The high performance liquid chromatography electrospray ionization mass spectrometry analysis of diverse basic pharmaceuticals on cyanopropyl and pentafluorophenylpropyl stationary phases

https://doi.org/10.1016/S0731-7085(00)00348-4Get rights and content

Abstract

Cyanopropyl (CN) and pentafluorophenylpropyl (PFPP) modified silica columns give good retention and good peak shape for the high performance liquid chromatography/electrospray ionization/mass spectrometry (HPLC/ESI/MS) analysis of several classes of basic drugs. These phases enhance the ESI-MS signal by providing good retention of basic drugs with a mobile phase containing 90% acetonitrile. With C18 columns, in order to achieve good retention of basic drugs, only a mobile phase containing less than 40% acetonitrile can be used. Higher concentrations of acetonitrile produce a larger MS signal in the ESI process; the MS signal was a factor of 9 and 12 times greater on the CN and PFPP phases when compared with the C18 phase for the analysis of codeine. The C18 phase required only 4.0–6.0% acetonitrile to obtain the same retention time for codeine. The CN and PFPP stationary phases can be used for the analysis of a range of basic drugs, including many compounds which are poorly retained on the popular C18 and C8 stationary phases. Applications of CN and PFPP columns in the HPLC/ESI/MS of basic drugs include the analysis of antimalarials, such as quinine, bronchodialators, such as salbutamol and tulobuterol, cardioactive drugs, such as procainamide and β-blockers, tricyclic antidepressants (TCAs), such as protriptyline and trimipramine and alkaloids, such as morphine and codeine. The CN and PFPP phases are also useful for the analysis of bufuralol and its metabolite, hydroxy-bufuralol. All the above analyses were performed using the same mobile phase, 90% acetonitrile; thus the HPLC method development process was expedited. The CN and PFPP phases also gave reproducible retention times and peak shape after more than 8 h of analyses.

Introduction

Bonded C8 and C18 phases have been widely used for the HPLC analysis of basic drugs [1], [2]. However, buffer salts and various additives, such as pairing- or counter-ions, which include alkylsulphonates, alkylamines or quaternary ammonium compounds, are often needed with these hydrophobic phases to achieve good peak shape and good retention [3], [4]. In ESI-MS, volatile buffers and additives must be used; thus many conventional additives and ion-pairing agents must be avoided. Even when ion-pair agents, such as trifluoroacetic acid [5], or ion suppressing agents, such as triethylamine [6], are used the ESI-MS signal is decreased. To analyze drugs such as codeine or morphine by HPLC without the use of silanol suppressing agents is problematic because these basic molecules (pKa 8.2 and 9.9, respectively) interact with the silanols to produce tailing peaks [7]. In addition, drugs such as these are hydrophilic; thus, with reversed-phase columns (C8 or C18), low concentrations (<10%) of organic solvent or ion-pairing agents must be used to provide adequate retention of the solutes [8]. However, it has been reported that when the concentration of the aqueous solvent in the mobile phase is increased, the ESI-MS signal is decreased [9], [10], [11]. Therefore, because of the poor peak shape and inefficient desolvation in the ESI process due to high concentrations of aqueous solvent when C18 columns are used, the overall MS response is highly reduced.

Ion suppression in the ESI interface has recently received much attention [12], [13], [14], [15]. Ion suppression can occur when endogenous interferences are co-eluted with the analytes. These endogenous compounds can diminish the ionization of the analyte in the ESI interface. The final result is a reduced and imprecise MS signal. Improvements in the sample preparation [12] or the HPLC separation [14] or a combination of both can reduce ion suppression. It has been reported that capacity factors (k′) greater than 4 were necessary to separate analytes from endogenous interferences and thus significantly decrease ion suppression [16]. Our goal is, therefore, to find HPLC stationary phases that provide good peak shape and good retention (k′>4) for the HPLC/ESI/MS analysis of basic drugs using a mobile phase with high concentrations of organic solvent. We have investigated a number of stationary phases and found that cyanopropyl (CN) and pentafluorophenylpropyl (PFPP) stationary phases provided good peak shape and good retention for basic drugs with the use of 90% acetonitrile in the mobile phase [17], [18]. By retaining the drugs with a high concentration of acetonitrile in the mobile phase, the CN and PFPP stationary phases provided signal enhancements greater than a factor of 10 when compared with a C18 stationary phase. We found other hydrophobic phases such as C8 and C4 had retention characteristics similar to C18 phases. However, only two classes of basic drugs were tested and it was not known whether the analyses obtained could be universally applied to all basic drugs. Thus, we report on the results of the use of a CN and PFPP stationary phase for the HPLC/ESI/MS analysis of a variety of basic drugs. Fig. 1 shows the structures of the CN and PFPP stationary phases.

Section snippets

Reagents and standards

The drug, gepirone was obtained from Mr Bob Behme of Scientific Resources, Inc. (Evansville, IN). Sumatriptan, bufuralol and hydroxy-bufuralol were obtained from Ms Jessica Dunn of Pfizer Inc. (Groton, CT). All other compounds were obtained from Sigma Chemical (St Louis, MO). Standard stock solutions (1.0 mg/ml) were prepared by dissolving a weighed amount of the compounds in H2O/MeOH (90:10 v/v%). The solutions were sonicated in an Ultrasonicating Bath 3200 (Bransonic, Danbury, CT) for 10 min.

Results and discussion

Mass spectral, drug class and pKa information for the solutes are shown in Table 1. All the solutes formed predominant protonated molecules [M+H]+ in the ESI source.

To demonstrate the broad applicability of the CN and PFPP stationary phases for the HPLC/ESI/MS analysis of basic compounds, we chose basic drugs with a wide-range of characteristics. The basic drugs ranged in polarities, had a pKa range from 7.2 to 10.7 and ranged in molecular weight from 227.7 to 385.5 Daltons. In addition, we

Conclusions

The CN and PFPP phases offer significant advantages when compared with C18 or C8 phases for the HPLC/ESI/MS of basic drugs. By retaining polar, basic drugs with 90% acetonitrile in the mobile phase, the ESI-MS signal is enhanced by over a factor of 9 when compared with a C18 phase which required 6.0% acetonitrile for the analysis of a model solute, codeine. The improvement in signal is due to more efficient desolvation in the ESI source with the use of higher concentrations of acetonitrile.

Acknowledgements

The authors thank Mr Keith Duff of Restek Corporation for the synthesis of the stationary phases and his insightful discussions. The authors would also like to thank Mr Bob Behme of Scientific Resources, Inc. for his donation of the drug, gepirone and Ms Jessica Dunn of Pfizer Inc for the use of the standards of sumatriptan, bufuralol and hydroxy-bufuralol.

References (21)

  • D McCalley et al.

    J. Chromatogr. A.

    (1998)
  • A Argekar et al.

    J. Pharm. Biomed. Anal.

    (1999)
  • S Tanase et al.

    J. Chromatogr. B Biomed. Sci. Appl.

    (1998)
  • F Kuhlmann et al.

    J. Am. Soc. Mass Spectrom.

    (1995)
  • S Needham et al.

    J. Chromatogr. B Biomed. Sci. Appl.

    (1998)
  • G Schanzle et al.

    J. Chromatogr. B Biomed. Sci. Appl.

    (1999)
  • K Otsuka et al.

    J. Chromatogr. A

    (1998)
  • D Buhrman et al.

    J. Am. Soc. Mass Spectrom.

    (1996)
  • R Wieboldt et al.

    J. Chromatogr. B Biomed. Sci. Appl.

    (1998)
  • I Fu et al.

    J. Pharm. Biomed. Anal.

    (1998)
There are more references available in the full text version of this article.

Cited by (31)

  • Development of a rapid method for the simultaneous separation and determination of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol and its N- and O-glucuronides in human urine by liquid chromatography-tandem mass spectrometry

    2013, Analytica Chimica Acta
    Citation Excerpt :

    These endogenous compounds can diminish the ionization of the analytes in the ESI interface. Improvement in the sample preparation or the LC separation or a combination of both can reduce ion suppression [26]. Moreover, in the MS/MS mode, the major product ions of NNAL-N-Gluc and NNAL-O-Gluc were all obtained by the fragmentation of the [M+H]+ ion m/z 386.0.

  • Synthesis, characterisation and chromatographic evaluation of pentafluorophenyl and phenyl bonded silica phases prepared using supercritical carbon dioxide as a reaction solvent

    2013, Journal of Chromatography A
    Citation Excerpt :

    Zhang [32] suggested that fluorinated phases could show different separation mechanisms depending on the type of analytes. “U shape” retention profile for polar and basic analytes with increased amount of organic modifier in the mobile phase [40,46,47] have been observed on PFPP. The “U-shape” behaviour has also been observed on non-fluorinated phases, e.g. polar-embedded stationary phases [48,49].

  • Optimization of a high-performance liquid chromatography method for the analysis of complex polyphenol mixtures and application for sainfoin extracts (Onobrychis viciifolia)

    2010, Journal of Chromatography A
    Citation Excerpt :

    The combination of reversed- and normal-phase behavior forms a “U-shape” relationship between retention and organic modifier percentage and can be rationalized by the presence of hydrophobically assisted ion-exchange mechanism or additional independent interactions due to the presence of the pentafluorophenyl ligands [20–22]. Needham et al. found that the pentafluorophenylpropyl modified silica columns gave good retention of several kinds of basic drugs with a mobile phase containing 90% acetonitrile, whereas, to achieve good retention on C18 columns, 40% acetonitrile has to be used [23–25]. In a similar study from Marín and Barbas, both reversed-phase and normal-phase-like characteristics for certain analytes have been observed [17].

View all citing articles on Scopus
View full text