Supercritical fluid chromatography–mass spectrometry
Jeffrey P. Kiplinger, president, Paul M. Lefebvre, director of laboratory operations, Michael J. Rego, staff scientist, and
John H. Tipping, staff scientist, Averica Discovery Services
Supercritical fluid chromatography (SFC) is a well-characterized technology for analytical and preparative chiral resolution
(1). It is useful for small- to mid-scale production of single enantiomers for pharmaceutical discovery, where competitive
assays of enantiomers can help validate mechanisms and improve the precision of lead-compound assessment (2). Current FDA
guidance speaks to the desirability of comparing enantiomers early in pharmaceutical R&D (3). SFC, however, has seen limited
use in areas dominated by highly selective high-performance liquid chromatography assays due to perceptions of low sensitivity,
interfacing difficulties with detectors such as mass spectrometers, and incompatibility with hydrophilic analytes and matrices.
The authors present an example in which chiral SFC–mass spectrometry (MS) is shown to be an expedient analytical approach
to solving a bioanalysis problem.
Materials and methods.
A compound developed in a pharmaceutical lead-optimization project as a racemic mixture was separated into constituent Enantiomers
A and B using SFC with ultraviolet (UV) detection of 230 nm. The enantiomers were tested competitively in rats, and plasma
was drawn from the animals for pharmacokinetic assays. In the course of the efficacy study, unexpected off-target effects
were observed, and the team suspected in vivo racemization. Unfortunately, insufficient compound remained for further in vivo work. Only residual samples in storage vials of a mixture of the two enantiomers (not racemic, simply a mixture generated
for testing) and of the active Enantiomer A were available.
Because an SFC method for rapid separation of the enantiomers had already been developed, a rapid SFC survey of the plasma
samples for enantiomeric excess was desired. Unfortunately, detection by UV absorbance is problematic with plasma extracts
due to interferences. Extensive sample preparation was undesirable because a low recovery might jeopardize detection in the
analytical SFC systems used. SFC with selective MS detection was therefore attempted on crude extracts from the remaining
rat-plasma samples.
The residual samples of the mixtures of Enantiomers A and B and Enantiomer A were used as standards, and plasma samples were
treated only by deproteinization with acetonitrile and centrifugation prior to analysis. The standards, estimated at approximately
100 μg, were taken up in 1.0 mL of acetonitrile for analysis by SFC–MS. Injection volumes were 10 μL. For sample preparation,
450 μL acetonitrile were added to 150 μL of plasma (containing 1–5 mg/mL drug, as estimated by liquid chromatography–mass
spectrometry–mass spectrometry. The tubes were sonicated and centrifuged, and 550 μL of supernatant were removed. The samples
were dried under nitrogen and reconstituted with 500 μL of acetonitrile for analysis by SFC–MS. Injection volumes for samples
were 50 μL.
Figure 1 (SFC–MS): Analysis of a mixture of the enantiomers by a supercritical fluid chromatographic–mass spectrometric method.
(FIGURES 1–3 (SFC–MS) ARE COURTESY OF THE AUTHORS (KIPLINGER ET AL.))
|
The SFC analytical method (see Figures 1-3[SFC–MS) used a 4.6 × 100 mm RegisPack 5μm Kromasil column (Regis Technologies) using a 5-min isocratic elution with 60% carbon dioxide,
40% cosolvent (methanol:isopropyl alcohol (1:1) w/ 0.1% isopropylamine) at 4.0 mL/min.
Figure 2 (SFC–MS): (a) Comparison of the supercritical fluid chromatography (SFC)–ultraviolet (UV) (320 nm) and (b) SFC–mass
spectrometric method (mass-to-charge ratio of 330) using the plasma sample and the same SFC method. The UV signal at this
wavelength, selected for noninterference by other components in plasma, is below the detection limit. (FIGURES 1–3 (SFC–MS)
ARE COURTESY OF THE AUTHORS (KIPLINGER ET AL.))
|
The custom SFC–MS interface in the laboratory used in this study split the flow from the SFC system 20:1 immediately after
the system's backpressure regulator, and a makeup solvent (methanol: water (1:1) 0.5% formic acid, 1.0 mL/min) that facilitated
electrospray ionization was added post split. The mass spectrometer (ZQ mass spectrometer, Waters) operated in a positive
electrospray ionization mode using a selected ion monitoring mode at a mass-to-charge ratio (m/z) of 330.
Figure 3 (SFC–MS): The lack of appearance of the other enantiomer in this plasma sample chromatogram indicates no in vivo
racemization is occurring. (FIGURES 1–3 (SFC–MS) ARE COURTESY OF THE AUTHORS (KIPLINGER ET AL.))
|
Results and discussion
. Standard samples of the mixture of Enantiomers A and B (see Figure 1 [SFC–MS]) and of the active Enantiomer A (not shown) were analyzed using the previously developed resolution method to define enantiomer
retention times and to verify detection by MS. Sequential injections of the racemate and Enantiomer A indicated that carryover
of Enantiomer B (and thus presumably A) was negligible. Figure 2 (SFC–MS) compares detection of the compound in rat plasma at 1.85 min by MS and UV detection at 320 nm. As predicted, with UV detection
the signal is below the limit of detection. Nine plasma samples were analyzed using this method and detection methodology. Figure 3 (SFC–MS) shows that the drug did not racemize to a detectable degree in the in vivo study. SFC–MS, used with a chiral separation method identical to the one used to produce the tested enantiomers proved useful
for studying their potential racemization in pharmacokinetic studies.
References (SFC–MS)
1. M. Venturea et al., J. Chromatogr. A
1036 (1), 7–13 (2004).
2. J.D. Pinkston et al., Anal. Chem.
78 (21), 7467–7472 (2006).
3. FDA, Guidance for Industry: Drugs: Development of New Stereoisomeric Drugs (Rockville, MD, April 2011).
|
