The Impact of Phosphorylation on Electron Capture Dissociation of Proteins: A Top-Down Perspective
- 661 Downloads
Electron capture dissociation (ECD) is well suited for the characterization of phosphoproteins, with which labile phosphate groups are generally preserved during the fragmentation process. However, the impact of phosphorylation on ECD fragmentation of intact proteins remains unclear. Here, we have performed a systematic investigation of the phosphorylation effect on ECD of intact proteins by comparing the ECD cleavages of mono-phosphorylated α-casein, multi-phosphorylated β-casein, and immunoaffinity-purified phosphorylated cardiac troponin I with those of their unphosphorylated counterparts, respectively. In contrast to phosphopeptides, phosphorylation has significantly reduced deleterious effects on the fragmentation of intact proteins during ECD. On a global scale, the fragmentation patterns are highly comparable between unphosphorylated and phosphorylated precursors under the same ECD conditions, despite a slight decrease in the number of fragment ions observed for the phosphorylated forms. On a local scale, single phosphorylation of intact proteins imposes minimal effects on fragmentation near the phosphorylation sites, but multiple phosphorylations in close proximity result in a significant reduction of ECD bond cleavages.
KeywordsPhosphorylation Mass Spectrometry Electron Capture Dissociation Top-down MS
Since the introduction of electron capture dissociation (ECD) in 1998 , this powerful tandem mass spectrometry (MS/MS) technique has been widely used to elucidate amino acid sequences, structures, and, particularly, to localize post-translational modifications (PTMs) of proteins [2, 3, 4, 5, 6]. During ECD, ion analytes (proteins or peptides) are irradiated by low-energy electrons, cleaving N-Cα bonds and producing c-type and z-type ions [1, 2, 7, 8]. Compared to traditional “slow heating” MS/MS methods such as collisionally activated dissociation (CAD) [9, 10] and infrared multi-photon dissociation (IRMPD) , which usually cleave the weakest bonds in a given protein/peptide ion and lead to the preferred loss of labile PTMs, ECD causes fast fragmentation prior to energy randomization over the whole protein or peptide [1, 12, 13, 14]. This characteristic of ECD generally results in preservation of labile PTMs [2, 15], such as phosphorylation  and O-GlcNAc .
Phosphorylation, a highly prevalent PTM, plays a critical role in regulating various complex cellular processes, and altered phosphorylation is implicated in many diseases [5, 6, 18]. During CAD, the phosphorylated ions often experience 98 and 80 Da neutral losses, hindering the interrogation of phosphorylation sites . In contrast, ECD preserves phosphorylation while providing rich fragmentation patterns, presenting a valuable option to characterize both phosphopeptides and phosphoproteins . For phosphopeptides, it has been demonstrated that ECD can be applied in data dependent online LC-MS/MS experiments for large scale identification and localization of phosphorylation sites . However, because of the elongated ion accumulation time, utilization of ECD in online LC-MS/MS is not widely utilized . On the other hand, ECD is particularly powerful in characterizing phosphoproteins in an offline mode. Over the past years, our group has successfully applied Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) equipped with ECD to comprehensively characterize biologically important phosphoproteins in the heart such as myosin binding protein C , cardiac troponin I [23, 24, 25], tropomyosin [26, 27], and AMP-activated protein kinase catalytic domain . Furthermore, we were able to discover reduced phosphorylation of a previously unknown phosphoprotein, enigma homolog isoform 2, from swine heart tissue after acute myocardial infarction and localize the phosphorylation site with ECD , which demonstrated the power of ECD for characterizing phosphoproteins.
In addition to localizing phosphorylation sites, ECD has also proven feasibility for quantitative analysis. The Kelleher group has pioneered and demonstrated the use of ECD spectra to quantify isomeric methylated and acetylated proteoforms from histone H4 . Our group has also developed a novel strategy to quantify isomeric phosphorylated proteins that utilized only the abundance of unphosphorylated fragment ions with intraspectrum normalization, which avoids confronting potential influence from the phosphate groups on the ECD . However, the influence of phosphorylation on intact protein ECD fragmentation is still largely unknown. Although Creese and Cooper showed that the addition of phosphorylation has deleterious effect on ECD sequence coverage for doubly charged peptides , and later demonstrated that the structure rather than the sequence of phosphopeptides influences ECD efficiency , these results could not be generalized to proteins. Compared with peptides, the physiochemical properties for larger intact proteins tend to be much less affected by the addition of phosphorylation, thus prompting further investigation.
Here, we have systematically investigated the potential effect of phosphorylation of intact proteins on ECD behavior using mono-phosphorylated α-casein, multi-phosphorylated β-casein, and immunoaffinity-purified cardiac troponin I from swine and rhesus monkey tissues. We have localized the phosphorylation sites and statistically compared the ECD fragments of the unphosphorylated and the phosphorylated forms of these proteins. The results collectively show that the phosphorylation has considerably less deleterious influence on ECD fragmentation of intact protein compared to peptides. Globally, the total number of fragment ions produced by ECD was slightly decreased for the phosphorylated precursor ions, but the fragmentation pattern remained highly comparable between the unphosphorylated and the phosphorylated proteins. Locally, near the amino acid residues adjacent to the single phosphorylation site, fragmentation patterns remained largely unchanged; however, multiple phosphorylations in close proximity led to a decrease in protein backbone cleavages. These findings provide a contemporary top-down perspective on ECD fragmentation of phosphoproteins.
Chemicals and Reagents
Dephosphorylated α-casein (bovine milk) was purchased from Sigma-Aldrich (St. Louis, MO, USA). β-Casein (bovine pancreas) standard was purchased from Protea Biosciences (Morgantown, WV, USA). Lambda protein phosphatase (λPP) was purchased from New England Biolabs (Ipswich, MA, USA). Pierce Protein Concentrator PES, 10K MWCO, high-performance liquid chromatography (HPLC)-grade water (H2O), and acetonitrile (ACN) were purchased from Fisher Scientific (Waltham, MA, USA). All other reagents were purchased from Sigma-Aldrich unless otherwise specified.
In Vitro Dephosphorylation of β-Casein
For every 20 μL of dephosphorylation reaction, 10 μg of β-casein was incubated with 100 units of lambda protein phosphatase (λPP) with the supplement of 1 mM MnCl2 in 1× protein metallophosphatase buffer. The mixture was incubated at 30 °C for 0.5 h. Subsequently, dephosphorylated β-casein was desalted using 0.1% formic acid in H2O wash through 10 K MWCO protein concentrators prior to top-down MS analysis. The final concentrations of the phosphorylated and the unphosphorylated β-casein were about 0.25 μg/μL in ACN: H2O (50:50) and 0.1% formic acid.
Immunoaffinity Purification of Swine and Rhesus Monkey cTnI
Heart tissue was obtained from juvenile Yorkshire domestic pig and rhesus monkey hearts as approved by the University of Wisconsin Animal Care and Use Committee. The excised heart tissue was immediately frozen in liquid nitrogen and stored in a – 80 °C freezer. The whole cardiac troponin complex was purified by the immunoaffinity purification method as previously described [34, 35]. Briefly, 1.5 g of heart tissue was homogenized in tissue wash buffer (NaH2PO4 500 mM, Na2HPO4 100 mM, MgCl2 100 mM, EGTA 100 mM, NaCl 0.1 M, Triton X-100 1%, DTT 5 mM, protease and phosphatase inhibitor cocktail tablet, PMSF 1 mM, leupeptin 2 μg/mL, pH 7.4) using a Polytron electric homogenizer (Model PRO200, PRO Scientific Inc., Oxford, CT, USA) for 30 s on ice. The homogenate was centrifuged at 8000 rpm at 4 °C for 10 min; the pellet was collected and resuspended in 6 mL protein extraction buffer (0.7 M LiCl, 25 mM Tris, 5 mM EGTA, 0.1 mM CaCl2, 5 mM DTT, 1 mM PMSF, 2 μg/mL leupeptin, and 0.75 mg/mL protease and phosphatase inhibitor cocktail, pH 8.0) and protein extraction was performed with agitation on a nutating mixer (Fisher Scientific Inc., Pittsburgh, PA, USA) at 4 °C for 45 min. The sample was then centrifuged at 16,000 rpm (Centrifuge 5415R; Eppendorf, Hamburg, Germany) for 5 min. The supernatant was further centrifuged at 55,000 rpm (Beckman L-55 ultracentrifuge; Beckman Coulter, Fullerton, CA, USA) for 45 min to completely remove any particulate. The supernatant was incubated with 0.25 mL of CNBr-activated Sepharose CL-4B conjugated with 1.25 mg monoclonal cTnI antibody (anti-troponin I monoclonal antibody MF4 and 14G5; HyTest Ltd., Turku, Finland) for 35 min at 4 °C to ensure complete binding of troponin complex to the antibody. After washing the column with 2 mL of extraction buffer, the bound troponin complex was eluted with 100 mM glycine at pH 2 into four 0.4 mL fractions and neutralized immediately with 40 μL of 1M MOPS (pH 9). Fractions were further separated and desalted using an offline reverse phase protein microtrap (Michrom Bioresources Inc., Auburn, CA, USA) to obtain purified cTnI.
Top-Down MS Analysis
A 7T linear ion trap/Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer (MS) (LTQ/FT Ultra, Thermo Scientific, Bremen, Germany) and a 12T SolariX FTICR-MS (Bruker, Bremen, Germany) equipped with ECD were used for the investigation. Triversa NanoMate (Advion, Ithaca, NY, USA), a chip-based nano-electrospray ionization (nanoESI) system delivered all samples into the mass spectrometers. For the LTQ/FT Ultra, the resolving power was set to 200,000 at 400 m/z. During ECD, individual precursor ions with the same charge state of the unphosphorylated and the phosphorylated proteins were isolated and fragmented using 2.2%–3.3% electron energy with a 50 ms duration and no delay; 1000 to 9000 transients were averaged to ensure high quality of ECD spectra. For the SolariX FTICR-MS, the mass spectra were collected over a 202.76 to 3000 m/z range with 2 mega-words (resolution 270,000 at 400 m/z) transient length (1.1534 s) and a pulse at 20% excitation power. All MS/MS spectra were accumulated for 1000 scans unless specified otherwise. For the “standard” parameters, ECD pulse length was set to 0.05 s; ECD bias was set to 0.8 V (0.8 eV); and ECD lens was 10 V for ECD fragmentation.
MASH Suite  and MASH Suite Pro  were used to analyze MS/MS data generated from 7T LTQ/FT Ultra and 12 T SolariX FT-ICR, respectively. All peaks were extracted with a signal-to-noise ratio set to 3 and a minimum fit of 60%, followed by manual validation. Fragment ions were assigned by matching the experimental fragment ions to the calculated ones from theoretical amino acid sequence with a 10 ppm mass tolerance. The abundance of each ion is the sum of the intensities from the top five most abundant isotopomers. Subsequently, the quantitation value was normalized to the charge of the ion. The normalized values from the same ions with different charge states were then summed. The relative abundance of each fragment ion was calculated by normalizing to the most abundant fragment ion within the same spectrum.
Results and Discussion
To evaluate the phosphorylation effect on ECD fragmentation of intact protein, it is critical to maintain phosphorylation as the only changing factor and minimize other variations arising from instrument parameters and sample introduction. ECD parameters were kept constant between ECD of unphosphorylated and phosphorylated counterparts of the same protein. Because the precursor ion signal in the FTMS instrument may introduce biased fragment ion intensity, the ion accumulation was adjusted between the isolated unphosphorylated and phosphorylated precursor so that their intensities are comparable in a single scan. Furthermore, fragment ion intensity was normalized to the most abundant fragment ion in each spectrum to minimize the variation caused by the difference in precursor ion intensity. Here, three different protein species, including single and multiple phosphorylation sites with multiple replicates, were used to investigate the influence of phosphorylation on ECD of intact protein.
ECD of Mono-Phosphorylated Versus Unphosphorylated α-Casein
ECD of Multi-Phosphorylated Versus Unphosphorylated β-Casein
ECD of Mono-Phosphorylated Versus Unphosphorylated Cardiac Troponin I from Swine and Rhesus Monkey
The Effect of Phosphorylation on ECD of Protein Versus Peptide Ions
Compared with peptides, intact proteins are much larger in size and, therefore, the addition of one or more phosphorylations is less likely to perturb the overall ion behavior during ECD. We demonstrated that, locally, adjacent to the phosphorylation site, ECD cleavages remain similar between mono-phosphorylated and unphosphorylated proteoforms. However, due to the introduction of a negatively charged moiety (deprotonated phosphate group), phosphorylated protein ions still experience different fragmentation than unphosphorylated counterparts even with selection of the same charge state for fragmentation. This becomes more apparent when multiple phosphorylations are present in close proximity on a protein. For the case of the multiply phosphorylated versus unphosphorylated β-casein, ECD fragmentation was greatly reduced at the region surrounding the multiple phosphorylation sites. In a previous study, Creese and Cooper investigated the ECD efficiency on phosphopeptides using a peptide from β-casein (Phe33 to Lys48; phosphorylated Ser35) . They found that the phosphorylated peptide did not fragment as well as the unphosphorylated counterpart, and postulated that the phosphorylation alters peptide folding through the formation of salt-bridges between the glutamine and the glutamic acid residues. Using a very high electron energy, they were able to cleave 13 out of the 15 N–Cα bonds on the phosphopeptide. In our case, ECD cleaved all of the previously mentioned 15 N-Cα bonds around Ser35 (Figure 4b) of the phosphorylated β-casein with low electron energy. Nevertheless, in contrast to peptide ions, protein ions carry significantly more charges in denatured conditions, which not only disturbs electrostatic interaction (salt bridges) [39, 40], but also increases ECD reaction cross-section , resulting in minor differences in the overall ECD cleavages between the phosphorylated and the unphosphorylated species without the requirement of additional electron energy as for phosphopeptides . Our data also suggests that globally, ECD of the phosphorylated form generally produces slightly fewer fragment ions; however, it is drastically less affected by phosphorylation compared with the marked effects observed for phosphopeptides.
In this study, we systematically investigated the influence of phosphorylation on ECD fragmentation of intact proteins, including mono-phosphorylated α-casein, multiply phosphorylated β-casein, and immunoaffinity-purified cardiac troponin I from swine and rhesus monkey, by analyzing the relative ion abundance from the unphosphorylated and the phosphorylated forms under the same ECD conditions. The results suggest that compared with phosphopeptides, phosphorylation has little deleterious effect on ECD fragmentation of intact proteins. Globally, phosphorylation on intact proteins slightly decreases the total number of fragment ions, but the pattern of fragmentation remains highly similar. Locally, phosphorylation on intact proteins tends not to significantly affect the fragmentation and cleavages of the amino acid residues adjacent to the phosphorylation sites; however, when multiple phosphorylations are present in close proximity, cleavages near the phosphorylation sites decrease.
This work is dedicated to Professor Kristina Håkansson, the recipient of the 2016 Biemann Medal of the ASMS, for her outstanding achievement in the field of mass spectrometry related to electron-based activation methods. The authors thank Fangmin Xu, Han Zhang, and Huseyin Guner on the initial studies on the LTQ/FT. They thank Jeremy Wolff from Bruker for technical assistance on the Solarix FTMS. They also thank Dr. Barbara Lewis and Zachery Gregorich for critical reading of the manuscript. Financial support was kindly provided by NIH R01HL109810 and HL096971 (to Y.G.) and GM117058 (to Song Jin and Y.G.). They also acknowledge NIH high-end instrument grant S10OD018475 (to Y.G.). T.T. acknowledges support of the NIH Chemical Biology Interface Training grant NIH T32GM008505.
- 6.Cai, W.X., Tucholski, T.M., Gregorich, Z.R., Ge, Y.: Top-down proteomics: technology advancements and applications to heart diseases. Expert Rev. Proteomics. 13, 717–730 (2016)Google Scholar
- 7.Zubarev, R.A., Kruger, N.A., Fridriksson, E.K., Lewis, M.A., Horn, D.M., Carpenter, B.K., McLafferty, F.W.: Electron capture dissociation of gaseous multiply-charged proteins is favored at disulfide bonds and other sites of high hydrogen atom affinity. J. Am. Chem. Soc. 121, 2857–2862 (1999)CrossRefGoogle Scholar
- 19.Steen, H., Jebanathirajah, J.A., Rush, J., Morrice, N., Kirschner, M.W.: Phosphorylation analysis by mass spectrometry: myths, facts, and the consequences for qualitative and quantitative measurements. Mol. Cell. Proteomics. 5, 172–181 (2006)Google Scholar
- 20.Sweet, S.M.M., Bailey, C.M., Cunningham, D.L., Heath, J.K., Cooper, H.J.: Large scale localization of protein phosphorylation by use of electron capture dissociation mass spectrometry. Mol. Cell. Proteomics. 8, 904–912 (2009)Google Scholar
- 23.Zhang, J., Guy, M.J., Norman, H.S., Chen, Y.C., Xu, Q.G., Dong, X.T., Guner, H., Wang, S.J., Kohmoto, T., Young, K.H., Moss, R.L., Ge, Y.: Top-down quantitative proteomics identified phosphorylation of cardiac troponin I as a candidate biomarker for chronic heart failure. J. Proteome Res. 10, 4054–4065 (2011)CrossRefGoogle Scholar
- 25.Ayaz-Guner, S., Zhang, J., Li, L., Walker, J.W., Ge, Y.: In vivo phosphorylation site mapping in mouse cardiac troponin I by high resolution top-down electron capture dissociation mass spectrometry: Ser22/23 are the only sites basally phosphorylated. Biochemistry 48, 8161–8170 (2009)CrossRefGoogle Scholar
- 29.Peng, Y., Gregorich, Z.R., Valeja, S.G., Zhang, H., Cai, W.X., Chen, Y.C., Guner, H., Chen, A.J., Schwahn, D.J., Hacker, T.A., Liu, X.W., Ge, Y.: Top-down proteomics reveals concerted reductions in myofilament and z-disc protein phosphorylation after acute myocardial infarction. Mol. Cell. Proteomics. 13, 2752–2764 (2014)Google Scholar
- 35.Xu, F.M., Xu, Q.G., Dong, X.T., Guy, M., Guner, H., Hacker, T.A., Ge, Y.: Top-down high-resolution electron capture dissociation mass spectrometry for comprehensive characterization of post-translational modifications in rhesus monkey cardiac troponin I. Int. J. Mass Spectrom. 305, 95–102 (2011)CrossRefGoogle Scholar
- 37.Cai, W.X., Guner, H., Gregorich, Z.R., Chen, A.J., Ayaz-Guner, S., Peng, Y., Valeja, S.G., Liu, X.W., Ge, Y.: MASH Suite Pro: a comprehensive software tool for top-down proteomics. Mol. Cell. Proteomics. 15, 703–714 (2016)Google Scholar