A systematic approach to selenium speciation in selenized yeast

Abstract

A systematic approach to the characterization of selenized yeast supplements in terms of the speciation of selenium was developed. The optimized fractionation procedures included the sequential leaching of water soluble, cell-wall bound, and membrane-protein selenium followed by a further fractionation of each extract by high-resolution size-exclusion chromatography. The stability of fractions collected as chromatographic peaks was investigated in the presence of a proteolytic enzyme (pronase XIV) and trypsin in order to discriminate between selenium-containing peptides and other selenocompounds. Reversed-phase HPLC of tryptic digests of size-exclusion chromatographic fractions allowed the identification of selenopeptides by MALDI and electrospray MS. The complexity of the speciation of the water-soluble selenium in yeast was confirmed. Surprisingly, selenomethionine in the water insoluble fraction was found to be bound physically to cell wall constituents rather than being incorporated chemically into the protein structure, in contrast to former studies.

---

Introduction

Selenized yeast has probably been the most widely investigated natural product containing selenium.¹ The interest in these studies was triggered off by the study of Clark et al.,² who indicated a possible role of selenized yeast (one of the most economic sources of organic forms of Se) in cancer prevention.³ During the growth of Saccharomyces cerevisiae yeast, selenite, which is a potentially toxic and poorly bioavailable species, is converted to safer and highly bioactive species with improved nutritional properties.⁴ The knowledge of the identity of these biosynthesized compounds is of paramount importance for the understanding of the mechanisms of selenium incorporation, assuring the batch-to-batch reproducibility of the produced selenized yeast, prevention of fraud and, especially, for the understanding of the beneficial and anticancer activity of selenized yeast.

The literature on the selenium speciation in selenized yeast is fairly abundant and has been reviewed recently.^1,5 The analytical methods used have been based on the coupling of chromatography,^6–10 capillary electrophoresis,¹¹ and 2-D gel electrophoresis (for Se-containing proteins),^12,13 with selenium-specific detection, usually by inductively coupled plasma mass spectrometry (ICP-MS). Electrospray MS, either as an HPLC detector,^14–16 or in infusion mode using triple quadrupole^7,10 and quadrupole-TOF tandem systems,⁶ have been used for the identification of selenocompounds detected with mixed success.

Despite a large number of works, the information available on the identity of the molecules incorporating or binding selenium in selenized yeast is still scarce. Solutions to a number of analytical problems are required. Indeed, most of the to-date characterization studies have focused on the water extract that accounts only for 10–15% of the total selenium in yeast.^{6,7,17–19,23} The results were presented in the form of chromatograms or electropherograms with very limited information regarding the peak purity or species identity, especially concerning polypeptide compounds. The water-insoluble selenium fraction has usually been determined as selenomethionine based on proteolytic (pronase–lipase) digestion procedures.^{14,16,20–22} It was assumed that this selenomethionine was bound to proteins; no ultimate proof for this, however, has ever been provided. Multidimensional chromatographic approaches were recommended to prove the chromatographic purity of a selenium peak and to isolate a selenocompound prior to electrospray MS/MS.¹⁸ The sample preparation procedure by multidimensional chromatography often includes freeze drying and redissolution of the selenium containing fractions.²³ The stability of the different compounds is unsure and some losses of selenium from selenoproteins during purification were reported.²⁴

Solubilization of selenocompounds without destroying their original identity (or with their controlled degradation) is the prerequisite for their further characterization by chromatography combined with MS detectors. In our former work a step of pectinolytic digestion with a mixture of cellulases and hemicellulases was introduced to recover selectively selenium bound to cell walls. However, the chromatographic resolution was far from optimal and the quality of chromatograms did not allow a detailed characterization of the fractions obtained.²⁵

In this paper the above procedure was revisited with high-resolution size-exclusion HPLC followed by a second chromatographic dimension in the form of reversed-phase HPLC. It was preceded, if possible, by controlled degradation of the isolated fractions with trypsin which allowed the preservation of most of the structural information. A deeper insight into the identity of selenocompounds in yeast was possible by verification of the stability of selenium-containing fractions isolated in the presence of different proteolytic enzymes, followed by mass spectrometry of the reaction products.

Experimental

Apparatus

An ICP-MS (Elan 6000, PE-SCIEX, ON, Canada) fitted with a cross-flow nebulizer and double-pass Scott spray chamber was used for the analysis of fractions from size-exclusion chromatography and for on-line SEC-ICP-MS measurements. For on-line RP HPLC-ICP-MS analysis the interface consisted of a cooled low-volume cyclonic spray chamber (Glass Expansion, Romainmotier, Switzerland), a 0.85 mm id alumina torch injector and a set of platinum cones. The use of an auxiliary oxygen flow of 15 mL min⁻¹ allowed the introduction into the plasma of up to 60% methanol at 0.75 mL min⁻¹ with a negligible loss of sensitivity. All the connections were made of PEEK tubing (id 0.17 mm) for the purpose of fraction collection; elution profiles were constructed with Microsoft Excel software using the intensity of ⁸²Se obtained for each fraction. ⁷⁷Se was measured in parallel to verify the absence of polyatomic interferences.

For MALDI-TOFMS a Voyager-DE STR (Applied Biosystems, PE-SCIEX, ON, Canada) was used. The combination of delayed extraction and reflectron mode allows a mass measurement accuracy of 0.01 Da in the lower (<5000 Da) m/z range that is particularly attractive for peptide mapping. BioAnalyst software was used to process data. Electrospray MS experiments were performed using a PE-SCIEX API 300 Ion-spray triple-quadrupole mass spectrometer (Thornhill, ON, Canada).

Semi-preparative size-exclusion chromatography was carried out in a HiLoad 26/60 Superdex 30 Prep (Pharmacia, Uppsala, Sweden) using an AKTA-Prime system (Pharmacia). Reversed-phase HPLC was performed in a C₄ column (250 mm × 4.6 mm id) with 5 µm and 300 Å particle and pore size, respectively (Supelco, Bellefonte, PA, USA) using a 1100 Series pump (Agilent). Lyophilization was carried out using a Model LP3 lyophilizer (Jouan, France). Analytical size-exclusion chromatography was carried out using a Superdex Peptide column (300 × 10 mm × 14 µm) with an exclusion limit of 20 kDa and an optimum separation range between 0.1 and 7 kDa. The size-exclusion columns were calibrated using albumin (67 kDa), metallothionein (7 kDa), Cd complexed with phytochelatin (3.5 kDa) and glutathion (in the reduced form, 0.3 kDa).

Materials and reagents

A commercial preparation of selenized yeast (Alltech, Lexington, KY) was used as the example sample. Chemicals were from Sigma–Aldrich (St. Quentin Fallavier, France) unless stated otherwise. Water from a Milli-Q system (Millipore, Bedford, MA) was used throughout. Solutions of trypsin (1.3 mg mL⁻¹) and dithiothreitol (DTT, 36 mg mL⁻¹) were prepared by dissolving the corresponding reagent in 0.1 M Tris buffer (pH 7.8).

Procedures

The general outline of the analytical procedure used was presented in Fig. 1.

Fig. 1 General outline of the analytical procedure used.

Extraction of selenocompounds. 12 portions of 0.3 g of selenized yeast were each extracted with 10 mL of water containing 0.1 mM phenylmethylsulfonyl fluoride (PMSF) and 0.1% DTT in an ultrasonic bath for 3 h. The extracts were centrifuged at 2000 rpm for 30 min and decanted into a 500 mL bottle. This procedure was repeated three times using fresh leaching solution each time. The residue was submitted to an extraction with 5 ml of 4% (w/v) Driselase solution in 30 mM Tris–HCl (pH 7.5) containing 1 mM PMSF for 18 h. The supernatant was centrifuged and the residue was extracted with 4% (w/v) SDS in 30 mM Tris–HCl (pH 7.5) for 1 h.

Semi-preparative size-exclusion LC. The combined water extracts were lyophilized. A sample of 80 mg was dissolved in 4 mL of water and injected onto the preparative size exclusion column. The elution was carried out with 10 mM ammonium acetate (pH 9.5) at 2 mL min⁻¹. The eluate was collected every 1 min over 3 h. Selenium was selectively determined in each fraction by ICP-MS. The Se containing fractions within chromatographic peaks were pooled and freeze-dried.

Tryptic digestion. The lyophilizate from a SEC fraction was dissolved in 120 µL of 0.1 M Tris buffer (pH 7.8). It was incubated at 37 °C for 1.5 h with 15 µL of the DTT solution to reverse the oxidation, leading to the formation of Se–Se and Se–S bonds and selenoxides, and then with 50 µL of the trypsin solution at 37 °C for a further 15 h. The reaction was stopped by bringing the pH down to 5 by the addition of acetic acid. A 30 µL portion of the digest was analysed by HPLC-ICP-MS.

Pronase digestion. The samples were digested using 2 ml of a 30 mM TRIS-HCl solution (pH 7.5) containing 40 mg of protease and 20 mg of lipase for 16 h. The conditions were the same for the peaks of the aqueous extract and the SDS fraction.

Chromatographic conditions. Selenocompounds were eluted from the analytical size-exclusion column (Superdex Peptide) using 10 mM ammonium acetate at 0.7 ml min⁻¹. The injection volume was 100 µl. For reversed-phase HPLC, a 30 µL-portion of the digest was analysed and elution was carried out at the flow rate of 0.75 ml min⁻¹ in the gradient mode. Phases A and B were 0.1% TFA in water and in methanol, respectively. The gradient conditions used are specified in the figure captions. In the preparative mode, reversed-phase HPLC was repeated, for a 140 µl-injection with fraction collection every 22 s. Each fraction was diluted with 1.5 mL of water and analysed by ICP-MS. The Se containing fractions within chromatographic peaks were pooled and freeze-dried. The fractions were then analysed by MALDI-TOFMS and, subsequently, by electrospray MS/MS.

MALDI-TOFMS and electrospray MS analysis. The lyophilisate from a fraction isolated by reversed-phase HPLC was dissolved in 100 µL of water. A 20 µl-aliquot was mixed with 20 µL of 0.1 M 2,5-dihydroxybenzoic acid. A 1 µL aliquot of the mixture was placed onto the target plate, allowed to dry, inserted into the MALDI source and analysed in conditions described elsewhere.⁷ External calibration for tryptic peptides was carried out using polyethylene glycol peaks providing a suitable mass calibration in the m/z range up to 2500.

For electrospray MS/MS measurements the sample solution in 50% methanol was introduced into the ion spray source at a flow rate of 3 µL min⁻¹. The ion-spray voltage was 5000 V. The mass range was selected according to the targeted selenospecies and the CID MS spectra were obtained for three Se isotopes (⁷⁸Se, ⁸⁰Se and ⁸²Se). The collision energy was optimized for each selenopeptide and varied between 20 and 50 eV. The TOF mass analyser was calibrated using the polypropylene glycol peaks.

Results and discussion

The yeast analysed contained 2.1 ± 0.1 mg g⁻¹ selenium as determined by HNO₃ and H₂O₂ digestion followed by ICP-MS. The analytical procedure for a systematic comprehensive speciation study of selenium was based on a sequential extraction with water, mixture of cellulases and hemicellulases (Driselase), and a chaotropic agent (sodium dodecylsulfonate) solution. The extracts were examined in terms of stability in the presence of proteolytic enzymes, fractionated by SEC and RP HPLC and submitted to attempts at identification of selenized species by MALDI and electrospray MS, as outlined in Fig.1.

The fractions were isolated and re-chromatographed after lyophilisation and redissolution to produce a series of chromatograms, shown in Figs. 2(b)–2(h). The chromatograms show in each case one peak with elution times matching well those observed in Fig. 2(a), which indicates the stability of fractions in view of degradation/polymerization reactions during the freeze-drying process. The reactivity of selenium compounds in fractions collected as peaks in Figs. 2(b)–2(h) in the presence of proteolytic enzymes was further investigated.

Fig. 2 Size-exclusion LC-ICP-MS chromatograms of: (a) an aqueous extract of selenized yeast; (b–h) fractions 1–7 after collection, freeze-drying and redissolution in the buffer, respectively. The recovery at these conditions was 98.2 ± 2% and can be considered quantitative.

Stability of selenocompounds in size-exclusion LC fractions in the presence of proteolytic enzymes

Enzymatic degradation of selenized yeast has often been used as a means of recovery of selenomethionine, and if present, selenocysteine, from selenized yeast.^{14,16,20–22} A systematic intercomparison study of the different enzymes used for this purpose was reported.⁸ However, proteolysis carried out on a whole sample is not informative regarding which species present have undergone degradation and which have not. Therefore, in this study the stability of selenocompounds in seven individual fractions obtained by size-exclusion LC of a yeast water extract (Fig. 2) were investigated in the presence of proteolytic enzymes. The enzymes chosen were pronase XIV, a non-specific protease from Streptomyces griseus, which is recognised as the most effective enzyme to break the peptides’ bonds down to selenomethionine,⁸ and trypsin, which is widely used for controlled degradation of proteins prior to peptide mapping in proteomics.

High molecular weight fractions. Fig. 3 (left panel) shows analytical size-exclusion chromatograms obtained after enzymatic digestion by pronase and trypsin for the high molecular fractions (>10 kDa, elution volume below 15 min) isolated as shown in Fig. 2 (a). Reversed-phase HPLC-ICP-MS chromatograms of the tryptic digests in peptide mapping conditions were presented for comparison in the right panel of the figure.

Fig. 3 Left panels: size-exclusion LC-ICP-MS chromatograms of the high-molecular weight fractions (fractions 1–3 in Fig. 2) after digestion with trypsin (broken line) and pronase (solid line). The arrows indicate the position of the initial peak (prior to digestion, cf. Fig. 2). Vertical lines mark the elution volume of selenomethionine and selenomethionine oxide under the same chromatographic conditions. Right panels: reversed-phase HPLC-ICP-MS chromatograms of the tryptic digest. The gradient used was: 5% B (5 min), 10% B 5–10 min), 45% B (10–55 min), 70% B (55–60 min), 80% B (60–75 min). See the Procedure section for the composition of the mobile phases A and B. Peak identification: 1, XGHDQSGTK; 2, XNAGR; 3, TYENXKK; 4, TYENXK; 5, DYXGAAK; 6, Ac-SNXMNK; 7, Ac-SNXXNK; 8, ERDDXNXDMGMGHDQSEGGMK and ERDDXNXDXGMGHDQSEGGMK; 9, ERDDXNXDMGMGHDQSEGGMK and ERDDXNXDXGMGHDQSEGGMK; 10, ERDDXNXDXGMGHDQSEGGMK; 11, ERDDXNXDXGXGHDQSEGGMK and DDXNXDXGMGHDQSEGGMK; 12, ERDDXNXDXGXGHDQSEGGMK and DDXNXDXGXGHDQSEGGMK; 13, DDXNXDXGXGHDQSEGGXK, where X denotes selenomethionine.

The fraction 1 (Fig. 3(a)), excluded from the size-exclusion column, is readily digested by pronase to produce a characteristic pair of peaks corresponding to selenomethionine and selenomethionine oxide. This does not say, however, whether the selenomethionine found replaces methionine in the primary structure of a protein, or whether it is non-specifically bound by coordination bonds. Indeed, the size exclusion chromatogram of the tryptic digest shows a continuum in the range 15–25 min. No peak can be recognized, which would suggest that the protein fragments created bind selenomethionine in an unspecific way. This hypothesis is corroborated by reversed-phase HPLC, which shows only some small peaks of selenium-containing peptides, accounting for less than 5% of the total selenium. It can be therefore concluded that there is some high molecular mass (eluting in the void of the SEC column) selenium-containing protein present but the great majority of the selenomethionine is non-specifically bound.

Selenium present in fraction 2 (Fig. 3(b)) is converted quantitatively with pronase into selenomethionine and its oxide. With trypsin it is converted to a single selenium-containing compound identified by MALDI-TOFMS and electrospray MS/MS as a peptide DYXGAAK (where X denotes selenomethionine), possibly a fragment of a heat-shock protein HSP 12, recently identified in yeast.²³ As to fraction 3 (Fig. 3(c)), the chromatogram of the pronase digest shows selenomethionine and some selenomethionine oxide, whereas those of the tryptic digest show a number of peaks. This suggests that selenomethionine present is chemically incorporated into a selenium-containing protein or proteins. An analytical protocol for identification of these peaks has been discussed elsewhere²³ and the results of the identification were given in the caption to Fig. 3(c) (right panel).

Low-molecular weight fractions. The low molecular fractions (<10 kDa, elution time above 15 min) turn out to be remarkably stable in the presence of pronase and trypsin. This may suggest that the selenocompounds present do not contain peptide bonds. Fraction 4 (Fig. 4(a)) contains a selenocompound which has a hydrodynamic volume close to that of a 1000 Da selenopeptide. The chromatographic purity of this peak is poor. At least three equally abundant selenospecies can be detected by reversed-phase HPLC-ICP-MS. Attempts to identify the peaks present by MALDI-TOFMS and electrospray tandem MS were unsuccessful.

Fig. 4 Left panels: size-exclusion chromatograms of the low-molecular weight fractions (4–7 in Fig. 1) after digestion with trypsin (broken line) and pronase (solid line). The arrows indicate the position of the initial peak (prior to digestion, cf. Fig. 1). The vertical lines mark the elution volume of selenomethionine and selenomethionine oxide under the chromatographic conditions. Right panels: reversed-phase HPLC-ICP-MS chromatograms showing the chromatographic purity of the size-exclusion LC-ICP-MS peaks. The gradient used in reversed-phase HPLC was: 0 –15 min 100% A, 50–55 min 40% B, 60 min 100% A.

The digestion of fraction 5, both with pronase and with trypsin, releases some selenomethionine regardless of the enzyme used (Fig. 4(b)). This would suggest that there is some selenomethionine present in this fraction but the binding is more physical than chemical because the chromatograms after digestion with pronase and trypsin are identical. A trace of selenomethionine is also observed in the reversed-phase chromatogram of this fraction. The majority of selenium elutes, however, in the void. Fraction 6 seems to contain selenomethionine (Fig. 4(c)). Indeed, the elution volume of the peak in size-exclusion LC corresponds to oxidized selenomethionine, whereas both forms, oxidized and non-oxidized, can be observed in the reversed-phase chromatogram.

The peak corresponding to Fraction 7 (Fig. 4(d)) is chromatographically pure, as demonstrated by the reversed-phase chromatogram. Electrospray MS/MS confirms it to be Se-adenosyl-homoselenocysteine, the compound first identified in our previous work.¹⁰

Identification of Se-species in the Driselase extract

Driselase is a multicomponent enzyme preparation from a Basidiomycetes sp. that is usually used in the preparation of protoplasts. The mixture contains cellulase, pectinase, laminarinase, xylanase, and amylase, and is known to selectively digest cell wall components. The amount of selenium extracted with a Driselase solution corresponds to ca. 60% of the total selenium present in the yeast sample. This (100 ± 2%) elutes quantitatively as a single peak at the elution volume of selenomethionine. In order to confirm the identity of the compound unambiguously by electrospray MS/MS a method for its purification needed to be developed. Indeed, in a size-exclusion chromatogram selenomethionine co-elutes with compounds containing sulfur and phosphorus which suppress the electrospray ionization. However, by changing the pH of the mobile phase the selenium peak can be separated from the matrix components and interferents, as demonstrated in Fig. 5. The tandem mass spectrum (not shown) contains a number of fragments characteristic of selenomethionine, including m/z of 198, 181, 152, 135, 109 (for ⁸⁰Se), 102, 84 and 74 u. This identification was confirmed by anion-exchange and reversed-phase HPLC-ICP-MS, which showed that selenomethionine was the only selenium-containing compound in this fraction. This finding is important since it shows that a considerable amount of selenomethionine present in yeast may not be incorporated into proteins but rather simply bound to cell walls.

Fig. 5 Optimization of the size-exclusion mobile phase (ammonium acetate–ammonia, 10 mM) pH for the purification of the selenocompound in the Driselase extract. (a) pH 9.5, (b) pH 8.5, (c) pH 7.5, (d) pH 6.68.

Investigation of the sodium dodecylsulfonate extract

The residue after extraction with water and Driselase solution still contains 8% of selenium bound to macromolecular edifices that can be recovered by leaching with a chaotropic agent, such as, for example, sodium dodecylsulfonate (SDS). A chromatogram (Fig. 6(a)) of the SDS extract of the residue after triple aqueous and Driselase extraction shows a single peak excluded from the column. The recovery from the column was quantitative, 98 ± 3%. A proteolytic digest again shows a single peak, eluting at the elution volume of selenomethionine, both in size-exclusion and in reversed-phase LC. Interestingly, the size-exclusion chromatogram of the tryptic digest (Fig. 6(c)) shows a pair of peaks corresponding to selenomethionine and its oxide in a similar way to the chromatogram of the tryptic digest of the void peak of the aqueous extract (cf. Fig. 2(a)). This would suggest that a considerable part of selenomethionine may not be incorporated in the protein structure but bound non-specifically by coordination bonds. The continuum between 15–25 min is more characteristic of cellular debris carrying selenomethionine than a mixture of selenopeptides, as already shown in Fig. 2(a). The reversed-phase HPLC-ICP-MS chromatogram shows a number of selenopeptides, products of the degradation of selenium-containing proteins, but the overall recovery from a reversed-phase column is low, below 50%.

Fig. 6 Size-exclusion chromatograms of the SDS extract. (a) Extract; (b) extract after digestion with pronase; (c) extract after digestion with trypsin. The corresponding reversed-phase HPLC-ICP-MS chromatograms of the enzymatic digests are shown in the insets.

Conclusions

The combination of a sequential extraction with high resolution size-exclusion LC-ICP-MS, supported by peptide mapping approaches and atmospheric pressure ionization MS, allowed a comprehensive study of the speciation of selenium in selenized yeast. Selenomethionine is the major selenium species but, in contrast to the previous findings, a large part of it may not be bound into selenium-containing proteins but rather physically associated with a number of macromolecules, especially cell wall constituents. Orthogonal size exclusion-reversed phase chromatography allows a comprehensive overview of the selenocompounds present and assures their purity sufficient for a successful identification by MALDI-TOFMS or electrospray MS/MS, especially in the case of selenopeptides.

References

1. P. C. Uden, Anal. Bioanal. Chem., 2002, 373, 422.
2. L. C. Clark, J. G. Combs, B. W. Turnbill, E. H. Slate, D. K. Chalker, J. Chow, K. S. Davis, R. A. Glover, G. F. Graham, E. G. Gross, A. Krongrad, J. L. Lesher, K. Park, B. B. Sanders, C. L. Smith and J. R. Taylor, J. Am. Med. Assoc., 1996, 276, 1957.
3. M. P. Rayman, Lancet, 2000, 356, 233.
4. M. Korhola, A. Vainio and K. Edelman, Ann. Clin. Res., 1986, 18, 65.
5. R. Lobinski, J. S. Edmonds, K. T. Suzuki and P. C. Uden, Pure Appl. Chem., 2000, 72, 447.
6. S. McSheehy, J. Szpunar, V. Haldys and J. Tortajada, J. Anal. At. Spectrom., 2002, 17, 507.
7. J. Ruiz Encinar, R. Ruzik, W. Buchmann, J. Tortajada, R. Lobinski and J. Szpunar, Analyst, 2003, 128, 220.
8. C. B′Hymer and J. A. Caruso, J. Anal. At. Spectrom., 2000, 15, 1531.
9. S. M. Bird, P. C. Uden, J. F. Tyson, E. Block and E. Denoyer, J. Anal. At. Spectrom., 1997, 12, 785.
10. C. Casiot, V. Vacchina, H. Chassaigne, J. Szpunar, M. Potin-Gautier and R. Lobinski, Anal. Commun., 1999, 36, 77.
11. L. Bendahl, PhD Thesis, The Danish University of Pharmaceutical Sciences, Copenhagen, March 2003.
12. C. C. Chery, E. Dumont, R. Cornelis and L. Moens, Fresenius’ J. Anal. Chem., 2001, 371, 775.
13. C. C. Chery, H. Chassaigne, L. Verbeeck, R. Cornelis, F. Vanhaecke and L. Moens, J. Anal. At. Spectrom., 2002, 17, 576.
14. M. Kotrebai, M. Biringer, J. F. Tyson, E. Block and P. C. Uden, Anal. Commun., 1999, 36, 249.
15. M. Kotrebai, J. F. Tyson, E. Block and P. C. Uden, J. Chromatogr. A, 2000, 866, 51.
16. M. Kotrebai, M. Birringer, J. F. Tyson, E. Block and P. C. Uden, Analyst, 2000, 125, 71.
17. J. Zheng, W. Goessler and W. Kosmus, Trace Elem. Electrolytes, 1998, 15, 70.
18. S. McSheehy, F. Pannier, J. Szpunar, M. Potin-Gautier and R. Lobinski, Analyst, 2002, 127, 223.
19. K. Wrobel, K. Wrobel and J. A. Caruso, J. Anal. At. Spectrom., 2002, 17, 1048.
20. K. Wrobel, S. S. Kannamkumarath, K. Wrobel and J. A. Caruso, Anal. Bioanal. Chem., 2003, 375, 133.
21. M. Kotrebai and P. C. Uden, Spectrochim. Acta, Part B, 1999, 54, 1573.
22. N. Gilon, A. Astruc, M. Astruc and M. Potin-Gautier, Appl. Organomet. Chem., 1995, 9, 623.
23. J. Ruiz Encinar, L. Ouerdane, W. Buchmann, J. Tortajada, R. Lobinski and J. Szpunar, Anal. Chem., 2003, 75, 3765–3774.
24. S. Ma and R. M. Caprioli, J. Am. Soc. Mass Spectrom., 2003, 14, 593.
25. C. Casiot, J. Szpunar, R. Lobinski and M. Potin Gautier, J. Anal. At. Spectrom., 1999, 14, 645.

---

Footnotes

† On leave from: Department of Chemistry, Technical University of Gdansk, Gdansk, Poland.

‡ On leave from: Department of Applied Chemistry, Szent István University, Budapest, Hungary.

§ On leave from: Department of Analytical Chemistry, Warsaw University of Technology, Warsaw, Poland.

---