Determination of α‐hydroxy acids and their enantiomers in fruit juices by ligand exchange CE with a dual central metal ion system

The content of α‐hydroxy acids and their enantiomers can be used to distinguish authentic and adulterated fruit juices. Here, we investigated the use of ligand exchange CE with two kinds of central metal ion in a BGE for the simultaneous determination of enantiomers of dl ‐malic, dl ‐tartaric and dl ‐isocitric acids, and citric acid. Ligand exchange CE with 100 mM d ‐quinic acid as a chiral selector ligand and 10 mM Cu(II) ion as a central metal ion could enantioseparate dl ‐tartaric acid but not dl ‐malic acid or dl ‐isocitric acid. Addition of 1.8 mM Sc(III) ion to the BGE with 10 mM Cu(II) ion to create a dual central metal ion system permitted the simultaneous determination of these α‐hydroxy acid enantiomers and citric acid. The proposed ligand exchange CE was thus well suited for detecting adulteration of fruit juices.     

Effect of analyte lipophilicity on the resolution of α‐ and β‐amino acids on liquid chromatographic ligand exchange chiral stationary phases

Separation of the two enantiomers of racemic α‐ and β‐amino acids on two ligand exchange chiral stationary phases (CSPs) prepared previously by covalently bonding sodium N‐((S)‐1‐hydroxymethy‐3‐methylbutyl)‐N‐undecylaminoacetate or sodium N‐((R)‐2‐hydroxy‐1‐phenylethyl)‐N‐undecylaminoacetate on silica gel was studied with variation of the organic modifier (methanol) concentration in the aqueous mobile phase. In particular, the variation of retention factors with changing organic modifier concentration in the aqueous mobile phase was found to be strongly dependent on both the analyte lipophilicity and the stationary phase lipophilicity. In general, the retention factors of relatively lipophilic analytes on relatively lipophilic CSPs tend to increase with increasing organic modifier concentration in the aqueous mobile phases while those of less lipophilic or hydrophilic analytes tend to increase. However, only highly lipophilic analytes show decreasing retention factors with increasing organic modifier concentration in the aqueous mobile phase on less lipophilic CSPs. The contrasting retention behaviors on the two CSPs were rationalized by the balance of the two competing interactions, viz. hydrophilic interaction of analytes with polar aqueous mobile phase and the lipophilic interaction of analytes with the stationary phase.     

Chiral capillary electrophoresis with cationic pyrrolidinium‐β‐cyclodextrin derivatives as chiral selectors

New single‐isomer, cationic β‐cyclodextrins, including mono‐6‐deoxy‐6‐pyrrolidine‐β‐cyclodextrin chloride (pyCDCl), mono‐6‐deoxy‐6‐(N‐methyl‐pyrrolidine)‐β‐cyclodextrin chloride (N‐CH₃‐pyCDCl), mono‐6‐deoxy‐6‐(N‐(2‐hydroxyethyl)‐pyrrolidine)‐β‐cyclodextrin chloride (N‐EtOH‐pyCDCl), mono‐6‐deoxy‐6‐(2‐hydroxymethyl‐pyrrolidine)‐β‐cyclodextrin chloride (2‐MeOH‐pyCDCl) were synthesized and used as chiral selectors in capillary electrophoresis for the enantioseparation of carboxylic and hydroxycarboxylic acids and dansyl amino acids. The unsubstituted pyCDCl exhibited the greatest resolving ability. Most analytes were resolved over a wide range of pH from 6.0 to 9.0 with this chiral selector. In general, increasing pH led to a decrease in resolution. The effective mobilities of all the analytes were found to decrease with increasing CD concentration. The optimal concentration for most carboxylic acids and dansyl amino acid was in the range 5–7.5 mM and >15 mM for hydroxycarboxylic acids. ¹H NMR experiments provided direct evidence of inclusion in the CD cavity.     

Enantioseparation of hydroxyeicosatetraenoic acids by hydroxypropyl‐γ‐cyclodextrin‐modified micellar electrokinetic chromatography

Complete resolution of hydroxyeicosatetraenoic acid (HETE) enantiomers was achieved using hydroxypropyl‐γ‐cyclodextrin (HP‐γ‐CD)‐modified MEKC. The optimum running conditions were determined to be utilizing a 30 mM phosphate–15 mM borate buffer (pH 9.0) containing 30 mM HP‐γ‐CD and 75 mM SDS as the BGE, application of +30 kV as the effective voltage, and carrying out the experiment at 15°C. The eluents were detected at 235 nm. The method was used successfully for the simultaneous separations of (S)‐ and (R)‐enantiomers of regioisomeric 8‐, 11‐, 12‐, and 15‐HETEs. Subsequently, the optimized method was applied to evaluate the stereochemistry of 8‐ and 12‐HETEs from the marine red algae, Gracilaria vermiculophylla and Gracilaria arcuata, respectively. The 8‐HETE was found to be a mixture of 98% (R)‐enantiomer and 2% (S)‐enantiomer, while the 12‐HETE was a mixture of 98% (S)‐enantiomer and 2% (R)‐enantiomer. The present study demonstrates that the HP‐γ‐CD‐modified MEKC method is simple and sensitive and provides unambiguous information on the configuration of natural and synthetic HETEs.     

Enantioseparation of underivatised amino acids by ligand exchange capillary electrophoresis in a counter-electroosmotic mode

Enantiomer separations of underivatised amino acids were carried out by using ligand exchange capillary electrophoresis (LECE). Chiral discrimination is based on the formation of ternary complexes between copper(II), a chiral selector (L-proline or trans-4-hydroxy-L-proline) and an amino acid. All amino acids containing aromatic moieties or not were detected at 214 nm because of their interactions with copper(II). In order to reduce copper(II) adsorption onto capillary walls, we used hexadimethrine bromide to reverse the electroosmotic flow. Using this original strategy, the studied enantiomers migrated in the opposite direction of the anodic electroosmosis. After optimising the analytical conditions taking into account the chiral resolution and the detection sensitivity, we performed very satisfactory enantioseparations not only of aromatic amino acids (tryptophan, tyrosine, phenylalanine and histidine) but also of aliphatic amino acids (threonine, serine, isoleucine and valine). These enantioseparations were performed in a short analysis time at 35 °C. In order to rationalise the obtained results, we evaluated the complexation constants corresponding to the formed ternary complexes by capillary electrophoresis and we used molecular mechanics modelling.     

Enantioseparation of dansyl amino acids by ligand‐exchange capillary electrophoresis with zinc(II)‐L‐phenylalaninamide complex

A novel method of chiral ligand‐exchange CE was developed with L ‐amino acylamides as a chiral ligand and zinc(II) as a central ion. It has been demonstrated that these chiral complexes, such as Zn(II)‐L ‐alaninamide, Zn(II)‐L ‐prolinamide, and Zn(II)‐L ‐phenylalaninamide, are suitable for use as chiral selectors for the enantioseparation of either individual pair of or mixed dansyl amino acids. The optimal separation running buffer consisted of 5 mM ammonium acetate, 100 mM boric acid, 4 mM ZnSO₄·7 H₂O, and 8 mM L ‐amino acylamides at pH 8.2. The experiments showed that apart from the effect of the concentration of the complexes on the resolution and the migration time, the buffer pH also had a sharp influence on resolution. The employed chiral ligands exhibited different enantioselectivities and enantiomer migration orders. D ‐Amino acids migrate faster than L ‐amino acids when Zn(II)‐L ‐alaninamide and Zn(II)‐L ‐phenylalaninamide are used as chiral selectors, but it was observed that the migration order is reversed when Zn(II)‐L ‐prolinamide is used as the chiral selector. Furthermore, the amount of dansylated amino acids is found to be highly dependent on the labeling temperature.     

Enantioseparation of citalopram analogues with sulfated β‐cyclodextrin by capillary electrophoresis

Capillary electrophoresis methods were developed for the enantiomeric separation of 27 citalopram analogues. Sulfated β‐cyclodextrin was the most broadly selective and useful chiral selector. The separations of most of the citalopram analogue compounds reported in this work have not been reported previously. Excellent enantiomeric separations were obtained for 26 out of 27 compounds, and most of the separations were achieved within 10 min. The effects of chemical parameters such as chiral selector types, buffer types, chiral selector and buffer concentrations, buffer pH and organic modifiers on the separation were investigated. The influence of analyte structure on separation also was examined and discussed.     

Effect of the residual silanol group protection on the liquid chromatographic resolution of α‐amino acids and proton pump inhibitors on a ligand exchange chiral stationary phase

A new ligand exchange chiral stationary phase (new CSP) containing residual silanol group‐protecting n‐octyl groups on the silica surface was prepared by treating a ligand exchange CSP (original CSP) based on sodium N‐[(R)‐2‐hydroxy‐1‐phenylethyl]‐N‐undecylaminoacetate bonded to silica gel with excess n‐octyltriethoxysilane. The new and original CSPs containing an identical amount of chiral selector were applied to the resolution of α‐amino acids and proton pump inhibitors (PPIs) including omeprazole, pantoprazole, lansoprazole, and rabeprazole. The separation factors (α) and resolutions (R_S) were greater on the new CSP than on the original CSP except for the resolution of asparagine. The trends of the retention factors (k₁) for the resolution of α‐amino acids on the new and original CSPs with the variation of the organic modifier content in aqueous mobile phase were opposite to those for the resolution of PPIs. Removal of the nonenantioselective interactions between the residual silanol groups and the analytes and the improved lipophilicity of the new CSP were proposed to be responsible for the improved chiral recognition ability of the new CSP and the different retention behaviors of the enantiomers between the new and original CSPs.     

Enantioseparation of tartaric acid by ligand‐exchange capillary electrophoresis using contactless conductivity detection

A method for the determination of tartaric acid enantiomers using CE with contactless conductivity detection has been developed. Cu(II) as a central metal ion together with l ‐hydroxyproline were used as a chiral selector, the BGE was composed of 7 mM CuCl_2, 14 mM trans‐4‐hydroxy‐l ‐proline, and 100 mM ε‐aminocaproic acid; the pH was adjusted to 5 by hydrochloric acid. Separation with a resolution of 1.9 was achieved in 9 min in a polyacrylamide‐coated capillary to suppress the EOF. Various counterions of the BGE were studied, and migration order reversal was achieved when switching from ε‐aminocaproic acid to l ‐histidine. With detection limits of about 20 μM, the method was applied to the analysis of wine and grape samples; only l ‐tartaric acid was found.     

Enantioseparation of nicotine alkaloids in cigarettes by CE using sulfated β‐CD as a chiral selector and a capillary coated with amino groups

Nicotine (NC) and its related compounds (cotinine (CN), nornicotine (NN), anatabine (AT) and anabasine (AB)) were simultaneously enantioseparated by CE using a capillary with amino groups and sulfated β‐CD as a chiral selector. The optimum running conditions were found to be 30 mM acetate buffer (pH 5.0) containing 8% sulfated β‐CD with an applied voltage of +15 kV at 30°C using direct detection at 260 nm. Using a capillary coated with amino groups, the EOF migrates toward the positive pole. However, when sulfated β‐CD was added to the BGE, it was found that the EOF migrated toward the negative pole due to ionic adsorption of sulfated β‐CD to amino groups on the capillary inner wall. All the cationic analytes migrated as anions, suggesting that they formed stable anionic complexes with sulfated β‐CD. With this system and a simple pretreatment with mini‐cartridges, NC alkaloids in five cigarette samples were enantioseparated. As a result, each of the compounds except for CN was detected. In the case of NC, only (S)‐NC was detected (more than 99.9%), but in the case of NN, AT and AB, the ratios of (S)‐isomer to total isomers were in the ranges 58–70, 81–85 and 59–65%, respectively. On the other hand, only NC was detected in cigarette smoke and the ratio of (S)‐ and (R)‐NCs was 96:4. The amounts of NC alkaloids in cigarettes suggest that the production of (R)‐NC resulted from racemization due to the high temperature/burning of the cigarette.     

Enantioseparation of 3‐phenyllactic acid by chiral ligand exchange countercurrent chromatography

3‐Phenyllactic acid is an antimicrobial compound with broad‐spectrum activity against various bacteria and fungus. The observed difference in pharmacological activity between optical isomeric 3‐phenyllactic acid necessitates a method for enantioseparation. Chiral ligand exchange countercurrent chromatography was investigated for the enantioseparation of 3‐phenyllactic acid with a synthesized chiral ligand. A two‐phase solvent system was composed of n‐butanol/hexane/water (0.4:0.6:1, v/v/v) to which N‐n‐dodecyl‐l ‐hydroxyproline was added to the organic phase as chiral ligand and cupric acetate was added in the aqueous phase as a transitional metal ion. The influence factors were optimized by enantioselective liquid–liquid extraction. Baseline enantioseparation of racemic 3‐phenyllactic acid by analytical high‐speed countercurrent chromatography was achieved. The optical purities of enantiomeric 3‐phenyllactic acid reached 99.0%, as determined by chiral high‐performance liquid chromatography.     

Combined use of ionic liquid and β‐CD for enantioseparation of 12 pharmaceuticals using CE

In this study, the enantioseparation of zopiclone, repaglinide, chlorphenamine maleate, brompheniramine maleate, dioxopromethazine hydrochloride, promethazine hydrochloride, liarozole, carvedilol, homatropine hydrobromide, homatropine methylbromide, venlafaxine, and sibutramine hydrochloride has been investigated using β‐CD in combination with a chiral ionic liquid (IL), 1‐ethyl‐3‐methylimidazolium‐L‐lactate. The influence of the type of IL and its concentration, BGE pH, and chain length of the IL cations on the resolution are discussed. Finally, the proposed method was successfully applied for the chiral impurity determination of eszopiclone in pharmaceutical tablets after validation with respect to accuracy and precision, linearity range, selectivity, repeatability, LOD and LOQ. It is assessed that the chiral impurity determination in the commercial tables is fewer than 0.1%.     

New single isomer negatively charged β‐cyclodextrin derivatives as chiral selectors in capillary electrophoresis

To improve resolution power of chiral selector and enantiomeric peak efficiency in CE, single isomer negatively charged β‐CD derivatives, mono(6‐deoxy‐6‐sulfoethylthio)‐β‐CD (SET‐β‐CD) bearing one negative charge and mono[6‐deoxy‐6‐(6‐sulfooxy‐5,5‐bis‐sulfooxymethyl)hexylthio]‐β‐CD (SMHT‐β‐CD) carrying three negative charges, were synthesized. The structure of these two β‐CD derivatives was confirmed by ¹H NMR and MS. SET‐β‐CD and SMHT‐β‐CD successfully resolved the enantiomers of several basic model compounds. SMHT‐β‐CD provided for a significantly greater enantioseparation than SET‐β‐CD at lower concentrations. This appears to be due to the higher binding affinity of SMHT‐β‐CD to the model compounds and the wider separation window resulting from an increased countercurrent mobility of the selector. Overall, the new chiral selectors provided enantioseparations with high peak efficiency while avoiding peak distortion due to polydispersive and electrodispersive effects. The information obtained from an apparent binding constant study suggested that the enantioseparation of the model compounds followed the predictions of charged resolving agent migration model and that the observed degree of enantioseparation difference were due to the magnitude of differences in both enantiomer‐chiral selector binding affinities (ΔK) and the mobilities of the complexed enantiomers (Δμ_c).     

HPLC enantioseparation of α,α‐diphenyl‐2‐pyrrolidinemethanol and methylphenidate using a chiral fluorescent derivatization reagent and its application to the analysis of rat plasma

Enantioseparation of α,α‐diphenyl‐2‐pyrrolidinemethanol (D2PM) and methylphenidate (MPH; Ritalin^®) using (R)‐(?)‐4‐(N,N‐dimethylaminosulfonyl)‐7‐(3‐isothiocyanatopyrrolidin‐1‐yl)‐2,1,3‐benzoxadiazole as the chiral derivatization reagent has been achieved for the first time, and a simple, reliable detection method using HPLC with fluorescence detection has been developed. D2PM and MPH have been derivatized with (R)‐(?)‐4‐(N,N‐dimethylaminosulfonyl)‐7‐(3‐isothiocyanatopyrrolidin‐1‐yl)‐2,1,3‐benzoxadiazole at 55°C for 15 min. The derivatives of D2PM and MPH have been separated, completely and rapidly, using a reversed‐phase system within 16 min (resolution factor (R_s)=1.60 and 2.53, respectively). The detection limits of (R)‐ and (S)‐D2PM were found to be 6.8 and 13 ng/mL, respectively, and those of D ‐ and L ‐threo‐MPH were 61 and 66 ng/mL, respectively (S/N=3). The proposed method was successfully applied to the analysis of rat plasma, where the rats were separately dosed with D2PM and MPH (Ritalin).     

Enantioseparation of DL-isocitric acid by a chiral ligand exchange CE with Ni(II)-D-quinic acid system

The ratio of citric acid to D ‐isocitric acid can be used to distinguish authentic and adulterated fruit juices. To separate DL ‐isocitric acid enantiomers, we used ligand exchange CE. D ‐Quinic acid was used as a chiral selector ligand and Mn(II), Fe(III), Co(II), Ni(II), Cu(II), and Zn(II) ions were used as the central ions of the chiral selector in the BGE. DL ‐Isocitric acid was found to be enantioseparated with the above metal ions except for Mn(II) ion. The optimum running conditions for the analysis of D ‐ and L ‐isocitric acids along with citric acid, an isomer of isocitric acid, were found to be a BGE (pH 5.0) containing 30% ACN, 20 mM acetic acid, 20 mM NiSO₄, and 80 mM D ‐quinic acid. Under these conditions, DL ‐isocitric and citric acids in fruit juices were analyzed successfully.     

Combined use of hydroxypropyl‐β‐cyclodextrin and ionic liquids for the simultaneous enantioseparation of four azole antifungals by CE and a study of the synergistic effect

A CE method employing a dual system of hydroxypropyl‐β‐cyclodextrin (HP‐β‐CD) and ionic liquids (ILs) has been developed for the simultaneous enantioseparation of four azole antifungals for the first time. In this study, three different types of ILs were employed as modifiers and among them dodecyl trimethyl ammonium chloride was found to be the most effective. The effects of the concentration, cations, and anions of ILs on the enantioseparation were investigated. With the developed dual system, all the enantiomers were well separated in resolutions of 3.8, 3.5, 2.8, and 2.5 for miconazole, econazole, ketoconazole, and itraconazole, respectively. The interactions between dodecyl trimethyl ammonium chloride and HP‐β‐CD were also studied using a neutral polyacrylamide coated capillary and ¹H NMR spectroscopy to further explore the synergistic effect involved. It was found that ILs improved the enantioseparation not only by changing the EOF, but also by interactions with HP‐β‐CD that could change its ability of forming inclusion complex with the enantiomers.     

Enantioseparation of chiral aromatic acids by multiple dual mode counter‐current chromatography using hydroxypropyl‐β‐cyclodextrin as chiral selector

This work concentrates on extending the utilization of multiple dual mode (MDM) counter‐current chromatography in chiral separations. Two aromatic acids, 2‐(6‐methoxy‐2‐naphthyl)propionic acid (NAP) and 2‐phenylpropionic acid (2‐PPA), were enantioseparated by MDM counter‐current chromatography using hydroxypropyl‐β‐cyclodextrin (HP‐β‐CD) as chiral selector. The two‐phase solvent systems consisting of n‐hexane/ethyl acetate 0.1 mol/L phosphate buffer pH 2.67 containing 0.1 mol/L HP‐β‐CD (7.5:2.5:10 for NAP and 7:3:10 for 2‐PPA, v/v/v) were used. Conventional MDM and modified MDM were compared according to peak resolution under current separation mechanism. The influence of elution time after the first‐phase inversion and number of cycles for MDM were investigated. Peak resolution of NAP and 2‐PPA increased from 0.62 to 1.05 and 0.72 to 0.84, respectively, using optimized MDM conditions. Being an alternative elution method for counter‐current chromatography, MDM elution greatly improved peak resolution in chiral separations.     

Development of a fast and efficient CE enzyme assay for the characterization and inhibition studies of α‐glucosidase inhibitors

The inhibition of the α‐glucosidase enzyme plays an important role in the treatment of diabetes mellitus. We have established a highly sensitive, fast, and convenient CE method for the characterization of the enzyme and inhibition studies of α‐glucosidase inhibitors. The separation conditions were optimized; the pH value and concentration of the borate‐based separation buffer were optimized in order to achieve baseline separation of p‐nitrophenyl‐α‐d ‐glucopyranoside and p‐nitrophenolate. The optimized method using 25 mM tetraborate buffer, pH 9.5, was evaluated in terms of repeatability, LOD, LOQ, and linearity. The LOD and LOQ were 0.32 and 1.32 μM for p‐nitrophenyl‐α‐d ‐glucopyranoside and 0.83 and 3.42 μM for p‐nitrophenolate, respectively. The value of the Michaelis–Menten constant (K_m) determined for the enzyme is 0.61 mM, which is in good agreement with the reported data. The RSDs (n = 6) for the migration time was 0.67 and 1.83% for substrate and product, respectively. In the newly established CE method, the separation of the reaction analytes was completed in <4 min. The developed CE method is rapid and simple for measuring enzyme kinetics and for assaying inhibitors.     

Differentiation of α‐COOH from β‐COOH in aspartic acids by N‐phosphorylation

The biomimic reactions of N‐phosphoryl amino acids, which involved intramolecular penta‐coordinate phosphoric‐carboxylic mixed anhydrides, are very important in the study of many biochemical processes. The reactivity difference between the α‐COOH group and β‐COOH in phosphoryl amino acids was studied by experiments and theoretical calculations. It was found that the α‐COOH group, and not β‐COOH, was involved in the ester exchange on phosphorus in experiment. From MNDO calculations, the energy of the penta‐coordinate phosphoric intermediate containing five‐member ring from α‐COOH was 35 kJ/mol lower than that of the six‐member one from β‐COOH. This result was in agreement with that predicted by HF/6‐31G** and B3LYP/6‐31G** calculations. Theoretical three‐dimensional potential energy surface for the intermediates predicted that the transition states 4 and 5 involving α‐COOH or β‐COOH group had energy barriers of ΔE=175.8 kJ?mol^?1 and 210.4 kJ?mol^?1, respectively. So the α‐COOH could be differentiated from β‐COOH intramolecularly in aspartic acids by N‐phosphorylation. © 2001 John Wiley & Sons, Inc. Int J Quant Chem 83: 41–51, 2001     

Enantioselective Syntheses of Substituted γ‐Butyrolactones

The previously described chiral 2‐acyloxathianes 5 (Scheme I) are used in two different enantioselective syntheses of γ‐butyrolactones. In one synthesis, Grignard addition, cleavage and reduction to carbinols RR'C(OH)CH₂OH is followed by tosylation, malonate homologation, lactonization, and removal of the carbomethoxy group to give optically active γ‐lactones. A modification of this synthesis (Scheme I) leads to optically active α‐methylene‐γ‐lactones. In the second synthesis, reaction of a bromomagnesium enolate with ketones 5 leads to β‐hydroxyesters, which, by appropriate sequences of reduction and cleavage (Scheme II) are converted to optically active α‐ or β‐hydroxy‐γ‐lactones.