Ernesto F. Simó-Alfonso

Micellar liquid chromatography: suitable technique for screening analysis.

The screening capability of micellar liquid chromatography (MLC) is discussed using the reported chromatographic data of several sets of compounds (amino acids, beta-blockers, diuretics, phenethylamines, phenols, polynuclear aromatic hydrocarbons, steroids and sulfonamides) and new results (sulfonamides and steroids). The chromatographic data are treated with an interpretive optimisation resolution procedure to obtain the best separation conditions. Usually, the pH and the concentration of surfactant (sodium dodecyl sulfate, SDS, or cetyltrimethylammonium bromide) for the optimal mobile phase were 2.5-3 and < 0.12 M, respectively. The nature and concentration of organic solvent depended on the polarity of the eluted compounds: a low volume fraction of propanol (approximately 1%, v/v) was useful to separate the amino acids, with log P(o/w) < -1 (where P(o/w) is the octanol-water partition coefficient). A greater concentration of this solvent (approximately 5-7%) was needed for compounds in the range -1 < log P(o/w) < 2, as with the studied diuretics and sulfonamides, and a high concentration of propanol (approximately 15%) or a low concentration of butanol (< 10%) had to be used for less polar compounds with 1 < log P(o/w) < 3, such as the beta-blockers. Pentanol (< 6%) was more suitable for the even less polar compounds with log P(o/w) > 3, such as the steroids. For basic drugs such as the phenethylamines (0 < log P(o/w) < 1.7), eluted with a micellar eluent of anionic SDS, propanol was too weak. A study is also shown for mixtures of sulfonamides (log P(o/w) = -1.2 to 1.7) and steroids (log P(o/w) = 3.0-8.1) eluted from conventional C18 columns with SDS mobile phases containing acetonitrile and 1-pentanol, respectively, which are compared with classical acetonitrile-water and methanol-water mixtures. The results complement a previous study on beta-blockers (log P(o/w) = -0.03 to 2.8) and reveal that MLC is a very competitive technique for the screening of compounds against conventional RPLC, due to its peculiar behaviour with regard to the selectivity and elution strength. The concentration of organic solvent needed to obtain sufficiently low retention times (even for highly hydrophobic steroids with log P(o/w) = 7-8) is also appreciably smaller for MLC, which reduces the environmental impact of the mobile phases.

Optimised procedures for the reversed-phase liquid chromatographic analysis of formulations containing tricyclic antidepressants

The chromatographic behaviour (retention, selectivity, peak shape and resolution) of seven tricyclic antidepressants (TCAs), amitryptiline, clomipramine, doxepin, imipramine, maprotiline, nortryptiline and trimipramine, was examined. Conventional unendcapped Cs and C18 columns and an endcapped XTerra MS C18 column recommended for the analysis of basic compounds were used together with acetonitrile-water and micellar sodium dodecylsulfate (SDS)-pentanol mobile phases. The two best combinations were XTerra C18/acetonitrile, which yielded the largest efficiencies and resolution, and C8/SDS-pentanol, which eliminated the peak tails that were still observed with the XTerra C18 column. Both the systems were used to develop simple chromatographic procedures for the control of TCAs in pharmaceutical formulations using UV detection. The selected mobile phase compositions were 35% (v/v) acetonitrile (XTerra C18 column) and 0.075 M SDS-6% (v/v) pentanol (C8 column), both at pH 3. Satisfactory recoveries were achieved in both cases, with intra- and inter-day relative standard deviations (RSDs) always below 0.6 and 2.0%, respectively. The preparation of the samples was simple in both modes, since a previous extraction of the drugs was not needed. The micellar mode has, however, the advantage of using a smaller amount of organic solvent, which is retained in the micellar SDS solution. The C8 column is also less expensive.

Use of triacylglycerol profiles established by high performance liquid chromatography with ultraviolet–visible detection to predict the botanical origin of vegetable oils

A method for the determination of triacylglycerols (TAGs) in vegetable oils from different botanical origins by HPLC with UV-vis detection has been developed. Using a core-shell particle packed column (C18, 2.6 μm), TAG separation was optimized in terms of mobile phase composition and column temperature. Using isocratic elution with acetonitrile/n-pentanol at 10 °C, excellent efficiency with good resolution between most of the TAG peak pairs, within a total analysis time of 15 min, was achieved. Using mass spectrometry detection, a total of 15 peaks, which were common to oils of six different botanical origins (corn, extra virgin olive, grapeseed, hazelnut, peanut and soybean) were identified. These peaks were used to construct linear discriminant analysis (LDA) models for botanical origin prediction. Ratios of the peak areas selected by pairs were used as predictors. All the oils were correctly classified with assignment probabilities higher than 95%.

Determination of cow's milk in non-bovine and mixed cheeses by capillary electrophoresis of whey proteins in acidic isoelectric buffers.

An improved method for the determination of cows milk in non-bovine cheese is reported: electrophoresis of whey proteins in acidic, isoelectric buffers. Two background electrolytes (BGEs) have been tested: (i) 50 mM iminodiacetic acid (pH=isoelectric point=2.30 at 25 degrees C), 0.5% hydroxyethylcellulose, 0.1% Tween 20 and 6 M urea (apparent pH 3.1), E=300 V/cm, for the separation of alpha-lactalbumins (alpha-LAs); (ii) a BGE with the same composition, but supplemented with 10% Tween 20, E=450 V/cm, for the fractionation of beta-lactoglobulins (beta-LGs). Surfactants have a discriminating effect on the retention behaviour of the bovine alpha-LA and beta-LG proteins, owing to the different strength of the protein-surfactant association complexes, and are needed for separating these two proteins from small peaks in the electropherograms generated by degradation of casein during cheese ripening. Novel equations are given for deriving the ratio of the area (or height) of bovine alpha-LA, or beta-LG, to the area (or height) of ovine or caprine alpha-LA or beta-LG (such ratios being typically used to determine the percentage of cows milk in dairy products), since previous equations had marked drawbacks, such as non-linearity of the plots with increasing slopes at high cows milk percentages, and too broad confidence limits at high cows milk contents, where the peak area (or height) ratio tends asymptotically to infinite. With the novel procedures reported, contents of cows milk as low as 1% can be quantified in goats and ewes cheeses. The present protocols give lower detection limits, are cheaper and more rapid than any other methodology reported in the literature, and can be easily applied to the routine quality control of binary and ternary cheeses.

Micellar-organic versus aqueous-organic mobile phases for the screening of β-blockers

A comparative study of the performance of reversed-phase liquid chromatography with micellar-organic (MLC) and aqueous-organic (RPLC) mobile phases is reported for the separation of 16 -blockers (acebutolol, alprenolol, atenolol, bisoprolol, carteolol, celiprolol, esmolol, labetalol, metoprolol, nadolol, oxprenolol, pindolol, practolol, propranolol, sotalol, and timolol). MLC with hybrid mobile phases of sodium dodecyl sulfate (SDS) and propanol is revealed as a very competitive technique for the screening of these drugs. Using a conventional Spherisorb C18 column, the theoretical plates ( N) and asymmetry factors (B/A) for the optimal mobile phase compositions were in the ranges N = 2200–4400 and B/A = 1.0–1.3 for SDS–propanol against N = 290–960 and B/A = 2.2–3.6, and N = 1700–5100 and B/A = 1.5–2.2 for acetonitrile–water in the absence and presence of 0.1% (v/v) triethylamine. With a base-deactivated XTerra MS C18 column and acetonitrile–water, peak shape was improved (N = 1100–8800 and B/A = 1.1–2.5), but the elution pattern (large differences in retention between polar and low polar compounds) was similar to that achieved with the Spherisorb column. A set of 14 -blockers showing a wide range of polarities (log Po/w =− 0.03 for atenolol to 2.60 for propranolol) could be resolved isocratically with a mobile phase of 0.10 M SDS–15% (v/v) propanol with an analysis time of 32 min. The resolution was better than with aqueous-organic RPLC using the conventional and base-deactivated C18 columns, where only 10–12 compounds could be separated in a single chromatogram. Using specific mobile phases of SDS–propanol, all possible binary mixtures of the 16 -blockers could be resolved fully (peak purity of the peak pairs, R> 0.999), except practolol–sotalol (R = 0.992), nadolol–pindolol (R = 0.959), and esmolol–metoprolol (R = 0.956), which showed a small overlapping. The resolution of the individual peaks with the aqueous-organic mixtures was more problematic.

Variety identification in maize lines via capillary electrophoresis of zeins in isoelectric acidic buffers

Zeins (the prolamins or seed storage proteins in maize) have been used to characterize and identify different genotypes. Zeins were fractionated by capillary zone electrophoresis in acidic, amphoteric buffers, which represent a medium of moderate conductivity and are thus compatible with higher voltage gradients. The running buffer consisted of 40 mM isoelectric aspartic acid, in presence of 6 M urea and 0.5% hydroxyethyl cellulose (apparent pH: 3.8; pI in the absence of urea: 2.77). Thirty‐one different zein peaks were mapped out of a total of 21 different maize genotypes. Each of them typically exhibited seven to twelve peaks, with some genotypes showing up to 20 zein bands. Due to slightly changing elution times, caused by a lack of reproducibility of the electroendoosmotic flow in uncoated silica surfaces, correct peak assignment and alignment among different runs was obtained by multivariate statistical analysis. The present method compares well, both in resolution and total number of peaks, with current protocols adopted for screening of maize inbreds, which consist of isoelectric focusing in agarose gels.

Sodium dodecyl sulfate capillary electrophoresis of proteins in entangled solutions of poly(vinyl alcohol)

Abstract A novel polymer network is described for efficient sieving of sodium dodecyl sulfate (SDS)-protein complexes: poly(vinyl alcohol) (PVA; average Mr 133 000). The entanglement threshold of these solutions was found to be at 3% (w/v) PVA. Solutions from 4 to 6% PVA offer excellent resolution in the 14 400-94 000 protein molecular mass interval. Ferguson plot analysis showed that the separation is indeed based on mass discrimination, as it should in SDS electrophoresis, with extrapolated (at 10% polymer) limit values of mobility for all particles in the range (2.34-2.87) · 10−8 m2 V−1s−1. The advantages of PVA are full transparency in the UV region down to 200 nm and extremely low viscosities (e.g., a 5% PVA solution has a viscosity 25 times greater than that of buffer at 30°C). A unique wall effect was found, by which, on decreasing the inner diameter of the capillary from 75 to 25 μm, the apparent entanglement threshold was shifted to extremely dilute PVA solutions, since in 25-μm capillaries efficient sieving was obtained below 1% PVA, i.e., at concentrations well below the entangled regime. It is hypothesized that residual, free silanols present (even in a coated capillary) act as nucleation sites for H-bond formation and aggregation of free PVA molecules.

Prediction of the genetic variety of Spanish extra virgin olive oils using fatty acid and phenolic compound profiles established by direct infusion mass spectrometry

The genetic varieties of Spanish extra virgin olive oils (Arbequina, Hojiblanca and Picual) were predicted by direct infusion of the samples in the electrospray ionization source of a mass spectrometer, followed by linear discriminant analysis of the spectral data. The samples were 1:50 diluted (v/v) with an 85:15 propanol/methanol (v/v) mixture containing 40mM KOH and infused. The abundances of the [M-H](-) peaks of the free fatty acids (7 peaks) and 28 phenolic compounds (20 peaks) were measured. Ratios of pairs of peak abundances were used as predictors in the construction of the linear discriminant analysis models. An excellent resolution between the three genetic varieties was achieved.

High-performance micellar liquid chromatography determination of sulphonamides in pharmaceuticals after azodye precolumn derivatization

A chromatographic procedure with precolumn derivatization to form the N-(1-naphthyl)ethylenediamine dihydrochloride azodyes is proposed for the analysis of several sulphonamides (sodium sulphacetamide, sulphadiazine, sulphaguanidine, sulphamerazine, sulphamethizole, sulphamethoxazole, sulphanilamide and sulphathiazole) in pharmaceutical preparations (tablets, pills, capsules, suspensions and drops). The separation is performed with a 0.05 M sodium dodecyl sulphate/2.4% pentanol eluent at pH 7. The precolumn derivatization improved the resolution in the chromatograms and increased the selectivity in the determination of mixtures of sulphonamides and in preparations where other drugs were present. The derivatization reaction was readily performed in a micellar medium of SDS at pH 1, leading to a rapid and simple procedure. The recoveries were in the 97-104% range with relative standard deviations usually below 3%.

Determination of tocopherols and tocotrienols in vegetable oils by nanoliquid chromatography with ultraviolet-visible detection using a silica monolithic column.

A method for the determination of tocopherols and tocotrienols in vegetable oils by nanoliquid chromatography with UV-vis detection has been developed. The separation of tocopherols was optimized in terms of mobile phase composition on the basis of the best compromise between efficiency, resolution, and analysis time. The optimal conditions were achieved using a C18 silica monolithic column (150 mm x 0.1 mm) with an isocratic elution of acetonitrile/methanol/water (acidified with 0.2% acetic acid) at a flow rate of 0.5 microL min(-1), giving a total analysis time below 18 min. The method has been applied to the quantification of tocopherols and tocotrienols present in several vegetable oils with different botanical origins.