The interaction between human serum albumin (HSA) and apremilast (APR), a novel antipsoriatic drug, was characterized by multimodal analytical techniques including high-performance liquid chromatography (HPLC), fluorescence spectroscopy and molecular docking for the first time. Using an HSA chiral stationary phase, the APR enantiomers were well separated, indicating enantioselective binding between the protein and the analytes. The influence of chromatographic parameters—type and concentration of the organic modifier, buffer type, pH, ionic strength of the mobile phase, flow rate and column temperature—on the chromatographic responses (retention factor and selectivity) was analyzed in detail. The results revealed that the eutomer S-APR bound to the protein to a greater extent than the antipode. The classical van ’t Hoff method was applied for thermodynamic analysis, which indicated that the enantioseparation was enthalpy-controlled. The stability constants of the protein–enantiomer complexes, determined by fluorescence spectroscopy, were in accordance with the elution order observed in HPLC (K_R-APR-HSA = 6.45 × 10³ M⁻¹, K_S-APR-HSA = 1.04 × 10⁴ M⁻¹), showing that, indeed, the later-eluting S-APR displayed a stronger binding with HSA. Molecular docking was applied to study and analyze the interactions between HSA and the APR enantiomers at the atomic level. It was revealed that the most favored APR binding occurred at the border between domains I and II of HSA, and secondary interactions were responsible for the different binding strengths of the enantiomers.

The enantioselective binding of three proton pump inhibitors (PPIs)—omeprazole, rabeprazole, and lansoprazole—to two key plasma proteins, α1-acid glycoprotein (AGP) and human serum albumin (HSA), was characterized. The interactions between PPI enantiomers and proteins were investigated using a multifaceted analytical approach, including high-performance liquid chromatography (HPLC), fluorescence and UV spectroscopy, as well as in silico molecular docking. HPLC analysis demonstrated that all three PPIs exhibited enantioseparation on an AGP-based chiral stationary phase, suggesting stereoselective binding to AGP, while only lansoprazole showed enantioselective binding on the HSA-based column. Quantitatively, the S-enantiomers of omeprazole and rabeprazole showed higher binding affinity to AGP, while the R-enantiomer of lansoprazole displayed greater affinity for AGP, with a reversal in the elution order observed between the two protein-based columns. Protein binding percentages, calculated via HPLC, were greater than 88% for each enantiomer across both transport proteins, with all enantiomers displaying higher affinity for AGP compared to HSA. Thermodynamic analysis indicated that on the HSA, the more common, enthalpy-controlled enantioseparation was found, while in contrast, on the AGP, entropy-controlled enantioseparation was observed. The study also identified limitations in using fluorescence titration due to the high native fluorescence of the compounds, whereas UV titration was effective for both proteins. The determined logK values were in the range of 4.47–4.83 for AGP and 4.02–4.66 for HSA. Molecular docking supported the experimental findings by revealing the atomic interactions driving the binding process, with the predicted enantiomer elution orders aligning with experimental data. The comprehensive use of these analytical methods provides detailed insights into the enantioselective binding properties of PPIs, contributing to the understanding of their pharmacokinetic differences and aiding in the development of more effective therapeutic strategies.

A reversed-phase high-performance liquid chromatographic (HPLC) method was developed for the simultaneous determination of the potential impurities of dexketoprofen, including the distomer R-ketoprofen. After screening the separation capability of four polysaccharide columns (Lux Amylose-1, Lux Amylose-2, Lux Cellulose-1 and Lux Cellulose-2) in polar organic and in reversed-phase modes, appropriate enantioseparation was observed only on the Lux Amylose-2 column in an acidified acetonitrile/water mixture. A detailed investigation of the mobile phase composition and temperature for enantio- and chemoselectivity showed many unexpected observations. It was observed that both the resolution and the enantiomer elution order can be fine-tuned by varying the temperature and mobile phase composition. Moreover, hysteresis of the retention times and enantioselectivity was also observed in reversed-phase mode using methanol/water mixtures on amylose-type columns. This could indicate that the three-dimensional structure of the amylose column can change by transitioning from a polar organic to a reversed-phase mode, which affects the enantioseparation process. Temperature-dependent enantiomer elution order and rare enthalpic/entropic controlled enantioseparation in the operative temperature range were also observed in reversed-phase mode. To find the best methodological conditions for the determination of dexketoprofen impurities, a full factorial optimization design was performed. Using the optimized parameters (Lux Amylose-2 column with water/acetonitrile/acetic acid 50/50/0.1 (v/v/v) at a 1 mL/min flow rate at 20 °C), baseline separations were achieved between all compounds within 15 min. Our newly developed HPLC method was validated according to the current guidelines, and its application was tested on commercially available pharmaceutical formulations. According to the authors’ knowledge, this is the first study to report hysteretic behavior on polysaccharide columns in reversed-phase mode.

Enantioseparation in HPLC is a considerable challenge in analytical chemistry, frequently requiring numerous trials with varying experimental conditions to achieve baseline separations. To address this issue, we propose a solution that utilizes consensus modelling based on partial least squares (PLS) regression method together with neural network (NN) algorithms and a graph neural network (GNN) method to predict the retention times of compounds on Lux Cellulose-1 chiral stationary phase under various polar organic mode mobile phases. A homogeneous dataset was collected for the developed machine learning methods, consisting of 535 unique molecules and 1,414 retention time measurements under four polar organic mode conditions (acidic and basic methanol, acidic and basic acetonitrile). The PLS + NN consensus model showed outstanding results in condition-specific predictions, achieving R² values over 0.70 and RMSE values below 0.40 in most cases. Conversely, the GNN model excelled in combined predictions under all conditions, achieving a R² of 0.58 and RMSE_CV of 0.49 during cross-validation, as well as a R² of 0.85 and RMSE_Test of 0.25 on the test set. Our research presents a novel approach for predicting chiral separations, offering an easy-to-use, open-access web tool to the scientific community. The robust GNN model was used to create a web server (https://chiralscreen.com) that enables the prediction of retention times, separation capabilities, and elution orders for various compounds. Furthermore, the software helps determine optimal initial chromatographic conditions for separations.

Many drugs are chiral, and their enantiomers often display distinct pharmacological activities. Accurate determination of enantiomerization kinetics during chromatographic separations is therefore of critical importance. Batman peaks (plateau regions arising between partially resolved enantiomer peaks) carry valuable information about on-column interconversion dynamics. We developed an automated R framework that integrates both the unified equation and a stochastic model to extract kinetic and thermodynamic constants from chromatographic data. The workflow applies robust optimization algorithms to enhance flexibility, accelerate fitting, and improve accuracy relative to manual approaches. Quetiapine chromatograms were analyzed over a wide range of flow rates (0.1mL/minto3.0mL/min) and temperatures (10°C to 40°C). Both the unified equation and the stochastic model produced consistent rate constants, which were in good agreement with values obtained from optical rotation experiments. Notably, the stochastic model remained applicable even under extreme conditions where peaks coalesced. Linearized Eyring-Polányi analysis afforded the determination of thermodynamic parameters. The presented workflow provides a robust, flexible, and customizable platform for determining enantiomerization kinetics from Batman peaks. By combining the unified equation with stochastic modeling, it delivers reliable results across diverse chromatographic conditions. This framework is readily adaptable for routine analysis of dynamic enantiomerization in pharmaceutical and analytical chemistry applications.

The separation of three proton pump inhibitors (omeprazole, lansoprazole, and rabeprazole) as exemplified molecules containing chiral sulfoxide groups was investigated in polar organic liquid chromatographic mode on seven different polysaccharide stationary phases (Chiralcel OD and OJ; Chiralpak AD, AS, and IA; Lux Cellulose-2 and -4). Different alcohols, such as methanol, ethanol, 1-propanol, 2-propanol, and their combinations, were used as eluents. After method optimization, semi-preparative enantioseparation was successfully applied for the three proton pump inhibitors to collect the individual enantiomers. A detailed investigation was conducted into elution order reversal, thermodynamic parameters, the effect of eluent mixtures, and the hysteresis of retention time and selectivity. Using Chiralpak AS, containing the amylose tris[(S)-α-methylbenzylcarbamate] chiral selector, the separation of the investigated enantiomers was achieved in all four neat eluents, with methanol providing the best results. In many cases, a reversal of the enantiomer elution order was observed. In addition to chiral-selector-dependent reversal, eluent-dependent reversal was also observed. Notably, even replacing methanol with ethanol altered the enantiomer elution order. Both enthalpy- and entropy-controlled enantioseparation were also observed in several cases; however, temperature-dependent elution order reversal was not. The hysteresis of retention and selectivity was further investigated on amylose-type columns in methanol–2-propanol and methanol–ethanol eluent mixtures. The phenomenon was observed on all amylose columns regardless of the eluent mixtures employed. Hystereticity ratios were calculated and used to compare the hysteresis behaviors of different systems. Multivariate statistical analysis revealed that Chiralpak AS exhibited the most distinct enantioselective behavior among the tested columns, likely due to the absence of a direct connection between the carbamate moiety and the aromatic substituent. The present study aided in understanding the mechanisms leading to enantiomer recognition, which is crucial for developing new chiral stationary phases and chiral HPLC method development in general.