Electron-capture Negative-ion Chemical Ionization Research Articles

Temporal trends of lipophilic organic contaminants in blue mussel (1994–2017) and eelpout (1994–2017) from the southern Baltic Sea

A time-trend study was carried out for two important Baltic Sea species, blue mussel (1994–2017, 11 samples) and eelpout (1994–2017, 11 samples), to track the changes in levels of regulated persistent organic pollutants (POPs) and show potential increases in the levels of the contaminants of emerging concern (CECs). It was carried out utilizing gas chromatography–high-resolution mass spectrometry (GC-HRMS) based non-target screening (NTS). Data were acquired in two modes – electron ionization (EI) and electron capture negative ion chemical ionization (ECNI) – to widen the contaminant coverage, and treated using a fast semi-automated NTS data processing workflow. The study revealed that >250 tentatively identified compounds show statistically significant temporal trends in Baltic blue mussel and eelpout. A large number of regulated substances, including but not limited to PCBs, DDTs and other organochlorine pesticides (OCPs), chlorobenzenes, and many polybrominated diphenyl ethers (PBDEs), showed significant declining trends, as was expected. Their rates of decline were in good agreement with previously reported data. In contrast, increasing trends were observed for many CECs, some polycyclic aromatic compounds (PAHs), and hydrocarbons. The CEC group included, among others, four compounds, namely, one personal care product ingredient, 2-ethylhexyl stearate, one brominated compound 1,2,3,5-tetrabromobenzene and two intermediates 4-isopropoxyaniline and bilobol dimethyl ether, that were reported in marine biota for the first time to the best of our knowledge. Several compounds, including four CECs and two unknown brominated compounds, showed levels considerably higher than the common legacy pollutants (CB-153 and BDE-99), which might be taken into consideration for future monitoring and risk assessment. In addition, this work revealed the presence of a plethora of organoiodinated compounds that exhibited statistically significant temporal trends in the samples under study, which could be of future interest.

GC-MS Studies on the Conversion and Derivatization of γ-Glutamyl Peptides to Pyroglutamate (5-Oxo-Proline) Methyl Ester Pentafluoropropione Amide Derivatives.

Glutathione (γ-L-glutamyl-L-cysteinyl-glycine, γ-Glu-Cys-Gly) is the most abundant intra-cellular dicarboxylic tripeptide with multiple physiological roles. In biological samples, glutathione exists in its reduced form GSH and in two stable oxidized forms, i.e., in its symmetric disulfide form GSSG and as S-glutathionyl residue in proteins. S-Glutathionylation is a post-translational modification, which is involved in several pathophysiological processes, including oxidative stress. The GSH-to-GSSG molar ratio is widely used as a measure of oxidative stress. γ-Glutamyl is the most characteristic structural moiety of GSH. We performed gas chromatography-mass spectrometry (GC-MS) studies for the development of a highly specific qualitative and quantitative method for γ-glutamyl peptides. We discovered intra-molecular conversion of GSH, GSSG, γ-Glu-Cys and of ophthalmic acid (OPH; γ-glutamyl-α-amino-n-butyryl-glycine) to pyroglutamate (pGlu; 5-oxo-proline, also known as pidolic acid) during their derivatization with 2 M HCl/CH3OH (60 min, 80 °C). For GC-MS analysis, the methyl esters (Me) were further derivatized with pentafluoropropionic (PFP) anhydride in ethyl acetate (1:4, v/v; 30 min, 65 °C) to their PFP derivatives. At longer reaction times, pGlu is hydrolyzed to Glu. Internal standards were prepared by derivatizing GSH, GSSG, γ-Glu-Cys and OPH in 2 M HCl/CD3OD. Quantification of the Me-PFP derivative of pGlu was performed in the electron-capture negative-ion chemical ionization (ECNICI) mode by selected-ion monitoring (SIM) of the mass-to-charge (m/z) ions 269 for unlabeled pGlu (d0Me-PFP-pGlu) and m/z 272 for the in situ prepared deuterium-labeled pGlu (d3Me-PFP-pGlu). Although not inherent to the analysis of small peptides, the present GC-MS method is useful to study several biochemical aspects of GSH. Using pentafluorobenzyl bromide (PFB-Br) as the derivatization reagent, we found that synthetic pGlu is converted in aqueous acetone (60 min, 50 °C) into its pentafluorobenzyl (PFB) ester (PFB-pGlu). This derivatization procedure is useful for the GC-MS analysis of free pGlu in the ECNICI mode. Quantitative analysis of PFB-pGlu by GC-MS requires the use of stable-isotope labeled analogs of pGlu as an internal standard.

Comprehensive non-target screening of biomagnifying organic contaminants in the Baltic Sea food web

High-resolution mass spectrometry (HRMS) based non-target screening (NTS) is a powerful approach for the simultaneous determination of multiple environmental contaminant classes in complex biota samples. In this study, trophic biomagnification factor (TMF) directed NTS was performed to find and (tentatively) identify known, emerging, and new chemical contaminants that are persistent and biomagnify in Baltic Sea biota. The investigated food web included seven species: one filter feeder (blue mussel, Mytilus edulis), two fish (eelpout, Zoarces viviparous; herring, Clupea harengus), two marine mammals (harbor porpoise, Phocoena phocoena; grey seal, Halichoerus grypus) and two birds (guillemot, Uria aalge; white-tailed sea eagle, Haliaeetus albicilla). The NTS procedure included extraction with organic solvent mixtures, two-step high-resolution gel permeation chromatography clean-up, Florisil® fractionation, gas chromatography (GC) HRMS analysis in electron ionization (EI) and electron capture negative ion chemical ionization (ECNI) modes, and NTS data processing. The latter was performed differently for the EI and ECNI data: the EI data were treated using a flexible and highly automated TMF-directed NTS workflow, whereas the ECNI data were treated with a simpler and less automated workflow that specifically screened for brominated compounds. The two workflows collectively revealed biomagnification (statistically significant TMF values) of >250 tentatively identified compounds, including legacy persistent organic pollutants (POPs), such as PCBs and PCB-related compounds, DDT and its metabolites, and organochlorine pesticides (OCPs), contaminants of emerging concern (CECs), and halogenated natural products (HNPs). Among the tentatively identified CECs, nine have not previously been reported in environmental biota samples. These included four polymer additives (used as antioxidants, rubber additives or plasticizers) and two cosmetic product additives (ethyl myristate and isopropyl palmitate). The CECs should be prioritized for future structure verification and quantification using reference standards.

A time-trend guided non-target screening study of organic contaminants in Baltic Sea harbor porpoise (1988–2019), guillemot (1986–2019), and white-tailed sea eagle (1965–2017) using gas chromatography–high-resolution mass spectrometry

The rate of decline in regulated persistent organic pollutant (POP) concentrations in Baltic Sea biota has leveled off in recent years, with new contaminants frequently being discovered. There is, therefore, a need for comprehensive approaches to study occurrence and temporal trends of a wide range of environmental contaminants, including legacy POPs, contaminants of emerging concern (CECs), and new contaminants. In the current work, non-target screening (NTS) workflows were developed and used for, to the best of our knowledge, the first time-trend directed NTS of biota using gas chromatography–high-resolution mass spectrometry (GC-HRMS). To maximize contaminant coverage, both electron ionization (EI) and electron capture negative ion chemical ionization (ECNI) were used. The EI data were treated using highly automated workflows to find, prioritize, and tentatively identify contaminants with statistically significant temporal trends. The ECNI data were manually processed and reviewed prior to time-trend analysis. Altogether, more than 300 tentatively identified contaminants were found to have significant temporal trends in samples of Baltic guillemot, harbor porpoise, or white-tailed sea eagle. Significant decreases were found for many regulated chemicals, as could be expected, such as PCBs, polychlorinated terphenyls, chlorobenzenes, toxaphenes, DDT, other organochlorine pesticides, and tri- and tetra- bromodiphenyl ethers (BDEs). The rate of decline of legacy POPs agreed well with data reported from targeted analyses. Significant increases were observed for small polycyclic aromatic hydrocarbons, heptaBDEs, CECs, and terpenes and related compounds. The CECs included, among others, one plasticizer tributyl acetylcitrate (ATBC), two antioxidants 2,6-bis(1,1-dimethylethyl)phenol and 2,6-bis(tert-butyl)-4-(4-morpholinyl-methyl)phenol, and two compounds used in polymer production, trimethyl isocyanurate and 2-mercaptobenzothiazole, which had not previously been reported in biota. Their increased concentrations in biota indicate increased use and release. The increase in ATBC may be linked to increased use of it as a substitute for di-2-ethylhexyl phthalate (DEHP), which has been phased out over the last decade.

Non-targeted screening workflows for gas chromatography\u2013high-resolution mass spectrometry analysis and identification of biomagnifying contaminants in biota samples

The health of key species in the Baltic region has been impaired by exposure to anthropogenic hazardous substances (AHSs), which accumulate in organisms and are transferred through food chains. There is, thus, a need for comprehensive characterization of the occurrence and accumulation of AHSs in the ecosystem. In this study, we use a non-target screening (NTS) approach for this purpose. A major challenge in NTS of biological samples is the removal of matrix components such as lipids that may interfere with the detection and identification of compounds of interest. Here, we combine gel permeation chromatography with Florisil® column fractionation to achieve sufficient lipid removal for gas chromatography–high-resolution mass spectrometry analysis using electron ionization (EI) and electron capture negative ion chemical ionization (ECNI). In addition, we present new data processing workflows designed to systematically find and identify frequently occurring and biomagnifying AHSs, including known, emerging, and new contaminants. Using these workflows, we discovered a wide range of contaminants in tissue samples from blue mussels, fish, and marine mammals, and calculated their biomagnification factors (BMFs). Compounds with BMFs above 1 for herring and at least one marine mammal included legacy chlorinated pollutants (polychlorinated biphenyls, DDTs, chloro-benzenes/cyclohexanes, chlordanes, toxaphenes, dieldrin), polybrominated diphenyl ethers (PBDEs), and brominated biphenyls. However, there were also several halogenated natural products (halogenated methoxylated brominated diphenyl ethers, 1′-methyl-1,2′-bipyrroles, 1,1′-dimethyl-2,2′-bipyrroles, and the halogenated monoterpene mixed halogenated compound 1) as well as the novel flame retardant Dechlorane 602 and several polycyclic aromatic hydrocarbons, terpenoids, and steroids. The legacy pollutants exhibited the expected biomagnification behavior, demonstrating the utility of the unguided data processing workflow.Graphical abstract

Measurement of the tripeptides glutathione and ophthalmic acid by gas chromatography-mass spectrometry

Glutathione (GSH, γ-glutamyl-cysteinyl-glycine) is the most abundant intracellular dicarboxylic tripeptide with multiple physiological roles. GSH and its symmetric disulfide GSSG co-exist in biological samples. The GSH-to-GSSG molar ratio is high within cells and low in extracellular compartments. It is widely used as a biomarker of oxidative stress. Here, we report on a novel two-step derivatization procedure which allows for the specific quantitative measurement of GSH by gas chromatography-mass spectrometry (GC-MS) in the electron-capture negative-ion chemical ionization mode. The method is based on the derivatization of GSH first with pentafluoropropionic anhydride in ethyl acetate (30 min, 65 °C) and subsequently with 2 M HCl/CH3OH (30 min, 80 °C). Esterification in 2 M HCl/CD3OD is used for the in situ synthesis of deuterium-labelled GSH for use as internal standard. Quantification is performed by selected-ion monitoring of m/z 557 for GSH and m/z 560 for the internal standard (1 mM). Linear regression analysis between the peak area ratio of m/z 557 to m/z 560 (y) and the GSH concentration (x; range, 0–1 mM) resulted in the regression equations y = 0.02 + 0.62x (r2 = 0.976) in the absence and y = 0.02 + 0.63x (r2 = 0.997) in the presence of GSSG (range, 0–1 mM), underlying the linearity of the method for GSH and suggesting lack of interference by GSSG. γ-Glu-Cys and Cys-Gly also did not interfere. The GC-MS method demonstrated that in human hemolysate, nitrite at toxicological concentrations (range, 0–15 mM) depletes erythrocytic GSH. For the GC-MS analysis of the GSH tripeptide analogue ophthalmic acid, which lacks a Cys moiety, the derivatization order was HCl/CH3OH followed by pentafluoropropionic anhydride in ethyl acetate.

Development and validation of GC-MS methods for the comprehensive analysis of amino acids in plasma and urine and applications to the HELLP syndrome and pediatric kidney transplantation: evidence of altered methylation, transamidination, and arginase activity.

We developed and validated gas chromatography-mass spectrometry (GC-MS) methods for the simultaneous measurement of amino acids and their metabolites in 10-µL aliquots of human plasma and urine. De novo synthesized trideutero-methyl esters were used as internal standards. Plasma proteins were precipitated by acidified methanol and removed by centrifugation. Supernatants and native urine were evaporated to dryness. Amino acids were first esterified using 2M HCl in methanol and then amidated using pentafluoropropionic anhydride for electron-capture negative-ion chemical ionization. Time programmes were used for the gas chromatograph oven and the selected-ion monitoring of specific anions. The GC-MS methods were applied in clinical studies on the HELLP syndrome and pediatric kidney transplantation (KTx) focusing on L-arginine-related pathways. We found lower sarcosine (N-methylglycine) and higher asymmetric dimethylarginine (ADMA) plasma concentrations in HELLP syndrome women (n = 7) compared to healthy pregnant women (n = 5) indicating altered methylation. In plasma of pediatric KTx patients, lower guanidinoacetate and homoarginine concentrations were found in plasma but not in urine samples of patients treated with standard mycophenolate mofetil-based immunosuppression (MMF; n = 22) in comparison to matched patients treated with MMF-free immunosuppression (n = 22). On average, the global arginine bioavailability ratio was by about 40% lower in the MMF group compared to the EVR group (P = 0.004). Mycophenolate, the major pharmacologically active metabolite of MMF, is likely to inhibit the arginine:glycine amidinotransferase (AGAT), and to enhance arginase activity in leukocytes and other types of cell of MMF-treated children.

GC-MS quantification of urinary symmetric dimethylarginine (SDMA), a whole-body symmetric l-arginine methylation index

Circulating and excretory NG,N´G-dimethyl-l-arginine (symmetric dimethylarginine, SDMA) and NG,NG-dimethyl-l-arginine (asymmetric dimethylarginine, ADMA) are cardiovascular risk factors. Despite close chemical structures, the gas chromatography-mass spectrometry (GC-MS) measurement of SDMA is remarkably more difficult than that of ADMA for as yet unknown reasons. Here, we describe an improved GC-MS method for the quantitative determination of SDMA in human urine using commercially available NG,N´G-di-[2H3]methyl-l-arginine (d6-SDMA) as internal standard. The method is based on a single derivatization step with pentafluoropropionic anhydride (PFPA) in ethyl acetate (30 min, 65 °C) to N,N,N,O-tetrakis-pentafluoropropionyl derivatives, electron-capture negative-ion chemical ionization and selected-ion monitoring of the mass-to-charge (m/z) ions of m/z 456 for SDMA and m/z 462 for d6-SDMA.

GC-MS determination of nitrous anhydrase activity of bovine and human carbonic anhydrase II and IV

The most widely recognized activity of the large family of the metalloenzyme carbonic anhydrases (CAs) is the diffusion-controlled hydration of CO2 to HCO3− and one proton, and the less rapid dehydration of HCO3− to CO2: CO2 + H2O ⇆ HCO3− + H+. CAs also catalyze the reaction of water with other electrophiles such as aromatic esters, sulfates and phosphates, thus contributing to lending to CAs esterase, sulfatase and phosphatase activity, respectively. Renal CAII and CAIV are involved in the reabsorption of nitrite, the autoxidation product of the signalling molecule nitric oxide (NO): 4 NO + O2 + 2 H2O → 4 ONO− + 4 H+. Bovine and human CAII and CAIV have been reported to exert nitrite reductase and nitrous anhydride activity: 2 NO2− + 2 H+ ⇆ [2 HONO] ⇆ N2O3 + H2O. In the presence of L-cysteine, NO may be formed. In the literature, these issues are controversial, mainly due to analytical shortcomings, i.e., the inability to detect authentic HONO and N2O3. Here, we present a gas chromatography-mass spectrometry (GC-MS) assay to unambiguously detect and quantify the nitrous anhydrase activity of CAs. The assay is based on the hydrolysis of N2O3 in H218O to form ON18O− and 18ON18O−. After pentafluorobenzyl bromide derivatization and electron capture negative-ion chemical ionization of the pentafluorobenzyl nitro derivatives, quantification is performed by selected-ion monitoring of the anions with mass-to-charge (m/z) ratios of 46 (ONO−), m/z 48 (ON18O− and 18ONO−), m/z 50 (18ON18O−) and m/z 47 (O15NO−, internal standard).

GC-ECNICI-MS analysis of S-nitrosothiols and nitroprusside after treatment with aqueous sulphide (S2-) and derivatization with pentafluorobenzyl bromide: Evidence of S-transnitrosylation and formation of nitrite and nitrate.

A GC-MS method is reported for the quantitative analysis of S-nitrosothiols (RSNO) derived from endogenous low- and high-molecular mass thiols (RSH) including hemoglobin, cysteine, glutathione, N-acetylcysteine, and the exogenous N-acetylcysteine ethyl ester. The method is based on the conversion of RSNO to nitrite by aqueous Na2S (S2-). 15N-Labelled analogs (RS15NO) or 15N-labelled nitrite and nitrate were used as internal standards. The nitrite (14NO2- and 15NO2-) and nitrate (O14NO2- and O15NO2- anions were derivatised by pentafluorobenzyl (PFB) bromide (PFB-Br) in aqueous acetone and their PFB derivatives were separated by gas chromatography. After electron-capture negative-ion chemical ionization, the anions were separated by mass spectrometry and detected by selected-ion monitoring of m/z 46 for 14NO2-, m/z 47 for 15NO2-, m/z 62 for O14NO2-, and m/z 63 for O15NO2-. The expected thionitrites (-S14NO and -S15NO) were not detected, suggesting that they are intermediates and rapidly exchange their S by O from water, presumably prior to PFB-Br derivatization. The reaction of S2- with RSNO and sodium nitroprusside (SNP) resulted in the formation of nitrite and nitrate as the major and minor reaction products, respectively. The novel Na2S procedure was compared with established procedures based on the use of aqueous HgCl2 or cysteine/Cu2+ reagents to convert the S-nitroso group to nitrite. Our results provide evidence for an equilibrium S-transnitrosylation reaction between S2- with RSNO in buffered solutions of neutral pH. Use of Na2S in molar excess over RSNO shifts this reaction to the right, thus allowing almost complete conversion of RSNO to nitrite and nitrate. The Na2S procedure should be useful for the quantitative determination of RSNO as nitrite and nitrate after PFB-Br derivatization and GC-MS analysis. The Na2S procedure may also contribute to explore the complex reactions of S2- with RSNO, SNP and other NO-containing compounds.

Pentafluorobenzyl bromide-A versatile derivatization agent in chromatography and mass spectrometry: I. Analysis of inorganic anions and organophosphates.

Pentafluorobenzyl bromide (PFB-Br) is a versatile derivatization agent. It is widely used in chromatography and mass spectrometry since several decades. The bromide atom is largely the single leaving group of PFB-Br. It is substituted by wide a spectrum of nucleophiles in aqueous and non-aqueous systems to form electrically neutral, in most organic solvents soluble, generally thermally stable, volatile, strongly electron-capturing and ultraviolet light-absorbing derivatives. Because of these greatly favoured physicochemical properties, PFB-Br emerged an ideal derivatization agent for highly sensitive analysis of endogenous and exogenous substances including various inorganic and organic anions by electron capture detection or after electron-capture negative-ion chemical ionization in GC-MS. The present article attempts an appraisal of the utility of PFB-Br in analytical chemistry. It reviews and discusses papers dealing with the use of PFB-Br as the derivatization reagent in the qualitative and quantitative analysis of endogenous and exogenous inorganic anions in various biological samples, notably plasma, urine and saliva. These analytes include nitrite, nitrate, cyanide and dialkyl organophosphates. Special emphasis is given to mass spectrometry-based approaches and stable-isotope dilution techniques.

Simultaneous GC-MS/MS measurement of malondialdehyde and 4-hydroxy-2-nonenal in human plasma: Effects of long-term L-arginine administration

Here, we report the simultaneous derivatization and quantification of malondialdehyde (MDA) and 4-hydroxy-2-nonenal (HNE) in human plasma by GC-MS/MS using [1,3-2H2]-MDA (d2-MDA) and [9,9,9-2H3]-HNE (d3-HNE) as the internal standards, respectively. MDA, d2-MDA, HNE and d3-HNE were converted to their pentafluorobenzyl oximes (PFBOX) by pentafluorobenzyl hydroxylamine. Subsequently, the hydroxyl groups of the PFBOX of HNE and d3-HNE were trimethylsilylated with N,O-bis(trimethylsilyl)trifluoroacetamide/1% trimethylchlorosilane. GC-MS/MS analyses were performed in the electron-capture negative-ion chemical ionization mode. Quantification was performed by selected-reaction monitoring the mass transitions m/z 442 to m/z 243 for MDA, m/z 444 to m/z 244 for d2-MDA, m/z 403 → m/z 283 for HNE and m/z 406 → m/z 286 for d3-HNE. The method was applied to measure MDA and HNE in plasma of patients suffering from coronary artery disease (CAD) or peripheral artery occlusive disease (PAOD) before and after oral supplementation of L-arginine (3 g/day) or placebo for 3 (CAD and PAOD) and 6 months (PAOD). All plasma samples were analyzed after completion of the studies. Our results revealed that storage of plasma samples (at −80 °C) leads to lower MDA and HNE plasma concentrations in the plasma samples that were collected at the end of the studies as compared to those collected at the begin of the studies. Based on MDA and HNE measurements in plasma, L-arginine did not influence lipid peroxidation in CAD and PAOD patients. Long-term studies on lipid peroxidation are best performed by measuring oxidative stress biomarkers such as MDA and/or HNE in plasma samples immediately after their collection. Long-term storage of plasma samples even at −80 °C is not recommended.

Cross-validated stable-isotope dilution GC-MS and LC-MS/MS assays for monoacylglycerol lipase (MAGL) activity by measuring arachidonic acid released from the endocannabinoid 2-arachidonoyl glycerol.

2-Arachidonoyl glycerol (2AG) is an endocannabinoid that activates cannabinoid (CB) receptors CB1 and CB2. Monoacylglycerol lipase (MAGL) inactivates 2AG through hydrolysis to arachidonic acid (AA) and glycerol, thus modulating the activity at CB receptors. In the brain, AA released from 2AG by the action of MAGL serves as a substrate for cyclooxygenases which produce pro-inflammatory prostaglandins. Here we report stable-isotope GC-MS and LC-MS/MS assays for the reliable measurement of MAGL activity. The assays utilize deuterium-labeled 2AG (d8-2AG; 10μM) as the MAGL substrate and measure deuterium-labeled AA (d8-AA; range 0-1μM) as the MAGL product. Unlabelled AA (d0-AA, 1μM) serves as the internal standard. d8-AA and d0-AA are extracted from the aqueous buffered incubation mixtures by ethyl acetate. Upon solvent evaporation the residue is reconstituted in the mobile phase prior to LC-MS/MS analysis or in anhydrous acetonitrile for GC-MS analysis. LC-MS/MS analysis is performed in the negative electrospray ionization mode by selected-reaction monitoring the mass transitions [M-H]-→[M-H - CO2]-, i.e., m/z 311→m/z 267 for d8-AA and m/z 303→m/z 259 for d0-AA. Prior to GC-MS analysis d8-AA and d0-AA were converted to their pentafluorobenzyl (PFB) esters by means of PFB-Br. GC-MS analysis is performed in the electron-capture negative-ion chemical ionization mode by selected-ion monitoring the ions [M-PFB]-, i.e., m/z 311 for d8-AA and m/z 303 for d0-AA. The GC-MS and LC-MS/MS assays were cross-validated. Linear regression analysis between the concentration (range, 0-1μM) of d8-AA measured by LC-MS/MS (y) and that by GC-MS (x) revealed a straight line (r2=0.9848) with the regression equation y=0.003+0.898x, indicating a good agreement. In dog liver, we detected MAGL activity that was inhibitable by the MAGL inhibitor JZL-184. Exogenous eicosatetraynoic acid is suitable as internal standard for the quantitative determination of d8-AA produced from d8-2AG by hepatic MAGL activity. The formation of d8-prostaglandin E2 by the consecutive catalytic action of recombinant MAGL on d8-2AG and recombinant cyclooxygenase-2 (COX) on d8-AA was demonstrated by GC-MS/MS.

Simultaneous pentafluorobenzyl derivatization and GC-ECNICI-MS measurement of nitrite and malondialdehyde in human urine: Close positive correlation between these disparate oxidative stress biomarkers.

Urinary nitrite and malondialdehyde (MDA) are biomarkers of nitrosative and oxidative stress, respectively. At physiological pH values of urine and plasma, nitrite and MDA exist almost entirely in their dissociated forms, i.e., as ONO- (ONOH, pKa=3.4) and -CH(CHO)2 (CH2(CHO)2, pKa=4.5). Previously, we reported that nitrite and MDA react with pentafluorobenzyl (PFB) bromide (PFB-Br) in aqueous acetone. Here, we report on the simultaneous derivatization of nitrite and MDA and their stable-isotope labeled analogs O15NO- (4μM) and CH2(CDO)2 (1μM or 10μM) with PFB-Br (10μL) to PFBNO2, PFB15NO2, C(PFB)2(CHO)2), C(PFB)2(CDO)2 by heating acetonic urine (urine-acetone, 100:400μL) for 60min at 50°C. After acetone evaporation under a stream of nitrogen, derivatives were extracted with ethyl acetate (1mL). A 1-μL aliquot of the ethyl acetate phase dried over anhydrous Na2SO4 was injected in the splitless mode for simultaneous GC-MS analysis in the electron capture negative-ion chemical ionization mode. Quantification was performed by selected-ion monitoring (SIM) the anions [M-PFB]-m/z 46 for ONO-, m/z 47 for O15NO-, m/z 251 for -C(PFB)(CHO)2, and m/z 253 for -C(PFB)(CDO)2. The retention times were 3.18min for PFB-ONO2/PFB-O15NO2, and 7.13min for -C(PFB)(CHO)2/-C(PFB)(CDO)2. Use of CH2(CDO)2 at 1μM but not at 10μM was associated with an unknown interference with the C(PFB)2(CDO)2 peak. Endogenous MDA can be quantified using O15NO- (4μM) and CH2(CDO)2 (10μM) as the internal standards. The method is also useful for the measurement of nitrate and creatinine in addition to nitrite and MDA. Nitrite and MDA were measured by this method in urine of elderly healthy subjects (10 females, 9 males; age, 60-70 years; BMI, 25-30kg/m2). Creatinine-corrected excretion rates did not differ between males and females for MDA (62.6 [24-137] vs 80.2 [52-118]nmol/mmol, P=0.448) and for nitrite (102 [71-174] vs. 278 [110-721]nmol/mmol P=0.053). We report for the first time a close correlation (r=0.819, P<0.0001) between MDA and nitrite in human urine. This correlation is assumed to be due to involvement of myeloperoxidase which catalyzes the formation of hypochlorite (-OCl) from chloride and hydrogen peroxide. In turn, hypochlorite reacts both with nitrite and with polyunsaturated fatty acids such as arachidonic acid, with the later reaction generating MDA. The proposed mechanisms are supported by the literature but remain to be fully explored.

GC-ECNICI-MS/MS of eicosanoids as pentafluorobenzyl-trimethylsilyl (TMS) derivatives: Evidence of CAD-induced intramolecular TMS ether-to-ester rearrangement using carboxy-18O-labelled eicosanoids and possible implications in quantitative analysis.

GC-MS and GC-MS/MS of pentafluorobenzyl (PFB) ester trimethylsilyl (TMS) ether (PFB-TMS) derivatives of hydroxylated long-chain fatty acids including arachidonic acid metabolites, the eicosanoids, in the electron-capture negative-ion chemical ionization (ECNICI) mode are the most sensitive and accurate approaches to quantify carboxyl groups-containing compounds in complex biological fluids such as plasma and urine. Under ECNICI conditions, PFB-TMS derivatives of eicosanoids ionize to form very few ions, with the carboxylates [M-PFB]- being typically the most intense. Less intense ions may be additionally formed by consecutive neutral loss (NL) of trimethylsilanol (TMSOH, 90Da) groups ([M-PFB-(TMSOH)n]-). By using [1,1-18O2]- and [1,ω-18O2]-eicosanoids, we studied ion processes following collisionally activated dissociation (CAD) of the precursor ions [M-PFB]-. We found that CAD resulted in formation of product ions due to NL of a TMS18OH (92Da) group in monocarboxylic and of a PFB18OH (200Da) group in dicarboxylic eicosanoids. TMS18OH NL implies an intra-molecular transfer of the TMS group from hydroxyl groups to their carboxylate anions [M-PFB]-. From a mechanistic point of view, this rearrangement may explain formation of unique product ions in GC-MS/MS of eicosanoids under ECNICI conditions. From the quantitative point of view, quantification by GC-MS/MS of product ions due to [M-PFB-(TMSOH)n]- and [M-PFB-TMS18OH-(TMSOH)n-1]-would reveal incorrect data, if [1,1-18O2]-eicosanoids are used as internal standards and if no correction for the 18O-loss is performed. In 18O-labelled dicarboxylic eicosanoids, such as the major urinary metabolite (MUM) of E prostaglandins, i.e., [1,ω-18O2]-PGE-MUM), no TMS ester/TMS ether rearrangement was observed. Yet, 18O-loss occurred upon CAD of [M-PFB]- due to NL of PFB18OH (200Da). In both cases the extent of 18O-loss needs to be determined and considered for accurate quantification of monocarboxylic acids such as 8-isoprostaglandin F2α (8-iso-PGF2α) and dicarboxylic eicosanoids such as PGE-MUM.

GC-MS and GC-MS/MS measurement of ibuprofen in 10-μL aliquots of human plasma and mice serum using [α-methylo-2H3]ibuprofen after ethyl acetate extraction and pentafluorobenzyl bromide derivatization: Discovery of a collision energy-dependent H/D isotope effect and pharmacokinetic application to inhaled ibuprofen-arginine in mice.

GC-MS and GC-MS/MS methods were developed and validated for the quantitative determination of ibuprofen (d0-ibuprofen), a non-steroidal anti-inflammatory drug (NSAID), in human plasma using α-methyl-2H3-4-(isobutyl)phenylacetic acid (d3-ibuprofen) as internal standard. Plasma (10μL) was diluted with acetate buffer (80μL, 1M, pH 4.9) and d0- and d3-ibuprofen were extracted with ethyl acetate (2×500μL). After solvent evaporation d0- and d3-ibuprofen were derivatized in anhydrous acetonitrile by using pentafluorobenzyl (PFB) bromide and N,N-diisopropylethylamine as the base catalyst. Under electron-capture negative-ion chemical ionization (ECNICI), the PFB esters of d0- and d3-ibuprofen readily ionize to form their carboxylate anions [M-PFB]- at m/z 205 and m/z 208, respectively. Collision-induced dissociation (CID) of m/z 205 and m/z 208 resulted in the formation of the anions at m/z 161 and m/z 164, respectively, due to neutral loss of CO2 (44 Da). A collision energy-dependent H/D isotope effect was observed, which involves abstraction/elimination of H- from d0-ibuprofen and D- from d3-ibuprofen and is minimum at a CE value of 5eV. Quantitative GC-MS determination was performed by selected-ion monitoring of m/z 205 and m/z 208. Quantitative GC-MS/MS determination was performed by selected-reaction monitoring of the mass transitions m/z 205 to m/z 161 for d0-ibuprofen and m/z 208 to m/z 164 for d3-ibuprofen. In a therapeutically relevant concentration range (0-1000μM) d0-ibuprofen added to human plasma was determined with accuracy (recovery, %) and imprecision (relative standard deviation, %) ranging between 93.7 and 110%, and between 0.8 and 4.9%, respectively. GC-MS (y) and GC-MS/MS (x) yielded almost identical results (y=4.00+0.988x, r2=0.9991). In incubation mixtures of arachidonic acid (10μM), d3-ibuprofen (10μM) or d0-ibuprofen (10μM) with ovine cyclooxygenase (COX) isoforms 1 and 2, the concentration of d3-ibuprofen and d0-ibuprofen did not change upon incubation at 37°C up to 60min. The trough pharmacokinetics of an inhaled arginine-containing ibuprofen preparation in mice was studied after once-daily treatment (0.0, 0.07, 0.4 and 2.5mg/kg body weight) for three days. A linear relationship between ibuprofen concentration in serum (10μL) and administered dose 24h after the last drug administration was observed.

Simultaneous GC-ECNICI-MS measurement of nitrite, nitrate and creatinine in human urine and plasma in clinical settings.

Creatinine in urine is a useful biochemical parameter to correct the urinary excretion rate of endogenous and exogenous substances. Nitrite (ONO-) and nitrate (ONO2-) are metabolites of nitric oxide (NO), a signalling molecule with multiple biological functions. Under certain and standardized conditions, the concentration of nitrate in the urine is a suitable measure of whole body NO synthesis. The urinary nitrate-to-nitrite molar ratio (UNOxR) may indicate nitrite-dependent renal carbonic anhydrase (CA) activity. In clinical studies, urine is commonly collected by spontaneous micturition. In those cases the nitrate and nitrite excretion must be corrected for creatinine excretion. Pentafluorobenzyl (PFB) bromide (PFB-Br) is a useful derivatization reagent of numerous inorganic and organic compounds, including urinary nitrite, nitrate and creatinine, for highly sensitive and specific quantitation by GC-MS. Here, we report on the simultaneous PFB-Br derivatization (60min, 50°C) of ONO-, O15NO-, ONO2-, O15NO2-, creatinine (do-Crea) and [methylo-2H3]creatinine (d3-Crea) in acetonic dilutions of native human urine and plasma samples (4:1, v/v) and their simultaneous quantification by GC-MS as PFBNO2, PFB15NO2, PFBONO2, PFBO15NO2, do-Crea-PFB and d3-Crea-PFB, respectively. Electron capture negative-ion chemical ionization (ECNICI) of these derivatives generates anions due to [M-PFB]-, i.e., the starting analytes. Quantification is performed by selected-ion monitoring (SIM) of m/z 46 (ONO-), m/z 47 (O15NO-), m/z 62 (ONO2-), m/z 63 (O15NO2-), m/z 112 (do-Crea), and m/z 115 (d3-Crea). Retention times were 2.97min for PFB-ONO2/PFB-O15NO2, 3.1min for PFB-NO2/PFB-15NO2, and 6.7min for do-Crea-PFB/d3-Crea-PFB. We used this method to investigate the effects of long-term oral NaNO3 or NaCl (serving as placebo) supplementation (each 0.1mmol/kg body weight per day for 3 weeks) on creatinine excretion and UNOxR in 17 healthy young men. Compared to NaCl (n=8), NaNO3 (n=9) supplementation increased UNOxR (1709±355 vs. 369±77, P<0.05). Creatinine excretion did not differ between the groups (6.67±1.34mM vs. 5.72±1.27mM, P=0.57). The method is also applicable to human plasma. In 78 adults patients newly diagnosed for cerebrovascular disease (CVD), there was a close correlation (r=0.9833) between the creatinine concentrations measured in plasma by GC-ECNICI-MS and those measured in serum by an enzymatic assay. Creatinine-corrected plasma nitrate and nitrite concentrations (P=0.035 and P=0.004, respectively) but not their concentrations (P=0.68 and P=0.40, respectively) differ between male (n=54) and female (n=24) CVD patients. No such differences were found between preterm newborn boys (n=25) and girls (n=22). Like in urine, circulating creatinine may be useful to correct for gender-specific differences in plasma nitrite and nitrate in adults. Chronic NaNO3 supplementation to healthy young men does not affect renal CA-dependent nitrite excretion or creatinine synthesis and excretion.

Gas chromatographic-mass spectrometric analysis of the tripeptide glutathione in the electron-capture negative-ion chemical ionization mode.

The dicarboxylic tripeptide glutathione (GSH) is the most abundant intracellular thiol. GSH analysis by liquid chromatography is routine. Yet, GSH analysis by gas chromatography is challenged due to thermal instability and lacking volatility. We report a high-yield laboratory method for the preparation of (2)H-labeled GSH dimethyl ester ((d3Me)2-GSH) for use as internal standard (IS) which was characterized by LC-MS/MS. For GC-MS analysis, the dimethyl esters of GSH and the IS were derivatized with pentafluoropropionic (PFP) anhydride. Electron-capture negative-ion chemical ionization of the (Me)2-(PFP)3-GSH provided high sensitivity. We encourage increasing use of GC-MS in the analysis of amino acids as their Me-PFP derivatives in the ECNICI mode.

Utilizing immunoaffinity chromatography (IAC) cross-reactivity in GC–MS/MS exemplified at the measurement of prostaglandin E1 in human plasma using prostaglandin E2-specific IAC columns

Immunoaffinity chromatography (IAC) is an elegant and highly efficient method to isolate a particular compound from biological samples for measurement by mass spectrometry coupled to GC, CE, or LC. The utility of IAC for the quantitative determination of several prostaglandins including prostaglandin E2 (PGE2) by GC-MS/MS and LC-MS/MS has been demonstrated. The aim of the present work was to test whether the cross-reactivity of the antibody immobilized on an insoluble support can be utilized for the quantitative determination of biomolecules by stable-isotope dilution mass spectrometry. In this communication, we provide evidence that this is indeed possible for prostaglandin E1 (PGE1) in human plasma by GC-MS/MS using commercially available Sepharose 4-based IAC columns with immobilized mouse anti-PGE2 monoclonal antibody with a declared cross-reactivity of about 19% toward PGE1. Endogenous PGE1 and the internal standard [3,3',4,4'-(2)H4]-PGE1 (d4-PGE1) externally added to human plasma samples were extracted by IAC, converted to their pentafluorobenzyl ester-methoxime-trimethylsilyl ether derivatives and analyzed by GC-MS/MS in the electron-capture negative-ion chemical ionization mode. Quantification was performed by selected-reaction monitoring of the mass transition m/z 526→m/z 258 for PGE1 and m/z 530→m/z 262 for d4-PGE1. By this method we measured PGE1 concentrations in EDTA plasma samples (1mL) of six healthy volunteers in the range 10-25pg/mL (29-72pM). PGE1 plasma concentration showed a trend for positive correlation with plasma parameters such as low density lipoprotein (LDL)-cholesterol, total cholesterol and glucose. The method described here provides a novel tool to study the potential link of PGE1 formation to dyslipidemia, insulin resistance and related metabolic disorders.

GC-MS and GC-MS/MS measurement of the cardiovascular risk factor homoarginine in biological samples.

L-Homoarginine (hArg) has recently emerged as a novel cardiovascular risk factor and to herald a poor prognosis in heart failure patients. Here, we report on the development and thorough validation of gas chromatography-mass spectrometry (GC-MS) and gas chromatography-tandem mass spectrometry (GC-MS/MS) methods for the quantitative determination of hArg in biological samples, including human plasma, urine and sputum. For plasma and serum samples, ultrafiltrate (10µL; cutoff, 10kDa) was used. For urine samples, native urine (10µL) was used. For sputum, protein precipitation by acetone was performed. hArg is derivatized to its methyl ester tri(N-pentafluoropropionyl) derivative; de novo synthesized trideutero-methyl ester hArg is used as the internal standard (IS). Alternatively, [guanidino-(15)N2]-arginine can be used as an IS. Quantitative analyses were performed after electron-capture negative-ion chemical ionization by selected-ion monitoring in GC-MS and selected-reaction monitoring in GC-MS/MS. We obtained very similar hArg concentrations by GC-MS and GC-MS/MS, suggesting that GC-MS suffices for accurate and precise quantification of hArg in biological samples. In plasma and serum samples of the same subjects very close hArg concentrations were measured. The plasma-to-serum hArg concentration ratio was determined to be 1.12±0.21 (RSD, 19%), suggesting that blood anticoagulation is not a major preanalytical concern in hArg analysis. In healthy subjects, the creatinine-corrected urinary excretion of hArg varies considerably (0.18±0.22µmol/mmol, mean±SD, n=19) unlike asymmetric dimethylarginine (ADMA, 2.89±0.89µmol/mmol). In urine, hArg correlated with ADMA (r=0.475, P=0.040); in average, subjects excreted in the urine about 17.5 times more ADMA than hArg. In plasma of healthy humans, the concentration of hArg is of the order of 2µM. hArg may be a low-abundance constituent of human plasma proteins. The GC-MS and GC-MS/MS methods we report in this article are useful to study the physiology and pathology of hArg in experimental and clinical settings.