Determination of All-trans-Retinoic Acid in Rat Plasma by LC-MS-MS
Authors
Christopher J.L. Buggé, Michael P. Sullivan
Introduction
All- trans -retinoic acid (RA), (Figure 1) is a pharmacologically important endogenous retinoid used topically to treat dermatological disorders. Accurate and sensitive measurement of blood levels is required following such treatment to determine possible systemic exposure.
The determination of retinoids in biological fluids is notoriously challenging because of their sensitivity to light and oxygen, the problem of separating geometrical isomers and the presence of endogenous levels of the analytes.
These problems were overcome and we present here a rugged, sensitive method for the quantitation of retinoic acid in rat plasma by LC-MS-MS that is suitable for determinations of systemic exposure following topical dosing of RA.
Experimental
Chemicals
RA and acitretin (internal standard) were obtained commercially; All chemicals were AR grade and all solvents were HPLC grade. K 3 EDTA rat plasma was fortified with 0.5% w/v ascorbic acid as an antioxidant preservative.
Standards/Solutions
Stock solutions of the retinoids were prepared in ethanol, stored at 4 °C in the dark and were used to prepare a series of calibration standards. 13-cis-RA was also added at the same concentrations as RA in order to provide evidence of the separation of these isomers. The internal standard working solution was prepared at 2.00 µg/mL. Daily calibration standards were prepared by fortifying 0.200 mL of plasma with 10 µL of standards and 10 µL of working internal standard solutions. All operations were done under yellow lighting to prevent photoisomerization.
Blood/Plasma Collection
Whole blood was collected under yellow lighting from healthy rats into K 3 EDTA Vacutainers. The samples were immediately centrifuged and the plasma transferred to glass tubes containing sufficient ascorbic acid to provide a final concentration of 0.5% w/v. The samples were immediately frozen at -20 °C in the dark and transferred the next day to the analytical laboratory where they were stored at –70 °C until analysis.
- SCIEX API 4000 LC-MS-MS with heated nebulizer source in negative ion MRM mode.
- Mobile Phase: Ethyl acetate/hexanes
- Shimadzu LC 10 AD pump
- Monochrom SIL 4.6 x 150 mm, Varian
- Perkin-Elmer series 200 autosampler
Ions monitored: RA m/z 299 --> 255 , acitretin (int. std.) m/z 325 --> 281
Quantitation
Calibration curves and QC samples were prepared in rat plasma that was cleansed of endogenous retinoic acids. Weighted linear least squares regression (1/x 2 ) was used to determine the daily calibration curves for each analyte. In standard addition experiments, non-treated plasma was analyzed unfortified and after fortifying at 8.00 ng/mL; any endogenous concentrations were then subtracted from the total measured concentrations to determine the analytical accuracy, and matrix effect.
Results
Method Validation
The method was validated by analyzing plasma QC samples on each day. A minimum of three replicates from three QC pools were included on each day. These QC pools were stored in the dark at -70° C and were used to assess inter- and intra-day assay variation and short-term stability.
Specificity
Ion chromatograms showed excellent separation of the all- trans -RA from 13- cis -RA with sharp peaks and a total run time of 5 min between samples (Figure 2). The limit of detection for the retinoids was estimated at approximately 0.5 ng/mL for RA and 13- cis -RA.
Endogenous Retinoic Acid Removal
Ion chromatograms of normal untreated plasma pooled from rats showed peaks due to endogenous RA, and 13- cis -RA. Following the cleansing treatment, the same plasma showed no detectable retinoid (figure 1) and was used as a matrix to prepare calibration standards and QC samples.
Validation
The lower limit of quantitation was established at 2.00 ng/mL for RA and the method was validated over the range 2.0 to 50.0 ng/mL (Tables 1 and 2). Linearity, precision, accuracy, and ruggedness were all found to be acceptable within the FDA guidelines. Extraction recovery was > 70% for all compounds.
Stability of QC Samples (Yellow lighting)
Bench-top stability (27h at room temp.); freeze/thaw stability (4 cycles); and extract stability (21h at room temp) gave results that were within ±15% of theoretical (data not shown).
Extended stability of the retinoids in plasma for 8 weeks at –70° C in the dark was within ±15% of theoretical.
Conclusions
The problematic areas of oxidation and sensitivity to light; separation of geometrical isomers; and the presence of endogenous plasma levels of the retinoids, were all overcome with this method.
- LC-MS-MS in negative ion mode provided sufficient sensitivity; no analyte derivatization was necessary.
- Good chromatographic separation provided by the HPLC column enabled baseline resolution of geometrical isomers, and the mobile phase constituents allowed a simple isocratic system to be used with a run time of 5 minutes. RA was separated from its geometric isomer 13-cis-RA and other endogenous compounds in one run. The need for a gradient system was eliminated.
- Approximately 200 samples per day can be analyzed per analyst/instrument.
- Yellow lighting was used in all laboratory operations and amberized glassware reduced possible photoisomerization to insignificant levels. The extraction was carried out at close to neutral pH, thus hydrolysis of glucuronides or other conjugates was minimized.
- Plasma that was intended to be used for calibration and QC samples was first treated by a special process to remove endogenous retinoic acids. Other workers (references 1 and 2) have used water or phosphate-buffered saline or solvent for calibration standards which may have provided unreliable absolute accuracy of retinoic acid levels in plasma due to differential recovery. Our method used a plasma matrix, which results in superior accuracy. Standard addition experiments using untreated plasma validated this approach.