Increased Coverage in the Transmembrane Domain with Activated-Ion

Apr 19, 2007 - Nicholas M. RileyJacek W. SikoraHenrique S. SecklerJoseph B. GreerRyan T. FellersRichard D. LeDucMichael S. WestphallPaul M...
0 downloads 0 Views 363KB Size
Download PDF
Increased Coverage in the Transmembrane Domain with Activated-Ion Electron Capture Dissociation for Top-Down Fourier-Transform Mass Spectrometry of Integral Membrane Proteins Vlad Zabrouskov† and Julian P. Whitelegge*,‡ Thermo Fisher, 355 River Oaks Parkway, San Jose, California, California 95134, and The Pasarow Mass Spectrometry Laboratory, The Jane & Terry Semel Institute for Neuroscience and Human Behavior, David Geffen School of Medicine, The Brain Research Institute and The Molecular Biology Institute, 760 Westwood Plaza, University of California, Los Angeles, California 90024 Received December 30, 2006

The c-subunit of ATP synthase (AtpH) is an 8 kD integral membrane protein with two transmembrane domains; we set out to demonstrate it amenable to top-down electrospray-ionization Fourier-transform mass spectrometry (FT-MS) using both collision activated and electron capture dissociation (CAD/ ECD). Thermal activation concomitant with electron delivery was necessary for efficient ECD (activatedion ECD; aiECD), yielding complementary information and greater sequence coverage in the transmembrane domains in comparison with CAD. Keywords: electrospray-ionization mass spectrometry (ESI-MS) • Fourier-transform mass spectrometry (FT-MS) • Fourier-transform ion cyclotron resonance (FTICR) • collision activated dissociation (CAD) • electron capture dissociation (ECD) • activated-ion ECD (aiECD) • infrared multiphoton dissociation (IRMPD) • thylakoid membrane • membrane protein • ATP synthase

Introduction Around one-third of the proteome is made up of integral membrane proteins that are embedded in the biological membranes that compartmentalize living cells and perform vital functions that frequently make them important drug targets.1 Because of their unique physicochemical nature, these proteins are difficult to characterize with conventional enzymatic digestion and LC-MS/MS analyses of the subsequent peptides. Although specialized sample preparation methods are opening the bilayer proteome to “bottom-up” identification experiments,2,3 an alternative approach, “top-down” protein characterization, has been developed that promises coverage in the transmembrane domain. Top-down protein characterization relies on high-resolution Fourier-transform mass spectrometry (FT-MS) to accurately measure the mass of intact protein molecular ions, followed by their fragmentation in the mass spectrometer without prior digestion. This makes it possible to obtain primary structure information for unambiguous identification and characterization of protein covalent modifications.4,5 Incorporating the “top-down” approach, a suite of techniques has been developed using convenient purification and * To whom correspondence should be addressed. The Jane & Terry Semel Institute for Neuroscience and Human Behavior, 760 Westwood Plaza, Los Angeles, CA 90024. Tel, 310 794 5156; Fax, 310 206 2161; E-mail, jpw@ chem.ucla.edu † Thermo Fisher. ‡ University of California. 10.1021/pr0607031 CCC: $37.00

 2007 American Chemical Society

online electrospray-ionization (ESI) mass spectrometry to study intact integral membrane proteins.6-9 FT-MS was used to measure the mass of the seven transmembrane domain bacteriorhodopsin apoprotein with better than 10 ppm mass accuracy,10 whereas collision activated dissociation (CAD) with quadrupole time-of-flight mass analyzers was used to characterize the primary structure of small, single transmembrane helix, integral subunits of the cytochrome b6f complex.11 Described here are experiments demonstrating the top-down analysis of the c-subunit (also known as Subunit III) of the ATP synthase (AtpH) from chloroplasts of Arabidopsis thaliana, an integral membrane protein with two transmembrane domains with hydrophobicity similar to lipids such that it partitions into chloroform (proteolipid). The experiment was conducted using a hybrid linear ion-trap Fourier-transform ion cyclotron mass spectrometer equipped with electron capture dissociation (ECD) to achieve a consistent mass accuracy of better than 5 ppm for product ions from protonated protein molecular ions. Previously, activation of molecular ions in the gas phase (heat, collisions, IR laser) was required only on proteins larger than 20 kDa to increase ECD fragmentation efficiency (for review12). Here we demonstrate for the first time that by activating an 8 kDa membrane protein ions in vacuo using an IR laser at fluences beneath the threshold for infrared multiphoton dissociation (IRMPD), the efficiency of ECD was dramatically enhanced over conventional ECD, so that backbone cleavage was observed frequently in transmembrane R-helical domains, providing information complementary to CAD. This combined Journal of Proteome Research 2007, 6, 2205-2210

2205

Published on Web 04/19/2007

research articles

Zabrouskov and Whitelegge

Figure 1. Molecular ion mass spectrum of AtpH. Insets: (a) Molecular ion adducts; (b) Comparison of the charge state distribution between ubiquitin (8.6 kDa, top) and AtpH (bottom).

thermal ion activation and ECD technique permits analysis of primary structure in the transmembrane domains of membrane proteins that are otherwise challenging to measure.

Experimental Materials and Procedures Protein Samples. A suspension of Arabidopsis thaliana thylakoid membranes (50 µL of 1 mg chlorophyll/mL)13 was diluted with 200 µL of water, 600 µL of methanol, and 200 µL of chloroform respectively and mixed prior to the addition of 400 µL of water to induce phase separation. After vigorous mixing and centrifugation (5 min; 10 000× g), the lower chloroform enriched phase was recovered. An aliquot was injected into an HPLC system for immediate size-exclusion chromatography (Super SW2000, Tosoh Biosciences, Montgomeryville, PA) with a buffer containing chloroform/methanol/ 1% formic acid in water (4/4/1, v/v/v) at 250 µL/minute and 40 °C (7) to purify the ATP synthase c-subunit away from small molecule contaminants including chlorophyll. One-minute fractions were collected into glass vials and stored at -20 °C until they were analyzed by static-nanospray experiments. For direct infusion, sequential LC fractions were individually loaded into 2 µm i.d. externally coated nanospray emitters (New Objective Inc., Woburn, MA) and desorbed using a spray voltage of between 1.2 and 1.4 kV (versus the inlet of the mass spectrometer). These conditions produced a flow rate of 2050 nL/min. Mass Spectrometry. In this study, all protein samples were analyzed using a hybrid linear ion-trap/FTICR mass spectrometer (LTQ FT, Thermo Electron, Bremen, Germany). Ion transmission into the linear trap and further to the FTICR cell was automatically optimized for maximum ion signal. The ion count targets for the full scan FTICR and MS2 FTICR experiments were 2 × 106. The m/z resolving power of the FTICR mass analyzer was set at 100 000 (defined by m/∆m50% at m/z 400). Individual charge states of the multiply protonated protein molecular ions were selected for isolation and collisional activation in the linear ion trap followed by the detection of the resulting fragments in the FTICR cell. For the CAD studies, 2206

Journal of Proteome Research • Vol. 6, No. 6, 2007

the precursor ions were activated using 35% normalized collision energy at the default activation q-value of 0.25. Alternative studies were conducted in which the precursor ions were guided to the FTICR cell and further fragmented using ECD. The ECD fragmentation efficiency was optimized to maximize fragment ion signal. Activated ion ECD experiments were conducted in which precursor ions were excited with an infrared laser to an energy level just below their dissociation threshold while simultaneously being irradiated with electrons via ECD using the same conditions as previously mentioned for conventional ECD. All FTICR spectra were processed using XtractAll (Xcalibur 2.0, Thermo Electron, Bremen, Germany) to produce monoisotopic mass lists (s/n ) 1.1, fit 0%, remainder 0%, averagine table set to averagine), and the fragment assignments were manually validated. For clarity, the mass difference (in units of 1.00235 Da) between the most abundant isotopic peak and the monoisotopic peak is denoted in italics after each Mr value. FTMS data was derived from an average of between 50 and 500 transient signals. Protein identification was achieved by generating sequence tags (sequence tag compiler and sequence tag searching tool, ProSightPC). Fragment mass sequence assignments consistent with the sequence of AtpH were made using ProSightPC software (Thermo Electron, San Jose, CA) operated in single protein mode14 with a 5 ppm mass accuracy threshold. The sequence for AtpH was taken from Swiss-Prot (P56760) with retention of initiating formylmethionine. For manual ion assignments, mass was taken from the most abundant isotopomer, in a cluster with sufficient data to assign charge state (minimally 3 ions) and the monoisotopic mass estimated by calculating a series of decreasing masses from the most abundant isotopomer based upon the difference between the mass of 13C and 12C (1.00235). ProSightPC was used to determine mass accuracy of the assigned monoisotopic mass value. Safety Considerations. Use of chloroform/methanol/1% formic acid in water (4/4/1, v/v/v) under pressure in an HPLC is noted; safety glasses must be worn at all times. Electrospray ionization involves use of high voltages in exposed source

CAD and ECD of Transmembrane Domains

research articles

Figure 2. Fragmentation mass spectra of 5+ protonated molecule of AtpH (Top, inset). (Top) Collision-activated dissociation (CAD) spectrum. (Bottom) Zoom in region m/z 1700-2110; the fragment identities and charge states are indicated.

regions of the mass spectrometer so care must be taken not to touch live components.

Results Electrospray Spectra of the Intact Membrane Protein. A spectrum containing signals for the multiply protonated and adducted AtpH molecules is presented in Figure 1. The charge state distribution of this 8003.4-4 Da ion (this mass is consistent with the AtpH sequence plus one formyl modification, calculated to be 8003.4013 Da for M + 4H+) was dominated by lower charge states, with the 5+ charge state being the dominant signal. A number of oxidized and sodiated adduct signals are identified (Figure 1, inset a), a problem made worse due to the delay between chromatography and FTMS; online SEC-MS does not show any sodium adducts and less oxidation. In contrast, spectra of proteins of the similar size but lower hydrophobicity, such as Ubiquitin, typically show signals indicative of twice as many charges (Figure 1, inset b, top spectra) resulting in m/z signals with a distribution centered at significantly lower m/z

values. The lower number of charges in AtpH results in decreased sensitivity and lowered resolution in the ICR mass analyzer (resolution decreases linearly with increasing m/z). Integral membrane proteins typically have fewer basic residues and exhibit high mass-shifted charge-state distributions. Collisionally Activated Dissociation of AtpH. CAD fragmentation of the 5+ precursor resulted in seventeen amide bond cleavages (Figure 2, 4), with a mass accuracy of 2.9 ppm (RMS) for the corresponding b and y fragments. All cleavages were observed in two transmembrane domains resulting in a cumulative coverage of 28%. No b/y cleavages were observed in the polar loop region (Ala40-Gly47) shown to connect the two hydrophobic helixes in the solution phase.15 Because CAD breaks the weakest amides first, the number of the primary b and y fragments is difficult to increase to improve the efficiency of this dissociation technique. Electron Capture Dissociation of AtpH. The fragmentation by ECD of the 5+ protonated molecules (Figure 3A, top) produced a spectrum that is dominated by signals from reduced charge species, thus confirming that electron capture Journal of Proteome Research • Vol. 6, No. 6, 2007 2207

research articles

Zabrouskov and Whitelegge

Figure 3. (A) Fragmentation mass spectra of 5+ protonated molecule of AtpH. (Top) Electron capture dissociation (ECD) spectrum. (Inset a, b) Partially reduced molecular ions. (Middle) IR activation (5-7 ms) of 8003.4-4 Da molecular ion. No fragmentation or molecular ion depletion was observed. (Bottom) Activated ion electron capture dissociation (IR +ECD; aiECD) spectrum. (B) Expanded regions m/z 1800-1930 and m/z 2060-2600 from the aiECD spectrum (A, bottom panel); the fragment identities and charge states are indicated.

occurred. Surprisingly, only three c and z˘ -type fragments ions (Figure 3A, top, 4) were detected, and of that, only one fragment formed by cleavage at the entrance to the C-terminal transmembrane domain was observed. Such low efficiency in producing c and z˘ fragment ions likely results from a tightly folded tertiary conformation of the molecular ion such that after ECD fragments are formed, they are held together by noncovalent hydrogen bonds and thus have combined m/z of reduced odd electron molecular cation (M + nH)n-1+.; double electron capture leads to the formation of reduced even electron molecular cation (M + nH)n-2+ (Figure 3A top, insets a, b). 2208

Journal of Proteome Research • Vol. 6, No. 6, 2007

Activated Ion Electron Capture Dissociation of AtpH. Previous studies12,16,17 indicate that activation of the gas-phase protein protonated molecules by heat, collisions with neutrals, or IR photons led to temporary unfolding. The simultaneous electron-capture dissociation of such “activated” ions produces many more primary c and z˘ fragments, which led to significantly increased sequence coverage. In this study, we irradiated the precursor ions with the output from an infrared laser for 5-7 ms, well below the dissociation threshold for the precursor ions of interest, while simultaneously performing ECD (Figure 3A, bottom, 4). The resulting spectrum contained signals that were assigned as c (25 observed) and z˘ (30 observed) ions based

CAD and ECD of Transmembrane Domains

research articles

Figure 4. Fragment assignments from CAD (b and y ions), ECD and aiECD (c and z˙ ions) experiments to the sequence of AtpH. The c and z˙ fragments marked by @ symbol are also present in conventional ECD and activated ion ECD spectra. The b and y˙ fragments marked by * symbol are present in both CAD and aiECD spectra; those marked by # symbol are only present in aiECD spectra. The c and z˙ fragments in blue were manually annotated. Transmembrane domains are shaded. The numbering on the right-hand side of the figure is reversed for counting y and z˙ ions. The initiating Met1 retains its N-formyl modification (+27.9949 Da).

on their accurate mass values. Fifty out of 55 fragments were automatically recognized and converted by Xtract to the corresponding monoisotopic masses (2.5 ppm RMS) and assigned using ProSightPC software, whereas the rest were found manually. The latter five fragments exhibited low S/N and poorer ion statistics, contributing to the lower mass accuracy (4.7 ppm RMS). Of the total fragments, 50 were indicative of cleavages in the transmembrane domain (72% coverage) and two in the hydrophilic loop (Figure 3B shows m/z 1800-1930 and 2060-2600 regions of the aiECD spectra with fragment assignments). This is three times as large as sequence coverage in the transmembrane helices as that attained by CAD alone. When the protein molecular ions were irradiated with IR laser only at the same output and duration as was used for combined IR + ECD experiments neither molecular ion depletion nor fragment formation has been observed (Figure 3A, middle). The sequence coverage obtained with CAD and aiECD is summarized in Figure 4. As mentioned above, manual inspection of the aiECD spectrum revealed some ions not previously assigned automatically. Our manual inspections also revealed both b and y ions in the aiECD spectra despite using laser fluences that did not result in IRMPD under just laser excitation. Although we believe the presence of these ions is likely due to thermal processes (CAD), we note that the mechanism of ECD is under heavy discussion and b-ion production has been demonstrated in ECD under conditions where CAD can be eliminated.18 Raw data files(.raw), peaklists (.raw and .xls), and assigned mass lists from Prosight (.xls) are included in supplemental information.

Discussion For top-down mass spectrometry to be a viable proteomics technology, it is important that it be broadly applicable across different classes of proteins. Earlier experiments demonstrated the technique for model soluble proteins of increasing size and more recently human proteins with coding polymorphisms and alternative splicing.19 Large proteins (>200 kDa) have recently been shown compatible with top-down also.20 Efforts are now underway to extend top-down approaches to integral membrane proteins that have inconvenient physicochemical properties when removed from native biological membranes. By use of suitable sample preparation protocols combined with liquid chromatography in appropriate aqueous/organic mixtures, integral membrane proteins can be obtained in solutions compatible with both electrospray-ionization and MALDIbased mass spectrometry.9 In this study, the ATP synthase

c-subunit proteolipid, extracted with chloroform and purified by size-exclusion chromatography in chloroform/methanol/ aqueous formic acid, was subjected to ESI-MS allowing dissociation experiments for primary structure determination using tandem mass spectrometry. The great difficulty of handling the c-subunit of ATP synthase (as well as other hydrophobic proteins) is emphasized in bottom-up proteomics studies. Taylor and colleagues recovered a nine amino-acid CNBr peptide from human heart mitochondria c-subunit,21 derived from the protein’s C-terminus. The facilitating Met residues (equivalent of 68 and 77 in AtpH) are not conserved in the A. thaliana sequence however. Detailed bottom-up proteomic studies of the chloroplast organelle reported recovery of peptides from AtpH using limited acid hydrolysis (2 peptides) and pepsin (1 peptide).22 The retention of the initiating N-formyl group was only revealed by our topdown analyses. Retention of initiating N-formyl Met has been reported for other integral thylakoid membrane proteins,11 despite the presence of chloroplast deformylase activity, and appears to be correlated with chloroplast gene products that are translated directly into the membrane. The observed limited charging of the AtpH protein under standard ESI conditions is likely related to the lack of the basic residues (three in total) as others have discussed23 and likely contributes to the tight tertiary structure of this protein in the gas phase. This hypothesis is supported by published studies on gaseous structures of different charge states of ubiquitin, which is a protein of a similar size to AtpH16 but much greater hydrophylicity. Indeed, previous studies by McLafferty and coworkers16 indicated a presence of a significant tertiary noncovalent bonding in ubiquitin at charge states 9+, or lower, and the strength of such hydrogen bonding has been shown to increase with decreasing charge due to the increased proton affinity of the basic residues. When such an observation is considered in the context of the well-known stability of the tertiary structure of the two transmembrane alpha-helices of the conserved c-subunit from Escherichia coli FOF1 ATPsynthase in chloroform/methanol/water,15 it appears that two factors unite to challenge gas-phase dissociation studies. Limited charging of the molecular ions of AtpH, as well as their secondary and tertiary intramolecular bonding, likely contributed to the limited fragment yield from ECD. It was only when ECD was combined with IR irradiation that these noncovalent bonds were broken as a result of the thermal denaturation by IR laser and ECD yield increased dramatically. The lack of fragmentation when IR (at the same fluence) was used w/o ECD confirms that IR irradiation was only powerful enough to break Journal of Proteome Research • Vol. 6, No. 6, 2007 2209

research articles weaker noncovalent interactions and thus release c and z˘ fragments in the combined ECD and IR experiments (though we note b13 and y3 ions in the aiECD spectrum, Figure 4). Although not utilized in these experiments, other possibilities for denaturation of the gaseous ions such as blackbody infrared irradiation or collisions with neutrals (for review, see ref 12) exist. The order and the relative strength of the tertiary bonds can be estimated from such experiments and detailed studies will be necessary to clarify the relative contributions of limited charging versus stable tertiary/secondary structure in the case of AtpH and other membrane proteins. A soluble “control” the size of Ubiquitin, with less ionizable groups, would be a useful in this respect, but might be an anachronism. Despite limited charging due to less-abundant ionizable groups and strong tertiary bonding, we were able to demonstrate that activated ion ECD is a powerful and complementary tool alongside CAD for sequencing integral membrane proteins in their transmembrane domains, providing more frequent consecutive cleavages for precise primary structure determination. In the case of unknown proteins such increased coverage can be used for membrane protein identification and primary sequence characterization. Indeed, two extensive sequence tags (V[L/I+A]AGL/IAVGL/IA and AL/IAL/IL/IFA) formed by consecutive c and z˘ fragments easily allow unique identification of this protein in the Arabidopsis database. Furthermore, because ECD has been shown to preserve labile post-translational modifications including phosphorylation and glycosylation,24,25 it can be proposed that activated-ECD-aided top-down mass spectrometry will provide a window into posttranslational modifications in hard-to-analyze transmembrane regions; the Schiff-base linked retinal chromophore of bacteriorhodopsin, for example. Previous studies on glyco- and phosphopeptides demonstrated that activated ion techniques could be indispensable when combined with ECD26 to increase fragmentation yield. This is the first report that clearly demonstrates that ECDaided tandem mass spectrometry of an 8 kD intact membrane protein in the gas phase is feasible but hindered, presumably by limited charging as well as strong intramolecular noncovalent interactions, with low sequence coverage. By applying a combination of ECD and thermal denaturation, the fragment yield from such a molecular ion can be significantly increased however (3-fold in this case, compared with CAD), providing complementary information to collision-activated dissociation techniques. Although broad applicability of top-down proteomics in the bilayer domain remains to be demonstrated, we believe the presented experiments provide proof of principle for the technology and highlight issues for further investigation.

Acknowledgment. Financial support from NIH (U01 DE016275-01, P01 NS049134-01, P01 HL80111-01, 1U19 AI067769) is acknowledged. We thank Jimmy Ytterberg and Klaas van Wijk for providing details of their bottom-up work

2210

Journal of Proteome Research • Vol. 6, No. 6, 2007

Zabrouskov and Whitelegge

on AtpH and Bob Barkovich (Thermo Fisher) for software support.

Supporting Information Available: Files for CAD, ECD, and aiECD spectra. Peaklists created by Xtract. Measured fragment masses versus calculated values for AtpH, assigned by ProSightPC for CAD, aiECD, aiECD (b/y ions). This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Lehnert, U.; Xia, Y.; Royce, T. E.; Goh, C. S.; Liu, Y.; Senes, A.; Yu, H.; Zhang, Z. L.; Engelman, D. M.; Gerstein, M. Q. Rev. Biophys. 2004, 37, 121-146. (2) Wu, C. C.; MacCoss, M. J.; Howell, K. E.; Yates, J. R., 3rd. Nat. Biotechnol. 2003, 21(5), 532-538. (3) Blonder, J.; Conrads, T. P.; Yu, L. R.; Terunuma, A.; Janini, G. M.; Issaq, H. J.; Vogel, J. C.; Veenstra, T. D. Proteomics 2004, 4(1), 31-45 (4) Kelleher, N. L.; Lin, H. Y.; Valaskovic, G. A.; Aaserud, D. J.; Fridriksson, E. K.; McLafferty, F. W. J. Am. Chem. Soc. 1999, 121, 806-807. (5) Kelleher, N. L. Anal. Chem. 2004, 76, 197A. (6) Whitelegge, J. P.; Gundersen, C. B.; Faull, K. F. Protein Sci. 1998, 7, 1423-1430. (7) Whitelegge, J. P.; le Coutre, J.; Lee, J. C.; Engel, C. K.; Prive, G. G.; Faull, K. F.; Kaback, H. R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 10695-10698. (8) Whitelegge, J. P.; Go´mez, S. M.; Faull, K. F. Adv. Protein Chem. 2003, 65, 271-307. (9) Whitelegge, J. P. Methods Mol. Biol. 2004, 251, 323-339. (10) Whitelegge, J. P. Photosyn. Res. 2003, 78, 265-277. (11) Whitelegge, J. P.; Zhang, H.; Aguilera, R.; Williams, R.; Cramer, W. A. Mol. Cell Proteomics 2002, 1, 816-827. (12) Cooper, H. J.; Hakansson, K.; Marshall, A. G. Mass Spectrom. Rev. 2005, 24, 201-222. (13) Whitelegge, J. P.; Jewess, P.; Pickering, M. G.; Gerrish, C.; Camilleri, P.; Bowyer, J. R. Eur. J. Biochem. 1992, 207, 1077-1084. (14) LeDuc, R. D.; Taylor, G. K.; Kim, Y. B.; Januszyk, T. E.; Bynum, L. H.; Sola, J. V.; Garavelli, J. S.; Kelleher, N. L. Nucleic Acids Res. 2004, 32, W340-W345. (15) Girvin, M. E.; Rastogi, V. K.; Abildgaard, F.; Markley, J. L.; Fillingame, R. H. Biochemistry 1998, 37(25), 8817-8824. (16) Oh, H.; Breuker, K.; Sze, S. K.; Ge, Y.; Carpenter, B. K.; McLafferty, F. W. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15863-15868. (17) Horn, D. M.; Ge, Y.; McLafferty, F. W. Anal. Chem. 2000, 72, 47784785. (18) Cooper, H. J. J. Am. Soc. Mass Spectrom. 2005, 16(12), 1932-1940. (19) Roth, M. J.; Forbes, A. J.; Boyne, M. T., 2nd.; Kim, Y. B.; Robinson, D. E.; Kelleher, N. L. Mol. Cell. Proteomics 2005, 4(7), 1002-1008. (20) Han, X.; Jin, M.; Breuker, K.; McLafferty, F. W. Science 2006, 314(5796), 109-112. (21) Gaucher, S. P.; Taylor, S. W.; Fahy, E.; Zhang, B.; Warnock, D. E.; Ghosh, S. S.; Gibson, B. W. J. Proteome Res. 2004, 3, 495-505. (22) Peltier, J. B.; Ytterberg, A. J.; Sun, Q.; van Wijk, K. J. J. Biol. Chem. 2004, 279(47), 49367-49383. (23) Fearnley, I. M.; Walker, J. E. Biochem. Soc. Trans. 1996, 24, 912917. (24) Mirgorodskaya, E.; Hassan, H.; Clausen, H.; Roepstorff, P. Anal. Chem. 2001, 73, 1263-1269. (25) Shi, S. D.; Hemling, M. E.; Carr, S. A.; Horn, D. M.; Lindh, I.; McLafferty, F. W. Anal. Chem. 2001, 73, 19-25. (26) Hakansson, K.; Chalmers, M. J.; Quinn, J. P.; McFarland, M. A.; Hendrickson, C. L.; Marshall, A. G. Anal. Chem. 2003, 3256-3262.

PR0607031