The characterization of glycosylation in proteins by mass spectrometry (MS) is often impeded by strong suppression of ionization of glycopeptides in the presence of non-glycosylated peptides. Glycopeptides with a large carbohydrate part and a short peptide backbone are particularly affected by this problem. To meet the goal of generating mass spectra exhibiting glycopeptide coverages as complete as possible, derivatization of glycopeptides offers a practical way to increase their ionization yield. This paper investigated derivatization with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) which is a rapid labeling technique commonly used for fluorescence detection in high-performance liquid chromatography (HPLC) and capillary electrophoresis (CE). As test samples we used peptides and glycopeptides obtained by enzymatic digestion of three different glycoproteins, i.e., human antithrombin, chicken ovalbumin, and bovine alpha1-acid-glycoprotein. It was found that AQC derivatization resulted in strongly increased signal intensities when analyzing small peptides and glycopeptides by matrix-assisted laser desorption/ionization (MALDI)-MS. For these compounds the limit of detection could be reduced to low fmol amounts. Without derivatization only glycopeptides containing large peptide backbones were detected by MALDI-MS. This effect was even significant when glycopeptides were pre-separated and enriched by means of lectin affinity chromatography before MALDI-MS analysis and when using electrospray ionization (ESI). This labeling method, applied in combination with MS detection for the first time, was found to be well suited for the enhancement of detection sensitivity for small glycopeptides in MALDI-MS analysis and thus for reducing the need for pre-separation steps.

In this study, we investigated a novel ionic liquid matrix (ILM), namely, the 1,1,3,3-tetramethylguanidinium salt of 2,4,6-trihydroxyacetophenone (THAP). This matrix[1,1,3,3-tetramethylguanidinium 2,4,6-trihydroxyacetophenone (GTHAP)] turned out to be well suited for the matrix-assisted laser desorption/ionization mass spectrometric analysis of glycopeptides and glycans, and overcame the well-known ionization suppression of carbohydrate structures in the presence of peptides. The matrix was evaluated by two different series of experiments, in each case in comparison with the crystalline THAP matrix. In the first set of experiments, mass spectra were taken from unseparated tryptic digests of three glycoproteins taken as model compounds. Even glycopeptides containing short peptide backbones and large carbohydrate moieties gave high signal intensities when using the ILM though they did not appear in the THAP spectra. In the second set of experiments, the total tryptic digests were treated with endoglycosidase PNGase F to cleave off the N-linked glycans. When using the GTHAP matrix, it was possible to detect the glycans with high intensities in the presence of the tryptic peptides, whereas glycan ionization was completely suppressed when measured with the solid matrix THAP. The extent of metastable decay of glycopeptides was reduced when using the ILM. Altogether, GTHAP proved as a useful ILM particularly being superior to solid matrices in the context of glycosylation analysis.

The performances of several matrices were investigated for the accurate determination of the molecular mass distributions of pullulans by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS). The ionic liquid matrix (ILM) 2,5-dihydroxybenzoic acid butylamine (DHBB) gave better and more reliable results than the crystalline matrices 2,5-dihydroxybenzoic acid (DHB) and 2,4,6-trihydroxyacetophenone (THAP). With the ILM it was possible to obtain spectra of pullulans up to more than 100 kDa, the highest molar mass reported thus far. Owing to the known advantages of liquid matrices providing better spot-to-spot reproducibility, an almost noise-free spectrum and constant baselines were obtained when working under optimized conditions. In particular, the extent of in-source fragmentation occurring with this group of fragile polymers was considerably and decisively reduced. Thus, a more reliable representation of the true oligomer and polymer distributions is experimentally attainable, especially for distributions with small polydispersity values. The maximum error in the measured distribution associated with fragmentation was estimated by model calculations describing the changes in the polymer distribution upon different probabilities of fragmentation events. These simulation results indicated that the data obtained by MALDI-TOFMS using the liquid DHBB matrix were of high reliability. In particular, the average value of the distributions, M(w), and the polydispersity were obtained with predicted uncertainties of between 3 and 15% depending on the width of the distribution and the mass of the polymers.