Journal of Proteome Research. Jan/1/2015; 14(1): 557-566

Published online Oct/30/2014

PMID: 25361234

PMC: 4286171

doi: 10.1021/pr500782g

LC–MS Metabolomicsof Psoriasis Patients RevealsDisease Severity-Dependent Increases in Circulating Amino Acids ThatAre Ameliorated by Anti-TNFα Treatment

Abstract

Psoriasis is an immune-mediated highlyheterogeneous skin diseasein which genetic as well as environmental factors play important roles.In spite of the local manifestations of the disease, psoriasis mayprogress to affect organs deeper than the skin. These effects aredocumented by epidemiological studies, but they are not yet mechanisticallyunderstood. In order to provide insight into the systemic effectsof psoriasis, we performed a nontargeted high-resolution LC–MSmetabolomics analysis to measure plasma metabolites from individualswith mild or severe psoriasis as well as healthy controls. Additionally,the effects of the anti-TNFα drug Etanercept on metabolic profileswere investigated in patients with severe psoriasis. Our analysesidentified significant psoriasis-associated perturbations in threemetabolic pathways: (1) arginine and proline, (2) glycine, serineand threonine, and (3) alanine, aspartate, and glutamate. Etanercepttreatment reversed the majority of psoriasis-associated trends incirculating metabolites, shifting the metabolic phenotypes of severepsoriasis toward that of healthy controls. Circulating metabolitelevels pre- and post-Etanercept treatment correlated with psoriasisarea and severity index (PASI) clinical scoring (R² = 0.80; p < 0.0001). Although theresponsible mechanism(s) are unclear, these results suggest that psoriasisseverity-associated metabolic perturbations may stem from increaseddemand for collagen synthesis and keratinocyte hyperproliferationor potentially the incidence of cachexia. Data suggest that levelsof circulating amino acids are useful for monitoring both the severityof disease as well as therapeutic response to anti-TNFα treatment.

Introduction

Psoriasis is an immune-mediatedskin disease associated with significantmorbidity and mortality.¹ The ultimatecause of psoriasis remains unclear, but it involves triggering theimmune system, leading to sustained inflammation and dysregulationof keratinocyte differentiation.² Psoriasisis a heterogeneous disease with significant comorbidities including,in particular, psoriatic arthritis.³⁻⁵ Genome-wide associationstudies indicate a large genetic component in the pathogenesis ofpsoriasis.⁶ In particular, the HLA-C geneis estimated to contribute ∼50% to the heritability in earlyonset psoriasis.⁷ Once activated, dendriticcells stimulate differentiation and migration of Th1 and Th17 effectorT cells to the skin, which, through cytokine release, immune cellrecruitment, and keratinocyte proliferation, drive a sustained cycleof chronic inflammation.⁸ Patients withpsoriasis are also at a significant risk of developing metabolic syndrome,type 2 diabetes, hypertension, and obesity.⁹⁻¹¹ The underlyingmechanisms are still unclear, and, to date, comorbidities are supportedprimarily by epidemiological data.¹²

The cytokine tumor necrosis factor alpha (TNFα) is knownto play a major role in the pathophysiology of psoriasis,¹³ and anti-TNFα therapeutics are routinelyused to treat immune-mediated diseases including psoriasis as wellas psoriatic and rheumatoid arthritis in adult as well as pediatricpopulations.^5,14 One of the first-line anti-TNFαtreatments is the biologic Etanercept, which has demonstrated efficacyin resolving psoriatic lesions. Etanercept is a fusion protein thatbinds to the constant end of the IgG1 antibody^5,15 andacts as a competitive inhibitor of TNFα. Although it significantlysuppresses the associated inflammation, Etanercept, as well as otherbiologic therapeutics, does not cure the underlying disease.¹⁶ TNFα is also involved in other physiologicalprocesses such as muscle protein proteolysis and cachexia,¹⁷ thereby suggesting additional pathways by whichanti-TNFα therapeutics may affect the underlying pathophysiologyof the disease.

In order to investigate the systemic biochemicalshifts associatedwith disease severity, liquid chromatography high-resolution massspectrometry (LC–HRMS) metabolomic analysis was used to characterizecirculating metabolites in psoriasis patients. Metabolomics profileswere also compared before and after a 12 week treatment with Etanercept.A metabolomics-based research approach involves simultaneously analyzingthe complement of small molecules (metabolites) in a system. Thishas been shown to be useful for identifying metabolic traits thatrepresent intermediate phenotypes capable of linking genetic and environmentalfactors to heterogeneous diseases.^18,19 This studyis, to the best of our knowledge, the first to examine the metabolicprofile associated with plaque psoriasis severity and the effectsof Etanercept treatment. These results provide insight into the biochemicalpathways involved in the etiology of psoriasis and the systemic effectsof Etanercept treatment.

Experemintal Section

Study Design

Healthycontrols as well as patients withmild or severe psoriasis were recruited at the Karolinska UniversityHospital. Mild psoriasis patients were recruited from a cohort ofpatients with newly onset psoriasis who did not require systemic therapyand were therefore treated only topically. Severe psoriasis patientsrequired systemic therapy to control the skin manifestations of thedisease. None of the patients were on statins or prescribed anti-inflammatorydrugs. All samples were obtained prior to the commencement of anytreatment. The recruitment group consisted of 96 gender-balanced individuals(32 healthy controls and 32 mild and 32 severe psoriasis). For analysispurposes, the full cohort (n = 96) was subdividedinto two gender- and disease severity-balanced groups (n = 48 each), called exploratory and validation cohorts (Table 1). The exploratory cohort was used to identify metabolicmarkers for psoriasis severity, and the validation cohort as a confirmationof the identified trends. Additional plasma samples were taken fromthe severe psoriasis patients (n = 16) in the validationcohort following 12 weeks of Etanercept (Enbrel) treatment (50 mgonce per week subcutaneously), and this group is referred to as thetreatment cohort. For blood collection, 10 mL of whole blood was collectedin EDTA tubes after overnight fasting. Samples were left standingfor 1 h before centrifugation at room temperature for 20 min at 3100rpm. After centrifugation, samples were aliquoted and immediatelystored at −70 °C until use. Psoriasis disease was judgedas severe when it required systemic therapy and was evaluated by thepsoriasis area and severity index (PASI), which is an establishedmeasurement that quantifies the thickness, redness, scaling, and distributionof psoriasis lesions.²⁰ The study was approvedby the Regional Committee of Ethics and was performed according tothe Declaration of Helsinki Principles. Signed consent forms werecollected from all sample donors.

HILIC Mode Metabolomics

A cocktail of four internalstandards (10 μL; Table S1) was addedto 50 μL of EDTA plasma. Proteins were precipitated using 200μL of HPLC grade acetonitrile (Rathburn). Samples were vortexedfor 5 s and then left to stand on ice for 10 min followed by centrifugationat 15 000 rcf for 10 min at 4 °C. The supernatant (150μL) was transferred to a clean Eppendorf tube, and 20 μLof each sample was used to produce a pooled quality control. Sampleswere stored at −20 °C prior to analysis. Prepared sampleswere analyzed on a Thermo Ultimate 3000 HPLC and Thermo Q-Exactive(Orbitrap) mass spectrometer. Ten microliters of sample was injectedon a Merck Sequant ZIC-HILIC column (150 × 4.6 mm, 5 μmparticle size) coupled to a Merck Sequant ZIC-HILIC guard column (20× 2.1 mm). Mass spectrometry data were acquired (full scan mode)in both positive and negative ionization modes, using 140 000mass resolution.

Reversed-Phase (RP) Metabolomics

A cocktail of fiveinternal standards (10 μL; Table S1) was added to 50 μL of EDTA plasma followed by 150 μLof chilled (−20 °C) methanol (Rathburn) for protein precipitation.Samples were vortexed for 5 s and left to stand for 2 h at −20°C, followed by centrifugation at 15 000 rcf for 12 minat 4 °C. The supernatant (90 μL) was transferred to a cleanEppendorf tube, and 10 μL of each sample was used to producea pooled quality control. On the analysis day, samples were diluted1:1 with Milli-Q water (Millipore). Prepared samples were analyzedon a Thermo Ultimate 3000 HPLC and Thermo Q-Exactive (Orbitrap) massspectrometer. Twenty microliters of sample was injected on a ThermoAccucore aQ RP C18 column (150 × 2.1 mm, 2.7 μm particlesize). Mass spectrometry data were acquired (full scan mode) in bothpositive and negative ionization modes, using 70 000 mass resolution.Detailed methods are provided in the SupportingInformation.

Data Processing and Metabolite Annotation

RAW fileswere converted to universal mzXML file using MSconvert.²¹ All chromatograms were evaluated using the opensource software package XCMS²² performedin R.²³ For the preliminary analysis, metaboliteswere annotated by matching accurate mass and retention time (AMRT)to authentic chemical reference standards. Variables of importanceidentified from the multivariate analyses (see Statistical Analysis) were subjected to further identity confirmationby comparing fragmentation patterns to those of chemical standards.The MS/MS spectra of all reported metabolites matched those of thestandards with the exception of inosine (which was excluded from furtheranalysis). Data analysis was limited to metabolites matching the AMRTand MS/MS fragmentation spectra of standards except for sphingosine-1-phosphateand GlcCer(C16:0), which were identified only by AMRT. The coefficientof variance (CV) of the HILIC internal standard cocktail was <35%,and for reversed-phase, <15%. All CVs of the discussed metaboliteswere <30%, except for cystathionine and cytidine (exploratory andvalidation cohorts) and cysteine and proline (validation cohort only).The median CV for the identified metabolites was 14.8 and 16.0% forthe exploratory and validation cohorts, respectively.

StatisticalAnalysis

Statistical analysis was usedto identify significantly altered metabolites within the exploratoryand validation cohorts between (1) control and mild, (2) control andsevere, (3) mild and severe, and (4) severe pre- and post-treatmentwith Etanercept (validation cohort only) psoriasis patients. Comparisonsfor 1–3 were made using two-sample t-tests,and for 4, based on a paired t-test carried out inthe R statistical programming environment.²³ The false discovery rate (FDR) due to the multiple hypotheses testedwas adjusted according to Benjamini and Hochberg (q = 0.05)²⁴ and reported as p_adj. FDR was also directly estimated according to Dabneyand Storey²⁵ and reported as the q-value.

Multivariate analysis was performed usinga combination of principal component analysis (PCA) and orthogonalprojection to latent structures–discriminant analysis (OPLS-DA)using SIMCA-P 13 (Umetrics, Umeå, Sweden). OPLS-DA was conductedfollowing logarithmic transformation (base 10), mean centering, andscaling to unit variance (UV). OPLS-DA model performance was evaluatedbased on the cumulative coefficient of correlation between group labels(Y) and model projection of metabolites (X) (R²Y_cum) and 7-fold cross-validated model fit to the data (Q²_cum), the significance of whichwas assessed through cross-validation analysis of variance (CV-ANOVA).Model predictive and orthogonal components are reported as (predictive+ orthogonal). Iterative feature selection (2 rounds) was performedto identify important metabolic discriminants between the comparedpopulations. Metabolites were retained in the model based on a combinationof variable importance in projection (VIP) > 1.0 and absolute magnitudeof correlation with model scores (p_corr > 0.4).²⁶ SIMCA-P was used to calculatethe PLS inner relation between disease severity score (PASI) and correlatedmetabolites (|Pearson’s correlations| > 0.5).

Pathway EnrichmentAnalysis

Biochemical pathway enrichmentanalysis was used to identify psoriasis-dependent changes in globalbiochemical domains. MetaboAnalyst²⁷ wasused to test for significant enrichment in KEGG pathways (http://www.genome.jp/kegg/) among the noted metabolic perturbations in common to both the exploratoryand validation cohorts (Table 2). Significantenrichment was assessed on the basis of the false discovery rate-adjustedhypergeometric test statistic (p ≤ 0.05),and impact on pathway topology was defined based on relative-betweenesscentrality.

Partial Correlation Network Analysis

Gaussian graphicalmodel networks were calculated for metabolite relationships in thecontext of the identified differences between (1) control and severe,(2) control and treated severe, and (3) untreated and treated severepsoriasis patients from the validation cohort. q-orderpartial correlations (q = 1, 12, 24, 35)²⁸ were calculated (n = 1000)in R (v3.0.1)²³ between metabolites (n = 93) from the validation data set, excluding patientswith mild psoriasis (n = 48). To maximize networknode inclusion and minimize edge degree, a threshold of β =0.4 for the average nonrejection rate (β) for metabolite pairwiserelationships was selected. Using this approach, a smaller averagenonrejection rate corresponds to stronger q-orderpartial correlation. Spearman’s rank order coefficients ofcorrelation, p-values, and FDR-adjusted p-values (p_adj)²⁴ were calculated for all q-order selected relationships.Cytoscape²⁹ was used to generate networkvisualizations for all edges displaying p_adj ≤ 0.05 (83 edges or 90% of the original q-order calculated edges). Network mapping was used to encode anddisplay statistical and multivariate analysis results within the contextof the partial correlation defined relationships.

Results

Psoriasis patients and control characteristics were consistentbetween the exploratory and validation cohorts (Table 1). PASI scoring significantlyincreased (p < 0.05) with disease severity inboth cohorts and decreased following Etanercept treatment (Table 1). HILIC mode and reversed-phase metabolomic analysesof the exploratory and validation cohorts’ plasma were usedto identify 94 and 93 metabolites from the XCMS diffreport of allsamples, respectively (67 of which were in common), through matchingof retention time and spectra to external standards (Table S2). Statistical analyses of the metabolomic measurementswith adjustment for FDR were used to identify significantly alteredcompounds among control, mild, and severe psoriasis patients for boththe exploratory and validation cohorts (TablesS3 and S4) and between severe and severe Etanercept-treatedpsoriasis patients in the validation cohort (TableS4). Additionally, changes in >150 metabolite features putativelyidentified based on only accurate mass are reported for the comparisonsbetween severe psoriasis patients and controls in both cohorts (Tables S5 and S6) and between severe psoriasispatients at baseline and after Etanercept treatment (Table S7).

Table 1

Characteristics of the Study Cohortsa

	exploratory cohort
	control	mild	severe
gender	8/8d	8/8	8/8
age (years)	52 ± 9	52 ± 8	58 ± 10
BMIb	26.1 ± 4.1	25.0 ± 4.8	28.4 ± 3.5
PASIc	n/a	1.4 ± 0.7e	16.5 ± 7.4e
cholesterol	5.2 ± 0.8	5.1 ± 0.9	4.8 ± 0.9
triglycerides	1.0 ± 0.4	1.0 ± 0.56	1.3 ± 0.5

	validationcohort
	control	mild	severe	severe (treated)f
gender	8/8	8/8	8/8	8/8
age (years)	44 ± 13	42 ± 20	53 ± 13	53 ± 13
BMI	24.0 ± 3.1	25.2 ± 4.7	27.3 ± 4.8	27.2 ± 4.6
waistline (cm)	n/d	n/d	98 ± 13.8	97 ± 12.9
PASI	n/a	1.6 ± 1.0e	13.6 ± 4.5e	4.9 ± 3.4g
cholesterol	5.1 ± 0.8	4.9 ± 0.9	5.3 ± 0.6	6.57 ± 4.1h
triglycerides	1.1 ± 0.6	1.2 ± 1.0	1.0 ± 0.4	1.7 ± 0.9

aValues are reportedas the mean± SD. Units for cholesterol and triglycerides are mmol/L. n/dindicates that the value was not determined.

bBMI, body mass index.

cPASI, psoriasis area and severityindex. There is no PASI score for the control group (n/a).

dGender balance: male/female.

ep-value < 0.05for a two-sample t-test for the PASI score betweenmild and severe psoriasis subjects.

fSevere psoriasis patients treatedwith Etanercept for 12 weeks.

gp-value < 0.05for a paired two-sample t-test for the PASI scorefor treated vs untreated severe psoriasis.

hp-value < 0.05based on a paired t-test.

As expected, the largest effect size was observedbetween controland severe psoriasis patients, with 33 and 34 significantly (p_adj ≤ 0.05) perturbed plasma metabolitesin the exploratory and validation cohorts, respectively (Tables S3 and S4). Comparison of the psoriasis-associatedmetabolic alterations in common to both cohorts identified 20 significantly(p_adj ≤ 0.05) altered metabolites,17 of which increased with psoriasis severity in both cohorts (Table 2). In particular, ornithineand another urea cycle intermediate, citrulline, increased by 215and 90%, on average, respectively, in severe psoriasis patients comparedto that in controls (Table 2). Etanercept treatmentled to reductions in 10 of the 20 (50%) previously identified psoriasis-associatedmetabolic dysregulations (Table 2). Specifically,treatment resulted in significant reductions in amino acids, highlightedby 230, 233, and 150% decreases in threonine, ornithine, and methionine,respectively (Table 2). Comparison of Etanercept-treatedsevere psoriasis to controls revealed a normalization in the majority(89%) of metabolites previously shown to be increased with disease.Although cystine was significantly reduced by 10% following treatment,this amino acid remained 60% elevated in the treated group relativeto that in controls (Table 2). Cystathioninewas the only metabolite in common to both cohorts that was reduced(80%) in severe psoriasis compared to that in controls and was notsignificantly affected by Etanercept treatment (Table 2). Similarly, sphingosine-1-phosphate levels were not affectedby treatment and remained 70% elevated in the treated group (Table 2).

Table 2

Fold Changes in MetabolitesAssociatedwith Severe Psoriasis That Showed Similar Patterns in the Exploratoryand Validation Cohorts

		exploratory cohort		validation cohort		Etanercept treated cohort
		severevs controlb		severe vs control		treated vs severe		treated vs control
pathway	metabolitea	q-valuec	FCd	q-value	FC	q-value	FC	q-value	FC
arginine andproline pathwaye	arginineo	1.82 × 10^–2	2.47	1.43 × 10^–3	2.29	7.58 × 10^–3	0.45	7.48 × 10^–1	1.02
	citrullineo	4.09 × 10^–4	2.61	1.16 × 10^–5	1.88	1.46 × 10^–2	0.64	4.07 × 10^–1	1.19
	ornithineo	6.92 × 10^–3	3.37	1.77 × 10^–4	2.92	2.18 × 10^–3	0.41	4.07 × 10^–1	1.21
	prolineo	1.24 × 10^–2	1.83	1.40 × 10^–2	1.89	1.09 × 10^–1	0.65	4.16 × 10^–1	1.22
	hydroxyprolineo	2.16 × 10^–2	2.93	6.88 × 10^–3	1.81	7.87 × 10^–2	0.69	3.99 × 10^–1	1.25
glycine,serine,and threonine pathwayf	glycineo	9.42 × 10^–3	2.13	2.87 × 10^–3	1.68	2.18 × 10^–3	0.49	3.99 × 10^–1	0.83
	serineo	4.74 × 10^–3	1.96	1.15 × 10^–4	1.61	1.35 × 10^–1	0.78	3.58 × 10^–1	1.25
	threonineo	4.09 × 10^–4	2.68	3.99 × 10^–6	2.58	7.77 × 10^–7	0.32	3.85 × 10^–1	0.82
alanine,aspartate,and glutamate pathwayg	aspartateo	7.66 × 10^–3	2.43	2.63 × 10^–5	1.96	1.39 × 10^–1	0.69	3.99 × 10^–1	1.34
	glutamateo	1.26 × 10^–2	3.00	4.80 × 10^–3	2.22	3.33 × 10^–1	0.78	3.58 × 10^–1	1.72
	glutamineo	1.75 × 10^–3	2.03	4.26 × 10^–4	1.71	7.34 × 10^–3	0.59	7.48 × 10^–1	1.02
cysteineandmethionine pathwayh	cystineo	4.07 × 10^–3	2.81	1.77 × 10^–4	1.83	2.86 × 10^–1	0.9	6.07 × 10^–3	1.64
	cystathionineo	1.13 × 10^–2	0.83	3.99 × 10^–6	0.31	2.96 × 10^–1	1.51	1.94 × 10^–2	0.47
	methionineo	1.41 × 10^–2	2.15	1.77 × 10^–4	1.84	4.82 × 10^–5	0.42	3.58 × 10^–1	0.78
taurine and hypotaurinepathwayi	taurineo	2.70 × 10^–4	1.92	4.96 × 10^–4	1.47	1.79 × 10^–1	0.82	3.38 × 10^–1	1.21
phenylalaninepathwayj	phenylalanineo	1.90 × 10^–2	1.34	1.89 × 10^–4	1.48	4.52 × 10^–2	0.73	5.38 × 10^–1	1.07
pyrimidine pathwayk	cytidineo	7.80 × 10^–3	2.09	2.55 × 10^–2	2.47	2.42 × 10^–1	0.65	3.58 × 10^–1	1.6
amino sugar pathwayl	acetylglucosamineo	6.92 × 10^–3	2.36	2.90 × 10^–3	1.34	7.87 × 10^–2	0.72	7.12 × 10^–1	0.96
sphingolipidpathwaym	glucosylceramide (C16:0)p	4.29 × 10^–2	1.46	1.01 × 10^–2	1.59	4.34 × 10^–1	0.95	3.02 × 10^–1	1.51
sphingolipidpathwaym	sphingosine-1-phosphatep	3.54 × 10^–2	1.21	1.89 × 10^–4	1.93	3.06 × 10^–1	0.88	2.94 × 10^–2	1.69

aAll metabolites displayed were notsignificantly altered in mild vs control or mild vs severe psoriasispatients in either the exploratory or validation cohorts.

bSevere psoriasis patients vs healthycontrols.

cFalse discoveryrate (FDR) was directlyestimated according to the methods of Dabney and Storey.²⁵

dFoldchange between the two groups.

eKEGG Pathway map hsa00330: arginineand proline metabolism.

fKEGG Pathway map hsa00260: glycine,serine, and threonine metabolism.

gKEGG Pathway map hsa00250: alanine,aspartate, and glutamate metabolism.

hKEGG Pathway map hsa00270: cysteineand methionine metabolism.

iKEGG Pathway map hsa00430: taurineand hypotaurine metabolism.

jKEGG Pathway map hsa00360: phenylalaninemetabolism.

kKEGG Pathwaymap hsa00240: pyrimidinemetabolism.

lKEGG Pathwaymap hsa00520: aminosugar and nucleotide sugar metabolism.

mKEGG Pathway map hsa00600: sphingolipidmetabolism.

oResults obtainedfrom HILIC analysis.

pResultsobtained from reversed-phaseanalysis.

The relationshipbetween psoriasis disease severity score (PASI)and the metabolites identified in Table 2 wasfurther interrogated using correlation analysis and partial least-squares(PLS) inner relation. Of the metabolites presented in Table 2, 10 correlated with psoriasis disease severityscores (r ≥ 0.5) in either the validationor treated cohort (Table 3). A PLS inner relationwas calculated between these 10 metabolites and PASI scores (Figure 1), producing a significant multivariate association(R² = 0.80; p < 0.0001)between PASI scores (Figure 1A) and metaboliteabundances (Figure 1B). Threonine, citrulline,and ornithine displayed the highest positive correlation with PASIscores (Table 3), whereas threonine, glutamine,and ornithine were the most highly ranked multivariate predictorsof psoriasis severity (Figure 1B). FollowingEtanercept treatment, there was a significant reduction in PASI scoresassociated with a normalization in all but two of the metabolitesthat were altered in severe psoriasis patients within both cohorts(Table 2).

Figure 1

PLS inner relation between the 10 metabolitesidentified to correlate(r ≥ 0.5) with the psoriasis area and severityindex (PASI) in severe psoriasis patients pre- and post-treatmentwith Etanercept (Table 3). (A) The inner relationfor severe psoriasis and treated severe psoriasis patients from thevalidation cohort (R² = 0.80). For untreatedsevere psoriasis only, R² = 0.82; formild psoriasis only, R² = 0.32; and formild, severe untreated, and treated psoriasis combined, R² = 0.78. For the exploratory cohort, R² = 0.91 for severe psoriasis, R² = 0.49 for mild and severe psoriasis combined, and R² = 0.26 for mild psoriasis only. (B) The variableimportance in projection (VIP) plot displaying the relative contributionsof the individual metabolites to the inner relation. S1P=sphingosine-1-phosphate.

Table 3

Pearson’sCorrelations betweenDisease Severity (PASI) and Plasma Metabolite Levelsa

metabolite	exploratorycohortb	validationcohortb	treated cohortb^,c
arginine	0.52**	0.50***	0.60**
citrulline	0.73***	0.84***	0.70***
ornithine	0.43*	0.84***	0.75***
proline	0.77***	0.23	0.37*
hydroxyproline	0.20	0.43*	0.23
glycine	0.74***	0.47***	0.69***
serine	0.44**	0.73***	0.55**
threonine	0.84***	0.87***	0.88***
aspartate	0.05	0.26	0.17
glutamate	0.33	0.42*	0.09
glutamine	0.70***	0.66***	0.76***
cystine	0.79***	0.75***	0.56**
cystathionine	–0.27	–0.43	–0.19
methionine	0.39*	0.46**	0.74***
taurine	0.34	0.51**	0.22
phenylalanine	0.31	0.66**	0.37*
cytidine	0.63***	0.36*	0.34
acetylglucosamine	0.28	0.48**	0.45*
glucosylceramide(C16:0)	0.50**	0.34	0.19
sphingosine-1-phosphate	0.35	0.50*	0.16

aMetabolites in the validation andtreated cohorts with r ≥ 0.5 were used forthe regression with PASI score in Figure 1.Correlations include both mild and severe psoriasis patients for boththe exploratory and validation cohorts.

bThe significance level is indicatedas follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001. The confounding effectsof age and BMI were tested via a linear regression model using STATA11.

cSevere psoriasis patientsfrom thevalidation cohort were treated with Etanercept for 12 weeks. Correlationvalues are only for the treated patients.

Data sets were further interrogated using multivariatemethods.PCA identified no outliers in the exploratory or validation cohortsbased on Hotelling’s T² (i.e., 95% confidence interval)or DModX (data not shown). Multivariate classification modeling (OPLS-DA)followed by feature selection was implemented to identify top metabolicmarkers for disease-specific differences in plasma metabolite profilesin both cohorts (Figure 2 and Table S8). For the exploratory cohort, only severe psoriasisproduced significant models (p < 0.05; mild vssevere [Q² = 0.605]; control vs severe[Q² = 0.741]). For the validation cohort,all calculated models were significant (p < 0.001)and showed a disease severity-based increase in circulating metabolitesfrom control vs mild (Q² = 0.626) to mildvs severe (Q² = 0.794) to control vs severe(Q² = 0.891). Etanercept treatment resultedin metabolic profiles that were shifted from both severe untreated(Q² = 0.645) and controls (Q² = 0.534), giving a unique pharmacological phenotype.However, the Etanercept vs control OPLS-DA model was the weakest ofall generated models (p = 2.76 × 10^–4), indicating that the treated cohort had a metabolic profile mostsimilar to the controls. All model statistics are provided in Table S8.

Figure 2

OPLS-DA scores and variable importancein projection (VIP) plotsfrom the curated models following 2 rounds of variable selection asdescribed in the Experimental Section. (A)Scores plot of control vs severe psoriasis in the exploratory cohort(R²Y = 0.762 Q² = 0.741, CV ANOVA p = 3.2× 10^–9, 1 + 0 components); (B) VIP plot ofcontrol vs severe psoriasis in the exploratory cohort; (C) scoresplot of control vs severe psoriasis in the validation cohort (R²Y = 0.895 Q² = 0.891, CV ANOVA p = 1.1 × 10^–14, 1 + 0 components); (D) VIP plot of control vs severepsoriasis in the validation cohort. S1P=sphingosine-1-phosphate.

Biochemical pathway enrichmentanalysis of the psoriasis-associatedmetabolic perturbations in common to both cohorts (Table 2) was used to identify significant perturbations(p ≤ 0.05) in 10 major biochemical pathways(Table S9). Partial correlation networkswere calculated to analyze empirical metabolite–metaboliterelationships in the context of the identified psoriasis-associatedmetabolic perturbations (Figure 3). On thebasis of the network topology, the three most psoriasis-impacted pathwayswere those of alanine, aspartate, and glutamate metabolism (hsa00250);glycine, serine, and threonine metabolism (hsa00260); and arginineand proline metabolism (hsa00330). In particular, there was a dominantpsoriasis-dependent increase in the majority of urea cycle intermediatesincluding aspartate, arginine, ornithine, and citrulline. The confirmedchanges in metabolites (Table 2) are highlightedin these networks (thick borders) and can be classified into threemajor correlated clusters (Figure 3A): (1)cytidine, cystathionine, acetylglucosamine, hydroxyproline, and taurine;(2) ornithine, arginine, threonine, methionine, glutamine, glycine,citrulline, and proline; and (3) phenylanine, cystine, GlcCer(C16:0),aspartate, and glutamate. Metabolic changes within these three clusters,with the exception of cystathionine, were positively correlated andincreased with psoriasis severity (Table 3).Etanercept treatment of severe psoriasis patients predominantly impactedcluster 2 metabolites (Figure 3B). Comparisonof the Etanercept treated group to healthy controls in the validationcohort (Figure 3C) revealed normalization inthe majority of the previously identified psoriasis-associated metabolicperturbations with the exception of cystathionine and cystine.

Figure 3

Dependencynetwork displaying plasma metabolite relationships inpsoriasis in the context of the noted metabolic perturbations between(A) severe untreated psoriasis and control, (B) severe treated andsevere untreated psoriasis, and (C) severe treated psoriasis and controlpatients. Metabolites are connected based on partial correlation definedrelationships, and edge color and width display the direction andmagnitude of the FDR-adjusted Spearman rank order coefficient of correlation(p_adj ≤ 0.05). Vertices representmetabolites, with the shape and color displaying relative directionand statistical significance (p_adj ≤0.05) of the metabolic change for each respective comparison (i.e.,panel A displays changes in severe psoriasis patients relative tocontrols). Metabolites are sized according to each comparison’srespective OPLS-DA model VIP (Figure 2), andspecies in common in the exploratory and validation models’selected feature sets are highlighted with thick black borders.

Discussion

There is an extensivebody of literature on psoriasis; however,to date, there has been limited work investigating the underlyingmetabolic processes associated with the disease.³⁰ To the best of our knowledge, this is the first study utilizingnontargeted metabolomics to study the effect of psoriasis severityand the impact of Etanercept treatment on metabolism. Comparisonsof plasma metabolic profiles of psoriasis patients suggest that themild and severe disease states are not, from a metabolic perspective,distinct pathologies but a progression of the disease along a sharedmetabolic continuum. However, although distinct shifts in the circulatorymetabolic profiles of severe psoriasis patients were observed, itis unclear as to the corresponding mechanism responsible.

Psoriasisseverity was associated with an increase in three intermediatesof the urea cycle (citrulline, ornithine, and arginine). Shifts inarginine and urea cycle metabolism have been previously reported inpsoriatic skin lesions,³¹ and similar shiftsin urea cycle intermediates,³² as wellas changes in glutamine and glutamate, have been associated with woundhealing.³³ The commonality in markers forpsoriasis and wound healing is not surprising given that both psoriasis^34,35 and wound healing³⁶⁻³⁸ involve the production of new keratinocytes. Theurea cycle is an entry to the pathway for the synthesis of polayamines,which are essential hormones in cell proliferation, a hallmark ofkeratinocytosis in psoriasis.³⁹ The polyaminerequirement may promote the mobilization of the urea cycle intermediatearginine from its sites of synthesis to the skin, resulting in theobserved enhanced plasma levels.

Protein synthesis demand inthe proliferating skin could also explainthe elevated amino acid profile in plasma. Psoriasis is associatedwith changes in protein expression.⁴⁰ Cornificationof the epidermis requires different scaffolding proteins than thatin healthy cells, and a collection of support proteins such as smallproline-rich protiens (SPRP), hornirine (HNRN), and late cornifiedenvelope 3 A (LCE3A) were elevated up to 500 times in psoriasis skincompared to that in healthy skin.⁴⁰ Theproduction of these psoriasis-enriched proteins postulates an enhancedinflux of amino acids. The requirements of this process agree largelywith the observed increases in circulating amino acids. The most representedamino acids in the regulated proteins in psoriasis were serine, proline,glycine, and glutamine. This profile was not altered when a scoreof fold change (psoriasis vs healthy) of protein expression was calculated(Table S10). To add to the requirementburden, the major amino acids in human collagen 1 alpha are glycine(27%) and proline (18%). While collagen is produced in the dermis,which does not thicken in psoriasis, there are indications that collagenturnover is higher in psoriasis patients, with reported enhanced activityof collagen breakdown enzymes prolidase⁴¹ and matrix metalloproteasease MMP1⁴² (thelatter being 13 times higher in psoriasis patients compared to thatin healthy controls). Hydroxyproline, a marker for tissue collagendegradation,⁴³ was upregulated in severepsoriasis patients and normalized by Etanercept treatment. This agreeswith Garvican et al., who showed the ability of IL-1 or TNFαto promote collagen degradation in ovine cartilage.⁴⁴ The modest correlation of hydroxyproline with PASI scoremay indicate different susceptibility of subjects, and further investigationsare necessary to examine if this susceptibility is reflected withincidence of psoriatic arthritis.

Another potential explanationfor the observed increase of circulatingamino acids would be due to cachexia or wasting syndrome, which isthe loss of lean body mass that can accompany systemic inflammatorydiseases.⁴⁵ Cachexia has been linked withpsoriasis,⁴⁶ and cytokine inhibitors (e.g.,anti-TNFα) have been suggested in the treatment of cachexia.⁴⁷ TNFα in particular is thought to playa role in both the anorexic effect via neuronal leptin receptors⁴⁸ as well as muscle wasting by enhancing proteinubiquitation.⁴⁹ A cachectic state of psoriasispatients has been suggested due to a dual role of TNF, evidenced bythe increase in BMI during anti-TNFα treated patients.^50,51 Accordingly, it is possible that increased muscle protein breakdown,a hallmark of cachexia, may explain the higher levels of circulatingamino acids observed in subjects with severe psoriasis. Muscle wastinghas been reported in other autoimmune diseases such as rheumatoidarthritis;⁵² however, this may not be associatedwith changes in body weight.⁵³ In our study,BMI was marginally higher in severe psoriasis patients (Table 1; p = 0.07), whereas the BMI valuesof each subject did not change following treatment (p = 1.0). Treatment with anti-TNFα has previously been reportedto increase BMI,⁵⁴ but this was primarilydue to increases in fat-free mass, which was not examined in the currentcohort.

Cachexia has not been widely studied from a metabolicperspective,and there is no consensus on plasma levels of metabolites in cachecticpatients or animal models. Peters et al. showed that in a tumor-bearingmouse model plasma amino acids were upregulated.⁵⁵ O’Connell et al. reported changes in lipids, glycerol,and glucose, but not amino acids, in the plasma of a murine cancercachexia model.⁵⁶ A follow up study observedan increase in urea cycle amino acids and decreased glycine, alanine,and serine in skeletal muscle.⁵⁷ Ubhi etal. observed a slightly significant (p = 0.05–0.1)increase in plasma amino acids of cachectic compared to noncachecticCOPD patients.⁵⁸ Moreover, a number ofcachectic studies exhibited lower levels of circulating amino acidsduring cachexia in clinical as well as animal studies.⁵⁹⁻⁶¹ However, a downregulation of circulating branched-chain amino acids,which is thought to be a hallmark of cachexia,⁶² was not observed in the current study. Accordingly, althoughit is unlikely that the observed shifts in circulating amino acidsare due to cachexia, further evaluation is warranted.

The dominanteffect of Etanercept treatment was observed in normalizingthe plasma levels of a large cluster of positively correlated metabolitesconsisting of ornithine, arginine, proline, citrulline, glycine, glutamine,threonine, and methionine (Figure 3), specificallywithin the arginine/proline and glycine, serine, and threonine pathways(Table S9). The biochemical mechanism leadingto the Etanercept-dependent reduction in these metabolites is unclear.However, blocking the immune (autoimmune) response can lead to a reducedsignal for collagen and other keratinocyte-specific structural proteinproduction as well as keratinocytosis, resulting in a diminished requirementfor these metabolites. Etanercept acts by inhibiting the activityof the cytokine TNFα,¹⁵ which isinvolved in a wide range of biological activities. TNFα canmodulate the activity of nitric oxide synthase (NOS),⁶³ which is involved in the production of nitric oxide fromthe conversion of arginine to citrulline.⁶⁴ If Etanercept treatment, through inhibition of TNFα, was impactingNOS activity, then shifts would be expected in the ratio of the citrulline-to-arginineconcentration following treatment. However, a large reduction in allurea cycle metabolites, including arginine and citrulline, was observed,which does not support the modulation of NOS activity as a mode ofaction of treatment. Similarly, there are no reported mechanisms bywhich Etanercept may lead to reductions in intermediates of glycine,serine, and threonine metabolism. Etanercept could affect these metabolitesthrough modulation of NOS, for example, by modulating the flux ofaspartate between the urea cycle and threonine production. However,an analysis of metabolite partial correlations (Figure 3) reveals that aspartate levels are not directly linked tourea cycle intermediates but instead to glutamate. Of the 20 metabolitesidentified to shift with psoriasis, only cystathionine and cystinewere not normalized to healthy levels following Etanercept treatment(Table 2 and Figure 3). Cystathionine was not correlated with PASI scores prior to treatment(Table 3), which may explain the lack of responseand suggests a distinct mechanism. The fact that circulating aminoacid levels returned to normal following anti-TNFα treatmentin combination with the strong correlation to PASI score (before andafter treatment) indicates that amino acid metabolism is a good markerfor anti-TNFα responsiveness, as indicated by Kapoor et al.⁶⁵

Psoriasis patients are at increased riskof metabolic syndromeand diabetes,¹⁰ which share the commonelement of insulin resistance. The excess amino acid availabilitycan stimulate the nutrient-sensitive mTOR/S6K pathway and inhibitserine phosphorylation of insulin receptor substrate 1, which canlead to an impairment in insulin-stimulated glucose disposal in skeletalmuscles and insulin-mediated inhibition of glucose production.⁶⁶ Accordingly, although the chronic inflammatorystatus of psoriasis certainly plays a direct role in the developmentof insulin resistance, the observed enhanced circulating levels ofamino acids suggest the hypothesis that the mTOR/S6K pathway may contributeto this risk. However, the validation of this tentative hypothesiswarrants further investigation.

Conclusions

Althoughthe severity of psoriasis is clearly linked to levelsof circulating amino acids, the responsible mechanism(s) for the observedshifts are unclear. The observed increased levels may be due to keratinocytehyperproliferation, increased proteolysis due to cachexia, or otherunknown pathways. During hyperproliferation, the increased demandof protein building units, and specifically proline, may lead to astrong shift in amino acid profiles. Alternatively, it can be hypothesizedthat individuals with severe psoriasis are cachetic. There is a paucityof information on cachexia in psoriasis, but the majority of studiesreport an increase in BMI, which is not affected by Etanercept treatmentin this study. Accordingly, further investigations are required tounderstand the significance of the observed amino acid shifts. Itis clear that Etanercept treatment significantly shifts the metabolicprofiles of psoriasis patients, reversing the distinct psoriasis metabotypeto that observed in healthy individuals, suggesting that focused metabolicprofiling can be used to monitor patient response to therapeutic interventionsystematically. The strong correlation of disease severity scoringwith the metabolite levels indicates that the observed metabolic shiftreflects a trajectory of disease progress rather than distinct diseasepathologies. It is also possible that circulating amino acid profilescould be used as markers of both disease severity as well as responsivenessto treatment.

Acknowledgments

We thankresearch nurse Helena Griehsel for excellent technicalassistance. D.G. was supported by NIH Metabolomics Center grant no.DK097154. M.S. acknowledges support from the Swedish Research Council(K2012-57X-14202-11-6 and CERIC Linné Center), Stockholm CountyCouncil (20120059), Hudfonden, and Psoriasisfonden. C.E.W. was supportedby the Center for Allergy Research (Cfa) and the Karolinska Institutet.

Supporting Information Available

Materials and Methods: Detaileddescription of the methods used for analyzing the samples, data processing,safety considerations, and metabolite annotation. Table S1: Analyticalinternal standards used for HILIC and RP mode metabolomic analysis.Table S2: List of the chemical reference standards used for annotatingmetabolite features. Table S3: Summary of metabolic perturbationsassociated with psoriasis disease severity for the exploratory cohort.Table S4: Summary of metabolic perturbations associated with psoriasisdisease severity for the validation cohort. Table S5: Changes in putativelyidentified (accurate mass) metabolite features in severe psoriasisrelative to controls in the exploratory cohort. Table S6: Changesin putatively identified (accurate mass) metabolite features in severepsoriasis relative to controls in the validation cohort. Table S7:Changes in putatively identified (accurate mass) metabolite featuresin severe psoriasis relative to controls in the treatment cohort.Table S8: OPLS-DA model classification performance statistics forthe exploratory and validation cohorts. Table S9: Biochemical pathwayenrichment analysis of psoriasis-associated metabolic perturbationsin common to the exploratory and validation cohorts. Table S10: Requirementof amino acids for the regulated proteins in psoriasis. This materialis available free of charge via the Internet athttp://pubs.acs.org.

Supplementary Material

pr500782g_si_001.pdfpr500782g_si_001.pdf

Author Contributions

^⊥ M.A.K. and S.G.S. contributed equally to this work.

The authorsdeclare no competing financial interest.

References

1. RachakondaT. D.; SchuppC. W.; ArmstrongA. W.Psoriasis prevalence among adultsin the United States. J. Am. Acad. Dermatol.2014, 70, 512–516.[PubMed][Google Scholar]
2. BenoitS.; et al. Elevated serum levelsof calcium-binding S100 proteins A8 and A9reflect disease activity and abnormal differentiation of keratinocytesin psoriasis. Br. J. Dermatol.2006, 155, 62–66.[PubMed][Google Scholar]
3. HenselerT.; ChristophersE.Psoriasisof early and late onset: characterizationof two types of psoriasis vulgaris. J. Am. Acad.Dermatol.1985, 13, 450–456.[PubMed][Google Scholar]
4. McGonagleD.; LoriesR. J. U.; TanA. L.; BenjaminM.The concept of a “synovio-enthesealcomplex” and its implications for understanding joint inflammationand damage in psoriatic arthritis and beyond. Arthritis Rheum.2007, 56, 2482–2491.[PubMed][Google Scholar]
5. MeaseP. J.; et al. Etanercept in the treatment of psoriatic arthritis and psoriasis:a randomised trial. Lancet2000, 356, 385–390.[PubMed][Google Scholar]
6. A genome-wideassociation study identifies new psoriasis susceptibility loci andan interaction between HLA-C and ERAP1. Nat.Genet.2010, 42, 985–990.[PubMed][Google Scholar]
7. EnerbäckC.; MartinssonT.; InerotA.; WahlströmJ.; EnlundF.; YhrM.; SwanbeckG.Evidence that HLA-Cw6determines early onset of psoriasis, obtained using sequence-specificprimers (PCR-SSP). Acta Derm.-Venereol.1997, 77, 4.[PubMed][Google Scholar]
8. NestleF. O.; KaplanD. H.; BarkerJ.Psoriasis. N. Engl. J. Med.2009, 361, 496–509.[PubMed][Google Scholar]
9. CarrascosaJ. M.; et al. Obesity and psoriasis: inflammatory nature of obesity, relationshipbetween psoriasis and obesity, and therapeutic implications. Actas Dermo-Sifiliogr.2014, 105, 31–44.[Google Scholar]
10. AzfarR. S.; et al. Icreased risk of diabetes mellitus and likelihood of receiving diabetesmellitus treatment in patients with psoriasis. Arch. Dermatol.2012, 148, 995–1000.[PubMed][Google Scholar]
11. MaC.; HarskampC. T.; ArmstrongE. J.; ArmstrongA. W.The associationbetween psoriasis and dyslipidaemia: a systematic review. Br. J. Dermatol.2013, 168, 486–495.[PubMed][Google Scholar]
12. AzfarR. S.; GelfandJ. M.Psoriasis and metabolic disease: epidemiology and pathophysiology. Curr. Opin. Rheumatol.2008, 20, 416–422.[PubMed][Google Scholar]
13. SchönM. P.; BoehnckeW.-H.Psoriasis. N. Engl. J. Med.2005, 352, 1899–1912.[PubMed][Google Scholar]
14. IyerS.; YamauchiP.; LoweN. J.Etanerceptfor severe psoriasis andpsoriatic arthritis: observations on combination therapy. Br. J. Dermatol.2002, 146, 118–121.[PubMed][Google Scholar]
15. SprottH.; GlatzelM.; MichelB. A.Treatment of myositis with etanercept(Enbrel®), a recombinant human soluble fusion protein of TNF-αtype II receptor and IgG1. Rheumatology2004, 43, 524–526.[PubMed][Google Scholar]
16. LynchM.; KirbyB.; WarrenR. B.Treatingmoderate to severe psoriasis—bestuse of biologics. Expert Rev. Clin. Immunol.2014, 10, 269–279.[PubMed][Google Scholar]
17. OnestiJ. K.; GuttridgeD. C.Inflammationbased regulation of cancer cachexia. BioMedRes. Int.2014, 2014, 7.[Google Scholar]
18. PetersenA.-K.; et al. Epigenetics meets metabolomics:an epigenome-wide association studywith blood serum metabolic traits. Hum. Mol.Genet.2014, 23, 534–545.[PubMed][Google Scholar]
19. SuhreK.; GiegerC.Genetic variation inmetabolic phenotypes: study designsand applications. Nat. Rev. Genet.2012, 13, 759–769.[PubMed][Google Scholar]
20. FredrikssonT.; PetterssonU.Severe psoriasis—oraltherapy with a new retinoid. Dermatologica1978, 157, 238–244.[PubMed][Google Scholar]
21. ChambersM. C.; et al. A cross-platform toolkitfor mass spectrometry and proteomics. Nat. Biotechnol.2012, 30, 918–920.[PubMed][Google Scholar]
22. SmithC. A.; WantE. J.; O’MailleG.; AbagyanR.; SiuzdakG.XCMS: processingmass spectrometry data for metabolite profiling using nonlinear peakalignment, matching, and identification. Anal.Chem.2006, 78, 779–787.[PubMed][Google Scholar]
23. R Development Core TeamR: A languageand environment for statistical computing; R Foundation for Statistical Computing; Vienna, Austria, 2011;http://www.R-project.org/.
24. BenjaminiY.; HochbergY.Controlling the falsediscovery rate—a practicaland powerful approach to multiple testing. J.R. Stat. Soc. B1995, 57, 289–300.[Google Scholar]
25. DabneyA.; StoreyJ. D.qvalue:Q-value estimation for false discovery rate control, 2013;http://www.bioconductor.org/packages/release/bioc/html/qvalue.html.
26. WheelockA. M.; WheelockC. E.Trials and tribulationsof ’omics data analysis:assessing quality of SIMCA-based multivariate models using examplesfrom pulmonary medicine. Mol. BioSyst.2013, 9, 2589–2596.[PubMed][Google Scholar]
27. XiaJ.; MandalR.; SinelnikovI. V.; BroadhurstD.; WishartD. S.MetaboAnalyst 2.0—a comprehensiveserver formetabolomic data analysis. Nucleic Acids Res.2012, 40, W127–W133.[PubMed][Google Scholar]
28. CasteloR.; RoveratoA.Reverse engineeringmolecular regulatory networks frommicroarray data with qp-graphs. J. Comput. Biol.2009, 16, 213–227.[PubMed][Google Scholar]
29. ShannonP.; et al. Cytoscape: a softwareenvironment for integrated models of biomolecularinteraction networks. Genome Res.2003, 13, 2498–2504.[PubMed][Google Scholar]
30. LuC.; DengJ.; LiL.; WangD.; LiG.Applicationof metabolomics on diagnosis and treatment of patients with psoriasisin traditional Chinese medicine. Biochim. Biophys.Acta2014, 1844, 280–288.[PubMed][Google Scholar]
31. AbeyakirthiS.; et al. Arginaseis overactive in psoriatic skin. Br.J. Dermatol.2010, 163, 193–196.[PubMed][Google Scholar]
32. AlbinaJ.E.; MillsC. D.; HenryW. L.; CaldwellM. D.Temporal expressionof different pathways of l-arginine metabolism in healingwounds. J. Immunol.1990, 144, 3877–3880.[PubMed][Google Scholar]
33. AlbinaJ. E.; AbateJ. A.; MastrofrancescoB.Role of ornithineas a proline precursorin healing wounds. J. Surg. Res.1993, 55, 97–102.[PubMed][Google Scholar]
34. PrinzJ. C.; et al. T cell clones from psoriasisskin lesions can promote keratinocyteproliferation in vitro via secreted products. Eur. J. Immunol.1994, 24, 593–598.[PubMed][Google Scholar]
35. SchönM. P.; RuzickaT.Psoriasis: the plot thickens. Nat. Immunol.2001, 2, 91.[PubMed][Google Scholar]
36. MartinP.Wound healing—aimingfor perfect skin regeneration. Science1997, 276, 75–81.[PubMed][Google Scholar]
37. BakerE. A.; LeaperD. J.Proteinases, theirinhibitors, and cytokine profilesin acute wound fluid. Wound Repair Regener.2000, 8, 392–398.[Google Scholar]
38. MannA.; BreuhahnK.; SchirmacherP.; BlessingM.Keratinocyte-derivedgranulocyte-macrophage colony stimulating factor accelerates woundhealing: stimulation of keratinocyte proliferation, granulation tissueformation, and vascularization. J. Investig.Dermatol.2001, 117, 1382–1390.[PubMed][Google Scholar]
39. WeiL. H.; WuG.; MorrisS. M.; IgnarroL. J.Elevated arginase I expression inrat aortic smooth muscle cells increases cell proliferation. Proc. Natl. Acad. Sci. U.S.A.2001, 98, 9260–9264.[PubMed][Google Scholar]
40. deKoningH. D.; et al. Expression profile of cornified envelope structuralproteins and keratinocyte differentiation-regulating proteins duringskin barrier repair. Br. J. Dermatol.2012, 166, 1245–1254.[PubMed][Google Scholar]
41. GuvenB.; CanM.; GencM.; KocaR.Serum prolidase activity in psoriasispatients. Arch. Dermatol. Res.2013, 305, 473–476.[PubMed][Google Scholar]
42. StarodubtsevaN. L.; SobolevV. V.; SobolevaA. G.; NikolaevA. A.; BruskinS. A.[Expressionof genes for metalloproteinases (MMP-1, MMP-2, MMP-9, and MMP-12)associated with psoriasis]. Genetika2011, 47, 1254–1261.[PubMed][Google Scholar]
43. KindtE.; Gueneva-BouchevaK.; RekhterM. D.; HumphriesJ.; HallakH.Determination of hydroxyprolinein plasma and tissueusing electrospray mass spectrometry. J. Pharm.Biomed. Anal.2003, 33, 1081–1092.[PubMed][Google Scholar]
44. GarvicanE.R.; Vaughan-ThomasA.; RedmondC.; GabrielN.; CleggP. D.MMP-mediatedcollagen breakdown induced by activated protein C in equine cartilageis reduced by corticosteroids. J. Orthop. Res.2010, 28, 370–378.[PubMed][Google Scholar]
45. SchiaffinoS.; DyarK. A.; CiciliotS.; BlaauwB.; SandriM.Mechanismsregulating skeletal muscle growth and atrophy. FEBS J.2013, 280, 4294–4314.[PubMed][Google Scholar]
46. YamanakaK.; et al. Persistent release ofIL-1s from skin is associated with systemiccardio-vascular disease, emaciation and systemic amyloidosis: thepotential of anti-IL-1 therapy for systemic inflammatory diseases. PLoS One2014, 9, e104479.[PubMed][Google Scholar]
47. StraubR.H.; CutoloM.; ButtgereitF.; PongratzG.Energy regulation andneuroendocrine–immune control in chronic inflammatory diseases. J. Int. Med.2010, 267, 543–560.[Google Scholar]
48. GanL.; et al. TNF-αup-regulates protein level and cell surface expressionof the leptin receptor by stimulating its export via a PKC-dependentmechanism. Endocrinology2012, 153, 5821–5833.[PubMed][Google Scholar]
49. LIY.-P.; et al. TNF-α increases ubiquitin-conjugating activityin skeletalmuscle by up-regulating UbcH2/E220k. FASEB J.2003, 17, 1048–1057.[PubMed][Google Scholar]
50. FlorinV.; CottencinA. C.; DelaporteE.; Staumont-SalléD.Body weightincrement in patients treated with infliximab for plaque psoriasis. J. Eur. Acad. Dermatol. Venereol.2013, 27, e186–e190.[PubMed][Google Scholar]
51. SaracenoR.; et al. Effect of anti-tumor necrosis factor-αtherapies on body massindex in patients with psoriasis. Pharmacol.Res.2008, 57, 290–295.[PubMed][Google Scholar]
52. SummersG.D.; MetsiosG. S.; Stavropoulos-KalinoglouA.; KitasG. D.Rheumatoidcachexia and cardiovascular disease. Nat. Rev.Rheumatol.2010, 6, 445–451.[PubMed][Google Scholar]
53. LloveraM.; López-SorianoF. J.; ArgilésJ. M.Effectsof tumor necrosis factor-α on muscle-protein turnover in femaleWistar rats. J. Natl. Cancer Inst.1993, 85, 1334–1339.[PubMed][Google Scholar]
54. RenzoL. D.; et al. Prospective assessmentof body weight and body composition changesin patients with psoriasis receiving anti-TNF-α treatment. Dermatol. Ther.2011, 24, 446–451.[PubMed][Google Scholar]
55. PetersS.; van HelvoortA.; KeglerD.; ArgilèsJ.; LuikingY.; LavianoA.; van BergenhenegouwenJ.; DeutzN.; HaagsmanH.; GorselinkM.; van NorrenK.Dose-dependent effects of leucine supplementation onpreservation of muscle mass in cancer cachectic mice. Oncol. Rep.2011, 26, 247–254.[PubMed][Google Scholar]
56. O’ConnellT.; et al. Metabolomic analysis of cancer cachexia reveals distinct lipid andglucose alterations. Metabolomics2008, 4, 216–225.[Google Scholar]
57. Der-TorossianH.; et al. Metabolic derangementsin the gastrocnemius and the effect of CompoundA therapy in a murine model of cancer cachexia. J. Cachexia Sarcopenia Muscle2013, 4, 145–155.[PubMed][Google Scholar]
58. UbhiB. K.; et al. Targeted metabolomics identifies perturbations in amino acid metabolismthat sub-classify patients with COPD. Mol. BioSyst.2012, 8, 3125–3133.[PubMed][Google Scholar]
59. PearlstoneD. B.; et al. Effect of enteral andparenteral nutrition on amino acid levels incancer patients. JPEN, J. Parenter. EnteralNutr.1995, 19, 204–208.[PubMed][Google Scholar]
60. BlumbergD.; HochwaldS.; BurtM.; DonnerD.; BrennanM. F.Tumor necrosisfactor alpha stimulates gluconeogenesis from alanine in vivo. J. Surg. Oncol.1995, 59, 220–225.[PubMed][Google Scholar]
61. PrestonT.; et al. Fibrinogen synthesis is elevated in fasting cancer patients withan acute phase response. J. Nutr.1998, 128, 1355–1360.[PubMed][Google Scholar]
62. UbhiB. K.; et al. Metabolic profilingdetects biomarkers of protein degradation inCOPD patients. Eur. Respir. J.2012, 40, 345–355.[PubMed][Google Scholar]
63. GazzinelliR. T.; EltoumI.; WynnT. A.; SherA.Acute cerebral toxoplasmosisis induced by in vivo neutralization of TNF-alpha and correlates withthe down-regulated expression of inducible nitric oxide synthase andother markers of macrophage activation. J. Immunol.1993, 151, 3672–3681.[PubMed][Google Scholar]
64. CastilloL.; et al. Plasma arginine andcitrulline kinetics in adults given adequateand arginine-free diets. Proc. Natl. Acad. Sci.U.S.A.1993, 90, 7749–7753.[PubMed][Google Scholar]
65. KapoorS.R.; et al. Metabolicprofiling predicts response to anti-tumor necrosis factorα therapy in patients with rheumatoid arthritis. Arthritis Rheum.2013, 65, 1448–1456.[PubMed][Google Scholar]
66. TremblayF.; et al. Overactivation of S6 kinase 1 as a cause of human insulin resistanceduring increased amino acid availability. Diabetes2005, 54, 2674–2684.[PubMed][Google Scholar]