Approach for Determining Protein Ubiquitination Sites by MALDI-TOF

This provides a platform for mapping ubiquitination sites using mass ... sulfonated tryptic peptides can provide a highly effective method for the det...
0 downloads 0 Views 311KB Size
Download PDF
Anal. Chem. 2005, 77, 1458-1466

Approach for Determining Protein Ubiquitination Sites by MALDI-TOF Mass Spectrometry Dongxia Wang and Robert J. Cotter*

Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205

Protein ubiquitination plays an important role in the degradation and other functional regulation of cellular proteins in organisms ranging from yeasts to mammals. Trypsin digestion of ubiquitin conjugated proteins produces diglycine branched peptides in which the C-terminal Gly-Gly fragment of ubiquitin is attached to the E-amino group of a modified lysine residue within the peptide. This provides a platform for mapping ubiquitination sites using mass spectrometry. Here we report the development of a novel strategy for determining posttraslational protein ubiquitination based on the N-terminal sulfonation of diglycine branched peptides. In contrast to conventional tandem MS spectra of native tryptic peptides, MALDI MS/MS analysis of a sulfonated tryptic peptide containing a diglycine branch generates a unique spectrum composed of a signature portion and a sequence portion. The signature portion of the spectrum consists of several intense ions resulting from the elimination of the tags, the N-terminal residues at the peptide and the branch, and their combination. This unique ion distribution pattern can distinguish ubiquitination modificatons from others and can identify the first N-terminal residues of the peptides as well. The sequence portion consists of an exclusive series of y-type ions and y′ ions (differing by the loss of one glycine residue from the sulfonated diglycine branch) that can directly reveal the amino acid sequence of the peptide and the precise location of the ubiquitination site. The technique is demonstrated for a series of synthetic peptides and is validated by a model protein, tetraubiquitin. Our results show that the MALDI MS/MS analysis of sulfonated tryptic peptides can provide a highly effective method for the determination of ubiquitination substrates, ubiquitination sites on protein targets, and modification sites on ubiquitins themselves. Ubiquitin is a 76-amino acid globular protein that can modify target proteins via a covalent attachment through an isopeptide bond between the C-terminal glycine residue of ubiquitin and the -amino group of the lysine residue on a target protein. Like other posttranslational modifications, ubiquitination has recently emerged as an important regulation mechanism for the degradation and functional regulation of cellular proteins in organisms ranging from * To whom correspondence should be addressed. E-mail: [email protected].

1458 Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

yeasts to mammals.1-4 A cellular protein is attached to ubiquitin by a multienzymatic system consisting of activating (E1), conjugating (E2), and ligating (E3) enzymes. While a target protein can be modified by a single ubiquitin molecule, the substrate can become tagged by a polyubiquitin chain where one ubiquitin is conjugated to another through lysine-glycine isopeptide bonds. A vast majority of misfolded or short-lived proteins are degraded by the ubiquitin-proteasome pathway. Information to date has revealed that protein ubiquitination not only leads to protein degradation but also participates in many aspects of cellular signaling pathways such as protein sorting, membrane trafficking, gene expression, DNA repair, IKK activation, and endocytosis. Significant technical advances are needed in the determination of ubiquitination substrates, ubiquitination sites on cellular protein target, and modification sites on ubiquitins themselves. In the traditional mutagenesis approach for determining ubiquitination, mutant forms of substrate, ubiquitin, or both are used to address which of the lysine residues are used in the modification or polyubiquitin chain assembly. This is an indirect method and is usually labor-intensive. Ubiquitin has a C-terminal Arg-Gly-Gly moiety and is covalently attached to its target through an isopeptide bond between the carboxy group of its last C-terminal Gly residue and the -amino group of a modified lysine residue. Trypsin digestion of ubiquitin conjugates results in cleavage of the Arg-Gly bond in ubiquitin, resulting in uniquely tagged peptides in which the two terminal glycine residues from ubiquitin remain attached to the target lysine residue in a digested peptide derived from substrates (Scheme 1A). The formation of these branched peptides provides a platform for the direct determination of ubiquitination by mass spectrometry because the diglycine branch can be simply viewed as a small modifier with a mass addition of 114 Da during MS/MS sequencing. Recently, Gygi and co-workers reported a proteomic approach to detecting protein ubiquitination.5 Using an approach combining affinity protein purification and tandem mass spectrometry analysis for tryptic peptides digested from ubiquitin-conjugated proteins, they have identified 1075 candidate ubiquitin-modified proteins from yeast cells. In addition, they have detected 110 ubiquitination sites from 72 ubiquitin-tagged proteins. (1) Finley, D.; Ciechanover, A.; Varshavsky, A. Cell 2004, S116, S29-S32. (2) Pickart, C. M. Cell 2004, 116, 181-190. (3) Weissman, A. M. Nat. Rev. Mol. Cell Biol. 2001, 2, 169-178 (4) Glickman, M. H.; Ciechanover, A. Physiol. Rev. 2002, 82, 373-428 (5) Peng, J.; Schwartz, D.; Elias, J. E.; Thoreen, C. C.; Cheng, D.; Marsischky, G.; Roelofs, J.; Finley, D; Gygi, S. P. Nat. Biotechnol. 2003, 21, 921-926. 10.1021/ac048834d CCC: $30.25

© 2005 American Chemical Society Published on Web 02/03/2005

Scheme 1. (a) Trypsin Digestion of an Ubiquitinated Protein and (b) the Sequences of the Synthetic Linear and Branched Peptides Used in This Study

Interpretation of the MS/MS spectra of native peptides is often complicated by the complexity of the resulting fragmentation pattern, which can include product ions from one or more of the different N-terminal and C-terminal product ion series. To direct peptide fragmentation toward a single type of product ion and simplify the fragmentation spectra, a novel derivatization strategy, N-terminal sulfonation, was developed recently by Keough and co-workers.6 In this method, a sulfonic acid group was introduced at the N-terminus of a peptide. This leads to the formation (almost exclusively) of a series of y-series ions in the positive ion mode, as b-series (and other N-terminal) products are neutralized or negatively charged by the presence of the sulfonic acid group. The amino acid sequence can then be obtained by the mass differences between adjacent y ions, enabling sequences (rather than masses) to be used in database searching. While the sulfonation reagent developed by this group has been commercialized and is available to investigators, another reagent, 4-sulfophenyl isothiocyanate (SPITC), has been used by several research groups for N-terminal sulfonation,7-10 and we recently reported a modification to the reaction utilizing this reagent that enables the reaction to be carried out more efficiently and in aqueous solution.8 Here we present the development of a novel mass spectrometric strategy for the determination and site mapping of posttranslational ubiquitination in proteins. In this study, a series of Gly-Gly branch-containing peptides (similar to those that would be obtained from the tryptic digest of a ubiquitinated protein) are first derivatized by N-terminal sulfonation using SPITC and then subjected to MS/MS analysis using a MALDI-TOF mass spectrometer. Fragmentation of these peptide derivatives produces a unique product ion spectrum consisting of two distinct regions: a group of high-mass and high-abundant product ion peaks that serve as a signature portion to distinguish branched peptides from others, and a group of y-series ion peaks that act as sequence (6) Keough, T.; Youngquist, R. S.; Lacey, M. P. Anal. Chem. 2003, 75, 157A165A. (7) Marekov, L. N.; Steinert, P. M. J. Mass Spectrom. 2003, 38, 373-377 (8) Wang, D.; Kalb, S. R.; Cotter, R. J. Rapid Commun. Mass Spectrom. 2004, 18, 96-102 (9) Chen, P.; Nie, S.; Mi, W.; Wang, C.-C.; Liang S.-P. Rapid Commun. Mass Spectrom. 2004, 18, 191-198 (10) Lee, Y. H.; Kim, M.-S.; Choie, W.-S.; Min, H.-K.; Lee, S.-W. Proteomics 2004, 4, 1684-94

portion for amino acid sequencing and ubiquitination site mapping. In this way, the peptides containing ubiquitination sites can be readily identified and sequenced. EXPERIMENTAL SECTION Materials. All chemicals used in this study were of analytical grade. SPITC, O-methyl isourea, sodium bicarbonate, and ammonium bicarbonate were purchased from Sigma (St. Louis, MO). Bovine pancreas modified trypsin was supplied by Roche Diagnostics Corp. (Indianapolis, IN). R-Cyano-4-hydroxycinnamic acid was from Aldrich (Milwaukee, WI). Tetraubiquitin (Ub4) was purchased from Affinity Research Products, Ltd. (Exeter, U.K.). All synthetic peptides were synthesized and purified by the Synthesis & Sequencing Facility at the Johns Hopkins University School of Medicine (Baltimore, MD). Sulfonation. The reagent solution was prepared by dissolving SPITC (10 mg/mL) in 20 mM NaHCO3 (pH ∼9.0). The sulfonation reaction was carried out in a 0.6-mL Eppendorf tube by mixing 9 µL of reagent solution with 1 µL of peptide solution (∼10-100 pmol). After incubation for 30 min at 55 °C, the reaction was terminated by adding 1 µL of 1% trifluoracetic acid (TFA). The sample was then loaded onto a micropipet tip (C18 OMIX, Varian, Lake Forest, CA), washed with 3 × 10 µL of 0.1%TFA, and followed by eluting with 10 µL of 75% acetonitrile/0.1% TFA. The solution was taken to dryness by SpeedVac and then resuspended with 10 µL of ddH2O. For tetraubiquitin, 4 µg of the protein was digested with trypsin (50:1 w/w) in 20 µL of reaction solution containing 25 mM ammonia bicarbonate at 37 °C for 18 h. A 10µL aliquot of the reaction was then mixed with 5 µL of SPITC solution and underwent derivatization following the same procedure described. Guanidination. The reagent, O-methyl isourea, was freshly prepared in ddH2O to the concentration of 1 mg/µL. The reaction was conducted in a mixture of 2 µL of peptide (∼100 pmol), 1.5 µL of the reagent solution, and 5.5 µL of ammonia hydroxide (30%) at 55 °C for 30 min. The reaction was terminated with 10 µL of 10% TFA and followed by purification using micropipet tips as described. Mass Spectrometry. All MS and MS/MS spectra were acquired in the positive ion mode using a Kratos Analytical (Manchester, U.K.) AXIMA-CFR MALDI-TOF high-performance mass spectrometer equipped with a pulsed extraction source, a 337-nm pulsed nitrogen laser, and a curved-field reflectron. The acceleration voltage was 20 kV. The matrix solution was prepared by dissolving R-cyano-4-hydroxycinnamic acid (10 µg/µL) in 50% acetonitrile containing 0.1% TFA. A thin layer of the matrix was applied onto the sample plate first, followed by the addition of 0.5 µL of sample and 0.5 µL of matrix, and the resultant mixture was allowed to dry at room temperature. RESULTS AND DISCUSSION The Gly-Gly branched peptide, LIFAGK(GG-)QLEDGR (Scheme 1, Pep1), was prepared by solid-phase peptide synthesis. The sequence of this peptide corresponds to that of a tryptic ubiquitin peptide (Ub43-54) containing a ubiquitination site at its K48 residue, a typical modification site in polyubiquitin chains. The linear part of this peptide was used as a template for the other peptides in the present study, which differed from this sequence only in the number and positions of lysine residues. The peptide was Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

1459

Figure 1. MALDI MS/MS spectrum of the SPITC sulfonation derivative of the peptide, LIFAGK(GG-)QLEDGR. The explanations for the letter labels above some peaks are listed in Scheme 2b.

derivatized by N-terminal sulfonation using the SPITC reagent, and resulted in the addition of two sulfonic acid groups as tags on the amino termini of both the peptide and the Gly-Gly branch. The MS/MS spectrum of this peptide derivative provided several unique features that are illustrated in Figure 1. In that spectrum, two groups of fragment ions are observed. The sequence portion consists of a somewhat lower intensity series of peaks (m/z 232.31234.5) corresponding to the sequence-specific product ions y2y10. In this series, the site of the diglycine modified lysine residue can be precisely determined by the mass difference between the y6 and y7 ions. The signature portion in the upper mass range of the spectrum (m/z 1290.7-1718.1) consists of several higher intensity peaks, four of which can be assigned to fragments formed by the elimination of one or two sulfonation tags. In particular, our previous work8 has shown that SPITC-derivatized peptides produce two characteristic peaks: a more abundant peak corresponding to the release of the entire sulfonation tag (HO3S-C6H4NCS group, 215 Da) from the peptide and a less abundant peak corresponding to only partial loss of the tag molecule (HO3SC6H4-NH2 group, 173 Da). Thus, the peaks at m/z 1676.0 (P1) and 1461.1 (P2) correspond to the loss of one or two sulfonation tags, respectively. The peaks at m/z 1718.1 (P1p) and 1503 (P2p) correspond to the partial loss of a sulfonation tag, without and with an additional loss of an additional tag. A smaller peak at m/z 1544.9 corresponds to the loss of two partial tags. The remaining peaks in this region can be attributed to the losses of one or more tags along with an adjacent amino acid residue. For example, the peak at m/z 1618.9 (P1′) corresponds to the loss of one tag plus a glycine residue (differing from 1676.0 by 57.1 mass units). The peak at m/z 1562.8 (P1′′) corresponds to the loss of one tag plus a leucine residue (differing from 1676.0 by 113.2 mass units). These peaks, of course, identify which of the tags is lost. In addition, because of its closeness in mass, we also considered the possibility that the peak at m/z 1562.8 might include some unresolved loss of a tag plus two glycine residues. The mass spectra of Pep2 and Pep3 in Figure 2a and b, 1460

Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

respectively, show that this is not the case. The sequence tagGLIFAGAQLEDGR produces the loss of the tag plus glycine (but not GL), and the sequence tag-GGIFAGAQLEDGR shows a loss of the tag plus glycine, but not GG). Furthermore, the branched peptide sequence tag-LIFAGK(tag-GA-)QLEDGR, in which the GG modification is changed to GA, showed losses of one or two tag moieties plus leucine, glycine, or both, but not loss of GA (Figure 2c). Thus, in Figure 1 the peak at m/z 1404.1 corresponds to the loss of two tags and a glycine, the peak at m/z 1347.9 corresponds to the loss of two tags and a leucine, and the peak at m/z 1290.7 corresponds to the loss of two tags, a glycine and a leucine. Note that the peak at m/z 1347.9 corresponds to the loss of tag-L from the sequence and is therefore the y11 ion. The peak at m/z 1290.7, which we have designated as y′11, is similar except that there is an additional loss of glycine on the branch. A similar pair of sequence ions is observed at m/z 1234.5 (y10) and 1177.5 (y′10). Gaskell and co-workers11,12 investigated the fragmentation of the N-terminal amino acid in the gas phase promoted by collisionactivation dissociation of a peptide derivatized on its N-terminus by the Edman reagent, phenylisothiocyanate (PITC). The formation of abundant complementary b1 and yn-1 ion pairs resulting from the dissociation of doubly charged [M + 2H]2+ ions formed by electrospray was proposed to follow a mechanism similar to that of condensed-phase Edman degradation. That is, cleavage of the first peptide bond is promoted by nucleophilic attack by the thiocarbonyl moiety of the tag on the carbonyl group of the N-terminal peptide bond, and protonation by the second proton, while the strong nucleophilic character of the thiocarbonyl group contributes to the preferred fragmentation. The dominant formation of yn-1 ions has also been observed in the fragmentation of SPITC-derivatized peptides.8,10 SPITC attaches to the peptide by (11) Summerfield, S. G.; Bolgar, M. S.; Gaskell, S. J. J. Mass Spectrom. 1997, 32, 225-231. (12) van der Rest, G.; He, F.; Emmett, M. R.; Marshall, A. G.; Gaskell, S. J. J. Am. Soc. Mass Spectrom. 2001, 12, 288-295.

Figure 2. MS/MS spectra of the sulfonation derivatives of the linear peptides: (a) GLIFAGKQLEDGR, (b) GGLIFAGKQLEDGR, and (c) the branched peptide, LIFAGK(GA-)QLEDGR.

Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

1461

Figure 3. MS/MS spectrum of the sulfonation derivative of the linear peptide, LIFAGKQLEDGR, containing an internal unmodified lysine residue.

Scheme 2. (a) Proposed Mechanism for Edman Cleavage of Doubly Tagged Peptide Containing a Diglycine Branch, by N-terminal Sulfonation and (b) Possible Fragments Formed from Branched Peptide Derivatives during MS/MS Process

the same chemistry as the PITC reagent, but the labile proton from the sulfonic acid group of SPITC derivatives might serve as a mobile proton and promote the preferential cleavage of the first 1462

Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

peptide bond from the singly charged ions normally generated by MALDI mass spectrometry. This is consistent with our observation of abundant peaks formed by the removal of N-

Figure 4. Effect of the diglycine branch positions on the fragmentation of sulfonated peptide derivatives: (a) LK(GG-)FAGAQLEDGR, (b) LIFK(GG-)GAQLEDGR, (c) LIFAGAQLK(GG-)DGR, and (d) LIFAGAQLEDK(GG-)R.

Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

1463

Figure 5. MS/MS spectra of sulfonation derivatives of (a) the peptide, LK(GG-)FAGK(GG-)QLEDGR, containing two Gly-Gly branches, and (b) the peptide, LIFAGK(GG-)QLEDGK, containing a branch and a C-terminal lysine, but this residue was converted to homoarginine by guanidination reaction prior to sulfonation.

terminal residues from derivatized peptides and suggests that the same mechanism of Edman-type cleavage should occur on these doubly tagged SPITC derivatives (Scheme 2a). Note that the signature portion of the spectrum (peaks P1, P2, P1p, P2p, P1′, P2′, P1′′, and y11, identified in Scheme 2b) identifies the first N-terminal amino acids and distinguishes sulfonated GlyGly branched peptides from other singly tagged linear peptides, and from doubly tagged linear peptides containing an internal unprotected lysine residue (resulting from a missed cleavage) that can be sulfonated on its free -amino group. As illustrated in Figure 3, the lack of a second N-terminal amino acid residue results in a product ion cluster that displays the losses of one and two tag molecules (m/z 1676.2 and 1461.3), one and two partial tags (m/z 1718.2 and 1503.2), but only one and two tags with the N-terminal residue (glycine in this case yielding m/z 1619.1 and 1404.1, which begin the y-series ions). Note also (in Figure 3) that the absence of N-terminal residues on the branch results in only a single y-series of ions, and the location of the sulfonated lysine is not identified from the sequence portion of the spectrum. It is also 1464

Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

possible, of course, to protect lysine residues prior to sulfonation, and this is described below. To examine the effect of the position in the sequence of a tagged-GG-lysine residue on fragmentation, we synthesized a set of peptides (Pep6-Pep9) using the native ubiquitin peptide (Pep1) as a template. Specifically, we substituted the residues at positions 2, 4, 9, or 11 with GG-branched lysine and replaced the lysines at position 6 with alanines to eliminate an unprotected -amino group. The MS/MS spectra of these peptide derivatives after SPITC sulfonation are shown in Figure 4. Not surprisingly, the signature portions of these spectra are similar, showing the losses of one and two tags, along with their two proximal N-terminal residues. Except for the differences in masses of these proximal residues, the signature portion is in effect independent of the position and sequence. In the sequence portion of the spectra, the ion masses do, of course, reflect the differences in the sequence, and the position of the tag-GG modified lysine is easily determined. In addition, there are again two sets of y-series ions for all those that include

Figure 6. MALDI MS spectra of the peptide mixture, digested from tetraubiquitin by trypsin, (a) before and (b) after N-terminal sulfonation. Derivatized peptide are labeled by asterisks. Insets are the MS/MS spectra of the derivatized and underivatized peptide43-54 containing a GlyGly branch modified lysine residue (K48).

the modified lysine residue. For example, the peptide tag-LK(tagGG-)FAGAQLEDGR (Figure 4a), in which the tagged lysine is on the second residue, shows a pair of ions at m/z 1305.4 (y11) and 1249.1 (y′11) as seen for Pep1, but only the single series of ions below the modified site. The peptide tag-LIFK(tag-GG-)GAQLEDGR (Figure 4b), in which the tagged lysine is on the fourth residue, shows two y-series ions from y8/y′8 to y11/y′11. For the peptide tag-LIFAGAQLK(tag-GG-)DGR, the series extends from y4/y′4 to y11/y′11, and for the peptide tag-LIFAGAQLEDK(tag-GG-)R, it extends from y2/y′2 to y11/y′11. These series of ions differ in mass by 57.1 mass units, corresponding to the mass of the N-terminal glycine on the peptide branch. Multiple ubiquitination sites can occur within a single tryptic peptide obtained from a protein modified by more than one ubiquitin or ubiquitin chain. Indeed, in Gygi’s pioneering work,5 several tryptic peptides were found to contain two ubiquitination sites. To investigate the application of the sulfonation method to multiply ubiquitinated peptides, the synthetic peptide, LK(GG-)FAGK(GG-)QLEDGR (Pep 10), was prepared. After sulfonation,

peptide derivatives having two or three tags were detected by the mass spectrometer but the quality of the positive ion MS/MS spectrum of the triply tagged derivative was poor (data not shown), presumably due to the presence of three sulfonic acid groups. Interestingly, the doubly tagged peptide gave rise to a nice fragmentation spectrum (Figure 5a) in which two distinct groups of the fragment ions were clearly displayed, similar to the ones in the spectra described above. The presence of the y′11 ion at m/z 1419.9 implies that one of the two tags was attached to the branch at the position 2 while another one was on the N-terminal of the peptide. These data suggest that peptides with multiple modifications can be analyzed through incomplete derivatization. Tryptic digestion generally produces peptides containing arginine or lysine on their C-termini. Although experimental conditions for the sulfonation reaction with SPITC can be controlled to restrict labeling to the N-terminus of a peptide with a C-terminal lysine,10 protection of free (untagged) lysine residue may be necessary because of the lower ionization efficiency of Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

1465

lysine-terminated peptides than arginine-terminated peptides.13 Guanidination is a widely used method for modifying lysineterminated tryptic peptides that converts the lysine side chain to homoarginine. The guanidinium group increases the basicity of the modified lysine side chain and so is used to increase sensitivity in the positive ion mode.14,15 To test the effect of guanidination on the fragmentation of a sulfonated derivative, a branched peptide, LIFAGK(GG-)QLEDGK, with a C-terminal lysine residue was prepared (Pep11). Guanidination of the peptide was followed by N-terminal sulfonation of the derivative having homoarginine at its C-terminus. The fragmentation pattern (Figure 5b) similar to that of the Pep1 derivative indicates that derivatization by guanidination prior to sulfonation did not impact the fragmentation of derivatized peptides. It should be noted that guanidination can also occur at the N-terminal amino group when that is a glycine, and therefore, the diglycine branch would also be modified if guanidination is carried out on tryptic peptides. However, this problem can be overcome by applying this reaction directly to intact ubiquitinated proteins prior to trypsin digestion. Finally, our method was applied to a model ubiquitinated protein: tetraubiquitin, in which each ubiquitin molecule is conjugated to the next through its K48 residue. After digestion by trypsin, the peptides derived from this protein conjugate were derivatized by N-terminal sulfonation using SPITC as described and were then subjected to mass spectrometric analysis without further chromatography separation. Figure 6 shows the mass spectra of the samples before and after sulfonation. The peptide covering the ubiquitin sequence 43-54 and containing a ubiquitination site at K48 was easily identified by the MS/MS analysis of the peptide at m/z 1890.9 (Figure 6a, inset). In contrast, the MS/MS spectrum (Figure 6b, inset) of its underivatized counterpart contains fragment ions from several series and cannot be as easily interpreted. CONCLUSIONS Here we have extended our N-terminal sulfonation procedure for the analysis of ubiquitinated proteins. MALDI MS/MS analysis (13) Krause, E.; Wenschuh, H.; Jungblut, P. R. Anal. Chem. 1999, 71, 41604165 (14) Brancia, F. L.; Oliver, S. G.; Gaskell, S. J. Rapid Commun. Mass Spectrom. 2000, 14, 2070-2073 (15) Beardsley, R. J.; Reilly, J. P. Anal. Chem. 2002, 74, 1884-1890.

1466

Analytical Chemistry, Vol. 77, No. 5, March 1, 2005

of the SPITC-derivatized Gly-Gly branched peptides produces a unique fragmentation distribution pattern that consists of a signature group and a sequence group of product ions. In addition to providing y-series ions exclusively and simplifying the determination of sequence, the resulting MS/MS spectra provide peaks that determine the number of tagged N-terminal peptides/ branches, the N-terminal residue for each, and the location of ubiquitination site. The latter is also distinguished by the presence of multiple y-series ions for those sequence-specific fragment ions that include the Gly-Gly branch. This strategy is being used currently to map the ubiquitination sites in several ubiquitin-modified proteins, and to date, we have successfully identified a previously unknown site on a ubiquitinmodified protein (manuscript in preparation). We are also developing similar strategies for elucidation of modifications to proteins by small ubiquitin-related modifier (SUMO) and other ubiquitinlike proteins (Ubls). Among nine families of Ubls identified to date, none of the mature form of these protein modifiers except Nedd contain the RGG moiety at their C-termini although the last two residues are always glycines. Branched peptides resulting from these modification sites can be generated with appropriate proteases, but the branches may have more than two amino acid residues. While this makes these peptides more difficult to characterize by MS/MS analysis of the native peptides, our approach can be used in these cases to provide similar signature peaks and simplify the sequence portion to enable the determination of targets for these Ubls modifications. ACKNOWLEDGMENT The authors acknowledge the helpful discussions and the contribution of the tetraubiquitin from Wanping Xuand Len Neckers, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Rockville, MD 20805. This work was supported by a contract N01 HV28180 from the National Heart Lung and Blood Institute (Jennifer Van Eyk, PI) and a grant U54 RR020839 from the National Institutes of Health (Jef Boeke, PI). Mass spectral analyses were carried out at the Middle Atlantic Mass Spectrometry Laboratory. Received for review August 6, 2004. Accepted December 7, 2004. AC048834D