Mass Spectrometric Analysis of Glyoxal and Methylglyoxal-Induced

Subscriber access provided by University of Otago Library

Article

Mass Spectrometric Analysis of Glyoxal and MethylglyoxalInduced Modifications in Human Hemoglobin from Poorly Controlled Type 2 Diabetes Mellitus Patients Hauh-Jyun Candy Chen, Yu-Chin Chen, Chiun-Fong Hsiao, and Pin-Fan Chen Chem. Res. Toxicol., Just Accepted Manuscript • DOI: 10.1021/acs.chemrestox.5b00380 • Publication Date (Web): 30 Oct 2015 Downloaded from http://pubs.acs.org on November 2, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Chemical Research in Toxicology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemical Research in Toxicology

Mass Spectrometric Analysis of Glyoxal and Methylglyoxal-Induced Modifications in Human Hemoglobin from Poorly Controlled Type 2 Diabetes Mellitus Patients

Hauh-Jyun Candy Chen,*1 Yu-Chin Chen,1 Chiung-Fong Hsiao,1 and Pin-Fan Chen2

1

Department of Chemistry and Biochemistry, National Chung Cheng University, 168 University Road, Ming-Hsiung, Chia-Yi 62142, Taiwan. 2Buddhist Dalin Tzu Chi General Hospital, No.2, Minsheng Rd., Dalin, Chia-Yi 622, Taiwan

*To whom correspondence should be addressed. Phone: (886) 5-242-8176. Fax: (886) 5-272-1040. E-mail: [email protected]. Keywords: blood, glyoxal, hemoglobin, mass spectrometry, methylglyoxal, PTMs Running title: modifications of hemoglobin by glyoxal and methylglyoxal

This article is dedicated for Prof. Iwao Ojima on the occasion of his 70th birthday.

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC graph

2


Page 2 of 42

Page 3 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Abstract

Glyoxal and methylglyoxal are oxoaldehydes derived from the degradation of glucose-protein conjugates and from lipid peroxidation, and they are also present in the environment. This study investigated the site-specific reaction of glyoxal and methylglyoxal with the amino acid residues on human hemoglobin using a shot-gun proteomic approach with nanoflow liquid chromatography/nanospray ionization tandem mass spectrometry (nanoLCNSI/MS/MS). In human hemoglobin incubated with glyoxal, modification on 8 different sites, including lysine residues at -Lys-11, -Lys-16, -Lys-56, -Lys-17, -Lys-66, -Lys-144 and arginine residues at -Arg-92 and -Arg-30, was observed using the data-dependent scan. In methylglyoxal-treated hemoglobin, there were specific residues, namely -Arg-92, -Lys-66, -Arg-30, and -Lys-144, and forming carboxyethylation as well as the dehydrated product hydroimidazolone at -Arg-92 and -Arg-30. These lysine and arginine modifications were confirmed by accurate mass measurement and the MS2 and MS3 spectra. The most intensive signal of each modified peptide was used as the precursor ion to perform the product ion scan. The relative extent of modifications was semi-quantified simultaneously relative to the native reference peptide by the nanoLCNSI/MS/MS under the selected reaction monitoring (SRM) mode. The extent of these modifications increased dose-dependently with increasing concentrations of glyoxal or methylglyoxal. Six out of the 8 modifications induced by glyoxal and three out of the six modifications induced by 3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 42

methylglyoxal were detected in hemoglobin freshly isolated from human blood samples. The relative extent of modification of these post-translational modifications was quantified in poorly controlled type 2 diabetes mellitus patients (n= 20) and in non-diabetic control subjects (n=21). The results show that the carboxymethylated peptides at α-Lys-16, α-Arg-92, β-Lys-17,

β-Lys-66,

and

the

peptide

at

α-Arg-92

with

methylglyoxal-derived

hydroimidazolone are significantly higher in diabetic patients than in normal individuals (p value < 0.05). This report identified and quantified glyoxal- and methylglyoxal-modified hemoglobin peptides in human and revealed the association of the extent of modifications at specific sites with T2DM. Only one drop (10 L) of fresh blood is needed for this assay and only an equivalent of 1 g of hemoglobin was analyzed by the nanoLC-NSI/MS/MS-SRM system. These results suggest the potential use of these specific post-translational modifications in hemoglobin as feasible biomarker candidates to assess protein damage induced by glyoxal and methylglyoxal.

4


Page 5 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction The non-enzymatic conjugation of glucose or reducing sugars with proteins (Maillard reaction) and advance glycation reactions in vivo leads to formation of reactive oxoaldehydes, such as glyoxal (gx) and methylglyoxal (Mgx).1-4 These oxoaldehydes react with biological targets forming irreversible advanced glycation end-products (AGEs), which are implicated in complications originated from hyperglycemia.2,3,5 While the concentration of glucose is high in patients with diabetes mellitus (DM), the plasma concentrations of glyoxal and methylglyoxal are higher than those in non-diabetic individuals.3,5 Glyoxal and methylglyoxal arise endogenously from oxidation of nucleic acids6,7 and lipids,8-11 and metabolism of nitrosamines.12 They are also found in the environment as well as in beverages, foods, and cigarette smoke.13-15 Glyoxal is mutagenic16,17 and it reacts with biomolecules, modifying DNA and proteins and causing cross-links between them.18-26 Glyoxal and methylglyoxal are substrates of glyoxalase 1, part of an important defense system, and they are thus implicated in diabetes, uremia, aging, and various cancers.2,27-30 These oxoaldehydes are also known to induce signals for tyrosine phosphorylation and trigger protein tyrosine kinase cascades.31,32 Post-translational modifications (PTMs) of a protein lead to alteration of the protein’s structural integrity and its biological functions.33-35 The N-termini, cysteine, arginine, and lysine residues are possible sites of protein modification by glyoxal and methylglyoxal.2,36-38 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For instance, glyoxal and methylglyoxal react with lysine, forming Nε-carboxymethyllysine (CML) and Nε-carboxyethyllysine (CEL), respectively.36,39 These AGEs are potential biomarkers of diabetic complications, including atherosclerosis, retinopathy, nephropathy, and neuropathy.2,40,41 It was shown that the major modification in human Hb by methylglyoxal was N-(5-hydro-5-methyl-4-imidazolon-2-yl)ornithine (H-Mgx), the cyclic dehydration product of addition of one methylglyoxal molecule on arginine residues (Scheme 1).42-44 Characterization of modified peptides in enzyme digest has been performed on glycated human serum albumin (HSA) using MALDI-TOF MS and LC/ESI-MS/MS, and amino groups on lysine residues of HSA were found to be most exposed to the glycation reaction.45 However, MALDI-TOF MS is not suitable for quantitative analysis. Measurement of plasma CML and CEL was performed after acid or enzyme hydrolysis to release the modified amino acids with or without derivatization before analysis by liquid chromatography tandem mass spectrometry (LC-MS/MS) 30,37,39,46,47 or by ELISA.48 However, acid hydrolysis of proteins to amino acids is accompanied by decomposition of the modified amino acids to great extents.38 Quantification of modified amino acids does not provide insight of the site of modification on the proteins. Identification and quantification of post-translationally modified proteins by mass spectrometric techniques facilitate the discovery of modified peptides as disease biomarkers. 6


Page 6 of 42

Page 7 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The shot-gun approach with nanoflow liquid chromatography nanospray ionization tandem mass spectrometry (nanoLC-NSI/MS/MS) is a highly sensitive and specific method in proteomic research, allowing qualitative as well as quantitative analysis for low abundance proteins. It is especially valuable when the amount of biological samples is limited. Biological fluids, such as urine and blood, are often used as surrogates for tissues in biomonitoring studies. However, compared to urine collection, blood drawing is invasive to an individual. In this study, sites and types of glyoxal and methylglyoxal-induced modifications on the tryptic peptides of human hemoglobin (hHb) were identified. The extent of these modifications can be quantified in as little as one drop (10 L) of blood. This nanoLC-NSI/MS/MS assay is applied to clinical samples to investigate the association with diabetes.

7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

EXPERIMENTAL SECTION

Materials. Human hemoglobin, dithiothreitol (DTT), glyoxal, methylglyoxal, and 2-vinylpyridine were from Sigma Chemical Co. (St. Louis, MO). Trypsin was obtained from Promega Corporation (Madison, WI). All reagents used in this study were of reagent grade or above.

Incubation of Human Hemoglobin with Glyoxal or Methylglyoxal. A solution containing commercial hHb (100 M) and glyoxal or methylglyoxal (0, 10, 20, 200, 400 M or 50 mM) in potassium phosphate buffer (0.1 M, pH 7.4) in 0.1 mL (total volume) was incubated at 37 C for 48 h under the nitrogen atmosphere. Isolation and Quantification of Hemoglobin. Blood was freshly collected and contained in a tube containing the anticoagulant (10% (v/v) citrate-dextrose solution), which was centrifuged at 800g for 10 min at 10 C to isolate red blood cells from plasma. The hemoglobin isolated was as reported and quantified by intrapolation into a calibration curve constructed from solutions of standard hHb in 50 mM of HCl measuring fluorescence at 280 nm (excitation) and 353 nm (emission).49,50 Enzyme Digestion. An aliquot equivalent

to 50 g of Hb (8 L) was added to cold

acetone (80 L), and kept at -20 C for 15 min, followed by centrifugation at 23,000 ×g at 0 C for 20 min to remove excess glyoxal or methylglyoxal. The supernatant was discarded and 8


Page 8 of 42

Page 9 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the precipitate was air-dried. The precipitate was dissolved in 80 L of double distilled water, SDS (1.0%, 10 L) and DTT (100 mM, 10 L) added, and then incubated at 95 C for 10 min, followed by treatment with 2-vinylpyridine (50 mM) with shaking in the dark at room temperature for 1 h. Next, the mixture was added to cold acetone (900 L) and kept at 20 C for 15 min, followed by centrifugation at 23,000 ×g for 20 min to remove the reagents. To the precipitate was added a solution of trypsin (10:1, w/w, 50 L, 5 g) in ammonium bicarbonate (100 mM, pH 8.0) and it was incubated at 37 °C for 18 h. A solution of trifluoroacetic acid (0.1%, 50 L) was added to the reaction mixture to terminate the digestion reaction and the trypsin digest was passed through a 0.22 m Nylon syringe filter. Two microliters of the solution was injected into the nanoLC-NSI/MS/MS system for analysis of the modified peptides described below. Accurate Mass Measurement. The hemoglobin solution incubated with glyoxal or methylglyoxal (5 mM) went through trypsin digestion and precipitation as described above, and it was reconstituted in 100 μL of trifluoroacetic acid (0.1%) before analysis by a reversed phase nanoLC system coupled with LTQ Orbitrap XL (Thermo Fisher Scientiﬁc, San Jose, CA). The peptides were analyzed by electrospray ionization in the positive ion mode with a spray voltage of 1.6 kV. The mass spectrometry was performed in the data-dependent scan mode. Using a rate of 30 ms/scan, one full scan with m/z 3002000 in the Orbitrap was at a resolution of 60,000 at m/z 400. In the LTQ, the five most intense peaks for fragmentation 9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 42

were selected with a normalized collision energy value of 35%. To exclude the same m/z ions from the reselection for fragmentation, a repeat duration of 180 s was applied. Modified peptides were identified using the in-house MASCOT v2.3.02 search engine on the Swiss-Prot 56 human protein database. The mass tolerance was defined as less than 5 ppm for the precursor ions and 0.8 Da for the product ions. All the MS/MS spectra were searched

against

the

database

for

detecting

variable

modifications,

including

pyridinylethylation (+ 105), the addition of one glyoxal (+ 58) or methylglyoxal (+ 72) molecule to cysteine, histidine, lysine, and arginine residues, as well as formation of hydroimidazolone derived from glyoxal (+ 40) or methylglyoxal (+ 54) to arginine residues. One missed cleavage was allowed on trypsin. The Mascot cutoff score was set to 20 (p < 0.05) to exclude the low score peptides, and only allowed “rank1” (best match for each MS/MS) peptides to be included. NanoLC-NSI/MS/MS Analysis. All the peptide quantification experiments were performed using an LTQ linear ion trap mass spectrometer (Thermo Electron Corp., San Jose, CA) equipped with a nanospray ionization (NSI) source. The source was connected online to an UltiMate 3000 Nano LC system (Dionex, Amsterdam, Netherlands). Two microliters of each sample was manually injected onto a C18 tip column (75 μm 120 mm, 5 μm, 100 Å ) packed in-house (Magic C18, Michrom BioResource, Auburn, CA). The mobile phases A and B were composed of 5 % and 80 % acetonitrile in 0.1% formic acid (pH 2.6), 10


Page 11 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


respectively. The elution was employed at a flow rate of 300 nL/min eluting at 4% B for the first 3 min, followed by a linear gradient from 4% B to 40% B in the next 37 min, and from 40% B to 90% B in the next 20 min, maintained at 90 % B for another 10 min and equilibrated with 4 % B for 20 min before the next run. The column was coupled to an LTQ linear ion trap mass spectrometer (Thermo Electron Corp., San Jose, CA) fitted with a nanospray ionization source. All MS/MS experiments for peptide identification were performed at a heated capillary temperature of 200 C with a capillary voltage of 2.0 V, a source voltage of 1.5 kV, a tube lens voltage of 70 V, a source current of 100 A, a normalized collision energy setting of 35%, and the ion gauge pressure of 6.4  10-6 torr. The MS3 experiments were performed by acquiring the product ion scan spectra of the most abundant b or y ion containing the modification in the MS2 spectra with 35% normalized collision energy and at the precursor ion at 2 m/z isolation width. Activation Q for collision-induced dissociation (CID) was at 0.25 and the activation time was 30 ms. Semi-quantification of Glyoxal- or Methylglyoxal-Modified Peptides. The selected reaction monitoring (SRM) experiments were performed by selection of the precursor ion and acquisition of product ion scan spectra. The formation of a specific fragment ion from each precursor ion was used to construct the chromatogram. The specific SRM conditions for lysine- and arginine-containing peptides with modifications and their reference peptides are listed in Table 3. The relative extent of modification on a specific peptide was calculated as 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 42

the peak area ratio of the modified peptide versus the sum of the peak areas of the modified peptide and the corresponding reference peptide in the SRM chromatograms. Dose-dependency of the relative extent of PTM by glyoxal or methylglyoxal concentrations. The solution 0.1 mM of human hemoglobin was incubated with various concentrations of glyoxal (10, 20, 200, or 400 M) or methylglyoxal(10, 20, 200 M) in potassium phosphate buffer (0.1 M, pH 7.4) at 37 ℃ for 48 h and quantified for the extent of modifications described above. For each concentration, the experiments were performed in triplicates. The dose-dependency was plotted as the peak area ratios versus the concentration of glyoxal or methylglyoxal. Study-Subjects. This study was approved by the Institutional Review Boards (IRB) of of Buddhist Dalin Tzu Chi General Hospital (IRB No. B10203014-1) to recruit type 2 diabetic patients (12 male and 9 female). The average age was 54.5  14.4 ( SD) years and their HbA1c levels averaged 10.2%  1.4% ( SD) ranging between 8.0% and 13.1%. Nondiabetic subjects were recruited under the approval of the IRB of the National Chung Cheng University (IRB No. 100112902). The nondiabetic subjects included 15 male and 6 female workers and students of NCCU with the mean age of being 26.1  8.5 ( SD) years. Statistical Analysis. Statistical analysis was performed by GraphPad InStat version 3.00 for Windows 95, GraphPad Software (San Diego, CA, www.graphpad.com). The 12


Page 13 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


nonparametric MannWhitney test was used to analyze the difference in level of each modification between the diabetics and the normal control subjects. The correlation between the extent of each modification and the HbA1c was performed by the nonparametric Spearman correlation.

13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 42

RESULTS AND DISCUSSION

The aim of this study is to characterize and quantify glyoxal- and methylglyoxal-modified peptides in human hemoglobin and to explore the possibility of using them as biomarker candidates for glyoxal- and methylglyoxal-induced protein damage in vivo. First, the sites and types of glyoxal-induced modifications was identified in the tryptic digest of glyoxal-

and

methylglyoxal-treated

hHb

by

accurate

mass

measurement

with

nanoLC-NSI/MS/MS under the data-dependent scan mode and confirmed by the MS2 and MS3 spectra. Second, semi-quantification of the modified peptides was established based on the peak areas of the modified peptides and their reference peptides by the selected reaction monitoring (SRM) transitions. Third, the suitability of modified peptides as biomarker candidates was validated by their dose-dependent formation in glyoxal- or methylglyoxal-treated human Hb. Finally, hemoglobin samples freshly isolated from blood samples of diabetic patients and normal subjects were analyzed and the extent of these modifications were compared (Figure 1).

Characterization of Glyoxal- and Methylglyoxal-Induced Modifications on Hemoglobin by Accurate Mass Analysis. In order to obtain high extent of modification for characterization purpose, glyoxal-induced post-translational modifications of hHb were investigated in hHb incubated with glyoxal or methylglyoxal (5 mM) at 37 C for 48 h. After 14


Page 15 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the excess glyoxal or methylglyoxal was removed, the treated hHb was denatured, alkylated by vinylpyridine, and digested with trypsin before analysis by nanoLC-NSI/MS/MS using high-resolution LTQ Orbitrap under the data-dependent scan mode to identify the reaction sites of the peptides. Modification at lysine and arginine sites led to miscleavage when digested with trypsin. Using the conventional setting for trypsin with cleavage at lysine and arginine residues and allowing two miscleavage sites, the sequence coverage of - and -globin was 96% and 89%, respectively.

Characterization of these PTMs used the following criteria. The mass tolerance of the parent ions should be below 5 ppm and that for the fragment ions should be below 0.8 Da. In addition, the Mascot scores for the peptides should be greater than 20, and those for the gxand Mgx-modified peptides were between 22 and 109 (Table 1). Each CID spectrum of each peptide ion was manually examined to confirm its identity and the site of modification. Both modified and unmodified peptides were observed and their fragmentation patterns were similar. The most intensive molecular ion of each modified peptide was used as the precursor ion to obtain the product ion scan spectra. The flanking b- or y-ions of modification-containing fragments were present in the MS2 and MS3 spectra of the modified peptides (Figure S1S3, Supporting Information). Table 1 lists the types and sites of glyoxal-modified peptides and alkylated peptides in human hemoglobin. Addition of a glyoxal molecule to the reactive sites leads to a mass 15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 42

increase of 58 Daltons. To distinguish it from carboxymethylation (+58 Da) by the commonly used alkylating agent iodoacetic acid or carboxyamidomethylation (+57 Da) by iodoacetamide, 2-vinylpyridine was used to alkylate the cysteine residues, which leads to a mass increase of 105 Daltons. Carboxymethylation (+58 Da) on 8 different sites, including 6 lysine residues (-Lys-11, -Lys-16, -Lys-56, -Lys-17, -Lys-66, -Lys-144) and 2 arginine

residues

(-Arg-92,

-Arg-30),

was

observed.

No

dehydrated

product

hydroimidazolone formation (+40 Da) derived from glyoxal on arginine residues was detected. Figure 2A showed the representative collision-induced dissociation spectrum of peptide AAWG16KgxVGAHAGEYGAEALER at the doubly charged molecular ion of m/z 1051.10. A mass difference of 186 Da was observed between the b5 ion at m/z 572.42 and the b4 ion at m/z 386.06 as well as between the y16 ion at m/z 1716.13 and the y15 ion at m/z 1529.67, supporting the presence of carboxymethyllysine at the 5th residue of this peptide. Several modification-containing fragment ions (b5, b6, b9, b10, b12, b13, b15, b16, b16+2, b18, b19, y16, y17+2, y18, y18+2) are present in this spectrum. The MS3 spectrum obtained by fragmenting the y18+2 ion at m/z 981.10 gives ions with carboxymethyl modification, which further confirmed its identity (Figure 2B). In methylglyoxal-treated hemoglobin, carboxyethylation (+72 Da) was observed on 2 lysine residues (-Lys-66, -Lys-144) and 2 arginine residues (-Arg-92, -Arg-30). Also 16


Page 17 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


detected was the hydroimidazolone formation (+54 Da) on -Arg-92 and -Arg-30 (Table 2). At -Arg-30, both peptides with miscleavage and cleavage were detected. It was reported that the major form of methylglyoxal induced modification of hemoglobin35 and other proteins38,44,51-57 was the hydroimidazolone on arginine residues. Our results showed carboxyethylation at lysine residues in addition to formation of the hydroimidazolone. No modification on cysteine residues was observed in gx- or Mgx-treated hemoglobin, contrary to what was detected in plasma proteins.58 It might be due to the fact that modification of methylglyoxal was shown to be reversible.36 In addition, there are only three cysteine residues in human hemoglobin, one in -chain and two in -chain, and their solvent accessibility is very low. Semiquantification of the Extent of Modifications. After characterization of PTMs induced by glyoxal and methylglyoxal, simultaneous quantification of these modified peptides is achieved under the SRM mode. The SRM experiments were performed by selecting the precursor ion of a peptide and acquiring the full-scan product ion spectra. Monitoring the fragmentation of each precursor ion to a specific fragment ion was used to construct the chromatogram. Adopting the concept of native reference peptide method,59 an unmodified peptide sharing partially the same sequence as the modified peptide was used as internal standard to adjust for the variations in peptide recovery during the sample preparation procedures prior to LC-MS/MS analysis. The relative extent of modification of a 17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 42

peptide was quantified as the peak area ratio, namely, the peak area of the modified peptide over the sum of the peak areas of the modified peptide and the unmodified reference peptide on their SRM chromatograms. There might be difference in the charge state distribution for the modified and unmodified peptides, but all the peptide ions we found and chose were mostly doubly charged ions as shown in Table 3. Nevertheless, quantification based on different ionic states was reported to be similar.60 The product ions selected for the modified peptides were those containing the modification sites, although they might not be the ions with the highest intensity. However, when the intensity of the fragment ions with modification was too low, the product ions without modification in high intensity were chosen. Because modification by glyoxal or methylglyoxal at lysine and arginine residues normally causes miscleavage during trypsin digestion, the unmodified peptides used as reference peptides are hence shorter than the corresponding modified peptides. The only exception was glyoxal-modified -Arg-30-containg peptide which both miscleaved and cleaved peptides were identified and quantified. In chromatograms with multiple peaks, the one with its peak area increases with increasing concentrations of glyoxal or methyglyoxal is chosen (described below). The identity of the peak was confirmed with its MS2 spectrum. Figure 3 shows representative nanoLC-NSI/MS/MS chromatograms of glyoxal- and methylglyoxal-modified and unmodified -Arg-92- containing peptides in the tryptic digest 18


Page 19 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of hemoglobin from a diabetic patient (No. 18) analyzed under the SRM mode, including the modified peptides L92RgxVDPVNFK, L92RMgxVDPVNFK and L92RH-MgxVDPVNFK, and their unmodified reference peptide

93

VDPVNF99K. Peptide L92RH-MgxVDPVNFK eluted at

34.67 min, slightly earlier than L92RMgxVDPVNFK eluting at 35.03 min. The modified peptides

eluted

later

than

the

reference

VNVDEVGGEALG30RH-MgxLLVVYPWTQR

peptide eluted

(21.29 slightly

min). earlier

Similarly, than

VNVDEVGGEALG30RMgxLLVVYPWTQR (Figure S4, Supporting Information). The ideal method to obtain absolute quantitation of a modification is to synthesize isotopically labelled protein and mix it with the target protein before digestion. Because of the high cost and difficulty in the synthesis, it is rarely performed. Our method is semi-quantitative, which does not provide absolute quantitation. We used native peptides from the same protein as reference peptide, which can adjust for the variation in digestion efficiency between samples, and thus the relative extent of modification can be compared between samples. This approach has been used for relative quantification of PTMs including phosphorylation,

nitration,

oxidation,

glutathionylation,

and

hydroimidazolone

formation.42,49,59,61-64 Dose-Dependent Modifications of Hemoglobin by Glyoxal or Methylglyoxal. To identify the peak of interest in the chromatogram with multiple peaks and to investigate whether the modified peptides may be useful for biomonitoring purposes, commercial hHb 19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 42

was incubated with various concentrations of glyoxal (10400 M) or methylglyoxal (10200 M), digested with trypsin, and analyzed by nanoLC-NSI/MS/M under the SRM mode as described above. The results showed that the glyoxal- and methylglyoxal-induced modifications increased dose-dependently with glyoxal or methylglyoxal concentrations (Figure 4), suggesting the validity of these peptides as protein damage by these oxoaldehydes. Using trypsin as the digestive enzyme offers high sequence coverage, although it gives miscleavage on modification at Lys and Arg sites, which is expected and can be recognized. Only modification at -Arg-30 is both cleaved and miscleaved peptides identified in gx-treated hemoglobin. As shown in Figure 4A, the extent of modification in -Arg-30-containing peptide with miscleavage is several thousand times higher than that without miscleavage. The Extent of Modifications in Hemoglobin Isolated from Poorly Controlled T2DM Patients and Nondiabetic Individuals. Because glyoxal and methylglyoxal are present in human plasma and cells at concentrations of submicromolar range,65 we examined the presence of these gx- and Mgx-induced modifications in human hemoglobin and investigated the relationship between the extent of these modifications with diabetes mellitus. Hemoglobin samples from 20 patients with poorly controlled T2DM and 21 nondiabetic individuals (normal controls) were isolated from blood, trypsin-digested, and analyzed by nanoLC/NSI/MS/MS using the SRM transitions listed in Table 3. The results showed that six 20


Page 21 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of the eight gx-modified peptides were detected and quantified except peptides containing -Arg-30 (both miscleaved and cleaved) and -Lys-144. On the other hand, among the six Mgx-modified peptides, only the ones with carboxyethylation at -Arg-92 and -Lys-66 and hydroimidazolone at -Arg-92 were detected and quantified. The extents of modification of these peptides are listed in Table S1 of Supplementary Information. As summarized in Table 4A, the extent of modification for -Lys-16, -Arg-92, -Lys-17, and -Lys-66 in gx-modified peptides is significantly higher in poorly controlled T2DM patients than in nondiabetic controls with p values of 0.0196,

Mass Spectrometric Analysis of Glyoxal and Methylglyoxal-Induced

Recommend Documents