Article pubs.acs.org/ac
Differentiating Sulfopeptide and Phosphopeptide Ions via Resonant Infrared Photodissociation Amanda L. Patrick, Corey N. Stedwell, and Nicolas C. Polfer* Department of Chemistry, University of Florida, P.O. Box 117200, Gainesville, Florida 32611, United States S Supporting Information *
ABSTRACT: The post-translational modifications sulfation and phosphorylation pose special challenges to mass spectral analysis due to their isobaric nature and their lability in the gas phase, as both types of peptides dissociate through similar channels upon collisional activation. Here, we present resonant infrared photodissociation based on diagnostic sulfate and phosphate OH stretches, as a means to differentiate sulfated from phosphorylated peptides within the framework of a mass spectrometry platform. The approach is demonstrated for a number of tyrosine-containing peptides, ranging from dipeptides (YG, pYG, and sYG) over tripeptides (GYR, GpYR, and GsYR), to more biologically relevant enkephalin peptides (YGGFL, pYGGFL, and sYGGFL). In all cases, the diagnostic ranges for sulfate OH stretches are established as 3580−3600 cm−1 and can thus be distinguished from other characteristic hydrogen stretches, such as carboxylic acid OH, alcohol OH, and phosphate OH stretches.
A
tyrosine upon sequencing the peptide.10 Additionally, some attempts at sequencing sulfopeptides by electron-mediated (e.g., electron capture or electron transfer) dissociation methods have been made with some success.11−13 However, while these techniques are promising for sequencing, high mass accuracy and resolution would still be required to determine definitively whether an unknown is a sulfopeptide or a phosphopeptide. An alternative approach to detecting peptide modifications is through spectroscopic techniques.6 For example, synthetic sulfopeptides can be characterized by condensed-phase infrared (IR) spectral measurements based on the presence of the symmetric SO3 stretch (∼1060 cm−1) and the antisymmetric SO3 stretches (∼1230 and 1270 cm−1).14 In the hydrogen stretching region, these condensed-phase IR measurements are more challenging as a result of vibrational band shifting and broadening, primarily due to solvent interactions (e.g., hydrogen bonding).15,16 Conversely, gas-phase IR measurements, which are performed in solvent-free conditions, are expected to provide enhanced spectral resolution and predictability in band position, allowing for more robust and reliable distinction of various OH moieties. The IR spectra of mass-separated ions in mass spectrometers can be obtained via “action” spectroscopy approaches, such as infrared multiple photon dissociation (IRMPD) spectroscopy. In IRMPD spectroscopy, an ion of interest is photodissociated with a
s the analysis of proteins, peptides, and post-translational modifications (PTMs) shift toward mass spectrometry (MS)-based platforms,1−4 it is imperative that MS technologies and innovations adapt to more challenging problems. One such problem involves the detection and analysis of sulfopeptides, which exhibit similar mass spectral behavior to phosphorylated peptides,5,6 potentially confounding correct interpretation. This challenge is 2-fold: (1) sulfotyrosine and phosphotyrosine are often found to be isobaric (with only a 9.5 mDa mass difference) and (2) both undergo similar neutral losses upon collision-induced dissociation (CID). The need for sulfopeptide characterization methods is underscored by the fact that sulfation is thought to be an important modification of tyrosine residues, occurring in Golgi apparatus-containing animal cells.7,8 Thus far, some MS-based methods have been proposed for the differentiation of sulfopeptides and phosphopeptides, but no single best method has yet been determined. Perhaps the most obvious approach to distinguishing between sulfopeptides and phosphopeptides is the use of ultrahigh resolution, high mass accuracy instruments. For example, a 9.4 T Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer was successfully used to differentiate between the phosphorylated and sulfated versions of C-terminal amidated cholecystokinin fragment 26−33.9 However, this dependence on high-resolution instruments is prohibitive to many, as FTICR instruments are very costly and not widely available for routine use. Another approach relies on the chemical modification (i.e., acetylation) of free tyrosine residues (after any phosphate groups have been enzymatically removed, if needed); since sulfate groups are lost quantitatively during CID, the original sulfation site can be readily localized as a free © 2014 American Chemical Society
Received: March 18, 2014 Accepted: May 13, 2014 Published: May 13, 2014 5547
dx.doi.org/10.1021/ac500992f | Anal. Chem. 2014, 86, 5547−5552
Analytical Chemistry
Article
nature of the sulfate modification, sulfopeptides were cleaved under carefully controlled conditions (i.e., 0 °C, 2 h).46 The resin was then removed by filtration and the peptide was precipitated from the cleavage cocktail with ice-cold ethyl ether and collected via centrifugation. The synthesized peptides, along with their respective calculated m/z values ([M + H]+), are summarized in Table 1. All samples were prepared to a
tunable infrared laser by virtue of resonant absorption of multiple IR photons.17−19 While the absorption of multiple photons is subject to some nonlinear processes, a number of studies have shown that IRMPD spectra can serve as analogues of IR absorption spectra.20 IRMPD spectroscopy has been previously implemented in a number of fundamental studies on biomolecular ions related to structural questions, such as elucidating cation binding motifs21−27 and rationalizing pathways in peptide dissociation chemistry.28−36 More recently, the technique has also been applied to answering bioanalytical questions related to distinguishing carbohydrate isomers37−39 and identifying characteristic moieties in peptides by virtue of diagnostic vibrational modes.40,41 In addition, fixed-wavelength IRMPD (e.g., using CO2 lasers) has been implemented successfully in the detection of phosphopeptides.42,43 Here, we present proof-of-concept IRMPD spectroscopy studies to distinguish between sulfo- and phosphopeptides based on characteristic vibrations in the OH stretching region (3500−3700 cm−1). This methodology is demonstrated for a series of tyrosine-containing peptides. As shown for the example of the amino acid variants (Figure 1), three possible
Table 1. Peptides Synthesized and Analyzed in This Study, with Their Corresponding Calculated m/z peptide series
sequence
m/z [M + H]+
dipeptides
YG pYG sYG GYR GpYR GsYR YGGFL pYGGFL sYGGFL
239.1026 319.0690 319.0595 395.2037 475.1701 475.1606 556.2766 636.2429 636.2334
tripeptides
Leu-enkephalin
concentration of approximately 10−4 M in a solvent of 70% methanol/29% water/1% formic acid for MS analysis. As an aside, control experiments indicate that mass spectra can be obtained on our laboratory-constructed system for peptide solutions down to concentrations of approximately 10−7 M. Here, a higher concentration was used to obtain higher-quality reference IRMPD spectra for these proof-of-concept studies. Mass Spectrometry and IRMPD Spectroscopy. All mass spectrometry and IRMPD spectroscopy experiments were performed on a custom-built hybrid mass spectrometer described elsewhere.47 Briefly, singly charged positive ions were generated by electrospray ionization (ESI) using a custom source, mass selected by a quadrupole mass filter (QMF), and trapped in a reduced-pressure quadrupole ion trap (QIT), where they were irradiated with focused IR beams from a tunable optical parametric oscillator/amplifier (OPO/A) (LINOS Photonics OS4000) to induce photodissociation. This OPO provides continuous wave (cw) output at a power between 20 and 50 mW, depending on the wavelength. By focusing the IR beams down to
