Scope and application of a commercial chiral anion-exchange LC stationary phase with 'hydro-organic' mobile phases

Experimental

An Agilent 1290 Infinity (Agilent Technologies, (76337) Waldbronn, Germany) was used for the study. Itconsisted of a G4220A 1290 binary pump, G4226A 1290 sampler, G1316C Infinity TCC (Thermostatted ColumnCompartment) and a G4212A Infinity DAD detector. Data was collected using Agilent ChemStation software.For the evaluation of enantioseparation, enantioselectivity and structure activity relationship of chiral acidicdrugs, Lindner and Commercial quinidine anion exchange and quinine anion exchange columns were used(QD-AX and QN-AX columns). Mobile phases were prepared using HPLC – grade methanol (Fisher chemicals,UK). Mobile phase additives ammonium formate (NH₄HCOOH), formic acid (HCOOH) 98/100% and ammoniumacetate (NH₄OAc) acetic acid (HOAc), were of analytical/ HPLC grade (Fisher chemicals, UK). All the drugcompunds used in this research were provided by University of Sunderland chemical store or were purchasedfrom Sigma Aldrich (UK)

Conclusions

Low pH (when the analytes would not be fully ionised) proved to give the best retention and enantioresolution Selectivity was remarkably independent of mobile phase variables (other than at loss of selectivity and retention at pH > 5 – 6) In general, improved resolution could only be obtained at the expense of analysis time. Enantioselectivity was obtained for a range of drug and related acids including NSAIDs, substituted mandelc acids and acids of interest for pesticides. Good chemical selectivity was obtained but enantioselectivity was not sufficient to warrant exploring achiral – chiral separations in a single column (2) Retention on all three columns was very similar, thus suggesting a robust manufacturing process. QD-AX and QN-AX did NOT always act as “psuedo-enantiomers”; in some cases showing the same retention order with and without the same selectivity. While one set of conditions were validated, many would be suitable for monitoring the appearance of mandelic acids during enzymatic hydrolysis.

Introduction

Much of the work on chiral stationary phases based on immobilized quinine and quinidine moieties has involved the analysis of derivatised amino acids andthe use of a polar organic, or more recently super-critical fluid, mobile phase (1). However, because of an interest in the resolution of acidic chiral drugswith ‘reversed-phase’ mobile phases and the study of the enzymatic hydrolysis of mandelic esters in aqueous buffers with added organic solvent,carbamate-based chiral anion-exchange LC stationary phases based on the pseudoenantiomers cinchona alkaloids, quinidine (QD) (8R, 9S) and quinine (QN)(8S, 9R), were fully evaluated using hydro – organic mobile phases that would be more conventionally used in ion-exchange LC. The aim of the study wasto further explore the effect of experimental variables on retention, selectivity and resolution for the LC of chiral acidic drugs and related compounds oncommercially-available carbamate-based chiral anion-exchange LC stationary phases based on the pseudoenantiomers cinchona alkaloids, quinidine(QD) (8R, 9S) and quinine (QN) (8S, 9R), when using hydro – organic (i.e. polar organic solvents mixed with aqueous buffers) as would be used under‘reversed-phase’ LC conditions) mobile phases. In so doing, a secondary aim was to build knowledge on the full scope of conditions that might besuitable for separating the enantiomers of mandelic acid and its substituted analogues using hydro – organic mobile phases on these CSP.

References1. M. Lämmerhofer, M., J. Chromatogr. A (2010) 1217(6), p.814-856. 2. R. Wimal H. Perera and W. J. Lough, J Chromatogr A (2011) 1218(48):8655-63.

Nadia Jawaid and W John LoughSunderland Pharmacy School, University of Sunderland, UK

All the drug and compunds used in this research were from the University of Sunderland chemical store or were purchased from Sigma Aldrich (UK), Illustrative chromatograms shown below:

N

N

OCH

3

O

O

NH H

SH

H

silica3R

4S

1S

8

9XO+ -O

CHIRALPAK® QN-AX: (8S,9R)CHIRALPAK® QD-AX: (8R,9S)

0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40 0 5 10 15 20

Results and Discussion

Prompted by a suggestion that in comparing the use of QD-AX with other CSP with ‘hydro – organic’ mobile phases (2), it was sought first to fully explore the effect of the full range of mobile phase variables in order to findoptimal conditions for separating the enantiomers of acidic drugs and related compounds. In so doing a lack of robustness was found but this could be remedied by equilibration with 50 – 100 ml mobile phase after thecolumn had been washed with methanol – water after the end of autosampler sequences. It was then possible to proceed to look at variables such as %aqueous buffer (Fig. 4a), mobile phase pH (Fig. 4b). Enantioselectivitywas studied for a very wide range of compounds including substituted mandelic acids (Fig. 5). In some cases it was possible to study retention order (Fig. 6a, b, c). Finally, attempts were made to develop and validate amethod suitable for determining the enantiomer ratio of mandelic acid itself (just about the worst-case scenarion for this set) following enzymatic hydrolysis of esters.

Figure 1 CSP structures

2-chlorophenoxy propionic acid

4-bromomandelic acid suprofen

Figure 2 CSP drug related compounds separations

time (min) time (min) time (min)

Figure 3 Equilibration time needed

Figure 4a (top) Effect of %aqueous component4b (bottom) Effect of pH

Table 1 Mandelic acid analogues analyzed on ChiralPak QD-AX (150 mm x 4.6 mm I.D.) mobile phase 0.02 M ammonium formate, 2 ml/L formic acid 70:30 v/v, MeOH-H₂O)

Figure 5 “Psuedo-enantiomeric CSP?

Flurbiprofen:opposite retention order

Mandelic acid:same order and magnitude

Naproxen:Same order, different magnitude

Figure 6 CSP consistent retention time across three batches of CSP

Figure 7 Determination of enantiomer ratio (from phosphate buffer reaction mixture)

Robust wrt mobile phase batch, temperature; precision %RSD 1.4%; linear (r2 = ).999 and accurate in the range ratio 0.3 to 1.6 R:S