Anal Bioanal Chem (2010) 397:3209–3212 DOI 10.1007/s00216-009-3372-x

TRENDS

Stepchild phosphohistidine: acid-labile phosphorylation becomes accessible by functional proteomics Ulli Martin Hohenester & Katrin Ludwig & Josef Krieglstein & Simone König

Received: 27 October 2009 / Revised: 30 November 2009 / Accepted: 30 November 2009 / Published online: 10 January 2010 # Springer-Verlag 2010

Abstract Bioanalytical techniques were preferentially developed for the investigation of phosphohydroxyamino acids in the past and there is a wealth of information on the detection of serine, threonine and tyrosine phosphorylation in functional proteomics. However, similarly important for protein regulation and signalling is the phosphorylation of other amino acids such as histidine, but its detection is hampered by the sensitivity to acid. Mass spectrometry in conjunction with chromatographic methods is allowing us to start to get a handle on phoshohistidine. 32P-labelling and amino acid analysis for phosphorylation site determination is increasingly complemented by typical proteomic approaches based on reversed-phase peptide separation and gas-phase fragmentation. Chemical phosphorylation of peptides is a valuable tool, therefore, for the generation of analytical standards for use in method development. Keywords Phosphohistidine . Protein phosphorylation . Fragmentation . Mass spectrometry . Chromatography

U. M. Hohenester : S. König (*) Integrierte Funktionelle Genomik, Interdisziplinäres Zentrum für Klinische Forschung, Westfälische Wilhelms-Universität Münster, Roentgenstraße 21, 48149 Münster, Germany e-mail: [email protected] URL: http://ifg-izkf.uni-muenster.de K. Ludwig : J. Krieglstein Institut für Pharmazeutische und Medizinische Chemie, Westfälische Wilhelms-Universität Münster, Roentgenstraße 21, 48149 Münster, Germany

Abbreviations CID collision-induced dissociation HPLC high-performance liquid chromatography LC liquid chromatography MS mass spectrometry MS/MS tandem mass spectroemtry pHis phosphohistidine

Owing to its enormous importance in signal transduction and energy transfer [1], reversible protein phosphorylation has been the subject of mass spectrometry (MS)-based analyses for many years [2, 3]. However, mainly the hydroxyamino acids Ser, Thr and Tyr have been studied [2–4], although kinases and phosphatases that act on other amino acids such as His [5], Arg [6], Asp [7], Cys [8], Glu [9] and Lys [10] have also been described. This was partially due to the better accessibility of phosphomonoesters by typical analytical handling procedures, whereas phosphoamidates and acylphosphates have been shown to be labile in acidic solvents [11]. Phosphohistidine (pHis) is estimated to be 10–100-fold more abundant in nature than phosphotyrosine [12] but its bioanalytical exploration lags behind. As reviewed by Besant and Attwood [11] in 2009, radiolabelling methods and phosphoamino acid analysis are still an integral part of analysis strategies for the investigation of His phosphorylation. Nevertheless, specific purification techniques known from peptide phosphoester analysis such as ion metal affinity chromatography are increasingly being transferred to pHiscontaining peptide preparation [13]. MS is also pushing into the field, but it has had its difficulties. Initial studies using common tandem MS (MS/MS) noted the ready loss of the phosphate group from pHis-containing peptides, a fact which effectively prevented the generation of phosphate-containing

3210

U.M. Hohenester et al.

Fig. 1 Reversed-phase separation of chemically phosphorylated angiotensin II forms containing 1phosphohistidine (1-pHis), 3-phosphohistidine (3-pHis) and 1,3diphosphohistidine (1,3-pHis) (for the structures, see Fig. 3). The total ion chromatogram is shown in the upper-right corner; the corresponding retention times (RT) are given in the respective tandem mass spectrometry (MS/MS) spectra of the doubly charged peptides. The major peptide (middle spectrum) exhibits phosphatecontaining y and b ions as well as the histidine phosphoimmonium ion [19]

fragment ions [14–16]. Conclusively, phosphorylation site assignment using widely applied techniques of proteomic MS based on collision-induced dissociation (CID) seemed not to be very promising. More elaborate fragmentation techniques appeared to be necessary to obtain structural information. Medzihradszky et al. [14] resorted to dedicated matrix-assisted laser desorption/ionization–high-energy CID experiments and detected phosphate-containing b and y ions for peptide AcSFTNPLpHSAAW-NH2 in 1997. Electron capture dissociation, electron detachment dissociation and electron transfer dissociation proved to be soft enough to retain the phosphate group on backbone fragments for some peptides 10 years later [15, 16]. Recently, Zu et al. [17] demonstrated the complex investigation of histone tryptic peptides involving negative ion MS and precursor ion scanning for the loss of PO3−, as well as positive ion MS and MS/MS. As a side note, the

method using an inductively coupled plasma (ICP) which targets only the phosphorus and is destructive for the peptide is an interesting alternative to classic MS/MS and it was successful in finding the phosphorylation site of CheA [18]. However, the low success of MS in pHis-containing peptide analysis so far may have simply been due to the physicochemical and structural properties of the peptides investigated. A comprehensive study with synthetically His phosphorylated peptides [19] clearly showed that it is possible to obtain site-specific information by MS fragmentation analysis using both manual nanoMS/MS and low-pH NH2

O

NH2

NH2

O

H HO

HO

HO N

NH N

N

histidine

O

NH N

OH

1-phosphohistidine

NH2

O

P

histidine immonium ion +

NH2

O

NH2 H

HO

HO N

O

N

N P

OH

3-phosphohistidine

Fig. 2 Chemical phosphorylation of histidine-containing peptides using potassium phosphoramidate [6]. Since it is not commercially available, the agent needs to be synthesized first using phosphoryl chloride and a base. Potassium phosphoramidate crystals are stable in a desiccator for several months and aqueous solutions phosphorylate peptides in yields up to 80% [14]

+

O

P HO

HO

HO

O

N

P

OH

O

-

1,3-diphosphohistidine

N

O

N P

HO

OH

3-phosphohistidine immonium ion

Fig. 3 Structures of histidine and phosphohistidine and their immonium ions. Histidine can carry a phosphate on two possible sites. 1-pHis and 3-pHis are biologically relevant, whereas there are no reports to this effect of 1,3-pHis [11]

Stepchild phosphohistidine: acid-labile phosphorylation becomes accessible by functional proteomics

nano liquid chromatography (LC)–MS/MS with CID in an ion trap or a quadrupole time of flight analyser as is exemplified for the angiotensin II mono-His-phosphorylated and di-His-phosphorylated forms in Fig. 1. This is also true for the substrates of protein His phosphatase peptide AHPF [20] or the CheA peptide IFRAAHSIKGG [21] which were treated in the same manner [19]. Interestingly, the immonium ion of pHis in addition to that of His can be observed in the lower mass region of the CID spectra, providing strong evidence for the presence of a phosphorylation site. Despite these positive results, a considerable dependence of the phosphate neutral loss on peptide sequence, size and charge was noted, so generic analysis protocols can hardly be established. Thus, the requirements for MS-based pHiscontaining peptide analysis are not so unlike those for Ser/ Thr/Tyr phosphopeptides. Although phosphopeptides are occasionally found in routine LC-MS/MS runs, more often they are not. Dedicated analyses involve, e.g., the specific purification of phosphopeptides to reduce suppression effects in complex protein digests and the variation of the cleavage enzyme to generate peptides suitable for the detection window (e.g., peptides of molecular weights between 1,000 and 3,000 Da). It has also been discussed that the location of the phosphorylation site within the peptide may be of importance for ionization and fragmentation. A summary of the current state of the art has recently become available [22]. A major tool for peptide separation before MS is reversed-phase high-performance LC (HPLC) and it is often used in conjunction with mass spectrometers. For online electrospray ionization experiments of pHiscontaining peptides, Kleinnijenhuis et al. [15] recommended either non-acidic solvents or fast LC methods to accommodate the acid lability of these compounds. However, in tests of a number of various synthetic His-phosphorylated peptides [19], it was observed that pHis-containing peptides are sufficiently stable during purification procedures and measurement at low pH to allow functional investigations. Both His-phosphorylated proteins and peptides can be measured at low pH in positive ion mode MS and the halflives—at least of the ones investigated so far—are long enough (several hours) to comfortably perform the experiments. Being able to investigate pHis-containing peptides with some of the standard proteomic procedures is important in particular in laboratory settings such as core facilities where it is not possible to change instrumental parameters often. For routine tasks the instruments need to be on call, so, e.g., exchanging the solvents for HPLC runs from acidic to neutral solvents is not an option. Reversed-phase HPLCMS/MS allows the separation and fragmentation of pHiscontaining peptides in data-dependent scans and MS-based techniques will increasingly be employed for that purpose. Nevertheless, as discussed above, the preparation of the

3211

phosphopeptides before MS remains the limiting step for any type of phosphorylation analysis. For testing purposes and method validation, chemical phosphorylation using potassium phosphoramidate (PPA) is recommended (Fig. 2) [23]. Potassium phosphoramidate [24] is described as a mild phosphorylating reagent which does not modify the hydroxyamino acids [14, 25, 26]. Its use generates 1,3-pHis (Fig. 3), but mostly with low yield [19]. For the H18-histone peptide, the authors observed the formation of about 20% 1,3-pHis and its formation was also proven by NMR experiments [27]. 3-pHis is preferentially formed, being the more stable isomer [26]. In summary, the structural analysis of pHis-containing proteins and peptides has matured. Although this modification is sensitive to acid, the presence of weak acids as typically used in purification protocols and MS-based analysis does not lead to an immediate loss of the phosphate group. In fact, procedures such as solid-phase extraction and HPLC-MS/MS can be performed with no rush and pHis-containing peptides also can be stored frozen for some time. Moreover, the investigation of pHiscontaining peptides by CID is not so different from that of O-linked phosphopeptides. Phosphate neutral losses are abundant, but phosphate-containing fragments can still be measured. The necessary efforts for this purpose vary with the structural properties of the peptide, the biochemical matrix in which they are delivered and the quantity available. Taken together, phosphorylation analysis will not become an everyday task in the near future and will continue to require dedicated research efforts to be successful. Acknowledgement S.K. is highly indebted to Prof. Susanne Klumpp (died 17 June 2009), who guided her into the field of histidine phosphorylation in a joint project investigating the interaction of ATP with growth factors.

References 1. Paradela A, Albar JP (2008) J Proteome Res 7:1809–1818 2. Boersema PF, Mohammed S, Heck AJR (2009) J Mass Spectrom 44:861–878 3. Zeller M, König S (2004) Anal Bioanal Chem 378:898–909 4. DeGnore JP, Qin J (1998) J Am Soc Mass Spectrom 9:1175–1188 5. Klumpp S, Krieglstein J (2009) Sci Signal 2:pe13 6. Dipolo R, Beaugé L (2006) Physiol Rev 86:155–203 7. Fraser JS, Merlie JP, Echols N, Weisfield SR, Mignot T, Wemmer DE, Zusman DR, Alber T (2007) Mol Microbiol 65:319–332 8. Keyhan NO, Bacia K, Roseman S (2000) J Biol Chem 275:33102–33109 9. Collet JF, Stroobant V, Pirard M, Delpierre G, Van Schaftingen E (1998) J Biol Chem 273:14107–14112 10. Yokoi F, Hiraishi H, Izuhara K (2003) J Biochem 133:607–614 11. Besant PG, Attwood PV (2009) Mol Cell Biochem 329:93–106 12. Klumpp S, Krieglstein J (2002) Eur J Biochem 269:1067–1071 13. Napper S, Kindrachuk J, Olson DJH, Ambrose SJ, Dereniwsky C, Ross ARS (2003) Anal Chem 751741–1747

3212 14. Medzihradszky KF, Phillipps NJ, Senderowicz L, Wang P, Turck CW (1997) Protein Sci 6:1405–1411 15. Kleinnijenhuis AJ, Kjeldsen F, Kallipolitis B, Haselmann KF, Jensen OL (2007) Anal Chem 79:7450–7456 16. Chi A, Huttenhower C, Geer LY, Coon JJ, Syka JEP, Bai DL, Shabanowitz J, Burke DJ, Troyanskaya OG, Hunt DF (2007) Proc Natl Acad Sci USA 104:2193–2198 17. Zu X-L, Besant PG, Imhof A, Attwood PV (2007) Amino Acids 32:347–357 18. Wind M, Wegener A, Kellner R, Lehmann WD (2005) Anal Chem 77:1957–1962 19. Hohenester U-M, Ludwig K, Krieglstein J, König S (2010) Biomacromol Mass Spectrom (in press) 20. Ek P, Pettersson G, Ek B, Gong F, Li JP, Zetterqvist Ö (2002) Eur J Biochem 269:5016–5023

U.M. Hohenester et al. 21. Ludwig K (2008) Thesis, University of Münster 22. Lehman WD (2010) Protein phosphorylation analysis by mass spectrometry: a guide to concepts and practice. Royal Society of Chemistry, Cambridge 23. Hultquist DE, Moyer RW, Boyer PD (1966) Biochemistry 5 (1):322–331 24. Cameron TS, Chan C, Chute WJ (1980) Acta Crystallogr B 36:2391–2393 25. Wei YF, Matthews HR (1991) Methods Enzymol 200:388– 414 26. Besant PG, Byrne L, Thomas G, Attwood PV (1998) Anal Biochem 258:372–375 27. Attwood PV, Ludwig K, Bergander K, Besant PG, Adina-Zada A, Krieglstein J, Klumpp S (2010) Biochem Biophys Acta 1804 (1):199–205