Asms2000_proceedings.pdf

Fast-gradient Microbore Column-Switching LC-MS/MS for the Quantitative G. Hopfgartner, C. Husser, M. Zell, Department of Non-Clinical Drug Safety,F. Hoffmann-La Roche Ltd, Pharmaceuticals Division, CH-4070 Basel, SwitzerlandProceedings 48th ASMS Conference on Mass Spectrometry and Allied Topics, Long Beach, CA, USA, 2000 Introduction
The examination of pharmacokinetic properties of drugs in discovery and development require fast,sensitive and accurate assays for simultaneous determination of these compounds and their metabolitesin biological fluids. In recent years, HPLC combined with mass spectrometric detection (LC-MS/MS)played a key role in accelerating the sample throughput and providing outstanding sensitivity andselectivity. However, high-speed analysis with narrow-bore columns resulted in a substantial loss insensitivity due to dilution effects. This phenomenon is caused by the ion spray mass spectrometerbecause its response is proportional to the concentration of a compound (amount per peak volume). Onthe other hand, this feature allows tuning of sensitivity by using microbore or packed capillary columnsinstead of narrow-bore HPLC columns. The goal of the poster is to demonstrate the speed of analysis withmicrobore and capillary columns rather than to exploit the sensitivity to its full potential. The increasingtendency to accommodate this analytical technique to the analysis of micro samples demands a furtherminiaturization of the analytical system. The design of new HPLC pumps capable of providing accurategradients in splitless mode at flow rates down to 5 µl/min opens up a new field for packed capillary HPLCcolumns. This new gradient capability is particularly useful for multiple component determination (cocktailsor metabolites) to obtain good sensitivity and selectivity as well as short chromatographic run times. Theloading of samples onto the capillary column might be one of the reasons for the somewhat reluctant useof this technique for quantitative analysis. This supposed shortcoming could be nicely circumvented ifcolumn-switching HPLC was used.
Experimental
The column-switching system consisted of a microbore trapping column (Inertsil ODS-3, 5 µm, 0.8 mm i.d.
* 5 mm) and a capillary column (PepMap C18, 3 µm, 300 µm i.d.). The columns were connected by amicrobore switching valve H.S. 7000E from Valco (Prolab Instruments GmbH, CH-4153 Reinach,Switzerland). After precipitation of the plasma proteins of a 50-µl plasma sample with perchloric acid, a 50-µl supernatant aliquot (HTS PAL autosampler from CTC Analytics AG, CH-4222 Zwingen, Switzerland)was directly loaded onto the trapping column at a flow rate of 0.2 ml/min in 1.5 min. The retained analyteswere then transferred onto the capillary column at a flow rate of 20 µl/min for 1 min using gradient elutionin the backflush mode. The binary eluent was supplied by a high pressure micro pump ‘µHPLC-CECSystem Evolution 200’ (Prolab) capable of providing fast gradients in the range 5 to 100 µl/min withoutsplitting due to its novel design. The eluent was 5 mM ammonium formate/methanol/acetonitrile. Thelinear gradient was raised from 10 % to 90 % acetonitrile/methanol in 1 min. The analytes were elutedwithin 2 min and were detected in the positive ion mode with SRM utilizing ion spray tandem massspectrometry (PE Sciex API IIIplus (Concord, ON, Canada). The run cycle time was 3 min.
A comparison was made between two different drug assays using column-switching on-line solid phase
extraction combined with fast gradient elution on microbore and capillary columns. The more classical
approach using 2 mm i.d. trapping columns coupled by a 10-port switching valve to a 1 mm i.d. analytical
column provided a run time of 3 min (Fig.1). The N-oxide metabolite (IV), which was prone to form the
secondary amine metabolite (II) caused by chemical decomposition, could be separated from this
compound (II) on the analytical column (YMC AQ, 3 µm, 1.0 mm id * 30 mm). Faster analysis times could
not be achieved without compromising the resolution between these analytes. The quantification limit was
0.1 ng/ml using a 100-µl plasma aliquot. Owing to the higher flow rate of the 1.0 mm i.d. column (100
µl/min) compared to that of the capillary column, a four-fold bigger volume could be loaded compared to
the capillary column enhancing the sensitivity. The accuracy of the assay ranged from 88.4 % to 107 % for
all analytes while the precision was in the range 1.1 % to 12 % (Table 1).
Fig. 1 SRM chromatograms of compound I and its
metabolites using dual column-switching HPLC
SRM chromatograms of drug I and its metabolites using fast gradient elution with capillary HPLC in the splitless mode Plasma sample taken 20 h after administration Calibration sample of 2.5 ng/ml using a 100-µl aliquot of human plasma from 10 % to 90 % organic solvent in 60 s Precision and accuracy with microbore dual column-switching HPLC Precision and accuracy with capillary column-switching HPLC using using splitless gradient elution and IonSpray SRM/MS splitless gradient elution and IonSpray SRM/MS Metabolite (II)
Metabolite (III)
Metabolite (IV)
Metabolite (II)
Metabolite (III)
The column-switching method utilizing a microbore trapping column for SPE and a capillary HPLC column
for separation of the analytes provided a similar chromatographic performance (Fig 2) as the
aforementioned assay method using a 1 mm i.d. analytical column. Besides other aspects, this finding
also underscores the accuracy of the fast gradient elution when using a flow rate as low as 20 µl/min. It
can also be derived that a shorter run time might be achieved with the capillary method by using the dual
rather than the single trapping column-switching approach. The quantification limit was found to be 0.1
ng/ml using a 50-µl plasma aliquot. In order to achieve a short chromatographic run time, only a 50-µl
supernatant aliquot was injected onto the microbore trapping column. This corresponded only to 20 % of
the supernatant. Provided sensitivity was the main focus, the injection of a 250-µl volume would increase
the sensitivity correspondingly, but at the expense of a much longer run time. As can be derived from
Table 2, the accuracy and precision is very similar to that of Table 1 underlining the performance of this
method.
Conclusion
Capillary HPLC appears to be sufficiently mature for fast multiple component determination of drugs andtheir metabolites from biological fluids using ion spray SRM/MS. The main advantage of capillary HPLC,compared to microbore or narrow-bore columns, is its capability to provide excellent sensitivity with microsamples. Fast analysis with capillary HPLC using gradient elution is feasible, but there is bound to be atrade-off between analysis speed and sensitivity. The advent of novel pumping systems allows fast andreproducible gradient elution at flow rates compatible with capillary columns (300 µm i.d.). The precisionand accuracy of the drug assay with on-line column-switching SPE and capillary HPLC is comparable tothat of microbore columns. The main restriction for achieving run times below 1 min was the occurrence ofmatrix effects compromising accuracy and precision of the assay. Surprisingly, capillary HPLCdemonstrated robustness. More than 20 ml plasma equivalents could be injected before thechromatographic performance started to degrade.

Source: http://www.prolab.ch/PDFs/ASMS2000.pdf