1H NMR Based Targeted Metabolite Profiling for Understanding the

Hindawi Publishing CorporationBioMed Research InternationalVolume 2013 Article ID 329058 9 pageshttpdxdoiorg1011552013329058

Research Article1H NMR Based Targeted Metabolite Profiling forUnderstanding the Complex Relationship ConnectingOxidative Stress with Endometriosis

Saikat K Jana1 Mainak Dutta1 Mamata Joshi2 Sudha Srivastava2

Baidyanath Chakravarty3 and Koel Chaudhury1

1 School of Medical Science and Technology Indian Institute of Technology Kharagpur 721302 India2National Facility for High-Field NMR Tata Institute of Fundamental Research Mumbai 400005 India3 Institute of Reproductive Medicine Kolkata 700106 India

Correspondence should be addressed to Koel Chaudhury koeliitkgpgmailcom

Received 28 April 2013 Accepted 3 July 2013

Academic Editor Nikhat J Siddiqi

Copyright copy 2013 Saikat K Jana et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Accumulating evidence indicates the active role of oxidative stress in the development of endometriosis however the mechanismof reactive oxygen species generation is poorly understood Metabonomicsmetabolomics is a scientific discipline that can beused to study changes in metabolite ensembles associated with disease pathophysiology The present study focuses on the useof proton nuclear magnetic resonance spectroscopy based targeted metabolite profiling approach to explore dysregulation inmetabolites expression in women with endometriosis Further association of oxidative stress with the metabolite ensembles if anyis investigated Using multivariate statistics partial least square discriminant analysis model was generated which could classifyendometriosis patients with sensitivity and specificity of 9283 and 100 respectively and with a classification rate of 964In conjunction with increased glucose metabolism citrate and succinate were found to be elevated in endometriosis patientsHigher levels of reactive oxygen species lipid peroxidation and advanced oxidation protein products and lower levels of totalantioxidant capacity superoxide dismutase catalase and glutathione were also observed Increased glucosemetabolism and defectsin the mitochondrial respiratory system are suggested to be the possible sources of excessive reactive oxygen species generation inendometriosis

1 Introduction

Endometriosis is classically defined as the growth of endome-trial glands and stroma at extrauterine sites most commonlyimplanted over visceral and peritoneal surfaces within thefemale pelvis In India approximately 1 women under-going major gynecological surgery 6ndash43 undergoingsterilization 12ndash32 undergoing laparoscopy for pelvicpain of 21ndash48 undergoing laparoscopy for infertility arediagnosed with endometriosis [1] From a global perspectiveit is a fairly common gynecological disorder affecting almost10 of women of reproductive age [2] The correct approachfor endometriosis management is still unclear The risks andthe diagnostic limitations of laparoscopy and the inaccuracyof clinical examination justify the considerable efforts made

to improve the diagnosis with noninvasive techniques In ourpreviously published paper we have reported a metabolicfingerprint which may be used for non-invasive diagnosis ofthe disease [3] Also the therapeutic approach still focuseson management of clinical symptoms of the disease ratherthan on the disease itself A thorough understanding of thepathophysiology of endometriosis is therefore essential to thedevelopment of novel diagnostic and treatment approachesfor this debilitating condition

Several early studies in 1990s showed no significantcorrelation between oxidative stress (OS) and endometriosis[4ndash6] Murphy and coworkers way back in late 1990s wereone of the first groups to emphasize on the active role ofOS in the pathogenesis and development of endometriosis[7] Since then there has been increasing evidence suggesting

2 BioMed Research International

that OS plays a key role in the pathogenesis and progressionof endometriosis [8 9] Recently a systematic review of 19original articles from 1990 to 2011 on OS biomarkers inpatients with endometriosis has been reported [10] Despitesuch long-term research activity in this field it is still not veryclear as to when and why OS may occur and its relation toendometriosis possibly because of limited data and complexenvironment of the peritoneal cavity

During OS in addition to insufficient antioxidants oxy-gen radicals are also formed at a rate greater than the rateof consumption Changes in the composition of metaboliteensembles are responsible for physiological changes associ-ated withOSThus it is reasonable to conjecture that targetedmetabolite profiling will provide information on quantitativechanges in metabolites of interest based on prior knowledgeof the biological function or metabolic pathway

Metabolomic profiling has emerged as a powerful andreliable tool for the identification of total metabolites presentin the biological system under a given physiological con-dition Metabolites represent the final products of cellsrsquoregulatory processes and act as communicators betweeninformation-rich genome and the functional phenotype ofthe cell Nuclear magnetic resonance (NMR) spectroscopyis the only technique which can identify and quantifycomplex mixtures of metabolites with little or no samplepreparation [11] Furthermore only small sample volumesare required and the analysis is nondestructive [12] Anadded advantage of this technique is that the metaboliteprofile of a biological sample can be acquired rapidly (1ndash15min) with sufficient sensitivity to differentiate even subtlebiological differences Chemometrics and targeted profil-ing are the two distinct approaches that have been estab-lished for processing NMR spectra Chemometrics involvesseparation of NMR spectra into different groups with noassumptions about the identity and quantity of metabolitesin the spectra while targeted profiling involves the iden-tification and quantification of each metabolite in everyNMR spectrum so that metabolite concentrations are thevariables [13] In both the approaches various multivariatestatistical methods such as principal components analysis(PCA) partial least squares discriminant analysis (PLS-DA)are used to search for meaningful differences among thespectra

Herein we have used 1H NMR based targeted metabo-lite profiling followed by multivariate statistical analysis toexplore metabolite imbalances which in turn would providea better understanding of oxidative protection injury andrecovery in endometriosis Further association of OS withthe metabolite ensembles if any is investigated and aprobable mechanism of ROS generation in endometriosis ishypothesized

2 Material and Methods

21 Subject Selection and Sample Collection One hundredand thirty five women (24ndash40 years BMI lt 25) reporting atthe Institute of Reproductive Medicine Salt Lake KolkataIndia for infertility treatment volunteered to participatein this study The study group consisted of 75 women

with endometriosis confirmed by diagnostic laparoscopyand 60 women with tubal factor infertility were consid-ered as controls Tubal factor infertility refers to womenwho had salpingectomy for ectopic pregnancy and proximaltubal obstruction because of low-grade infection or fimbrialocclusion with or without mild peritubal adhesions Tubalinfertility associated with gross hydrosalpingeal changesdense pelvic adhesions because of endometriosis or pelvicinflammatory diseases were excluded As such in this studytubal factor infertility refers to women who had fallopiantube(s) removed for tubal pregnancy Women included inthe study did not receive any kind of medical or hormonaltreatment during the last three months Women with historyof removal of chocolate cysts previous history of any kind ofgynecological surgery including lower pelvic and abdominalsurgery with other possible causes of pain or pelvic pathologyincluding pelvic tuberculosis were excluded The study wasapproved by the Institutional ethics committee and writteninformed consent was taken from all women enrolled in thestudy

Venous blood was drawn in sterile containers from allwomen during the day time in fasting state and in theirearly follicular phase Based on the effects of estrogen itwas hypothesized that the magnitude of the responses wouldbe greater when measured during the early follicular phasecompared to the rest of the menstrual cycle The blood werethen allowed to clot and serum separated by centrifugation at3000 rpm for 5min at 4∘C ROS was measured immediatelyin the serum of all subjects Remaining serum samples werestored at minus20∘C until further analysis for total antioxidantcapacity (TAC) lipid peroxidation (LPO) superoxide dismu-tase (SOD) glutathione (GSH) and catalase levels Further26 samples from endometriosis cases and 24 from controlswere randomly selected from these stored serum samples andused for metabolic profiling using NMR

22 Measurement of Reactive Oxygen Species Free radi-cals generation was determined in serum samples fromendometriosis and controls bymonitoring luminol (5-amino-23-dihydro-14-phthalazinedione)-mediated chemilumines-cence (CL) [14] Luminol reacts with reactive oxygen species(ROS) present in serum resulting in a luminophore that hasan emission peak at sim425 nm The intensity of the CL isproportional to the amount of ROS in the serumThe reactionmixture contained 25120583L of serum and 01mM luminol in10mM sodium phosphate buffer (pH 74) The reaction wasinitiated by the addition of H

2O2at a final concentration

of 1mM CL was measured for 10min using a luminometer(Berthold Sirius Single tube Luminometer Model no 0727)H2O2was added as an activator of luminol Background

(blank) was determined in each experiment utilizing H2O2

luminol and sodium phosphate buffer without samples Thereactionwas performed at 37∘C and expressed in relative lightunits (cps)

23 Measurement of Lipid Peroxidation The slightly modi-fied thiobarbituric acid (TBA) method [15] was used to mea-sure malonaldehyde (MDA) content in the serum sample

BioMed Research International 3

Fifty microliter aliquot of frozen serum was thawed andimmediately used for lipid peroxidation (LPO) estimationHundredmicroliter of stock reagent (12 wv trichloroaceticacid 0375 wv TBA and 025molL HCl warmed todissolve the TBA) was mixed thoroughly with 50 120583L of thesample and heated for 15min in a boiling water bath Aftercooling the flocculent precipitate was removed by centrifu-gation at 1000timesg for 10min and the OD of the supernatantdetermined at 535 nm inmultiplate reader (Victor X3 PerkinElmer USA) against a blank containing all the reagents LPOvalues were expressed as 120583MMDA

24 Measurement of Total Antioxidant Capacity TAC wasmeasured in serum using a modified enhanced CL assay [16]Serumwas thawed at room temperature and diluted 1 10withdeionized water Signal reagent was prepared using a CL kit(GE Healthcare UK) A constant peak of ROS was producedusing HRP conjugated IgG (Santa Cruz Biotechnology SantaCruz CA) HRP was diluted with deionized water to givea desired CL output Luminometer was set in the kineticmode and 900120583L of reaction mixture (100 120583L signal reagent+ 700120583L deionized water + 100 120583L 1 350 diluted HRP) wasadded to the cuvette After 100 seconds of reaction mixtureaddition 50120583L of the diluted serum sample was added tothe mixture and 10 recovery of CL was recorded Troloxa water soluble tocopherol analogue was used as a standardThe antioxidant capacity of serum samples was expressed in120583M Trolox equivalents

25 Determination of the Superoxide Dismutase EnzymaticActivity Serum superoxide dismutase (SOD) activity wasmeasured utilizing the inhibition of auto-oxidation of pyro-gallol by SOD [17] Tris buffer (containing 50mM of Trisbuffer and 1mM of ethylene diamine tetraacetic acid EDTA)was prepared and pH was adjusted to 85 using hydrochloricacid One hundred microliter of 20mM pyrogallol solutionwas added to 29mL of Tris buffer and mixed for controlsmeasurement Reading was taken at 420 nm after 1min 30 sand 3min 30 s using a UV-visible spectrophotometer (PerkinElmer USA) Serum SOD activity was measured by adding01mL of diluted serum sample to 28mL of Tris buffer Thereaction was started by adding 01mL of 20mM pyrogallolsolution Reading was taken at 420 nm exactly after 1min30 s and 3min 30 s and the absorbance recorded per 2minThe percentage of inhibition of pyrogallol autoxidation wascalculated and 1 unit of enzymatic activity was defined as thequantity of enzyme necessary to achieve a 50 inhibition ofautoxidation at 25∘C

26 Determination of the Catalase Enzymatic Activity Theactivity of catalase enzyme was determined according toAebi [18] The capacity of catalase to transform H

2O2was

measured by monitoring the decomposition of H2O2at

240 nm The final reaction volume of 3mL contained 005MTris-buffer 5mM EDTA (pH 70) and 10mM H

2O2(in

01M potassium phosphate buffer pH 70) Fifty microliterof serum sample was added to the above mixture Oneunit of enzymatic activity was considered as the quantity

of enzyme necessary to transform 1 120583mol of H2O2in 1min

at 37∘C Catalase activity was calculated using the molarextinction coefficient of 00436 (mmol Lminus1)minus1 cmminus1 andexpressed in terms of mM H

2O2consumedminmg protein

used

27 Determination of the Glutathione Glutathione (GSH) inthe serum samples was estimated by the method of Moron etal [19] The required amount of serum was mixed with 25of trichloroacetic acid (TCA) and centrifuged at 2000timesg for15min to settle the precipitated proteinsThe supernatant wasaspirated and diluted to 1mL with 02M sodium phosphatebuffer (pH 80) Later 2mL of 06mM 551015840-dithiobis-2-nitrobenzoic acid (DTNB) was added After 10min the ODof the yellow-colored complex formed by the reaction ofGSH and DTNB was measured at 405 nm A standard curvewas obtained with standard GSH The levels of GSH wereexpressed as 120583molmL

28 Determination of Advanced Oxidation Protein Prod-ucts Determination of advanced oxidation protein products(AOPP) was based on spectrophotometric detection accord-ing to Witko-Sarsat et al [20] Two hundred microliter ofserum diluted 1 5 in PBS 200120583L of chloramine-T standardsolution (0 to 100 120583molliter) for calibration and 200120583L ofPBS as blank were placed in each well of a 96-well microliterplate (PerkinElmer USA) Tenmicroliter of 116M potassiumiodide (KI Sigma) was then added followed by 20 120583L ofacetic acid The absorbance of the reaction mixture wasimmediately read at 340 nm in a microplate reader againsta blank The chloramine-T absorbance at 340 nm was linearwithin the range of 0 to 100120583molliter AOPP concentrationswere expressed in 120583molliter of chloramine-T equivalents

29 Nuclear Magnetic Resonance Analysis Prior to nuclearmagnetic resonance (NMR) analysis serum samples werethawed and homogenized using a vortex mixer Two hun-dred microliter of serum was mixed with 400 120583L deu-terium oxide (D

2O) containing 1mM sodium salt of 3-

(trimethylsilyl)propionic-221015840331015840d4 acid (TSP) After cen-trifugation (8000 rpm 5min) 600 120583L of each sample wastransferred to 5mm NMR tubes Proton NMR spectra wererecorded at 298K using a 700MHz Bruker Avance AV IIIspectrometer The resulting spectra were phased and base-line corrected by Bruker TOPSPIN 21 software Individualmetabolites were identified from various sources includingearlier published articles and literature and cross checkedfrom the Human Metabolome Database (HMDB)

210 Statistical Analysis Two types of statistical analysis wereperformed for analysis of the entire data presented in thispaper

2101 Student 119905-Test Data were analyzed using analysis of119905-test as appropriate Data analyses were performed with(GraphPadQuickCalcs 2002ndash2005GraphPad Software IncUSA) Statistical significance was defined as 119875 le 005


2102 Multivariate Statistical Analysis Expression of differ-ent metabolites in endometriosis compared to controls wasanalyzed using multivariate statistical analysis The analysiswas applied to a total of 50 spectra from 26 endometriosis and24 control serum samples After acquisition data matrix ofpeak integral values corresponding to identified compoundswas built Data were preprocessed using normalization andscaling to remove possible bias arising due to sample handlingand sample variability Normalization (by sum) was per-formed in order to minimize possible differences in concen-tration between samples Following normalization scaling(mean-centering and division by the square root of standarddeviation of each variable) was done to give all variablesequal weight regardless of their absolute value After datapreprocessing PCA and PLS-DAwere performed using web-based metabolomic data processing tool MetaboAnalyst 20(Canada) (accessible at httpwwwmetaboanalystca) [21]In MetaboAnalyst statistical computing and visualizationoperations are performed using functions from the R andBioconductor packages [22 23] PCA is used to detectintrinsic clusters and outliers within the data set while PLS-DA maximizes class discrimination Model robustness wasassessed using another web-based ROC curve analysis toolROCCET (accessible at httpwwwroccetca) ROC curvesin ROCCET are generated by Monte Carlo Cross Validation(MCCV) [24] using balanced subsampling In each MCCVtwo-third (23) of the samples were used to evaluate thefeature importanceThe top 100 important features were thenused to build classification models which were validated onone-third of the samples that were left out The procedureswere repeated multiple times to calculate the performanceof the PLS-DA model Sensitivity and specificity of themodel were calculated from the confusion matrix generatedfrom the multiple MCCV iterations Further validation wasperformed with MetaboAnalyst 20 using permutation testsconsisting of 1000 permutations

2103 Metabolic Pathway Analysis Detailed analysis of themost relevant metabolic pathways and networks in womenwith endometriosis was performed by MetaboAnalyst thatcombines results frompowerful pathway enrichment analysisinvolved in the conditions under study MetaboAnalyst useshigh-quality KEGG metabolic pathways as the back endknowledge base It integrates many well-established methods(ie univariate analysis over-representation analysis) andnovel algorithms and concepts (ie Global Test Global-Ancova network topology analysis) with pathway analysis[25]

3 Results

OS markers in serum of women with endometriosis andcontrols are summarized in Table 1 ROS LPO and AOPPwere observed to be increased significantly whereas TACSOD catalase and GSH levels were significantly less inendometriosis women as compared to controls Severalamino acids organic acids and other molecules wereidentified in serum using 1H NMR metabolic profilingA representative metabolic fingerprint of serum from

women with endometriosis is shown in SupplementaryFigure 1 in Supplementary Materials available online athttpdxdoiorg1011552013329058 Exploratory PCA ofidentified metabolites was employed to detect intrinsic clus-tering and possible outliers Figure 1(a) depicting PC1 versusPC2 scores scatter plot shows a trend for unsupervised sepa-ration between endometriosis and controls PLS-DA furthermaximized the group separation (Figure 1(b)) Leave one outcross-validation (LOOCV) was employed from which 1198762and 1198772 values (representing the predictive capability and theexplained variance resp) were extracted The model with 1198772close to 083 and 1198762 well above 082 showed a very goodpredictive ability To better assess the predictive ability ofthe PLS-DA classification model MCCV was applied as dis-cussed in Section 2 Sensitivity (percentage of endometriosissamples correctly classified as true positives) and specificity(percentage of control samples correctly classified as truenegatives) of the PLS-DA model were found to be 100 and9167 respectively with a classification rate (total numberof samples correctly classified) of 9583 Area under theROC curve (Figure 1(c)) was found to be 099 denoting highpredictive accuracy of the model Further model validationrelied on permutation analysis The analysis was performedonly for the best model If the performance score of theoriginal data lies outside the distribution then the resultis significant Using 1000 permutation tests the 119875 value isreported as 119875 le 0001 denoting that none of the resultsare better than the original one (Figure 1(d)) The furtheraway from the plot origin a variable lies the stronger impactthe variable has on the model From PLS-DA loading plotmetabolites with higher loadings were identified (Supple-mentary Figure 2) On comparing the loading and scores plotit becomes evident that lactate 2-hydroxybutyrate succinatelysine glycerophosphocholine citric acid pyruvate adipicacid and lipids are overexpressed in endometriosis whereasisoleucine arginine asparagine glucose creatine alanineleucine and fatty acid have higher expression in controlsSignals with high VIP values (Supplementary Figure 3) areconsidered to be significant and therefore validated using 119905-test (Table 2) With pattern recognition analysis of profilesof metabolites a clear separation between endometriosisand controls was achieved As a consequence of dysregu-lation of specific metabolites metabolic pathway analysiswas performed using MetaboAnalyst The affected pathways(Impact-value ge 010) in women with endometriosis arerepresented in Supplementary Figure 4

4 Discussion

The present study is an attempt to measure levels of variousOS parameters in serum of endometriosis patients andidentify differently expressed metabolites using 1H-NMRbased targeted profiling to have an improved understandingof the disease pathophysiology

Several studies have reported ROS concentration to behigher in women with endometriosis [6 26] which is inagreement with our findings Endometriosis has also beenassociated with significantly higher levels of lipid peroxide-modified rabbit serum albumin malondialdehyde-modified


Table 1 Levels of oxidative stress parameters in endometriosis and controls

Parameters Endometriosis (119899 = 60) Controls (119899 = 60) 119875 valueROS (cps) 1295 plusmn 19 7384 plusmn 17 119875 le 0001

LPO (120583MMDA) 1366 plusmn 004372 1014 plusmn 003146 119875 le 0001

TAC (120583M Trolox equivalent) 7336 plusmn 4989 9363 plusmn 4141 119875 le 0001

SOD (unitsmL) 2319 plusmn 043 3923 plusmn 048 119875 le 0001

Catalase (nmolminmg) 3063 plusmn 07389 4909 plusmn 1148 119875 le 0001

GSH (120583molmL) 306 plusmn 006898 701 plusmn 008175 119875 le 0001

AOPP (120583molL) 1551 plusmn 4250 9060 plusmn 3445 119875 le 0001

Mean plusmn SEM

0

10

20

30

0 10 20 30

PC2

(18

2)

PC1 (290) minus30 minus20 minus10

minus40

minus30

minus20

minus10

(a)

0

5

10

15

0 10 20 30

Com

pone

nt 2

(84

)

Component 1 (289) minus30 minus20 minus10

minus15

minus10

minus5

(b)

0 02 04 06 08 10

0

02

04

06

08

10

Sens

itivi

ty (t

rue p

ositi

ve ra

te)

Area under the curve (AUC) = 0998

95 CI 0972-1

1minus specificity (false positive rate)

(c)

0

50

100

150

Freq

uenc

y

06 07 08 09

Permutation test statistics

ObservedstatisticP lt 0001

(d)

Figure 1 Scores scatter plot of (a) PC1 versus PC2 and (b) component 1 versus component 2 resulting from PCA and PLS-DA respectivelyapplied to 1H NMR spectra of serum of endometriosis patients (black diamond) and controls (white diamond) Validation of the PLS-DAmodel was performed by (c) receiver operating characteristic (ROC) analysis where area under the curve (AUC) was found to be 099 (d)Permutation test statistics at 1000 permutations with observed statistic at 119875 lt 0001

low-density lipoprotein and oxidized low-density lipopro-tein as measured in serum and compared to tubal ligationcases [27] Furthermore lipid peroxide concentrations havebeen reported to be highest in women with endometriosisindicating the involvement of ROS in the development of

the disease [9] These observations support the results ofthe present study where significant increase in LPO levels inendometriosis women is observed

Catalase SOD and GSH are utilized to keep ROS incheck It is thus imperative that lower levels of these enzymes


Acetyl-CoA

Iso-citrate

Fumarate

Malate

Oxaloacetate

I

II

III

Signaling pathway

CoQ

TCA cycle

SODCatalase

InflammationAngiogenesis

Matrix DegradationGrowth modular

Succinyl-CoA

AO

ROSNADH

FAD

Glycolysis

O2

O2

O2∙minus

O2∙minus

NAD+

FADH2

120572-ketoglutarate

Glucose darr

Pyruvate uarr

Citrate uarr

Succinate uarr

darr

Figure 2 Figure depicting the role of metabolites in generation of free radicals Lower levels of glucose along with elevated levels of pyruvatecitrate and succinate indicate enhanced glucose metabolism along with impaired mitochondrial respiration I II and III represent complexI II and III respectively of electron transport chain Complex I and III are the primary sites of free radical generation Elevated levels ofsuccinate indicate complex III to be the most probable site for ROS generation in endometriosisThese free radicals are further removed fromthe system by various ROS scavengers such as SOD and catalase An imbalance in the rate of formation and removal of free radicals resultsin oxidative stress In endometriosis lower levels of these enzymes lead to continued prevalence of oxidative stress

Table 2 Main serummetabolites contributing towards discrimina-tion between endometriosis and controls

Metabolites 120575 1H (ppm) Fold changes(relative to controls)

Glucose 326 026Alanine 15 032Creatine 306 0132-hydroxybutyrate 396 155Lactate 411 192L-lysine 304 172Pyruvate 247 158Succinic acid 239 148Adipic acid 154 161L-leucine 099 051Citric acid 254 167L-iso leucine 103 082Glycerophosphocholine 366 185L-asparagine 284 041L-arginine 164 094

would indicate increased ROS production which is evi-denced by the findings of the present study (Table 1) Asignificant decrease in TAC level in endometriosis womenwas also observed in the present study (Table 1) This is inaccordance with the findings of Szczepanska and coworkerswho have documented that mean activity of SOD and total

antioxidant status are lowest among infertile patients withendometriosis [9] Ota et al reported a higher expressionof SOD in endometrial tissues of endometriotic women ascompared to controls by immunohistochemistry which is asemiquantitative method [28] On the contrary Szczepanskaet al observed reduced SOD level in peritoneal fluid inendometriosis as compared to controls using quantitativespectrophotometric assay [9] This is in accordance with ourpresent study where reduced SOD level were observed inserum of endometriotic women From the present studyOS condition is indicated in endometriosis as evidenced bylower expression of TAC SOD GSH catalase and higherROS generation

Protein cross-linking products formed due to oxidationof amino acids by ROS in the plasma are designated as AOPP[29] AOPP are formedduringOSby the action of chlorinatedoxidants mainly hypochlorous acid and chloramines [30]AOPP are predominantly aggregates of albumin damaged byOS [29] Increased levels of AOPP in endometriosis observedin the present study are in agreement with increased OSreported in these women

Multivariate analysis of the identified NMR spectraclearly distinguishes endometriosis cases from the controlsOn plotting PC1 (290) times PC2 (182) the plots forendometriosis are observed to lie in a principal componentplane which is significantly different from the plane whereplots for controls are clustered (Figure 1(a)) As evidencedin Figure 1(b) PLS-DA shows statistically significant sepa-ration between endometriosis and control cases Our valuescorresponding to 1198772 and 1198762 close to 08 indicate that the


model is valid and can predict better than chance Sensitivityspecificity and classification rates of 100 9167 and9583 respectively further validate the model robustnessAlso permutation test statistics result (119875 lt 0001) validatesthe predictive ability of the real model Metabolites with highVIP scores including lactate L-alanine glycerophosphol-choline glucose L-leucine L-lysine creatine L-argininesuccinic acid adipic acid lipid pyruvate 2-hydroxybutyrateL-isoleucine L-asparagine and citric acid were considered assignificant

The tripeptide glutathione is reported to be critical inthe detoxification of ROS [31] Glutathione is considered asone of the most abundant antioxidants present in cells [3233] Reduced glutathione donates electron to ROS and getsoxidized thus helping in scavenging unstable ROS IncreasedGSSGGSH ratio is indicative of OS [34]This is in agreementwith our results which indicate a significant decrease inGSH Ophthalmate has been indicated as a biomarker ofOS as insufficient levels of GSH results in ophthalmatesynthesis [35] 2-hydroxybutyrate is released as a by-productduring ophthalmate synthesis Metabolite profiling of serumrevealed increased levels of 2-hydroxybutyrate which furthersupport the prevalence of OS in endometriosis Unlike ROSdetoxifying agents like SODand catalase that only decomposeROS without having an impact on their production theTCA cycle can regulate both their formation and decom-position [35] Glucose levels were found to be significantlyless in endometriosis as compared to controls Increasedmetabolism of glucose in conjunction with up-regulation ofpyruvate suggests enhanced glycolysis in these women It iswell known that all aerobic organisms rely predominantly onthe TCA cycle to generate NADH and FADH2 from acetylCoA for the synthesis of ATP during oxidative phosphory-lation Other than the production of ATP mitochondria arealso involved in the cellrsquos response to OS Several steps in thepath of oxygen reduction in mitochondria have the potentialto produce highly reactive free radicals that can damage cells01ndash4 of the O

2used by actively respiring mitochondria

forms ∙O2

minus which can have lethal effects on a cell unlessthe free radical is quickly disposed of In the present studyin conjunction with increased glucose metabolism citrateand succinate two TCA cycle intermediate products werealso found to be elevated in endometriosis patients Thesedata support the notion that a possible defect is associatedwith the mitochondrial respiratory system with an increasedprobability of ROS generation from the ETC A probablehypothesis summarizing the possible route of free radicalgeneration in endometriosis patients is depicted in Figure 2While in normal cells various antioxidants dispose of ROSfrom the system insufficient levels of SOD and catalase inendometriosis are indicative of inefficient scavenging effectleading to a continued prevalence of ROS in the system

Leucine and isoleucine are essential amino acids Leucinecatabolism terminates at acetyl CoA whereas isoleucine givesrise to acetyl CoA and propionyl CoA Moreover acetyl CoAprovides the carbon atoms of its acetyl group to the TCAcycle for energy production Decreased serum levels of bothleucine and isoleucine indicate enhanced catabolism of theseamino acids leading to elevated TCA cycle intermediates

Adipic acid is a byproduct of LPO which is nonenzymati-cally mediated by ROS NMR metabonomic data indicatinga significant decrease in lipid and increase in adipic acidlevels supported by the higher levels of LPO estimatedbiochemically reconfirms the involvement of OS L-lysine isan essential amino acid which is known to reduce OS [35]Since lysine uses the same intracellular transporter system asarginine hence lysine competeswith arginine for its transportwithin the cells Elevated lysine and lower arginine levelsin serum of women in endometriosis may be attributed toincreased uptake of arginine by the cells

Glycerophosphocholine (GPC) plays a pivotal role as anosmotic pressure regulator and metabolic antitoxin It is amajor reservoir for cell membrane omega-3 phospholipidsthe major building blocks for cell membranes It helps insupporting membrane fluidity enabling membrane proteinsto perform with better efficiency Under OS condition GPCis reported to get oxidized [36] and reduced GPC levelsin serum may lead to reduced cell membrane fluidity andefficiency

Asparaginase is an enzyme which converts asparagineinto ammonia and aspartate Aspartate transaminates tooxaloacetate which follows the gluconeogenic pathway toform glucose Lower serum levels of gluconeogenic aminoacids including asparagine and alanine in endometriosis maybe due to its increased utilization as a major gluconeogenicprecursor to meet the high glucose uptake and demand byfast growing endometrial cells outside the uterus

5 Conclusion

This is the first attempt to identify the potential causesof ROS generation in women with endometriosis NMRbased targeted metabolic profiling method was used for thispurpose The preliminary findings suggest increased glucosemetabolism and defects in the mitochondrial respiratorysystem to be possible sources of excessive ROS generation inthese women It may be argued that these preliminary resultsare not reliable enough to functionally associate OS withmetabolite pathways Follow-up laboratory efforts leading toexperimental validation targeted specifically to the predictedpathway are therefore necessary

6 Limitation

Multivariate statistical analysis is useful in providing asso-ciation between multiple risk factors and disease conditionIt was therefore used to identify the potential causes ofROS generation in endometriosis cases However thesepreliminary results require further validation to functionallyassociate OS with the proposed metabolite pathway

Abbreviations

OS Oxidative stressROS Reactive oxygen speciesLPO Lipid peroxidationOD Optical densityTBA 3-(methyl thio)butyric acid


MDA MalondialdehydeTAC Total antioxidant capacityHRP Horseradish peroxidaseSOD Superoxide dismutaseGPC GlycerophosphocholineGSH GlutathioneAOPP Advanced oxidation protein productsNMR Nuclear magnetic resonanceD2O Heavy water

Conflict of Interests

Authors have no conflict of interests

Authorsrsquo Contribution

Saikat K Jana and Koel Chaudhury conceived and designedthe experiments Saikat K Jana Mainak Dutta and MamataJoshi performed the experiments Saikat K Jana and MainakDutta analyzed the data Koel Chaudhury and Sudha Sri-vastava contributed in providing reagentsmaterialsanalysistools Baidyanath Chakravarty and Saikat K Jana contributedin patient selection and sample collection Saikat K JanaMainak Dutta and Koel Chaudhury drafted the paper Theauthors Saikat K Jana and Mainak Dutta have contributedequally to the work and may be considered as joint firstauthors

Acknowledgments

The authors would like to acknowledge Department of Sci-ence and Technology Government of India India (researchproject SRSOHS-00222008) for funding The authors aregrateful to all the volunteers who participated in this studyas well as to Shanawaz Yasmin from IRM Kolkata andIIT Kharagpur respectively for helping in the collectionof samples The facilities provided by the National Facilityfor High Field NMR located at TIFR and supported bythe Department of Science and Technology (DST) are alsogratefully acknowledged

References

[1] S Mittal ldquoEndometriosis-clinical perspectivesrdquo in Proceedingsof the INDOmdashUS Workshop on Microarray in EndometriosisClinics to Research Laboratory All India Institute of MedicalSciences New Delhi India 2009

[2] L C Giudice and L C Kao ldquoEndometriosisrdquo The Lancet vol364 no 9447 pp 1789ndash1799 2004

[3] M Dutta M Joshi S Srivastava I Lodh B Chakravartyand K Chaudhury ldquoA metabonomics approach as a meansfor identification of potential biomarkers for early diagnosisof endometriosisrdquo Molecular BioSystems vol 8 pp 3281ndash32872012

[4] K Arumugam and Y C Y Dip ldquoEndometriosis and infertilitythe role of exogenous lipid peroxides in the peritoneal fluidrdquoFertility and Sterility vol 63 no 1 pp 198ndash199 1995

[5] H-N Ho M-Y Wu S-U Chen K-H Chao C-D Chen andY-S Yang ldquoTotal antioxidant status and nitric oxide do not

increase in peritoneal fluids from women with endometriosisrdquoHuman Reproduction vol 12 no 12 pp 2810ndash2815 1997

[6] YWang J Goldberg R K Sharma A Agarwal and T FalconeldquoImportance of reactive oxygen species in the peritoneal fluidof women with endometriosis or idiopathic infertilityrdquo Fertilityand Sterility vol 68 no 5 pp 826ndash830 1997

[7] A A Murphy N Santanam and S Parthasarathy ldquoEndomet-riosis a disease of oxidative stressrdquo Seminars in ReproductiveEndocrinology vol 16 no 4 pp 263ndash273 1998

[8] A Agarwal A Aponte-Mellado B J Premkumar A Shamanand S Gupta ldquoThe effects of oxidative stress on female repro-duction a reviewrdquo Reproductive Biology and Endocrinology vol10 article 49 2012

[9] M Szczepanska J Kozlik J Skrzypczak and M MikołajczykldquoOxidative stress may be a piece in the endometriosis puzzlerdquoFertility and Sterility vol 79 pp 1288ndash1293 2003

[10] L F Carvalho A N Samadder A Agarwal L F Fernandesand M S Abrao ldquoOxidative stress biomarkers in patients withendometriosis systematic reviewrdquo Archives of Gynecology andObstetrics vol 286 pp 1033ndash1040 2012

[11] O Beckonert H C Keun T M D Ebbels et al ldquoMetabolicprofiling metabolomic andmetabonomic procedures for NMRspectroscopy of urine plasma serum and tissue extractsrdquoNature protocols vol 2 no 11 pp 2692ndash2703 2007

[12] W S Price ldquoApplications of pulsed gradient spin-echo NMRdiffusion measurements to solution dynamics and organiza-tionrdquo in Diffusion Fundamentals J Karger F Grinberg and PHeitjans Eds pp 490ndash508 Leipzig University Press LeipzigGermany 2005

[13] J Liu L Litt M R Segal M J S Kelly J G Pelton andM Kim ldquoMetabolomics of oxidative stress in recent studiesof endogenous and exogenously administered intermediatemetabolitesrdquo International Journal of Molecular Sciences vol 12no 10 pp 6469ndash6501 2011

[14] K Chaudhury K N Babu A K Singh S Das A Kumar andS Seal ldquoMitigation of endometriosis using regenerative ceriumoxide nanoparticlesrdquo Nanomedicine vol 9 pp 439ndash448 2012

[15] J A Buege and S D Aust ldquoMicrosomal lipid peroxidationrdquoMethods in Enzymology vol 52 pp 302ndash310 1978

[16] I L C Chapple ldquoReactive oxygen species and antioxidants ininflammatory diseasesrdquo Journal of Clinical Periodontology vol24 no 5 pp 287ndash296 1997

[17] S Marklund and G Marklund ldquoInvolvement of the superoxideanion radical in the autoxidation of pyrogallol and a convenientassay for superoxide dismutaserdquo European Journal of Biochem-istry vol 47 no 3 pp 469ndash474 1974

[18] H Aebi ldquoCatalase in vitrordquoMethods in Enzymology vol 105 pp121ndash126 1984

[19] M S Moron J W Depierre and B Mannervik ldquoLevels of glu-tathione glutathione reductase and glutathione S-transferaseactivities in rat lung and liverrdquo Biochimica et Biophysica Actavol 582 no 1 pp 67ndash78 1979

[20] V Witko-Sarsat M Friedlander C Capeillere-Blandin et alldquoAdvanced oxidation protein products as a novel marker ofoxidative stress in uremiardquo Kidney International vol 49 pp1304ndash1313 1996

[21] J Xia R Mandal I V Sinelnikov D Broadhurst and D SWishart ldquoMetaboAnalyst 2 0mdasha comprehensive server formetabolomic data analysisrdquo Nucleic Acids Research vol 40 pp127ndash133 2012


[22] J Xia and D S Wishart ldquoChapter 14 metabolomic dataprocessing analysis and interpretation using MetaboAnalystrdquoin Current Protocols in Bioinformatics 2011

[23] J Xia N Psychogios N Young and D S Wishart ldquoMetabo-Analyst a web server for metabolomic data analysis andinterpretationrdquoNucleic Acids Research vol 37 no 2 pp W652ndashW660 2009

[24] J Carrola CM Rocha A S Barros et al ldquoMetabolic signaturesof lung cancer in biofluids NMR-based metabonomics ofurinerdquo Journal of Proteome Research vol 10 no 1 pp 221ndash2302011

[25] J Xia and D S Wishart ldquoWeb-based inference of biologicalpatterns functions and pathways frommetabolomic data usingMetaboAnalystrdquo Nature Protocols vol 6 no 6 pp 743ndash7602011

[26] M A Bedaiwy T Falcone R K Sharma et al ldquoPredictionof endometriosis with serum and peritoneal fluid markers aprospective controlled trialrdquo Human Reproduction vol 17 no2 pp 426ndash431 2002

[27] A Shanti N Santanam A J Morales S Parthasarathy and AA Murphy ldquoAutoantibodies to markers of oxidative stress areelevated in women with endometriosisrdquo Fertility and Sterilityvol 71 no 6 pp 1115ndash1118 1999

[28] H Ota S Igarashi J Hatazawa and T Tanaka ldquoImmunohisto-chemical assessment of superoxide dismutase expression in theendometrium in endometriosis and adenomyosisrdquo Fertility andSterility vol 72 no 1 pp 129ndash134 1999

[29] F Yuan S-X Liu and J-W Tian ldquoAdvanced oxidation pro-tein products induce reactive oxygen species production inendothelial cellsrdquo Di Yi Jun Yi Da Xue Xue Bao vol 24 no 12pp 1350ndash1352 2004

[30] V Witko-Sarsat V Gausson A-T Nguyen et al ldquoAOPP-induced activation of human neutrophil and monocyte oxida-tive metabolism a potential target for N-acetylcysteine treat-ment in dialysis patientsrdquo Kidney International vol 64 no 1pp 82ndash91 2003

[31] DM Townsend K D Tew andH Tapiero ldquoThe importance ofglutathione in human diseaserdquo Biomedicine and Pharmacother-apy vol 57 no 3 pp 145ndash155 2003

[32] A Meister and M E Anderson ldquoGlutathionerdquo Annual Reviewof Biochemistry vol 52 pp 711ndash760 1983

[33] C Kerksick and D Willoughby ldquoThe antioxidant role ofglutathione and N-acetyl-cysteine supplements and exercise-induced oxidative stressrdquo Journal of the International Society ofSports Nutrition vol 2 pp 38ndash44 2005

[34] J B Owen and D A Butterfield ldquoMeasurement of oxi-dizedreduced glutathione ratiordquoMethods in Molecular Biologyvol 648 pp 269ndash277 2010

[35] T Soga R Baran M Suematsu et al ldquoDifferential metabolo-mics reveals ophthalmic acid as an oxidative stress biomarkerindicating hepatic glutathione consumptionrdquo Journal of Biolog-ical Chemistry vol 281 no 24 pp 16768ndash16776 2006

[36] I Milic M Fedorova K Teuber J Schiller and R HoffmannldquoCharacterization of oxidation products from 1-palmitoyl-2-linoleoyl-sn- glycerophosphatidylcholine in aqueous solutionsand their reactions with cysteine histidine and lysine residuesrdquoChemistry and Physics of Lipids vol 165 no 2 pp 186ndash196 2012

Submit your manuscripts athttpwwwhindawicom

PainResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom

Volume 2014

ToxinsJournal of

VaccinesJournal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

AntibioticsInternational Journal of

ToxicologyJournal of


StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Drug DeliveryJournal of



Advances in Pharmacological Sciences

Tropical MedicineJournal of


Medicinal ChemistryInternational Journal of


AddictionJournal of



BioMed Research International

Emergency Medicine InternationalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Autoimmune Diseases


Anesthesiology Research and Practice

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of


Pharmaceutics


MEDIATORSINFLAMMATION

of


that OS plays a key role in the pathogenesis and progressionof endometriosis [8 9] Recently a systematic review of 19original articles from 1990 to 2011 on OS biomarkers inpatients with endometriosis has been reported [10] Despitesuch long-term research activity in this field it is still not veryclear as to when and why OS may occur and its relation toendometriosis possibly because of limited data and complexenvironment of the peritoneal cavity

During OS in addition to insufficient antioxidants oxy-gen radicals are also formed at a rate greater than the rateof consumption Changes in the composition of metaboliteensembles are responsible for physiological changes associ-ated withOSThus it is reasonable to conjecture that targetedmetabolite profiling will provide information on quantitativechanges in metabolites of interest based on prior knowledgeof the biological function or metabolic pathway

Metabolomic profiling has emerged as a powerful andreliable tool for the identification of total metabolites presentin the biological system under a given physiological con-dition Metabolites represent the final products of cellsrsquoregulatory processes and act as communicators betweeninformation-rich genome and the functional phenotype ofthe cell Nuclear magnetic resonance (NMR) spectroscopyis the only technique which can identify and quantifycomplex mixtures of metabolites with little or no samplepreparation [11] Furthermore only small sample volumesare required and the analysis is nondestructive [12] Anadded advantage of this technique is that the metaboliteprofile of a biological sample can be acquired rapidly (1ndash15min) with sufficient sensitivity to differentiate even subtlebiological differences Chemometrics and targeted profil-ing are the two distinct approaches that have been estab-lished for processing NMR spectra Chemometrics involvesseparation of NMR spectra into different groups with noassumptions about the identity and quantity of metabolitesin the spectra while targeted profiling involves the iden-tification and quantification of each metabolite in everyNMR spectrum so that metabolite concentrations are thevariables [13] In both the approaches various multivariatestatistical methods such as principal components analysis(PCA) partial least squares discriminant analysis (PLS-DA)are used to search for meaningful differences among thespectra

Herein we have used 1H NMR based targeted metabo-lite profiling followed by multivariate statistical analysis toexplore metabolite imbalances which in turn would providea better understanding of oxidative protection injury andrecovery in endometriosis Further association of OS withthe metabolite ensembles if any is investigated and aprobable mechanism of ROS generation in endometriosis ishypothesized

2 Material and Methods

21 Subject Selection and Sample Collection One hundredand thirty five women (24ndash40 years BMI lt 25) reporting atthe Institute of Reproductive Medicine Salt Lake KolkataIndia for infertility treatment volunteered to participatein this study The study group consisted of 75 women

with endometriosis confirmed by diagnostic laparoscopyand 60 women with tubal factor infertility were consid-ered as controls Tubal factor infertility refers to womenwho had salpingectomy for ectopic pregnancy and proximaltubal obstruction because of low-grade infection or fimbrialocclusion with or without mild peritubal adhesions Tubalinfertility associated with gross hydrosalpingeal changesdense pelvic adhesions because of endometriosis or pelvicinflammatory diseases were excluded As such in this studytubal factor infertility refers to women who had fallopiantube(s) removed for tubal pregnancy Women included inthe study did not receive any kind of medical or hormonaltreatment during the last three months Women with historyof removal of chocolate cysts previous history of any kind ofgynecological surgery including lower pelvic and abdominalsurgery with other possible causes of pain or pelvic pathologyincluding pelvic tuberculosis were excluded The study wasapproved by the Institutional ethics committee and writteninformed consent was taken from all women enrolled in thestudy

Venous blood was drawn in sterile containers from allwomen during the day time in fasting state and in theirearly follicular phase Based on the effects of estrogen itwas hypothesized that the magnitude of the responses wouldbe greater when measured during the early follicular phasecompared to the rest of the menstrual cycle The blood werethen allowed to clot and serum separated by centrifugation at3000 rpm for 5min at 4∘C ROS was measured immediatelyin the serum of all subjects Remaining serum samples werestored at minus20∘C until further analysis for total antioxidantcapacity (TAC) lipid peroxidation (LPO) superoxide dismu-tase (SOD) glutathione (GSH) and catalase levels Further26 samples from endometriosis cases and 24 from controlswere randomly selected from these stored serum samples andused for metabolic profiling using NMR

22 Measurement of Reactive Oxygen Species Free radi-cals generation was determined in serum samples fromendometriosis and controls bymonitoring luminol (5-amino-23-dihydro-14-phthalazinedione)-mediated chemilumines-cence (CL) [14] Luminol reacts with reactive oxygen species(ROS) present in serum resulting in a luminophore that hasan emission peak at sim425 nm The intensity of the CL isproportional to the amount of ROS in the serumThe reactionmixture contained 25120583L of serum and 01mM luminol in10mM sodium phosphate buffer (pH 74) The reaction wasinitiated by the addition of H

2O2at a final concentration

of 1mM CL was measured for 10min using a luminometer(Berthold Sirius Single tube Luminometer Model no 0727)H2O2was added as an activator of luminol Background

(blank) was determined in each experiment utilizing H2O2

luminol and sodium phosphate buffer without samples Thereactionwas performed at 37∘C and expressed in relative lightunits (cps)

23 Measurement of Lipid Peroxidation The slightly modi-fied thiobarbituric acid (TBA) method [15] was used to mea-sure malonaldehyde (MDA) content in the serum sample






2O2was



2O2(in





used











3 Results



4 Discussion












Mean plusmn SEM

0

10

20

30

0 10 20 30

PC2

(18

2)


minus40

minus30

minus20

minus10

(a)

0

5

10

15

0 10 20 30

Com

pone

nt 2

(84

)


minus15

minus10

minus5

(b)

0 02 04 06 08 10

0

02

04

06

08

10

Sens

itivi

ty (t

rue p

ositi

ve ra

te)


95 CI 0972-1


(c)

0

50

100

150

Freq

uenc

y

06 07 08 09



(d)






Acetyl-CoA

Iso-citrate

Fumarate

Malate

Oxaloacetate

I

II

III

Signaling pathway

CoQ

TCA cycle

SODCatalase



Succinyl-CoA

AO

ROSNADH

FAD

Glycolysis

O2

O2

O2∙minus

O2∙minus

NAD+

FADH2


Glucose darr

Pyruvate uarr

Citrate uarr

Succinate uarr

darr













forms ∙O2






5 Conclusion


6 Limitation


Abbreviations








Acknowledgments


References











































Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of






2O2was



2O2(in





used











3 Results



4 Discussion












Mean plusmn SEM

0

10

20

30

0 10 20 30

PC2

(18

2)


minus40

minus30

minus20

minus10

(a)

0

5

10

15

0 10 20 30

Com

pone

nt 2

(84

)


minus15

minus10

minus5

(b)

0 02 04 06 08 10

0

02

04

06

08

10

Sens

itivi

ty (t

rue p

ositi

ve ra

te)


95 CI 0972-1


(c)

0

50

100

150

Freq

uenc

y

06 07 08 09



(d)






Acetyl-CoA

Iso-citrate

Fumarate

Malate

Oxaloacetate

I

II

III

Signaling pathway

CoQ

TCA cycle

SODCatalase



Succinyl-CoA

AO

ROSNADH

FAD

Glycolysis

O2

O2

O2∙minus

O2∙minus

NAD+

FADH2


Glucose darr

Pyruvate uarr

Citrate uarr

Succinate uarr

darr













forms ∙O2






5 Conclusion


6 Limitation


Abbreviations








Acknowledgments


References











































Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of




3 Results



4 Discussion












Mean plusmn SEM

0

10

20

30

0 10 20 30

PC2

(18

2)


minus40

minus30

minus20

minus10

(a)

0

5

10

15

0 10 20 30

Com

pone

nt 2

(84

)


minus15

minus10

minus5

(b)

0 02 04 06 08 10

0

02

04

06

08

10

Sens

itivi

ty (t

rue p

ositi

ve ra

te)


95 CI 0972-1


(c)

0

50

100

150

Freq

uenc

y

06 07 08 09



(d)






Acetyl-CoA

Iso-citrate

Fumarate

Malate

Oxaloacetate

I

II

III

Signaling pathway

CoQ

TCA cycle

SODCatalase



Succinyl-CoA

AO

ROSNADH

FAD

Glycolysis

O2

O2

O2∙minus

O2∙minus

NAD+

FADH2


Glucose darr

Pyruvate uarr

Citrate uarr

Succinate uarr

darr













forms ∙O2






5 Conclusion


6 Limitation


Abbreviations








Acknowledgments


References











































Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of










Mean plusmn SEM

0

10

20

30

0 10 20 30

PC2

(18

2)


minus40

minus30

minus20

minus10

(a)

0

5

10

15

0 10 20 30

Com

pone

nt 2

(84

)


minus15

minus10

minus5

(b)

0 02 04 06 08 10

0

02

04

06

08

10

Sens

itivi

ty (t

rue p

ositi

ve ra

te)


95 CI 0972-1


(c)

0

50

100

150

Freq

uenc

y

06 07 08 09



(d)






Acetyl-CoA

Iso-citrate

Fumarate

Malate

Oxaloacetate

I

II

III

Signaling pathway

CoQ

TCA cycle

SODCatalase



Succinyl-CoA

AO

ROSNADH

FAD

Glycolysis

O2

O2

O2∙minus

O2∙minus

NAD+

FADH2


Glucose darr

Pyruvate uarr

Citrate uarr

Succinate uarr

darr













forms ∙O2






5 Conclusion


6 Limitation


Abbreviations








Acknowledgments


References











































Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of


Acetyl-CoA

Iso-citrate

Fumarate

Malate

Oxaloacetate

I

II

III

Signaling pathway

CoQ

TCA cycle

SODCatalase



Succinyl-CoA

AO

ROSNADH

FAD

Glycolysis

O2

O2

O2∙minus

O2∙minus

NAD+

FADH2


Glucose darr

Pyruvate uarr

Citrate uarr

Succinate uarr

darr













forms ∙O2






5 Conclusion


6 Limitation


Abbreviations








Acknowledgments


References











































Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of





forms ∙O2






5 Conclusion


6 Limitation


Abbreviations








Acknowledgments


References











































Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of







Acknowledgments


References











































Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of





















Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of





Volume 2014

ToxinsJournal of

VaccinesJournal of















AddictionJournal of






Autoimmune Diseases




Journal of


Pharmaceutics



of

Documents

1H NMR Based Targeted Metabolite Profiling for Understanding the