INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

Zidovudine (ZDV; 3'-azido-3'-deoxythymidine) was the first drug approved for the treatment of human immunodeficiency virus (HIV) infection. The parent compound (ZDV) is not active against HIV, since it needs to be metabolized in the host cells to 5'-ZDV triphosphate (ZDV-TP), which acts as a competitive inhibitor of HIV reverse transcriptase or is incorporated into the viral genome, terminating DNA chain elongation (7, 9). Recently, several studies have shown that the intracellular concentration of the triphosphate moieties of the nucleoside reverse transcriptase inhibitors (NRTIs) correlated better with the virologic response than did levels of the parent compound in plasma (20, 21). Furthermore, Fletcher et al. demonstrated a strong relationship between intracellular ZDV-TP concentrations and an increase in the percent change in CD4⁺ cells from the baseline (5).

In recent years, the quantitation of ZDV-TP and other NRTI triphosphates in peripheral blood mononuclear cells (PBMC) has been a high priority in clinical pharmacology. However, one of the major problems in the measurement of intracellular ZDV-TP is the small amounts present in patients (low numbers of femtomoles per 10⁶ cells). Thus, specific and sensitive analytical methodologies need to be developed in order to measure these small quantities without drawing large amounts of blood from patients.

Several approaches have been reported for the determination of ZDV-TP, including a combination of high-performance liquid chromatography (HPLC) and radioimmunoassay (8, 17, 22). This methodology accomplished the separation of ZDV monophosphate (ZDV-MP), ZDV diphosphate (ZDV-DP), and ZDV-TP by strong anion-exchange HPLC. However, the limit of detection (LOD) was only 200 fmol/10⁶ cells and the method was also time consuming since it took more than 45 min for each sample to be processed (3, 11). Recently, a variation of this approach was developed in which the separation method was replaced with anion-exchange solid-phase extraction (AX-SPE) (14). This method achieved a LOD of 20 fmol/10⁶ cells. However, the method did not use an internal standard in the quantitation process, assuming complete recovery from all of the samples analyzed (14). In addition, the method cannot determine two or more NTRI triphosphates simultaneously.

Several liquid interfaces for mass spectrometry have been used to study nucleosides and nucleotides (1, 4, 6, 12, 16, 25). Electrospray ionization (ESI) is one of the most common interfaces used to obtain structural information and fragmentation patterns of nucleosides (1, 6, 12, 25). ESI-tandem mass spectrometry (MS/MS) provides even better specificity and selectivity when daughter ion experiments are performed in which a specific fragment ion is selected for quantitation analyses (18, 24).

In this report, we describe a method combining an AX-SPE procedure with HPLC-MS/MS to measure intracellular ZDV-TP in PBMC using azidodeoxyuridine (AZdU; 3'-azido-3'-deoxyuridine) as the internal standard. The LOD by this methodology is 4.0 fmol/10⁶ cells with a linear concentration range of 4 orders of magnitude (4.0 to 10,000 fmol/10⁶ cells). This method was successfully applied to measure ZDV-TP in PBMC obtained from HIV-infected patients. Furthermore, this methodology could be used to perform simultaneous detection of two or more NRTIs, such as ZDV-TP and lamivudine triphosphate (3TC-TP).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

Chemicals. ZDV, AZdU, sodium acetate, and acid phosphatase (type XA) were obtained from Sigma Chemical Co., St. Louis, Mo. [³H]ZDV, [³H]ZDV-TP, ZDV-TP, 3TC, and 3TC-TP were purchased from Moravek Biochemicals, Brea, Calif. Potassium chloride, acetonitrile, methanol, and glacial acetic acid (American Chemical Society certified) were obtained from Fisher Scientific, Fair Lawn, N.J. Anion-exchange Sep-Pak plus QMA cartridges were purchased from Waters Co., Milford, Mass. XAD resin was obtained from Serva, Heidelberg, N.Y. RPMI 1640 medium, glutamine, nonessential amino acids, penicillin-streptomycin, and fetal calf serum were obtained from BioWhittaker, Baltimore, Md.

Cell culture and incubation. CEM_ss cells, obtained from the National Institutes of Health AIDS Research and Reference Reagent Program, were grown at a density of 0.6 × 10⁶ to 0.8 × 10⁶/ml in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, 1% L-glutamine, and 1% penicillin-streptomycin. These cells were incubated for 24 h with ZDV at final concentrations of 5.0 and 10.0 µM. After incubation, the cells were pelleted by centrifugation and then subjected to extraction with 70% methanol (20 µl/10⁶ cells). The samples were centrifuged, and the supernatants were collected and stored at 80°C until analysis.

Sample collection from HIV-infected patients. Two tubes of venous blood (16 ml) were sampled in Vacutainer CPT tubes at different times after oral dosing. PBMC were separated from erythrocytes by centrifugation at 1,500 × g for 20 min at room temperature. PBMC were recovered and counted in a Z2 series system (Coulter, Hialeah, Fla.), subjected to extraction with 70% methanol, and stored at 80°C until analysis. All patients provided informed consent, and the Institutional Review Board of the Medical Sciences Campus at the University of Puerto Rico approved the protocol.

Intracellular ZDV-TP isolation. AX-SPE was used to separate ZDV nucleotides as previously described (14). Briefly, the cartridges were preconditioned with 1.0 M KCl and washed with 5 mM KCl. The sample was loaded onto the cartridges, and 11 ml of 74.5 mM KCl was added. The fraction containing ZDV-TP was eluted with 3.3 ml of 1.0 M KCl. Recovery was performed with [³H]ZDV-TP and determined to be >95%. Cleavage of phosphate groups was accomplished by the addition of 2 U of acid phosphatase per ml and incubation for 30 min at 37°C and adjustment of the fraction to pH 4 with sodium acetate. After enzyme digestion, AZdU (internal standard; 100 ng/ml) was added to the extract and recovered simultaneously with ZDV by using an XAD column. The XAD column was preconditioned with water before sample loading. The sample was desalted with 20 ml of water. ZDV and AZdU were eluted with 2 ml of acetonitrile. Desalting and recovery of ZDV and AZdU using XAD columns was >99%. Samples were dried in a CentriVap console (Labconco, Kansas City, Mo.) and reconstituted with 200 µl of the mobile phase prior to HPLC-MS/MS analysis.

HPLC-MS/MS. HPLC analysis was performed on a 1050 system (Hewlett-Packard, San Fernando, Calif.) using a Hypersil C₁₈ reversed-phase column (100 by 2.1 mm; 3 µm). The mobile phase consisted of an acetonitrile-methanol mixture (10:30, vol/vol) with 0.25% acetic acid at a flow rate of 0.20 ml/min. The sample volume injected was 20 µl. A Quattro II triple quadrupole mass spectrometer (Micromass, Manchester, United Kingdom) was used for the analysis in the multiple reaction monitoring (MRM) mode. Sample introduction was through ESI in the positive-ion mode. The cone voltage was set between 15 and 20 V, and the source temperature was 120°C. Ions were collisionally activated at a collision energy of 7 eV with the cell pressure set to 7.0 × 10¹ Pa of argon. MRM data were acquired and analyzed by using MassLynx software (v. 2.0).

Data analysis. Concentrations of analytes were determined by using peak-area ratios for ZDV and AZdU. A calibration curve from ZDV-TP solutions was prepared every time a series of samples were analyzed. Linear regression analyses were performed by using seven ZDV-TP concentrations distributed between 4.0 and 10,000 fmol/10⁶ cells. Regression coefficients (r²) were better than 0.997 for all calibration curves.

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

The objective of this study was to develop a sensitive and selective method of intracellular ZDV-TP quantitation suitable for the determination of the intracellular pharmacokinetics of this anabolite in ZDV-treated patients. In this method, nucleotides from PBMC were extracted, ZDV-TP was separated from ZDV-MP and ZDV-DP through AX-SPE, and the phosphate groups were removed from ZDV-TP with acid phosphatase. The sample was desalted through an XAD column and quantified by HPLC-MS/MS.

Mass spectrum characterization. Figure 1 shows the ESI mass spectra of ZDV and AZdU, which are the protonated molecules (M + H)⁺ at m/z = 268 for ZDV and m/z = 254 for AZdU, the most abundant ions. Furthermore, the fragmentation of both nucleosides occurred through the glycosidic bond producing daughter ions at m/z = 127 and m/z = 113 for ZDV and AZdU, respectively. The glycosidic bond cleavage is induced by protonation of the base moiety, a general mechanism observed in the fragmentation of nucleosides in the ESI positive-ion mode (12). For the determination of ZDV, MRM only allows the ZDV protonated molecule (m/z = 268) to pass through the first quadrupole (MS-1). The daughter ion at m/z = 127 is formed in the collision cell and extracted for quantitation in the second quadrupole (MS-2). We used the same procedure for the determination of AZdU (MS-1, m/z = 254; MS-2, m/z = 113). MRM provides better selectivity and a better signal-to-noise ratio than a single mass spectrometer, enhancing the sensitivity and LOD of the method.

View larger version (21K): [in this window] [in a new window]   FIG. 1.   Electrospray single mass spectra of AZdU (top) and ZDV (bottom). Experimental conditions were as described in Materials and Methods. The protonated molecular ion for AZdU appears at m/z = 254, and the fragment ion at m/z = 113 is assigned to the protonated base due to the cleavage of the glycosidic bond as illustrated by the dashed line. The protonated molecule of ZDV appears at m/z = 268, and the fragment ion at m/z = 127 is also due to the cleavage of the glycosidic bond. On the y axis is the normalized ion intensity from the fragment ions.

Interference from endogenous nucleotides. We obtained ESI-MS/MS chromatograms from 10⁷ CEM_ss cells without ZDV with AZdU as the internal standard and ZDV-TP from 10⁷ CEM_ss cells incubated with 5 µM ZDV for 24 h (data not shown). For the CEM_ss cells without ZDV, no appreciable signal appears in the chromatogram. Thus, endogenous nucleotides or other compounds from cell samples do not interfere with the ZDV-TP quantitation process. The chromatogram of AZdU (internal standard) showed a strong signal with a retention time of approximately 2.00 min. A similar chromatogram was obtained for ZDV with a retention time of about 2.20 min. We determined that AZdU did not interfere with the ZDV signal despite the proximity of the retention times. Moreover, the short retention times of both compounds increase the throughput of samples (15 samples/h). This is a critical consideration for the use of this method in large clinical trials.

Standard curve and statistics. Figure 2 shows a typical calibration curve constructed from the ratio between the areas of ZDV and AZdU chromatograms versus ZDV-TP concentration. This calibration curve was obtained by using different ZDV-TP standard concentrations to spike PBMC and passing them through the entire methodology (AX-SPE, enzyme digestion, and XAD column chromatography). The LOD obtained with this methodology is 4.0 fmol/10⁶ cells (coefficient of variation [CV], <8%, n = 3; accuracy, <10%, n = 3), with a linear range of 4.0 to 10,000 fmol/10⁶ cells. By using [³H]ZDV-TP, we determined the recovery of ZDV-TP to be >95% for the complete method at different ZDV concentrations (n = 3).

View larger version (26K): [in this window] [in a new window]   FIG. 2.   Calibration standard curve for ZDV-TP constructed from standard solutions (4, 20, 59, 200, 590, 2,000, and 10,000 fmol/106 cells) passed through the complete methodology. The data are for two determinations for each concentration. The equation describing the complete range is y = 0.002x  0.008 with a regression coefficient (r2) of 0.998.

ZDV-MP and ZDV-TP in CEM_ss cells. Table 2 shows the concentrations of ZDV-MP and ZDV-TP obtained from the incubation of CEM_ss cells in 5.0 and 10.0 µM ZDV for 24 h. As shown in previous studies (13-15, 23), intracellular ZDV-MP concentrations were higher than those of ZDV-TP. This is due to inefficient activity of thymidylate kinase, the enzyme responsible for the conversion of ZDV-MP to ZDV-DP. As part of the same effect, the ZDV-MP concentration almost doubled (from 34.3 to 51.2 pmol/10⁶ cells) when the extracellular ZDV concentration increased from 5.0 to 10.0 µM, whereas the ZDV-TP concentration changed insignificantly (from 1.18 to 1.04 pmol/10⁶ cells).

Patient samples. The methodology described here was used to measure the intracellular concentrations of ZDV-TP in six samples from Hispanic HIV-infected patients treated with the standard ZDV therapy (300 mg twice a day). Blood samples were drawn at different time points, depending on the availability of the patients in the clinic. Figure 3 shows a typical chromatogram for ZDV obtained from one HIV-infected patient, showing an excellent signal above the background even at this low concentration (41 fmol/10⁶ cells). Table 3 shows intracellular ZDV-TP concentrations in HIV-infected patients that ranged between 41 and 193 fmol/10⁶ cells. These values are not different from those obtained previously by using other methodologies (2, 10, 13-15). In addition, the interpatient variability in our population is similar to that of other populations (2, 10, 13-15). The highest ZDV-TP concentration was observed in patient 1 (193 fmol/10⁶ cells), who was sampled 2 h after drug administration. This ZDV-TP concentration is similar to that reported by Fletcher et al. for patients who had an increase in the percent change of CD4⁺ cells over the baseline (5). The only other patient who was sampled before 6 h also had ZDV-TP concentrations above 100 fmol/10⁶ cells (patient 2; 137 fmol/10⁶ cells). All other patients were sampled more than 10 h after drug administration and had measurable intracellular ZDV-TP concentrations.

View larger version (15K): [in this window] [in a new window]   FIG. 3.   Chromatograms for AZdU (internal standard) and ZDV obtained from patient 3. The ZDV signal was obtained from an intracellular ZDV-TP sample passed through the method. Chromatographic conditions are as described in Materials and Methods. Retention times for AZdU and ZDV were 2.00 and 2.20 min, respectively. The y axis is the normalized ion intensity obtained from daughter ions.

ZDV-TP and 3TC-TP. Although the aforementioned methodology is a step forward for the quantitation of ZDV-TP in clinical trials, the use of ZDV as monotherapy is no longer the best therapeutic option. Therefore, we need to develop new methodologies to quantify different NRTI triphosphates simultaneously. Recently, two methods were reported that measure intracellular 3TC-TP concentrations in HIV-infected patients (13, 19). Both methods used AX-SPE to separate 3TC-TP from other nucleotides, and the quantitation process was performed by either radioimmunoassay (13) or HPLC with UV detection (19). However, both methods could not measure ZDV-TP and 3TC-TP simultaneously. Figure 4 shows the chromatograms for the simultaneous determination of ZDV-TP and 3TC-TP in the same sample. ZDV-TP and 3TC-TP standard solutions were used together to spike 10⁷ CEM_ss cells which were passed through the complete methodology with minor modifications. It is clear from Fig. 4 that ZDV and 3TC can be detected in the same sample. The 3TC-TP quantitation process is still being developed. This approach provides the advantage of allowing two or more NRTI triphosphates to be monitored without the need to obtain additional blood from the patient.

View larger version (12K): [in this window] [in a new window]   FIG. 4.   Chromatograms for 3TC and ZDV obtained from standard 3TC-TP and ZDV-TP solutions used together to spike CEMss cells. The standard solutions were processed throughout the complete procedure. Retention times for ZDV and 3TC were 2.20 and 5.10 min, respectively. Experimental conditions were the same as for Fig. 3.