sign in sign up
<<
Home , Fludrocortisone

5 181

L S Nguyen and others Interactions between hormones 181:5 481–488 Clinical Study and HLA-G

Influence of hormones on the immunotolerogenic molecule HLA-G: a cross-sectional study in patients with congenital adrenal hyperplasia

Lee S Nguyen1,2, Nathalie Rouas-Freiss3, Christian Funck-Brentano1, Monique Leban4, Edgardo D Carosella3, Philippe Touraine4,5, Shaida Varnous2, Anne Bachelot4 and Joe-Elie Salem1,6

1Department of Pharmacology, Sorbonne Université, INSERM CIC Paris-Est, UNICO AP-HP.6 Cardio-Oncology Program, Pitié-Salpêtrière Hospital, Paris, France, 2Department of Cardiothoracic Surgery, Sorbonne Université, Institute of Cardiology, Pitie-Salpetriere University Hospital, ICAN, Paris, France, 3Department of Research in Hemato-Immunology (SRHI), CEA, Saint-Louis Institute, UMR U976, Paris, France, 4Department of Endocrinology, Pitie-Salpetriere University Hospital, Sorbonne Universite, Correspondence ICAN, Paris, France, 5Center for Rare Endocrine Disorders and center for Rare Gynecological Disorders, Paris, France, should be addressed and 6Department of Pharmacology and Medicine, Cardio-Oncology Program, Vanderbilt University Medical Center, to L S Nguyen Nashville, Tennessee, USA Email [email protected]

Abstract

Background: HLA-G is an immune checkpoint molecule, naturally expressed during , playing a critical role in the tolerance of the fetal semi-allograft from the maternal immune system. While HLA-G expression levels are associated with , the influence of other hormones is still unclear. Congenital adrenal hyperplasia (CAH) represents an adequate model to study the hormonal influence on biomarkers as it leads to impaired biosynthesis and increased progesterone and androgens production due to 21-hydroxylase enzyme deficiency. Methods: In a cross-sectional study of CAH patients matched on sex and age with healthy control, the association between circulating levels of soluble HLA-G and hormones was assessed by use of non-parametric analyses tests. European Journal of Endocrinology Multivariable linear regressions were performed on normalized data. Results: Overall, 83 CAH patients and 69 healthy controls were included. Among CAH patients, all were under and 52 (62.6%) were under supplementation. Compared to controls, CAH patients had increased HLA-G levels (15 vs 8 ng/mL, P = 0.02). In controls, HLA-G level was independently associated with progesterone and estradiol (β = 0.44 (0.35–1.27) and −0.44 (−0.94, −0.26) respectively, both P values = 0.001). In CAH patients, HLA-G level was independently associated with mineralocorticoid supplementation dosage (β = 0.25 (0.04–0.41), P = 0.001) and estradiol (β = −0.22 (−0.57, −0.02), P < 0.001). Conclusion: CAH patients had higher HLA-G levels than healthy controls. HLA-G level was positively associated with progesterone and supplementation, and negatively with estradiol. The association between mineralocorticoid, renin and HLA-G levels may suggest a role of the renin-angiotensin system in the expression of soluble HLA-G.

European Journal of Endocrinology (2019) 181, 481–488

https://eje.bioscientifica.com © 2019 European Society of Endocrinology Published by Bioscientifica Ltd. https://doi.org/10.1530/EJE-19-0379 Printed in Great Britain Downloaded from Bioscientifica.com at 09/29/2021 10:42:10PM via free access

-19-0379 European Journal of Endocrinology https://eje.bioscientifica.com study comparing early cardiovascular damage in adult NCT01807364), amulticenter prospectiveobservational (CARDIOHCS) study( in PatientsWith CongenitalAdrenalHyperplasia tothe Cardiovascular Risk Profile This study is ancillary Study design Patients andmethods HLA-G levels. relationships betweensteroidhormonesandsoluble healthy controls,weaimedtoseparatelyanalyzethe study ofCAHpatientsmatchedonsexandagewith cohortof HLA-G. Thus, in a prospective observational supplementation hadaninfluenceoncirculating levels supplementation. on glucocorticoid and insomecases mineralocorticoid production. Moreover, treatmentofCAHpatientsrelies leading toincreasedprogesteroneandandrogens cortisol biosynthesis due to 21 over biomarkers.Indeed,CAHleadstoimpaired an adequatemodeltostudythehormonalinfluence capabilities( immunomodulatory bound orinsolubleisoform,HLA-Gdisplayedthesame or solubleproteins,respectively. Whethermembrane surface or circulating in body fluids as membrane-bound on HLA-Gisunclear. HLA-Gcanbefoundatthecell ( preeclampsia in patients with lowerHLA-Gexpression progesterone, withanincreasedriskofmiscarriageand of HLA-Gandastrongassociationbetween Studies inobstetricswerethefirsttoshowexistence considered asanimmunecheckpointmolecule( propertiesandis associated withimmunomodulatory toward the fetal semi-allograft. Hence HLA-G is closely is acharacteristicofpregnancy, asituationoftolerance activating non-self-recognitioncascades.Thelatterfeature (iii) theabilitytoinduceimmunotoleranceinsteadof HLA-G transcript, and alternative splicing of the primary isoforms (solubleortransmembrane)generatedfrom 19 proteinvariants,(ii)theexistenceofseven more restricted polymorphism than other HLA, with only HLA-G isatypeIbnon-classicalmoleculefeaturing(i) coded bythemajorhistocompatibilitycomplex(MHC). Human leucocyteantigens(HLA)arecell-surfaceproteins Introduction 4 , Clinical Study 5 We hypothesizedthathormonesand corticosteroid Congenital adrenalhyperplasia(CAH)represents , 6 ). However, the influence of other hormones clinicaltrials.gov α L SNguyenandothers -hydroxylase deficiency 7 ).

identifier 1 , 2 , 3 ). Clinical InvestigationCenterParis-Est(CIC-1421,Pitie- with CAHandhealthycontrolswereallassessedatthe between May2011andAugust2015( included intheCARDIOHCSstudy(NCT01807364) matched forsex,age,andsmokingstatuswasprospectively Kremlin-Bicetre, France)). France); andKremlin-BicetreUniversityHospital(Le Hospital andSaint-AntoineUniversity(Paris, endocrinology departments(Pitie-SalpetriereUniversity than 18years,referredbyoneofthreeuniversityhospital Inclusion criteriawerepatientswithCAH,agedmore Study populationandsetting study wasapprovedbyeachhospitalethicscommittee. provided writteninformedconsenttoparticipate,andthe deficiency andhealthycontrolsubjects. Allpatients men andwomenwithCAHdueto21 mineralocorticoid supplementation) wascomputedas the mineralocorticoideffect (hereafternamedtotal . equivalentand 0.0154timesthedosageof supplementation) wascomputedasthesumofdosage of glucocorticoid effect(hereafternamedtotal Some ofthemreceivedfludrocortisoneinassociation. standard ofcare,withhydrocortisoneordexamethasone. All patientswithCAHweretreated,asneededbytheir CAH group Exposure tocorticosteroidsupplementationinthe patients withCAH. discontinuationwaspromotedin subjects, andtemporary previous monthwasanexclusioncriterioninhealthy Estradiol and/orprogesteronecontraceptioninthe ofcardiovascular disease andpregnancy.known history or non-classicalform( presentation types: classic salt-wasting, simple virilizing the study). physical examination,standardbloodsamplingrelatedto fullmedicalcheck-up(medicalhistory,All underwent gene. Healthycontrolswererecruitedviaspecializedads. childhood andconfirmedbygenetictestingof patients hadbeenidentifiedpriortothestudy, during Salpetriere UniversityHospital,Paris,France).AllCAH and HLA-G Interactions betweenhormones For every patientwithCAH,ahealthycontrol For every Dosage offludrocortisone equivalentregarding Dosage of hydrocortisone equivalent regarding the Exclusion criteria for CAH andhealthy subjects were CAH patientswerecategorizedinoneofthreeclinical 11 Downloaded fromBioscientifica.com at09/29/202110:42:10PM ). 181 8 :5 , 9 , α -hydroxylase 10 ). Patients CYP21A2 482 via freeaccess

European Journal of Endocrinology associations weretestedusing Spearmancorrelations. parametric testsduetodata distribution.Consequently, (proportions). Comparisons were performed using non- (interquartile range)andcategorical variablesasnumber Hence, continuous variables are presented as median normally distributedwereheightandbloodpressure. omnibus normalitytests,theonlyvariableswhichwere Using Shapiro–Wilk andD’AgostinoPearson Statistical analysis using ELISAhasbeenvalidatedpreviously( detection is1 and theresultswereexpressedasng/mL.Thelimit of purified from M8-HLA-G5 cell line culture supernatant, (12.5–200 levels weredeterminedfromafive-pointstandardcurve detection antibodies, respectively ( -G5) andanti-human recognizes the mostabundantsoluble isoforms (sHLA-G1, soluble HLA-GwasquantifiedusingMEM-G/9,which HLA-G wasanalyzedinplasmaofpatientsasfollows: HLA-G analysis respectively). and RIAdelta4–Androstenedione;Beckman-Coulter© DIAsource© ImmunoAssays, Louvain-la-Neuve, Belgium 4 androstenedionebyradioimmunoassay(KIP1409; (Liaison XL; Diasorin©) and 17-OH progesterone, delta (Modular-E170; Roche©),reninbychemiluminescence ACTH, FSH,LHwereassayedusingchemiluminescence Hospital. Estradiol,progesterone,SHBG,testosterone, ofPitie-SalpetriereUniversity immunology laboratory tube and further assayed in the were collected in a dry stimulating hormone(FSH),luteinizing(LH), (SHBG), adrenocorticotropichormone(ACTH),follicle- 4 androstenedione,sexhormone–bindingglobulin progesterone, estradiol,totaltestosterone,renin,delta serum concentrationsof,progesterone,17-OH product foratleast24 for atleast12 in themorningandpatientswererequiredtobefasting All bloodsampleswerecollectedatonefixedtimepoint Hormone analyses dosage ofhydrocortisoneequivalent( the sumofdosagefludrocortisoneandfivetimes Clinical Study Blood samplesforthedeterminationof ng/mL) using dilutions of calibrated HLA-G5 ng/mL. ThemethodologytomeasureHLA-G h andtorefrainfromanycoffee-containing h priortobloodsampling. β 2-microglobulin ascaptureand L SNguyenandothers 13 12 ). Total sHLA-G ). 14 ). and healthycontrolswereadequatelymatchedsimilar study, 83ofwhomhadCAH(54.6%).Patientswith One hundredfifty-twosubjectswereincludedinthis Overall population Results analyses andfigureswereexecutedusingSPSS25.0(IBM). 0.05 and type II errorrate of 0.2.Allcalculations, statistical achieve statisticalsignificance,givenatypeIerrorrateof 0.35, aminimumof62patientswouldberequiredto a correlation between progesterone and HLA-G levels of probability ofF andremovalcriteriaa a stepwisemodelwithentry Multivariable linearregressionwasperformedusing data werenormalizedusinglogarithmictransformation. than inhealthycontrols(8( CAH (medianvalue:15(interquartile range:4–25)ng/mL) bioavailable andfree),ACTH andrenin( progesterone and17-OHprogesterone, testosterone(total, to healthycontrols,CAHpatientshadincreasedlevels of than healthycontrols(25.6vs23.2, compared to3(4.3%)inhealthycontrols. Two (2.4%)CAHwomenwereundermenopauseas men under (for testicular adrenal rest tumors). on combinedoralcontraceptivewithdesobelandtwo progestin ( and one chlormadinone), one progestin onlypills(),twoonsequential treatmentforanti-androgenicproperty, twoon receiving othersteroids:sixwomenwereoncyproterone under corticoid treatment. Fifteen patients (18.1%) were daily. Incomparison,noneofthehealthycontrolswas supplementation, mediandosagewas200( in the 52 patients (62.6%) under mineralocorticoid was 21(interquartilerange:15–32)mgdailyand tumors. Mediandosageofglucocorticoidsupplementation CAH men ( normal androstenedionelevels(below4 patients, 50(60.2%)maybeconsideredcontrolledwith form and26(31.3%)werenon-classical.AmongCAH classic salt-wastingform,16(19.3%)thesimplevirilizing were nomissingdata. including hormonelevelsarepresentedin in ageandsex.Clinicalbiologicalcharacteristics and HLA-G Interactions betweenhormones Circulating HLA-Glevelswerehigherinpatientswith CAH patientshadhigherbody-massindex(BMI) Among theCAHpatients,41(49.4%)presented In ordertoperformmultivariablelinearregressions, n

= 25), 9 (36%) had testicular adrenal rest < .5 and 0.05 Downloaded fromBioscientifica.com at09/29/202110:42:10PM > 3 0.10, respectively. Assuming – 22 https://eje.bioscientifica.com ) ng/mL)( 181 P .2. Compared =0.02). :5 Table 1 ng/mL). Among P Table 1 =0.02). 75 ). –245) µg . There 483 via freeaccess European Journal of Endocrinology https://eje.bioscientifica.com rgseoe (rho =0.25, progesterone In healthycontrols, HLA-G levelscorrelated with Relationship betweenhormones andHLA-G on Table17-OH progesterone(seeSupplementary 1,seesection cortisol, ACTH,totaltestosterone,estradiol,FSH,LHand (rho = (rho = progesterone (rho =0.27, renin supplementation dosagewaspositivelycorrelatedwith (rho = P correlated withrenin(correlationcoefficientrho Glucocorticoid supplementationdosagewaspositively CAH patients Relationship betweentreatmentandhormonesin low-density lipoprotein;LH,luteinizinghormone;SBP,systolicbloodpressure;SHBG,sexhormone-bindingglobulin. C-reactive protein;DBP,diastolicbloodpressure;FSH,follicle-stimulatinghormone;HDL,high-densitylipoprotein;HLA,humanleukocyteantigen; LDL, 17-OH progesterone,17-hydroxyprogesterone;ACTH,adrenocorticotropichormone;BMI,bodymassindex;CAH,congenitaladrenalhyperplasia; CRP, HLA-G (ng/mL) Cortisol (µg/L) Renin (pg/mL) ACTH (pg/mL) Free androgenindex SHBG (ng/mL) Total testosterone(ng/mL) 17-OH progesterone(ng/mL) Androstenedione (ng/mL) LH (IU/mL) FSH (IU/mL) Progesterone (ng/mL) Estradiol (pg/mL) Triglycerides (mmol/L) LDL cholesterol(mmol/L) HDL cholesterol(mmol/L) Total cholesterol(mmol/L) Glycemia (mmol/L) CRP (u/L) DBP (mmHg) SBP (mmHg) BMI Weight (kg) Height (cm) Age (years) Women variables. interquartile range(firstquartile:Q1,thirdQ3)forcontinuousvariablesandnumber(proportion)categorical Table 1

= Clinical Study 0.009) and negatively correlated with androstenedione supplementary data supplementary There were no correlations between treatment and There werenocorrelationsbetweentreatmentand − − 0.35, 0.34, Clinical andbiologicalcharacteristicsofCAHhealthycontrolpatients.Dataarepresentedasmedian P =0.001). P

= − 0.016) andnegativelycorrelatedwith 0.23, P .0) Mineralocorticoid =0.001). givenattheendofthisarticle). P P .3) n androstenedione and =0.038)

= 0.04) and estradiol levels L SNguyenandothers Median 23.23 97.3 11.8 2.32 58.1 0.36 1.39 0.72 2.45 1.55 4.45 0.71 27.8 111 170 2.2 5.2 4.8 0.8 4.5 19 59 69 69 49 8 Healthy controls (71%) 28.4 0.55 1.24 3.91 20.7 23.4 65.2 11.1 1.01 36.3 0.26 0.86 103 163 1.7 4.3 3.5 0.5 2.1 4.3 0.4 8.8 Q1 63 61 3 = 0.29, ( n = 69) 25.85 44.23 0.92 1.89 4.95 15.8 140 118 178 138 8.8 7.6 1.2 2.8 4.7 1.4 4.5 2.2 Q3 72 77 35 21 29 79 (0.04–0.41), ( with mineralocorticoid supplementation ( showed thatHLA-Glevelswere independentlyassociated ACTH, renin, andglucocorticoids) gender, estradiol, androstenedione,testosterone,cortisol, Table 3 and androstenedionelevels(rho (rho =0.23, renin (rho =0.25, supplementation (rho =0.27, mineralocorticoid supplementation glucocorticoid P 0.35–1.27) and coefficient with progesterone and estradiollevels(adjusted beta and renin)showedthatHLA-Glevelswereassociated estradiol, androstenedione,testosterone,cortisol,ACTH regression (accountingforage,gender, progesterone, (rho = and HLA-G Interactions betweenhormones 3 − .0) (see values =0.001) 0.57, In CAHpatients,HLA-Glevelscorrelatedwith − ). Multivariable linear regression (accounting for age, ). Multivariable linearregression (accounting for age, 0.25, − Median 25.64 0.02), 0.87 2.76 1.48 4.56 0.53 27.4 21.7 31.1 6.04 48.7 0.81 14.3 110 162 108 2.8 5.2 4.8 4.4 47 68 67 58 15 2 β

= P P

0.44 (95% confidence interval for 0.44 (95% confidence interval β = =0.04; = 0.001) andwithestradiol ( CAH patients P P − .3) etail (rho = estradiol =0.036),

< 0.44 ( 0.001)(see Fig. 1 (69.9%) Downloaded fromBioscientifica.com at09/29/202110:42:10PM Table 2 21.97 0.62 2.29 1.14 4.04 22.6 63.7 13.8 10.4 1.59 30.1 0.26 − 103 156 1.1 2.7 0.7 4.2 0.3 4.3 Q1 31 59 57 ). 3 4 0.94, ( n = 83) − ). Multivariablelinear Fig. 1 0.26), respectively; both = − 181 30.23 0.23, 1.13 3.35 1.79 5.46 1.39 28.6 42.7 81.9 117 123 168 143 6.3 7.7 6.5 6.4 4.7 2.9 75 74 36 25 89 49 Q3 ). Asadditional :5 − P 0.3, .4 (see =0.04) < < < < P P P β P value 0.02 0.16 0.001 0.003 0.03 0.19 0.001 0.001 0.33 0.91 0.88 0.001 0.27 0.02 0.02 0.53 0.5 0.21 0.46 0.88 0.95 0.02 1 0.001 1 0.88 = =0.03), =0.01), β =0.01) =0.25 − 484 0.22 via freeaccess β :

European Journal of Endocrinology 17-OH progesterone,17-hydroxyprogesterone;ACTH,adrenocorticotropichormone;FSH,follicle-stimulatingLH,luteinizingSHBG, sex hormone-bindingglobulin. P (rho =0.26, renin analysis: drug, associationsremainedunchangedinunadjusted of patients under anticontraceptive or antiandrogen sensitivity analysis,inCAHpatients,afterexclusion Renin ACTH FAI SHBG Total testosterone 17-OH progesterone Androstenedione LH FSH Progesterone Estradiol correlations). Table 2 patients ( mineralocorticoid supplementation andHLA-GlevelsinCAH − levels inCAHpatients( P in healthycontrols( P in healthycontrols( levels. Topleft:scatterplotofprogesteroneandHLA-Glevels Association ofsolubleHLA-Gwithhormonesandsteroids Figure 1 = 0.001); bottomleft:scatterplot ofestradiolandHLA-G = 0.001); topright:scatterplotofestradiolandHLA-Glevels .3, nrseein (rho =-0.30, androstenedione =0.03), 0.02), Clinical Study P

< n Association betweenHLA-G,hormonesandcorticosteroidsupplementationinhealthycontrols( = 83) (adjusted 0.001) andbottomright:Tukey plotof n n = 68) (adjusted = 68) (adjusted n = 83) (adjusted β = 0.25 (0.04–0.41), P Correlation coefficient .3, srdo (rho = estradiol =0.03), Healthy controlsall − − − − L SNguyenandothers β β 0.04 0.04 0.1 0.07 0.13 0.15 0.06 0.12 0.16 0.25 0.25 = = 0.44 (0.35–1.27), β − = 0.44 ( − 0.22 ( P − = 0.001). 0.94, − ( 0.57, n = 69) P P − 0.76 0.76 0.4 0.59 0.29 0.21 0.65 0.33 0.2 0.04 0.04 value =0.01), 0.26), − 0.26, Correlation coefficient After adjustment,theonlyremainingindependent (rho =0.34, supplementation mineralocorticoid (rho =0.35, supplementation glucocorticoid and coronary endothelial cells, witha reversible and coronary could induce expression ofHLA-G by cardiomyocytes and notably in cardiac transplantation, progesterone described previouslyinobstetrics ( ( statistically significantinnon-parametriccomparison between thetwogroupswashoweversufficientto be similar outerboundvaluesinbothgroups.Difference close betweenthetwogroups;hence,patientspresented as highthatofhealthycontrols, interquartile rangewas median valueofHLA-GlevelsinCAHpatientswastwice controls issurprising.Ithastobenotedthatwhilethe express morecirculating HLA-Gthanmatchedhealthy was examinedinCAHpatients,andthefindingthatthey was independentlyassociatedwithHLA-Glevels. mineralocorticoid supplementation,butonlythelatter correlated withglucocorticoidsupplementationand renin levelsand(iv)inCAHpatients,HLA-Gwere estradiol levels;(iii)HLA-Glevelswereassociated to were independently associated with progesterone and healthy controls;(ii)incontrols,HLA-Glevels patients expressmorecirculating HLA-Gthanmatched The mainfindingsofthisstudyareasfollows:(i)CAH Discussion ( association waswithmineralocorticoidsupplementation and HLA-G Interactions betweenhormones P β .4 (0.11–0.53), =0.34 =0.02). Healthy men − − − − − − − − − − The linkbetweenprogesterone andHLA-Glevelswas To ourknowledge,thisisthefirsttimethatHLA-G 0.08 0.24 0.29 0.06 0.25 0.22 0.19 0.21 0.2 0.08 0.26 ( n = 20) P 0.75 0.32 0.21 0.79 0.29 0.35 0.43 0.37 0.41 0.74 0.27 value P =0.004). Downloaded fromBioscientifica.com at09/29/202110:42:10PM Correlation coefficient https://eje.bioscientifica.com Healthy women 5 − − − − n , 0.02 0.13 0.15 0.027 0.19 0.24 0.11 0.07 0.11 0.34 0.25 = 68) (non-parametric 6 181 , 15 :5 ). Inotherfields, P 0.003) and =0.003) ( n P = 49) =0.004). P 0.89 0.39 0.31 0.85 0.18 0.1 0.46 0.66 0.45 0.02 0.09 value 485 via freeaccess European Journal of Endocrinology https://eje.bioscientifica.com both glucocorticoid and mineralocorticoid andmineralocorticoids. Unsurprisingly, positively associatedwithsubstitutive treatment,mainly to beconfirmedbythepresentwork( estradiol comparedwithprogesterone,whichwouldtend and mayevenbeinterpretedasanantagonisteffect of expression ofHLA-Ginadipose-derivedstemcells, demonstrate a beneficial effect of estradiol toward the hormones areelevated.Experimentalstudiesfailed to progesterone. Duringearlyphasesofpregnancy, both between estradiolandHLA-Gislessclearthanwith mild progestative effect (below 1%). The relationship very or the factithasa correlations whichwereobserved), of 17-OH-progesterone, as attestedbythenon-significant (with a non-linear effect of corticosteroid effect on levels by thecounter-regulationeffectofsubstitutivetreatment 17-OH-progesterone levels inourcohortmaybeexplained lack ofcytotrophoblasts(i.e.inmen). understand howHLA-Gmaybeincreasedeveninspiteof endothelial andsmoothmusclecells)whichmayhelp throughout heart wall and vessels (vascular observed expression. Incidentally, thispromoteractivitywas enhancement ofbothsteady-statemRNAandprotein receptor ( progesterone-responsive element,boundbyanactivated progesterone stimulation goes through apromoter, the first-trimester cytotrophoblasts,expressionofHLA-Gvia ( whenaddingprogesteroneantagonist process observed ACTH FAI SHBG Total testosterone 17-OH progesterone Androstenedione LH FSH Estradiol Mineralocorticoid Glucocorticoid correlations). Table 3 sex hormone-bindingglobulin. 17-OH progesterone,17-hydroxyprogesterone;ACTH,adrenocorticotropichormone;FSH,follicle-stimulatingLH,luteinizing SHBG, Renin Clinical Study supplementation (µg) supplementation (mg) In CAHpatients,soluble HLA-Glevelswere The lackofassociationbetweenHLA-Gand Association betweenHLA-G,hormonesandcorticosteroidsupplementationinCAHpatients( 6 ). Response is dose dependent with maximal 16 ). InJEG-3choriocarcinoma celllineand Correlation coefficient L SNguyenandothers − − − − − − 17 CAH all 0.23 0.03 0.14 0.06 0.13 0.08 0.23 0.05 0.10 0.30 0.25 0.27 ). ( n = 83) P 0.04 0.80 0.19 0.56 0.23 0.50 0.04 0.65 0.39 0.01 0.03 0.01 value Correlation coefficient supplementation wereassociatedwithplasmarenin. explain twoaspectsofimmunomodulation inCAH inCAHpatientsthiscohortdonot HLA-G observed be investigatedfurtherinaspecific experimentalsetting. renin-angiotensin system,HIFandHLA-Gwouldneed to although therelationshipbetweenmineralocorticoids, present withincreasedsolubleHLA-G( failure withreducedejectionfractionandhypertension) 26 cells tohypoxia-responsive elements (namely HIF-2)( previous experimentalstudieslinkedrenin-producing factor 1alpha expression reliesonthemetabolicsensor areas ( or during invasion of metastatic tumor cells in hypoxic implantation phase in the early placenta development hypoxia, whichnaturallyoccursduringthetrophoblastic the wholecohort. ( propertiesassociatedtothissystem immunomodulatory renin-angiotensin system (RAS), through means of of HLA-Gexpressionmayalsobeinfluencedbythe mineralocorticoids (i.e.fludrocortisone),promotion a five-foldhighermineralocorticoideffectthanpure mineralocorticoids hasnot. been demonstratedpreviously( While theeffectofglucocorticoidsonHLA-Glevelshas and HLA-G Interactions betweenhormones 19 ). Moreover, diseasesinwhichRASisactivated(i.e.heart , − − − Clinical consequencesofthe increaseofcirculating Hence, RASmaybeassociatedwithHLA-GthroughHIF, On theonehand,HLA-Gexpressionisinfluencedby Because glucocorticoids(i.e.hydrocortisone)have -0.65 -0.05 -0.32 20 0.27 0.003 0.34 0.21 0.03 0.15 0.31 0.08 0.11 CAH men 21 ). Incidentally, reninwasassociatedwithHLA-Gin , 22 ). Thishypoxia-relatedmodulationofHLA-G (HIF-1)signaling( ( n = 25) P 0.19 0.51 1 0.31 0.76 0.83 0.12 0.87 0.46 0.14 0.71 0.59 value Downloaded fromBioscientifica.com at09/29/202110:42:10PM Correlation coefficient 23 18 n , = 83) (non-parametric − − − − − − − − 24 ), thespecificeffectof CAH women 0.23 0.11 0.059 0.065 0.15 0.098 0.16 0.029 0.18 0.13 0.1 0.16 ). Ontheotherhand, 181 27 :5 hypoxia-inducible , 28 ). ( n = 68) P 0.08 0.42 0.66 0.63 0.27 0.46 0.23 0.83 0.18 0.35 0.48 0.22 value 486 25 via freeaccess , European Journal of Endocrinology controls. Association with progesterone was confirmed CAH patientshadhigher HLA-G levelsthanhealthy Conclusion nature ofthestudygroupsisacknowledged. nature ofthesefinding,duetothepurelyobservational substitutive treatmentonHLA-G.Finally, the exploratory contributed tomorepreciseresultsregardingtheeffects of in the course of the CARDIO-HCS study but may have mass spectrometeranalysiswasnotroutinelyperformed of hormoneassay(i.e.),forwhich,atandem discussed earlier. Anotherlimitationliesin themethod the associations which resisted multivariate analyses were into mineralocorticoideffect.To adjustforthese,only Biases includetheoverlappingeffectofglucocorticoids factors whichmayalterhormonelevelsinCAHpatients. explained by the important number of confounding and hormonesweremoderate,althoughthismaybe the results.Moreover, associationsbetweenHLA-G any dynamic analysis which may have strengthened were onlyavailableatonetimepoint,whichexcludes not assayedtobecorrectlyinterpreted. bias of being under the influence ofprogestative effect, HLA-G inCAHpatientswouldhavebeensubjectedtothe relationship betweenendogenousprogesteroneleveland drugs with progestative effect. Therefore, studying the proportion were under contraceptive pills or antiandrogen not thecaseofCAHpatients,inwhom,asubstantial were excludedifundercontraceptive pills, whichwas analyzed withoutbiasgiventhathealthycontrolwomen Analysis ofprogesteroneinCAHpatientscouldnotbe Limitations identifying patientsathigherrisk. with increasedriskofmortality, HLA-Gmaybekeyto chronic exposuretocorticosteroidtreatmentisassociated the contrary, legitimatesfutureinvestigations.Given does notrefutethefindingsofpresentstudy, buton overexpression may have an incidence in CAH patients in CAHpatients.ThelackofclinicalproofthatHLA-G available ontheriskofrejectionafterorgantransplantation imputable tocorticoidsupplementation( however, haveanincreasedrateofgestationaldiabetes, particularly withregardtopreeclampsiaevents;theydo with differentoutcomesascomparedtogeneralpopulation, patients. Duringpregnancy, CAHwomendonotpresent Clinical Study Other study limitations include the fact that samples L SNguyenandothers 29 ). Nodataare the public,commercialornot-for-profitsector. This research did not receive any specific grant from any funding agency in Funding be could that interest of conflict perceived asprejudicingtheimpartialityofthisstudy. no is there that declare authors The Declaration ofinterest EJE-19-0379 at paper the of version online the to linked is This Supplementary data expression maybesuggested. Renin-angiotensin systeminvolvementinHLA-G supplementation (glucocorticoidandmineralocorticoid). levels includedestradiol,reninandcorticosteroid in healthycontrols.OtherfactorsassociatedwithHLA-G References and Salenave S Leger, J Dulon, Polak, J M including Gallo, N Bourcigaut.Theyarealsoextremelygratefultothestudyparticipants. A group, Bruckert, study E Dahmoune, CARDIOHCS N the thank authors The Acknowledgments critical revisiontothemanuscript. provided and study the supervised Salem the J-E and to Bachelot A insight manuscript. critical provided Varnous S and Touraine P Carosella, D E Leban, M Funck-Brentano, C Rouas-Freiss, N manuscript. the analyses wrote statistical and performed study, ancillary the designed Nguyen S L Author contributionstatement and HLA-G Interactions betweenhormones 6 5 4 3 2 1 Yie SM, Li LH,Li GM,Xiao R&Librach CL.Progesterone enhances Szekeres-Bartho J, Polgar B,Kozma N,Miko E,Par G,Szereday L, Szekeres-Bartho J, Varga P, Kinsky R&Chaouat G.Progesterone- Rouas-Freiss N, Goncalves RM,Menier C,Dausset J&Carosella ED. Rebmann V, daSilvaNardi F, Wagner B &Horn PA. HLA-G Ferreira LMR, Meissner TB,Tilburgs T &Strominger JL.HLA-G:atthe (https://doi.org/10.1093/humrep/dei305 cytotrophoblasts invitro. HLA-G geneexpressioninJEG-3choriocarcinoma cells andhuman 118–125. immunomodulation. Barakonyi A, Palkovics T, Papp O&Varga P. Progesterone-dependent org/10.1016/0923-2494(90)90139-P Research inImmunology mediated immunosuppressionandthemaintenanceofpregnancy. 11520–11525. fetus frommaternaluterinenaturalkillercytolysis. Direct evidencetosupporttheroleofHLA-Ginprotecting org/10.1155/2014/297073 ofImmunologyResearchJournal as atolerogenicmoleculeintransplantationandpregnancy. 272–286. interface ofmaternal-fetaltolerance. . (https://doi.org/10.1159/000087953 (https://doi.org/10.1016/j.it.2017.01.009) (https://doi.org/10.1073/pnas.94.21.11520 Chemical ImmunologyandAllergy 1990 Human Reproduction ) Downloaded fromBioscientifica.com at09/29/202110:42:10PM 141 2014 175–181. ) 2014 https://eje.bioscientifica.com Trends inImmunology ) 297073. 181 (https://doi. https://doi.org/10.1530/ 2006 ) :5 (https://doi. PNAS 21 2005 46–51. 1997 ) 2017 89 487

94 38 via freeaccess

European Journal of Endocrinology https://eje.bioscientifica.com

15 17 16 14 13 11 10 12 9 8 7 Clinical Study Moslehi A, Hashemi-Beni B,Akbari MA&Adib M.The Sheshgiri R, Rao V, Tumiati LC, Xiao R,Prodger JL,Badiwala M, Ragusa A, deCarolis C,dalLago A,Miriello D,Ruggiero G, Rebmann V, Lemaoult J, Rouas-Freiss N,Carosella ED&Grosse- Menier C, Saez B,Horejsi V, Martinozzi S,Krawice-Radanne I,Bruel S, El-Maouche D, Arlt W&Merke DP. Congenitaladrenalhyperplasia. Salem JE, Nguyen LS,Hammoudi N,Preud’homme G,Hulot JS, Goodman L, Gilman A, Brunton L, Lazo J & Parker K. Goodman & Goodman L,GilmanA,BruntonLazoJ&ParkerK. Abehsira G, Bachelot A,Badilini F, Koehl L,Lebot M,Favet C, Rosenbaum D, Gallo A,Lethielleux G,Bruckert E,Levy BI, Carosella ED, Rouas-Freiss N,Tronik-Le Roux D,Moreau P, kjpp.2016.20.4.341 and Pharmacology HLA-G inadiposederivedstemcells. effect ofprogesteroneand17-betaestradiolonmembrane-bound CIRCULATIONAHA.107.757781 cells. antigen-g expressioninvascularendothelialandsmoothmuscle Librach C &Delgado DH.Progesteroneinduceshumanleukocyte 07oa Lupus Progesterone supplementinpregnancy:animmunologictherapy? Brucato A, Pisoni MP, Muscara M,Merati R,Maccario L (https://doi.org/10.1016/j.humimm.2005.05.003) HLA-G inEssen,2004. Wilde H. Reportofthewetworkshopfor quantificationofsoluble 8859(02)00821-2 Immunology analyze theexpressionofnonclassicalHLAclassImolecules. monoclonal antibodiesrecognizingHLA-GorHLA-E:newtoolsto Le Danff C,Reboul M,Hilgert I,Rabreau M McGraw-Hill. Brunton L. Gilman’s thepharmacologicalbasisoftherapeutics.2006NewYork: 6736(17)31431-9 Lancet org/10.1016/j.echo.2017.10.017 Society ofEchocardiography function: insightintosexualdimorphism. Complex associationofsexhormonesonleftventricularsystolic Leban M, Funck-Brentano C,Touraine P, Isnard R,Bachelot A (https://doi.org/10.1210/jc.2016-1877 ofClinicalEndocrinologyand Metabolism Journal duration. gonadotropins andsexsteroidhormonesonQTinterval Touraine P, Funck-Brentano C&Salem JE.Complexinfluenceof (https://doi.org/10.1097/HJH.0000000000001850 21-hydroxylase deficiency. Early centralbloodpressureelevationinadultpatientswith Tanguy ML, Dulon J,Dahmoune N,Salem JE,Bittar R bs.ai.2015.04.001 in Immunology LeMaoult J. HLA-G:animmunecheckpointmolecule. ) Circulation 2004 2017 2003 13 390 2015 639–642. 2008 2194–2210. ) ) 2016 ) 64 ) 127 315–326. 118 Human Immunology 20 33–144. The Pharmalogical BasisofTherapeutics The Pharmalogical (https://doi.org/10.1191/0961203304lu20 2018 341–346. S58–S64. Journal ofHypertension Journal (https://doi.org/10.1016/S0140- (https://doi.org/10.1016/s0198- 31 ) ) (https://doi.org/10.1016/ 231–240.e1. (https://doi.org/10.1161/ (https://doi.org/10.4196/ Korean Journal ofPhysiology Korean Journal L SNguyenandothers ) 2005 Journal oftheAmerican Journal et al. 2016 (https://doi. Characterizationof 66 2019 ) 853–863. 101 et al. Advances 37 2776–2784. et al. 175–181.

Human et al

, 2006. . Accepted 9September2019 Revised versionreceived5September2019 Received 20May2019

and HLA-G Interactions betweenhormones 25 24 23 22 21 19 18 29 28 27 26 20 Kurt B, Gerl K,Karger C,Schwarzensteiner I&Kurtz A.Chronic Satou R, Penrose H & Navar LG. Inflammation as a regulator of the Satou R, Penrose H&Navar LG.Inflammationasaregulatorofthe Garziera M, Scarabel L&Toffoli G. HypoxicmodulationofHLA-G Mouillot G, Marcou C, Zidi I,Guillard C,Sangrouber D,Carosella ED Kilburn BA, Wang J, Duniec-Dmuchowski ZM,Leach RE,Romero R& Rodesch F, Simon P, Donner C&Jauniaux E.Oxygenmeasurements Capettini LS, Montecucco F, Mach F, Stergiopulos N,Santos RA Moreau P, Faure O,Lefebvre S,Ibrahim EC,O’Brien M, Hagenfeldt K, Janson PO,Holmdahl G,Falhammar H,Filipsson H, Garcia-Gonzalez IJ, Valle Y, Rivas F, Figuera-Villanueva LE, Munoz- Olesen LL &Hviid TV. UpregulationofsolubleHLA-Ginchronicleft Stefanska A, Eng D,Kaverina N,Pippin JW, Gross KW, Duffield JS& juxtaglomerular renincellsintofibroblast-likeinvivo. hypoxia-inducible transcriptionfactor-2activationstablytransforms doi.org/10.1155/2017/4587520 expression throughthemetabolicsensorHIF-1inhumancancer org/10.1016/j.humimm.2006.10.016 tumor cells. & Moreau P. HypoxiamodulatesHLA-Ggeneexpressionin org/10.1095/biolreprod62.3.739 cell line. the expressionofHLA-Gandintegrinsinahumantrophoblast Armant DR. Extracellularmatrixcompositionandhypoxiaregulate Obstetrics andGynecology in endometrialandtrophoblastictissuesduringearlypregnancy. Reports renin-angiotensin systemandbloodpressure. (https://doi.org/10.2174/138161212799436593 immunity andaging. & daSilva RF. Roleofrenin-angiotensinsystemininflammation, org/10.1016/s0041-1345(01)01990-x Transplantation Proceedings hormones upregulatelevelsofHLA-Gtranscriptsintrophoblasts. Gourand L, Dausset J,Carosella ED&Paul P. Glucocorticoid (https://doi.org/10.1093/humrep/den118 21-hydroxylase deficiency. outcome inwomenwithcongenitaladrenalhyperplasiadueto Frisen L, Thoren M&Nordenskjold A.Fertilityandpregnancy 2014 syndrome andtype2diabetesmellitus. polymorphism isrelatedwithhighbloodpressureinacutecoronary Rodriguez NO &Padilla-Gutierrez JR.The14bpDel/InsHLA-G Valle JF, Flores-Salinas HE,Gutierrez-Amavizca BE, Davalos- 2016 ventricular systolicdysfunction. 2016 2alpha followingexperimentalureteralobstruction. Shankland SJ. Cellsofreninlineageexpresshypoxiainduciblefactor org/10.1681/ASN.2013111152 of theAmericanSocietyNephrology cells. 8417190. Journal ofImmunologyResearchJournal 17 2014 2018 5. Biology ofReproduction 898159. (https://doi.org/10.1186/s12882-015-0216-0 Human Immunology 20 100–100. (https://doi.org/10.1155/2016/8417190 (https://doi.org/10.1155/2014/898159 Current Pharmaceutical Design Current Pharmaceutical 1992 (https://doi.org/10.1007/s11906-018-0900-0 2001 Downloaded fromBioscientifica.com at09/29/202110:42:10PM Human Reproduction ) 80 2000 ) 2007 ) 33 Journal ofImmunologyResearchJournal 283–285. 2277–2280. 2015 2017 ) ) 62 68 739–747. BioMed Research International 277–285. ) 26 2017 181 587–596. Current Current Hypertension 2008 4587520. ) :5 (https://doi. (https://doi. (https://doi. 2012 BMC Nephrology 23 (https://doi. ) ) 1607–1613. 18 (https:// ) 963–970. Journal Journal 488 2016 via freeaccess

)