Development of an LC-ESI-MS-MS Method With Formate Anion Attachment for Detecting Neutral Molecules in Rat Plasma

Quantitative bioanalysis is the most important application area of liquid chromatography coupled with atmospheric pressure ionization (API) tandem mass spectrometric detection (LC-API-MS-MS) in terms of the number of instruments employed and the number of analyses performed.1 The most commonly used API sources are electrospray ionization (ESI) or Ionspray (pneumatically assisted electrospray) (Applied Biosystems/MDS Sciex, Foster City, CA) and atmospheric pressure chemical ionization (APCI). While ESI is suitable for polar and chargeable molecules and APCI expands the range of application to less polar compounds, the analysis of neutral and nonpolar compounds is challenging due to their poor ionization efficiency.

Several strategies to overcome the limitations of ESI and APCI sources have evolved, including chemical derivatization,2 coordination-ion-spray,3 and atmospheric pressure photon ionization.4 In another approach, the attachment of small organic or inorganic ions to a neutral molecule is an important ionization mechanism in both positive and negative ion ESI mass spectrometry, and adduct ion formation is a frequently observed phenomenon in mass spectra originating from the ESI process. This article presents the application of an anion attachment strategy for detecting two neutral molecules (Pfizer compound A and Pfizer compound B) in biological matrices. Although the compounds give poor ionization efficiencies with both ESI and APCI sources, they form anionic adduct ions with a variety of anions, including formate and acetate, when an ESI source is used.

In negative ion multiple reaction monitoring (MRM) mode, the formate adduct ions of the compounds are selected in the first quadrupole (Q1), fragmented in the second quadrupole (Q2), and the dissociated formate ions are detected in the third quadrupole (Q3). The anion attachment strategy shown here offers a very simple and effective way to improve the performance of the LC-ESI-MS-MS methods for certain neutral molecules without derivatization reactions to enhance analyte ionization efficiencies. A sensitive, high-throughput LC-ESI-MS-MS method has been developed for detecting ESI mass spectrometry-insensitive compounds in biological matrices using an anion attachment strategy.

Experimental

Chemicals and reagents

Pfizer compound A (MW 287) and Pfizer compound B (MW 294) were synthesized by Pfizer Global Research and Development (Ann Arbor, MI). Formic acid (88%, analytical grade [AR]) and 50% ammonium hydroxide (AR grade) were obtained from Mallinckrodt Baker (Phillipsburg, NJ). Protein precipitation filter plates (96-well) were obtained from Whatman Inc. (Florham Park, NJ). All HPLC-grade solvents were obtained from Mallinckrodt Baker (Paris, KY). Various lots of rat and monkey plasma with K3EDTA as the anticoagulant were purchased from Bioreclamation (East Meadow, NY).

LC-MS-MS system

The LC-MS-MS system consisted of a CTC PAL autosampler (CTC Analytics, Zwingen, Switzerland), an integrated pump system consisting of LC-10AD pumps with an SCL-10A system controller and a DGU-14A degasser (Shimadzu, Kyoto, Japan), and an API 4000 triple quadrupole mass spectrometer (Applied Biosystems/MDS Sciex). Chromatographic separations were performed on a 50 mm × 2.1 mm i.d., 3-μm, Thermo Hypersil Gold C18 column (Thermo Electron Corp., San Jose, CA) using gradient elution at a flow rate of 300 μL/min. Mobile phase A and B were 2 mM formic acid and methanol, respectively. All tubing connections were polyetherether ketone (PEEK) with 0.005-in. i.d. The mass spectrometer was operated in negative ion MRM mode for quantitative work. In order to establish the appropriate MRM conditions for Pfizer compounds A and B, solutions of the standards (1.00 μg/mL in methanol) were infused into the mass spectrometer, and mass spectrometric parameters such as declustering potential (DP) and collision energy (CE) were optimized to maximize the intensities of the proposed anionic adduct ions and the corresponding product ions. Collision-induced dissociation (CID) mass transitions of m/z 332 → 45 and m/z 339 → 45 were monitored for Pfizer compound A and Pfizer compound B, respectively. The dwell time was 200 msec for each mass transition.

Preparation of calibration standards and validation samples

The stock solutions of Pfizer compounds A and B were prepared by dissolving appropriate amounts of solid material (typically 10 mg) in dimethyl sulfoxide (DMSO) to give concentrations of 1.0 mg/mL, respectively. The stock solutions were used to prepare the calibration standards in rat plasma at concentrations of 0.100, 0.200, 1.00, 5.00, 10.0, 50.0, 90.0, and 100 ng/mL for Pfizer compound A, and concentrations of 0.400, 0.800, 2.00, 20.0, 100, 200, 360, and 400 ng/mL for Pfizer compound B. Similarly, quality control samples (QCs) were prepared from separate weighings at concentrations of 0.100, 0.300, 5.00, and 80.0 ng/mL (Pfizer compound A), and concentrations of 0.400, 1.00, 20.0, and 320 ng/mL (Pfizer compound B), respectively. For the Pfizer compound A assay, an acetonitrile solution of 10.0 ng/mL Pfizer compound B was used as the working internal standard solution (WISS) and vice versa. The DMSO stock solutions were stored at 4 °C, and the plasma samples were stored frozen at –20 °C.

Sample preparation

Plasma samples were thawed to room temperature, then vortexed to render the thawed sample homogeneous . Plasma samples (100 μL each) were transferred to a 96-well protein precipitation filter plate (Whatman) containing 300 μL of the working internal standard solution. The filtrates were evaporated to dryness under a stream of nitrogen at 37 °C, and the dry residues were reconstituted in 100 μL of methanol /water (30/70, v/v) solution. Sample extracts (10.0 μL) were injected for LC-MS-MS analysis.

Results and discussion

ESI mass spectrum

Figure 1 - General chemical structure of Pfizer compounds A and B.

Pfizer compound A and Pfizer compound B are among a group of analogous molecules that have the following common chemical structure (Figure 1): two electron withdrawing groups (EWGs) on a phenyl ring and an ether linkage to an alkyl group. They do not contain ionic or chargeable functional groups such as amino and carboxylic acid groups that can be ionized in ESI or APCI sources.

Figure 2 - Negative ion mass spectra of Pfizer compound A using a) ESI at DP 5 V, b) ESI at DP 110 V, and c) APCI at 350 °C.

Figure 2 shows typical mass spectrometric full scans of Pfizer compound A in negative ion mode under various ionization conditions including ESI and APCI . The molecule has a molecular weight of 287 (MW 287) . Molecular ions [M+1]+ in positive ion mode (MS spectra not shown) and [M–1] in negative ion mode were not observed. This is not surprising, since the compound contains no chargeable functional groups. However, using ESI in negative ion mode and at low declustering potential (DP 5 V), characteristic ions at m/z 322, 332, 349, 366, and 550 (Figure 2a) were observed. These negative ions appeared as the compound was infused and disappeared as the infusion was stopped. It was later confirmed that these negative ions were adduct ions formed between Pfizer compound A with residue anions such as Cl, HCOO, NO3, Br, and C4F9COO, respectively. The presence of two EWG groups (EWG1 and EWG2) on the aromatic ring makes the molecule a good π -acid for forming charge transfer complexes with small anions. Trace amounts of anions (~1.0 ppm) are commonly present in water and other solvents. The signal intensities of these anionic adduct ions decreased as the DP was increased from 5 V to 110 V (Figure2b), and no adduct ions were observed when an APCI source was used (Figure 2c). The intense peak of m/z 263 (Figure 2a and 2c) is from residue nonafluoropentanoic acid (C4F9COOH, MW 264) that was an ion-pairing reagent used for the analysis of a polar compound in another LC-MS-MS assay. Nonafluoropentanoic acid is difficult to remove completely from the LC-MS-MS system, even with extensive cleanup procedures.

Figure 3 - a) CID mass spectrum of formate adduct ions of Pfizer compound A, b) CID dissociation profiles, and c) MRM chromatograms of various anionic adduct ions of Pfizer compound A.

Figure 3a shows a CID spectrum of the adduct ions of Pfizer compound A with formate. CID dissociation profiles of the adduct ions of Pfizer compound A with both formate and C4F9COOanions are shown in Figure 3b. MRM chromatograms of several anionic adduct ions monitored simultaneously are shown in Figure 3c. The intensity of [M–1] ions (m/z 286) is very weak compared to that of formate anions since the ion–molecule association between the formate anion and the neutral molecule is noncovalent and weaker than a typical covalent bond (Figure 3a). Low collision energies (CE less than 20 V) were needed to break the ion–molecule association (Figure 3b). As shown in Figure 3c, anionic adduct ions of Cl, NO3, HCOO, CF3COO, and C4F9COO can be used for the chromatographic detection of Pfizer compound A by an alternative means of tandem mass spectrometric detection. The chromatographic separation was carried out on a reversed-phase column with gradient elution using water and acetonitrile as mobile phases. Initially, C4F9COO adduct ions with Pfizer compound A for the MRM detection (m/z 550 → 263) and subsequent quantitation were chosen because theoretically there are fewer mass spectral interferences in the higher mass range. The concentration of the ion pairing reagent (C4F9COOH) in the mobile phase had to be within the range of micromolar (μM) in order to have low and stable background noises, and it was found that the analyte peak responses decreased after using the LC-MS system for real plasma samples or switching the LC-MS system between different assays. This might be caused by competition from other residue anions including formate, since the concentration of the ion-pairing reagent (C4F9COOH) was low in the μM range. Formate adduct ions were chosen for the MRM detection (m/z 332 → 45) and the subsequent quantitation of Pfizer compound A since formic acid is among the most common buffers in LC-MS-MS analysis. It was found that 2 mM formic acid added to the mobile phase provided enough formate anions for the adduct ion formation without elevating background noise.

LC-ESI-MS-MS method

Figure 4 - LC-ESI-MS-MS chromatograms of a) blank rat plasma, b) Pfizer compound A at 0.100 ng/mL in rat plasma, and c) internal standard (Pfizer compound B), 10.0 ng/mL in rat plasma.

Figure 5 - LC-ESI-MS-MS chromatograms of a) blank rat plasma, b) Pfizer compound B at 0.400 ng/mL in rat plasma, and c) internal standard (Pfizer compound A), 10.0 ng/mL in rat plasma.

Previously, for Pfizer compound A, an LC-ESI-MS-MS method using a conventional approach could only achieve a limit of quantitation (LOQ) of 200 ng/mL in rat plasma, and a GC-MS method with an LOQ of 8.00 ng/mL required liquid–liquid extraction for sample cleanup. The use of the formate adduct ions provides an alternative means of tandem mass spectrometric detection of the nonionic compound, Pfizer compound A. As shown in Figure 4, using formate anion attachment, an LOQ of 0.100 ng/mL of Pfizer compound A in rat plasma was achieved. With a similar formate anion attachment strategy, an LC-ESI-MS-MS assay was developed for Pfizer compound B, a structural analog of Pfizer compound A, at an LOQ of 0.400 ng/mL in rat plasma (Figure 5).

Figure 6 - Calibration standard curves for a) Pfizer compound A (0.100–100 ng/mL) and b) Pfizer compound B (0.400–400 ng/mL) in rat plasma.

These two LC-ESI-MS-MS methods were successfully validated under the authors’ Standard Operating Procedures (SOPs), which are in conformance with the current FDA guidelines for bioanalytical method validation.5Figure 6 shows typical calibration standard curves of these two compounds in rat plasma, and the correlation coefficients (R2) were >0.999 using a 1/concentration2 weighed quadratic regression model. Using 100-μL aliquots of rat plasma, assay concentration ranges were validated from 0.100 to 100 ng/mL for Pfizer compound A and from 0.400 to 400 ng/mL for Pfizer compound B (MW 294). Plasma sample preparation was carried out using protein precipitation with acetonitrile in a 96-well filter. Some of the validation results are presented in Table 1. The data show that the two methods are consistent and reliable with acceptable values of precision and accuracy.

Application to preclinical samples

Figure 7 - LC-ESI-MS-MS chromatograms of a) a solvent blank in an unprinted tube, b) a solvent blank in a printed tube, and c) a study sample extract (0.160 ng/mL).

Using the above LC-ESI-MS-MS methods, over 1000 plasma samples were analyzed from several preclinical and toxicokinetic studies for Pfizer compounds A and B. When analyzing Pfizer compound A in rat plasma study samples, two interfering peaks were observed close to the analyte peak. Later it was possible to conclude that the interfering peaks are related to the printing ink materials on a typical polypropylene sample tube (Figure 7a and b). By carefully selecting sample tubes that have no printed marks for the work and by using a better chromatographic separation, the authors were able to keep the interfering effects to a minimum. Figure 7c shows an MRM chromatogram of a study sample containing 0.160 ng/mL of Pfizer compound A in rat plasma. For those study samples with concentrations close to the LOQ (0.100 ng/mL in rat plasma), the interfering peaks might affect the analyte integration and quantitation.

Conclusion

LC-MS-MS analysis of neutral or nonionic compounds is a challenging area in a bioanalytical laboratory supporting drug discovery and development. The above two examples demonstrate that an alternative approach of using formate attachment can lead to sensitive and selective LC-ESI-MS-MS methods without involving time-consuming procedures such as sample derivatization or GC-MS-MS analysis. Adduct ion formation is very common in the ESI process, and the use of adduct ions for LC-ESI-MS-MS quantitation can be applied to other neutral compounds and nonionic compounds.

References

  1. Niessen, W.M.A. J. Chromatogr. A1999, 856, 179–97.
  2. Van Berkel, G.J.; Quirke, J.M.E.; Tigani, R.A.; Dilley, A.S.; Covey, T.R. Anal. Chem. 1998, 70, 1544–54.
  3. Bayer, E.; Gfroer, P.; Rentel, C. Angew. Chem. Int. Ed.1999, 38, 992–5.
  4. Robb, D.B.; Covey, T.R.; Bruins, A.P. Anal. Chem.2000, 72, 3653.
  5. U.S. Food and Drug Administration/Center for Drug Evaluation and Research, Guidance for Industry Bioanalytical Method Validation,www.fda.gov/cder/guidance/index.htm.

The authors are with Pharmacokinetics, Dynamics & Metabolism, Pfizer Global Research and Development, Michigan Laboratories, Pfizer Inc., 2800 Plymouth Rd., Ann Arbor, MI 48105, U.S.A.; tel.: 734-622-4117; fax: 734-622-5115; e-mail: [email protected]. The authors thank Mr. David Weller, Ms. Laura Penn, and Mr. Michael Bass for their help in the early stages of assay development. Part of this work was first presented at the 29th International Symposium on High Performance Liquid Phase Separations and Related Techniques, Stockholm, Sweden, June 2005.