Covalent binding of styrene and styrene-7,8-oxide to plasma proteins, hemoglobin and DNA in the mouse

The extent of covalent binding to plasma proteins, hemoglobin and guanine-N-7 in DNA was determined after intraperitoneal administration of radiolabelled styrene and styrene-7,8-oxide to mice. The degree of alkylation increased non-linearly with the dose. It was proportionally higher after the highest doses of styrene-7,8-oxide while the reverse was observed with respect to the ability of styrene to alkylate plasma proteins and DNA. Thus, a dose dependence was indicated in the elimination of both styrene and styrene-7,8-oxide. A comparison of the degree of alkylation of plasma proteins, hemoglobin and guanine-N-7 in DNA suggests that the two compounds are about equally effective as alkylating agents in vivo at moderate dose levels. At high doses styrene-7,8-oxide is the more effective alkylator. The alkylation of DNA in liver, brain and lung after administration of styrene-7,8-oxide exceeded that in spleen and testis.     

Adenine N3 is a main alkylation site of styrene oxide in double-stranded DNA

Styrene 7,8-oxide (SO), a major metabolite of styrene, is classified as a probable human carcinogen. In the present work, salmon testis DNA was reacted with SO and the alkylation products were analysed after sequential depurination in neutral or acidic conditions followed by HPLC separation and UV-detection. A novel finding was that the N-3 position of adenine was the next most reactive alkylation site in double-stranded DNA, comprising 4% of the total alkylation, as compared to alkylation at the N-7 position of guanine, 93% of the total alkylation. Both alpha- and beta-products of SO were formed at these two sites. Other modified sites were N2-guanine (1.5%, alpha-isomer), 1-adenine (0.4%, both isomers) and N6-adenine (0.7%, both isomers) as well as 1-hypoxanthine (0.1%, alpha-isomer), formed by deamination of the corresponding 1-adenine adduct. The results indicated that in double-stranded DNA N-7 of guanine and N-3 of adenine account for 97% of alkylation by SO. However, these abundant adducts are not stable, the half-life of depurination in DNA for 3-substituted adenines being approximately 10 and approximately 20 h, for alpha- and beta-isomers, respectively, and 51 h for both isomers of 7-substituted guanines.     

Isotopically labeled crosslinking reagents: resolution of mass degeneracy in the identification of crosslinked peptides

Mass spectrometry in three dimensions (MS3D) is a newly developed method for the determination of protein structures involving intramolecular chemical crosslinking of proteins, proteolytic digestion of the resulting adducts, identification of crosslinks by mass spectrometry (MS), peak assignment using theoretical mass lists, and computational reduction of crosslinks to a structure by distance geometry methods. To facilitate the unambiguous identification of crosslinked peptides from proteolytic digestion mixtures of crosslinked proteins by MS, we introduced double 18O isotopic labels into the crosslinking reagent to provide the crosslinked peptides with a characteristic isotope pattern. The presence of doublets separated by 4 Da in the mass spectra of these materials allowed ready discrimination between crosslinked and modified peptides, and uncrosslinked peptides using automated intelligent data acquisition (IDA) of MS/MS data. This should allow ready automation of the method for application to whole expressible proteomes.     

The formation of styrene glutathione adducts catalyzed by prostaglandin H synthase. A possible new mechanism for the formation of glutathione conjugates

The metabolism of styrene by prostaglandin hydroperoxidase and horseradish peroxidase was examined. Ram seminal vesicle microsomes in the presence of arachidonic acid or hydrogen peroxide and glutathione converted styrene to glutathione adducts. Neither styrene 7,8-oxide nor styrene glycol was detected as a product in the incubation. Also, the addition of styrene 7,8-oxide and glutathione to ram seminal vesicle microsomes did not yield styrene glutathione adducts. The peroxidase-generated styrene glutathione adducts were isolated by high pressure liquid chromatography and characterized by NMR and tandem mass spectrometry as a mixture of (2R)- and (2S)-S-(2-phenyl-2-hydroxyethyl)glutathione. (1R)- and (1S)-S-(1-phenyl-2-hydroxyethyl)glutathione were not formed by the peroxidase system. The addition of phenol or aminopyrine to incubations, which greatly enhances the oxidation of glutathione to a thiyl radical by peroxidases, increased the formation of styrene glutathione adducts. We propose a new mechanism for the formation of glutathione adducts that is independent of epoxide formation but dependent on the initial oxidation of glutathione to a thiyl radical by the peroxidase, and the subsequent reaction of the thiyl radical with a suitable substrate, such as styrene.     

Styrene oxide-induced 2'-deoxycytidine adducts: implications for the mutagenicity of styrene oxide

The reaction between 2'-deoxycytidine and styrene 7,8-oxide (SO) resulted in alkylation at the 3-position and at the O(2)-position through the alpha- and beta-carbons of the epoxide but at the N(4)-position only through the alpha-carbon. The 3-alkylated adducts were found to deaminate to the corresponding 2'-deoxyuridine adducts (37 degrees C, pH 7.4) with half-lives of 6 min and 2.4 h for the alpha- and beta-isomers, respectively. The N(4)-alkylated products were stable at neutral pH. The O(2)-alkylated products were unstable being prone to depyrimidation and to isomerisation between alpha- and beta-isomers. In SO-treated double-stranded DNA, enzymatic hydrolysis allowed the identification of the beta3-deoxyuridine and alphaN(4)-deoxycytidine adducts (1.9 and 0.5% of total alkylation, respectively), in addition to the previously identified DNA-adducts. The 3-substituted uracil may have implications for the mutagenicity of SO.     

Ring test for the determination of N-terminal valine adducts of styrene 7,8-oxide with haemoglobin by the modified Edman degradation technique

A ring-test was organised between three laboratories using different versions of the modified Edman degradation technique for the gas chromatographic-mass spectrometric determination of N-terminal valine adducts of styrene 7,8-oxide. The analyses were performed on a sample of human haemoglobin reacted in vitro with styrene 7,8-oxide and on a set of five haemoglobin samples from mice dosed by i.p. injection of styrene. Strong correlations between the haemoglobin adduct determinations of the different laboratories were observed. However, covariance analysis revealed different slopes for the dose-response curves, indicating differences for the calibration of the reference globin or reference peptide.     

Investigation of styrene in the liver perfusion/cell culture system. No indication of styrene-7,8-oxide as the principal mutagenic metabolite produced by the intact rat liver

Mutagenic effect of styrene and styrene-7,8-oxide was studied with the isolated perfused rat liver as metabolizing system and Chinese hamster V79 cells as genetic target cells. Styrene-7,8-oxide which is mutagenic per se was rapidly metabolized by the perfused rat liver. Thus no mutagenic effect was detected neither in the perfusion medium nor in the bile. However when styrene was added to the perfusion system, an increase in V79 mutants was observed regardless of where in the circulating perfusion medium the V79 cells were placed: the same effect was obtained with V79 cells close to the liver as well as at a distance from the liver. No mutagenic effect was observed in the bile. Simultaneous analysis of the styrene-7,8-oxide concentration in the perfusion medium, suggest that this metabolite is not the cause of the mutagenic effect observed during perfusion with styrene.The effect of the two test compounds on some liver functions was also studied. Both styrene and styrene-7,8-oxide changed the bile flow without affecting bile acid secretion: styrene caused a reduction in bile flow as compared to control perfusions and styrene-7,8-oxide increased the bile flow. Styrene, but not styrene-7,8-oxide, reduced gluconeogenesis from lactate. Styrene had no effect on the liver's capacity to incorporate amino acids into plasma proteins, whereas styrene-7,8-oxide reduced the amino acid incorporation. The microsomal cytochrome P-450 content was not affected by the two test compounds. No alteration in microsomal N- and C-oxygenation of N, N-dimethylaniline (DMA) was observed with styrene-7,8-oxide or the lower styrene dose used (240 μmol), whereas the higher styrene concentration (480 μmol) reduced N-oxygenation and thus also the total DMA metabolism.It is suggested that the results on styrene and styrene-7,8-oxide found here using the liver perfusion/cell culture system mimic the metabolism expected to be found in the intact animal, thus indicating that styrene-7,8-oxide is not the principal mutagenic metabolite of styrene in vivo.     

Spectrum of styrene-induced DNA adducts: the relationship to other biomarkers and prospects in human biomonitoring

Styrene is an important industrial chemical that has shown genotoxicity in many toxicology assays. This is believed to be related to the DNA-binding properties of styrene-7,8-oxide (SO), a major metabolite of styrene. In this review, we have summarized knowledge on various aspects of styrene genotoxicity, especially in order to understand the formation and removal of primary DNA lesions, and the usefulness of biomarkers for risk assessment. Biological significances of specific DNA adducts and their role in the cascade of genotoxic events are discussed. Links between markers of external and internal exposure are evaluated, as well as metabolic aspects leading to the formation of DNA adducts and influencing biomarkers of biological effect. Finally, we suggest a design of a population study, which may contribute to our understanding genotoxic events in the exposure either to single xenobiotic or complex mixture.     

Kinetics of formation of specific styrene oxide adducts in double-stranded DNA.

The possible carcinogenicity of styrene is believed to be related to the DNA-binding properties of styrene 7,8-oxide (SO). In order to compare the intrinsic reactivity of the different nucleophilic sites in DNA towards SO and to evaluate the candidates for human biomonitoring we have determined the second-order rate constants and stabilities of several SO-adducts in double-stranded DNA. These include alpha- and beta-isomers of N7-substituted and alphaN(2)-substituted guanines, alpha- and betaN3-substituted and alphaN(6)-substituted adenines as well as betaN3- and alphaN(4)-substituted cytosines. The highest rate constants were found for the spontaneously depurinating N7-guanines being ca. 3-15-fold higher than those for the stable adducts. When the relative proportions of different alkylation products were determined in course of time, after a single addition of SO, the labile N7-guanines and N3-adenines were the major products at early time points. After 144 h of incubation at 37 degrees C, alphaN(6)-SO-adenine and alphaN(2)-SO-guanine as well as betaN3-SO-uracil were the major adducts. Regarding human biomonitoring, the N7-substituted guanines should be one of the main targets because of the high reactivity of the N7-atom of guanine. However, in the case of chronic styrene exposures the chemically more stable DNA adducts may become important.     

Site-selective modifications of arginine residues in human hemoglobin induced by methylglyoxal

Methylglyoxal (MG) is an important glycating agent produced under physiological conditions. MG could react with DNA and proteins to generate advanced glycation end products. Human hemoglobin, the most abundant protein in blood cells, has not been systematically investigated as the target protein for methylglyoxal modification. Here we examined carefully, by using HPLC coupled with tandem mass spectrometry (LC-MS/MS), the covalent modifications of human hemoglobin induced by methylglyoxal. Our results revealed that hemoglobin could be modified by methylglyoxal, and the major form of modification was found to be the hydroimidazolone derivative of arginine residues. In addition, Arg-92 and Arg-141 in the alpha chain as well as Arg-40 and Arg-104 in the beta chain were modified, whereas two other arginine residues, that is, Arg-31 in the alpha chain and Arg-30 in the beta chain, were not modified. Semiquantitative measurement for adduct formation, together with the analysis of the X-ray structure of hemoglobin, showed that the extents of arginine modification were highly correlated with the solvent accessibilities of these residues. The facile formation of hydroimidazolone derivatives of arginine residues in hemoglobin by methylglyoxal at physiologically relevant concentrations suggested that this type of modification might occur in vivo. The unambiguous determination of the sites and extents of methylglyoxal modifications of arginines in hemoglobin provided a basis for understanding the implications of these modifications and for employing this type of hemoglobin modification as molecular biomarkers for clinical applications.     

Characterization of Hydrophobic Peptides in the Presence of Detergent by Photoionization Mass Spectrometry

The characterization of membrane proteins is still challenging. The major issue is the high hydrophobicity of membrane proteins that necessitates the use of detergents for their extraction and solubilization. The very poor compatibility of mass spectrometry with detergents remains a tremendous obstacle in studies of membrane proteins. Here, we investigated the potential of atmospheric pressure photoionization (APPI) for mass spectrometry study of membrane proteins. This work was focused on the tetraspanin CD9 and the multidrug transporter BmrA. A set of peptides from CD9, exhibiting a broad range of hydropathicity, was investigated using APPI as compared to electrospray ionization (ESI). Mass spectrometry experiments revealed that the most hydrophobic peptides were hardly ionized by ESI whereas all peptides, including the highly hydrophobic one that corresponds to the full sequence of the first transmembrane domain of CD9, were easily ionized by APPI. The native protein BmrA purified in the presence of the non-ionic detergent beta-D-dodecyl maltoside (DDM) was digested in-solution using trypsin. The resulting peptides were investigated by flow injection analysis of the mixture followed by mass spectrometry. Upon ESI, only detergent ions were detected and the ionic signals from the peptides were totally suppressed. In contrast, APPI allowed many peptides distributed along the sequence of the protein to be detected. Furthermore, the parent ion corresponding to the first transmembrane domain of the protein BmrA was detected under APPI conditions. Careful examination of the APPI mass spectrum revealed a-, b-, c- and y- fragment ions generated by in-source fragmentation. Those fragment ions allowed unambiguous structural characterization of the transmembrane domain. In conclusion, APPI–MS appears as a versatile method allowing the ionization and fragmentation of hydrophobic peptides in the presence of detergent.     

Evaluation of the possible proteomic application of trypsin from Streptomyces griseus

Trypsin (EC 3.4.21.4) is the protease of choice for proteome analysis using mass spectrometry of peptides in sample digests. In this work, trypsin from Streptomyces griseus (SGT) was purified to homogeneity from pronase. The enzyme was evaluated in in-gel digestion of protein standards followed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) analyses of the digests. We recognized a remarkable cleavage performance of SGT. The number of produced and matching tryptic peptides was higher than in the case of commonly used bovine trypsin (BT) and allowed us to obtain higher identification scores in database searches. Interestingly, SGT was found to also generate nonspecific peptides whose sequencing by MALDI-TOF/TOF tandem mass spectrometry (MS/MS) revealed a partial F-X, Y-X, and W-X cleavage specificity. To suppress autolysis, either arginine or arginine plus lysine residues in SGT were modified by chemical reagents. In consequence, the autolytic pattern of SGT was reduced significantly, but specific activity dropped dramatically. As demonstrated by relative quantification of peptides at different times, SGT is more stable at 37 °C than is its bovine counterpart. We conclude that SGT represents a convenient alternative for proteomic applications involving protein digestion. Moreover, parallel digestions of sample aliquots by SGT and BT provide the possibility of combining partially different results (unique matching peptides) to improve protein identification.     

Identification of the carbohydrate moieties and glycosylation motifs in Campylobacter jejuni flagellin.

Flagellins from three strains of Campylobacter jejuni and one strain of Campylobacter coli were shown to be extensively modified by glycosyl residues, imparting an approximate 6000-Da shift from the molecular mass of the protein predicted from the DNA sequence. Tryptic peptides from C. jejuni 81-176 flagellin were subjected to capillary liquid chromatography-electrospray mass spectrometry with a high/low orifice stepping to identify peptide segments of aberrant masses together with their corresponding glycosyl appendages. These modified peptides were further characterized by tandem mass spectrometry and preparative high performance liquid chromatography followed by nano-NMR spectroscopy to identify the nature and precise site of glycosylation. These analyses have shown that there are 19 modified Ser/Thr residues in C. jejuni 81-176 flagellin. The predominant modification found on C. jejuni flagellin was O-linked 5,7-diacetamido-3,5,7,9-tetradeoxy-l-glycero-l-manno-nonulosonic acid (pseudaminic acid, Pse5Ac7Ac) with additional heterogeneity conferred by substitution of the acetamido groups with acetamidino and hydroxyproprionyl groups. In C. jejuni 81-176, the gene Cj1316c, encoding a protein of unknown function, was shown to be involved in the biosynthesis and/or the addition of the acetamidino group on Pse5Ac7Ac. Glycosylation is not random, since 19 of the total 107 Ser/Thr residues are modified, and all but one of these are restricted to the central, surface-exposed domain of flagellin when folded in the filament. The mechanism of attachment appears unrelated to a consensus peptide sequence but is rather based on surface accessibility of Ser/Thr residues in the folded protein.     

Comparison of peptides in the phloem sap of flowering and non-flowering Perilla and lupine plants using microbore HPLC followed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

Physiological evidence indicates that flower formation is hormonally controlled. The floral stimulus, or florigen, is formed in the leaves as a response to an inductive photoperiod and translocated through the phloem to the apical meristem. However, because of difficulties in obtaining and analyzing phloem sap and the lack of a bioassay, the chemical nature of this stimulus is one of the major unsolved problems in plant biology. A combination of microbore high-performance liquid chromatography (HPLC) and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) was used to compare the contents of the phloem sap from flowering and non-flowering plants. Instead of using one- or two-dimensional gel electrophoresis, microbore HPLC separations allowed us to detect proteins/peptides that were very small and present at very low levels. We detected more than 100 components in the phloem sap of Perilla ocymoides L. and Lupinus albusL. Sequences for 16 peptides in a mass range from 1 to 9 kDa were obtained. Two of these could be identified, 11 showed similarity to known or deduced protein sequences, and three showed no similarity to any known protein or translated gene sequence. Four of these peptides were specific to, modified, or increased in plants that were flowering, indicating their possible role in flower induction. The sequences of these peptides showed similarities to two purine permeases, a protein with similarity to protein kinases, and a protein with no similarities to any known protein.     

Determination of styrene and styrene-7,8-oxide in human blood by gas chromatography–mass spectrometry

Methods of isotope-dilution gas chromatography–mass spectrometry (GC–MS) are described for the determination of styrene and styrene-7,8-oxide (SO) in blood. Styrene and SO were directly measured in pentane extracts of blood from 35 reinforced plastics workers exposed to 4.7–97 ppm styrene. Using positive ion chemical ionization, styrene could be detected at levels greater than 2.5 μg/l blood and SO at levels greater than 0.05 μg/l blood. An alternative method for measurement of SO employed reaction with valine followed by derivatization with pentafluorophenyl isothiocyanate and analysis via negative ion chemical ionization GC–MS–MS (SO detection limit=0.025 μg/l blood). The detection limits for SO by these two methods were 10–20-fold lower than gas chromatographic assays reported earlier, based upon either electron impact MS or flame ionization detection. Excellent agreement between the two SO methods was observed for standard calibration curves while moderate to good agreement was observed among selected reinforced plastics workers (n=10). Levels of styrene in blood were found to be proportional to the corresponding air exposures to styrene, in line with other published relationships. Although levels of SO in blood, measured by the direct method, were significantly correlated with air levels of either styrene or SO among the reinforced plastics workers, blood concentrations were much lower than previously reported at a given exposure to styrene. The two assays for SO in blood appear to be unbiased and to have sufficient sensitivity and specificity for applications involving workers exposed to styrene and SO during the manufacture of reinforced plastics.     

Bone marrow cell chromosomal aberrations and styrene biotransformation in mice given styrene on a repeated oral schedule

Styrene's capacity to induce chromosomal aberrations was studied in bone marrow cells of CD1 male mice. No mutagenic effect could be detected after either a 4-day treatment course with daily oral doses of 500 mg/kg or a 70-day course with daily oral doses of 200 mg/kg. Urinary elimination of styrene metabolites related to styrene-7,8-oxide formation (i.e. phenylethylene glycol, mandelic acid, benzoic acid, phenylglyoxylic acid and total mercapturic acids) was quantitatively evaluated in the group of mice given the 200 mg/kg dose. In parallel, kinetic studies were made on styrene and styrene-7,8-oxide blood concentrations in the same group of animals. These determinations were carried out on days 1 and 70 of treatment by spectrophotometric, gas chromatographic and mass fragmentographic procedures.Not even nanograms of styrene-7,8-oxide were found in the blood of styrene-treated mice. This suggests that the metabolite does not migrate from the cellular compartment where it is formed being immediately metabolized or irreversibly bound to cellular structures.This observation could well explain the lack of mutagenic effects observed.     

Synthesis of peptides containing 2‐oxohistidine residues and their characterization by liquid chromatography‐tandem mass spectrometry

Protein oxidation by reactive oxygen species has been associated with aging and neurodegenerative disorders, and histidine is one of the major oxidation targets due to its metal‐chelating property and susceptibility to metal‐catalyzed oxidation. 2‐Oxohistidine, the major product of histidine oxidation, has been recently identified as a stable marker of oxidative damage in biological systems, but its biophysical and biochemical properties are understudied, partly because of difficulties in its chemical synthesis. We developed an efficient method to generate a 2‐oxohistidine side chain using metal‐catalyzed oxidation, applicable to both monomers and peptides. By optimizing reagent ratios and pH buffering in Cu²⁺/ascorbate/O₂ reaction system, we improved the yield more than tenfold compared to reported conditions, which allowed us to obtain homogeneously modified 2‐oxohisidine peptides for further studies. Analysis of 2‐oxohistidine‐containing model peptides by liquid chromatography‐tandem mass spectrometry demonstrated increased retention time in reverse‐phase chromatography and general stability of 2‐oxohistidine under electrospray ionization and collision‐induced dissociation. Thus, large‐scale analysis of 2‐oxohistidine‐modified proteome should be feasible using shotgun protein mass spectrometry, and we were able to observe such peptides in proteomics datasets. The feasibility of acquiring purified peptide probes and peptide antigens containing 2‐oxohistidine will help advance the study of this non‐enzymatic posttranslational modification. Copyright © 2014 European Peptide Society and John Wiley & Sons, Ltd.     

Identification of C8-methylguanine in the hydrolysates of DNA from rats administered 1,2-dimethylhydrazine. Evidence for in vivo DNA alkylation by methyl radicals.

C8-Methylguanine was identified in the neutral hydrolysates of DNA isolated from the liver or colon tissue of rats administered 1,2-dimethylhydrazine. In all the samples examined, the biologically isolated adducts were characterized by co-elution with synthetic C8-methylguanine under different high pressure liquid chromatography conditions. The sample isolated from liver DNA was also identified by UV spectroscopy at different pH values and by mass spectrometry. The estimated yields of C8-methylguanine obtained in hydrolysates of DNA from the liver or colon tissue were comparable to those of O6-methylguanine. C8-Methylguanine was not detected when the spin trap alpha-(4-pyridyl-1-oxide)-N-tert- butylnitrone was administered together with 1,2-dimethylhydrazine. The spin trap also inhibited N7-methylguanine and O6-methylguanine yields, although to a lesser extent. These results constitute the first evidence that DNA alkylation by carbon-centered radicals can occur in vivo.     

Some properties of vicinal diol-epoxides derived from benzo(a)anthracene and benzo(a)pyrene

The alkylating properties of pairs of syn- and anti-isomers of 2 diol-epoxides derived from benzo(a)pyrene (BP) and of 1 derived from benz(a)anthracene (BA) have been investigated. Of the anti-diol-epoxides, anti-BP 7,8-diol-9,10-oxide was the most reactive compound towards DNA, towards sodium p-nitrothiophenolate in a non-aqueous solvent system, and towards 4-(p-nitrobenzyl)pyridine in aqueous solution; anti-BP 9,10,-diol-7,8-oxide was of intermediate reactivity and anti-BA 8,9-diol-10,11-oxide was least reactive. The syn-diol-epoxides gave unsatisfactory results with DNA and 4-(p-nitrobenzyl)pyridine because of their rapid solvolysis in aqueous solution, but with sodium p-nitrothiophenolate showed the order of reactivity syn-BP 7,8-diol-9,10-oxide greater than syn-BA 8,9-diol-10,11-oxide greater than syn-BP 9,10-diol-7,8-oxide. The products of the reaction between diol-epoxides and nucleic acids were examined by Sephadex LH-20 chromatography followed by high-pressure liquid chromatography (HPLC) and the diol-epoxides were shown to react principally with the guanosine and adenosine moieties of RNA.     

Isolation and characterization of porcine somatotropin containing a succinimide residue in place of aspartate129.

Aspartate129 in porcine somatotropin was converted into a cyclic imide residue (succinimide) under acidic solution conditions. Reversed-phase high performance liquid chromatography was utilized to isolate and quantitate this altered species, which accounted for approximately 30% of the total protein. The molecular mass of this modified species was determined by electrospray mass spectrometry to be 18 Da less than normal porcine somatotropin, indicative of a loss of 1 H2O molecule. Tryptic peptide mapping demonstrated that the peptide composed of residues 126-133 was altered in this modified protein. Amino acid analysis, amino acid sequencing, mass spectrometry, and capillary zone electrophoresis were used to demonstrate that aspartate129 in this peptide had been converted into a succinimide residue. Further confirmation that this peptide contained a succinimide was obtained by hydrolyzing the modified peptide at pH 9.0, which yielded both the aspartate and isoaspartate peptides.