Effect of metal salts on UV-MALDI-MS and ESI-MS carbohydrate analysis

M. KANIZ FATEMA; SILVANA L. GIUDICESSI; ROSA ERRA-BALSELLS; HIROSHI NONAMI

Matsuyama, Japon.

Congreso; The annual meeting of the Japanese Society of Agricultural, Biological and Environmental Engineers and Scientists; 2008

Environmental Control in Biology

UV-MALDI- and ESI-MS are almost rutine techniques for the determination of high molecular mass when they are used with proteins while for carbohydrates still the techniques faces serious experimental problems. This limitation is probably related, at least in part, to the structural complexity and to the molecular interactions often found in the oligosaccharide moieties of these biopolymers. Proteins are detected as the species [M+H]+ while carbohydrates are in general detected as cationized species [M+Cat]cat+. Taking into account that the isolated oligosaccharides include in its structure salts naturally present in cells and tissues, they can modify the results of the MS analysis. Furthermore, salt impurities present in the sample originated also from the matrix (salts present as impurities in commercial compounds), the analyte, solvents, glassware, surface of sample carriers, etc. may result in alkali metal adducts appearing (or not) in the UV-MALDI-MS. This is sometimes undesirable and can complicate the mass spectra. However, these adducts are necessary because some species such as synthetic polymers and carbohydrates are primarily ionized by the formation of metal ion adducts (cationization). For this reason selected metal salts are often added to samples in order to enhance the analyte signals. Furthermore, there are several question to discuss not only related with the stability of the species [M+Cat]cat+ in gas state, but also with the fact that cationization might be in some way matrix dependent too: (a) chemical matrix effect on the formation of the species [M+Cat]cat+ and (b) matrix provision of the enough energy required for desorption/ionization of the species [M+Cat]cat+. Understanding cationization mechanisms and the combined salt-matrix effects on it, are obviously required for better control of ion formation in carbohydrate UV-MALDI process. UV-MALDI-TOF-MS measurements were performed with: (i) a Shimadzu Kratos, Kompact MALDI 4, laser desorption time-of-flight mass spectrometer, equipped with pulsed nitrogen laser (em=337 nm), tunable PDE and PSD (MS/MS) device; (ii) Ultraflex II TOF/TOF mass spectrometer equipped with a high performance solid state laser (λ=355 nm), reflector (Bruker Daltonics GmbsH, Bremen, Germany), tuneable pulsed-delayed extraction time (PDE) and MS/MS (Lift) and CID (Ar). All mass spectra were taken in positive and negative ion modes, in the linear and reflectron modes. Stock matrix solutions (9H-pirido[3,4-b]indole, nHo; 2,5-dihidroxybenzoic acid, GA; 2,4,6-trihydroxyacetophenone, THAP) 1 mg/0.5 mL and analyte solutions 0.05 mg/0.025 mL in the appropriate solvent were prepared. CNTs (10-20 nm) were purified by successive rinsing-centrifuging with acetone and water/methanol, following by suspending in water/methanol 1:9 (v/v). Salt solutions were prepared in water and then mixtures carbohydrate:salt 1:1 to 1:100 (m/m) were prepared. For MALDI-MS two different sample preparation methods were used: Sandwich method, 0.5 *L of the matrix solution was placed on the sample probe tip and the solvent was removed by blowing air. Subsequently, 0.5 *L of the analyte solution was placed on the same probe tip covering the matrix. Then, two additional portions (0.5 *L) of the matrix solution were deposited on the same sample probe tip. Mixture method, mixtures 10/1 to 1/1 (v/v) of the corresponding matrix and analyte solutions were prepared and transferred (0.5 *L x2) on the probe tip. The ESI-TOF mass spectra were acquired by directly infusing an aqueous analyte solution into the ESI ion source of the Mariner ESI-TOF mass spectrometer (Applied Biosystems, USA), equipped with a loop (0.5-50 *L) for direct micro injection and a Harvard PHD 2000 syringe infusion pump at a flow-rate of 5 *L/min. MeOH-H2O 9:1 was used as solvent stream. Spectra were recorded in a m/z range comprised between 100 and 4000 Da, in positive-ion mode. Analyte solutions were prepared as 1.0 mg/100 L H2O. Experiments were conducted in each case with the above concentrations and with 1/10 dilution. We have recently described the analysis of neutral polysaccharides, a family of xylans, that could only be efficiently desorption/ionizated by using nHo as UV-MALDI matrix. It was quite surprisingly that although by adding extra NaCl to the sample the spectra were quite similar but after addition of CaCl2 practically all the signals of the oligomers disappeared. At first sight a problem in the formation and stability of the analyte adducts [M+CaCl]+ / [M+Ca]++ could be the cause. Thus, in order to shed some light to this problem a comparative study of the UV-MALDI-MS was conducted with a family of neutral cyclic and acyclic sugar as well as with more complex systems as inuline and dextran, using two matrices in comparative way (GA and nHo), using several doping salts (NaCl, KaCl, NH4Cl, CaCl2 and MgCl2) in linear positive and negative ion modes, preparing the samples by using the sandwich and the mixture methods. The critical effect observed on the experiments conducted by using nHo but practically absent when GA was used encourage us to perform additional UV-MALDI experiments by using as matrix carbon nanotubes (CNTs) and THAP as well as to conduct thermal and photophysical studies. Thus, in parallel experiments the thermal stability of the selected doping salts, the used UV-MALDI matrices and carbohydrates, each compound alone and in different binary and ternary mixtures, were studied by classical melting point analysis, thermogravimetric analysis and differential scanning calorimetry as well as the effect of the salts on the UV-vis absorption, fluorescence emission and excitation spectra and acid-base character of matrices in ground and electronic excited states. Although the species [M+CaCl]+ / [M+Ca]++ for some carbohydrates ([M]) had been detected in previous ESI-MS studies, only we were able to detect the species [M+Ca]++ and/or [M+Mg]++ in ESI-TOF-MS studies too. On the contrary, we never detected these species by UV-MALDI-TOF-MS studies of any carbohydrate when added Ca++ and Mg++ salts inhibits almost completely the desorption/ionization of the carbohydrate [M] as [M+Na]+ when nHo and CNTs are used as matrices. Results obtained as well as a model to explain then will be presented and discussed.