Enhanced Separation and Characterization of Deamidated Peptides

Article pubs.acs.org/jpr

Enhanced Separation and Characterization of Deamidated Peptides with RP-ERLIC-Based Multidimensional Chromatography Coupled with Tandem Mass Spectrometry Piliang Hao,† Jingru Qian,† Bamaprasad Dutta,† Esther Sok Hwee Cheow,† Kae Hwan Sim,† Wei Meng,† Sunil S. Adav,† Andrew Alpert,‡ and Siu Kwan Sze*,† †

School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, 637551, Singapore PolyLC Inc., 9151 Rumsey Road, Suite 180, Columbia, Maryland 21045, United States

‡

S Supporting Information *

ABSTRACT: Deamidation of asparaginyl residues in proteins produces a mixture of asparaginyl, n-aspartyl, and isoaspartyl residues, which affects the proteins’ structure, function, and stability. Thus, it is important to identify and quantify the products to evaluate the effects in biological systems. It is still a challenging task to distinguish between the n-Asp and isoAsp deamidation products in a proteomewide analysis because of their similar physicochemical properties. The quantification of the isomeric deamidated peptides is also rather difficult because of their coelution/poor separation in reverse-phase liquid chromatography (RPLC). We here propose a RP-ERLIC− MS/MS approach for separating and quantifying on a proteome-wide scale the three products related to deamidation of the same peptide. The key to the method is the use of RPLC in the first dimensional separation and ERLIC (electrostatic repulsion−hydrophilic interaction chromatography) in the second, with direct online coupling to tandem MS. The coelution of the three deamidation-related peptides in RPLC is then an asset, as they are collected in the same fraction. They are then separated and identified in the second dimension with ERLIC, which separates peptides on the basis of both pI and GRAVY values. The coelution of the three products in RPLC and their efficient separation in ERLIC were validated using synthetic peptides, and the performance of ERLIC−MS/MS was tested using peptide mixtures from two proteins. Applying this sequence to rat liver tissue, we identified 302 unique N-deamidated peptides, of which 20 were identified via all three deamidation-related products and 70 of which were identified via two of them. KEYWORDS:

■

INTRODUCTION Nonenzymatic deamidation of asparaginyl residues occurs spontaneously in proteins in vivo, resulting in a time-dependent change in protein structure and function. It has been proposed as a “molecular clock” in studies of aging since the deamidation products accumulate over time.1,2 Isoaspartyl residues are also reported to be related to Alzheimer’s disease (AD) and cataracts.3−5 In vitro, deamidation affects the purity and stability of therapeutic proteins during production and storage.6,7 As shown in Figure 1, under physiological conditions or at elevated pH, Asn deamidation proceeds mainly through the formation of a succinimide ring intermediate that is quickly hydrolyzed to n-Asp and isoAsp at a ratio of about 1:3.8 Isomerization of Asp residues occurs through the dehydration of Asp via a similar mechanism at neutral or acidic pH.9 It would be helpful to have an accurate and reliable analytical method to determine deamidation and isomerization sites in proteins in order to assess their biological significance. Mass spectrometry (MS) is a powerful tool for large-scale study of deamidation because of its sensitivity, speed, and © 2012 American Chemical Society

specificity of detection. It is routine to differentiate deamidated peptides from their undeamidated counterparts using highresolution MS on the basis of their mass difference of 0.984 Da.10,11 However, it is still a challenging task to distinguish between the n-Asp and isoAsp deamidation products since they have identical mass and charge (Figure 1). Different tandem MS peptide sequencing methods, such as collisional activated dissociation (CAD), high energy CID (HCD), and negative electrospray ionization have occasionally been observed to distinguish between the isomeric products on the basis of specific reporter ions, but the results are nonspecific, highly variable, and unreliable.12−14 The powerful soft peptide backbone-fragmenting techniques, electron capture dissociation (ECD) and electron transfer dissociation (ETD) are capable of distinguishing between the isomeric Asp and isoAsp peptides on the basis of a pair of c+57 and z−57 reporter ions.9 A proteome-scale deamidation study Received: October 19, 2011 Published: January 13, 2012 1804

dx.doi.org/10.1021/pr201048c | J. Proteome Res. 2012, 11, 1804−1811

Journal of Proteome Research

Article

mode of chromatography performed with an anion-exchange column has been reported to separate peptides on the basis of their pI,16 so the anticipated elution order of the triad of peptides in ERLIC should be Asn < n-Asp < isoAsp. Their separation and elution order may permit them to be distinguished even if they afford similar MS/MS spectra. Thus, ERLIC−MS/MS can potentially be used in the separation and characterization of the triad peptides instead of RPLC−MS/MS. To be compatible with mass spectrometry, a volatile elution solvent is necessary to minimize the ion suppression effect, clogging and adduct formation in the ionization source. Our recent development in ERLIC introduces such solvents that are fully compatible with MS analysis.16 For proteome-wide analysis, multidimensional liquid chromatography (MDLC) is commonly used to reduce sample complexity and to increase the dynamic range and sensitivity of peptide identification by minimizing undersampling and ion suppression problems.17−19 The inability of RPLC to separate the three deamidation-related products from the same peptide becomes an asset rather than a liability if it is used as the first dimension of MDLC since the triad of tryptic peptides will then be collected into the same fraction. This facilitates their separation and identification in the second dimension of ERLIC coupled to MS/MS and subsequent data analysis. Therefore, we propose here to develop RP-ERLIC−MS/MS for the proteome-scale study of deamidation and isomerization sites.

Figure 1. Deamidation of asparaginyl residues and isomerization of aspartyl residues through a succinimide intermediate.

has been attempted by applying LC−ECD−MS to cell lines;15 however, many of the initial hits based on the reporter ions have been found to be false positives. Additional criteria and manual validation of the results are necessary for positive assignment. The limitation is that the peak intensity of the reporter ions is only 5% that of other fragment ions. Thus, the reporter ions of isoAsp are usually buried in the background noise. More challenging is that under typical RP-MS/MS conditions, the isomeric Asp and isoAsp peptides coelute from RP columns, making their quantification rather difficult.15 Since the isomers have almost identical fragmentation patterns except for the weak reporter ions, they interfere with each other in identification by diluting the concentration of the reporter ions. Consequently, it is unlikely that deamidated peptides of low abundance can be identified on a proteome-wide scale. The triad of tryptic peptides (containing an Asn, n-Asp, or isoAsp residue at the same position) originating from the same protein can, in principle, be identified and quantified by LC− MS/MS if they can be separated by HPLC prior to tandem MS. These three peptides have similar or identical GRAVY values (−3.5 for the amino acids Asn, n-Asp, and isoAsp) and are unlikely to be separated well by RPLC in a typical SCX-RPbased shotgun proteomic approach. However, a peptide with an Asn to Asp conversion acquires an additional negative charge and is expected to elute later from an anion-exchange chromatography (AEX) column. Furthermore, n-Asp residues have a side-chain pKa around 3.9, while the corresponding pKa for isoAsp is around 3.1, resulting in different pI. The ERLIC

■

MATERIALS AND METHODS

Sample Preparation

Male Sprague−Dawley rats were handled in accordance with the guidelines of NTU Institutional Animal Care and Use Committee (NTU-IACUC), NTU, Singapore. Rat livers were snap-frozen in liquid nitrogen immediately after collection and kept at −80 °C until use. The tissue was cut into small pieces and ground into fine powder in liquid nitrogen with a pestle. The powder was then suspended in 8 M urea with protease inhibitor cocktail (P8340, Sigma) and 10 mM PMSF added in the ratio of 1:50 and 1:20 (v/v), respectively. The suspension was sonicated for 10 s thrice on ice and centrifuged at 20000g at room temperature (RT) for 30 min. The protein concentration of the supernatant was then determined by the bicinchoninic acid (BCA) assay. Trypsin digestion was done as previously

Figure 2. RPLC chromatogram of rat liver tryptic peptides. The UV absorption chromatogram was acquired at 280 nm. Twenty nine fractions were collected and pooled into 24 fractions for subsequent ERLIC−LC−MS/MS analysis. The inset shows the full scale chromatogram; the strong UV absorption at 2.5 and 7.8 min were due to impurities. 1805



Article

described.16 The obtained tryptic peptides were desalted using a Sep-Pak C18 cartridge (Waters, Milford, MA) and dried in a SpeedVac (Thermo Electron, Waltham, MA). The mixture of BSA (A2153, Sigma-Aldrich) and chicken ovalbumin (A5503, Sigma-Aldrich) was digested and desalted in the same way as with rat liver samples.

ERLIC−LC−MS/MS

ERLIC−LC−MS/MS was done on the same UFLC and mass spectrometer system with the following modifications: (1) A PolyWAX LP anion-exchange guard column (PolyLC, Columbia, MD) was used as the trap column, and 85% ACN/0.1% NH4OH was used as the desalting solvent. (2) The peptides were separated in a capillary column (200 μm × 15 cm) packed with PolyWAX LP anion-exchange bulk material (5 μm, 300 Å; PolyLC). (3) For the tryptic digest of BSA and chicken ovalbumin, mobile phase A (0.1% FA in ACN) and mobile phase B (0.1% FA in H2O) were used to establish the 60 min gradient comprised of 45 min of 0−30% B, 5 min of 30−50% B, 4 min of 50−90% B, and 1 min of 90−0% B, followed by reequilibration at 100% A for 5 min. (4) For synthetic peptides and rat liver samples, the same ERLIC mobile phases were used to establish a 90 min gradient comprised of 70 min of 0−30% B, 5 min of 30−50% B, 3 min of 50−90% B, and 2 min of 90− 0% B followed by re-equilibration at 100% A for 10 min.

RPLC Fractionation

Four triads of deamidation-related peptides with different pI and GRAVY values were synthesized at the Peptide Synthesis Core Facility, Nangyang Technological University, with the sequences DG(N/D/isoD)GYISAAELR, DS(N/D/isoD)GAILVYDVTDEDSFQK, VT(N/D/isoD)GAFTGEISPGMIK, and IEYDDQ(N/D/isoD)DGSCDVK. These peptides had been identified from a rat liver protein digest in a previous study.16 The mixture of each triad of peptides was fractionated using a BioBasic C18 column (4.6 × 250 mm, 5 μm, 300 Å, Thermo Scientific, West Palm Beach, FL) on a Shimadzu Prominence UFLC system monitored at 280 nm. Mobile phase A (0.1% formic acid (FA) in H2O) and mobile phase B (0.1% FA in ACN) were used to establish the 50 min gradient of 5% mobile phase B for 3 min, 5−40% B over 32 min, 40−60% B over 5 min, 60−100% B over 5 min, and 5 min at 100% B, at a flow rate of 1 mL/min. For the fractionation of rat liver whole proteome, 2 mg of peptides were fractionated using the same gradient, with 29 fractions collected. These were then dried in vacuo and combined into 24 fractions based on the amplitude of absorbance in the chromatogram as shown in Figure 2. Each fraction was redissolved in 200 μL of 85% ACN/0.1% NH4OH for ERLIC−LC−MS/MS analysis.

Data Analysis

The raw data were first converted into dta format using the extract_msn (version 4.0) in Bioworks Browser (version 3.3, Thermo Fisher Scientific, Inc.), and then the dta files were converted into Mascot generic file format using an in-house program. Intensity values and fragment ion m/z ratios were not manipulated. For BSA and chicken ovalbumin, the Swiss-Prot protein database (Release 57.11) was used for database searches. For synthetic peptides and rat liver samples, the IPI rat protein database (version 3.40, 40 381 sequences, 20 547 209 residues) and its reversed complement were combined and used for database searches. The database search was performed using an in-house Mascot server (version 2.2.04, Matrix Science, Boston, MA) with MS tolerance of 5.1 ppm and MS/MS tolerance of 0.8 Da. Two missed cleavage sites of trypsin were allowed. Carbamidomethylation (C) was set as a fixed modification, and oxidation (M), phosphorylation (S, T, and Y), and deamidation (N and Q) were set as variable modifications. The ion score significance threshold was set at 20. The obtained peptide/protein list for each fraction was exported to Microsoft Excel or processed using an in-house script for further analysis. For high confidence peptide identification, only peptides with an E-value of less than 0.05 were used for statistical calculation. The FDR of peptide identification was set to be less than 1% (FDR = 2.0 × decoy_hits/total_hits). The FDR rate is based on the assigned spectra.

ERLIC Fractionation

The mixture of the above-mentioned triad of peptides was fractionated using a PolyWAX LP anion-exchange column (4.6 × 200 mm, 5 μm, 300 Å, PolyLC, Columbia, MD) on a Shimadzu Prominence UFLC system monitored at 280 nm. Mobile phase A (85% ACN/0.1% acetic acid) and mobile phase B (30% ACN/0.2% FA) were used in a 30 min gradient of 5− 30% B over 15 min and 30−100% B over 10 min followed by 5 min at 100% B, at a flow rate of 1 mL/min. The volatile acid in the mobile phases is used to maintain the low pH. Thus, FA is in use when ERLIC is coupled to MS/MS. RP-LC−MS/MS

The mixture of the triad of synthetic peptides was separated and analyzed on a Shimadzu UFLC system coupled to a LTQFT Ultra (Thermo Electron, Bremen, Germany). It was injected onto a Zorbax peptide trap column (Agilent, CA) via the autosampler of the Shimadzu UFLC for concentration and desalting. The peptides were separated in a capillary column (200 μm × 10 cm) packed with C18 AQ (5 μm, 300 Å; BrukerMichrom, Auburn, CA). The flow rate was maintained at 500 nL/min. Mobile phase A (0.1% FA in H2O) and mobile phase B (0.1% FA in ACN) were used to establish the 60 min gradient comprised of 45 min of 5−35% B, 8 min of 35−50% B, and 2 min of 80% B followed by re-equilibration at 5% B for 5 min. Peptides were then analyzed on the LTQ-FT with an ADVANCE CaptiveSpray Source (Bruker-Michrom) at an electrospray potential of 1.5 kV. A gas flow of 2 L/min, ion transfer tube temperature of 180 °C, and collision gas pressure of 0.85 mTorr were used. The LTQ-FT was set to perform data acquisition in the positive ion mode as previously described except that the m/z range of 350−1600 was used in the full MS scan.20

■

RESULTS AND DISCUSSION

Proposed Strategy for Proteome-Scale Deamidation Study

We propose to study protein deamidation via the following shotgun proteomics arrangement: (1) Peptides in a tryptic digest are first separated on the basis of polarity using RPLC. All three peptides in a deamidation triad will presumably elute in the same fraction since they have similar polarity. (2) Each RPLC fraction is then run in the ERLIC mode in a capillary coupled to an MS/MS mass spectrometer. Volatile mobile phases can be used that are compatible with direct MS analysis. Tryptic peptides elute in ERLIC in order of decreasing pI value as well as increasing polarity,16 so the selectivity will be orthogonal to that of the RPLC step. This is appropriate for a 2-dimensional chromatography fractionation. In addition, the three peptides in a deamidation triad, all present in the same RPLC fraction, should be separated from each other at this 1806



Article

step. Figure 3 illustrates the flow scheme for the RP-ERLIC− MS/MS sequence. Accordingly, while studying this combina-

Figure 3. Analysis of whole proteome and deamidated peptides using RP-ERLIC-based two-dimensional chromatography coupled with MS/ MS. Two milligrams of rat kidney tissue extract were digested by trypsin, and the peptides were then separated using RP chromatography and combined into 24 fractions. Each fraction was analyzed using ERLIC chromatography coupled with MS/MS. The resulting spectra were searched against the rat IPI database with a Mascot algorithm.

tion for analysis of deamidation events, we also assessed its utility for peptide identification in general for proteomics applications using a complex tissue extract. Each step of the proposed strategy was tested using synthetic peptides and peptide mixtures from BSA and chicken ovalbumin before being applied to complex tissue samples.

Figure 4. Comparison of RPLC and ERLIC for fractionation of synthetic peptides: separation and identification of the peptide mixture of DG(N)GYISAAELR/DG(D)GYISAAELR/DG(isoD)GYISAAELR using C18 RP chromatography with a conventional column and absorbance detection (A) and with a C18 capillary coupled to MS/MS (B); separation of the peptide mixture DG(N)GYISAAELR/DG(D)GYISAAELR/DG(isoD)GYISAAELR using ERLIC chromatography with a conventional column (C) and with a capillary coupled to MS/ MS (D). The presence of Tyr residues in these peptides permitted their detection at 280 nm. The lack of separation of the three deamidation-related peptides by C18 columns makes this a good chromatography mode for the first dimension of MDLC since their elution in the same fraction will facilitate their identification as a set when resolved in the second dimension.

Comparison of RPLC and ERLIC for Fractionation of Synthetic Peptides

To evaluate the separation of the triads of deamidation-related peptides in RPLC, the peptide mixture was fractionated using a C18 column. As shown in Figure 4A, the peptides DG(N)GYISAAELR/DG(D)GYISAAELR/DG(isoD)GYISAAELR could not be separated from each other using a shallow gradient with normal RPLC conditions. When the RPLC fractionation was done on a nanoscale C18 column coupled to MS/MS, the extracted ion chromatograms (XICs) also showed similarly poor separation of the triad so that they interfered with each other’s identification and quantification (Figure 4B). Similar results were obtained from other peptide triads (data not shown). This seems to confirm the performance of RPLC in the strategy described above. It has been reported that RPLC could separate the three variant deamidation products with some modifications, but their separation in RPLC was poor, and the elution order was unpredictable and highly dependent upon chromatographic conditions.21 For comparison, ERLIC chromatography was also run on a conventional scale to separate the mixture of the triad of deamidation-related peptides. As shown in Figure 4C, they were well separated from each other using a normal ERLIC gradient. Each of the peptides was also run separately to confirm the elution order within the triad (Supporting Information, Figure S1). In addition, the identifications were verified with MALDI-TOF/ TOF based on the MS/MS spectra and the reporter ion of yn −

46.13 When the ERLIC fractionation was done on a nanoscale PolyWAX LP capillary coupled to MS/MS, the XICs also showed the same good peptide separation so that they did not affect each other’s identification and quantification (Figure 4D). Similar results were obtained from other peptide triads (data not shown). ERLIC appears to qualify as an appropriate second dimension of MDLC for analysis of deamidation events. Application of ERLIC−MS/MS to the Tryptic Digest of BSA and Chicken Ovalbumin

ERLIC−MS/MS was used to distinguish between the triad peptides from the tryptic digest of BSA and chicken ovalbumin in order to optimize the LC−MS/MS system before it was applied to complex proteomic samples. BSA and chicken ovalbumin fragments were identified with sequence coverage of 82 and 78%, respectively, suggesting that ERLIC−MS/MS has potential for use in analysis of whole proteomes as well as deamidation events. The database search result is available in 1807



Article

Figure 5. Separation and identification of two groups of deamidated peptides containing an asparaginyl, aspartyl, or isoaspartyl residue in the same position from the tryptic digest of BSA and chicken ovalbumin. (A) XICs of YNGVFQECCQAEDK and its deamidated products from ERLIC− LC−MS/MS; (B) XICs of YICDNQDTISSK and its deamidated products from ERLIC−LC−MS/MS; (C) MS/MS spectra for identification of Y(N)GVFQECCQAEDK; (D) MS/MS spectra for identification of Y(D/isoD)GVFQECCQAEDK. The amidated and deamidated peptides were differentiated on the basis of MS/MS spectra, while the two deamidated peptides containing aspartyl and isoaspartyl residues were differentiated on the basis of their elution times.

spectrum of the unmodified peptides (Figure 5C).22,23 The MS/MS spectra of the deamidated peptides containing Asp or isoAsp were identical in CAD analysis. They had to be distinguished instead on the basis of their elution times since these are predictable in ERLIC, as discussed in the Introduction. As shown in Figure 5A and 5B, the relative intensity of the last peak was 5.1 and 2.1 times that of the second peak, respectively, consistent with the theoretical ratio of 1:3 for n-Asp and isoAsp formed during nonenzymatic deamidation and isomerization.7,8,24 This further confirms their identities. It is obvious that the rate of deamidation is much higher in Figure 5A than that in Figure 5B, reflecting the rapid kinetics of Asn-deamidation at an -N-G- sequence.25,26 It

Supporting Information, Data S1. In total, 72 Asn-deamidated peptides were identified, 15 of them unique, and 4 of which were identified via all three deamidation-related products. The triads of deamidation-related peptides identified are listed in Supporting Information, Table S1. Again, the triad peptides were separated from each other in ERLIC with their retention time differing by at least 5 min as shown in Figure 5A and 5B. The deamidated peptides can be distinguished unambiguously from the corresponding nondeamidated counterparts on the basis of their differences in MS/MS spectra. The deamidation site is sandwiched between fragments with the deamidated asparagine (aspartic acid), i.e. b3−5, b7, b11−13, and y12 (Figure 5D), the mass of which was about 0.984 Da higher than that of the corresponding b or y ions detected in the MS/MS 1808



Article

appears that ERLIC−MS/MS can also be used to assess relative quantification of deamidated products from the same peptide.

for separating and characterizing the triads of deamidationrelated peptides has been demonstrated satisfactorily.

Application of RP-ERLIC−MS/MS to the Tryptic Digest of Rat Liver Tissue

Optimization of RP-ERLIC−MS/MS in the Future

As discussed above, ERLIC−MS/MS was shown to work efficiently on simple samples, e.g., the tryptic digest of two model proteins, but its sensitivity in analyzing complex tryptic digests was not good enough. On the basis of our preliminary experience in ERLIC−MS/MS, it can be further optimized in the following several aspects. In this study, we tried 90% ACN/ 0.1% FA, 85% ACN/1 mM NH4HCO3 and 85% ACN/0.1% NH4OH as the peptide solvent and desalting buffer. The last combination performed the best since the high pH improves the efficiency of peptide retention in the trap column. Other reasonable solutions may also be tried in the future for optimization of the trapping efficiency. Here we used a PolyWAX LP guard column as the trap column for ERLIC− MS/MS. It was not designed specifically for this purpose, and presumably a better-designed PolyWAX LP trap would perform better. Alternatively, ERLIC−MS/MS should be tested on LC−MS/MS systems without using trap columns. In addition, the mobile phases and gradients used in ERLIC−MS/MS can be optimized further. Although we have experience in using ERLIC for fractionation of peptides, this is the first time that ERLIC was used for a nanoscale separation coupled to MS/MS. With so many aspects to be optimized, it is expected that ERLIC−MS/MS can be improved to become sensitive enough for analysis of complex samples as the last dimension of LC before MS/MS.

To test whether the proposed RP-ERLIC−MS/MS sequence could be used in analyzing complex proteomic samples, the tryptic digest of 2 mg of rat liver extract was distributed into 24 fractions with RPLC. These were subjected to ERLIC−MS/MS for whole proteome and deamidation analysis with about 250 000 MS/MS spectra collected. The database search result is available in Supporting Information, Data S2. In summary, 1305 proteins were identified via at least 2 unique peptides; 23 658 peptides were identified, 3882 of them unique; 1408 Asndeamidated peptides were identified with 302 being unique, 20 of which were identified via all three deamidation-related products and 70 of which were identified via two of them. The identified triads of deamidation-related peptides are shown in Supporting Information, Table S2. For peptides identified with two deamidation-related products, all of them include the undeamidated peptides so that it is impossible to differentiate between Asp and isoAsp-containing deamidation products only on the basis of the elution time. However, their separation will surely facilitate their identification and quantification when ERLIC is coupled with ETD MS/MS. Some peptides are identified with only one or two deamidation-related products, possibly due to the following reasons: (1) Deamidation happens completely on the deamidation site either in vivo or during the process of sample preparation, and isomerization of deamidation products may also happen. (2) The sensitivity of the current ERLIC−MS/MS setup needs to be further optimized for detecting low-abundance deamidation products. As shown in Supporting Information, Figure S2A, the XICs show potential Asp products that cannot be identified by MS/ MS because of their intensity being lower than the threshold of 500 counts for MS/MS. (3) Any peptides can be missed in LC−MS/MS identification if they coelute with some peptides of higher abundance since most current mass spectrometry selects for peptides with highest intensity in a MS scan for further MS/MS analysis. As shown in Supporting Information, Figure S2B, the XICs show that potential Asp products are missing from MS/MS, possibly because of the coelution with peptides of higher intensity. Manual check of the MS full scan at the time points confirms our prediction (Supporting Information, Figure S2C,D). For peptides with multiple deamidation sites, the deamidation products are much more complex than those with only one deamidation site since the residue on each deamidation site could possibly be an asparaginyl, n-aspartyl, or isoaspartyl residue. Thus, the presented strategy has difficulties in accurately assigning all of these products. However, as shown in Supporting Information, Table S2, the peptides with two deamidated residues have longer retention times than those with only one deadmidated residue, such as GILGYTENQVVSCNFNSNSHSSTFDAGAGIALDDNFVK and SGTTWIQEIVNMIEQNGDVEK. It is expected that this strategy can be used in characterizing peptides with multiple deamidation sites when coupled with ECD. The number of proteins and peptides identified here was significantly lower in comparison with those identified from SCX-RP or ERLIC-RP sequences,16,27 and the number of identified Asn-deamidated peptides was also significantly lower than that from ERLIC-RP, indicating that our RP-ERLIC−MS/ MS system needs to be optimized further. However, its capacity

Calculation of the Intensity Ratio of Corresponding Isomeric n-Asp and isoAsp Peptides

If a n-Asp or isoAsp peptide is identified for over three times with LC−MS/MS, the average intensity of three highest ones is used for calculating the relative abundance of them. Otherwise, the average intensity of all identified peptides is used. This is to statistically reflect the highest intensity of the peptides. As shown in Supporting Information, Table S3, the ratio of corresponding isoAsp versus Asp peptides is between 2.09 and 5.90 for BSA and chicken ovalbumin. However, it is between 0.23 and 69.49 for rat liver tissue. It indicates that this strategy can be effectively used in evaluating the ratio of isoAsp versus Asp peptides in complex samples.

■

CONCLUSIONS The commonly used RP-MS/MS arrangement cannot be used effectively in separating and characterizing the structural isomers of deamidated peptides containing n-Asp and isoAsp because of their similar polarities, but the recently introduced ERLIC mode has been shown to be capable of separating them well because of their different pI. Thus, the novel RP-ERLIC− MS/MS sequence was proposed and validated for proteomewide deamidation characterization. Further optimization is needed before it can be used efficiently with complex samples. In the future, RP-ERLIC may be coupled to ECD for more reliable characterization of deamidation. This combination may also have advantages for whole proteome analysis due to its outstanding resolving power and compatibility with MS/MS.

■

ASSOCIATED CONTENT

S Supporting Information *

Data S1: The database search result of BSA and chicken ovalbumin. Data S2: The combined database search results of 1809



Article

contribute to crystallin insolubility? J. Proteome Res. 2006, 5 (10), 2554−66. (6) Solstad, T.; Flatmark, T. Microheterogeneity of recombinant human phenylalanine hydroxylase as a result of nonenzymatic deamidations of labile amide containing amino acids. Effects on catalytic and stability properties. Eur. J. Biochem. 2000, 267 (20), 6302−10. (7) Huang, L.; Lu, J.; Wroblewski, V. J.; Beals, J. M.; Riggin, R. M. In vivo deamidation characterization of monoclonal antibody by LC/MS/ MS. Anal. Chem. 2005, 77 (5), 1432−9. (8) Geiger, T.; Clarke, S. Deamidation, isomerization, and racemization at asparaginyl and aspartyl residues in peptides. Succinimide-linked reactions that contribute to protein degradation. J. Biol. Chem. 1987, 262 (2), 785−94. (9) Cournoyer, J. J.; Pittman, J. L.; Ivleva, V. B.; Fallows, E.; Waskell, L.; Costello, C. E.; O’Connor, P. B. Deamidation: Differentiation of aspartyl from isoaspartyl products in peptides by electron capture dissociation. Protein Sci. 2005, 14 (2), 452−63. (10) Robinson, N. E.; Lampi, K. J.; McIver, R. T.; Williams, R. H.; Muster, W. C.; Kruppa, G.; Robinson, A. B. Quantitative measurement of deamidation in lens betaB2-crystallin and peptides by direct electrospray injection and fragmentation in a Fourier transform mass spectrometer. Mol. Vision 2005, 11, 1211−9. (11) Robinson, N. E.; Zabrouskov, V.; Zhang, J.; Lampi, K. J.; Robinson, A. B. Measurement of deamidation of intact proteins by isotopic envelope and mass defect with ion cyclotron resonance Fourier transform mass spectrometry. Rapid Commun. Mass Spectrom. 2006, 20 (23), 3535−41. (12) Carr, S. A.; Hemling, M. E.; Bean, M. F.; Roberts, G. D. Integration of mass spectrometry in analytical biotechnology. Anal. Chem. 1991, 63 (24), 2802−24. (13) Gonzalez, L. J.; Shimizu, T.; Satomi, Y.; Betancourt, L.; Besada, V.; Padron, G.; Orlando, R.; Shirasawa, T.; Shimonishi, Y.; Takao, T. Differentiating alpha- and beta-aspartic acids by electrospray ionization and low-energy tandem mass spectrometry. Rapid Commun. Mass Spectrom. 2000, 14 (22), 2092−102. (14) Andreazza, H. J.; Wang, T.; Bagley, C. J.; Hoffmann, P.; Bowie, J. H. Negative ion fragmentations of deprotonated peptides. The unusual case of isoAsp: A joint experimental and theoretical study. Comparison with positive ion cleavages. Rapid Commun. Mass Spectrom. 2009, 23 (13), 1993−2002. (15) Yang, H.; Fung, E. Y.; Zubarev, A. R.; Zubarev, R. A. Toward proteome-scale identification and quantification of isoaspartyl residues in biological samples. J. Proteome Res. 2009, 8 (10), 4615−21. (16) Hao, P.; Guo, T.; Li, X.; Adav, S. S.; Yang, J.; Wei, M.; Sze, S. K. Novel application of electrostatic repulsion-hydrophilic interaction chromatography (ERLIC) in shotgun proteomics: Comprehensive profiling of rat kidney proteome. J. Proteome Res. 2010, 9 (7), 3520−6. (17) Hao, P.; Ren, Y.; Xie, Y. An improved protocol for Nglycosylation analysis of gel-separated sialylated glycoproteins by MALDI-TOF/TOF. PLoS One 2010, 5 (11), e15096. (18) Fang, X.; Balgley, B. M.; Wang, W.; Park, D. M.; Lee, C. S. Comparison of multidimensional shotgun technologies targeting tissue proteomics. Electrophoresis 2009, 30 (23), 4063−70. (19) Motoyama, A.; Yates, J. R. 3rd Multidimensional LC separations in shotgun proteomics. Anal. Chem. 2008, 80 (19), 7187−93. (20) Gan, C. S.; Guo, T.; Zhang, H.; Lim, S. K.; Sze, S. K. A comparative study of electrostatic repulsion-hydrophilic interaction chromatography (ERLIC) versus SCX-IMAC-based methods for phosphopeptide isolation/enrichment. J. Proteome Res. 2008, 7 (11), 4869−77. (21) Krokhin, O. V.; Antonovici, M.; Ens, W.; Wilkins, J. A.; Standing, K. G. Deamidation of -Asn-Gly- sequences during sample preparation for proteomics: Consequences for MALDI and HPLCMALDI analysis. Anal. Chem. 2006, 78 (18), 6645−50. (22) Zhang, H.; Guo, T.; Li, X.; Datta, A.; Park, J. E.; Yang, J.; Lim, S. K.; Tam, J. P.; Sze, S. K. Simultaneous characterization of glyco- and phosphoproteomes of mouse brain membrane proteome with

rat liver tissue. Table S1: The triads of deamidation-related peptides identified from BSA and chicken ovalbumin. Table S2: The triads of deamidation-related peptides identified from rat liver tissue. Table S3: The intensity ratio of corresponding isomeric n-Asp and isoAsp peptides for BSA, chicken ovalbumin, and rat liver tissue. Figure S1: ERLIC chromatograms of the peptides of DG(N)GYISAAELR (A), DG(D)GYISAAELR (B), and DG(isoD)GYISAAELR (C). Figure S2: XICs of the triads of deamidation-related peptides showing the reason why only one or two of them are identified. (A) XICs of SVTEF(N/D/isoD)GDTITNTMTLGDIVYK; (B) XICs of LWVACYN (N/D/isoD) GGR; (C) a full MS scan at 28.80 min shows that the peptide of LWVACYDGGR (mass = 598.78, double charge) is masked by other coeluting highabundance peptides; (D) zoom of the mass between 595 and 610 confirms the existence of the peptide LWVACYDGGR. The gray font in (A) and (B) signifies that the peptides are not identified by MS/MS. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: (+65)6514-1006. Fax: (+65)6791-3856. E-mail: sksze@ ntu.edu.sg. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by grants from Nanyang Technological University (RG 157/06, RG 61/06, and RG 51/10).

■

ABBREVIATIONS: RPLC, reverse-phase liquid chromatography; ERLIC, electrostatic repulsion-hydrophilic interaction chromatography; WAX, weak anion exchange; SCX, strong cation exchange; LC, liquid chromatography; LC−MS/MS, liquid chromatography coupled to tandem mass spectrometry; MDLC, multidimensional liquid chromatography; BSA, bovine serum albumin; PTM, post translational modifications; MS, mass spectrometry; LTQ, linear quadrupole ion trap; XIC, extracted ion chromatogram; AEX, anion exchange chromatography; DTT, dithiothreitol; IAA, iodoacetamide; FA, formic acid; FDR, false discovery rate; IPI, international protein index

■

REFERENCES

(1) Robinson, N. E.; Robinson, A. B. Molecular clocks. Proc. Natl. Acad. Sci. U. S. A. 2001, 98 (3), 944−9. (2) Robinson, N. E.; Robinson, A. B. Amide molecular clocks in drosophila proteins: Potential regulators of aging and other processes. Mech. Ageing Dev. 2004, 125 (4), 259−67. (3) Roher, A. E.; Lowenson, J. D.; Clarke, S.; Wolkow, C.; Wang, R.; Cotter, R. J.; Reardon, I. M.; Zurcher-Neely, H. A.; Heinrikson, R. L.; Ball, M. J.; et al. Structural alterations in the peptide backbone of betaamyloid core protein may account for its deposition and stability in Alzheimer’s disease. J. Biol. Chem. 1993, 268 (5), 3072−83. (4) Takata, T.; Oxford, J. T.; Demeler, B.; Lampi, K. J. Deamidation destabilizes and triggers aggregation of a lens protein, betaA3crystallin. Protein Sci. 2008, 17 (9), 1565−75. (5) Wilmarth, P. A.; Tanner, S.; Dasari, S.; Nagalla, S. R.; Riviere, M. A.; Bafna, V.; Pevzner, P. A.; David, L. L. Age-related changes in human crystallins determined from comparative analysis of posttranslational modifications in young and aged lens: Does deamidation 1810



Article

electrostatic repulsion hydrophilic interaction chromatography. Mol. Cell. Proteomics 2010, 9 (4), 635−47. (23) Hao, P.; Guo, T.; Sze, S. K. Simultaneous analysis of proteome, phospho- and glycoproteome of rat kidney tissue with electrostatic repulsion hydrophilic interaction chromatography. PLoS One 2011, 6 (2), e16884. (24) Wright, H. T. Nonenzymatic deamidation of asparaginyl and glutaminyl residues in proteins. Crit. Rev. Biochem. Mol. Biol. 1991, 26 (1), 1−52. (25) Bischoff, R.; Kolbe, H. V. Deamidation of asparagine and glutamine residues in proteins and peptides: Structural determinants and analytical methodology. J. Chromatogr., B: Biomed. Sci. Appl. 1994, 662 (2), 261−78. (26) Capasso, S. Estimation of the deamidation rate of asparagine side chains. J. Pept. Res. 2000, 55 (3), 224−9. (27) Peng, J.; Elias, J. E.; Thoreen, C. C.; Licklider, L. J.; Gygi, S. P. Evaluation of multidimensional chromatography coupled with tandem mass spectrometry (LC/LC−MS/MS) for large-scale protein analysis: The yeast proteome. J. Proteome Res. 2003, 2 (1), 43−50.

1811


Enhanced Separation and Characterization of Deamidated Peptides

Recommend Documents