Evaluation of chiral normal-phase liquid chromatography as a secondary tier in pharmaceutical chiral screening strategy.

Determination of stereoisomers is an integral part of pharmaceutical analysis. Chiral liquid chromatography (LC) method development is typically initiated through screening of chiral stationary phases (CSPs) and mobile phases (MPs) since chiral separation is difficult to predict. We have previously reported a screening strategy using chiral reversed-phase (RP) LC as two primary tiers due to its versatility for enantio­recognition and compatibility with diverse sample matrices. Here we focus on developing a normal-phase (NP) LC screening strategy as a secondary tier for chiral method screening. A database was constructed from 60 NPLC screens performed on up to 18 CSPs and 3 MPs using gradient elution. This was used to investigate the effectiveness of NPLC compared to RPLC screening, as well as the impact of MP composition and the selectivity of different CSPs in NPLC screening. A success hit rate of 90% was observed in NPLC compared to 84% in RPLC screening for Bristol Myers Squibb compounds. Importantly, NPLC screening generated successful hit(s) in 81% of the cases that failed in RPLC, demonstrating the value of NPLC as a complementary screening tier. After optimizing the CSP/MP selection, we proposed a NPLC screening workflow with several user-options according to method requirements and instrument capacity. Among these, the most comprehensive NPLC screening consisted of ten CSPs (AD, AS, AY, AZ, OD, OJ, IC, IE, IG, O1) with three MPs. When combined with RPLC, an overall success rate of 97% was achieved for the diverse set of pharmaceutical compounds.

Evaluation of a polysaccharide-based chiral reversed-phase liquid chromatography screen strategy in pharmaceutical analysis.

Chirality control plays a critical role in developing stereoisomeric drugs. Due to the complexity and lack of predictability in chiral separations, column screening remains the gold standard to initiate chiral method development for active pharmaceutical ingredients (APIs) and synthetic intermediates. Chiral reversed-phase (RP) liquid chromatography (LC) has gained favor over other modes due to its versatility and compatibility in analyzing a wide range of chiral compounds in various matrices. Herein, we established a tier-based chiral RPLC screen strategy by constructing and analyzing a database of 101 chiral screens with a total of 3,401 entries (unique LC runs) for proprietary APIs or intermediates at Bristol Myers Squibb. Up to 17 polysaccharide-based chiral stationary phases (CSPs) and four mobile phases (MPs) have been screened with gradient elution. A selection of ten CSPs with two MPs was found sufficient to achieve successful separation for 82% of the total screens. Two RPLC screen tiers (Tier 1: AZ, OD, ID, and IG) and (Tier 2: AY, OJ, OZ, IA, IC, and IH) were proposed along with two MPs (acidic and neutral) to target ~70% hit rate for Tier 1, and ~80% for the combined set. We also implemented a user-friendly workflow to enable walk-up chiral RPLC screening with automated reports and system suitability tests.

High-molecular weight impurity screening by size-exclusion chromatography on a reversed-phase column.

Monitoring polymerization events leading to the discovery of new high-molecular weight (MW) impurities is challenging during chemical syntheses of active pharmaceutical ingredients. Employing reversed-phase chromatography (RPC) stationary phases (SPs) in size-exclusion chromatography (SEC) mode could be a potential solution given their high efficiency, sensitivity, and extensive solvent compatibility. However, there is a lack of generalized means for trace polymeric impurities across a wide range of physicochemical properties. Herein, we developed a SEC-based approach with a C18 SP for screening such high-MW impurities. Seven polymer standards presenting a variety of functional groups, consisting of hydrophobic, heterocyclic, ionic, and neutral hydrophilic moieties, were utilized as model impurities to establish the screening conditions. Nine mobile phases (tetrahydrofuran-based, buffered methanol, and buffered acetonitrile) were proposed to cover all model polymers and a majority of potential high-MW impurities in small molecule chemical syntheses. The established screening system demonstrated a linearity of 0.05-1.0 % w/w (R2>0.99) for the selected model impurities with proper elution conditions. Two real high-MW impurities, BMT-041910 (polymeric degradation) and poly(phenyl thiirane) (by-product polymerization), were identified from the proposed high-MW impurity screening. The successful conditions yielded a quantitative limit better than 0.1 % w/w in both cases. We believe the developed screening platform is applicable to the analysis of a wide variety of unknown high-MW impurities of low abundance potentially generated during drug substance development.

Tunable normal phase enantioselectivity of amino acid esters via mobile phase composition.

The ability to tune chiral selectivity through mobile phase modifiers is a powerful tool in chiral separations. Beyond improving efficiency and/or resolution, some mobile phase systems can even invert elution order, a highly desirable result for trace analyses or preparative scale isolations. Previous work has demonstrated that acidic modifiers, such as ethanesulfonic acid (ESA), can greatly impact separations of enantiomers. However, prior studies were primarily performed on coated chiral stationary phases (CSPs), which limited the selection of the bulk mobile phase component. In this work, the effect of ESA modifier was studied for the enantioseparation of six pairs of amino acid esters on a CHIRALPAK® IA column, an immobilized amylose-based CSP, with different combinations of standard solvents (hexane and ethanol) as well as "non-standard" solvents, such as methyl t-butyl ether, ethyl acetate, tetrahydrofuran, acetone, or 1,4-dioxane. ESA generally improved selectivity, and multiple instances of elution order reversal were observed. A Van Deemter plot study reveals that ESA exerts its effect by pulling the enantiomer deeper into the chiral cavity of the chiral polymer to increase the interactions between the analytes and the stationary phase, which is the main reason for the increased enantioselectivity.

The effect of additives on the zinc carbenoid-mediated cyclopropanation of a dihydropyrrole.

The synthesis of a key intermediate in the preparation of oral antidiabetic drug Saxagliptin is discussed with an emphasis on the challenges posed by the cyclopropanation of a dihydropyrrole. Kinetic studies on the cyclopropanation show an induction period that is consistent with a change in the structure of the carbenoid reagent during the course of the reaction. This mechanistic transition is associated with an underlying Schlenk equilibrium that favors the formation of monoalkylzinc carbenoid IZnCH2I relative to dialkylzinc carbenoid Zn(CH2I)2, which is responsible for the initiation of the cyclopropanation. The factors influencing reaction rates and diastereoselectivities are discussed with the aid of DFT computational studies. The rate accelerations observed in the presence of Brønsted acid-type additives correlate with the minimization of the undesired induction period and offer insights for the development of a robust process.

NMR and computational studies of chiral discrimination by amylose tris(3,5-dimethylphenylcarbamate).

Proton NMR and simulations were combined to study the origin of chiral selectivity by a polysaccharide used in a commercial chromatographic stationary phase: amylose tris(3,5-dimethylphenylcarbamate). This material has unusually high enantioselectivity for p-O-tert-butyltyrosine allyl ester, which is activated by the presence of an acid. Proton NMR spectra agreed with the HPLC in showing that the l-enantiomer interacts much more strongly with the polysaccharide and that acidity switches on the selectivity. 2D NOESY spectra revealed which protons of each enantiomer and the polysaccharide were in proximity, and these spectra revealed folding of the l-enantiomer. Computations generated energy-minimized structures for the polysaccharide-enantiomer complexes, independently predicting folding of the l-enantiomer. Molecular dynamics simulations 2 ns in duration, repeated for three different energy-minimized structures, generated pair distribution functions that are in excellent agreement with the 2D NOESY spectra. The modeling studies revealed why acidity switches on chiral selectivity and minimally affects the chromatographic retention time of the unfavored d-enantiomer. The results comprise the first case of a chiral separation by a commercial polysaccharide stationary phase being explained using a combination of 2D NOESY and simulations, providing excellent agreement between experiment and computation and lending detailed molecular insight into enantioselectivity for this system.

The effect of acidic and basic additives on the enantioseparation of basic drugs using polysaccharide-based chiral stationary phases.

The enantioseparation of nine commercially available basic drugs was achieved on polysaccharide-based chiral stationary phases with the acidic additive ethanesulfonic acid and the basic additive butylamine. Seven different commercially available CSPs were used for the study (AD, AS, OD, OJ, OG, OB, and OC). Mobile phase additives have been proven to be essential in obtaining satisfactory enantio-resolution in terms of both efficiency and selectivity. Significantly improved selectivities were obtained for the basic probe drugs with the acidic additive, ethanesulfonic acid, rather than the basic additive, butylamine. This is best seen with Chiralpak AS CSP. No enantioseparation for the nine drugs was observed when butylamine was used as an additive; however, satisfactory enantioseparation for the nine drugs was achieved using ethanesulfonic acid. Higher column efficiencies were observed with the acidic additive, especially when isopropanol was used as a modifier. Higher sensitivity was also achieved with ethanesulfonic acid because of the significantly lower background at the UV detection wavelength. The acidic additive was demonstrated to be superior to the basic additive for the enantioseparation of basic drugs using seven different polysaccharide-based CSPs. These results are counterintuitive to the common "rule of thumb" in enantioseparation that states acidic additives work best for acidic analytes and basic additives work best for basic analytes. The beneficial effects of acidic additive in enantioseparations observed in this study could significantly improve the applicability of polysaccharide-based CSPs for the enantioseparation of basic analytes.

Chiral separation of amines by high-performance liquid chromatography using polysaccharide stationary phases and acidic additives.

A dramatic and beneficial effect of ethanesulfonic acid (ESA) on the chiral HPLC separation of basic compounds was found. Using a single chiral column and a starting mobile phase, more than half of a diverse set of amines was baseline separated. Changing alcohol content and alcohol type increased the success rate. Methanesulfonic acid (MSA) proved even more successful. The mechanism of this unexpected finding appears to be a combination of ion-pair salt formation in the mobile phase and increased binding with the chiral stationary phase (CSP) arising from a localized decrease in pH.

Origin of enhanced chiral selectivity by acidic additives for a polysaccharide-based stationary phase.

The effects of ethanesulfonic acid (ESA) and n-butylamine as additives were studied for a wide variety of chiral compounds using the polysaccharide chiral stationary phase (CSP), Chiralpak AD. The mobile phase consisted of hexane-ethanol (90:10, v/v). The additives typically had small effects, with one exception: the acidic additive had an enormous effect on the chiral selectivity of amino acid esters. The improved chiral selectivity was largely due to the longer retention of the later eluting enantiomer. Retention behavior of amines indicated that the higher selectivity for amino acid esters owes to increased hydrogen-bonding donation by the amine group of the analyte. Computation establishes the feasibility of a planar complex between the analyte and the cliral stationary phase, involving a pair of complementary hydrogen-bonding groups on each species, enabled by protonation of the analyte. Retention behaviors for a range of structures point to steric hindrance as the third interaction to comprise the requisite three interactions in chiral recognition.

Effect of amine mobile phase additives on chiral subcritical fluid chromatography using polysaccharide stationary phases.

Increased retention and selectivity in the subcritical fluid chromatography (SFC) of various amine compounds on polysaccharide chiral stationary phases (CSP) was observed upon incorporation of cyclic amines into the modifier. The retention increases are most pronounced with 2-propanol and are almost absent when methanol is used as modifier. This suggests that the effect may arise from a restriction to the modifier access to the binding site required to effect elution. The effect of the amine additives in SFC does not remain after their removal from the mobile phase. Findings were applied to the development of a 5 min separation of amphetamine and methamphetamine enantiomers.

Memory effect of mobile phase additives in chiral separations on a Chiralpak AD column.

Using chiral probes shown to be sensitive to the presence of mobile phase additives, a memory effect for these additives by an amylosic column was demonstrated. Exposure to these additives gave prolonged chromatographic performance changes even after their removal from the mobile phase. This finding is consistent with strong binding of the additives to the stationary phase. A procedure to remove bound additives was developed.

Enantioseparation of amino acids on a polysaccharide-based chiral stationary phase.

Sulfonic acids have been shown to be more effective than the commonly used trifluoroacetic acid (TFA) in the chiral resolution of underivatized aromatic amino acids on an amylosic column. Sulfonic acid additives give a more UV transparent mobile phase, possibly allowing the detection of non-aromatic analytes. Work presented demonstrates that through the combination of sulfonic acid mobile phase additives, amine mobile phase additives and solvent modifier variations, the enantiomers of 20 of 25 probe amino acids are fully resolved, four are partially resolved with only one failing to be separated on a common amylosic column.