Jose Carlos Diez-Masa

Capillary isoelectric focusing of erythropoietin glycoforms and its comparison with flat-bed isoelectric focusing and capillary zone electrophoresis

The influence of several operation conditions on separation of recombinant human erythropoietin glycoforms by capillary isoelectric focusing (cIEF) is explored. From this study it is deduced that in order to separate several glycoforms of erythropoietin, urea has to be added to sample, which should not be completely depleted of the excipients used in its formulation. On-line desalting does not provide separation enhancement for samples with high content of salt. Better resolution is obtained using a mixture of a broad and a narrow pH-range carrier ampholytes than with either one used separately. Under the experimental conditions, focusing voltages of 25 kV improve separation compared to lower and higher electric fields. Focusing times shorter than the time necessary for electric current to reach a minimum provide similar separations than longer focusing times at which a minimum value of the current has already been achieved. The optimized method allows the separation and quantitation in 12 min of at least seven bands containing glycoforms of recombinant erythropoietin with apparent isoelectric points in the range 3.78-4.69. Compared to flat-bed isoelectric focusing, cIEF provides better separation of bands of glycoforms in a shorter time, and allows quantitative determination. Capillary zone electrophoresis (CZE) gives rise to resolution of erythropoietin glycoforms similar to that obtained by cIEF. Although CZE requires a longer analysis time, its reproducibility in terms of peak area of glycoforms is better than in cIEF.

High-efficiency capillary electrophoretic separation of basic proteins using coated capillaries and cationic buffer additives: Evaluation of protein—capillary wall interactions

Abstract The joint use of basic cationic additives (morpholine and several tetraazamacrocycles) in the buffer and chemically bonded cross-linked polyacrylamide-coated capillaries was evaluated as a method for decreasing the adsorption of basic proteins on the fused-silica capillary wall. The superiority of the tetraazamacrocycle Cyclen (1,4,7,10-tetraazacyclododecane) over morpholine and other tetraazamacrocycles is demonstrated. Using 20 mM phosphate-60 mM Cyclen buffer (pH 5.5) and cross-linked polyacrylamide-coated capillaries, separation efficiencies in the range of 106 plates/m were obtained for basic proteins. A simplified model that allows the quantification of the interactions between proteins and the capillary wall was developed. The model was assessed using the different buffers and capillaries evaluated in the first part. As the model predicts, a straight line for the plot of the inverse of the migration time versus the electric field strength with an intercept different from zero was observed. The value of the intercept correlates with the separation efficiency observed for the basic proteins studied and, therefore, with the interaction strength between proteins and the capillary wall.

Preparation of linear polyacrylamide-coated capillaries - Study of the polymerization process and its effect on capillary electrophoresis performance

The effect of different parameters controlling the characteristics of linear polyacrylamide coatings deposited on the inner wall of fused-silica capillaries and their influence on capillary electrophoresis (CE) performance of these coated columns is investigated. To carry out this study, a reproducible procedure to obtain capillaries with similar extent of modification of the surface silanols with 7-oct-1-enyltrimethoxisilane was first approached. Next the polymer attachment to the silica wall, via covalent linkage to the silyl reagent grafted onto the silica, was investigated. In this way, by using columns with a similar silylation extent, differences in CE performance observed among capillaries coated under diverse conditions could be assigned to the characteristics of the polyacrylamide layer. It is demonstrated that the characteristics and reproducibility of these polymeric coatings depend on the adequate control of both the temperature of polymerization and the degassing of the polymerizing dissolutions used. More interestingly, it is also demonstrated that the quantities of monomer (acrylamide), initiator (ammonium persulfate) and activator (N,N,N′,N′-tetramethylethylenediamine), and the ratio among them used in the preparation of the coating polymer have a large influence on the performance of CE columns. The optimum conditions for preparing the polyacrylamide coatings are discussed. The applicability of these linear polyacrylamide-coated capillaries to the separation of basic and acidic proteins in free zone CE is demonstrated. Besides, the use of these coated columns in capillary gel electrophoresis for the separation of DNA fragments is shown.

Correlation between the logarithm of capacity factors for aromatic compounds in micellar electrokinetic chromatography and their octanol-water partition coefficients

The correlation between the logarithm of the capacity factors (log k′) and the logarithm of the octanol-water partition coefficients (log Pow) for a group of 20 organic compounds (ten benzene derivatives and ten polycyclic aromatic hydrocarbons) was studied in micellar electrokinetic chromatography. Sodium dodecyl sulphate micelles in different aqueous and hydro-organic (n-propanol and n-butanol) buffers were used as electrolyte solutions. A linear relationship between log k′ and log Pow was found. This correlation improved when the percentage of alcohol in the buffer system was increased and when the polarity of the alcohol was decreased.

Separation of basic proteins by capillary electrophoresis using cross-linked polyacrylamide-coated capillaries and cationic buffer additives

Abstract A method for the preparation of fused-silica capillaries with a cross-linked polyacrylamide coating bonded to the internal wall is described. When these capillaries are used with acidic separation buffers containing 0.25 M morpholine as cationic additive to mask the residual effect of the negative charges on the silica surface, efficiencies between 3 · 105 and 5 · 105 plates/m can be obtained for basic proteins (pI 7–11). The reproducibility of migration times for these capillaries is better than 1% run-to-run and 2% capillary-to-capillary. The capillaries can be used for more than 90 h over the pH range 2–10 without any substantial alteration of the migration time or efficiency for the basic proteins assayed. It is also shown that the capillaries can be stored for more than 90 days in air with no appreciable variation in the migration time or the efficiency for several basic proteins.

Treatments of fused-silica capillaries and their influence on the electrophoretic characteristics of these columns before and after coating

Abstract Making coated columns in capillary electrophoresis (CE) is a laborious task involving many preparation steps such as etching, leaching, dehydration, silylation and coating of the inner wall of fused-silica capillaries. In this work we demonstrate, by testing more than 250 columns, that it is possible to follow up the influence of the different steps on both the electrophoretic behavior of CE columns and the reproducibility of their preparation. This study was done by carrying out triplicate measurements of electroosmotic flow values from columns at four different pH values after each step. The effectiveness of the coatings was also investigated by injecting a test-group of basic proteins. It is demonstrated that etching the columns with sodium hydroxide followed by a leaching treatment with hydrochloric acid provides higher reproducibility than by either leaching or etching alone. Dehydration of the capillary tubing affects the yield of the subsequent silylation reaction while the best results seem to be obtained by dehydrating the columns overnight at 160°C. The silylation degree achieved is demonstrated to also depend on the time of reaction and the concentration of silyl reagent. Moreover, a conclusive demonstration about the effect of both silylation and polymerization reactions on the final coating performance is given.

Separation of chiral polychlorinated biphenyls by micellar electrokinetic chromatography using β- and γ-cyclodextrin mixtures in the separation buffer

Chiral polychlorinated biphenyls (PCBs) 45, 84, 88, 91, 95, 132, 136, 139, 149, 171, 183 and 196 were separated each in its two enantiomers by cyclodextrin-modified micellar electrokinetic chromatography (CD-MEKC). Mixtures of β- and γ-cyclodextrins were used as chiral modifiers in a 2-(N-cyclohexylamino)ethanesulphonic acid (CHES) buffer containing urea and sodium dodecyl sulphate (SDS) micelles. Separations of multicomponent mixtures of PCBs into their enantiomers were also performed. A mixture of PCBs 45, 88, 91, 95, 136, 139, 149 and 196 was separated into all sixteen enantiomers in an analysis time of approx. 35 min.

Perfusion liquid chromatography of whey proteins

A perfusion reversed-phase (RP) HPLC method was developed for the rapid separation of the main bovine whey proteins: alpha-lactalbumin (alpha-LA), serum albumin (BSA) and the genetic variants of beta-lactoglobulin (A and B) (beta-LG A and beta-LG B). For the method development, the influence of factors favouring structural changes of proteins (temperature and organic acid concentration in the mobile phase), gradient and other chromatographic conditions and the mass of protein injected was examined. The optimized method allowed the separation of proteins in about 1.5 min (cycle time 3.5 min) with resolution around 1.0 for the beta-lactoglobulins. The method was applied to the determination of proteins in a whey from raw bovine milk. The precision of the determinations was < or = 3.75 mg per 100 ml (S.D.). With respect to the accuracy, errors < or = 7.0% in the determination of alpha-LA, beta-LG A and beta-LG B were obtained, compared with an RP-HPLC reference method. However, higher errors in the quantification of BSA were found owing to the lack of purity of the peak assigned. In addition, the proposed method has proved to be very useful in the detection of homologous whey proteins from different species (cow, sheep and goat) in milk mixtures.

Separation of basic proteins by capillary zone electrophoresis with coatings of a copolymer of vinylpyrrolidone and vinylimidazole

Capillary zone electrophoretic (CZE) separation of basic proteins has been achieved with capillary columns modified with copolymers of vinylpyrrolidone (VP) and vinylimidazole (VI). The copolymerization reaction is performed inside the capillary column and involves chemical bonding of the polymer to silica. The electroosmotic flow (EOF) is greatly decreased by this surface modification. The presence of positive charges on the coating surface, due to the cationic property of vinylimidazole at pH below 7, reduces the adsorption of basic proteins onto the silanol groups of the capillary surface. Acidic proteins are irreversibly adsorbed, but rapid separation and good performance reproducibility are obtained with basic proteins. In the case of capillaries modified with VP, the acidic and basic proteins are eluted within 10 min. In this work, we studied the effects of pH and buffer concentration on the magnitude of the EOF, as well as the effect of copolymer composition on the separation efficiency.

Determination of solute-micelle association constants for a group of benzene derivatives and polycyclic aromatic hydrocarbons with sodium dodecyl sulphate by micellar electrokinetic chromatography

Abstract Micellar electrokinetic chromatography was applied to determine solute-micelle association constants for a group of benzene derivatives and polycyclic aromatic hydrocarbons using sodium dodecyl sulphate as surfactant. Among the different buffers studied, only those of pH⩾9 [2-(N-cyclohexylamino)ethanesulphonic acid and ammonium acetate] gave an electroosmotic flow high enough to allow the elution of all compounds studied. Determination of association constants was achieved under different experimental conditions and the resulting errors were evaluated. The effect of the nature and concentration of the buffer and the alcohol (n-propanol and n-butanol) on the values obtained by this technique for solute-micelle association constants was studied. It was observed that these factors do not affect the association constants under the experimental conditions used. The values of the solute-micelle association constants and the errors in their determination were also compared with those obtained previously by micellar liquid chromatography.