International Journal of Biological Macromolecules 66 (2014) 311–318
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules
j ourna l h o mepa ge: www.elsev ier .com/ locate / i jb iomac
tructural alterations of hemoglobin and myoglobin by glyoxal: comparative study
auradipta Banerjee, Abhay Sankar Chakraborti ∗
epartment of Biophysics, Molecular Biology & Bioinformatics, University of Calcutta, 92, Acharyya Prafulla Chandra Road, Kolkata 700009, India
r t i c l e i n f o
rticle history:eceived 5 November 2013eceived in revised form 2 February 2014ccepted 17 February 2014vailable online 5 March 2014
eywords:emoglobin
a b s t r a c t
Glyoxal, a highly reactive oxoaldehyde, increases in diabetic condition. It reacts with different proteinsto form advanced glycation end products (AGEs). Here we have studied the structural alterations aswell as the sites and nature of amino acid modifications of two heme proteins, hemoglobin and myo-globin on incubation with glyoxal for seven days at 25 ◦C. In comparison with normal hemoglobin (HbA0),glyoxal-treated hemoglobin (GHbA0) exhibits decreased absorbance around 280 nm, reduced intrinsicfluorescence and lower surface hydrophobicity. However, glyoxal-treated myoglobin (GMb) exhibitsthe opposite effects in these respects when compared to normal myoglobin (Mb). Glyoxal increasesthe thermal stability of hemoglobin, while it decreases the stability of myoglobin. Matrix-assisted laser
yoglobin
lyoxaldvanced glycation end productsydroimidazolonearboxymethyllysine
desorption ionization-time of flight (MALDI-TOF)–mass spectrometry reveals modifications of Arg-31�,Arg-40� and Arg-104� of hemoglobin by glyoxal to hydroimidazolone adducts. On the other hand, gly-oxal modifies Lys-133 and Lys-145 to carboxymethyllysine and Arg-31 to hydroimidazolone adducts inmyoglobin. Thus the same oxoaldehyde exerts different effects on hemoglobin and myoglobin and maybe associated with different structural properties of the proteins.
. Introduction
Reducing sugars react with amino groups of proteins, a processnown as non-enzymatic glycation (Maillard reaction) resultingn browning, fluorescence and crosslinking of proteins [1]. Theeaction consists of several steps, including Schiff’s base forma-ion, Amadori rearrangement etc. finally leading to formation ofdvanced glycation end products (AGEs). Formation of AGEs in vivoontributes to pathophysiologies associated with aging and com-lications of diabetes [2].
The �-oxoaldehydes namely, glyoxal, methylglyoxal and 3-eoxyglucosone are produced in different pathways includinglycation reactions [3]. These compounds are highly reactive andnown to cause protein modification and AGE formation moreffectively than the parent hexose sugars [4,5]. Methylglyoxal haseen reported to react with several proteins namely, cytochrome c,eruloplasmin, myoglobin, hemoglobin, etc., resulting in either pro-
ein cross-linking and aggregation or formation of non cross-linkingGE adducts [6–10]. Like methylglyoxal, glyoxal is another reactivexoaldehyde, and its concentration increases from 215–230 nM in
normal individuals to 350–470 nM in diabetic subjects [11]. Glyoxalis a major product of glucose degradation under oxidative condi-tions [3]. Fructose, arabinose and ascorbate may also degrade toglyoxal, possibly through intermediate adducts to proteins. Glyoxalis formed directly during oxidative degradation of polyunsaturatedfatty acids [12] and myeloperoxidase-mediated degradation ofserine at sites of inflammation [13]. It has been reported to interactwith several proteins, namely, �-crystallin [14], bovine serumalbumin [15], �-synuclein [16] and hemoglobin [17]. Glyoxalmodifies predominantly lysine and arginine residues of proteinsto form several products, such as carboxymethyllysine (CML) [18],carboxymethylarginine (CMA) [19], dihydroxyimidazolidines (G-DH1 and G-DH2) and hydroimidazolones (G-H1, G-H2 and G-H3).It enters into red blood cells (RBC) and reduces their deformabilityprobably by interacting with the cellular proteins [20].
Recent findings from our laboratory have shown that methylgly-oxal interacts with the heme proteins myoglobin and hemoglobinleading to their modifications [8,10]. However, methylglyoxal-induced modifications differ with respect to structural changes,site of modifications and nature of AGEs formed in hemoglobinand myoglobin. The findings have prompted us to undertakethe present study. Myoglobin, a simple monomeric protein,associated with storage and transfer of oxygen, exhibits higher
percentage of � helicity and greater thermal stability comparedto hemoglobin, a tetrameric protein associated with transport ofoxygen. Considering increased concentration of glyoxal in diabetic
12 S. Banerjee, A.S. Chakraborti / International Jour
ondition as well as its high reactivity, we have studied structurallterations of hemoglobin and myoglobin, after in vitro reactionith glyoxal under identical conditions. In a recent study [17],emoglobin–glyoxal interaction has been reported to promoteggregation and AGE cross-linking with quite high concentrationsf glyoxal (20–90%, v/v). In the present study, we have used muchow concentration of glyoxal (50 �M) to find how it affects thetructure and stability of hemoglobin and myoglobin together withhe sites and nature of adducts formed on the modified proteins.
equencing grade trypsin and �-cyano-hydroxy cinnamic acidatrix (CHCA) were purchased from Sigma Chemical Company,SA. Biorex-70 resin (200–400 mesh) was obtained from Bio-Rad,
ndia. All other reagents were AR grade and purchased locally.
.2. Separation of non-glycated hemoglobin (HbA0) from bloodample
Blood samples were obtained from healthy non-smoking humanubjects aged 25–30 years. Total hemoglobin was isolated and puri-ed from RBC by using Sephadex G-100 column chromatography
n 50 mM phosphate buffer (PB), pH 6.6. Non-glycated hemoglobinHbA0) was separated from total hemoglobin by cation exchangehromatography using Biorex-70 resin in elution buffer PB [1].he concentration of HbA0 was determined from Soret absorbancesing an extinction coefficient (ε415nm) of 125 mM−1 cm−1 (hemeasis) [21].
.3. In vitro reaction of HbA0 and myoglobin (Mb) with glyoxal
Mb was dissolved in PB and its concentration was determinedsing ε408nm = 116 mM −1 cm−1 [21]. 100 �M each of HbA0 and Mbere separately incubated with different concentrations of glyoxal
5, 10, 20 and 50 �M) under sterile conditions for 7 days at 25 ◦C.or control experiments, HbA0 or Mb solution was incubated in thebsence of glyoxal under identical conditions.
.4. Polyacrylamide gel electrophoresis (PAGE)
The incubated samples were subjected to native PAGE (10%)or 3 h at constant voltage (60 V). Control Mb and 50 �M glyoxal-reated Mb sample were also applied to SDS-PAGE (10%). Forlectrophoresis, 15 �l of each protein sample (15 �M) was loaded,ollowed by staining with Coomassie R250.
.5. Spectrophotometric study
The absorption spectra of control and glyoxal-incubated sam-les (6 �M each) were recorded in the region 250–650 nm in aV/VIS Spectrophotometer (Hitachi U 2000) using 1 ml quartzuvette of path length 1 cm.
.6. Spectrofluorimetric study
Fluorescence emission spectra of control and glyoxal-incubatedamples (6 �M each) were monitored in the region 300–400 nm
n a spectrofluorimeter (Hitachi F-3010) with excitation at 280 nmsing 3 ml quartz cuvette of path length 1 cm.
For ANS binding study, the samples (8 �M each) were incu-ated with 20 �M ANS for 10 min at room temperature and the
Biological Macromolecules 66 (2014) 311–318
fluorescence emission spectra (450–600 nm) were recorded withexcitation at 370 nm.
2.7. Circular dichroic (CD) study
CD spectra of control and glyoxal-treated samples (3 �M each)were recorded in a spectropolarimeter (Jasco 600) using 1 mm pathlength cuvette in the far UV region (190–250 nm). The �-helicalcontents of the proteins were determined according to the methodof Chen et al [22].
2.8. Thermal stability study
In a differential scanning calorimetric (DSC) study, the meltingprofiles of 100 �M (1 ml) each of glyoxal (50 �M)-treated HbA0and Mb samples (termed as GHbA0 and GMb, respectively) andthe control samples were obtained in a VP-DSC Microcalorimeterby heating the samples (1 ◦C/minute) over a definite temperaturerange. Before introduction into the calorimetric cells, the proteinsamples were thoroughly degassed.
Thermal stability of the protein samples was also studied byPAGE. After heating HbA0 and glyoxal-incubated HbA0 samples(15 �M each) at 75 ◦C for 10 min and Mb and glyoxal-incubated Mbsamples (15 �M each) at 80 ◦C for 10 min, the samples were sub-jected to native PAGE (10%), followed by staining with CoomassieR250. The temperatures of heating were kept around the meltingtemperatures of the heme proteins.
2.9. Matrix-assisted laser desorption ionization-time of flight(MALDI-TOF)–mass spectrometric study
The samples (50 �M glyoxal treated proteins GHbA0 andGMb and respective controls) were digested with sequencing-grade trypsin in solution at 37 ◦C for 16 h using enzyme:proteinratio 1:100 (w/w). For GMb, in-gel trypsin digestion was alsoperformed after excising the lower band from the native gel.The digested samples (0.5 �l each) were loaded directly to theMALDI plate, mixed with 0.5 �l of saturated CHCA solution(prepared in 50% acetonitrile and 0.1% trifluoroacetic acid) andallowed to dry and crystallize. Mass spectra were recorded ina 4800 Proteomics Analyzer (MALDI-TOF/TOF mass spectrome-ter, Applied Biosystems) using the linear positive ion mode ofMALDI-TOF MS at 20 kV acceleration voltage. To identify prob-able modified peptides and the specific glyoxal derived AGEsformed, theoretical digestion of the heme proteins was per-formed, considering up to two trypsin miscleavages (peptidemass,Expasy, http://www.expasy.ch/tools/peptide-mass.html) and pep-tide masses with specific mass increments due to AGE adductswere searched. The particular peptides of interest (i.e. having massconsistent with the mass increment due to AGE) were selectedfor MS/MS fragmentation by Collision Induced Dissociation (CID)using 1 kV collision energy. 1,000 laser shots were collected for eachMS/MS spectrum using a fixed laser intensity of 4,500 V. Raw datawere generated by using GPS Explorer Software. The identificationof AGE modified peptides were done by manual interpretation ofthe MS/MS spectra.
3. Results and discussion
Glycation of proteins and their subsequent modifications havebeen ascribed to play significant roles in different pathologicalcomplications. Here we have studied the structural alterations
of HbA0 and Mb on reaction with glyoxal under identical condi-tions. The highest concentration of glyoxal used in the study was50 �M. The concentration used was much higher than the plasmaconcentration of the dicarbonyl found in vivo [23]. However, in
S. Banerjee, A.S. Chakraborti / International Jour
iological systems, less than 10% of glyoxal is present in unboundorm, as most of the reactive carbonyl groups are bound to cys-einyl, lysyl and arginyl residues of proteins [24]. Thus, the fractionsf dicarbonyls bound to proteins significantly exceed their mea-ured concentrations in plasma. Much higher concentrations oflyoxal (150 �M–100 mM) have been used for in vitro reactionsith �-crystallin and �-synuclein [14,16]. In another study, gly-
xal as much as 20–90% (v/v) has been used in reaction with bovineemoglobin [17].
.1. Effect of glyoxal on heme proteins: structural modifications
Compared to control HbA0 (lane 1), HbA0 samples incubatedith different concentrations of glyoxal (5, 10, 20 and 50 �M)
xhibited gradually increased electrophoretic mobility in nativeAGE (lanes 2–5), as shown in Fig. 1A . Glyoxal-treated Mb sampleslso exhibited increased electrophoretic mobilities (Fig. 1B, lanes–5) in comparison with untreated Mb (lane 1). However, Mb sam-le incubated with 50 �M glyoxal showed the presence of a distinctand with higher electrophoretic mobility (Fig. 1B, lane 5), whichas not detected in other samples (Fig. 1B, lanes 1–4). The increasedobility of HbA0 or Mb on glyoxal modification may occur due to
oss of positive charge(s), as reported in case of modification oferuloplasmin by methylglyoxal [7] or glycation of hemoglobin byructose [25,26]. However, for 50 �M glyoxal-treated Mb sample,he lower band with higher mobility may appear due to glyoxal-
ediated degradation of the protein. It may also be due to internalross-linking, as reported in the reaction of ribose with Mb [27].owever, 50 �M glyoxal treated Mb sample showed only one band
n SDS-PAGE (Fig. 1C, lane 2), and was more or less at the sameosition of control Mb (lane 1). MALDI mass spectrum of the sam-le did not show the presence of any peak with m/z value lower than7 kDa (molecular weight of Mb) (data not shown). Thus increasedobility (lower band) of the modified myoglobin in native PAGEight be due to loss of positive charge of the protein, and not due
o degradation or internal cross-linking.The absorption spectra of control and glyoxal-treated HbA0
amples were found to be more or less identical (Fig. 1D). However,lyoxal-treated HbA0 samples exhibited slightly lower absorbanceround the aromatic region (inset of Fig. 1D, traces b–e) comparedo control HbA0 (trace a). On the other hand, the absorption spec-ra of control and glyoxal-treated Mb samples were also more oress similar (Fig. 1E), except around the aromatic region, where thelyoxal-treated samples exhibited slightly higher absorbance (insetf Fig. 1E, traces b–e) compared to control Mb (trace a). Changes inbsorbance around the aromatic region may indicate slight changesn tertiary structure of the heme proteins on glyoxal modification.
Emission fluorescence spectra of the protein samples (excita-ion 280 nm) were recorded. As shown in Fig. 1F, the intensity ofuorescence emission of glyoxal-incubated HbA0 samples (traces–e) were lower than that of control HbA0 (trace a). On the otherand, glyoxal-incubated Mb samples (Fig. 1G, traces b–e) exhibitedigher fluorescence emission compared to control Mb (trace a). Inoth cases, the observed effects vary gradually with concentrationf glyoxal, and there was no shift in the wavelength of maxi-um intensity (�max), which was around 330 nm with excitation at
80 nm. There are reports which suggest that a decrease in fluores-ence emission may occur due to increased exposure of aromaticesidues to solvent molecules that collide with the fluorophoresnd quench the fluorescence energy, without shift in �max [28,29].owever, in absence of shift in emission maxima, it is difficult to
uggest the change in solvent accessibility of tryptophan residues
f glyoxal-treated HbA0 and Mb. Decrease in fluorescence emissionithout shift in �max in glyoxal-treated HbA0 may also occur due to
ncreased energy transfer (quenching) to heme moiety or neighbor-ng amide groups, compared to control HbA0. In comparison with
Biological Macromolecules 66 (2014) 311–318 313
Mb, reduced fluorescence energy transfer to heme or neighboringamide groups in glyoxal-treated Mb may cause increased fluores-cence emission. An increase or decrease in distance between hememoiety and tryptophan residues due to modification of the pro-teins may be associated with reduced or enhanced energy transferto heme [28,30].
ANS binding was studied to determine the effect of glyoxal onthe surface hydrophobicities of the heme proteins. ANS is stronglyfluorescent when bound to proteins and non-fluorescent whensurrounded by water. Enhancement of its fluorescence when sur-rounded by nonpolar amino acid residues of proteins is useful instudying surface hydrophobicity of proteins [31]. The fluorescenceintensity of ANS was found to be lower when bound to glyoxal-treated HbA0 samples (Fig. 1H, traces b–e) than when boundto HbA0 sample (trace a), suggesting that glyoxal modificationreduced the surface hydrophobicity of HbA0. Glyoxal has also beenreported to decrease the surface hydrophobicity of �-crystallin[14]. However, when bound to glyoxal-treated Mb samples (Fig. 1I,traces b–e), ANS exhibited increased fluorescence intensity thanwhen bound to control Mb (trace a), indicating greater exposure ofhydrophobic amino acid residues of Mb after reaction with glyoxal.The changes in tertiary structure of proteins may alter exposureof hydrophobic amino acid residues and ANS binding [32,33]. Thedifference between glyoxal-treated HbA0 and glyoxal-treated Mbsamples with respect to ANS binding may also be due to the differ-ence in their tertiary structures and exposed hydrophobic aminoacids.
CD experiments were carried out with different samples and�-helix contents were measured. The results indicate that glyoxalexhibits almost no effect or very little effect on the secondary struc-tures of the heme proteins. The CD spectra of control HbA0 andGHbA0, and that of control Mb and GMb are shown. As shown inFig. 1J, GHbA0 exhibited slightly increased � helicity (78%) (traceb) compared to HbA0 (75%) (trace a). Both control Mb (trace a)and GMb (trace b) exhibited identical CD profiles (Fig. 1K) hav-ing 84% � helices. The findings are in agreement with other reportsalso. Methylglyoxal-induced modification of �A-crystallin [34] andglycation-induced modification of human serum albumin [35]appear to change the tertiary structures of the proteins, withoutaffecting their secondary structures.
3.2. Effect of glyoxal on heme proteins: thermal stabilities
Comparison of the DSC thermograms of HbA0 and GHbA0 indi-cated higher Tm value of the modified protein (77.2 ◦C, trace b) thanthe unmodified protein (73.1 ◦C, trace a), as shown in Fig. 2A . In con-trast, GMb exhibited lower Tm value (75.1 ◦C, trace b) compared tountreated Mb (82.6 ◦C, trace a) (Fig. 2B).
On heating at 75 ◦C for 10 min, HbA0 sample appeared as asmear in native PAGE (Fig. 2C, lane 1), instead of a distinct bandas seen earlier without heating (Fig. 1A, lane 1). However, HbA0samples treated with different concentrations of glyoxal (5, 10, 20and 50 �M) exhibited gradual resistance to thermal stress, as seenby gradually increased protein band intensity along with disap-pearance of smearing (Fig. 2C, lanes 2–5). Fig. 2D shows the effectof thermal stress on Mb and glyoxal-treated Mb samples. Whenheated at 80 ◦C for 10 min, control Mb (lane 1) and Mb sample incu-bated with 5 �M glyoxal (lane 2) were found to be fairly resistantto thermal stress as seen by the presence of distinct protein bands,with a slight decrease in band intensity of the latter. However, Mbsamples treated with 10, 20 and 50 �M glyoxal exhibited gradu-ally and significantly increased protein degradation (lanes 3–5).
Thus glyoxal exerts opposite effects on HbA0 and Mb with respectto thermal stability. In a recent study on methylglyoxal-inducedrbance around 280 nm, reduced tryptophan fluorescence andincreased �-helix content of the modified protein [10]. These find-
314 S. Banerjee, A.S. Chakraborti / International Journal of Biological Macromolecules 66 (2014) 311–318
Fig. 1. Effect of glyoxal on structures of HbA0 and Mb. HbA0 (100 �M) or Mb (100 �M) was incubated with different concentrations of glyoxal – 0 �M (a), 5 �M (b), 10 �M (c),20 �M (d) and 50 �M (e), and were used in different experiments, unless otherwise indicated. (A and B) Native PAGE of glyoxal-treated HbA0 and Mb samples, respectively.Lanes 1–5: HbA0/Mb incubated with 0, 5, 10, 20 and 50 �M glyoxal, respectively. Samples (15 �l each) having 15 �M concentration were loaded. (C) SDS-PAGE of control Mband GMb (Mb treated with 50 �M glyoxal). Lane 1: Control Mb, lane 2: GMb. Samples (15 �l each) having 15 �M concentration were loaded. (D and E) Absorption spectraof glyoxal-treated HbA0 and Mb samples (a–e), respectively. The spectrum of each sample (6 �M) was recorded in the region 250–650 nm. Inset: The same spectra wereshown in the region 250–300 nm with enlarged scale. (F and G) Fluorescence emission spectra of glyoxal-treated HbA0 and Mb samples (a–e), respectively. The spectrum ofeach sample (6 �M) was recorded in the region 300–400 nm with excitation at 280 nm. (H and I) Fluorescence emission spectra of ANS in the presence of glyoxal-treatedHbA0 and Mb samples (a–e), respectively. The protein samples (a–e) were incubated with 20 �M ANS for 10 min at room temperature before recording the fluorescenceemission. The spectrum of each sample (8 �M) was recorded in the region 450–600 nm with excitation at 370 nm. (J and K) CD spectra of hemoglobin and myoglobin samples,respectively. HbA0, GHbA0 (HbA0 treated with 50 �M glyoxal), Mb and GMb were used for recording the spectra in the far UV region (190–250 nm), taking 3 �M each of theprotein samples. The spectra of control and treated samples are shown as traces a and b, respectively.
S. Banerjee, A.S. Chakraborti / International Journal of
Fig. 2. Effect of glyoxal on thermal stabilities of HbA0 and Mb. (A and B) DSC ther-mograms of hemoglobin and myoglobin samples, respectively. The thermograms ofHbA0 (trace a), GHbA0 (trace b), Mb (trace a) and GMb (trace b) were recorded using100 �M each of the proteins. (C and D) Native PAGE of glyoxal-treated HbA0 and Mbsamples, respectively. HbA0 and glyoxal-treated HbA0 samples (15 �M each) wereheated at 75 ◦C for 10 min. Mb and glyoxal-treated Mb samples (15 �M each) wereh1r
icmtw
amaeeaighowowachiOpiGaolpp
3o
F(at
eated at 80 ◦C for 10 min. The protein samples were subjected to native PAGE. Lane: control HbA0/Mb, lanes 2–5: HbA0/Mb treated with 5, 10, 20 and 50 �M glyoxal,espectively.
ngs are more or less similar with those of GHbA0. However, inontrast with present study, methylglyoxal makes hemoglobinore thermolabile. It is not yet clear why the observed effects of
wo �-oxoaldehydes (methylglyoxal and glyoxal) on hemoglobinith respect to thermal stability are different.
Fructose-induced modification has been reported to increasend decrease the thermal stabilities of multimeric �-crystallin andonomeric �-crystallin, respectively [36]. These effects (increased
nd decreased thermal stabilities) are, in turn, associated withnhanced or decreased �-helical contents of the proteins. How-ver, in our study glyoxal induces thermal stability to hemoglobin,lthough marginal or almost no increase in �-helicity is exhib-ted by glyoxal-incubated HbA0. Reduction in thermal stability oflyoxal-incubated Mb is also not associated with any change in �elicity. Thermodynamic stability of �A-crystallin by methylgly-xal has been reported with subtle change of tertiary structureithout any change of secondary structure [34]. There are also
ther reports [28,37] suggesting increase in protein stabilityithout significant change in secondary structure, and are in
greement with our findings. Methylglyoxal causes no signifi-ant structural alterations but induces slight decrease in surfaceydrophobicities of insulin, �-lactalbumin and �-crystallin, result-
ng in their resistance to thermal and chemical stresses [38].n the other hand, enhancement in surface hydrophobicity ofroteins may be associated with decrease of stability due to unfold-
ng, as seen for lysozyme and human serum proteins [39,40].lucose-mediated reduction of thermal stability of myoglobin isssociated with significant enhancement of surface hydrophobicityf the protein [41]. However, further studies are needed to corre-ate between observed changes in hydrophobicities and thermalroperties of the control and glyoxal-treated HbA0 and Mb sam-les.
.3. Identification of the chemical nature and molecular locationf AGEs in GHbA0 and GMb
The tryptic mass fingerprints of HbA0 and GHbA0 are shown in
ig. 3A and B , respectively. In the peptide mass spectrum of GHbA0Fig. 3B), the peptides with m/z values 1569.65 Da, 1314.70 Dand 1166.75 Da were identified as the arginine containing pep-ides 17–31 (having theoretical m/z value 1529.73 Da) of the �
Biological Macromolecules 66 (2014) 311–318 315
chain, 31–40 (having theoretical m/z value 1274.72 Da) and 96–104(having theoretical m/z value 1126.56 Da) of the � chain withmodifications of Arg-31�, Arg-40� and Arg-104� to hydroimida-zolone (G-H1) adducts, resulting in mass increase of 40 Da in eachcase. Sequence information of the modified peptides of GHbA0was obtained by MS/MS fragmentation using CID. In CID frag-mentation technique, the peptide bond is fragmented to generateb-ions (charge retained by the amino-terminal fragment) and y-ions (charge retained by the carboxy-terminal fragment). Thus, ifan amino acid residue is modified, the particular y and b frag-ment ions containing the modified amino acid residue, should havethe particular amino acid mass value plus mass increment due tomodification. The MS/MS spectrum of the peptide with m/z value1569.65 Da is shown in Fig. 3C. As shown in the spectrum, the argi-nine containing y (y1, y4, y5, y7, y8, y9, y10 and y13) ions exhibitedmass increment of 40 Da confirming modification of Arg-31 of �chain of HbA0 to G-H1 forming GHbA0. The MS/MS spectra of othermodified peptides having m/z values 1314.70 Da and 1166.75 Daindicated modifications of Arg-40� and Arg-104� to G-H1, respec-tively, in GHbA0 (Supplemental Figs. S1A and S1B). Results aresummarized in Table 1.
The tryptic mass fingerprints of control Mb and GMb are shownin Fig. 4A and B , respectively. For GMb, the m/z values 1418.84 Daand 1560.86 Da were identified as the peptide sequences 134–145(having theoretical m/z value 1360.75 Da) and 119–133 (havingtheoretical m/z value 1502.66 Da) with modifications of Lys-145and Lys-133 to carboxymethyllysine (CML) adducts, resulting inmass increment of 58 Da in each case. From the spectrum of GMb(Fig. 4B), the m/z value 1646.87 Da was identified as the peptidesequence 17–31 (having theoretical m/z value 1606.85 Da) withmodification of Arg-31 to hydroimidazolone (G-H1) adduct, result-ing in mass increment of 40 Da.
Peptide mass fingerprint spectrum of the lower band of GMb(Fig. 1B, lane 5) was obtained by in-gel trypsin digestion, and isshown in Fig. 4C. The mass spectrum also revealed the presence ofthree peptides with modifications at Lys-145, Lys-133 and Arg-31to AGE adducts (confirmed by CID fragmentation), similar to thatobtained by in solution digestion (Fig. 4B). Thus the loss of positivecharge on lysine and arginine residues due to AGE modificationmight be responsible for the appearance of fast moving band innative PAGE.
For both GMb and GHbA0, AGE adducts at lysine or arginineresidues were not found to prevent trypsin digestion at thesesites. Methylglyoxal-derived AGEs at lysine and arginine residuesof proteins namely, cytochrome c, myoglobin, hemoglobin, enolase,aldolase, etc., prevent trypsin digestion [6,8,10,42]. On the otherhand, AGE adducts including those formed by glyoxal have beenreported not to prevent trypsin digestion of several modified pro-teins [43,44], as found in our study. However, it is not yet clear how,even after loss of positive charge on lysine and arginine residues,the proteins bind at the catalytic pocket of trypsin.
Sequence information of the modified peptides of GMb wasobtained by MS/MS fragmentation using CID. The MS/MS spec-trum of the peptide with m/z value 1418.84 Da is shown in Fig. 4D.As shown in the spectrum, the lysine containing y (y1, y2, y3,y4, y6, y7, y9 and y11) ions exhibited mass increments of 58 Daconfirming modification of Lys-145 of Mb to CML forming GMb.The MS/MS spectra of the peptides having m/z values 1560.86 Daand 1646.87 Da indicated modifications of Lys-133 and Arg-31 toCML and G-H1, respectively in GMb (Supplemental Fig. S2A and B).Results are summarized in Table 1.
The tryptic mass fingerprint of GMb (Fig. 4B) showed a peak
with m/z = 1400.86 Da, and was identified as the peptide sequence134–145 (theoretical m/z value 1360.75 Da) with modification ofLys-145 of Mb to Pyrrolidone-CML, resulting in mass increment of40 Da. The MS/MS spectrum of this peptide is shown in Supple-
316 S. Banerjee, A.S. Chakraborti / International Journal of Biological Macromolecules 66 (2014) 311–318
Fig. 3. Detection and molecular localization of AGEs in GHbA0.(A) Tryptic mass fingerprint spectrum of HbA0. (B) Tryptic mass fingerprint spectrum of GHbA0. (C) MS/MS spec-t fragmw
moi
pweimugc
TA
T
rum of the glyoxal-modified HbA0 peptide with m/z 1569.65 Da, showing the y and bith a G-H1 modification on the arginine residue denoted as R*.
ental Fig. S2C. Loss of water and rearrangement after formationf CML gives rise to Pyrrolidone-CML [45]. Results are summarizedn Table 1.
Both the AGEs (G-H1 and CML) induce oxidative stress [46],rotein cross-linking [47] etc., and are known to be associatedith pathological conditions, including diabetes, Alzheimer’s dis-
ase, multiple sclerosis etc. [46,48,49]. As found in the presentn vitro study, glyoxal-induced G-H1 and CML cause structural
odifications of the heme proteins, and may be significantnder in vivo conditions also. For example, increased level oflyoxal in diabetes mellitus as well as its high reactivity mayause structural modifications of the heme proteins by G-H1 and
able 1ssignment of modified amino acid residues in GHbA0 and GMb.
Name of protein Observed mass (Da) Theoretical mass (Da) Peptide sequenc
he specific AGEs (G-H1 and CML) are indicated in the table and the modified amino acid
ent ions. The detected fragment ions arise from the sequence VGAHAGEYGAEALER*
CML. In a recent study, we have shown methyglyoxal inducesmodification of arginine residues (Arg-92� and Arg-104�) form-ing hydroimidazolone (MG-H1) with consequent change in thestructural and functional properties of hemoglobin, includingenhanced free iron-mediated oxidative reactions [10]. Furtherstudies are, therefore, necessary to understand how the structuralmodifications of hemoglobin and myoglobin by glyoxal-inducedG-H1 and CML are associated with the functional properties
of the heme proteins, including any pathological complica-tion.
In most experiments, the observed effects on HbA0 and Mbchange gradually with the concentration of glyoxal, and may be
e Mass increase (Da) AGE identified Modified residue
S. Banerjee, A.S. Chakraborti / International Journal of Biological Macromolecules 66 (2014) 311–318 317
Fig. 4. Detection and molecular localization of AGEs in GMb. (A) Tryptic mass fingerprint spectrum of Mb. (B) Tryptic mass fingerprint spectrum of GMb. (C) A representatives afterw ns arim
raocoeotf
C
A
ENtDs
[
[
[
ection of the tryptic mass fingerprint spectrum of the lower band of GMb obtainedith m/z 1418.84 Da, showing the y and b fragment ions. The detected fragment ioodified to CML.
elated to the extent of AGE modification of the proteins. Tetramericnd monomeric nature of the heme proteins may account for thebserved effects, including AGE adducts formation and structuralhanges by glyoxal, as also seen for fructose-induced modificationsf multimeric �-crystallin and monomeric �-crystallin [36]. How-ver, this explanation is only a speculation at current stage andther differences between HbA0 and Mb may also contribute tohe opposite effects of glyoxal on the heme proteins, and requiresurther investigations.
onflict of interest
The authors declare that they have no conflict of interest.
cknowledgements
S.B. received a research fellowship [no. 09/028(0802)/2010-MR-1] from the Council of Scientific and Industrial Research,
ew Delhi. The study was supported by financial assistances from
he Department of Science and Technology, New Delhi [grant no.ST/SR/FST/LSI-286/2006(c)] and the University Grants Commis-
ion, New Delhi [grant no. UGC (DSA) F.4-1/2009 (SAP-II)].
[
[
[
in-gel trypsin digestion. (D) MS/MS spectrum of the glyoxal-modified Mb peptidese from the sequence ALELFRNDIAAK* with the lysine residue denoted as K* being
Appendix A. Supplementary data
Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.ijbiomac.2014.02.034.
References
[1] M.P. Cohen, V. Wu, Methods Enzymol. 231 (1994) 65–75.[2] R. Bucala, A. Cerami, Adv. Pharmacol. 23 (1992) 1–34.[3] P.J. Thornalley, A. Langborg, H.S. Minhas, Biochem. J. 344 (1999) 109–116.[4] N. Ahmed, D. Dobler, M. Dean, P.J. Thornalley, J. Biol. Chem. 280 (2005)
5724–5732.[5] M.P. Kalapos, Toxicol. Lett. 110 (1999) 145–175.[6] L.M.A. Oliviera, R.A. Gomes, D. Yang, S.R. Dennison, C. Familia, A. Lages, A.V.
[7] J.H. Kang, J. Biochem. Mol. Biol. 39 (2006) 335–338.[8] S. Banerjee, A.S. Chakraborti, Protein J. 32 (2013) 216–222.[9] Y. Gao, Y. Wang, Biochemistry 45 (2006) 15654–15660.10] T. Bose, A. Bhattacherjee, S. Banerjee, A.S. Chakraborti, Arch. Biochem. Biophys.
529 (2013) 99–104.11] A. Lapolla, R. Flamini, A.D. Vedova, A. Senesi, R. Reitano, D. Fedele, E. Basso, R.
Seraglia, P. Traldi, Clin. Chem. Lab. Med. 41 (2003) 1166–1173.12] M. Fu, J.R. Requena, A.J. Jenkins, T.J. Lyons, J.W. Baynes, S.R. Thorpe, J. Biol. Chem.
[47] G.A. Kleter, J.J.M. Damen, M.J. Buijs, J.M.T. Cate, J. Dent. Res. 77 (1998) 488–495.
18 S. Banerjee, A.S. Chakraborti / International Jour
16] D. Lee, C.W. Park, S.R. Paik, K.Y. Choi, Biochim. Biophys. Acta 1794 (2009)421–430.
17] A. Iram, T. Alam, J.M. Khan, T.A. Khan, R.H. Khan, A. Naeem, PLoS ONE 8 (2013)e72075.
18] M.U. Ahmed, S.R. Thorpe, J.W. Baynes, J. Biol. Chem. 261 (1986) 4889–4894.19] K. Iijima, M. Murata, H. Takahara, S. Irie, D. Fujimoto, Biochem. J. 347 (2000)
23–27.20] H. Iwata, H. Ukeda, T. Maruyama, T. Fujino, M. Sawamura, Biochem. Biophys.
Res. Commun. 321 (2004) 700–706.21] J. Bhattacharyya, M. Bhattacharyya, A.S. Chakraborti, U. Chaudhuri, R.K. Poddar,
Biochem. Pharmacol. 47 (1994) 2049–2053.22] Y.H. Chen, J.T. Yan, H.M. Martinez, Biochemistry 11 (1972) 4120–4131.23] T.W. Lo, M.E. Westwood, A.C. McLellan, T. Selwood, P.J. Thornalley, J. Biol. Chem.
269 (1994) 32299–32305.24] P.J. Thornalley, Crit. Rev. Oncol. Hematol. 20 (1–2) (1995) 99–128.25] R. GhoshMoulick, J. Bhattacharya, S. Roy, S. Basak, A.K. Dasgupta, Biochim. Bio-
phys. Acta 1774 (2007) 233–242.26] T. Bose, A.S. Chakraborti, Biochim. Biophys. Acta 1780 (2008) 800–808.27] M. Bokiej, A.T. Livermore, A.W. Harris, A.C. Onishi, R.K. Sandwick, Biochem.
Biophys. Res. Commun. 407 (2011) 191–196.28] L. Jian-Zhong, M. Wang, BMC Biotechnol. 7 (2007) 1–8.29] P. Rondeau, G. Navarra, F. Cacciabaudo, M. Leone, E. Bourdon, V. Militello,
Biochim. Biophys. Acta 1804 (2010) 789–798.
30] K. Chattopadhyay, S. Mazumdar, Biochemistry 39 (2000) 263–270.31] E. Bismuto, E. Gratton, D.C. Lamb, Biophys. J. 81 (2001) 3510–3521.32] E. Gabellieri, G.B. Strambini, Biophys. J. 85 (2003) 3214–3220.33] M.H. Gangadhariah, B. Wang, M. Linetsky, C. Henning, R. Spanneberg, M.A.
36] M. Luthra, D. Balasubramanian, J. Biol. Chem. 268 (1993) 18119–18127.37] V.T. Pham, E. Ewing, H. Kaplan, C. Choma, M.A. Hefford, Biotechnol. Bioeng. 101
(2008) 452–459.38] A. Biswas, B. Wang, M. Miyagi, R.H. Nagaraj, Biochem. J. 409 (2008) 771–777.39] C.L. Hawkins, M.J. Davies, Chem. Res. Toxicol. 18 (2005) 1669–1677.40] S. Gorinstein, I. Goshev, S. Moncheva, M. Zemser, M. Weisz, A. Caspi, I. Libman,
H.T. Lerner, S. Trakhtenberg, O.M. Belloso, J. Protein Chem. 19 (2000) 637–642.41] A. Roy, R. Sil, A.S. Chakraborti, Mol. Cell. Biochem. 338 (2010) 105–114.42] R.A. Gomes, H.V. Miranda, M.S. Silva, G. Graca, A.V. Coelho, A.E. Ferreira, C.
teome Res. 7 (2008) 2025–2032.45] S. Carulli, C.D. Calvano, F. Palmisano, J. Agric. Food Chem. 59 (2011) 1793–1803.46] X. Girones, A. Guimera, C.-Z. Cruz-Sanchez, A. Ortega, N. Sasaki, Z. Makita,
