Anal. Chem. 1998, 70, 4410-4416
Applications of In-Source Fragmentation of Protein Ions for Direct Sequence Analysis by Delayed Extraction MALDI-TOF Mass Spectrometry Viswanatham Katta,* David T. Chow, and Michael F. Rohde
Amgen Inc., Mail Stop 14-2-E, Thousand Oaks, California 91320
In matrix-assisted laser desorption/ionization of proteins, there exists a certain amount of fast metastable decay immediately after laser irradiation. The fragment ions thus formed can be resolved and their m/z values measured accurately by employing delayed extraction linear time-of-flight mass spectrometry. At higher than threshold laser fluences, proteins exhibit a series of fragment ions providing useful sequence information. We also observe that when moderate amounts of salts are present in the sample with sinapinic acid being the matrix, the intensities of cn ions (N-terminal fragments) are enhanced compared to other types of fragment ions. This enhancement in cn ion signals allows direct sequencing of proteins. The cn ions are completely absent when XxxPro bonds are encountered and are of lower intensity when Xxx-Gly bonds are involved. Further, the cn ion series is interrupted at Xxx-Cys, when the cysteine is involved in a disulfide bond. Upon reduction of the disulfide bonds, the series continues and information is available for longer stretches. Using 10-20 pmol of recombinant proteins, sometimes contiguous sequence information up to 70 residues is obtained in a matter of minutes. Applications of the technique to some recombinant proteins with intra- or interchain disulfide linkages are presented. Mass spectrometric analysis of peptides and proteins has been revolutionized by the advent of “soft” ionization techniques, namely, matrix-assisted laser desorption/ionization (MALDI)1 and electrospray ionization (ESI).2 For many biological applications, MALDI time-of-flight mass spectrometry (TOF MS) has turned out to be a very valuable tool because of its high sensitivity, tolerance to salts and buffers, and ability to analyze mixtures directly.3 Initially, MALDI-TOF MS had lower resolution, but improvements in understanding the role of matrixes and sample preparation techniques led to an increase in resolution and mass
accuracy.4 In the linear time-of-flight mode, the resolution is mainly limited by the kinetic energy spread caused by the initial distribution of velocities of the analyte ions.5 By using dual-stage acceleration, where the first stage enables the protein ions to be extracted “gently” out of the plume of desorbed matrix molecules, resolution could be improved by reducing the energy deficits caused by collisions. Still, the resolution for large proteins is limited by metastable decay, which causes skewing of the peak shapes. Recently, the introduction of pulsed ion extraction,6 or delayed extraction (DE),7 or similar approaches8 has temporally decoupled the process of ion production after laser irradiation from ion extraction into the flight tube. In delayed extraction, ions are initially produced in a nearly field free region by a laser pulse, and after a delay of a few hundred nanoseconds (enough time for the high-density plume to completely dissipate), they are extracted into the flight tube by a high-voltage pulse applied to the source. This decoupling has considerably improved the resolution of molecular ions, and two factors appear to be responsible for this: (a) appropriate values of the delay and amplitude of the high-voltage pulse compensate for the initial velocity spread by making the ions of identical m/z values arrive simultaneously at the linear detector, and (b) the molecular ion peak is sharpened by reducing the contribution from the metastable decay.9 In addition to accurately measuring the molecular weight of proteins, MALDI-TOF MS has been employed in obtaining partial sequence information using ladder sequencing approaches. In these experiments, the individual amino acids are serially removed from either end of the peptide/protein by chemical means10 or exopeptidase digestion.11 The mass spectrum of the resulting mixture will show peaks separated by the masses of the corresponding amino acids, thus providing a contiguous sequence.
* To whom correspondence should be addressed: (tel) 805-447-6534; (fax) 805-499-7464; (e-mail)
[email protected]. (1) Karas, M.; Bachman, D.; Bahr, U.; Hillenkamp, F. Int. J. Mass Spectrom. Ion Processes 1987, 78, 53-68. (2) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246, 64-71. (3) Beavis, R. C.; Chait, B. T. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 6873-7.
(4) Beavis, R. C.; Chait, B. T. Methods Enzymol. 1996, 270, 519-51. (5) Beavis, R. C.; Chait, B. T. Chem. Phys. Lett. 1991, 5, 479-84. (6) Brown, R. S.; Lennon, J. J. Anal. Chem. 1995, 67, 1998-2003. (7) Vestal, M. L.; Juhasz, P.; Martin, S. A. Rapid Commun. Mass Spectrom. 1995, 9, 1044-50. (8) Whittal, R. M.; Li, L. Anal. Chem. 1995, 67, 1950-4. Colby, S. M.; King, T. B.; Reilley, J. P. Rapid Commun. Mass Spectrom. 1994, 8, 865-8. (9) Bahr, U.; Stahl-Zeng, E.; Gleistmann, E.; Karas, M. J. Mass Spectrom. 1997, 32, 1111-6. (10) Chait, B. T.; Wang, R.; Beavis, R. C.; Kent, S. H. Science 1993, 262, 8992. (11) Patterson, D. H.; Tarr, G. E.; Reigner, F. E.; Martin, S. A. Anal. Chem. 1995, 67, 3971-8.
4410 Analytical Chemistry, Vol. 70, No. 20, October 15, 1998
S0003-2700(98)00034-1 CCC: $15.00
© 1998 American Chemical Society Published on Web 09/12/1998
Alternatively, the protein is digested with an endoproteinase and the digest is directly analyzed in the linear or reflector mode, providing a peptide mass map. Information about the sequence of individual peptides is then obtained by postsource decay analysis, where the fragments formed from metastable decay of a chosen peptide are measured using reflector based instruments.12 The postsource decay analysis, however, is still not effective for peptides with molecular weights greater than 3000 because of the extensive metastable decay. Though fragmentation by mass spectrometry is rapid and effective, interpretation of the data can be slow and is often plagued by difficulties such as (a) uncertainity of an observed fragment to be an N- or C-terminal ion, (b) multiple cleavages around each peptide bond, (c) formation of internal fragment ions, and (d) in some cases, ions resulting from small neutral losses being more intense than those formed by simple cleavage of an amide bond. Brown and Lennon13 have shown that, even with delayed extraction, a considerable amount of metastable decay occurs in the ion source prior to the ion extraction into the flight tube. They observed that, within 320 ns (minimum delay possible with their instrument) after laser irradiation, large proteins exhibit a wealth of sequence-specific fragmentation along the peptide backbone. They found that there is a preference for the formation of cn fragment ions (the nomenclature used is a modified version14 of the original notation proposed by Roepstorff and Fohlman15). Their spectra also showed a number of yn and zn ions, making data interpretation difficult but providing a considerable amount of sequence coverage. Factors influencing the fast fragmentation were thoroughly studied by Brown et al.,16 and it was concluded that the fragment ions observed are not due to prompt fragmentation,17 but are a result of metastable decay on a fast time scale (different from the metastables observed in postsource decay). This fast metastable process is also shown to be influenced by the laser fluence and the matrix used. Protein sequencing by mass spectrometry will greatly benefit, if one can control the fragmentation and limit it to only one kind of fragments as the generated ladder will be much easier to interpret. Here we report that the relative intensity of the cn ions was increased substantially by the presence of salts in the sample, thus making it easier to deduce the sequence. Using this approach, we were able to generate such continuous ladders, extending up to 70 residues, for some of the proteins. Recently, another article appeared about direct sequence analysis of proteins.18 Here we show applications to some recombinant proteins with intra- or interchain disulfide linkages. EXPERIMENTAL SECTION All mass spectra were acquired using a delayed extraction MALDI-TOF mass spectrometer (model Voyager DE-RP, Perseptive Biosystems Inc., Framingham, MA) operated in the linear (12) Spengler, B.; Kirsch, D.; Kaufmann, R. Rapid Commun. Mass Spectrom. 1991, 5, 198-202. (13) Brown, R. S.; Lennon, J. J. Anal. Chem. 1995, 67, 3990-9. (14) Johnson, R. S.; Martin, S. A.; Biemann, K. Int. J. Mass Spectrom. Ion Processes 1988, 86, 137-54. (15) Roepstorff, P.; Fohlman, J. Biomed. Mass Spectrom. 1984, 11, 601. (16) Brown, R. S.; Carr, B. L.; Lennon, J. J. J. Am. Soc. Mass Spectrom. 1996, 7, 225-32. (17) Patterson, S. D.; Katta, V. Anal. Chem. 1994, 66, 3727-32. (18) Lennon, J. J.; Walsh, K. A. Protein Sci. 1997, 6, 2446-53.
mode. The instrument is generally operated with an accelerating voltage (ACV) of 25 kV and an extraction pulse voltage of 2 kV. Even though acquisition methods allow a delay to be specified for the extraction pulse (e.g., 150 ns), because of an inherent delay of 180 ns that exists between the laser pulse (337-nm wavelength,
