GSSG by UHPLC-ESI-MSMS for human plasma

Accepted Manuscript A standardized protocol for comparable analysis of GSH/GSSG by UHPLC-ESI-MSMS for human plasma

Anna-Sara Claeson, Sandra Gouveia-Figueira, Hans Stenlund, Annika I. Johansson PII: DOI: Reference:

S1570-0232(18)31199-1 https://doi.org/10.1016/j.jchromb.2018.11.007 CHROMB 21428

To appear in:

Journal of Chromatography B

Received date: Revised date: Accepted date:

2 August 2018 15 October 2018 6 November 2018

Please cite this article as: Anna-Sara Claeson, Sandra Gouveia-Figueira, Hans Stenlund, Annika I. Johansson , A standardized protocol for comparable analysis of GSH/GSSG by UHPLC-ESI-MSMS for human plasma. Chromb (2018), https://doi.org/10.1016/ j.jchromb.2018.11.007

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT A standardized protocol for comparable analysis of GSH/GSSG by UHPLC-ESI-MSMS for human plasma

Anna-Sara Claeson1*, Gouveia-Figueira Sandra2, Stenlund Hans2, Johansson Annika I.2

Department of psychology, Umeå University, Sweden

2

Swedish Metabolomics Centre (SMC), Umeå, Sweden

RI

PT

1

SC

*Corresponding author: Corresponding author information:

NU

E-mail: [email protected] Full postal address: Umeå university, 901 87 Umeå

AC

CE

PT E

D

MA

Keywords: Glutathione; GSH; GSSG; plasma; sample pre-processing; UHPLC-ESI-MS

ACCEPTED MANUSCRIPT Abstract Variability in the levels of GSH and GSSG in plasma are suggested to derive from inadequate pre-processing methods. The aim of this study was to develop a protocol for comparable and reliable measurements of GSH/GSSG. Venous blood from 8 healthy individuals were collected and divided into 7 different pre-processing procedures. For three of the samples an

PT

extraction mixture was added after 0 (baseline), 4 and 8 minutes and for three of the samples

RI

the extraction mixture was added at different times during defrost. A worst case scenario

SC

where a sample was left in a cool box during 6 hours was also included. The samples were analysed with UHPLC-ESI-MSMS. A large difference in the levels of GSH and GSSG were

NU

identified and it was clearly associated with the sample handling procedures. A sample left untreated for 4 minutes will have significantly reduced amount of GSH. Stability tests showed

AC

CE

PT E

D

MA

that the level of GSH was reduced after 3 months in -80°C.

ACCEPTED MANUSCRIPT 1. Introduction Glutathione (GSH) is a tripeptide present throughout the human body that plays a central role in the defence against oxidative stress. Reactive oxygen/nitrogen species (ROS/RONS) formed as a consequence of acute or chronic inflammation or because of certain environmental exposures (e.g. O3 or NO2) lead to oxidation of GSH into glutathione disulfide

PT

(GSSG) and depending on the conditions, other oxidative products might also be formed (e.g.

RI

GSH-mixed disulfides) [1–3]. The reaction is crucial for maintaining redox balance and whole

SC

body homeostasis [4–6]. Oxidative stress has been shown to play an important role in many clinical conditions such as cancer, neurodegenerative disorders, cystic fibrosis (CF), HIV and

NU

ageing [7]. Measurement of GSH and GSSG levels together with the ratio between the two species are useful indicators of the redox status of the tissue. Depletion of GSH have been

MA

related to certain ear, nose and throat conditions and is also suggested to be of importance in the development of certain environmental intolerances [8–11]. Further, mental stress in daily

D

life and environmental exposures have been shown to decrease the level of GSH [3,12–14].

PT E

Despite numerous studies supporting the link between oxidative stress and a variety of human

CE

diseases, a solid connection to GSH deficiency is missing [2].

AC

The variability in the analysis of GSH from plasma has been pointed out as a significant source for the inconclusive results (1-20 µM; [2,6,15,16]). Hence, large and unforeseen analytical variations will complicate associating the oxidative stress effect to a specific disease or intervention, both within as well as between studies [2,17]. Differences in concentrations have been suggested to derive from inadequate pre-processing methods, analysing GSH in plasma [18]. The half-life of plasma-GSH in room temperature was estimated to be in the range between 5 to 20 minutes, meaning that the additional centrifugation step required for plasma preparation inevitably infer obvious oxidation,

ACCEPTED MANUSCRIPT reducing the levels of GSH [19,20]. The auto-oxidation of GSH is expected to occur during several steps of the analytical chain, as sampling, preservation, storage, preparation, extraction, and through the analysis on the instruments, affecting the ratio between the levels of GSH and GSSG [21]. Different strategies, such as derivatization [22–24] and acidification [25,26] have been used to prevent the conversion of GSH, however, this has introduced new

PT

problems [27]. Also, the absence of a derivatization step speeds up the analysis and makes it

RI

easier to perform, which is important for the measurement of the GSH/GSSG ratio. A method

SC

for the simultaneous measurement of GSH and GSSG in plant tissues by HPLC-ESI-MS, using meta-phosphoric acid (MPA) to stop the auto-oxidation of GSH, was developed and

NU

validated by Rellán-Álvarez et al., 2006. In order to monitor changes in redox status during the analysis, a stable isotope of GSH was added during samples extraction [28]. The

MA

introduction of LC-MS-MS as the analytical tool for the analysis of GSH and GSSG have improved selectivity, precision and accuracy of the analysis allowing quantification of low

D

concentrations of both analytes [26,29,30]. Nevertheless, the uncertainty during the pre-

PT E

processing still exist and there is an immediate need of a general procedure agreement in the aspect of standardized GSH/GSSG-measurements [4,31]. Hence, in order to increase the

CE

knowledge of the involvement of the oxidative stress related to studied phenomena the

studies.

AC

measurements of GSH/GSSG has to be comparable, both within studies as well as between

The aim of this study was to investigate how sample preparation procedures affect the level of GSH/GSSG in human plasma, developing a protocol for comparable measurement of GSH/GSSG by UHPLC-ESI-MS in plasma. The aim was also to evaluate the feasibility of the sampling procedure to be used in field studies, including ‘worst case scenario’ in order to

ACCEPTED MANUSCRIPT investigate the relevance of the GSH/GSSG analysis with no or little control of the preanalytical procedure.

2. Materials and methods

PT

2.1 Chemicals

RI

L-Glutathione reduced (GSH), L-Glutathione oxidized (GSSG) and meta-Phosphoric acid

SC

(MPA), formic acid (LC-MS grade) and EDTA were purchased form Sigma (Sigma-Aldrich Co., St Louis, MO, USA). Glutathione (glycine-13C2,15N) were obtained from Cambridge

NU

Isotope Laboratories (Cambridge Isotope Laboratories, Inc. , Tewksbury, MA, USA) . Acetonitrile (LC-MS grade) was acquired from Merck (Darmstadt, Germany) and water was

D

MA

purified by a Milli-Q Gradient system (Millipore, Milford, MA, USA).

PT E

Solutions of GSH, GSSG and IS-GSH were prepared in 2.5% (v/v) MPA, 1 mM EDTA, 0.1% formic acid, stored in -80°C as suggested by Rellán-Álvarez et al., 2006. A stock solution for

CE

each metabolite was prepared at 10 mM and twelve different calibration levels of GSH and GSSG (0.01-10 µM) containing 0.25 µM GSH-IS were prepared. An extraction mixture

AC

consisting of 5% MPA, 1 mM EDTA and 0.1% formic acid containing 0.25µM IS-GSH was prepared and used for blood collection within the next three days.

2.2 Preparation of plasma samples Venous blood was collected from eight healthy individuals (A-G) by a trained nurse in vacutainer tubes containing NA2 EDTA as an anticoagulant (Greiner Bio-One, Kremsmünster, Austria). The blood was immediately put on the ice and centrifuged in a

ACCEPTED MANUSCRIPT refrigerated centrifuge (10 min at 4000 rpm) with a yield of ~ 3000 µL of plasma. The extraction mixture (100 µL) was added either prior refrigerating (at three different time points T=0, T=4 and T=8) or after refrigerating (either during or after defrosting the sample or after defrosting the sample twice), shown in figure 1. The time from blood collection until the centrifugation was done took on average 16.25 min. For a detailed sampling protocol, see

PT

supporting information. For the baseline samples (T=0), acidified prior freezing, 100 µL

RI

plasma was added to the tubes containing ice-cold pre-aliquoted extraction solution and the

SC

tubes were mixed and snap-frozen in N2 (liquid), this step was finalized approximately one minute after the centrifugation was done. The T=4 and the T=8 samples were prepared in the

NU

same manner as for the baseline (T=0) samples, aliquoted 4 minutes and 8 minutes after the baseline samples respectively. The non-acidified prior freezing samples were aliquoted in 100

MA

µL and snap-frozen in N2 (liquid), this step was finalized approximately two to four minutes after the baseline samples and prior the T=4 samples. The ‘worst case’ scenario samples were

D

aliquoted in 100 µL, stored for six hours in a cool box and then snap-frozen in N2 (liquid). A

AC

CE

PT E

summarized sampling protocol is shown in table 1.

[Figure 1, about here]

[Table 1, about here] 2.3 Extraction The acidified plasma and the plasma aliquots were stored at -80˚C until the day of extraction. The aliquots of the “twice defrost” samples were let to defrost for 10 minutes at room

ACCEPTED MANUSCRIPT temperature, vortexed, snap-frozen once again and put back to -80˚C. Extraction and analysis on UHPLC-ESI-MSMS were performed within 2 weeks from the sampling time point. During extraction, the acidified samples were defrosted at room temperature, homogenized in a bead mill (Retsch MM400) at 30Hz for 1 minute. The proteins were let to precipitate on ice for 20 minutes followed by centrifugation at +4˚C, 14000 rpm for 20 minutes (Hettich Zentrifugen).

PT

The supernatants were collected and transferred to a spin filter (Ultrafree MC, 0.22µm,

RI

Millipore, Burlington, MA, USA) centrifuged at +4˚C, 14000 rpm for 10 min to remove any

SC

remaining protein precipitate. 50 µL of the supernatant was transferred to an LC-MS vial and 1 µL was injected and analysed by UHPLC-ESI-MSMS. The remaining supernatant was

MA

months to check the stability of the extracts.

NU

stored at -80°C. Baseline samples (T=0) from three participants were analysed after 3 and 9

D

2.4 UHPLC-ESI-MSMS

PT E

A 1290 Infinity system from Agilent Technologies (Waldbronn, Germany) coupled to an Agilent 6490 Triple quadrupole mass spectrometer was used. The chromatographic separation

CE

was achieved using an Acquity UPLC HSS T3 column (2.1 x 50 mm, particle size 1.8 µm, Waters Corporation, Milford, MA, USA), thermostated at 40 °C. The mobile phase consisted

AC

of solvent A (0.1 % formic acid) and solvent B (Acetonitrile, 0.1 % formic acid) and a linear gradient based on the method reported by Cao et al., 2013 was carried out: 0-1 min (0%B), 12 min (0%B-1%B), 2-2.5 min (1-5%B), 2.5-4 (5-99%B), 4-5 (99%B), 5-5.5 (99%B-0%B), 5.5-8 (0%B). The flow rate was kept constant at 400 µL/min.

The compounds were detected with an Agilent 6490 triple quadrupole mass spectrometer equipped with a jet stream electrospray source operating in positive ion mode. The capillary

ACCEPTED MANUSCRIPT voltage was set at 4000 V. The jet-stream gas temperature was 150°C with a gas flow of 11 L/min, sheath gas temperature of 360 °C, and sheath gas flow of 14 L min-1. The nebulizer pressure was set to 20 psi. The iFunnel parameters were set to 150 V and 60 V for highpressure RF and low-pressure RF, respectively. The fragmentor was set to 380 V and the cell accelerator voltage to 7 V.

PT

The LC-MS parameters for native and labelled internal standards are listed in Table 2

RI

including monitored transitions (precursor and product ions), collision energies and retention

SC

times and linearity. The auto-oxidation that GSH undergoes to GSSG is also expected to occur for the labelled GSH (internal standard) therefore transitions for single IS-GSSG* and

NU

double labelled IS-GSSG** were also monitored to detect any auto-oxidation of GSH-IS. Quality Control (QC) samples prepared by pooling 10 µL of extract from 120 randomly

MA

selected samples. The QC pool were split in 16 different LC-vials and analyzed throughout

PT E

D

the sample series to follow the stability.

CE

[Table 2, about here]

AC

2.5 Ethical approval

The study was conducted in accordance with the declaration of Helsinki. The participants signed an informed consent prior to blood sampling. The protocol was approved by the Ethics Committee of Umeå University (Dnr: 2015-226-31M).

3. Results

ACCEPTED MANUSCRIPT Different sample handling procedures were shown to have a large impact on the level of both GSH and GSSG in human plasma. The samples acidified immediately and prior freezing, i.e. the baseline samples (T=0), showed the highest concentration of GSH and the lowest concentration of GSSG, indicating a clear association between the degree of auto-oxidized GSH with time from sampling to acidification and freezing. The mean concentration of GSH

PT

at baseline (T=0, 1.53±0.04) was significantly reduced after 4 minutes (T=4, 1.36±0.05,

RI

p<0.001) and reduced by as much as 25% after 8 minutes (T=8, 1.13±0.05, p<0.001). Defrost

SC

under normal and acidic conditions did not indicate any significantly different GSH levels (p=0.06 and p=0.10, respectively). For the conditions where the samples were defrosted twice

NU

or left in a cool box, the levels of GSH were drastically reduced (0.42±0.03 and 0.01±0.02, respectively). The levels of GSSG were only significantly different (p=0.04) from baseline

MA

(T=0) in the two worst case scenarios (twice defrost and cool box). The concentration of GSH and GSSG for each participant for each of the 8 different sample handling procedures, are

PT E

D

shown in Figure 2a and b.

CE

[Figure 2a and b, about here]

AC

The reduction in the ratio between GSH and GSSG was clearly associated with the sample handling (Figure 3). The mean ratio among our participants was reduced by 20% up to 95%, depending on the different sample handling procedures and compared to the baseline samples (T=0). Hence, no procedure was able to fully mirror the ratio that was identified when the samples were acidified immediately after sampling (Figure 3).

[Figure 3, about here]

ACCEPTED MANUSCRIPT

In addition, a stability test was performed of the baseline samples (T=0) for three of the individuals. The additional extracts were stored in the freezer (-80°C) for either 3 months or 9 months before the analysis. The level of GSH and GSSG was significantly changed at both 3 and 9 months of storage (p<0.001) (table 3), this change resulted in a different GSH/GSSG

PT

ratio (Figure 3). Single IS-GSSG* and double labelled IS-GSSG** was detected in the 9

RI

months samples but not in the 3 months samples indicating that auto-oxidation was on-going

NU

SC

in storage at -80˚C.

D

MA

[Table 3, about here]

PT E

4. Discussion

The aim of this study was to investigate the impact of different pre-processing methods on the

CE

level of GSH and GSSG in plasma. Lower GSH levels and lower GSH/GSSG ratios have been indicated as an important measure in various diseases, and results were shown to differ

AC

between studies [2,15]. The pre-processing procedure has been pointed out to be a conceivable cause for the large variability in the results, and it is therefore of great importance to establish reliable pre-processing routines in order to make different measurements comparable [18,20,31]. Large differences in the levels of GSH and GSSG, as well as a difference in the GSH/GSSG ratios, were identified in this study and clearly associated with the sample handling procedures prior to the analysis (Figure 2 and 3). Leaving the plasma on ice for an additional 4 minutes, before adding the extraction mixture, showed a significant

ACCEPTED MANUSCRIPT reduction in the level of GSH, although the oxidation was not as fast as established by other studies [19,20]. An increase in GSSG was only observed form the two ‘worst case’ scenarios used in this study (i.e. twice defrost and left in a cool box). The increase in GSSG was not corresponding to the decrease in GSH, which also have been shown previously [19,20]. Worth mentioning, the individual ‘G’ in this study was found with clearly higher

PT

concentrations both for the GSH and the GSSG, compared with the other individuals, and

RI

indicated to bias the statistics due to supposed differences in total blood plasma concentration.

SC

Notably, the individual ‘G’ was the only participant with known issues regarding the rate of collecting blood, however if this was the true cause of the shifted levels is merely speculation.

NU

The reproducibility of the QC samples were 1.8% and 6.1 % RSD for GSH and GSSG

MA

respectively.

D

The IS-GSH was stable throughout the analysis and the relative standard deviation (RSD) of

PT E

the entire analysis (81 h) was 11.6%, and no labelled form of GSSG could be detected within the samples. Hence, once the samples have been acidified the auto-oxidation of GSH to

CE

GSSG can be neglected. Earlier studies have indicated that the level of GSH might increase under acidic conditions [4], however not observed in this study. Notably, the calibration

AC

samples, prepared once, used throughout the analysis and stored in the freezer in parallel with the real samples, indicated some degree of auto-oxidation (GSH being converted into GSSG) during the long-time storage. Hence, the auto-oxidation of GSH was also linked with the time of storage, observed both for the real samples as well as for the calibration samples. The GSH/GSSG ratio indicated to be fairly stable after 3 months of storage, however, after 10 months of storage the GSH/GSSG ratio was reduced (T=0 (immediately) = 12.3 compared to T=0 (after 10 months) = 6.5, see table 1).

ACCEPTED MANUSCRIPT

For the ‘worst case’ scenario, when the sample was left in a cool box during the day prior to analysis, more than 90% loss of GSH and increased levels of GSSG was observed, compared to the baseline samples (Figure 1). This is a common way of handling blood samples during collection in field studies. It is also common to use samples (e.g. biobank samples) that have

PT

been defrosted multiple times. We therefore created a scenario were we defrosted the samples

RI

twice prior to analysis and there was a vast loss of GSH also in this case and only about 25%

NU

SC

of the initial (at T=0) amount of GSH were retained.

5. Conclusions

MA

In this study set out to develop a pre-processing protocol for analysis of GSH and GSSG in plasma by UHPLC-ESI-MSMS, formulating a standardized protocol in order to make the

D

measurements comparable. The results show the importance of control over the whole process

PT E

from sampling and pre-processing procedure to the analysis. For compounds like GSH and GSSG every minute is of importance in order to get reliable results since the auto-oxidation of

CE

GSH cannot be expected to be converted into corresponding proportions of GSSG (e.g the breakdown of GSH into cysteinylglycine, CysGly or oxidation to GS-mixed disulfides [1–3]).

AC

The preferred way to handle samples, before the analysis of GSH/GSSG (besides analysing the plasma immediately), was to add the extraction mixture immediately after centrifugation and aliquotation, snap-freeze the samples, and then store them in a freezer (-80°C). If samples have to be stored or kept untreated for the purpose of analysing other analytes, the method where the aliquoted plasma will be defrosted under acidic conditions is recommended. It was practically possible to prepare the samples in 20 minutes, from the start time of blood sampling through the centrifugation and aliquotation, and til the snap-freezing of the samples.

ACCEPTED MANUSCRIPT The most important factor for comparable results of GSH/GSSG ratio is a standardized protocol for sample handling. Our study show that a consequent time from blood with-drawal to snap-frozen sample is a very important factor to get comparable results. This result might have implications for the analysis of other non-stable analytes in plasma as well, however that

PT

needs to be further evaluated.

RI

Acknowledgements

SC

The generous funding provided by the Swedish metabolomics Centre and Swedish Research

NU

Council FORMAS for this project is gratefully acknowledged (2014-1229 and 2016-1110). The authors also thank Helen Bertilsson, Inga-Britt Carlsson and Siv Sääf for excellent work

D

MA

with the collection of blood.

[1]

PT E

References

E. Camera, M. Picardo, Analytical methods to investigate glutathione and related

181–206.

D. Giustarini, I. Dalle-Donne, D. Tsikas, R. Rossi, Oxidative stress and human

AC

[2]

CE

compounds in biological and pathological processes, J. Chromatogr. B. 781 (2002)

diseases: Origin, link, measurement, mechanisms, and biomarkers, Crit. Rev. Clin. Lab. Sci. 46 (2009) 241–281. [3]

D.A. Dickinson, H.J. Forman, Cellular glutathione and thiols metabolism, Biochem. Pharmacol. 64 (2002) 1019–1026.

[4]

R. Rossi, A. Milzani, I. Dalle-Donne, D. Giustarini, L. Lusini, R. Colombo, P. Di Simplicio, Blood glutathione disulfide: In vivo factor or in vitro artifact?, Clin. Chem.

ACCEPTED MANUSCRIPT 48 (2002) 742–753. [5]

H. Sies, Oxidative stress: A concept in redox biology and medicine, Redox Biol. 4 (2015) 180–183.

[6]

G. Wu, Y. Fang, S. Yang, J.R. Lupton, N.D. Turner, Recent Advances in Nutritional

PT

Sciences Glutathione Metabolism and Its Implications for Health 1, Environ. Heal. 134

[7]

D.M. Townsend, K.D. Tew, H. Tapiero, The importance of glutathione in human

SC

disease, Biomed. Pharmacother. 57 (2003) 145–155.

B.F. Asher, F.T. Guilford, Oxidative stress and low glutathione in common ear, nose,

NU

[8]

RI

(2004) 489–492.

and throat conditions: A systematic review, Altern. Ther. Health Med. 22 (2016) 44–

C.-Y. Lu, Y.-C. Ma, J.-M. Lin, C.-Y. Li, R.S. Lin, F.-C. Sung, Oxidative Stress

D

[9]

MA

50.

PT E

Associated with Indoor Air Pollution and Sick Building Syndrome-Related Symptoms among Office Workers in Taiwan, Inhal. Toxicol. 19 (2007) 57–65.

CE

[10] M.L. Pall, Elevated nitric oxide/peroxynitrite theory of multiple chemical sensitivity: Central role of N-methyl-D-aspartate receptors in the sensitivity mechanism, Environ.

AC

Health Perspect. 111 (2003) 1461–1464. [11] C. de Luca, D. Raskovic, V. Pacifico, J.C.S. Thai, L. Korkina, The search for reliable biomarkers of disease in multiple chemical sensitivity and other environmental intolerances, Int. J. Environ. Res. Public Health. 8 (2011) 2770–2797. [12] B.F. Bessac, S.-E. Jordt, Breathtaking TRP Channels: TRPA1 and TRPV1 in Airway Chemosensation and Reflex Control, Physiology. 23 (2008) 360–370. [13] M. Sivonová, I. Zitnanová, L. Hlincíková, I. Skodácek, J. Trebatická, Z. Duracková,

ACCEPTED MANUSCRIPT Oxidative stress in university students during examinations., Stress. 7 (2004) 183–188. [14] J.F. Lesgards, P. Durand, M. Lassarre, P. Stocker, G. Lesgards, A. Lanteaume, M. Prost, M.P. Lehucher-Michel, Assessment of lifestyle effects on the overall antioxidant capacity of healthy subjects, Environ. Health Perspect. 110 (2002) 479–486.

PT

[15] D.P. Jones, V.C. Mody, J.L. Carlson, M.J. Lynn, P. Sternberg, Redox analysis of human plasma allows separation of pro-oxidant events of aging from decline in

RI

antioxidant defenses, Free Radic. Biol. Med. 33 (2002) 1290–1300.

SC

[16] M.A. Mansoor, A.M. Svardal, P.M. Ueland, Determination of the in vivo redox status

NU

of cysteine, cysteinylglycine, homocysteine, and glutathione in human plasma, Anal. Biochem. 200 (1992) 218–229.

MA

[17] D.P. Jones, J.L. Carlson, P.S. Samiec, P. Sternberg, V.C. Mody, R.L. Reed, L. Ann, S. Brown, Glutathione measurement in human plasma Evaluation of sample collection,

D

storage and devatization conditions for analysis of densyl derivatives by HPLC, Clin.

PT E

Chim. Acta. 275 (1998) 175–184.

[18] A. Pastore, G. Federici, E. Bertini, F. Piemonte, Analysis of glutathione: Implication in

CE

redox and detoxification, Clin. Chim. Acta. 333 (2003) 19–39.

AC

[19] M.E. Anderson, A. Meister, Dynamic state of glutathione in blood plasma, J. Biol. Chem. 255 (1980) 9530–9533. [20] K. Ngamchuea, C. Batchelor-McAuley, R.G. Compton, Rapid Method for the Quantification of Reduced and Oxidized Glutathione in Human Plasma and Saliva, Anal. Chem. 89 (2017) 2901–2908. [21] Å. Florholmen-Kjær, R.A. ndr. Lyså, O.M. Fuskevåg, R. Goll, A. Revhaug, K.E. rlen. Mortensen, A sensitive method for the analysis of glutathione in porcine hepatocytes,

ACCEPTED MANUSCRIPT Scand. J. Gastroenterol. 49 (2014) 1359–1366. [22] Y. Iwasaki, Y. Saito, Y. Nakano, K. Mochizuki, O. Sakata, R. Ito, K. Saito, H. Nakazawa, Chromatographic and mass spectrometric analysis of glutathione in biological samples, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 877 (2009)

PT

3309–3317. [23] X. Guan, B. Hoffman, C. Dwivedi, D.P. Matthees, A simultaneous liquid

RI

chromatography/mass spectrometric assay of glutathione, cysteine, homocysteine and

SC

their disulfides in biological samples, J. Pharm. Biomed. Anal. 31 (2003) 251–261.

NU

[24] D.E. Hammermeister, J. Serrano, P. Schmieder, D.W. Kuehl, Characterization of dansylated glutathione, glutathione disulfide, cysteine and cystine by narrow bore

MA

liquid chromatography/electrospray ionization mass spectrometry, Rapid Commun. Mass Spectrom. 14 (2000) 503–508.

D

[25] R.L. Norris, G.K. Eaglesham, G.R. Shaw, M.J. Smith, R.K. Chiswell, A.A. Seawright,

PT E

M.R. Moore, A sensitive and specific assay for glutathione with potential application to glutathione disulphide, using high-performance liquid chromatography-tandem mass

CE

spectrometry, J. Chromatogr. B Biomed. Sci. Appl. 762 (2001) 17–23.

AC

[26] F. Zhang, M.J. Bartels, D.R. Geter, Y.-C. Jeong, M.R. Schisler, A.J. Wood, L. Kan, B.B. Gollapudi, Quantitation of glutathione by liquid chromatography/positive electrospray ionization tandem mass spectrometry, Rapid Commun. Mass Spectrom. 22 (2008) 3608–3614. [27] D. Giustarini, G. Colombo, M.L. Garavaglia, E. Astori, N.M. Portinaro, F. Reggiani, S. Badalamenti, A.M. Aloisi, A. Santucci, R. Rossi, A. Milzani, I. Dalle-Donne, Assessment of glutathione/glutathione disulphide ratio and S-glutathionylated proteins in human blood, solid tissues, and cultured cells, Free Radic. Biol. Med. 112 (2017)

ACCEPTED MANUSCRIPT 360–375. [28] R. Rellán-Álvarez, L.E. Hernández, J. Abadía, A. Álvarez-Fernández, Direct and simultaneous determination of reduced and oxidized glutathione and homoglutathione by liquid chromatography-electrospray/mass spectrometry in plant tissue extracts,

PT

Anal. Biochem. 356 (2006) 254–264. [29] T. Moore, A. Le, A.K. Niemi, T. Kwan, K. Cusmano-Ozog, G.M. Enns, T.M. Cowan,

RI

A new LC-MS/MS method for the clinical determination of reduced and oxidized

SC

glutathione from whole blood, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 929

NU

(2013) 51–55.

[30] I. Squellerio, D. Caruso, B. Porro, F. Veglia, E. Tremoli, V. Cavalca, Direct glutathione

MA

quantification in human blood by LC-MS/MS: Comparison with HPLC with electrochemical detection, J. Pharm. Biomed. Anal. 71 (2012) 111–118.

D

[31] D. Giustarini, D. Tsikas, G. Colombo, A. Milzani, I. Dalle-Donne, P. Fanti, R. Rossi,

PT E

Pitfalls in the analysis of the physiological antioxidant glutathione (GSH) and its disulfide (GSSG) in biological samples: An elephant in the room, J. Chromatogr. B

CE

Anal. Technol. Biomed. Life Sci. 1019 (2016) 21–28.

AC

[32] L. Cao, D. Waldon, Y. Teffera, J. Roberts, M. Wells, M. Langley, Z. Zhao, Ratios of biliary glutathione disulfide (GSSG) to glutathione (GSH): A potential index to screen drug-induced hepatic oxidative stress in rats and mice, Anal. Bioanal. Chem. 405 (2013) 2635–2642.

ACCEPTED MANUSCRIPT Figure captions

Figure 1. Overview of the experimental procedure

Figure 2 a and 2b.

RI

PT

The concentration (µM±SD) of GSH and GSSG for each participant for each of the different sample handling procedures

Figure 3.

AC

CE

PT E

D

MA

NU

SC

GSH/GSSG ratio for each of the different sample handling procedures

ACCEPTED MANUSCRIPT

Table 1. Sampling protocol Blood collection start: 2016-06-13 at 07.50. Extraction solution Sample (µl) Time Volym (µl)

centrifugation: 07.54-08.04

N2-liq

-80°

cool bag

Comments

100

08.08

100 H2O

08.09

08.30

ex-sol.+H2O, frozen in N2

T=0

100

08.08

100 plasma

08.09

08.30

ex-sol.+plasma, frozen in N2

T=4

100

08.12

100 plasma

08.13

08.30

ex-sol.+plasma (+4min), frozen in N2

T=8

100

08.16

100 plasma

08.17

08.30

B3

08.10

100 plasma

08.12

08.30

B4

08.10

100 plasma

08.12

08.30

B5

08.10

100 plasma

08.12

08.30

B6

08.10

100 plasma

08.12

B7

08.14

100 plasma

14.15

08.30

13.55

ex-sol.+plasma (+8min), frozen in N2 immediately frozen in N2, addition of ex-sol. mix to frozen plasma immediately frozen in N2, addition of ex-sol. to defrozen plasma (RT) immediately frozen in N2, one freeze thaw cycle, addition of ex-sol. to frozen plasma (RT) immediately frozen in N2, two freeze thaw cycle, addition of ex-sol. to frozen plasma (RT) plasma in coldbag over a day, addition of ex-sol. to frozen plasma (RT)

RI

SC

NU

MA

D PT E CE AC

PT

Blank

08.14

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

Table 2. Mass spectrometry parameters for each compound analyzed: precursor and product ion, collision energy (CE), retention time, linearity and slope. Compound Precursor Product Collision Retention Linear Linearity Slope ion ion* energy time (min) range (V) (µM) GSH 308.1 179 10 1.6 0.01-10 0.9996 3.21 162 10 GSSG 307.1 130 10 3.7 0.01-10 0.9984 1.35 355 10 GSH-IS 311.1 182 10 1.6 NA NA NA 165 10 GSSG-IS* 308.5 309.5 10 3.7 NA NA NA GSSG-IS** 310.0 311.0 10 3.7 NA NA NA *Both the quantifying (in bold) and qualifying transitions are reported; NA, not applicable

ACCEPTED MANUSCRIPT Table 3. Concentration of GSH and GSSG (µM (SD)) at T=0 at the initial analysis, after 3 and 9 months of storage in -80°C.

T=0 Initial analysis

T=0 after 3 months

T=0 after 9 months

Participant, µM (SD) GSSG 0.34 (0.01) 0.18 (0.01)

GSH

GSSH

Not analyzed Not analyzed

Not analyzed Not analyzed

Not analyzed

Not analyzed

1.87 (0.04)

0.29 (0.02)

D

MA

NU

SC

RI

PT

GSH 3.26 (0.04) 1.91 (0.07)

PT E

E

GSSG 0.15 (0.01) 0.09 (0.00) 0.09 (0.00)

CE

B

AC

A

GSH 1.68 (0.07) 0.94 (0.05) 1.11 (0.03)

ACCEPTED MANUSCRIPT Highlights Seven different pre-processing scenarios for analysis of GSH/GSSG were investigated



A large difference in the levels of GSH and GSSG in plasma were identified



The difference in concentrations was associated with the pre-processing procedure



Pre-processing procedures are important for reliable plasma levels of GSH and GSSG



Extraction mixture should be added prior to freezing to -80°C.

AC

CE

PT E

D

MA

NU

SC

RI

PT



Figure 1

Figure 2

Figure 3