The characterization of the proteins adsorbed, or immobilized, on substrates by TOF-SIMS has contributed to the development of other analyti-
cal techniques. In 1986, the isolation of an apolipoprotein was studied with TOF-SIMS (Jabs et al. 1986), and an accurate evaluation of the molecular weight of peptides was obtained. Moreover, it became evident at that time that it was feasible using this method to monitor on a microscale basis the genetic polymorphisms of proteins and posttranslational modifications.
Kröger et al. (1998) reported the surface analysis of an optical immunosensor using TOF-SIMS; however, in this research, the role of TOF-SIMS was in fact subsidiary and quite limited, targeted to the mere finding of a product of a reaction. There have really been very few repots on the surface analysis of biomaterials carried out mainly by TOF-SIMS until the twenty-first century. Although TOF-SIMS is very useful for simple chemical analysis with x-ray photoelectron spectroscopy (XPS; Léonard et al. 2001; Lhoest et al. 1998; Wagner et al. 2003a), it has recently been shown that TOF-SIMS is in fact the most useful tool for characterizing complicated samples, including samples comprising of biological molecules.
In order to analyze the interactions between artificial surfaces and proteins, a spectral interpretation protocol is required. There are numerous possible fragment ions from proteins out of the various combinations of the constituent 20 amino acids. Since it is difficult to estimate every combination, homopolymers of 16 amino acids were examined with static SIMS by Mantus et al. (1993). The results were used to establish the first step of a spectral interpretation protocol for protein surfaces. Both positive and negative ion mass spectra were estimated by Mantus et al., and negative ion mass spectra were found to afford much less information than the mass spectra of the positive ions. Wagner and Castner (2001) reported the same result, confirming that positive ion mass spectra are the most suitable for protein analysis with SIMS. Subsequently, plasma proteins adsorbed to titanium substrate were analyzed with static SIMS based on this protocol. Moreover, the adsorption of fibronectin was investigated using TOF-SIMS based on this protocol (for proteins) in comparison with two other techniques, radiolabeling and XPS (Lhoest et al. 1998). As a result, it was found that the TOF-SIMS peaks characteristic of proteins exhibit differences in reduced intensity between the two substrates. It was suggested that these differences, which are not detectable by XPS, are attributable to different orientations and/or conformations of the proteins.
Investigation of the SIMS spectra for amino acid homopolymers can provide helpful information, but detailed studies of protein spectra have not been forthcoming because of the complexity of the mass spectra involved. Therefore, multivariate analysis techniques such as PCA have been applied to analyze the SIMS spectra. PCA reduces a large set of correlated variables, such as peak intensities in a mass spectrum, to a smaller number of uncor-related variables referred to as PCs, as described above. Lhoest et al. (2001) reported characterization of adsorbed proteins such as bovine serum albu min (BSA), bovine plasma fibrinogen, bovine plasma fibronectin (Fn), and chicken egg white lysozyme, on mica and polytetrafluoroethylene (PTFE) substrates using TOF-SIMS with PCA. According to this research, PCA was implemented to enable the classification of several reference, positive-ion protein spectra according to protein and substrate type, and the combination of TOF-SIMS with PCA indeed did enable such a classification. The combination proved capable of classifying proteins by type in cases of pure protein spectra based on the score plots for each sample, and the relative surface concentration in the case of the binary protein spectra for BSA and Fn was based on a ratio of the peak intensities. Since BSA and Fn have large differences in their respective amino acid compositions, it is relatively easy for PCA to differentiate between the two proteins on the surface. The relative concentrations of BSA and Fn can be tracked using the ratio of the intensities of certain peaks, some ratios are more prevalent in one protein than in the other; for example, the ratios of threonine to glutamic acid or phenylalanine to tryptophan (BSA: Glu 10.12%, Phe 4.63%, Thr 5.66%, and Trp 0.34%, Fn: Glu 5.96%, Phe 2.03%, Thr 10.68%, and Trp 1.68%). The position of each peak in the loading plot represents its influence on the score plot, and supported the amino acid composition of BSA and Fn.
A variety of proteins on mica and PTFE substrates have been evaluated by means of TOF-SIMS and multivariate analysis, for example the PCA and LDA carried out by Wagner and Castner (2001, 2003) and Wagner et al. (2002, 2003a,b). LDA enhanced both the discrimination between groups and classification of unknowns because of its supervised nature. The TOF-SIMS spectra for albumin, collagen, cytochrome C, fibrinogen, fibronectin, hemoglobin, lactoferin, lysozyme, myoglobin, and transferrin films adsorbed onto PTFE were classified by PCA and LDA (Wagner and Castner 2001). Furthermore, it has been shown that qualitative information is obtainable for complex plasma and serum protein films using the multivariate analysis results from single-protein TOF-SIMS spectra. Quantitative results are also obtainable for binary adsorbed protein films, although such results depend on the substrate chemistry and morphology (Wagner et al. 2003b).
The TOF-SIMS measurement of proteins on metallic substrates and PTFE is an easier application of TOF-SIMS to biomaterials (Aoyagi et al. 2004b; Davies et al. 1996; Heard et al. 2001; Poleunis et al. 2002; Pradier et al. 2002; Wagner and Castner 2004), because ions from metal substrates and polymer substrates free of nitrogen, such as PTFE, can be distinguished from protein fragment ions quite easily. In addition, the latest research has demonstrated that cluster primary ions, such as the gold cluster, the bismuth cluster ions, and C6o+ ions enhance peptide molecular ion yields. Tempez et al. (2004) reported that there are advantages of using large polyatomic gold clusters, Au400+, as the primary ion: (1) enhanced secondary ion yield of the intact molecule, (2) significantly improved S/N ratio, and (3) reduced fragmentation. Similar advantages are also indicated in the case of using smaller cluster ions such as Au3+. Examples with the cluster primary ions will be described here later.
Was this article helpful?