In addition, integration of the H-1 signal of the glucose moiety at δ 5.44 and the H-3 and/or H-4 signals of the preponderant fructosyl units between δ 4.27 and 4.11 respectively, mean DPs of 8–9 (RFOS) and 7–8 (LFOS) can be proposed, but in both 1HNMR spectra there are signals of minor impurities that reduce the precision of this procedure for DP determination. Other methodologies for DP determination are high-performance chromatography (HPLC), gas chromatography (GC), and principally high-performance anion-exchange http://www.selleckchem.com/products/cx-5461.html chromatography with

pulsed amperometric detection (HPAC-PAD), but the response for fructooligosaccharides (FOS) with HPAC-PAD can vary (Timmermans, van Leeuwen, Tournois, de Wit, & Vliegenthart, 1994) and analysis

often requires considerable sample purification. PD0332991 mw Particularly significant was the analysis of FOS or inulin from some plants, using MALDI-MS (Wang et al., 1999, Štikarovská and Chmelík, 2004, Lasytovicková and Chmelík, 2006 and Arrizon et al., 2010). We therefore applied this technique for determination of the DPs of fractions RFOS and LFOS. The spectrum for each FOS is shown in Fig. 3. RFOS and LFOS ions had a mass difference of 162 Da, which corresponds to fructose/glucose residues. It gave rise to [M + Na]+ and [M + K]+ ions as the main distribution obtained in the +LIN mode. Almost all spectra exhibited monomodal molecular 3-mercaptopyruvate sulfurtransferase mass distributions. Often

food samples, such as onions, shallots, and garlic, naturally contain a high concentration of potassium ions, and could be analysed without further addition of salts (Wang Sporn, & Low, 1999). The molecular ions seen in MALDI-MS for RFOS and LFOS samples were almost entirely the potassium adducts (Fig. 3). Under these conditions, the DP distribution of FOS obtained for MALDI-MS ranged from 5 to 16 for RFOS and 4 to 9 for LFOS. These were consistent with their average-DPs obtained by integrating the 1H signals from NMR spectra. Carbohydrates ionise in a MALDI-MS source, only after cationisation with alkali ions (Börnsen, Mohr, & Widmer, 1995). For simplification of analysis, it is desirable that the carbohydrate sample contains predominantly only one metal ion, resulting in a single molecular ion peak. With no addition of another ion, the matrix and sample contained both sodium and potassium ions (Fig. 3), resulting in multiple ion signals. There have been some sample treatments, as with ion exchange membrane (Börnsen et al., 1995) and by dissolution of carbohydrates in 0.01 M solution of a particular metal salt (Wang Sporn, & Low, 1999), resulting in detection of a single adduct signal. Since the DPs of RFOS and LFOS were between 4 and 16 units and ionic exchange can be readily carried out in the inlet system of ESI-MS equipment, we analysed the FOS samples by this technique (Fig. 4).