MS method for the clinical determination of reduced and oxidized glutathione from whole blood

Accepted Manuscript Title: A New LC-MS/MS Method for the Clinical Determination of Reduced and Oxidized Glutathione from Whole Blood Authors: Tereza Moore Anthony Le Anna-Kaisa Niemi Tony Kwan Krinstina Cusmano-Ozog Gregory M. Enns Tina M. Cowan PII: DOI: Reference:

S1570-0232(13)00214-6 http://dx.doi.org/doi:10.1016/j.jchromb.2013.04.004 CHROMB 18356

To appear in:

Journal of Chromatography B

Received date: Revised date: Accepted date:

7-2-2013 3-4-2013 6-4-2013

Please cite this article as: T. Moore, A. Le, A.-K. Niemi, T. Kwan, K. Cusmano-Ozog, G.M. Enns, T.M. Cowan, A New LC-MS/MS Method for the Clinical Determination of Reduced and Oxidized Glutathione from Whole Blood, Journal of Chromatography B (2013), http://dx.doi.org/10.1016/j.jchromb.2013.04.004 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.

Ac ce

pt

ed

M

an

us

cr

ip t

Highlights We present a one-step preanalytical procedure that prevents oxidation of blood GSH We present a LC-MS/MS method for detection and quantitation of blood GSH and GSSG We determine references ranges for whole blood GSH, GSSG, and the ratio of GSH/GSSG

1

Page 1 of 23

A New LC-MS/MS Method for the Clinical Determination of Reduced and Oxidized Glutathione

ip t

from Whole Blood

cr

Tereza Moorea, Anthony Lea, Anna-Kaisa Niemib, Tony Kwanc, Krinstina Cusmano-Ozoga,1,

a

M

an

us

Gregory M. Ennsd, Tina M. Cowana*

Department of Pathology, Stanford University School of Medicine, Stanford, CA, 94305

b

ed

Division of Medical Genetics, Department of Pediatrics, Stanford University School of Medicine, Stanford, CA 94305 Stanford University Medical Center, Stanford, CA 94305

d

Department of Pediatrics, Stanford University School of Medicine, Stanford, CA, 94305

1

pt

c

Ac ce

Present address: Department of Genetics and Metabolism, Children’s National Medical Center, Washington DC, 20010

* Corresponding author: Tina M. Cowan, Ph.D. Stanford University Clinical Laboratories 3375 Hillview Avenue Room 2010 Palo Alto, CA 94304 Office: 650-724-7858 Fax: 650-724-1567 E-mail: [email protected] 2

Page 2 of 23

ABSTRACT Reduced levels of glutathione (γ-glutamylcysteinylglycine, GSH) and the ratio of GSH to

ip t

glutathione disulfide (GSSG) can serve as important indicators of oxidative stress and disease risk. Measured concentrations of GSH and GSSG vary widely between laboratories, largely due

cr

to the instability of GSH during sample handling and variables arising from different analytical

us

methods. We have developed a simple and sensitive liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for measuring whole blood GSH and GSSG that minimizes

an

preanalytic and analytic variability, reliably eliminates interference from ion suppression, and can easily be implemented in clinical laboratories. Samples were deproteinized with

M

sulfosalicylic acid (SSA) and derivatized with N-ethylmaleimide (NEM) in a single preparative step, and the resulting supernatants combined with stable-isotope internal standards (GSH-13C, N-NEM and GSSG-13C,15N), subjected to chromatographic separation using a Hypercarb

ed

15

column, and analyzed by MS/MS in the positive-ion mode. Results showed excellent linearity

pt

for both GSH and GSSG over the ranges of physiologic normal, with inter- and intra-assay CV’s

Ac ce

of 3.1-4.3% and accuracy between 95-101%. The lower limits of detection (LLOD) were 0.4 µM for GSH and 0.1µM for GSSG and the lower limits of quantitation (LLOQ) were 1.5μM for GSH and 0.1μM for GSSG. Derivatized samples are stable for at least three years when stored at -80°C, and underivatized samples for at least 24 hours at either 4°C or room temperature. Reference intervals were determined for 59 control samples, and were (mean±SD): GSH, 900 ±140μM; GSSG, 1.17 ±0.43μM; GSH/ GSSG, 880 ±370.

Keywords: glutathione, glutathione disulfide, tandem mass spectrometry, oxidative stress, clinical testing 3

Page 3 of 23

1. INTRODUCTION Glutathione (γ-glutamylcysteinylglycine) is an endogenous antioxidant that plays a

ip t

central role in the cellular defense against oxidative damage. It occurs in virtually all mammalian tissues and exists either free or conjugated to proteins and other endogenous and

cr

exogenous compounds. Free glutathione is present mainly in its reduced form (GSH), and is

us

readily oxidized to glutathione disulfide (GSSG) under conditions of free radical accumulation. GSSG itself is a stable molecule requiring the action of glutathione reductase to regenerate

an

reduced GSH. Oxidative stress therefore leads to decreased GSH levels that in turn predispose to increased susceptibility to oxidative damage, a cycle which is thought to contribute to various

M

disease states. Consequently, a decrease in GSH and/or the ratio of GSH/GSSG have been used as indicators of oxidative stress and disease risk in a variety of conditions including diabetes

ed

mellitus, Parkinson’s disease, multiple sclerosis, cystic fibrosis, and mitochondrial disease [1-7]. The use of GSH, GSSG, and GSH/GSSG as biomarkers of oxidative stress has led to the

pt

development of various methods for measuring these compounds in biological samples,

Ac ce

including urine, plasma, whole blood, and extracts from erythrocytes and neutrophils [8-13]. Many of the assays involve separation by high-pressure liquid chromatography (HPLC) with various detection techniques, such as ultraviolet (UV) absorbance, fluorescence, electrochemical and mass spectrometry (MS) detection. Additional methods include gas chromatography with MS detection and capillary electrophoresis with UV absorbance or coulometric detection. Reported concentrations for GSH and GSSG vary widely; in particular, GSSG levels are strongly influenced by oxidative conditions of sample handling and can appear falsely elevated if conditions are not properly controlled [14-16].

4

Page 4 of 23

To address the need for a robust method for clinical GSH and GSSG determinations, we have developed a LC-MS/MS method using equipment standard in many clinical laboratories. This approach minimizes preanalytical variability through a one-step procedure of

ip t

deproteinization and derivatization that prevents artifactual oxidation of GSH, and is easily adapted to the clinical setting without requiring excessive constraints on sample handling or

cr

storage. Chromatographic conditions have been optimized to eliminate ion suppression and

us

increase precision and sensitivity, even at very low analyte concentrations. A reference range has been determined using a population of healthy individuals and controlled preanalytic

MATERIALS AND METHODS

M

2.

an

conditions.

Materials

ed

2.1

N-ethylmaleimide (NEM) (Sigma-Aldrich) was reconstituted in

pt

15% methanol to make a 3mg/ml (24mM) stock solution.

Precipitating

Ac ce

solution was prepared to a final concentration of 20mM NEM, 2% SSA and 2mM EDTA in 15% methanol. Unlabeled GSH and GSSG standards (Sigma-Aldrich) were reconstituted in 50% methanol to make stock solutions of 1mg/ml each. From these, working solutions of GSH (200μM) and GSSG (1.63μM) were prepared by diluting with 15% methanol.

Stable-isotope internal standards of GSH (GSH-13C) and

GSSG (15N and GSSG-13C,15N ) (AnaSpec Inc., Fremont, CA) were reconstituted in 50% methanol to make stock solutions of 0.95mg/ml, and were also diluted with 15% methanol to make internal standard 5

Page 5 of 23

working solutions of GSH-IS (200μM) and GSSG-IS (1.63μM). GSH and GSHIS working solutions also contained 12mM NEM.

Patient samples

ip t

2.2

Anonymized, residual blood samples from 59 healthy individuals (31 male, 28 female;

cr

age 1-87 years [mean 25 years]) were obtained from the Stanford Clinical Laboratory with IRB

us

approval. Blood was collected by venipuncture in heparin- or EDTA-containing tubes as part of routine healthcare evaluations, and was processed within 24 hours of collection. Samples were

an

pooled for use in precision and recovery studies, or analyzed individually for determination of

2.3

Sample preparation

M

reference intervals.

ed

50μl of whole blood was mixed with 200μl of precipitating solution (see above), vortexed briefly and allowed to incubate at room temperature for 45 minutes. Following derivatization, samples

pt

were centrifuged for 5 minutes at 14,000 rpm and the resultant supernatants transferred to new

Ac ce

tubes for either immediate liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis or storage at -80°C. For MS/MS analysis, 10μl of derivatized sample was mixed with 100μl of GSH-IS working solution, vortexed briefly and transferred into microplate inserts. A positive control sample of 10μl of 200μM GSH-NEM standard with 100μl GSH-IS was included in each run. For GSSG measurements, inserts were similarly prepared using 50μl of sample (supernatant or GSSG standard working solution) mixed with 50μl of GSSG-IS.

2.4 Chromatography

6

Page 6 of 23

Compounds were separated by liquid chromatography using a Hypercarb column (Thermo Scientific, 2.1mm X 50mm X 5µm) at room temperature. Eluent A was water containing 0.1% formic acid, and Eluent B was 100% acetonitrile containing 0.1% formic acid. The run consisted of

ip t

a linear gradient of 0% - 40% Eluent B over 3.3 min, a second linear gradient of 40% - 95% Eluent B over 0.9 min, isocratic delivery of 95% Eluent B for 0.7min, and a final wash with 100% Eluent A

cr

for 1.4 min. The eluent from the column was introduced into the mass spectrometer without

Tandem Mass Spectrometry

an

2.5

us

splitting between 3.2 and 4.2 min.

GSH and GSSG ions and fragments were detected using a triple-quadrupole mass

M

spectrometer (API 3000, Perkin-Elmer Sciex) with a Turbo Ion Spray Source, a Shimadzu System Controller (model SCL-10Avp), Shimadzu Solvent Delivery Module (model LC-10Advp) and a

ed

Shimadzu On-line Degasser (model DGU-14A). Instrument parameters for GSH and GSSG were as follows: declustering potential, 40; focusing potential, 200; collision energy, 22 (GSH) or 32

pt

(GSSG); entrance potential, 7.5; and collision cell exit potential, 18. Source and gas parameters for

Ac ce

both compounds were: nebulizer gas, 8; curtain gas, 8; ion spray voltage, 5500; collision gas, 8; and temperature, 360°C. A Leap Technology (CTC) Autosampler (model HTS PAL) using a Hamilton 100µl syringe delivered specimens with injection volumes of 10μl (GSH) or 15μl (GSSG). GSH-NEM and GSSG were monitored by their positive selected reaction monitoring (SRM) pairs using m/z 433 > 304 and m/z 613 > 355 respectively. Stable-isotope internal standards were monitored as m/z 435 > 306 (GSH-13C, 15N-NEM) and m/z 617 > 359 (GSSG-13C, 15N). The transition m/z 308 > 162 was also monitored for the presence of underivatized GSH, but none was detected. Data from MS/MS were acquired with Analyst software, version 1.4 and data

7

Page 7 of 23

calculations performed with Chemoview application, version 1.2b9. Results were multiplied by five to account for dilutions during sample preparation, yielding final concentrations of GSH and

ip t

GSSG in units of μmol/liter of whole blood.

2.6 Method performance evaluations

cr

Assay linearity was evaluated for GSH-NEM standard in 15% methanol at 25, 50, 100,

us

200, 400 and 500µM, and similarly for GSSG standard at 0.1, 0.2, 0.4, 0.8, 1.63, 8, 13 and 16.3µM. The lower limits of detection (LLOD) and quantitation (LLOQ) were assessed by

an

determining the lowest concentrations of GSH and GSSG that resulted in a signal-to-noise ratio of ≥3 (LLOD) and ≥10 (LLOQ). For precision assays, aliquots prepared from pooled blood were

M

analyzed 15 times in a single day (intra-assay-precision) or over 15 consecutive days (inter-assay precision). Recovery was assessed by preparing pooled blood from normal individuals with

ed

precipitating solution as described above (thereby stabilizing the endogenous GSH as its NEM adduct), and then spiking sample supernatants with known amounts of GSH-NEM (25, 50, 100,

pt

200, and 400µM) and GSSG (0.1, 0.2, 0.4, 0.8 and 1.6µM) standards prior to dilution with

Ac ce

internal standard and MS/MS analysis. GSH stability in whole blood was assessed by comparing levels from samples processed immediately after sample collection to those from samples stored at room temperature, 4ºC, and -80ºC. Statistical significance of the change in GSH concentrations at room temperature and 4ºC was assessed using Pearson correlation coefficients, and at -80ºC by matched pair analysis.

3. RESULTS

3.1

Chromatography 8

Page 8 of 23

GSSG and GSH-NEM were eluted from the column at 3.6 and 3.7 minutes respectively (Fig 1).

Negatively charged salts from the

sample matrix eluted in a single, sharp peak at 0.9 minutes (not

maximal ionization in the electrospray source.

ip t

shown), thereby eliminating sources of ion suppression and ensuring Analyses done

us

analyte sensitivity due to ion suppression.

cr

without column separation resulted in significantly decreased

Eluent B composition was evaluated at different concentrations

an

of acetonitrile (100% and 60%, with either 0.1% formic acid or 0.1% TFA) or methanol (15%, 20% and 50%, with 0.01% formic acid).

M

Eluent A was 100% water with 0.1% formic acid.

Optimal peak

quality together with lowest background noise level was achieved GSH and GSSG

ed

using Eluent B of 100% acetonitrile/0.1% formic acid.

were fragmented by collision-induced dissociation and the most

pt

intense fragment ion was used for quantification (GSH: m/z 433 >

Ac ce

304; GSSG: m/z 613 > 355). Parent-product ions of GSH-NEM, GSSG and their corresponding internal standards were easily detected without interfering molecular fragments.

No differences were appreciated

between blood samples obtained in either heparin or EDTA collection tubes.

3.2 Assay validation Linear coefficients >0.999 were obtained for GSH between 25-500µM and GSSG between 0.1-16 µM (Figure 2). Recoveries were evaluated over the concentration ranges 259

Page 9 of 23

400µM (GSH) or 0.1-1.6 µM (GSSG), with mean recoveries of 96% and 97% respectively (Table 1a). Intra-day imprecision was 3.3% and 3.1% for GSH and GSSG respectively, and inter-day imprecision 4.1% and 4.3%, respectively (Table 1b). The lower limits of detection

ip t

were 0.4 µM for GSH and 0.1µM for GSSG and the limits of quantitation were 1.5μM for GSH

cr

and 0.1μM for GSSG.

us

3.3 GSH stability

Experiments were performed to assess the effects of preanalytic sample handling on measured

an

GSH levels. Aliquots of whole blood from eight healthy controls were either (1) processed immediately for analysis, (2) stored at room temperature and processed at 1, 2, 3, 4, 5, 6, 7, 24 or

M

48 hours, or (3) stored at 4ºC and processed once daily over five days. For the samples kept at room temperature, GSH concentrations were not significantly altered after 24 hours but

ed

decreased by 52% after 48 hours, most likely a result of reactive oxygen species accumulation following prolonged oxygen exposure (Figure 3). Samples stored at 4°C also yielded

pt

GSH levels unchanged from baseline when measured after 24 hours,

Ac ce

with a more subtle 7% decrease seen after 48 hours, a result of slowed oxidation reactions at these lower temperatures. Samples compared at initial collection and after 3 years of storage at -80°C showed no significant change in GSH measurements (p<0.4)

3.4

GSH and GSSG levels in healthy individuals Whole blood GSH and GSSG concentrations were determined from

59 healthy individuals undergoing routine healthcare evaluations

10

Page 10 of 23

(31 male, 28 female; mean age 25 years [range 1-87 years]).

There

were no obvious differences in levels as a function of age, although the small sample sizes in each age category did not permit As a group, the mean

ip t

a more rigorous statistical comparison.

concentration (±1SD) for GSH was 900 + 140µM, GSSG 1.17 + 0.43µM,

us

cr

and GSH/GSSG 880 + 370.

an

4. DISCUSSION

Oxidative stress, or the accumulation of increased amounts of reactive oxygen species

M

with insufficient levels of antioxidants, is recognized as a contributing factor to the pathophysiology of a variety of inherited and acquired conditions [17-19]. Despite this widely-

ed

accepted view, supporting data from patient samples is limited in part because published methods evaluating oxidative stress frequently fail to control for factors affecting sample

pt

stability in a manner appropriate to testing clinical subjects. We describe a simple LC-MS/MS

Ac ce

assay for the quantification of whole blood GSH and GSSG that can easily accommodate common preanalytic effects of sample handling and storage. Whole blood can be collected and held at room temperature or 4ºC prior to transport, and shipped to the testing laboratory via overnight express at room temperature or on a cool pack. Once received, samples can be easily processed in a single precipitation/derivatization step, and the supernatants stably stored until time of analysis, or at least three years. Although many different analytic platforms have been described for measuring GSH and GSSG [reviewed in 8], LC-MS/MS is particularly well suited to the clinical setting for several reasons. Importantly, our method uses standard equipment already found in many clinical 11

Page 11 of 23

laboratories. Analysis by LC-MS/MS confers superior sensitivity and specificity and, unlike other methods such as gas chromatography, spectrophotometry, or coulometry, does not require additional derivatization steps [8,20]. Several other LC-MS/MS methods for GSH and GSSG

ip t

have been reported [e.g., 21,22], most using reversed-phase or hydrophobic interaction (HILIC) chromatography for separation. The Hypercarb column used here excels at the separation of

cr

small polar compounds, with the added advantages of rapid equilibration time and sharp focusing

us

of ion suppression agents (i.e., salts) in an early-eluting peak that is well separated from the rest of the chromatographic run. Harwood et al [10] also reported the use of Hypercarb separation in

an

the measurement of GSH and GSSG, but with longer retention times and poorer peak shape for GSSG, presumably due to specific differences in chromatographic conditions including the

M

mobile phase.

The wide range of reported GSH and GSSG concentrations from healthy individuals is

ed

well documented [14]. While this may in part be due legitimate variability within subjects [23] arising from physiologic and/or environmental factors, much of the variation likely is due to

pt

preanalytic or analytic conditions favoring autooxidation of GSH to GSSG, degradation of GSH

Ac ce

to derivative amino acids or dipeptides, or ion suppression from sample matrices. In particular, GSSG values vary widely depending on whether NEM or a similar alkylating agent is used to prevent oxidation of GSSG. In our study, we found whole blood GSH and GSSG levels that were generally comparable to the other studies in which samples were stabilized with NEM [8,9,24,25].

In addition to whole blood, the method described here can be applied to the measurement of GSH and GSSG in a variety of other sample types with essentially no modification. Preliminary studies of such samples include plasma, peripheral blood mononuclear cells, and

12

Page 12 of 23

liver tissue homogenates, but validated testing procedures and reference intervals for each tissue type are still needed. The increased sensitivity of our assay resulting from chromatographic removal of ion suppression makes it ideal for analysis of multiple sample types where GSH and

ip t

GSSG concentrations are very low and/or sample volume is very small. Such studies are necessary to more specifically evaluate the role of GSH and oxidative stress in disease states in

cr

the absence of potentially confounding influences from the relatively high GSH levels in

an

us

erythrocytes.

M

Conclusion:

We have developed a new LC-MS/MS method that minimizes preanalytic variability

ed

through a one-step procedure of deproteinization and derivatization, and chromatographic conditions that eliminate ion suppression and increase precision and sensitivity. Additional

pt

advantages of our method, including the small sample requirement, simple and rapid

Ac ce

preanalytical processing, and wide automation possibilities, make this method ideal for routine and large-scale clinical testing.

13

Page 13 of 23

Acknowledgements

Ac ce

pt

ed

M

an

us

cr

Pharmaceuticals, Inc. for providing support for this research.

ip t

The authors are grateful to Michael and Ellen Michelson, Bobbie and Mike Wilsey and Edison

14

Page 14 of 23

References

ip t

[1]Yoshida K, Hirokawa J, Tagami S, Kawakami Y, Urata Y, Kondo T. Diabetologia. 38(1995)201.

us

cr

[2]Sharma A, Kharb S, Chugh SN, Kakkar R, Singh GP. Metabolism. 49(2000)160.

an

[3]Sian J, Dexter DT, Lees AJ, Daniel S, Agid Y, Javoy-Agid F, Jenner P, Marsden CD. Ann Neurol. 36(1994)348.

M

[4]Pearce RK, Owen A, Daniel S, Jenner P, Marsden CD. J Neural Transm. 104(1997)661.

ed

[5]Roum JH, Buhl R, McElvaney NG, Borok Z, Crystal RG. J Appl Physiol. 75(1993)2419.

pt

[6]Hargreaves IP, Sheena Y, Land JM, Heales SJ. J Inherit Metab Dis. 28(2005)81.

Ac ce

[7]Atkuri KR, Cowan TM, Kwan T, Ng A, Herzenberg LA, Herzenberg LA, Enns GM. Proc Natl Acad Sci U S A. 106(2009)3941.

[8]Monostori P, Wittmann G, Karg E, Túri S. J Chromatogr B Analyt Technol Biomed Life Sci. 877(2009)3331.

[9]Iwasaki Y, Saito Y, Nakano Y, Mochizuki K, Sakata O, Ito R, Saito K, Nakazawa H. J Chromatogr B Analyt Technol Biomed Life Sci. 877(2009)3309.

15

Page 15 of 23

[10]Harwood DT, Kettle AJ, Brennan S, Winterbourn CC. J Chromatogr B Analyt Technol Biomed Life Sci. 877(2009)3393.

ip t

[11] Forman HJ, Zhang H, Rinna A. Mol Aspects Med. 30(2009)1.

cr

[12] Yap LP, Sancheti H, Ybanez MD, Garcia J, Cadenas E, Han D. Methods Enzymol. 473(2010)137.

an

us

[13] Steghens JP, Flourié F, Arab K, Collombel C. J Chromatogr B Analyt Technol Biomed Life Sci. 798(2003)343.

ed

M

[14] Rossi R, Milzani A, Dalle-Donne I, Giustarini D, Lusini L, Colombo R, Di Simplicio P. Clin Chem. 48(2002)742.

pt

[15] Rossi R, Dalle-Donne I, Milzani A, Giustarini D. Clin Chem. 52(2006)1406.

Ac ce

[16]Giustarini D, Dalle-Donne I, Milzani A, Rossi R. Anal Biochem. 415(2011)81.

[17] Ballatori N, Krance SM, Notenboom S, Shi S, Tieu K, Hammond CL. Biol Chem. 390(2009)191.

[18] Johnson WM, Wilson-Delfosse AL, Mieyal JJ. Nutrients. 4(2012)1399. [19]Marí M, Morales A, Colell A, García-Ruiz C, Kaplowitz N, Fernández-Checa JC. Biochim Biophys Acta. 4165(2012)304.

[20] Camera E, Picardo M. J Chromatogr B Analyt Technol Biomed Life Sci. 781(2002)181.

16

Page 16 of 23

[21] Squellerio I, Caruso D, Porro B, Veglia F, Tremoli E, Cavalca V. J Pharm Biomed Anal. 71(2012)111.

cr

ip t

[22]Norris RL, Eaglesham GK, Shaw GR, Smith MJ, Chiswell RK, Seawright AA, Moore MR. J Chromatogr B Biomed Sci Appl. 762(2001)17.

us

[23] Richie JP Jr, Skowronski L, Abraham P, Leutzinger Y. Clin Chem. 42(1996):64.

an

[24]Túri S, Friedman A, Bereczki C, Papp F, Kovàcs J, Karg E, Németh I. J Hypertens. 21(2003)145.

Ac ce

pt

ed

M

[25]Bramanti E, Vecoli C, Neglia D, Pellegrini MP, Raspi G, Barsacchi R. Clin Chem. 51(2005)1007.

17

Page 17 of 23

Figure and Table Legends

ip t

Figure 1. Representative MRM chromatograms of GSH and GSH-IS (a), and GSSG and GSSGIS (b).

us

cr

Figure 2. Linear responses of GSH (a) and GSSG (b).

an

Figure 3. Whole blood GSH stability in 8 control subjects following sample storage at 4°C (a) and at room temperature (b).

M

Figure 4. Concentrations of GSH (a), GSSG (b) and GSH/GSSG (c) in 59 control subjects plotted as a function of age.

Ac ce

pt

ed

Table 1. (a) Accuracy of GSH-NEM and GSSG measurements in spiked samples of varying concentrations; (b) Intra- and inter-day precision as determined in whole blood aliquots analyzed 15 times in a single day (intra-day) or over 15 consecutive days (inter-day). CV, coefficient of variation

18

Page 18 of 23

99+9 101+8 GSH-NEM 25

50

0.4

0.8

1.6

97+4

95+2

94+5

100

200

400

us

96+7 95+6 96+7 97+6 95+6

3.1

4.1

4.3

an

GSH GSSG 3.3

ip t

0.2

cr

0.1

Ac ce

pt

ed

B. Precision Intra-day CV (%) n=15 Inter-day CV (%) n=15

GSSG

M

A. Accuracy Spiked concentration (uM) recovery (%) n=4; mean+SD Accuracy Spiked concentration (uM) recovery (%) n=5; mean+SD

19

Page 19 of 23

Figure

A.

B. XIC of +MRM(2 pairs): 433.3/304.1 amu fromSample 14 (GSH std) of Data032113 GSH GSSG LOQ.wiff (Turbo Spray)

Max. 2.5e6 cps.

XIC of +MRM(2 pairs): 613.2/355.2 amu fromSample 21 (GSSG std) of Data091310 GSSG normal controls.wiff (Turbo Spray)

Max. 9648.3 cps. 3.75

4.2e6

9500

4.0e6 9000

3.8e6 8500

3.6e6

8000

3.4e6 3.2e6

7500

3.0e6

7000 6500

2.8e6

2.4e6 2.2e6 2.0e6 1.8e6

6000

ip t

Intensity, cps

3.87

In te n s ity , c p s

2.6e6 In te n s ity , c p s

Intensity, cps

5500 5000 4500 4000

1.6e6

3500

1.4e6 3000 2500

1.0e6 2000

8.0e5

1500

6.0e5

1000

4.0e5

500

0.0

us

2.0e5

0

0.5

1.0

1.5

2.0

2.5 Time, min

3.0

3.5

4.0

cr

1.2e6

4.5

0.5

1.0

1.5

2.0

2.5 Time, min

3.0

3.5

4.0

4.5

5.0

Time, min

Ac ce p

te

d

M

an

Time, min

Page 20 of 23

Figure

B. 20.00

450.00

18.00

400.00

16.00

350.00

14.00

250.00 200.00 150.00 100.00

10.00 8.00 6.00 4.00 2.00

50.00 0.00 0.00

12.00

ip t

300.00

100.00

200.00

300.00

400.00

500.00

0.00 0.00

600.00

2.00

cr

Actual value (uM)

500.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

18.00

20.00

Target value (uM)

te

d

M

an

us

Target value (uM)

Ac ce p

Actual vallue (uM)

A.

Page 21 of 23

Figure

Day 1

Day 2

Day 3

Day 4

Day 5

ip t

1400.00 1200.00 1000.00 800.00 600.00 400.00 200.00 0.00

cr

GSH (uM)

A.

us

Days at 4 degrees

an

B. 1400.00

M

1000.00 800.00 600.00

d

400.00 0.00 1

2

te

200.00 3

Ac ce p

GSH (uM)

1200.00

4

5

6

7

24

48

Hours at RT

Page 22 of 23

Figure

2000 1000 0 0

20

40

60

80

100

cr

Age (years)

ip t

GSH (uM)

A.

us

3 2 1 0 0

20

40

an

GSSG (uM)

B.

60

80

M

Age (years)

100

d te

3000 2000 1000

Ac ce p

GSH/GSSG

C.

0

0

20

40

60

80

100

Age (years)

Page 23 of 23