Contents lists available at SciVerse ScienceDirect
Journal of Inorganic Biochemistry
j ourna l homepage: www.e lsev ie r .com/ locate / j inorgb io
Copper(II) complexes of neurokinin A with point mutation (S5A) and products ofcopper-catalyzed oxidation; role of serine residue in peptides containingneurokinin A sequence
Elżbieta Jankowska a, Marta Błaszak b, Teresa Kowalik-Jankowska b,⁎a Faculty of Chemistry, University of Gdańsk, Sobieskiego 18, 80-952 Gdańsk, Polandb Faculty of Chemistry, University of Wroclaw, Joliot Curie 14, 50-383 Wroclaw, Poland
Apotentiometric, spectroscopic (UV-visible, CD and EPR) and electrospray ionizationmass spectrometric (ESI-MS)study of Cu(II) binding to the neurokinin Awith pointmutation (S5A) (ANKA), His-Lys-Thr-Asp-Ala5-Phe-Val-Gly-Leu-Met-NH2 and its N-acethyl derivative (Ac-ANKA), Ac-His-Lys-Thr-Asp-Ala5-Phe-Val-Gly-Leu-Met-NH2 werecarried out. For the ANKA and Ac-ANKA the additional deprotonation was not observed. It suggests that for thetachykinin peptides with C-terminal sequence of neurokinin A for the additional deprotonation the presence ofthe serine residue is necessary. For the Cu(II)-ANKA 1:2 system at physiological pH 7.4 the CuH2L2 species is pres-ent with histamine-like 4 N, 2×{NH2,NIm} coordination mode.With increasing pH the deprotonation and coordi-nation of amide nitrogen atoms occur and the CuH-2L, CuH-3L complexes are formed. In pH range 4.5 – 9.5the dimeric Cu2HL2, Cu2L2 and Cu2H-1L2 species in solution are also present. To elucidate the products of thecopper(II)- catalyzed oxidation of the ANKA and Ac-ANKA, the liquid chromatography-mass spectrometry(LC-MS) method and Cu(II)/hydrogen peroxide as a model oxidizing system were employed.In the presence of hydrogen peroxide with 1:1 peptide-H2O2 molar ratio for both peptides the oxidation ofthe methionine residue to methionine sulfoxide was observed. For the Cu(II)-peptide-hydrogen peroxidein 1:2:2 molar ratio systems oxidations of the histidine residues to 2-oxo-histidines and methionine sulfoxideto methionine sulfone were detected.
Tachykinins constitute a family of multifunctional neuropeptideswhose signaling mechanisms seem to be partially conserved throughevolution [1–5]. Although the tachykinin peptides display only limitedsequence identities when comparing invertebrates and mammals,their G-protein-coupled receptors (GPCRs) display more striking simi-larities, suggesting ancestral relationships [4,5]. The three principalmammalian tachykinins, substance P, neurokinin A and neurokinin B,are processed from two precursors, preprotachykinin A and B andthey act with preferential affinities on three different GPCRs, NK1 –
NK3 [5]. Tachykinins are a family of biologically active peptides distrib-uted in the central and peripheral nervous system. Tachykinins elicit awide and complex array of biological responses, such as the stimulationof extravascular smooth muscle, powerful vasodilation, hypertensiveaction, activation of immune system, regulation of pain transmission,and neurogenic inflammation [6–8]. The wide range of physiological
48 71 328 23 48.i.wroc.pl
rights reserved.
activity of tachykinins has been attributed to the lack of specificity oftachykinins for a particular receptor type [9].
Neurokinin A (NKA) is a decapeptide found in mammalianneuronal tissue, with the sequence His-Lys-Thr-Asp-Ser5-Phe-Val-Gly-Leu-Met-NH2. It was first isolated from the porcine spinalcord [10]. In water NKA prefers to be in an extended chain confor-mation whereas a helical conformation is induced in the centralcore and the C-terminal region (D4 – M10) of the peptide in thepresence of perdeuterated dodecylphosphocholine (DPC) micelles,a membrane model system [11]. The results of the structure – activitystudies performed by Rovero and collegues and those of otherworkers have indicated the Asp4 and Phe6 of NKA are importantfor activation of the NK2 receptor [12–15]. Results of analogues ofneurokinin A mediating contraction of rat uterus suggest that theHis at position one of NKA is of little or no importance in contractileactivity, while substitution or truncation of Lys at position two orsubstitution of Thr at position three causes a decrease in potency[16].
In rats, neurokinin A has been detected across brain regions [17]and some effects on motor behaviour and nociception have beendemonstrated [18,19]. It has been suggested that neurokinin A mayhave a neurotransmitter/neuromodulator role in the substantia nigra
2 E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
of rats [20]. Themeasurement of cerebrospinal fluid (CSF) neurokinin Aconcentration, in a larger sample of patients, together with brain tissuestudies of this neuropeptide in Parkinson's disease, may demonstrate asignificant neurokinin A reduction in Parkinsonian patients and indicatethe involvement of neurokinin A – containing neurons in this disease[21].
Oxidative stress is considered amajor contributor to the pathogenesisof a number of pathological processes leading to atherosclerosis, inflam-matory conditions, multiple system atrophy and several neurode-generative diseases [22–24]. The reactive oxygen species (ROS)are formed in biological systems by both enzymatic and metal-catalyzedoxidation (MCO) reactions [25,26]. A feature of metal-catalyzed oxida-tions is the site-specific nature of the reaction, i.e. specific amino acidresidues located at the metal-binding sites are generally altered [27].Therefore, the knowledge of the products of metal-catalyzed oxidationmay suggest the metal-binding sites [28–30]. Most susceptible to metal-catalyzed oxidation are His, Met, Cys [27,31] because of their abilityto form complexes with metals such as Cu(II) or Fe(III). The reactiveoxygen species are generated, and oxidation of specific amino acidresidues occurs in what is reffered to as a “caged” process [32]. TheMCO reaction involves reduction of Fe(III) or Cu(II) by a suitableelectron donor such as NADH, NADPH, mercaptane, or ascorbate.Fe(II) or Cu(I) ions bound to specific metal-binding sites on proteinsreact with H2O2 to generate .OH [33,34]. The mechanism of genera-tion for the active species from Cu(II)/peptide/ H2O2 system hasbeen considered (reactions (1) – (3)) [35,36].
The present paper reports the results of combined spectroscopic andpotentiometric studies on the copper(II) complexes of the modifiedneurokinin A and its analogue with N-terminal amine protected groupby acethylation. The peptides involved in the study are: neurokinin A(ANKA) HKTDA5FVGLM-NH2 and Ac-neurokinin A (Ac-ANKA) Ac-HKTDA5FVGLM-NH2, both peptides contain point mutation (S5A). Ourstudies on the acid-base properties and coordination abilities towardsto copper(II) of the neurokinin A [37], neuropeptide gamma and itsfragments [38–40], and neuropeptide K fragments [41] (all peptidescontain the neurokinin sequence in their C-terminal), clearly indicatethe presence of additional deprotonation of the ligands studied. TheCID MS/MS analysis (collision-induced dissociation tandem massspectrometry analysis) of the neurokinin A [30] indicates thatSer5 may be particularly responsible for this additional deproton-ation. To support this suggestion we synthesized the neurokininA and its N-acetyl derivative with point mutation (S5A). The acid-baseproperties of these ligands were determined by potentiometricstudies. The metal-catalyzed oxidation of the ANKA and Ac-ANKAby the Cu(II)/ H2O2 system demonstrate the relationship betweenthe binding sites of copper(II) ions and the oxidation products ofthe ligands studied.
2. Material and methods
2.1. Synthesis of the peptides
Synthesis of ANKA with point mutation (S5A), His-Lys-Thr-Asp-Ala5-Phe-Val-Gly-Leu-Met-NH2, and its acethyl derivative (Ac-ANKA)peptides were carried out using Millipore 9050 peptide synthesizerand continuous-flow methodology [42–44]. A polystyrene/polyethylene
glycol copolymer resin (TentaGel R RAM, Rapp Polymere) was used as asolid support. The peptidyl-resin was divided into two portions; one ofthem was subjected to 1 M acetylimidazole in DMF to acetylate theN-terminal amino group of the peptide and produce Ac-ANKA. Bothpeptides were cleaved from the resin and deprotected by treatmentwith a mixture containing 94.0% of trifluoroacetic acid (TFA), 2.5% ofH2O, 2.0% of triisopropylsilane (TIS) and 1.5% of 1,2-ethanedithiol (EDT).The cleavage reaction was carried out for 1.5 h at room temperature.
The crude peptides were purified by reversed-phase (RP) HPLCusing a Luna C8 semi-preparative column (21.2×250 mm, 100 Å,5 μm, Phenomenex). The purity of the peptides was confirmed bymatrix-assisted laser desorption/ionization time-of-flight mass spec-trometry (MALDI-TOF MS) and analytical RP-HPLC using a C8 Kromasilcolumn (4.6×250 mm, 5 μm) and 30 min. linear gradient of 5-80%acetonitrile (ACN) in 0.1% aqueous trifluoroacetic acid as a mobilephase.
Analytical data were as follows: ANKA: RT=15.4 min, MS: [L]+=1117.3, calc. 1117.3; Ac-ANKA: RT=15.6 min, MS: [L]+·=1159.3,calc. 1159.4.
The purity of the peptides was checked and the exact concentra-tion of their stock solutions was determined by the Gran method [45].
2.2. Potentiometric measurements
Stability constants for proton and Cu(II) complexes were calculatedfrom pH-metric titrations carried out in an argon atmosphere at 298 Kusing a total volume of 1.9 – 2 ml. Alkali was added from a 0.250 mlmicrometer syringe which was calibrated by both weight titration andthe titration of standardmaterials. Experimental details: ligand concen-tration 1.5×10-3 M, metal-to-ligand molar ratio 1:2; ionic strength0.10 M (KNO3); Cu(NO3)2 used as the source of the metal ions; pH-metric titration on a MOLSPIN pH-meter system using a RussellCMAW 711 semimicro combined electrode calibrated in concentra-tion using HNO3 [46] , number of titrations=2; method of calcula-tion SUPERQUAD [47]. The samples were titrated in the pH region2.5 – 10.5. Standard deviations (values) quoted were computed bySUPERQUAD and refer to random errors only. They are, however, agood indication of the importance of the particular species involvedin the equilibria. The precipitation for the Cu(II)-ANKA 1:1.1 systemwas observed, therefore, the 1:2 metal-to-ligandmolar ratio was inves-tigated. However, the Cu(II)-Ac-ANKA system could not be studied be-cause of precipitation for 1:1, 1:2 and 1:higher metal-to-ligand molarratios.
2.3. Spectroscopic measurements
Solutions were of similar concentrations to those used inpotentiometric studies. Absorption spectra (UV-visible) were recordedon a Cary 50 “Varian” spectrophotometer in the 850 – 300 nm range.Circular dichroism (CD) spectra were recorded on a Jasco J-715spectropolarimeter in the 750 – 250 nm range. The values of Δε(i.e., εl - εr) and ε were calculated at the maximum concentrationof the particular species obtained from potentiometric data. Electronparamagnetic resonance (EPR) spectra were performed in an ethyleneglycol-water (1:2, v/v) solution at 77 K on a Bruker ESP 300E spectrome-ter equipped with the ER 035 M Bruker NMR gaussmeter and the HP5350B Hewlett-Packard microwave frequency counter at the X-bandfrequency (~9.45 GHz). The spectra were analyzed by using Bruker'sWIN-EPR SimFonia software, version 1.25. Copper(II) stock solutionwas prepared from Cu(NO3)2×3 H2O.
2.4. ESI-MS measurement
The mass spectra were obtained on a Bruker MicrOTOF-Qspectrometer (BrukerDaltonik, Bremen, Germany), equippedwithApolloII electrospray ionization source. The mass spectrometer was operated in
Table 1Formation constants (logβ) and protonation constants (logK) for ANKA and Ac-ANKAand comparable peptides at 298 K and I=0.10 M (KNO3).
aref. [37];b pentapeptide with sequence His-Ser-Asp-Gly-Ile, ref. [49];c tripeptide with sequence His-Ala-Ala, ref. [50];d dipeptide His-Gly, ref. [59].
3E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
the positive or negative ion mode. The instrumental parameters were asfollows: scan range m/z 400–2300, dry gas - nitrogen, temperature200 °C, reflector voltage 1300 V, detector voltage 1920 V. The samples(1:2 Cu(II) – ANKA, Cu(II)-Ac-ANKA molar ratio) were dissolved inwater and pH value ~7 was adjusted by the addition of concetrated ofNaOH or HNO3, and infused at a flow rate of 3 μL/min. The instrumentwas calibrated externally with the Tunemix™mixture (Bruker Daltonik,Germany) in quadratic regression mode.
2.5. Materials used in the oxidation process
Deionized and triply distilled water was used, and theMOPS buffer atpH 7.4 (Sigma-Aldrich, MOPS [3-(N-morpholino)propanesulfonic acid][48] was treated with Chelex 100 resin (sodium form, Sigma-Aldrich) toremove trace metals. Hydrogen peroxide was purchased from Fluka(Perhydrol, 30%), and EDTA and Cu(NO3)2 were purchased from POCH.Stock solutions (0.10 M) of EDTA and hydrogen peroxide in MOPS bufferwere prepared.
2.6. Oxidation of the neurokinin A with point mutation (S5A) and itsacethyl-derivative and LCMS analysis
Copper(II)-catalyzed oxidation of the peptide in the presence ofhydrogen peroxide was monitored by analytical RP-HPLC on aVarian ProStar 240 station using an XTerra C 18 4.6×150 mmcolumn (Waters) at a 30 min linear gradient of 5-100% B, where Aused 0.1% aqueous TFA and B used 0.1% TFA in 80% ACN. A reactionmixture (0.2 cm3) containing 5×10-4 M peptide and a metal-to-ligandmolar ratio of 1:2 in a 0.02 M MOPS was incubated at 37 °C for 24 h inthe presence of hydrogen peroxide at a metal to hydrogen peroxidemolar ratio of 1:2 for the ANKA and Ac-ANKA. The reaction was startedby the addition of hydrogen peroxide, which was freshly prepared. Afterincubation, the reaction was stopped by the addition of EDTA to a finalcomplex at an EDTAmolar ratio of 1:5. The chelating agent EDTA inhibitsthe oxidation of the peptide by removing Cu(II) from the peptide.Oxidized and digested peptides were desalted on 10 μL ZipTipC18columns (Omnix, Varian). The columns were prepared by wetting with50% acetonitrile and equilibrated with 0.1% trifluoroacetic acid. Eachsample was loaded into a ZipTip column. The column was washed with0.1% TFA to remove salts, and then the peptides were eluted with 0.1%formic acid in 80% acetonitrile. The obtained samples were thenthe subject of LC-ESI-MS analysis. Acetonitrile, water, and formicacid of LC/MS grade were purchased from Sigma. Positive ionelectrospray mass spectrometric analysis was carried out using aShimadzu ion trap time-of-flight mass spectrometer (LC-MS ITTOF) at unit resolution. The source temperature was 200 °C, theelectrospray voltage was -1700 V. The separation and mass analy-sis of oxidized and digested peptides were carried out using aPhenomenex Jupiter Proteo90A analytical column (2×150 mm, 4 μm)with a linear gradient of 0-30% B for 12.5 min followed by a gradientof 30-100% B for 7.5 min (buffer A, 0.2% formic acid/water; buffer B,0.2% formic acid / ACN; flow rate 0.2 mL / min). The injection volumewas 80 μL, and the temperature in which the analysis proceeded was40 °C. Data were acquired and analyzed using LC Solution software pro-vided by Shimadzu.
3. Results and discussion
3.1. Protonation constants
Protonation constants (logβ, log K values) for the peptides studiedand comparable ligands are given in Table 1. The neurokinin A and itsN-acethyl derivative with point mutation (S5A) contain four (H4L)and three H3L dissociable protons, respectively (where L means ligandand HnL fully protonated ligand). For both peptides additional depro-tonation with pK value 9 – 10 was not observed in comparison to
those of neurokinin A and other peptides containing this peptide se-quence in C-terminal [37–41]. The protonation constants of ANKA forthe amine and the imidazole nitrogen atoms are comparable tothose of the peptides containing the N-terminal His residue (NKA [37],HSDGI [49], HAA [50]). However, the protonation constants (logK) ofthe imidazole nitrogen atoms for the N-acethyl derivatives of ANKAand NKA (Table 1) are one order of magnitude higher in comparison tothose with non-blocked amine group suggesting the involvement in hy-drogen bond the N-terminal amine group and the imidazole nitrogen ofhistidine residue. The protonation constants (log K 3.72 – 3.75, Table 1)the β-carboxylate of the Asp residue in both peptide amides are similarto each other, and they are close to those expected for comparable pep-tides [51,52].
As it is seen in Table 1, for the ANKA ligand the log K value for thelysyl amino group is one order of magnitude higher compared to thatof NKA. This difference may suggest the involvement of this group inthe formation of hydrogen bond. The protonation constants obtainedfor both peptides (ANKA and NKA) suggest the presence of hydrogenbonds in NKA peptide whichmost likely are going to the deprotonationof hydroxyl group of the serine residue. These results may also suggestthat for the tachykinin peptideswith C-terminal sequence of neurokininA for the additional deprotonation the presence of the serine residue isnecessary.
Serine proteases comprise nearly one-third of all known proteasesidentified to data and play crucial roles in a wide variety of cellular aswell as extracellular functions, including the process of blood clotting,protein digestion, cell signaling, inflammation, and protein processing[53]. The hallmark of these proteases is presence the so-called“classical” catalytic Ser/His/Asp triad. Although the classical serineproteases are the most widespread in nature, there exist a variety of“non-classical” serine proteases as the triads Ser/His/Glu, Ser/His/His,Ser/Glu/Asp and the dyads Ser/Lys and Ser/His. It seems that neurokininA may be one of the serine proteases. Spectroscopic properties ofchymotrypsin and model compounds indicate that a low-barrier hy-drogen bond participates in the mechanism of serine protease action[54]. Chemical exchange saturation transfer (CEST) is a unique NMRpulse sequence whose ability to detect rapidly exchanging protonshas been largely overlooked in biomolecular NMR. Using CEST J.T.Stivers and coworkers for the first time observed the Hγ proton ofSer195 in chymotrypsinogen at neutral and basic pH values [55].This proton is highly deshielded in the resting enzyme at this pHrange due to its hydrogen bond with His-Nε2, indicating that theSer195-Oγ is alkoxide – like and preactivated for nucleophilic attack in
Table 2Stability constants and calculated logK⁎e values of copper(II) complexes of ANKA and comparable peptides at 298 K and I=0.10 M (KNO3).
a,b,c,dReferences as in Table 1.eLog K*=log β (CuHjL) - log β (HnL) (where the index j corresponds to the number of protons in the coordinated ligand to metal ion and n corresponds to the number of protonscoordinated to the ligand).
4 E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
the free enzyme. It seems that also deprotonation of the Ser residue inthe tachykinin peptides may occur because of the involvement this res-idue in hydrogen bonds (NH2-terminal, NIm-His, NH2-Lys). It should bementioned that we start to perform theoretical calculations for ligandmolecules to obtain additional information about structures the moststable conformers in different pH values of solution.
Potentiometry detects a range of Cu(II) complexes of ANKA withthe formation constants reported in Table 2. The stoichiometry ofcopper(II) complexes as CupHqLr is given, where p is number ofmetal ions, q number of protons and r number of ligand moleculesin the complex formed. The values of log K* are the protonationcorrected stability constants which are useful to compare the abilityof ligands to bind a metal ion [56]. The spectroscopic properties ofthe major complexes are given in Table 3.
3.2. Copper(II) complexes
Because of precipitation in the solution containing Cu(II)-ANKA1:1 molar ratio the solutions 1:2 were investigated. According topotentiometric and spectroscopic results the neurokinin A with pointmutation (S5A) forms with Cu(II) ions the CuH2L, CuHL, CuH-2L,CuH-3L, CuH2L2, Cu2HL2, Cu2L2 and , Cu2H-1L2 complexes (charges omit-ted for simplicity, Table 2). The coordination of the metal ion starts atpH 2.5 (Fig. 1) and the CuH2L complex is formed. The spectroscopic pa-rameters for the CuH2L complex cannot be estimated because of theoverlap of several species and its low concentration. However, theCuH2L complex can be interpreted, on the basis of its stoichiometry
Table 3Spectroscopic data for copper(II) complexes of ANKA at 1:2 metal-to-ligand molar ratio.
LigandSpecies
UV-Vis
pH λ (nm) ε (M-1 cm-1)
ANKACuHL{NH2, NIm}
4.4 649a 75
CuH2L22x{NH2, NIm}
7.4 585a 119
CuH-2 LCuH-3 L{NH2, 3 N-}
9.6 506a 141
ad-d transition,b NIm→Cu(II) charge transfer transition,c NH2→Cu(II) charge transfer transition,d N-
amide→Cu(II) charge transfer transition.
and stability constants, either by coordination of the terminal amine orbymono-dentate coordination of an imidazole nitrogen atom (1 N com-plex). The log K* value for the CuH2L complex of the neurokinin A withpoint mutation (S5A) is comparable to those of the NKA, HSDGI, HAAcontaining His residue on first position of the peptide chain (Table 2).The next species, the CuHL, is detected in the wide 3 – 7 pH range(Fig. 1). The d-d transition energy at 649 nm and the presence in CDspectra of the NIm→Cu(II) and NH2→Cu(II) charge transfer transitionsat 300 nm and 268 nm, respectively, the EPR parameters AII 172 G, gII2.290 (Fig. 2, pH 4.1) are consistent with the histamine-like coordina-tion {NH2,NIm} of the ligand studied (Scheme 1) [37,49,50,57–59].Coordination of histidine normally gives charge transfer transitionsetc. at 330 nm [π1 NIm-Cu(II)]. The magnitude and precise energy ofthe charge transfer transitions for an imidazole nitrogen to Cu(II) arevery sensitive to the position of the ring plane relative to the complexplane, and thus to the possibility of its rotation [60]. Value of log K* forthe CuHL complex with {NH2,NIm} binding site of neurokinin A withpoint mutation (S5A) (-3.94) is comparable to those of the neurokininA, HSDGI, HAA, and HG (Table 2).
With increasing pH above 4.5 the CuH2L2 is formed with maximumof its concentration at pH 7. One protonation constant of the Lys residuewas detected, therefore, the CuH2L2 species will be complex with fournitrogen atoms 4 N coordinated to copper(II) ions by two ligand mole-culeswith histamine-like 2×{NH2,NIm} coordinationmode. The EPRpa-rameters AII 182 G, gII 2.233 obtained for the Cu(II) – ANKA system at1:2 metal-to-ligand molar ratio at pH 7.2 (Fig. 2d) are similar to thoseof the CuL2 complex of histidine (AII 184 G, gII 2.235). Moreover, in
Fig. 1. Species distribution curves for the Cu(II) complexes of neurokinin A with point mutation (S5A). Cu(II) to peptide molar ratio 1:2, [Cu(II)]=0.001 M.
5E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
UV-visible spectra the d-d transition energy at 585 nmand the presencein CD spectra of the d-d transition at 709 nm (708 nm for the CuL2 com-plex of histidine) and the charge transfer transitions of the NIm→Cu(II)and NH2→Cu(II) at 300 nm and 271 nm (Table 3), respectively, itstrongly supports the 4N 2×{NH2,NIm} binding sites (Scheme 1). ThelogK* value for the CuH2L2 complex is about 0.8 log units higher in com-parison to that of the CuL2 complex of the HG peptide (Table 2). Thisstabilization for the CuH2L2 complex of the ANKA peptide may resultfrom the presence of the inter- and/or intra molecular interaction(hydrogen bonds) [61]. At pH 8.5 (Fig. 1) the deprotonation andcoordination of the amide nitrogens occur and the CuH-2L complexis formed (Scheme 1). The d-d transition energy at 506 nm, the pa-rameters of CD spectra and AII 210 G, gII 2.170 (Fig. 2) correspondvery well to the four nitrogen atoms 4 N with {NH2,3N-} coordina-tion mode (Table 3). These parameters of UV-visible, CD and EPRspectra are not altered in basic solution supporting the same coordina-tion binding in the CuH-3L complex. Moreover, the deprotonation of
2400 2500 2600 2700 2800
Magnetic fi
ab
c
d
e
f
Fig. 2. X-band EPR spectra of 1:2 Cu(II)-neurokinin A with point mutation (S5A) frozen solmental; (d) 7.2, experimental; (e) 10.4, experimental; (f) 10.4, simulated.
the CuH-2L complex resulting in the CuH-3 L species with pK 10.68may correspond to deprotonation of non-coordinated ε-amino groupof the lysine residue (10.70 in free ligand, Table 1). The value of log K*for the 4N species with {NH2,3N-} binding site of modified neurokininA is comparable to that of neurokinin A but it is by 1.65 orders of magni-tude higher in comparison to that of pentaalanine amide [62]. This stabi-lization may result from the axial interaction of the histidine residue inthe CuH-2L complex as it was suggested for the HGG (His-Gly-Gly),HAA (His-Ala-Ala)[50] and neurokinin A [37].
The EPR spectra obtained for the system studied clearly indicatethat at pH 5 – 9 the dimeric species exist in the solution (Fig. 2).The EPR signal intensity is reduced and broad, which suggests apossible antiferomagnetic interaction among copper(II) ions in thecomplexes formed (Fig. 2c and d). According to the Cu(II) complexesof HSDGI (His-Ser-Asp-Gly-Ile) [49], HG (His-Gly) [59], HM (His-Met)[57] peptides in the potentiometric data calculations the dimericspecies the Cu2HL2, Cu2L2, Cu2H-1L2 were taken. These species
2900 3000 3100 3200 3300
eld [G]
ution at 77 K at different pH: (a) 4.1, experimental; (b) 4.1, simulated; (c) 5.5, experi-
NIm
H2N
Cu2+
N-
C=O
M10-NH2
N-
C=O
Cu2+
NIm
NH2H1
M10-NH2
Cu2L2
(e)H1
K2
K2
NH2
NIm
Cu2+
OH2
NIm
M10-NH2
C=O
N-
Cu2+
NH2
OH2
M10-NH2 H1
K2
Cu2HL2
(d)
H1
Cu2+
H2O
H2N
OH2
NIm
H1
M10–NH2
CuHL(a)
NH2
NIm
H1Cu2+
NH2
NIm
H1
M10-NH2
M10-NH2
CuH2L2
(b)
H2N
N-
H1Cu2+
N-
N-
K2
T3
M10-NH2
CuH-2LCuH-3L
(c)
Scheme 1. Binding modes of the species formed in the copper(II)-neurokinin A with point mutation (S5A) system.
6 E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
cannot be characterized by UV-vis and CD spectra because of theoverlap of several species and their low concentrations (Fig. 1). Omis-sion of these species from the ANKA equilibria for the 1:2 Cu(II)-ANKAmolar ratio caused a significant deterioration in the goodness of fit be-tween experimental and measured pH values in the pH range 4.5 – 9.5.The precipitation for the metal-to-ligand 1:1 molar ratio was observed,and it is most likely because of the dimeric species which are then dom-inant. According to the others peptides containing N-terminal histidineresidue in the peptide sequence (HG, HM, HGG, HAA) for the Cu2HL2(CuHL, CuH-1LH), Cu2L2 (CuH-1LH, CuH-1LH) and Cu2H-1L2 (CuH-1LH,CuH-2LH) species binding sites may be proposed (Scheme 1).
Mass spectra for the Cu(II)-ANKA 1:2 molar ratio solution at pH ~7revealed atm/z values 597.8 Da and 1194.5 Da doubly and singly chargedmolecular ions [CuL]2+ and [CuH-1L]+, respectively (Fig. 3).
The solution for MS measurements of Cu(II)-Ac-ANKA 1:2 systemwas prepared similar to that for the oxidation studies (0.00025 M ofCu(II) ions) at pH ~7. The obtained ESI-MS spectrum for this systemshows a dominant signals for the [CuL]2+ and [CuH-1L]+ specieswith m/z values 611.5 and 1221.9 Da, respectively (Fig. 4).
597.8
598.3
598.7
599.3
597.8
598.3 598.7
599.3
0.0
0.5
1.0
1.5
x108
Inte
ns.
0
500
1000
1500
2000
598.0 598.5 599.0
Fig. 3. ESI mass spectrum for the Cu(II)-ANKA at 1:2 molar ratio system in water solutio
ESI-MS has been used in a wide variety of fields to study formation,stoichiometry and speciation of complexes of metals and organic ligands[63,64].
3.3. Oxidation of neurokinin A with point mutation (S5A) and its N-acethylderivative
Oxidation of cellular components is believed to play a major role indestruction of cellular integrity, and has been demonstrated to causedamage to specific cellular components including lipids, DNA, andproteins [27,65]. Protein oxidation, in particular, has been reportedto alter susceptibility to proteolysis and decrease enzymatic activity,which certainly has profound physiological implications [27,66]. Thestudies that indicate accumulation of oxidized proteins with age [35]have led to the hypothesis that protein oxidation is an importantcontributor to the aging process [67]. Furthermore, in numerous dis-eases including Alzheimer's disease, senile cataract, muscular dystrophy,atherosclerosis, rheumatoid arthritis the products of protein oxidationhave been shown to be present [68–70]. Metal-catalyzed oxidation(MCO) is believed to be a primary cause of biomolecular oxidation
599.8 600.3 600.8
599.8
599.5 600.0 600.5 601.0 m/z
experimental
simulated
n at pH~7. The experimental and simulated spectra for the [CuL]2+ molecular ion.
Fig. 4. ESI mass spectrum for the Cu(II)-Ac-ANKA at 1:2 molar ratio system in water solution at pH~7. The experimental and simulated spectra for the [CuH-1 L] molecular ion.
7E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
occuring in many cells. Because MCO components are prevalent in bio-logical systems, it is important to understand their effects on biologicalstructure and function. In most cases, the interaction of a metal with aprotein is through the geometry of a specific metal-binding site createdthrough the coordination of the metal to selected amino acid residuessuch as His, Cys, and Met. The individual amino acids involved in metalbinding to a protein can be conveniently identified through site-specificmetal-catalyzed oxidation [71–73]. Although, the potentiometric andspectroscopic studies cannot be performed because of precipitation forthe Cu(II)-Ac-ANKA system (concentration of copper(II) ions 0.001 M)the metal-catalyzed oxidation with the concentration 0.00025 M of thecopper(II) with metal-to-ligand 1:2 molar ratio was made.
In the presence of copper(II) ions and the peptide alone the ANKApeptide was stable after 24 h incubation in 0.02 M MOPS buffer at pH
Table 4Products of copper(II)-catalyzed oxidation of ANKA and Ac-ANKA analyzed be LCMS spectr
Determination of peptide modification Charge
ANKA, 24 h 23
ANKA+H2O2 (1:1), 24 hOxidation of Met→sulfoxide 2
3Cu(II)-ANKA+H2O2 (1:2:2), 24 hOxidation of Met→sulfoxide 2
3Oxidation of Met→sulphone 2Oxidation of Met→sulphone and His→2-oxo-His 2
3Loss of CH3SOH fragment from peptide 2
3Loss of CH3SOH fragment from peptide and oxidation of His→2-oxo-His 2
Ac-ANKA, 24 h 2Oxidation of Met→sulfoxide 2Ac-ANKA+H2O2 (1:1), 24 hOxidation of Met→sulfoxide 2Cu(II)-Ac-ANKA+H2O2 (1:2:2), 24 hOxidation of Met→sulfoxide 2
3Oxidation of Met→sulphone 2Oxidation of Met→sulphone and His→2-oxo-His 2Loss of CH3SOH fragment from peptide 2Loss of CH3SOH fragment from peptide and oxidation of His→2-oxo-His 2Cleavage of H1-K2 peptide bond, (K2-M10) fragment 2
7.4. The Ac-ANKA derivative underwent the oxidation of the Metresidue to the Met-sulfoxide after 24 h incubation in 0.02 M MOPSbuffer at pH 7.4. For the Ac-ANKA peptide in a chromatographic frac-tion eluting at 9.8 min, a doubly charged molecular ion [L+2H]2+
withm/z 588.6 Da is present (Table 4). In the presence of the copper(II)ions and H2O2 the peptides studied underwent degradation. The spec-troscopic data for the copper(II) complexes of the ANKA inMOPS bufferat pH 7.4 are similar to those in aqueous solution at pH 7.4 (data notshown, Table 3). In the Cu(II)-ANKA 1:2 system at pH 7.4 the copper(II)ions are coordinated by the imidazole and amine nitrogen atoms of theHis1 residue, therefore, the modifications of this residue are expected.For the Ac-ANKA according to the N-blocked peptides containingthe histidine residue the coordination of copper(II) ions by this res-idue occurs [74,75]. For the 1:2:2 Cu(II)-ANKA (or Ac-ANKA) – H2O2
8 E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
system in chromatographic fraction eluting at 12.6 min (7.6 min) a dou-bly chargedmolecular ions [L+2H]2+withm/z 575.6 (596.5) Da are ob-served (Table 4). This +32 Da product results from the oxidation of Metresidue to sulfone. The enzyme methionine sulfoxide reductase selec-tively reduces the Met-sulfoxide and not the Met-sulfone [76]. Thus,any potential preferred formation of the Met-sulfone might lead to theaccumulation of peptide/protein oxidation products in tissue. The oxida-tion of methionine in peptides and proteins represents an importantposttranslational modification under conditions of oxidative stress [77]and aging [78]. To support the formation and presence of methioninesulfoxide the molecular ions with a loss of 64 Da (CH3SOH) wereobserved [71]. Mass spectrometry for the chromatographic fraction ofCu(II)-ANKA-H2O2 system (Cu(II)-Ac-ANKA-H2O2) eluted at 8.9 min(11.4 min) yielded doubly charged molecular ions with m/z 536.0(557.0) Da which may be assigned to the peptides with loss of CH3SOH.Histidine is susceptible to radical oxidation, leading to very complexprod-ucts (aspartic acid, 2-oxo-histidine) [79,80]. Mass spectrometry for thechromatographic fraction eluting at 8.9 min for the Cu(II)-ANKA-H2O2,1:2:2 system(and10.5 min for theCu(II)-Ac-ANKA-H2O2, 1:2:2) containsthe doubly charged molecular ions with m/z 583.6 (605.5) Da (Table 4,Fig. 5). The masses of these oxidation products are +32 Da higher thanthe parent oxidized with methionine to a sulfoxide peptide suggestingfurther oxidation of the methionine sulfoxide to sulfone and histidine to2-oxo-histidine. In experimental conditions for the Cu(II)-Ac-ANKA-H2O2
with molar ratio 1:2:2 the fragmentation by the cleavage of the peptidebonds near the His residue was observed (Table 4). Other molecularions of copper(II)-catalyzed oxidation products for both systems studiedwere also detected and the modifications of these peptides in Table 4are proposed.
4. Conclusions
For the Cu(II)-ANKA 1:1 metal-to-ligand molar ratio the precipitationat pH 4.5 – 5 was observed, therefore, the 1:2 system was studied. At pH3.5 – 5.5 the CuHL complex with histamine {NH2,NIm} type coordinationmode predominates. In the wide pH range 5.5 – 9.5 the formation of theCuH2L2 species was found in which the coordination of copper(II) ionsis 4N, 2×{NH2,NIm} binding sites. With increasing pH the deprotonationand coordination of amide nitrogen atoms occur and the CuH-2L andCuH-3L complexes with 4 N {NH2,3 N-} coordination mode are formed.In pH range 4 – 9.5 the dimeric Cu2HL2, Cu2L2, and Cu2H-1L2 specieswere also found, and it is most likely that because of these dimericcomplexes the precipitation was observed for 1:1 metal-to-ligand molarratio. Although, the spectroscopic characterization of the dimericcomplexes cannot be made (low concentration), by analogy toother systems containing the His residue on first position in thepeptide chain, in the dimeric complex Cu2L2 the copper(II) ions
575.0 580.0 585.0 590.00.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
Inte
ns.
(x1
00,0
00)
580.545
589.045588.547
[L+2H]2+
[L+O+2H]2+
Fig. 5. MS spectrum of chromatographic fraction eluting at retention time of 10.5 min of thewith oxidations of Met to sulfone and His to 2-oxo-His are detected.
are coordinated like to Gly-Gly dipeptide while fourth coordinationsite is occupied by the imidazole nitrogen atom, forming a bridgebetween two copper(II) ions. The Cu(II)-Ac-ANKA system cannotbe studied by the potentiometric and spectroscopic methods becauseof the precipitation observed even for 1:2 and higher metal-to-ligandmolar ratio. However, mass spectrometry (concentration of Cu(II) ions0.00025 Mand Cu(II)-Ac-ANKA1:2molar ratio) revealed the formationof the complexes.
For the neurokinin A with point mutation (S5A) and its N-acethylderivative the additional deprotonation was not detected. It suggeststhat the deprotonation at pH 9 – 10 observed for the neurokinin A andother tachykinin peptides containing C-terminal amino acid sequenceof neurokinin A may derive from the OH-hydroxyl group of the serineresidue. The deprotonation of this residue was already suggested andobserved for the serine proteases.
We can also conclude that the presence of the serine residue in thetachykinin peptides with the C-terminal sequence of neurokinin A isnecessary to have additional deprotonation.
For the neurokinin A with point mutation (S5A) and its acethylderivative - hydrogen peroxide at 1:1 molar ratio the oxidation ofthe methionine residue (M10) to methionine sulfoxide was observed.In the Cu(II)-peptide-hydrogen peroxide 1:2:2 molar ratio systems,oxidations of histidine to 2-oxo-histidine and the methionine sulfoxideto the methionine sulfone were observed. For the Cu(II)-Ac-ANKA-H2O2
1:2:2 system fragmentations by the cleavage of the peptide bonds nearthe His residue were also detected.
Acknowledgment
Financial support from the National Science Centre (NCN) forScientific Grant 2011/01/N/ST/02563 has been gratefully acknowledged.
References
[1] R.J. Nachman, G. Moyna, H.J. Williams, J. Zabrocki, J.E. Zadina, G.M. Coast,J. Vanden Broeck, Ann. N. Y. Acad. Sci. 897 (1999) 388–400.
[2] D.R. Nassel, Peptides 20 (1999) 141–158.[3] J. Vanden Broeck, H. Torfs, J. Poels, W. Van Poyer, E. Swinnen, K. Ferket, A. De Loof,
Ann. N. Y. Acad. Sci. 897 (1999) 374–387.[4] H. Satake, T. Kawada, K. Nomoto, H. Minakata, Zool. Sci. 20 (2003) 533–549.[5] J.N. Pennefather, A. Lecci, M.L. Candenas, E. Patak, F.M. Pinto, C.A. Maggi, Life Sci.
74 (2004) 1445–1463.[6] D.A. Groneberg, S. Harrison, Q.T. Dinh, P. Geppetti, A. Fischer, Curr. Drug Targets 7
(2006) 1005–1010.[7] C. Palma, Curr. Drug Targets 7 (2006) 1043–1052.[8] Y. Zhang, A. Berger, C.D. Milne, C.J. Paige, Curr. Drug Targets 7 (2006) 1011–1020.[9] S. Nakanishi, Annu. Rev. Neurosci. 14 (1991) 123–136.
[10] S. Kimura, M. Okada, Y. Sugita, I. Kanazawa, E. Munekata, Proc. Jpn. Acad. 59(1983) 101–104.
[11] I.R. Chandrashekar, S.M. Owsik, Biophys. J. 85 (2003) 4002–4011.[12] P. Rovero, V. Pestellini, N.-E. Rhaleb, S. Dion, N. Rouissi, C. Tousignant, S. Telemaque,
G. Drapeau, D. Regoli, Neuropeptides 13 (1989) 263–270.
595.0 600.0 605.0m/z
596.547595.539
605.541
[L+2O+2H]2+
[L+3O+2H]2+
Ac-ANKA after Cu(II)-catalyzed oxidation. The products of modified Ac-ANKA peptide
9E. Jankowska et al. / Journal of Inorganic Biochemistry 121 (2013) 1–9
[13] M.A. Cascieri, R.-R.C. Huang, T.M. Fong, A.H. Cheung, S. Sadowski, E. Ber, C.D. Strader,Mol. Pharmacol. 41 (1992) 1096–1099.
[14] S. Lovas, D.D. Smith, M. Murnin, J. Allen, R.F. Murphy, Biochem. Soc. Trans. 22(1993) 5S.
[15] D. Regoli, F. Nantel, Biopolymers 31 (1991) 777–783.[16] L. Fisher, J.N. Pennefather, Neuropeptides 32 (1998) 405–410.[17] H. Arai, P.C. Emson, Brain Res. 399 (1986) 240–249.[18] M.E. Hall, P. Grantham, J. Limoni, J.M. Stewart, Brain Res. 420 (1987) 82–94.[19] O. Laneuville, J. Dorais, R. Couture, Life Sci. 42 (1988) 1295–1305.[20] F.J. Diez Guerra, D.J.S. Sirinathsinghji, P.C. Emson, Neuroscience 27 (1988)
527–536.[21] R. Galard, R. Catalan, J. Montalban, A. Molines, F. Miquel, J.M. Castellanos, Eur. J. Clin.
Chem. Clin. Biochem. 30 (1992) 281–284.[22] K. Jomova, D. Vondrakova, M. Lawson, M. Valko, Mol. Cell. Biochem. 345 (2010)
91–104.[23] I. Dalle-Donne, D. Giustarini, R. Colombo, R. Rossi, A. Milzani, Trends Mol. Med. 9
(2003) 169–176.[24] S. Rivera-Mancia, I. Perez-Neri, C. Rios, L. Tristan-Lopez, L. Rivera-Espinosa,
S. Montes, Chem. Biol. Interact. 186 (2010) 184–199.[25] A.I. Bush, Curr. Opin. Chem. Biol. 4 (2000) 184–191.[26] H.R. Griffiths, Free Radic. Res. 33 (2000) S47–S58.[27] E.R. Stadtman, Free Radic. Biol. Med. 9 (1990) 315–325.[28] R. Srikanth, J. Wilson, C.S. Burns, R.W. Vachet, Biochemistry 47 (2008) 9258–9268.[29] J.D. Bridgewater, J. Lim, R.W. Vachet, Anal. Chem. 78 (2006) 2432–2438.[30] V. Sadineni, N.A. Galeva, Ch. Schoneich, Anal. Biochem. 358 (2006) 208–215.[31] A. Amici, R.L. Levine, I. Tsai, E.R. Stadtman, J. Biol. Chem. 264 (1989) 3341–3346.[32] E.R. Stadtman, Annu. Rev. Biochem. 62 (1993) 797–821.[33] E.R. Stadtman, B.S. Berlett, Drug Metab. Rev. 30 (1998) 225–243.[34] B.S. Berlett, E.R. Stadtman, J. Biol. Chem. 272 (1997) 20313–20316.[35] E.K. Hodgson, I. Fridovich, Biochemistry 14 (1975) 5294–5299.[36] J.I. Ueda, Y. Shimazu, T. Ozawa, Biochem. Mol. Biol. Int. 34 (1994) 801–808.[37] T. Kowalik-Jankowska, E. Jankowska, Z. Szewczuk, F. Kasprzykowski, J. Inorg.
Biochem. 104 (2010) 831–842.[38] M. Pietruszka, E. Jankowska, T. Kowalik-Jankowska, Z. Szewczuk, M. Smużyńska,
Inorg. Chem. 50 (2011) 7489–7499.[39] T. Kowalik-Jankowska, E. Jankowska, F. Kasprzykowski, Inorg. Chem. 49 (2010)
2182–2192.[40] E. Jankowska, M. Pietruszka, T. Kowalik-Jankowska, Dalton Trans. 41 (2012)
1683–1694.[41] M. Pietruszka, E. Jankowska, T. Kowalik-Jankowska, Inorg. Chem. http://
Rockford, IL, 1993.[43] The Milipore 9050 Plus Pepsynthesizer Operator's Guide, Milipore Corp., Miliford,
1992.[44] N. Sole, G. Barany, J. Org. Chem. 57 (1992) 5399–5403.[45] G. Gran, Acta Chem. Scand. 4 (1950) 559–577.[46] H. Irving, M.G. Miles, L.D. Pettit, Anal. Chim. Acta 38 (1967) 475–488.[47] G. Gans, A. Sabatini, A. Vacca, J. Chem. Soc. Dalton Trans. (1985) 1195–1199.
[48] E.N. Good, G.D. Winget, W. Winter, T.N. Connolly, S. Izawa, M.M.R. Singh,Biochemistry 5 (1966) 467–477.
[49] T. Kowalik-Jankowska, M. Jasionowski, L. Łankiewicz, J. Inorg. Biochem. 76 (1999)63–70.
[50] N.I. Jakab, B. Gyurcsik, T. Kortvelyesi, I. Vosekalna, J. Jensen, E. Arsen, J. Inorg.Biochem. 101 (2007) 1376–1385.
[51] J.F. Galey, B. Decock-Le Reverend, A. Lebkiri, L.D. Pettit, S.I. Pyburn, H. Kozłowski,J. Chem. Soc. Dalton Trans. (1991) 2281–2287.
[52] C. Kallay, Z. Nagy, K. Varnagy, G. Malandrinos, N. Hadjiliadis, I. Sovago, J. Bioinorg.Chem. Appl. (2007), http://dx.doi.org/10.1155/2007/30394, (Article ID 30394,9 pages).
[53] O.D. Kici, M. Paetzel, R.E. Dalby, Protein Sci. 17 (2008) 2023–2037.[54] P.A. Frey, S.A. Whitt, J.B. Tobin, Science 264 (1994) 1927–1930.[55] C.B. Lauzon, P. van Zijl, J.T. Stivers, J. Biomol. NMR 50 (2011) 299–314.[56] W. Bal, M. Dyba, H. Kozłowski, Acta Biochim. Pol. 44 (1997) 467–470.[57] I. Sovago, G. Potocz, J. Chem. Soc. Dalton Trans. 7 (1987) 1717–1720.[58] C.E. Livera, L.D. Pettit, M. Bataille, B. Perly, H. Kozlowski, B. Radomska, J. Chem.
Soc. Dalton Trans. 3 (1987) 661–666.[59] I. Sovago, E. Farkas, A. Gergely, J. Chem. Soc. Dalton Trans. (1982) 2159–2163.[60] T.G. Fawcett, E.E. Bernarducci, K. Krogh-Jespersen, H.J. Schugar, J. Am. Chem. Soc.
102 (1980) 2598–2604.[61] T. Kowalik-Jankowska, M. Ruta-Dolejsz, K. Wisniewska, L. Lankiewicz, J. Inorg.
Biochem. 92 (2002) 1–10.[62] W. Bal, H. Kozłowski, G. Kupryszewski, Z. Mickiewicz, L.D. Pettit, R. Robbins,
J. Inorg. Biochem. 52 (1993) 79–87.[63] M.J.A. Keith-Roach, Anal. Chim. Acta 678 (2010) 140–148.[64] M. Moini, Rapid Commun. Mass Spectrom. 24 (2010) 2730–2734.[65] J.A. Imlay, S. Linn, Science 240 (1988) 1302–1309.[66] E.R. Stadtman, Science 257 (1992) 1220–1224.[67] C.D. Smith, J.M. Carney, P.E. Starke-Reed, C.N. Oliver, E.R. Stadtman, R.A. Floyd,
W.R. Markedbery, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 10540–10543.[68] R.T. Dean, F. Shanlin, R. Stocker, M.J. Davies, Biochem. J. 324 (1997) 1–18.[69] D. Garland, Exp. Eye Res. 50 (1990) 677–682.[70] M.E. Murphy, J.P. Kehrer, Biochem. J. 260 (1989) 359–364.[71] J.D. Bridgewater, R.W. Vachet, Anal. Biochem. 341 (2005) 122–130.[72] J.D. Bridgewater, R.W. Vachet, Anal. Chem. 77 (2005) 4649–4653.[73] F. Zhao, E. Ghezzo-Schoneich, G.I. Aced, J. Hong, T. Milby, C. Schoneich, J. Biol.
Chem. 272 (1997) 9019–9029.[74] S. Timari, C. Kallay, K. Osz, I. Sovago, K. Varnagy, Dalton Trans. (2009) 1962–1971.[75] G. Di Natale, K. Osz, Z. Nagy, D. Sanna, G. Micera, G. Pappalardo, I. Sovago,
E. Rizzarelli, Inorg. Chem. 48 (2009) 4239–4250.[76] V.S. Sharov, D.A. Ferrington, T.C. Squier, C. Schoneich, FEBS Lett. 455 (1999)
247–250.[77] J. Moskovitz, B.S. Berlett, J.M. Poston, E.R. Stadtman, Proc. Natl. Acad. Sci. U. S. A.
94 (1997) 9585–9589.[78] M.H. Garner, A. Spector, Proc. Natl. Acad. Sci. U. S. A. 77 (1980) 1274–1277.[79] E.R. Stadtman, Methods Enzymol. 258 (1995) 379–393.[80] G. Xu, M.R. Chance, Chem. Rev. 107 (2007) 3514–3543.
