Ultraviolet Photodissociation of Native Proteins Following Proton

File failed to load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... Publication Date (Web): November 22, 2016 ..... Mass spec...
2 downloads 10 Views 1MB Size
Subscriber access provided by Macquarie University

Article

Ultraviolet Photodissociation of Native Proteins Following Proton Transfer Reactions in the Gas Phase Dustin D. Holden, and Jennifer S. Brodbelt Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b03565 • Publication Date (Web): 22 Nov 2016 Downloaded from http://pubs.acs.org on November 30, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Ultraviolet Photodissociation of Native Proteins Following Proton Transfer Reactions in the Gas Phase Dustin D. Holden, Jennifer S. Brodbelt Department of Chemistry, The University of Texas at Austin, Austin, Texas 78712 *

Corresponding author: Jennifer S. Brodbelt, [email protected]

Abstract The growing use of mass spectrometry in the field of structural biology has catalyzed the development of many new strategies to examine intact proteins in the gas phase. Native mass spectrometry methods have further accelerated the need for methods that can manipulate proteins and protein complexes while minimizing disruption of non-covalent interactions critical for stabilizing conformations. Proton transfer reactions (PTR) in the gas phase offer the ability to effectively modulate the charge states of proteins, allowing decongestion of mass spectra through separation of overlapping species. PTR was combined with ultraviolet photodissociation (UVPD) to probe the degree of structural changes that occur upon charge reduction reactions in the gas phase. For protein complexes myoglobin●heme (17.6 kDa) and dihydrofolate reductase●methotrexate (19.4 kDa), minor changes were found in the fragmentation patterns aside from some enhancement of fragmentation near the N- and C-terminal regions consistent with slight fraying. After finding little perturbation was caused by charge reduction using PTR, homodimeric superoxide dismutase/CuZn (31.4 kDa) was subjected to PTR in order to separate overlapping monomer and dimer species of the protein that were observed at identical m/z values.

1 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

Introduction With the advent of new strategies for unravelling details about protein structures, the applications of mass spectrometry in the realm of structural biology have accelerated significantly.1–5 Mass spectrometry methods based on hydrogen-deuterium exchange,

6,7

chemical cross-linking,8–11 and covalent labelling 12–14 have been widely adopted, and the advent of native electrospray ionization methods have allowed native-like proteins to be transferred to the gas phase for analysis as intact species using a myriad of methods like ion mobility15–20 and tandem mass spectrometry. In the context of structural biology, tandem mass spectrometry (MS/MS) has played an increasingly important role.21–38 Electron-based activation methods (electron capture dissociation (ECD) and electron transfer dissociation (ETD)) have been used to characterize native proteins and protein-ligand complexes.23–32 Some non-covalent protein-ligand interactions are retained during the electron-mediated activation process, thus allowing sites of ligand binding to be localized. ECD in particular causes preferential cleavage of backbone bonds in the more flexible regions of the protein, thus supporting correlation of ECD efficiency and Bfactors determined from crystal structures.26–30 Surface induced dissociation (SID), a higher energy activation method, offers an innovative way to map contacts between protein sub-units, thus shedding light on quaternary structures and interfaces of multimeric protein complexes.33–35 Ultraviolet photodissociation (UVPD) is another new way to characterize native protein structures and protein-ligand complexes.36–38 UVPD produces very rich fragmentation patterns that provide exceptional sequence coverage for intact proteins.39–42 In the context of native-like proteins in low charge states, conformational information and insight into conformational changes upon ligand binding are obtained based on variations in fragmentation of the protein

2 ACS Paragon Plus Environment

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

backbone upon UVPD.36–38 Moreover, UVPD produces fragment ions that retain the bound ligand that are useful for pinpointing the ligand binding sites.36–38 Adapting this methodology for even more challenging biological problems requires the ability to analyze multi-component mixtures, often ones containing ions with overlapping mass-to-charge (m/z) values adding to the already complex and information rich UVPD spectra. One way to manipulate ions to simplify spectral analysis is through gas-phase proton transfer reactions (PTR), utilizing ion-ion chemistry in the vacuum chamber of a mass spectrometer. PTR, pioneered by the McLuckey group, has been useful for decongesting complex mixtures of proteins.43,44 Other groups have also recently used proton transfer reactions to increase spectral quality and information for various ion activation methods and mass spectrometer platforms.45–47 A recent study has also sought to elucidate the structure of a PTR reagent and potential mechanism through infrared spectroscopy.48 McLuckey also integrated PTR with ion parking, a method for concentrating ion signal into specific charge states based on strategic control of ion kinetic energies during the proton transfer reactions.49 Utilizing these concepts we recently coupled UVPD with PTR to examine variations in fragmentation and the resulting sequence coverages of proteins as a function of charge state.50 The Hunt group has utilized a similar technique, termed parallel ion parking, to separate overlapping protein species and concentrate fragment ions during liquid chromatographic separation.51 The ability to manipulate and focus charge states of proteins offers potential benefits for native-spray UVPD studies, especially considering the relatively minimal dependence on charge state and proton mobility of UVPD. Other groups have recently studied the effects and benefits of reducing native protein charge state to increase resolution between overlapping non-covalent protein complexes 15

and also to increase the charge state of native proteins to observe the denaturing effect of

3 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

solvent additives useful for supercharging.52 More recently, the Bush group examined the impact of charge reduction reactions on the collision cross sections of denatured and partially denatured ubiquitin in comparison to cross sections of ubiquitin in native-like conformations.16 They concluded that the cross-sections of ubiquitin ions depended most significantly on the final charge state regardless of the initial charge state (prior to reduction by PTR), and that the final structures were not identical despite the similar cross-sections. In this study we establish that charge reduction via PTR is successful for native proteins and produces minimal disruption to ligand binding and noncovalent protein-protein interactions, and observe the minimal effects of charge reduction upon UVPD spectra of native-like proteins. Finally, we apply PTR to a case in which the ions produced by various protein species (monomers and dimers) overlap in order to separate the ions based on charge state to allow confident MS/MS characterization. Experimental Section Materials Equine myoglobin and methotrexate (MTX) were purchased from Sigma-Aldrich (St. Louis, MO), and bovine superoxide dismutase (SOD) was purchased from MP Biomedicals (Santa Ana, CA). Dihydrofolate reductase (DHFR) stock was prepared as previously described;38 briefly DHFR was expressed in E.coli to include a His6-tag and purified into NH4OAc buffer. Solvents were purchased from Thermo Fisher Scientific (Pittsburgh, PA). Mass Spectrometry Proteins were diluted to 10 µM in 200 mM ammonium acetate buffer (pH 6.5) and infused using Au Pd metal-coated (20 nm) borosilicate emitters with applied voltage ranging from 0.9 – 1.2 kV. DHFR was prepared in 50 mM ammonium acetate buffer with a 2X excess of

4 ACS Paragon Plus Environment

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

MTX and incubated for 30 minutes at room temperature. Apo-myoglobin was prepared by first denaturing the protein in 50/49.9/0.5 (ACN/water/formic acid) and incubating at room temperature for 30 minutes followed by two washes with the same denaturing solution using 10 KDa molecular weight cutoff (MWCO) filters from EMD Millipore (Billerica, MA) to remove excess heme. This was followed by three buffer exchanges in 50 mM ammonium acetate solution using 10 KDa MWCO filters. To retain native-like protein structure the heated capillary temperature was maintained at 150°C, following a prior study of source temperature effects on native ESI of ubiquitin.53 Source ion transfer optics were optimized to reduce the kinetic energy of ions during analysis. A Thermo Fisher Scientific Orbitrap Elite mass spectrometer modified to allow UVPD in the HCD cell as described previously39 was used for all experiments, with the resolving power set to 240 K and HCD nitrogen collision gas pressure reduced to δ 0.3 x 10-10 Torr from base pressure of the Orbitrap mass analyzer. 250 scans were averaged per spectrum in triplicate. A single 2.5 mJ pulse from a Coherent ExciStar excimer laser (Santa Clara, CA) at 193 nm was used for UVPD. MSn and reagent ion automatic gain control (AGC) targets were set to 5E5 ions to produce as close to a 1:1 cation (i.e. protein cations) to anion (nitrogen adducted fluoranthene anions of m/z 216) ratio as possible. Ion selection for performing PTR as well as ion parking have been described previously.50 Isolation widths of 30, 50, and 60 were used for myoglobin, DHFR, and SOD, respectively, in order to minimize isolation-induced activation. Data Analysis Spectra were deconvoluted using Xtract (Thermo Fisher Scientific) with a S/N threshold of 2. For construction of total ion current (TIC) abundance plots, absolute abundances for ion types a, a+1, b, c, x, x+1, y, y-1, and z were converted into percentages of TIC. Then fragment

5 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

ions were binned and summed according to the location along the protein backbone. In brief, all N-terminal product ions (an, bn, and cn ions) arising from backbone cleavages that occur Cterminal to a specific amino acid were summed with all the C-terminal product ions (xR-n+1, yRn+1 ,

and zR-n+1 ions) arising from cleavages that occur N-terminal to the same amino acid, where

n is the residue number and R is the total number of amino acids in the protein. For example, for peptide sequence “P-E-P-T-I-D-E” all fragment ions mapped to amino acid “I” would include the summation of ion abundances from N-terminal fragments a5, a5+1, b5, c5, and C-terminal fragments x3, y3, y3-1, and z3. All fragment ions were assigned using ProteinProspector v5.17.1 (http://prospector.ucsf.edu) with a 10 ppm error tolerance. Sequence coverage maps were generated using ProSight Lite v1.3 (Northwestern University) with a 10 ppm error tolerance using UVPD mode. Results and Discussion Native tandem mass spectrometry provides the opportunity to study non-covalent proteinligand and multimeric protein complexes. Owing to the fragility of the individual non-covalent interactions that stabilize the complexes, techniques must be employed to minimize their disruption during analysis. At the same time, conditions used to preserve native-like protein structures in solution and facilitate their transfer to the gas phase results in formation of salt adducts that can congest spectra. Taken together, these issues pose challenges for MS/MS analysis of native-like proteins in low charge states and also provide compelling reasons for the development of ways to manipulate protein ions to alleviate spectral congestion and overlap that might otherwise cause ambiguity or mis-assignment of ions. In this study we integrate PTR with UVPD to allow manipulation of charge states of native-like proteins and separation of overlapping species prior to MS/MS characterization. 6 ACS Paragon Plus Environment

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

As with any study of native-like proteins, it is important to evaluate parameters that may disrupt the native conformations after transfer of the proteins to the gas phase. The impact of proton transfer reactions using nitrogen-adducted fluoranthene anions (m/z 216) was examined for three well-characterized protein-ligand and multimeric systems: myoglobin●heme, DHFR●MTX, and dimeric SOD. The loss of the ligand or dimeric interactions during PTR is one means to assess the disruption of the non-covalent interactions. Native ESI of myoglobin results predominantly in formation of myoglobin●heme complexes in the 7+, 8+, and 9+ charge states along with trace quantities of heme-free myoglobin (apo) ions (Figure 1A). Isolation of the 9+ charge state of myoglobin●heme using an isolation width of 30 produced minimal ejection of heme (Figure 1B), an outcome also observed for the other charge states (8+ and 7+). As expected, isolation and collisional activation of myoglobin●heme (9+) resulted in dominant production of apo-myoglobin ions with a small amount of free heme (m/z 616) observed in the low m/z range (Figure 1C). This result clearly illustrates that even low energy activation disrupts the non-covalent interactions needed to retain the heme group. Myoglobin●heme ions (9+) subjected to 65 ms of proton transfer reactions produced charged-reduced species from 8+ to 5+ (Figure 1D). No apo-myoglobin or free heme ions were generated during the PTR process. The same types of PTR experiments were performed for the two other proteins: DHFR●MTX (8+) and dimeric SOD (11+), shown in Figure S1A and S1B, and in neither case did the protontransfer reactions cause detachment of the ligand or disruption of the dimeric complexes, thus supporting the use of charge-reduction strategies to manipulate charge states in native MS. Ion signal can be further concentrated into a single charge state via “ion parking”, pioneered by McLuckey et al.,49 in which a resonance frequency near the theoretical secular frequency of the targeted parked ion is applied during the PTR. Application of a resonance

7 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

frequency to an ion increases the kinetic energy of the ion, not only decreasing the probability of undergoing further proton transfer reactions but also increasing the probability of collisional activation (an undesirable outcome). In addition to allowing ions to be focused in a single charge state, parking also gives greater control over selection of the charge state. The successful parking of myoglobin●heme in a specific charge state is illustrated in Figure 1E. For this example, the myoglobin●heme complex (9+) was parked in the 7+ charge state during a 150 ms proton transfer period. PTR of the 9+ charge state of myoglobin●heme generates the 7+ complex exclusively when using a moderate resonance amplitude of 2.5 Vpp applied at a frequency 1.5 kHz above the theoretical secular frequency calculated for the 7+ complex. Application of a higher amplitude (3.2 Vpp) caused some parking in both the 8+ (