Benchmarking Multiple Fragmentation Methods on an Orbitrap Fusion

Mar 24, 2015 - Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, California 95134, United ..... An optimum is achieved at a NCE of 12%, with...
0 downloads 0 Views 1MB Size
Subscriber access provided by SUNY DOWNSTATE

Article

Benchmarking multiple fragmentation methods on an Orbitrap Fusion for top-down phospho-proteoform characterization Andrea M. Brunner, Philip Lossl, Fan Liu, Romain Huguet, Christopher Mullen, Masami Yamashita, Vlad Zabrouskov, Alexander Makarov, A. F. Maarten Altelaar, and Albert J.R. Heck Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b00162 • Publication Date (Web): 24 Mar 2015 Downloaded from http://pubs.acs.org on March 31, 2015

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 26

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

Benchmarking multiple fragmentation methods on an Orbitrap Fusion for top-down phosphoproteoform characterization

Andrea M. Brunner a,b, Philip Lössl a,b, Fan Liu a,b, Romain Huguet c, Christopher Mullen c, Masami Yamashita d, Vlad Zabrouskov c, Alexander Makarov a, e , A. F. Maarten Altelaar* a,b, Albert J. R. Heck* a,b a

Biomolecular Mass Spectrometry and Proteomics, Bijvoet Centre for Biomolecular

Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Padualaan 8, 3584CH Utrecht, The Netherlands b

c

Netherlands Proteomics Center, Padualaan 8, 3584CH Utrecht, The Netherlands

Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, California 95134, United States.

d

Forschungsgruppe Zelluläre Strukturbiologie, Max-Planck-Institut für Biochemie, Am Klopferspitz 18, 82152 Martinsried (Germany) e

Thermo Fisher Scientific (Bremen) GmbH, Hanna-Kunath-Str. 11, 28199 Bremen, Germany

* Corresponding Author, E-mail: [email protected]; [email protected]

ACS Paragon Plus Environment

1

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 26

ABSTRACT

Top-down analysis of intact proteins by mass spectrometry provides an ideal platform for comprehensive proteoform characterization, in particular, for the identification and localization of post-translational modifications (PTM) co-occurring on a protein. One of the main bottlenecks in top-down proteomics is insufficient protein sequence coverage caused by incomplete protein fragmentation. Based on previous work on peptides, increasing sequence coverage and PTM localization by combining sequential ETD and HCD fragmentation in a single fragmentation event, we hypothesized that protein sequence coverage and phosphoproteoform characterization could be equally improved by this new dual fragmentation method termed EThcD, recently been made available on the Orbitrap Fusion. Here, we systematically benchmark the performance of several (hybrid) fragmentation methods for intact protein analysis on an Orbitrap Fusion, using as a model system a 17.5 kDa N-terminal fragment of the mitotic regulator Bora. During cell division Bora becomes multiply phosphorylated by a variety of cell cycle kinases, including Aurora A and Plk1, albeit at distinctive sites. Here, we monitor the phosphorylation of Bora by Aurora A and Plk1, analyzing the generated distinctive phospho-proteoforms by top-down fragmentation. We show that EThcD and ETciD on a Fusion are feasible and capable of providing richer fragmentation spectra compared to HCD or ETD alone, increasing protein sequence coverage, and thereby facilitating phosphosite localization and the determination of kinase specific phosphorylation sites in these phospho-proteoforms. Data are available via ProteomeXchange with identifier PXD001845.

ACS Paragon Plus Environment

2

Page 3 of 26

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

TEXT Introduction

Phosphorylation is a fundamental cellular process for the regulation of protein activity, localization, conformation, and oligomeric state, and key to ubiquitous signaling events. In recent years, mass spectrometry (MS)-based phosphoproteomics has yielded a wealth of information on the dynamics of protein phosphorylation through global high-throughput analyses, determining over ten thousand phosphorylation sites per experiment. However, in these approaches information on the defined succession of phosphorylation events and the cooccupation of phosphorylation sites on one protein is largely lost through the digestion of the protein into peptides. In top-down proteomics, intact proteins are analyzed and fragmented by MS-based approaches without the digestion step. This makes top-down proteomics an attractive method to characterize proteoforms, including proteins with post-translational modifications (PTMs)1. However, in contrast to the remarkable total number of proteins identified in top-down experiments, comprehensive proteoform characterization is generally still more limited in specificity and throughput2. In top-down proteomics near complete sequence coverage is still hardly achieved for proteins larger than 10 kDa. This is mainly due to the inability to efficiently fragment larger proteins. Collision induced dissociation (CID) and higher-energy collisional dissociation (HCD) selectively cleave the most labile bonds in a protein, resulting typically in limited sequence coverage3. Use of the alternative fragmentation method electron capture/transfer dissociation (ECD/ETD) was shown to improve sequence coverage as the dissociation occurs prior to energy randomization4,5. Moreover, ETD preserves labile PTMs during fragmentation, facilitating confident PTM site localization assignment6. However, ETD strongly depends on charge density, and ETD tandem mass spectra are often dominated by unreacted and charge reduced precursors 7,8.

ACS Paragon Plus Environment

3

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 26

Recently, UV photo-dissociation has been shown to yield extensive fragmentation generating multiple different ion types, thus leading to improved protein sequence coverage 9. However, potentially the complexity of coupling a laser source to MS instrumentation to date somewhat limits the application of this attractive fragmentation method. A more widely used approach to increase fragmentation efficiency for peptides is the combination of ETD with supplemental activation (ETciD, sometimes also called ETcaD), which targets the chargereduced precursors 10. Moreover, although not commonly applied in top-down proteomics, the collisional activation applied during ETciD could disrupt the tertiary structure of larger proteins 11,12. Another alternative to increase ETD fragmentation efficiency is combining ETD and HCD sequentially in a single event, a fragmentation method termed EThcD13,14. In EThcD, the initial ETD event is followed by HCD fragmentation of all formed ions after electron transfer, including the charged reduced and unreacted precursor ions, yielding c/z and b/y fragment ions in a single MSMS spectrum. The generated dual fragmentation ion series provide richer fragmentation spectra compared to HCD and ETD alone. As HCD and ETD also generally generate fragment ions of complementary parts of the peptide/protein sequence, EThcD leads to increased sequence coverage. In recent work, we increased peptide sequence coverage and PTM localization confidence on peptides using this hybrid fragmentation method on a custom modified Orbitrap Elite 13-16. Now EThcD has been implemented next to ETciD on the new commercially available Orbitrap Fusion platform17, allowing, for the first time, a direct comparison of HCD, ETD, ETciD, and EThcD. Here, we hypothesize that the hybrid methods ETciD and EThcD should improve sequence coverage not only for peptides but also for intact proteins, thereby potentially facilitating phosphoproteoform characterization for medium sized proteins. To test our hypothesis, we systematically optimized and evaluated the performance of EThcD and ETciD for intact protein analysis in comparison to HCD and ETD on an Orbitrap

ACS Paragon Plus Environment

4

Page 5 of 26

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

Fusion. As a model system, we study the phosphorylation of Protein aurora borealis (Bora) by Aurora kinase A (AurA) and Polo-like kinase 1 (Plk1), respectively. Acting in concert with the Ser/Thr kinases AurA and Plk1, Bora is involved in regulating the mitotic entry of eukaryotic cells after the G2/M DNA damage checkpoint18. The 17.5 kDa N-terminal fragment of Bora that we use here (from here on termed Bora), has been described as a kinase substrate of both AurA and Plk119,20. Distinct AurA phosphorylation sites on Bora have recently been determined by bottom-up MS21, but it is not known at which sites Plk1 phosphorylates the Bora N-terminus. A comprehensive top-down analysis of all Bora proteoforms and their phosphorylation sites is challenging as it requires sound sequence coverage and exact phosphosite localization. Here, we unambiguously map the distinctive phosphorylated residues on three Plk1-specific and three AurA-specific phosphoproteoforms of Bora.

Our data clearly demonstrate the benefit of EThcD and ETciD for top-down analyses on an Orbitrap Fusion. They show that ETD-based methods outperform HCD for intact protein analysis over a range of precursor charge states and fragmentation parameters, and allow determination of specific phosphosite localization and defined phosphorylation successions in the multiply phosphorylated protein Bora, showcasing the potential of hybrid fragmentation methods for top-down proteomics.

Experimental section Materials Formic acid (FA) was purchased from Merck (Darmstadt, Germany), acetonitrile (ACN) and Methanol (MeOH) from Biosolve (Valkenswaard, The Netherlands). All materials for the

ACS Paragon Plus Environment

5

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 26

kinase phosphorylation assay were kindly provided by E. Conti (MPI of Biochemistry Martinsried). Kinase Phosphorylation assay The N-terminal fragment of Bora (156 amino acids, 17.5 kDa), the kinase domain of AurA and full-length Plk1 were expressed in and purified from E.coli and prepared as described previously 21. Bora and AurA were incubated for 20 min and 2 h, Bora and Plk1 for 2 h, at a 2:1 molar ratio in 20 mM HEPES pH 7.5, 100 mM sodium chloride at room temperature in the presence of 40 mM magnesium chloride and 40 mM ATP. The kinase reaction was quenched by addition of 80 mM EDTA and placed on ice. Prior to top-down MS analysis, the buffer was exchanged to 0.1% FA. Mass Spectrometry All data was acquired on a Thermo Scientific Orbitrap Fusion mass spectrometer (Thermo Fisher Scientific, San Jose, CA). For direct infusion, Bora samples were diluted to a final concentration of 2 µM in 40% MeOH, 0.1% FA. A 100 µL Hamilton syringe was used for direct infusion, and the flow rate was set to 2 µL/min. Instrument control software Tune 1.1.982 with EThcD and ETciD as built-in methods was used. All data were acquired in the Orbitrap mass analyzer at a resolution of 120,000 (full width at half maximum, FWHM) in intact protein mode (3mTorr ion-routing multipole (IRM) pressure). Ten scans, each consisting of 10 microscans were averaged. The 18, 19, 22 and 24+ charge states of intact, unphosphorylated Bora and the most abundant precursor charge state (19+) of mono-, di- and tri-phosphorylated Bora were subjected to fragmentation. Spectra for all fragmentation methods were acquired using a mass range of 150–2000 m/z. Precursor ion isolation was performed with mass selecting quadrupole and the isolation window was set to 2.0 m/z. The precursor automatic gain control (AGC) target value was 5e5, maximum injection time 200

ACS Paragon Plus Environment

6

Page 7 of 26

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

ms. For HCD, normalized collision energy (NCE) was set to 15, 20 and 25%. ETD ion/ion reaction times (anion AGC 1e6, 200 ms maximum injection time), supplemental collisional activation (SA) in ETciD, and NCE in EThcD varied as indicated. Data analysis Intact protein spectra were deconvoluted with XTRACT with a signal-to-noise ratio (S/N) threshold of 3, a fit factor of 80% and a remainder threshold of 15% using Protein Deconvolution 3.0 (Thermo Scientific). Phosphosite localization was determined using ProSight lite (Proteomics Center of Excellence, Evanston, IL) in combination with an adapted version of the in-house developed previously described program SlinkS15 for fragment ion matching, higher throughput and statistical data analysis. SlinkS can take into account b, y, c, z, b-H2O, y-1, y-H2O, and z+1 ions. Single protein searches were performed with 10 ppm fragment mass tolerance for all fragmentation methods. Manual interpretation was performed by fragment comparison of unprocessed raw spectra to theoretical masses generated by ProteinProspector v5.10.3 (http://prospector.ucsf.edu). Percentage sequence coverage was calculated as the total number of non-redundant inter-residue cleavages divided by the maximum number possible. Phosphosite localization and positional phospho-isomers were further validated by manual inspection of the spectra.

ACS Paragon Plus Environment

7

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 26

Results and Discussion AurA and Plk1 kinase assays to monitor Bora phosphorylation Taking advantage of previously described MS-based kinase assays that allow the monitoring of substrate phosphorylation21, Bora was incubated separately with the kinases AurA and Plk1 to determine phosphate incorporation and occupied phosphorylation sites. Incubation of Bora with AurA in a 2:1 ratio in the presence of Mg2+ and ATP at room temperature resulted in Bora being phosphorylated up to three times after 10 min, with monophosphorylated Bora being the most abundant proteoform (Figure 1B and S1B). After two hours of incubation, Bora was phosphorylated up to four times (Figure 1C and S1C). Unphosphorylated Bora was no longer detected and the most abundant phosphostate observed was the doubly phosphorylated protein, almost equal in intensity to the triply phosphorylated Bora. Two hours of incubation with Plk1 under the same conditions led to a slightly different distribution of also four phospho-states on Bora (Figure 1D and S1D), the most abundant proteoform being the doubly phosphorylated protein, equal intensities of the singly and triply phosphorylated and complete loss of unphosphorylated Bora. The MS spectra in Figure 1C and 1D look very similar, raising the interesting question whether AurA and Plk1 indeed do phosphorylate different Bora residues. In the MS spectra the precursor charge states ranged approximately from 9+ to 27+ for both unphosphorylated Bora and Bora phosphorylated by AurA or Plk1, with the 19+ being the most abundant charge state. At this charge state, phosphoforms differ by