LC-MALDI

In direct insertion techniques, reproducibility is the main obstacle in developing a reliable analytical technique. One of the many variables to take into account is sample shape. A compact sample with minimal surface area is ideal [64]. Direct mass-spectrometric characterisation in the direct insertion probe is not very quantitative, and, even under optimised conditions, mass discrimination in the analysis of polydisperse polymers and specific oligomer discrimination may occur. For nonvolatile additives that do not evaporate up to 350 °C, direct quantitative analysis by thermal desorption is not possible (e.g. Hostanox 03, MW 794). Good quantitation is also prevented by contamination of the ion source by pyrolysis products of the polymeric matrix. For polymer-based calibration standards, the homogeneity of the samples is of great importance. Hyphenated techniques such as LC-ESI-ToFMS and LC-MALDI-ToFMS have been developed for polymer analyses in which the reliable quantitative features of LC are combined with the identification power and structure analysis of MS. 

Many excellent reviews on the development, instrumentation and applications of LC-MS can be found in the literature [560-563]. Niessen [440] has recently reviewed interface technology and application of mass analysers in LC-MS. Column selection and operating conditions for LC-MS have been reviewed [564]. A guide to LC-MS has recently appeared [565]. Voress [535] has described electrospray instrumentation, Niessen [562] reviewed API, and others [566,567] have reviewed LC-PB-MS. For thermospray ionisation in MS, see refs [568,569]. Nielen and Buytenhuys [570] have discussed the potentials of LC-ESI-ToFMS and LC-MALDI-ToFMS. Miniaturisation (reduction of column i.d.) in LC-MS was recently critically evaluated [571]. LC-MS/MS was also reviewed [572]. Various books on LC-MS have appeared [164,433,434,573-575], some dealing specifically with selected ionisation modes, such as CF-FAB-MS [576] or API-MS [577],... 

Chen, H.S., Rejtar, T., Andreev, V., Moskovets, E., Karger, B.L. (2005). High-speed, high-resolution monolithic capillary LC-MALDI MS using an off-line continuous deposition interface for proteomic analysis. Anal. Chem. 77, 2323-2331. 

Reference 5. With the advent of commercially available MALDI-TOF-TOF instruments the combination of off-line one- or two-dimensional LC-MALDI-MS/MS has become a popular alternative or rather a complement to LC-ESI-MS/MS in the proteomics community. 

Hofmann, S., Gluckmaim, M., Kausche, S., Schmidt, A., Corvey, C., Lichtenfels, R., Huber, C., Albrecht, C., Karas, M., and Herr, W., Rapid and sensitive identification of major histocompatibility complex class 1-associated tumor peptides by nano-LC MALDI MS/MS, Molecular and Cellular Proteomics 4(12), 1888-1897, 2005. 

Wu WW, Wang G, Baek SJ et al. Comparative study of three proteomic quantitative methods, DIGE, cICAT, and iTRAQ, using 2D gel- or LC-MALDI TOF/TOF. J Proteome Res 2006 5 651-658. 

G. Lochnit and R. Geyer, An optimized protocol for nano-LC-MALDI-TOF-MS coupling for the analysis of proteolytic digests of glycoproteins, Biomed. Chromatogr., 18 (2004) 841-848. 

The speed of MALDI analysis depends on the laser pulse rate. With the recent introduction of 200-Hz lasers, samples can be analyzed ten times faster than before. This development is especially advantageous for offline liquid chromatography (LC)/MALDI applications (Ericson et al., 2003). MALDI mass analysis is performed considerably faster than LC separation, allowing for chromatographic... 

The complementarity of all the data obtained in this strategy (LC-MALDI-MS and nanoLC-MS/MS) allowed to increase significantly the coverage percentage for all the gel slices. Figure 3 summarizes this strategy and technical details are presented below. 

Figure 3. LC-MALDI strategy. Fractionation of the peptide mixture and MALDI-MS analysis.

LC-MALDI-MS experiments are also reported. The combination of both ESI and MALDI experiments shows well the advantage in term of coverage of this double ionization process. Indeed, some peptides are identified specifically from the ESI ionization process and other from the MALDI process, increasing the whole number of identified peptides. 

Moreover, we have determined the false positive rate for this approach. Many tryptic peptides originated from different proteins can be attributed to a single mass (e.g. HQHPLQCVMEK 1364.63 Da and EADFINCVIWR 1364.65 Da AM < 20 ppm). Thereby false positive identification may occur. To evaluate the false positive rate, we have selected three common proteins which were not identified during the nanoLC-MS/MS analysis. These proteins (tubulin, actin and myosin) were digested in-silico, and the generated mass lists were compared to the LC-MALDI-MS peaklist. A total of only five masses were attributed to the three... 

Table 1. Sununary of the proteins identified by nanoLC-MS/MS from a single 1D gel slice, and the increase of the coverage using LC-MALDI-MS strategy...

To conclude, we have developed a strategy which combines LC-MALDI-MS and nanoLC-MS/MS in order to identify proteins originating from ID gel with a high coverage compatible with plasma membrane study (CD98, CD71, CD44). 

In the off-line approach, analyte and matrix depositiorr are decoupled from MS analysis and only instrument-specific plate formats have to be considered in interface design. In this case interface design focuses on the development of techniques for a perfect transfer of LC-effluent and matrix to the surface from which laser desorption takes place at a later stage. The development of suitable techniques started very early in conjunction with capillary electrophoresis (CE) and some of the methods which are applied nowadays for low-flow LC-MALDI are included in the following compilation. 

Most of the currently applied LC-MALDI interfaces use off-line spot deposition and a number of commercial instruments are available (Mukhopadhyay 2005). Compared to on-line techniques, this interface type is technically less demanding but only careful adjustment guarantees optimal performance. 

To achieve the best performance in protein identification or quantitation, the extent of protein or peptide separation should match the capabilities of the technique applied in the identification or quantitation step. With LC-MALDI MS and MS/MS as analysis technique, the number of good-quality MS/MS spectra, which can be acquired from one sample spot (LC-fraction) represents one limitation the number of components in one fraction should therefore not exceed this maximum. By increasing matrix concentration more laser shots could be acquired from one spot, but analyte concentration in the crystals and detection sensitivity would... 

Figure 4. Standard LC-MALDI setup with sample concentration/desalting step, nanocolumn LC separation and off-line deposition of LC-eluent after admixture of matrix solution.

Compared to yeast with about 5800 ORFs which code for proteins, samples from higher organisms can be much more complex and the separation scheme has to be adjusted accordingly. For very simple species as the other extreme, already subcellular fractionation can provide the appropriate pre-separation. For example only one LC-MALDI MS/MS run of a peptide mixture derived from the separated E-coli 50S ribosomal subunit allowed to identify 30 of the 33 expected proteins (Mirgorodskaya et al. 2005a). 

Figure 6. Chromatographic resolution, MS- and MS/MS sensitivity achieved with the LC-MALDI setup described in Fig. 4. Trace of m/z 927 ([M + H]+ of peptide 137-143) from 10 fmol of a tryptic BSA hydrolysate using 12 sec wide fractions and a matrix concentration of 3 mg/mL (A). MALDI MS (B) and MS/MS spectrum (C) of the fraction with the highest intensity of m/z 927 (fraction 79).

To cope with the high number of RP-HPLC runs required to analyze fractions in the two-dimensional LC-MS approach, multiplexed LC-separation and spotting for LC-MALDI analyses was applied very early. In quantitative analysis of differential protein expression by ICAT technology, total analysis time could be significantly reduced with four parallel separation lines (Lee et al. 2002). Prerequisite is of course... 

---