In-Trap Cleanup of Proteins from Electrospray Ionization Using Soft

ICR trap control, data acquisition, and storage were provided by an Odyssey data ... using our own custom software package, ICR2LS.24 Although time do...
0 downloads 0 Views 168KB Size
Download PDF
Anal. Chem. 1998, 70, 405-408

In-Trap Cleanup of Proteins from Electrospray Ionization Using Soft Sustained Off-Resonance Irradiation with Fourier Transform Ion Cyclotron Resonance Mass Spectrometry Ljiljana Pasˇa Tolic´, James E. Bruce, Q. Paula Lei, Gordon A. Anderson, and Richard D. Smith*

Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington 99352

Electrospray ionization Fourier transform ion cyclotron resonance (FTICR) mass spectrometry is capable of producing high mass resolving power and improved mass accuracy for large proteins and noncovalent complexes when coupled with collisionally induced dissociation (CID) of noncovalent adducts and consequent minimization of ion charge density in the ICR trap during measurements. This work demonstrates the application of in-trap cleanup to several biologically relevant systems, including carbonic anhydrase, 4-oxalocrotonate tautomerase (4OT) analogue, and SecB, a chaperone from Escherichia coli. In-trap cleanup yields improved mass measurements for these systems and is expected to further enable measurements for even more complex systems where adduction levels have precluded study of intact complexes. Although electrospray ionization (ESI)1 Fourier transform ion cyclotron resonance (FTICR)2-5 mass spectrometry is increasingly utilized for biomolecule characterization, its limited m/z range is problematic for studies of biologically relevant high molecular mass and/or high m/z species (i.e., noncovalent complexes). The complexity of biological samples and frequent incompatibility of biological buffers with direct electrospray (even with significant sample cleanup) additionally complicate ESI-MS analysis. Although the gentle nature of ESI allows preservation of noncovalent complexes, it may also seriously complicate the spectrum, due to the presence of the noncovalent adducts from (often unknown) impurities, with cationic adduction (i.e., Na+/K+) being the most common. As a result, a wide variety of methods for minimizing cationic adduction, especially for DNA analysis, have been reported.6-11 Unfortunately, complete noncovalent adduct removal becomes a more difficult problem with increasing molecular mass, * Address correspondence to this author. Phone: (504) 376-0723. E-mail: rd [email protected]. (1) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Mass Spectrom. Rev. 1990, 9, 37-70. (2) Comisarow, M. B.; Marshall, A. G. Chem. Phys. Lett. 1974, 25, 282-283. (3) Comisarow, M. B.; Marshall, A. G. Chem. Phys. Lett. 1974, 6, 489-490. (4) Marshall, A. G.; Verdun, F. R. Fourier Transforms in NMR, Optical, and Mass Spectrometry; Elsevier: New York, 1990. (5) Marshall, A. G.; Grosshans, P. B. Anal. Chem. 1991, 63, 215A-229A. (6) Stults, J. T.; Marsters, J. C. Rapid Commun. Mass Spectrom. 1991, 5, 359363. S0003-2700(97)00828-7 CCC: $15.00 Published on Web 01/15/1998

© 1998 American Chemical Society

due to the higher number of functional groups prone to adduction and generally greater difficulties associated with the cleanup of high molecular mass samples. McLafferty and co-workers have demonstrated efficient cleanup of proteins and DNAs by use of infrared multiphoton dissociation (IRMPD) to “boil off” the noncovalent adducts with minimal fragmentation.12,13 They have also reported the use of collisionally induced dissociation (CID) techniques to detach the noncovalent adducts as a far less efficient cleaning procedures.13 Herein, we demonstrate efficient removal of adducted impurities from large protein ions by use of low-energy sustained offresonance irradiation collisionally induced dissociation (SORICID). Importantly, careful control of irradiation amplitude allows mild removal of adducts without significant dissociation of covalent or specific noncovalent interactions. Efficient in-trap cleanup of proteins by SORI-CID simplifies the spectrum and allows significant improvement in mass measurements accuracy. Examples presented in this paper include noncovalent complexes, thus stressing the gentleness of the technique: preferential dissociation of adducted species, even over cleavage of the weak noncovalent bonds between subunits of large multimeric proteins. Since the energy needed for the breakage of the stronger (i.e., covalent) bonds should be significantly higher, SORI-CID heating is an efficient cleanup procedure alternative to IRMPD; it is also simpler, since it does not require any instrument modifications, and is selective in application. Such in-trap cleanup procedures (IRMPD or SORI-CID) are uniquely applicable in an ion trap and are extremely useful in simplifying ESI mass spectra where adduction is problematic. (7) Little, D. P.; Chorush, R. A.; Speir, J. P.; Senko, M. W.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1994, 116, 4893-4897. (8) Greig, M.; Griffey, R. H. Rapid Commun. Mass Spectrom. 1995, 9, 97102. (9) Muddiman, D. C.; Cheng, X. H.; Udseth, H. R.; Smith, R. D. J. Am. Soc. Mass Spectrom. 1996, 7, 697-706. (10) Wu, Q.; Liu, C. L.; Smith, R. D. Rapid Commun. Mass Spectrom. 1996, 10, 835-838. (11) Liu, C.; Wu, Q.; Harms, A. C.; Smith, R. D. Anal. Chem. 1996, 68, 32953299. (12) Little, D. P.; McLafferty, F. W. J. Am. Soc. Mass Spectrom. 1996, 7, 209210. (13) Speir, J. P.; Senko, M. W.; Little, D. P.; Loo, J. A.; McLafferty, F. W. J. Mass Spectrom. 1995, 30, 39-42.

Analytical Chemistry, Vol. 70, No. 2, January 15, 1998 405

EXPERIMENTAL SECTION Materials. Bovine carbonic anhydrase II was obtained from Sigma Chemical (St. Louis, MO) and was electrosprayed from 10 mM ammonium citrate solution (1 mg/mL). 4-Oxalocrotonate tautomerase (4OT) analogue, in which Met45 was replaced with norleucine, (Nle45)4OT, was prepared by total chemical synthesis as described previously.14,15 SecB from Escherichia coli was purified as described earlier.16 (Nle45)4OT and SecB were desalted by the use of an off-line microdialysis procedure, described elsewhere,10,11 and were electrosprayed from 10 mM NH4OAc solutions, in the concentration range 0.1-1 mg/mL. ESI-FTICR. The 7 T FTICR mass spectrometer has been described elsewhere,17 as well as a custom ESI source and interface incorporating an rf quadrupole for collisional focusing.18 Briefly, the ESI source consists of a heated stainless steel “desolvation” inlet capillary, a 1 mm orifice diameter skimmer, and a short quadrupole segment added to the set of two quadrupole ion guides used in the original configuration and operated in rf-only mode (∼750 kHz, ∼500 Vpp). Mass spectra were obtained utilizing standard experimental sequences employing selected ion accumulation (SIA)19,20 that utilized either broadband (BBQE) or single-frequency (SFQE) quadrupolar excitation. Ion accumulation was accomplished at ∼10-5 Torr of N2, injected into the trap via a piezoelectric pulse valve (Lasertechniques Inc., Albuquerque, NM). Colored noise QE waveforms,21 generated by use of a PC board (PCIP-AWFG, 5 MHz, 12 bit, Keithley Metrabyte Co., Cleveland, OH), were sequentially repeated during the QE event, and typically involved 20 and 1 Vpp quadrupolar fields for BBQE and SFQE, respectively. ICR trap control, data acquisition, and storage were provided by an Odyssey data station (Finnigan, Madison, WI). The high-resolution (Nle45)4OT mass spectra were obtained by the use of time domain data sampling: zeroing22,23 or inverse apodizing (Welch) of the noise between beats prior to Fourier transformation, using our own custom software package, ICR2LS.24 Although time domain data sampling significantly improves the spectral quality for the designated species, these procedures may suppress signals from other species with beat patterns (or molecular masses) which differ from that of the ion of interest. (14) Fitzgerald, M. C.; Chernushevich, I. V.; Standing, K. G.; Kent, S. B. H.; Whitman, C. P. J. Am. Chem. Soc. 1995, 117, 11075-11080. (15) Fitzgerald, M. C.; Chernushevich, I. V.; Standing, K. G.; Whitman, C. P.; Kent, S. B. H. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 6851-6856. (16) Smith, V. F.; Schwartz, B. L.; Randall, L. L.; Smith, R. D. Protein Sci. 1996, 5, 488-494. (17) Winger, B. E.; Hofstadler, S. A.; Bruce, J. E.; Udseth, H. R.; Smith, R. D. J. Am. Soc. Mass Spectrom. 1993, 4, 566-577. (18) Pasa Tolic, L.; Anderson, G. A.; Smith, R. D. Unpublished results, 1997. (19) Bruce, J. E.; Van Orden, S. L.; Anderson, G. A.; Hofstadler, S. A.; Sherman, M. G.; Rockwood, A. L.; Smith, R. D. J. Mass Spectrom. 1995, 30, 124133. (20) Bruce, J. E.; Anderson, G. A.; Hofstadler, S. A.; Van Orden, S. L.; Sherman, M. S.; Rockwood, A. L.; Smith, R. D. Rapid Commun. Mass Spectrom. 1993, 7, 914-919. (21) Bruce, J. E.; Anderson, G. A.; Smith, R. D. Anal. Chem. 1996, 68, 534541. (22) Senko, M. W.; Guan, S.; Huang, Y.; Marshall, A. G.; McLafferty, F. W. Proceedings of the 43rd ASMS Conference on Mass Spectrometry and Allied Topics, Atlanta, GA, 1995; p 806. (23) Kelleher, N. L.; Senko, M. W.; Siegel, M. M.; McLafferty, F. W. J. Am. Soc. Mass Spectrom. 1997, 8, 380-383. (24) ICR2LS; Anderson, G. A., Bruce, J. E., Eds.; Pacific Northwest National Laboratory: Richland, WA, 1995.

406 Analytical Chemistry, Vol. 70, No. 2, January 15, 1998

Figure 1. Positive ion ESI-FTICR mass spectra of bovine carbonic anhydrase II electrosprayed from a citric buffer solution. (A) Spectrum obtained employing BBQE (2000 < m/z < 4000) shows heavily adducted [M + 10H + n(C6H6O8)]10+ molecular ions. (B) SFQE at m/z ) 3250, corresponding to [M + 10H + 18(C6H6O8)]10+ molecular ion, gave an isotopically resolved spectrum (see inset) but was accompanied by partial dissociation, indicating relatively low binding strength of BCA II/citric acid adducts. (C) Application of SORI (1 kHz off-resonance; ∼40 Vpp) resulted in partial dissociation of adducted species. (D) increase in the amplitude of applied irradiation (to ∼50 Vpp) leads to complete dissociation of adducted citric acid molecules, and thus the molecular ion becomes the most abundant species in the spectrum.

SORI of low amplitude (typically 15-50 Vpp) at a frequency ∼1 kHz lower than the reduced ICR frequency of the selected ions has been used for collisional cleanup. Additionally, we note that broadband cleanup of adducted ions was also feasible with the application of low-energy chirp excitation covering the m/z region of interest (i.e., conventional frequency sweep excitation from 10 to 110 kHz at 10 ms-1 sweep rate, ∼100 Vpp). CID events were performed at ∼10-5 Torr of N2, injected into the trap via a piezoelectric pulse valve. RESULTS AND DISCUSSION The utility and efficiency of the in-trap cleanup technique are illustrated below on a few biologically relevant systems. Carbonic Anhydrase. Maintenance of native tertiary structure of biomolecules in the solution to be electrosprayed is obviously essential to studies of noncovalent complexes with mass spectrometry. This often requires solution and buffer conditions that are not optimal for ESI-MS. For example, Figure 1 shows mass spectra of bovine carbonic anhydrase II (BCA II) electrosprayed from a citric buffer, a commonly used buffer in biological research. Typical spectra obtained with BBQE (2000 < m/z < 4000) show heavily adducted molecular ions [M + 10H + n(C6H6O8)]10+, C6H6O8 being citric acid. Extensive adduction

precluded observation of the molecular ion (Figure 1A), even when SFQE at m/z ) 2910, corresponding to nonadducted [M + 10H]10+ molecular ion, was employed during prolonged accumulation events. SFQE at m/z ≈ 3250, corresponding to [M + 10H + 18(C6H6O8)]10+ ion, resulted in the loss of one citric acid molecule (Figure 1B), indicating relatively low binding strength for BCA II/citrate adducts (i.e., conditions used to obtain these spectra are mild enough for nondissociative accumulation of nearly all noncovalent complexes studies thus far). The addition of dipolar irradiation (SORI) resulted in progressive dissociation of adducted species (Figure 1C), until finally, with the increase of the SORI amplitude to ∼50 Vpp, the molecular ion became the most abundant species in the spectrum (Figure 1D). Observation of higher and lower charge states (i.e., [M + 11H]11+ and [M + 9H]9+) in these spectra is likely due to the dissociation of dimeric species present in low abundance due to relatively high analyte concentration (∼30 µM).25 (Nle45)4OT. The quaternary structure of 4-oxalocrotonate tautomerase (4OT) and its analogues obtained by total chemical synthesis has been recently studied by ESI-TOF mass spectrometry.14,15 These studies have confirmed that 4OT from Pseudomona putida mt-2, a part of a set of inducible enzymes used by certain soil bacteria to convert aromatic hydrocarbons to intermediates in the Krebs cycle, is a noncovalent homohexameric complex. Similarly, mutant (Nle45)4OT (monomeric Mr,th ) 6791.75 amu), analyzed herein, folds to form a fully active hexameric enzyme complex resembling 4OT. ESI-TOF investigation of different 4OT mutants allows insight into the structure and function of this highly effective enzyme and the effects that subtle differences in primary structure might have on an enzyme’s higher order structure.14,15 A typical ESI-FTICR spectrum of (Nle45)4OT, obtained using standard sample cleanup procedures prior to MS (i.e., microdialysis),10,11 is shown in Figure 2A. The highest magnitude signals in the spectrum acquired under “native” ESI conditions (i.e., 10 mM ammonium acetate) corresponded to multiply charged monomer and hexamer. The absence of trimers, tetramers, and pentamers suggests that hexamer signals result from structurally specific complexation, and not from random aggregation of the monomer; dimeric species, detected in very low abundance, are likely related to the nonspecific associations due to relatively high analyte concentration (∼10 µM for the monomer). These observations are consistent with ESI-TOF MS results.14,15 BBQE SIA spectrum yielded Mr,exp ) 40 900 ( 350 amu (error is reported as the 95% confidence limit of the mean obtained from different charge states) that compares well with the theoretical Mr,th ) 40 756.5 amu; short-lived ICR signals, commonly observed for high molecular mass species, precluded resolution of the isotopic envelope. SIA employing SFQE at m/z ) 3150 gave the spectrum shown in Figure 2B. As previously demonstrated,19 by the use of SFQE SIA, it is possible to fill the trap to capacity with the ions of a chosen charge state. Thus, enough ions of a particular charge state can be accumulated to allow long-term ion cloud stability resulting in prolonged time domain signal, i.e., high mass resolution. Since SFQE has been centered at m/z ) 3150, and not at m/z ) 3136, corresponding to H13+ molecular ion, adducted (25) Smith, R. D.; Light-Wahl, K. J.; Winger, B. E.; Loo, J. A. Org. Mass Spectrom. 1992, 27, 811-821.

Figure 2. Positive ion ESI-FTICR mass spectra of (Nle45)4OT obtained employing SIA by the use of (A) BBQE (1000 < m/z < 5000) resulting in a low-resolution spectrum; (B) SFQE (m/z ) 3150), which provided higher resolution (see inset); (C) SFQE (m/z ≈ 3150) followed by relatively mild SORI-CID (1 kHz off-resonance, ∼17 Vpp), which gave a clean spectrum with an isotopic envelope that closely matches the theoretical distribution (see inset); and (D) SFQE (m/z ≈ 3150) followed by somewhat harsher SORI-CID (1 kHz offresonance, ∼22 Vpp), which induced the asymmetric breakup of hexamer (H) into monomer (M) and pentamer (P).

ions have been more effectively axialized than nonadducted H13+ molecular ion, and there seem to be more adducts present in the SFQE SIA spectrum (Figure 2B) than in the BBQE SIA spectrum (Figure 2A). The large number of (mostly unidentified) adducts present makes it difficult to assign peaks corresponding to the molecular ion (and, consequently, determine Mr) Our attempts to clean the sample by the use of microdialysis were unsuccessful. Low-amplitude SORI (100 ms, 1 kHz off-resonance, ∼17 Vpp) was found to dissociate most of the adducts, giving a clean spectrum with an isotopic envelope that closely matches the theoretical distribution (Figure 2C); note that molecular ion dissociation during this process is practically negligible. Increasing the amplitude of the applied irradiation (up to ∼22 Vpp) ultimately leads to the fragmentation of hexamer to yield monomer and pentamer (see Figure 2D). Dimeric species have not been detected, although recent X-ray data suggested that 4OT is actually a trimer of dimers. The observed dissociation pattern is in agreement with nozzle-skimmer dissociation ESI-TOF results and illustrates the gentleness of the SORI cleanup procedure, since all higher mass species are removed before any dissociation products are detected. Preferential asymmetric breakup of noncovalent protein complexes into unequal portions have been observed in earlier studies with tetrameric complexes.26-28 SecB. ESI-MS has been successfully used to show that SecB, a chaperone protein in E. coli dedicated to the facilitation of Analytical Chemistry, Vol. 70, No. 2, January 15, 1998

407

Due to the apparent variability in adduct molecular weights and the resulting spread in ion population, resolution sufficient for the definition of these species was not possible. However, SIA of the 16+ charge state, followed by in-trap cleanup, this time with a low-amplitude broadband chirp waveform covering the m/z range 3000-6000, removed much of this adduction, allowing higher resolution detection of the tetrameric complex ions and observation of complexes with various levels of acetylation incorporation (see Figure 3B). The observed Mr of the non-acetlyated complex (68 600 amu) also more closely matches the expected value (68 588 amu), showing the improvement in capability feasible with in-trap cleanup.

Figure 3. Positive ion ESI-FTICR mass spectra of protein SecB: BBQE low-resolution mass spectrum (top) produces a mass measurement roughly 200 amu higher than expected for the nonacetylated tetramer; SFQE of the 16+ charge state, followed by intrap SORI cleanup (low-energy broadband chirp waveform covering the m/z range 3000-6000), removed much of the adduction, allowing observation of complexes with various levels of acetylation (bottom).

polypeptide export, is a homotetramer.16 Low-resolution high m/z spectra of the SecB tetramer, obtained by the use of an extended mass range single-quadrupole mass spectrometer, gave Mr ) 68 610 ( 70 amu, which correlates equally well with the predicted mass of 68 588 amu for the nonacetylated tetramer and 68 672 amu for a 1:1 ratio of the acetylated and nonacetylated forms.16 Figure 3 shows ESI-FTICR mass spectra of the protein SecB (10 mM in 10 mM NH4OAc). Under these conditions, tetramer ions were the predominantly observed species, and the low-resolution mass spectrum in the top of Figure 3 illustrates the narrow charge state distribution that is typically seen with these complexes. Broad peaks, indicative of fairly high levels of adduction, are commonly observed and produce a mass measurement roughly 200 amu higher than expected for the non-acetylated tetramer. (26) Tang, X. J.; Brewer, C. F.; Saha, S.; Chernushevich, I.; Ens, W.; Standing, K. G. Rapid Commun. Mass Spectrom. 1994, 8, 750-754. (27) Light-Wahl, K. J.; Schwartz, B. L.; Smith, R. D. J. Am. Chem. Soc. 1994, 116, 5271-5278. (28) Schwartz, B. L.; Bruce, J. E.; Anderson, G. A.; Hofstadler, S. A.; Rockwood, A. L.; Smith, R. D.; Chilkoti, A.; Stayton, P. S. J. Am. Soc. Mass Spectrom. 1995, 6, 459-465.

408 Analytical Chemistry, Vol. 70, No. 2, January 15, 1998

CONCLUSION In-trap cleanup for ESI-FTICR by use of low-amplitude SORICID has been demonstrated for three biologically relevant systems (bovine carbonic anhydrase II, homohexameric 4-oxalocrotonate tautomerase analogue, homotetrameric SecB). SORI cleanup of noncovalent adducts in an FTICR trap simplifies the mass spectra and yields high mass resolving power and improved mass accuracy for larger proteins and noncovalent complexes. Careful control of irradiation amplitude has been shown to allow removal of adducts without significant dissociation of covalent or specific noncovalent interactions. While it is premature to suggest that such discrimination will always occur, the present results indicate significant utility. In comparison to IRMPD, SORI cleanup is simpler, since it does not require any hardware modifications, and is m/z selective in application. Such in-trap cleanup procedures, uniquely applicable in ion traps due to the extended irradiation periods possible, should be extremely useful in simplifying ESI mass spectra and thus enabling FTICR MS studies of more complex biological systems. ACKNOWLEDGMENT The authors would like to thank Dr. Michael C. Fitzgerald, The Scripps Research Institute, and Drs. Linda L. Randall and Virginia F. Smith, Washington State University, for providing (Nle45)4OT and SecB samples used in this study. This work was supported by a grant from the National Institutes of Health (GM 53558) to Pacific Northwest National Laboratory (PNNL). PNNL is multiprogram national laboratory operated by Battelle Memorial Institute for the U.S. Department of Energy under Contract DEAC06-76RLO 1830.

Received for review July 30, 1997. Accepted October 28, 1997.X AC970828C X

Abstract published in Advance ACS Abstracts, December 15, 1997.