative of phenylboronic acid to isolate and preconcentrate diols from 2 μL of complex matrixes (Figure 1). Column elution with a small plug of acidic solution allowed FASS of the analytes prior to their CE separation, ensuring LODs in the ng/mL range. This inline coupling was subsequently successfully used for the fully automated analysis of catecholamines (neurotransmitters) in urine samples.
Figure 1.
Miscellaneous
An online sample preconcentration method was developed exclusively for catecholamines that were fluorogenically derivatized with naphthalene-2,3-dicarboxaldehyde. It takes advantage of diol-borate complexation, through which one negative charge is added to the analytes. The sample was electrokinetically introduced via flow-gated injection. The analytes were selectively focused to a narrow zone by reversible complexation, leading to 100-fold preconcentration for catecholamines in artificial cerebrospinal fluid.
Shimura and Nagai combined immunoaffinity chromatography (IAC) with isoelectric focusing (IEF) in a single capillary. IAC was ensured by immobilizing an anti-E-tag antibody at the inlet of the capillary. The remainder of the capillary was coated with neutral polydimethylacrylamide to ensure efficient IEF separations. Fluorescently labeled recombinant Fab with an E-tag spiked at 16 pM to 10 nM in 50% serum was separated and detected with high precision.
An online high-throughput microdialysis-capillary electrophoresis (MD-CE) assay was designed to investigate branched-chain amino acids as possible biomarkers. Analytes were sampled using microdialysis, fluorescently labeled in an online reaction, separated using CE and detected using laser-induced fluorescence (LIF) in a sheath flow cuvette. CE separations were performed in less than 30 s, and the temporal resolution of the online MD-CE assay was within 60 s. In a next study, the MD-CE assay was used to monitor in vivo dynamics, achieving a temporal resolution of 22 s for small bioamines.
Coatings
When conventional bare fused-silica capillaries are used in CE, resolution and peak widths and shapes may be compromised by adverse interactions of the analytes with the inner capillary surface. Furthermore, adsorption of sample matrix components, e.g., proteins, may cause uncontrollable changes of the electroosmotic flow (EOF) and poor migration-time reproducibility. In order to avoid unwanted adsorptions, coating of the capillary wall is a common strategy which remains the subject of research.
Poulsen et al. posed new capillary coating procedures using polyethylene glycol (PEG). These include in-capillary surface-initiated atom transfer radical polymerization ensuring covalent binding to the capillary wall and an electrostatic adsorption process. Coating procedures were followed by monitoring adsorption of 2-propylisochinolinium bromide and Sunset Yellow as a positive and negative marker, respectively. Multiple injections of high concentrations of proteins covering a pI range of 3.4–8.4 could be performed without depletion of capillary performance, indicating coating stability of at least 100 days.
A capillary coating procedure allowing regulation of the magnitude and direction of the EOF was proposed by Fu and co-workers. This was achieved by coadsorption of polydopamine and polyethylenimine of different molecular weights in variable mass ratios. The polymer chains were stabilized by complexation with Fe3+. The obtained coatings were further characterized by field emission scanning electron microscopy and attenuated total reflection Fourier-transform infrared spectroscopy analysis.
Moreno-Gordaliza et al. used pretreated surface layer protein A from Lactobacillus acidophilus bacteria as a capillary coating, which was characterized by contact angle, fluorescence, and atomic force microscopy (AFM) analysis. The new coatings were used for analysis of lipoproteins from human serum with capillary ITP (cITP). The coating could be used for over 100 injections without loss of separation performance with coefficients of variation of 3% for protein migration times over a period of 7 days.
AFM with an adhesive tip was used by Stock et al. to assess topographic and charge-induced features on capillary coatings. The charge distribution of different successive multiple ionic polymer layer (SMIL) coatings was assessed with nanometer resolution employing avidin as a single molecule sensor. The acquired surface properties of a four-layer SMIL with poly(acrylamide-co-2-acrylamido-2-methyl-1-propansulfonate) as the terminal layer were related to the observed EOF and CE performance for model proteins and peptides on the same capillary.
Optimization of capillary coating procedures and their tuning toward specific applications commonly is time-consuming and a trial-and-error process. Montferrat et al. proposed a method to describe the EOF behavior of a polymer coating as a function of pH, allowing predictive analysis of electroosmosis under different polymeric coating conditions. By means of a theoretical argument and numerical simulations involving the linearized Poisson–Boltzmann equation and the Lattice Boltzmann scheme, the experimental curve for the EOF of an acrylamide/methacrylate coating is assessed.
Separation Media
Pseudostationary Phases
Pseudostationary phases (PSP) enable separations not achievable with regular CZE. Micellar electrokinetic chromatography (MEKC) was used for the resolution of insulin and closely related peptides. The use of neutral surfactants such as Thesit and Tween20 increased selectivity by reducing adverse interactions with the capillary wall, avoiding the use of a capillary coating while using an aqueous background electrolyte (BGE). For separation of different insulins, negatively charged surfactants were required, from which perfluorooctanoic acid was found to provide the best resolution.
The new chiral ionic ligand 1-ethyl-3-methyl imidazole L-tartrate ([EMIM][L-Tar]) was introduced for the separation of tryptophan, tyrosine, and phenylalanine enantiomers by chiral-ligand-exchange CE. A comparison with L-tartaric and [EMIM]l-proline indicated the potential of [EMIM][L-Tar] for the enantioseparation of amino acids (AAs). Liu et al. synthesized and used the sugar-based surfactant poly(sodium N-alkenyl-α-d-glucopyranoside) with various size and headgroup functionalities for chiral separation of ephedrine alkaloids and β-blockers by MEKC analysis. Polymers as compared to monomers showed to strongly enhance separation, while sulfate groups gave less resolution enhancement than phosphate head groups.