Mass spectrometry plays currently a major role in the qualitative and quantitative analysis of low molecular weight compounds and macromolecules in life sciences. Quantitation of pharmaceutical compounds, their metabolites and endogenous metabolites in biological matrices, such as plasma and urine, is nowadays mostly done with liquid chromatography coupled with atmospheric pressure tandem mass spectrometry (LC-MS/MS) [79]. Gas chromatography coupled with electron impact ionization mass spectrometry (GC-MS) remains an important analytical tool in forensic sciences, doping control and toxicology. For this purpose quadrupole or ion trap mass analyzers are typically used. In contrast, triple quad-rupole instruments have become more the working horse for quantitative pharmaceutical bioanalysis. While quantitative analysis is already well established, many of the new developments in the field of mass spectrometry will contribute to improve metabolites identification, metabolomics and proteomics analysis. Automated computerized data handling (bioinformatics) has become mandatory to cope with the large amount of data generated by the various systems. Mass spectrometers are, from a software point of view, becoming more user friendly while the expanding analysis capabilities of hybrid systems may require more fundamental user training. Due to the enhanced scan possibilities of MS, data dependent acquisition (DDA) has become state of the art for qualitative analysis. A DDA experiment includes a survey scan, a dependent scan and a selection criterion. Typically a survey scan is a full-scan MS and the dependent scan is a MS/ MS scan. The selection criterion requires to record a MS/MS spectrum of the most abundant ion in the survey scan which is above a certain threshold and taking into account the inclusion of ions of interest and exclusion of background ions.

One critical feature of mass spectrometry when combined with chromato-graphic or electrophoretic separation techniques remains the duty cycle of the mass analyzer. A conventional LC chromatographic peak lasts about 10 s, which is sufficient to perform various MS and MS/MS experiments on various types of instruments. In the case of fast LC, the peak width can be in the range 1-2 s which is too fast for most mass analyzers except for TOF mass spectrometers.

42 | 1 Mass Spectrometry in Bioanalysis - Methods, Principles and Instrumentation 1.6.2

Quantitative Analysis in Biological Matrices

Due to its high selectivity and sensitivity LC-MS with quadrupole mass analyzers has almost completely replaced traditional UV detection in many bioanalytical laboratories. ESI, APCI and APPI have become the ionization techniques of choice, covering a large variety of analytes. One limitation with API techniques is that the ionization response factor is compound-dependent and thus requires the use of an internal standard. Isotopically labeled (2H or 13C) internal standards have become very popular because they are capable of compensating for losses during sample preparation, HPLC and ion evaporation due to co-elution with the analyte. In the early days of LC-MS, analysis was mostly performed on QqQ instruments. Quantitative LC-MS analysis can also be performed on single quadrupole instruments, in particular when the Mr of the analyte is higher than 400 and when the limit of quantification is not below the ng ml-1 level. Figure 1.36 shows the total ion current (TIC) chromatogram of the LC-MS analysis of a cyclo-hexanediol derivative analyzed in human plasma after liquid-liquid extraction. It demonstrates clearly the selectivity of triple quadrupole compared to single quadrupole MS. Because this analyte does not have an appropriate chromophore, UV detection would not have been suitable. In contrast to GC, LC is not a high resolution separation technique and co-elution with endogenous compounds may require longer analysis time or improved sample preparation.

Fig. 1.36 Comparison of the LC-MS and LC-MS/MS analysis of a cyclohexanediol derivative in human plasma. (A) Selected ion monitoring mode m/z 443. (B) Selected reaction monitoring m/z 443 ! m/z 373. Ions were detected in the negative mode.

Fig. 1.36 Comparison of the LC-MS and LC-MS/MS analysis of a cyclohexanediol derivative in human plasma. (A) Selected ion monitoring mode m/z 443. (B) Selected reaction monitoring m/z 443 ! m/z 373. Ions were detected in the negative mode.

An important issue with quantitative LC-MS analysis concerns the matrix effects which need to be addressed during method development and validation. Matrix effects are caused by the co-elution of endogenous analytes which either enhance or suppress the analyte signal [80]. The major concern is that matrix effects are sample-dependent and may vary from one sample to another. It is also believed that ESI is more prone to matrix effects than APCI. Various approaches were devised and applied to investigate matrix effects. However, adequate sample preparation and selection of an appropriate internal standard generally provide the key to success. For multicomponent assays it is also important to use the internal standards most appropriate for the respective analyte. Offline and online solid phase extraction, column switching and automated liquid-liquid extractions are the most used sample preparation techniques. Online SPE combined with column switching are particularly attractive because they allow direct analysis of plasma in an automated and high throughput setup. With the high sensitivity of modern triple quadrupole instruments, protein precipitation of plasma in 96-well plate format followed by dilution and direct injection of the eluent has also become a viable approach. Shortterm matrix effects due to different samples may be relatively simple to monitor while longterm matrix effects are very difficult to monitor. Table 1.5 shows the calibration and quality control (QC) results obtained in human plasma of a cyclohexanediol derivative analyzed by LC-MS/MS. At a first glance the calibration seems to be very good. However, when the 10 ng mL_1 calibration sample is reanalyzed (n = 35) and declared as a quality control sample the accuracy becomes disastrous.

The explanation of this result is illustrated in Fig. 1.37, which shows selected reaction-monitoring traces of the sample at 10 ng mL_1. It becomes obvious that the response ratio between the analyte and the IS has dramatically changed. On one side there is enhancement of the analyte's response and on the other side suppression of the internal standard (IS) signal. These effects are mainly caused

Table 1.5 Calibration and QC data for a cyclohexanediol derivative in human plasma.

Sample n spiked ng/ml found ng/ml Accuracy %

00 Plasma

Was this article helpful?

0 0

Post a comment