Anal. Chem. 2001, 73, 1016-1022
Localization of Analyte Molecules in MALDI Preparations by Confocal Laser Scanning Microscopy Verena Horneffer,* Andre Forsmann, Kerstin Strupat, Franz Hillenkamp, and Ulrich Kubitscheck
Institute of Medical Physics and Biophysics, University of Mu¨nster, Robert-Koch-Strasse 31, 48149 Mu¨nster, Germany
In this study, the incorporation of Texas Red-labeled avidin into crystals of 2,5-dihydroxybenzoic acid (2,5DHB) and 2,6-DHB (used as matrixes for matrix-assisted laser desorption/ionization (MALDI)) was investigated by fluorescence spectrophotometry and confocal laser scanning microscopy (CLSM). The analyte distribution in crystals, grown slowly under controlled conditions, was compared to the analyte localization in different standard preparations (dried-droplet and thin-layer preparation). Texas Red turned out to be a useful fluorescence label in the acidic environments of typical matrixes. Earlier results by absorption spectrophotometry could be confirmed by fluorescence measurements; 2,5-DHB incorporates the analyte proportionally, while 2,6-DHB excludes the protein from its crystal lattice. It is found that the analyte distribution can be analyzed well in both single crystals and standard preparation, by CLSM using Texas Redlabeled analytes. The present study allows for a conclusive and consistent interpretation of analyte incorporation into MALDI preparations. Since their introduction in the late 1980s, matrix-assisted laser desorption/ionization (MALDI) and electrospray ionization (ESI) have revolutionized biological mass spectrometry (MS).1,2 Since its invention, significant improvements in preparation procedures, the identification of better matrixes for a given application, and, last but not least, instrumental developments have made MALDI an indispensable tool for the analysis of a large variety of biological compounds. However, progress in MALDI-MS derives still mostly from empirical testing rather than from a basic understanding of the underlying physical and chemical processes. It is well documented in the literature that different classes of analytes require different matrixes, and a number of analyte/ matrix combinations have been identified for optimal performance.3,4 Specific sample preparation procedures are also associated with many given analyte/matrix combinations and specific * Corresponding author: (tel) xx49-251-8355116; (fax) xx49-251-8355121; (e-mail)
[email protected]. (1) Gross, J.; Strupat, K. Trends Anal. Chem. 1998, 17, 470. (2) Cole, R. Electrospray Ionization Mass Spectrometry; John Wiley & Sons: New York, 1997. (3) Nordhoff, E.; Kirpekar, F.; Roepstorff, P. Mass Spectrom. Rev. 1997, 15, 67. (4) Harvey, D. J. Mass Spectrom. Rev. 1999, 18, 349.
1016 Analytical Chemistry, Vol. 73, No. 5, March 1, 2001
applications.5-8 Matrix isolation to prevent analyte cluster formation in the preparation is generally considered to be one of the important functions of a successful matrix. It is well known that for many if not most matrixes the analyte distribution within a given standard preparation can be quite inhomogeneous on a ∼100-µm scale and is largely dependent on the preparation procedure. Little is known, however, about the distribution of analyte molecules within or at the surface of the matrix crystals on a microscopic level and the mechanisms that determine it. It is also prudent to assume that the speed of matrix crystallization, which varies largely for the different preparation procedures, may have a significant influence on the analyte distribution in the sample. It was shown already early on by redissolving single crystals of 2,5-dihydroxybenzoic acid (2,5-DHB), grown from saturated protein-doped solutions, that this matrix incorporates macromolecules into its matrix lattice.5 Incorporation was also demonstrated for the cinnamic acid derivative sinapinic acid9,10,11 and for the IRMALDI matrix succinic acid.12 Surprisingly enough, other position isomers of benzoic acid show a quite different incorporation behavior. It was shown in particular that 2,6-DHB does not incorporate analyte molecules into slowly grown single crystals at all.13 Yet, Schlunegger and co-workers14 and, later, Horneffer at al. reported a quite satisfactory MALDI performance of this matrix when prepared as a microcrystalline thin-layer preparation as introduced originally by Vorm et al. for the 4-hydroxy-R-cyano cinnamic acid (HCCA) matrix,8,13 or, even better, by vacuum evaporation of the solvent. On the basis of these observations, Horneffer et al. suggested 13 a model for this matrix, according to which analytes get precipitated and isolated from each other at the surface of matrix crystals; the larger the strongly preparation (5) Strupat, K.; Karas, M.; Hillenkamp, F. Int. J. Mass Spectrom. Ion Processes 1991, 111, 89. (6) Beavis, R. C.; Chait, B. T. Rapid Commun. Mass Spectrom. 1989, 3, 432. (7) Beavis, R. C.; Chait, B. T. Rapid Commun. Mass Spectrom. 1989, 3, 436. (8) Vorm, O.; Roepstorff, P.; Mann, M. Anal. Chem. 1994, 66, 3281. (9) Beavis, R. C.; Bridson, J. N. J. Phys. D: Appl. Phys. 1993, 26, 442. (10) Xiang, F.; Beavis, R. C. Org. Mass Spectrom. 1993, 28, 1424. (11) Fournier, I.; Beavis, R. C.; Blais, J. C.; Tabet, J. C.; Bolbach, G. Int. J. Mass Spectrom. Ion Processes 1997, 169/170, 19. (12) Strupat, K.; Kampmeier, J.; Horneffer, V. Int. J. Mass Spectrom. Ion Processes 1997, 169/170, 43. (13) Horneffer, V.; Dreisewerd, K.; Lu ¨ demann, H.-C.; Hillenkamp, F.; Lage, M.; Strupat, K. Int. J. Mass Spectrom. 1999, 185/186/187, 859. (14) Krause, J.; Stoeckli, M.; Schlunegger, U. P. Rapid Commun. Mass Spectrom. 1996, 10, 1927. 10.1021/ac000499f CCC: $20.00
© 2001 American Chemical Society Published on Web 01/24/2001
dependent surface-to-volume ratio, the better the MALDI performance. It was the goal of the present study to close the gap between the observations for slowly grown single crystals and the microcrystalline preparations and thereby to substantiate this model by measuring the localization of fluorescence-tagged proteins in different preparations of 2,5- and 2,6-DHB by confocal laser scanning microscopy (CLSM). Proteins labeled with Texas Red (TR) were used for these studies, because its solution fluorescence declines only moderately in a saturated matrix solution to 25% of its value at neutral pH. CLSM was used before to study the incorporation of analyte molecules into matrix crystals by Dai et al.15,16 using fluorescein isothiocyanate (FITC) as fluorescent label. Their results should, however, be interpreted with caution because neither the pH dependence of the FITC fluorescence17,18 (in a saturated matrix solution
