Zurich Open Repository and
Archive
University of Zurich
Main Library
Strickhofstrasse 39
CH-8057 Zurich
www.zora.uzh.ch
Year: 2017
Interdependence of thyroglobulin processing and thyroid hormone export in
the mouse thyroid gland
Weber, Jonas ; McInnes, Joseph ; Kizilirmak, Cise ; Rehders, Maren ; Qatato, Maria ; Wirth, Eva K ;
Schweizer, Ulrich ; Verrey, François ; Heuer, Heike ; Brix, Klaudia
Abstract: Thyroid hormone (TH) target cells need to adopt mechanisms to maintain sufficient levels of
TH to ensure regular functions. This includes thyroid epithelial cells, which generate TH in addition to
being TH-responsive. However, the cellular and molecular pathways underlying thyroid auto-regulation
are insufficiently understood. In order to investigate whether thyroglobulin processing and TH export
are sensed by thyrocytes, we inactivated thyroglobulin-processing cathepsins and TH-exporting
monocar-boxylate transporters (Mct) in the mouse. The states of thyroglobulin storage and its protease-mediated
processing and degradation were related to the levels of TH transporter molecules by immunoblotting
and immunofluorescence microscopy. Thyroid epithelial cells of cathepsin-deficient mice showed increased
Mct8 protein levels at the basolateral plasma membrane domains when compared to wild type controls.
While the protein amounts of the thyroglobulin-degrading cathepsin D remained largely unaffected by
Mct8 or Mct10 single-deficiencies, a significant increase in the amounts of the thyroglobulin-processing
cathepsins B and L was detectable in particular in Mct8/Mct10 double deficiency. In addition, it was
observed that larger endo-lysosomes containing cathepsins B, D, and L were typical for Mct8- and/or
Mct10-deficient mouse thyroid epithelial cells. These data support the notion of a crosstalk between
TH transporters and thyroglobulin-processing proteases in thyroid epithelial cells. We conclude that a
defect in exporting thyroxine from thyroid follicles feeds back positively on its cathepsin-mediated
prote-olytic liberation from the precursor thyroglobulin, thereby adding to the development of auto-thyrotoxic
states in Mct8 and/or Mct10 deficiencies. The data suggest TH sensing molecules within thyrocytes that
contribute to thyroid auto-regulation.
DOI: https://doi.org/10.1016/j.ejcb.2017.02.002
Posted at the Zurich Open Repository and Archive, University of Zurich
ZORA URL: https://doi.org/10.5167/uzh-139734
Journal Article
Published Version
The following work is licensed under a Creative Commons: Attribution-NonCommercial-NoDerivatives
4.0 International (CC BY-NC-ND 4.0) License.
Originally published at:
thyroglobu-lin processing and thyroid hormone export in the mouse thyroid gland. European Journal of Cell Biology,
96(5):440-456.
DOI: https://doi.org/10.1016/j.ejcb.2017.02.002
ContentslistsavailableatScienceDirect
European
Journal
of
Cell
Biology
j ou rn a l h o m epa g e :w w w . e l s e v i e r . c o m / l o c a t e / e j c b
Research
paper
Interdependence
of
thyroglobulin
processing
and
thyroid
hormone
export
in
the
mouse
thyroid
gland
Jonas
Weber
a,
Joseph
McInnes
a,1,
Cise
Kizilirmak
a,
Maren
Rehders
a,
Maria
Qatato
a,
Eva
K.
Wirth
b,
Ulrich
Schweizer
c,
Francois
Verrey
d,
Heike
Heuer
e,
Klaudia
Brix
a,∗ aJacobsUniversityBremen,DepartmentofLifeSciencesandChemistry,CampusRing1,D-28759Bremen,GermanybCharité-UniversitätsmedizinBerlin,InstitutfürExperimentelleEndokrinologie,AugustenburgerPlatz1,D-13353Berlin,Germany cUniversitätBonn,InstitutfürBiochemieundMolekularbiologie,Nußallee11,D-53115Bonn,Germany
dUniversitätZürich,PhysiologischesInstitut,Winterthurerstrasse190,CH-8057Zürich,Switzerland
eIUF–LeibnizInstitutfürumweltmedizinischeForschung,Auf’mHennekamp50,D-40225Düsseldorf,Germany
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:
Received22December2016
Receivedinrevisedform8February2017 Accepted9February2017 Key-words: Cathepsins Endo-lysosomes Mct8 Mct10 Thyroidauto-regulation Thyroidstates Thyroxine
Translocationacrossmembranes
a
b
s
t
r
a
c
t
Thyroidhormone(TH)targetcellsneedtoadoptmechanismstomaintainsufficientlevelsofTHto
ensureregularfunctions.Thisincludesthyroidepithelialcells,whichgenerateTHinadditiontobeing
TH-responsive.However,thecellularandmolecularpathwaysunderlyingthyroidauto-regulationare
insufficientlyunderstood.InordertoinvestigatewhetherthyroglobulinprocessingandTHexportare
sensedbythyrocytes,weinactivatedthyroglobulin-processingcathepsinsandTH-exporting
monocar-boxylatetransporters(Mct)inthemouse.Thestatesofthyroglobulinstorageanditsprotease-mediated
processinganddegradationwererelatedtothelevelsofTHtransportermoleculesbyimmunoblottingand
immunofluorescencemicroscopy.Thyroidepithelialcellsofcathepsin-deficientmiceshowedincreased
Mct8proteinlevelsatthebasolateralplasmamembranedomainswhencomparedtowildtypecontrols.
Whiletheproteinamountsofthethyroglobulin-degradingcathepsinDremainedlargelyunaffectedby
Mct8orMct10single-deficiencies,asignificantincreaseintheamountsofthethyroglobulin-processing
cathepsinsBandLwasdetectableinparticularinMct8/Mct10doubledeficiency.Inaddition,itwas
observedthatlargerendo-lysosomescontainingcathepsinsB,D,andLweretypicalforMct8-and/or
Mct10-deficientmousethyroidepithelialcells.Thesedatasupportthenotionofacrosstalkbetween
THtransportersandthyroglobulin-processingproteasesinthyroidepithelialcells.Weconcludethata
defectinexportingthyroxinefromthyroidfolliclesfeedsbackpositivelyonitscathepsin-mediated
prote-olyticliberationfromtheprecursorthyroglobulin,therebyaddingtothedevelopmentofauto-thyrotoxic
statesinMct8and/orMct10deficiencies.ThedatasuggestTHsensingmoleculeswithinthyrocytesthat
contributetothyroidauto-regulation.
©2017TheAuthors.PublishedbyElsevierGmbH.ThisisanopenaccessarticleundertheCC
BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Thyroiddysfunctionscausedbyalteredthyroidhormone(TH)2 generationordisturbedTHactionareconsideredtobeamongthe mostfrequentendocrinedisorders(Brixetal.,2011;Führeretal.,
∗ Correspondingauthor.
E-mailaddress:[email protected](K.Brix).
1 Presentaddress:VIBCenterforBrainandDiseaseResearchandKULeuven
DepartmentforHumanGenetics,3000Leuven,Belgium.
2 Abbreviations:CMF-PBS,calcium-andmagnesium-freephosphatebuffered
saline;Cy3,indocarbocyanine;HPT,hypothalamus-pituitary-thyroid;Mct, mono-carboxylatetransporters;PTC,papillarythyroidcarcinoma;Tg,thyroglobulin;TH, thyroidhormones;TRH,thyrotropin-releasinghormone;TSH,thyroidstimulating hormone;WT,wildtype.
2015).THareofcrucialimportancefortheproperfunctioningof nearlyeveryorgan,includingthethyroidglanditself.Hence,allTH targetcellsneedtoadoptmechanismstomaintainsufficientlevels ofTHtoensureregularfunctions.Sincethethyroidglanditselfnot onlyproducesTHbutalsorespondstoTHlevelsinaself-sustaining manner,auto-protectionofthethyroidglandisvitalnotonlyfor thyrocytesbutalsoforallotherTH-dependentcells.Properthyroid statesofanorganismarecentrallyregulatedbythe hypothalamus-pituitary-thyroid(HPT)–axis,wherebylowserumTHlevelstrigger hypothalamicthyrotropin-releasinghormone(TRH)releasewhich thenpromotesreleaseofthyroidstimulatinghormone(TSH)from thepituitary(Fliersetal.,2014),therebytriggeringTHgeneration inthethyroidgland,andTHreleaseintotheblood.
Tofulfillandmaintainitstasks,thethyroidglandiscomposed offunctionalunits,theso-calledthyroidfollicles,thatarebuiltby http://dx.doi.org/10.1016/j.ejcb.2017.02.002
Table1
SpecificfunctionsofdifferentcathepsinsexpressedinthethyroidregardingtheirprocessingofTg. extracellularTg solubilisationfrom cross-linked storageforms extracellularTg processing intracellular limitedTg processing (endosomal) intracellularTg degradation (lysosomal) directT4liberation fromTg exhaustiveTg degradationforTg utilization (endo-lysosomalTH liberation) References
CathepsinB x x x x involvedtomajor
extent
Brixetal.,1996;Dunn etal.,1996;Tepeletal., 2000;Friedrichsetal., 2003;Jordansetal., 2009
CathepsinD x involvedtominor
extent
Brixetal.,1996;Dunn etal.,1996;Friedrichs etal.,2003
CathepsinK x x x Tepeletal.,2000;
Friedrichsetal.,2003; Jordansetal.,2009
CathepsinL x x x x involvedtomajor
extent
Brixetal.,1996;Dunn etal.,1996;Tepeletal., 2000;Friedrichsetal., 2003;Jordansetal., 2009
Fig.1. Cathepsin-mediatedsolubilization,processing,anddegradationofTg.
SchematicdrawingsummarizingthesequentialstepsofTgprocessinganddegradationbyendo-lysosomalcathepsins(Brixetal.,1996;Tepeletal.,2000;Brixetal.,2001; Linkeetal.,2002a;Friedrichsetal.,2003;Jordansetal.,2009;Dauthetal.,2011).CathepsinsBandLinvolveinextracellularTgsolubilizationfromitscovalentlycross-linked storageforms,whilecathepsinKisinstrumentalinTgutilizationforT4liberation.Uponre-internalizationofintactandpartiallydegradedTg,itisfurtherprocessedand degradedbyasparticandcysteinecathepsinswithinendo-lysosomesofthyrocytes.
amonolayerofthyroidepithelialcells(Fujita,1988;Nilssonand Fagman,2013).ThemacromolecularTHprecursorthyroglobulin (Tg)is storedin covalently cross-linkedformin the extracellu-larthyroidfolliclelumen(Herzogetal.,1992;Berndorferetal., 1996;Kleinetal.,2000;Saber-Lichtenbergetal.,2000).Covalently cross-linkedTgisprocessedforsolubilizationanddegradationby proteases,whicharesecretedinaTSH-regulatedfashionintothe extra-cellular folliclelumenupon a demand of TH inthe body periphery(forreviews,seeBrixetal.,2001;Dauthetal.,2011). Thus,solubilizationofTgfromitscovalentlycross-linkedstorage formsisenabledthroughlimited,extracellularproteinprocessing facilitatedbycathepsinsBandL,whileTHliberationfromTgis mediatedbycathepsinsKandL,andalsoinitiatedinthethyroid folliclelumen(Brixetal.,1996;Linkeetal.,2002a;Friedrichsetal., 2003;Tepeletal.,2000).Subsequently, partiallydegradedTgis internalizedandreachesendo-lysosomalcompartments,whereT3
andT4liberationiscompletedforTgutilizationthroughexhaustive
proteolysismediatedbyavarietyofhydrolyticenzymesincluding asparticandcysteinecathepsinsB,D,K,L,andS(Brixetal.,1996; Tepeletal.,2000;Linkeetal.,2002a;Friedrichsetal.,2003;Jordans etal.,2009).CathepsinKisspecialamongtheTg-processing pro-teasesasitisabletoliberateT4directlyfromTg(Tepeletal.,2000).
Inbriefsummary,thyroidfunctionsareenabledbysequential pro-teolyticprocessingofTg(Table1,Fig.1),resultingintheliberation ofT4and,toalowerextent,T3.
Considering that thyroid epithelial cells need to release TH throughtransmembranetranslocationintothebloodcirculation, andkeepinginmindthatthyrocytesareTHtargetcellsthemselves, aregulatorymechanismmustexistthatdetermineshowmuchof theT4 liberated fromTgis eitherretainedin thethyrocytesor
transportedbackintothesame.Weproposethatintra-thyroidal THtransporterslikethemonocarboxylatetransportermolecules Mct8andMct10mayactassuchsensorsof(i)theextentof pro-teolyticTHliberation,(ii)theamountsofTHreleasedfromthyroid follicles,and(iii)thelevelsofre-uptakeofcirculatingTHbackinto thyroidfollicles.
Mice with a global Mct8-deficiency (Mct8 knock-out mice) exhibitcentralTH-resistance,i.e.TRHlevelsaredramatically ele-vateddespite highlyincreasedserum T3 levels,whereas serum
Table2
Thyroidfunctionparameters,peripheralandintra-thyroidalthyroidstates,THgenerationandTHreleaseinMct8/10-deficientmice.
Mct8−/y Mct10−/− Mct8−/y/Mct10−/− References
Thyroidfunctionparameters,HPTaxis
• serumT4 ↓ = = Dumitrescuetal.,2006;Trajkovicetal.,2007;DiCosmoetal.,2010;
Liaoetal.,2011;Wirthetal.,2011;Trajkovic-Arsicetal.,2010b; Mulleretal.,2014
• serumT3 ↑↑↑ = ↑↑↑ Dumitrescuetal.,2006;Trajkovicetal.,2007;DiCosmoetal.,2010;
Liaoetal.,2011;Wirthetal.,2011;Trajkovic-Arsicetal.,2010b; Mulleretal.,2014
• serumTSH,orTSH mRNA(pituitary)
↑ = ↑ Dumitrescuetal.,2006;Trajkovicetal.,2007;Trajkovic-Arsicetal., 2010b;Liaoetal.,2011;Wirthetal.,2011;Mulleretal.,2014
• TRHmRNA (hypothalamus)
↑↑ = = Trajkovicetal.,2007;Trajkovic-Arsicetal.,2010b;Mulleretal.,2014
peripheralDio1activity (liver,kidney)
↑ = ↑↑ Dumitrescuetal.,2006;Trajkovicetal.,2007;Liaoetal.,2011; Trajkovic-Arsicetal.,2010a;Trajkovic-Arsicetal.,2010b;Mulleretal., 2014
intra-thyroidalthyroidstates • intra-thyroidalTH • (non-Tg-boundand
Tg-bound)
↑↑ = ↑↑↑ DiCosmoetal.,2010;Trajkovic-Arsicetal.,2010b;Liaoetal.,2011; Mulleretal.,2014
• intra-thyroidalDio1 activity
= DiCosmoetal.,2010;Wirthetal.,2011;Mulleretal.,2014
• auto-thyrotoxicstate,i.e. papillarystructureswith nuclearchanges;or alteredtumourmarker likeGalectin-3
↑↑ = ↑↑↑ Wirthetal.,2011;Mulleretal.,2014
• thyroidfollicleand/or luminalareas
↑ = ↑↑ DiCosmoetal.,2010;Trajkovic-Arsicetal.,2010b;Wirthetal.,2011; Mulleretal.,2014
• thyroidepithelial extensions
↑ DiCosmoetal.,2010;Wirthetal.,2011
THgeneration • Nis(sodiumiodide
symporter),Ttf1 (transthyretin)
↑notsignifi-cant = ↑↑ DiCosmoetal.,2010;Mulleretal.,2014
• Tshr(TSHreceptor),Tpo (thyroidperoxidase),Tg (thyroglobulin),Dehal1 (dehalogenase1)
= = ↑↑ DiCosmoetal.,2010;Trajkovic-Arsicetal.,2010b;Mulleretal.,2014
THreleasefromthyroidfollicles • TSH-stimulatedT4
release
↓ DiCosmoetal.,2010;Trajkovic-Arsicetal.,2010b
• TSH-stimulatedT3 release ↑ Trajkovic-Arsicetal.,2010b • Mct10,Lat1,Lat2(TH transporters) = DiCosmoetal.,2010
2014).Interestingly,micedeficientinbothMct8andMct10 pro-teins(Mulleretal.,2014)showedapartialrescueoftheunusual serumTHstatusobservedinMct8-deficientanimals,whereasthe intra-thyroidalauto-thyrotoxicstateswereevenmorepronounced thaninMct8 deficiencyalone.Therefore, bothMct8 andMct10 appeartoberesponsiblefor THexportfromthyrocytes(Muller etal.,2014).
Theastonishingchangesintissue-specificTHstatesinmicewith THtransporter-deficienciesindicatethatTHlevelsareregulated inanorgan-specific manner,andnot entirelygovernedby sys-temicfactorssuchasregulationthroughtheHPTaxis.Withregard tothethyroidglanditself,suchorgan-specificmechanismsmust compriseofintra-thyroidalself-regulatorypathwaystomanageits thyroidstate.Inthisstudy,pathwaysofthyroidauto-regulation werestudiedinMct8-and/orMct10-deficientmalemicein
com-parisontowildtype(WT)controls.Inaddition,Ctsk−/−malemice
were used, because they provide a suitable animal model for analyzingintra-thyroidalpathwaysofthyroidauto-regulation,in particularwithregardtoTgproteolysis(Dauthetal.,2011).Serum TH levelsare not altered in cathepsin Kdeficiency, and hence, TSHlevelsarelikelynotenhancedinthismousemodel,despite developinganenlargedthyroidglandandfeaturingchangesin Tg-processingabilitiesinthatcathepsinLisup-regulated(Friedrichs etal.,2003).ThismeansthatCtsk−/−miceshowsignsofthyroid
2. Materialsandmethods 2.1. Animals
All studies were performed with 10–14 weeks and 20–28 weeksoldmaleCtsk−/−,Mct8−/y,Mct10−/−,Mct8−/y/Mct10−/− or
WT C57Bl/6J mice. The mice were back-crossed to a congenic C57Bl/6JbackgroundintheanimalfacilityofJacobsUniversity Bre-men,Germany,licencedunderSfAFGJSAz.513-30-00/2-15-32and 522-27-11/3-1,05-A20andA21.Thegenerationandgenotyping proceduresofmousestrainslackingMct8(Trajkovicetal.,2007; Wirthetal.,2009),Mct10(Mariottaetal.,2012),Mct8andMct10 (Muller etal.,2014), orCtsk(Saftig etal.,1998)weredescribed previously.Ctsb−/−,Ctsk−/−,Ctsl−/−,andCtsb−/−/Ctsk−/−malemice
werebredand housedaspreviouslydescribed(Friedrichsetal., 2003).Ingeneral,micewerehousedunderstandardconditions, witha12h/12hlight/darkcyclewithlightsoutat07:00p.m.and adlibitumwaterandfood.
2.2. Tissuesamplingandsectioning
MiceweredeeplyanesthetizedbyCO2inhalationorbyusingthe
inhalation-anestheticIsoFlou(AbbottGmbH&Co.KG, Wiesbaden-Delkenheim,Germany)beforeintra-peritonealinjectionof100l of10%ketamine(CEVASanteAnimale,Düsseldorf,Germany).The abdominal and thoracic cavities wereopenedand the abdomi-nalaortawascut. Perfusion wascarriedout through theheart with0.9%NaClsupplementedwith200IUheparin(Braun Mel-sungenAG,Melsungen,Germany).Thethyroidwasimmediately removed,whilestill attachedtothetrachea,either snap-frozen inliquidnitrogen,orincubatedin4%PFAin200mMHEPES,pH 7.4,and leftovernightat4◦C. Then,cryo-preservation was
car-riedoutbyembeddinginTissueFreezingMedium(Jung,through LeicaMicrosystems,Wetzlar,Germany)overnight,beforetissue wasfrozen in thegasphase of liquid nitrogenand stored at -80◦Cuntilsectioning.Thyroidsampleswerethenmountedonthe
sectioningtableauofthecryotome,andsectionsof5mwere pre-paredat−20◦C(LeicaMicrosystems).
2.3. Indirectimmunofluorescence
Embedding material was washed out of the tissue sections withPBSpriortothestainingprocedure.Blockingwasperformed with3%bovineserumalbumin(AlbuminFractionV,Roth, Karl-sruhe, Germany) for 60min at 37◦C. Samples were incubated
overnightwithspecificantibodiesdilutedin0.1%BSAin calcium-andmagnesium-freePBS(CMF-PBS)at4◦Cinamoisturized
cham-ber.Specificantibodieswererabbitanti-Mct8(HPA003353;1in 250ATLASAntibodies,Stockholm,Sweden),sheepanti-cathepsin L (06-0008; 1 in 20, RD antibodies through Merck-Millipore, Darmstadt,Germany),goatanti-cathepsinB(GT15047;1in200, Neuromics, through Acris, Himmelstadt, Germany), sheep anti-cathepsinD(06-0003; 1in50, Bio-Ass,Diessen, Germany),and rabbitanti-bovineTg(Brixetal.,1998).Secondaryantibodiesand counter-stainswereincubatedfor60minat37◦C.Assecondary
antibodies,indocarbocyanine-(Cy3)orAlexa488-conjugatedgoat anti-rabbitIgG,donkeyanti-sheepand rabbitanti-goatIgG,and AlexaFluor546coupledtodonkeyanti-sheepwereused.Draq5TM
(DR05500, 1 in 500, BioStatus, Leicestershire, UK) wasused as counter-staintovisualizenuclearDNA.
2.4. Imageacquisitionofmousethyroidtissueanddensitometry ConfocalimagesweretakenusingaZeissLSM510METAlaser scanningmicroscopeequippedwithArgonandHelium-Neonlasers (CarlZeissJenaGmbH,Jena,Germany).Opticalsectionsweretaken with a pinhole opening of 1 Airy unit and at a resolution of
1024×1024pixelsor2048×2048,usingLSM5software(version 3.2;CarlZeissGmbH).Forquantitationofcysteinecathepsinand Mct8levels,thyroidtissuefromatleastthreedifferentanimalswas analyzedsuchthatimagesfromtissuesectionswereobtainedat thesamedetectorgain andpinholesettings.Thenumberof fol-liclesanalyzedisgivenintherespectivefigurelegends.Byusing CellProfilersoftware,signalsofimmunostainingintensitieswere quantifiedandnormalizedtothenumberofcells by enumerat-ingthecorrespondingDraq5TM-positivenuclei(Kamentskyetal., 2011).
2.5. Proteinextraction
Thyroidtissue sampleswere homogenizedand lysedin ice-coldPBScontaining0.5%TritonX-100,1mMEDTA,andprotease inhibitors(0.2g/mlaprotinin,Sigma-Aldrich,1153;EDTA;10M E-64, Enzo Life Science, Farmingdale, NY, USA, ALX 260007-M005;1MpepstatinA,Sigma-Aldrich,77170).Afterclearingat 10,000×gand4◦Cfor10min,thelysateswerestoredat−20◦C.The
proteincontentofallsampleswasdeterminedusingtheNeuhoff assay(Neuhoffetal.,1979)asdescribedbefore(Arampatzidouetal., 2012).
2.6. Gelelectrophoresisandimmunoblotting
Tissue lysates were normalized to equal amounts of pro-tein,heatedat95◦Cinsamplebufferandseparatedonself-cast
12.5% acrylamide gels or horizontal SDS Gradient 8–18 Excel-Gelgels(GEHealthcare,Uppsala,Sweden).Insomeexperiments, proteinsoftissuelysateswereseparatedundernon-reducing con-ditions.Gelswereeitherstainedfortotalproteinbysilverstaining (HeukeshovenandDernick,1988),orseparatedproteinswere fur-thertransferred onto a nitrocellulosemembrane by a semi-dry blottingprocedure(Kyhse-Andersen,1984).
Immunodetectionwasperformedwithprimaryantibodies rab-bitanti-ß-tubulin(ab6046,1in500,Abcam),goatanti-cathepsinB (GT15047;1in1000,NeuromicsthroughAcris,Herford,Germany), rabbitanti-cathepsinD(IM16;1in80,CalbiochemthroughMerck Millipore),goatanti-cathepsinL(GT15049;1in1000,Neuromics through Acris),rabbit anti-Mct8(HPA003353; 1in 2500ATLAS Antibodies),andrabbitanti-bovine Tg(W.S.3,Brixetal.,1998) appliedfor1.5h(see,Friedrichsetal.,2003).Therespective sec-ondaryantibodies,HRP-conjugatedgoatanti-rabbitIgG(H+Lsp.) andrabbitanti-goatIgG(H+Lsp.)wereappliedfor1hatroom tem-perature.VisualizationwasperformedbychemiluminescenceECL WesternBlottingsubstrateontoXPosureTMfilm(Piercethrough
ThermoScientific,Schwerte,Germany),whichwerescannedusing a transmittedlightscannerdevice.Densitometryanalyseswere performedbyusingImageJasdescribedelsewhere(TanandNg, 2008;Gassmannetal.,2009).Sampleswerenormalizedtoequal amountsofproteinandbandswerenormalizedtoWTcontrolsor toß-tubulinloadingifnotstatedotherwise.
2.7. Statisticalanalyses
Levelsofstatisticalsignificanceofnon-pairedobservationswere calculatedbytwo-tailedStudent’st-testorone-wayANOVAwith Dunnett’scorrectionformultiple comparisonsusingMicrosoft® Excel® 2016(MicrosoftCorp.)orGraphPadPrismTM (GraphPad
SoftwareInc.,SanDiego,CA,USA). 3. Results
tissueof Mct8-and Mct10-double-deficientmice(Mulleret al., 2014)can,in principle,beexplainedby theinabilityof thyroid folliclestoproperlyreleaseTHintothebloodstreamdueto lack-ingMct8-and/orMct10-mediatedtransportacrossthebasolateral plasmamembranedomainofthyroidepithelialcells,leadingtoTH accumulationwithinthesecells.Inaddition,reducedT4levelsin
thebloodstreamstimulateTSHproduction,inprinciple,possibly affectingTgsolubilizationandutilizationbymeansofitsextra-and intracellularproteolysismediatedbythecysteinecathepsins.Thus, alteredTgturnoveranddegradationmayfurthercontributetothe enhancedlevelsof THdetectableinthyrocytesofMct8-and/or Mct10-deficientmice.Therefore,thestatusofTgcross-linkingand degradationwascorrelatedtotheamountsofTg-processing pro-teasesinthyroidtissueofTHtransporter-deficientmice,andvice versa,theexpressionandlocalizationoftheTHtransporterMct8 wasdeterminedinthyroidtissueofcathepsin-deficientmice. 3.1. AlteredthyroglobulinstatusinMct8-and/orMct10-deficient mousethyroidtissue
TheTgstatusofMct8and/orMct10-deficientthyroidglandswas analyzedmorphologicallybystainingthyroidtissuesectionswith antibodiesdirectedagainstTg.LuminalTgdepositionsare hetero-geneousbutshowthetypicalmultilayeredappearance(Berndorfer etal.,1996;Friedrichsetal.,2003)(Fig.2).Regionsrepresenting sol-ubilizedTgcanbeidentifiedbyhigherimmuno-stainingintensities, becausetheseTgformsaremoreeasilyaccessible,whereas com-pact,covalentlycross-linkedformsofTgarestainedlessintense bytheantibodies(Friedrichsetal.,2003).Whilecovalently cross-linkedformsofTgareprevalentinthecentreoffolliclelumina,the solubleandmoreeasilyaccessibleTgformsaretypicallydetected nexttotheapicalplasmamembrane(Fig.2,panelsAandB,WT;cf. panelC).ThedifferencesinappearanceofluminalTg-stainingthus dependontheaccessibilityofTgfortheantibodies,andweused thateffecttoquantifytheproportionoffolliclescontainingtightly compactedTg(Fig.2C,a,black),multilayeredTg(Fig.2C,b,grey) andhighlysolubilized,multilayeredTgapicallywithdense, com-pactTginthecentreofthatfollicle(Fig.2C,c,lightgrey;diamonds). Manualcountingrevealedthattherewasatrendofadecline infolliclecountswithtightlycompactedTginMct8−/y/Mct10−/−
mice(60.8±5.7%)comparedtotheWT(87.4±2.5%),whileno sig-nificantdifferenceswereobservedinMct10−/−mice(72.0±11.3%)
andinMct8−/ymice(69.4±7.7%)(Fig.2,barchartsbeneathpanel
B).Additionally,follicleswithperipheralregionsofwell-accessible TgwerenotdetectedinWT(0.3%)orMct10−/−mice(0.0%),but4.3%
ofallfolliclesinMct8-deficientmicecontainedthoseTg appear-anceswhichwereevenmorepronouncedinMct8/Mct10-deficient animals(9.0%)(Fig.2,barchartsbeneathpanelB).Thus,asdeduced fromantibodyaccessibility,Tgstatesarealteredwithregardtothe well-accessibleTgappearancesintheperi-cellularregionatthe apexofthyrocytes,inparticularinMct8−/yandinMct8−/y/Mct10−/−
mice,indicatingmoresolubilizedTgthaninWTanimals(Fig.2, panelsAandB,cf.Mct8−/yandMct8−/y/Mct10−/−withWT).
3.2. ThyroglobulindegradationisenhancedinMct8-and/or Mct10-deficientmice
TofurtherinvestigatetheeffectsofTHtransporterdeficiencyon thestatusoftheTHprecursorTg,weanalyzedwholethyroidlysates preparedfromyoung(<2.5month,seeFig.3)andadult(5–8month, seeFig.4)mice.
Immunoblots revealed an increase of soluble Tg fragments (<330kDa)inyoungMct8−/y/Mct10−/−micecomparedtoallother
genotypes(Fig.3AandB),andlessabundantmultimericformsofTg insilver-stainedSDS-gelsofthesamegenotype(Fig.3C),indicating enhancedsolubilizationofTginMct8-andMct10-double-deficient
animals.Thepresence ofmore Tgfragments pointstowardsan enhancedutilizationoftheTHprecursorTgbythyrocytesupon deficienciesin Mct8and Mct10.Sincea decreaseofmono- and dimericTgformsinMct8−/y/Mct10−/−miceincomparisonto
Mct8-deficientmicewasdetectedunderreducingconditions(Fig.3D), andatthesametime,thesolubledimericformofTgwasincreased whenthyroidslysateswereseparatedundernon-reducing condi-tions(Fig.3C),itmustbeassumedthatTgcross-linkingisalteredin thyroidfolliclesofyoungMct8-andMct10-double-deficientmice. Toconfirm thisinterpretation,adult animalswereanalyzed. InaccordancewiththeTgstatusobservedinyounganimals(see
Fig. 3), significantly enhanced amounts of Tg fragments were detectedinbothMct8-andMct8/Mct10-deficientadultmicewhen the protein lysates were separated under reducing conditions (Fig.4A).However,itwasfoundthattheeffectofincreaseddimeric TginMct8/Mct10-deficiencyof younganimals wasnot present anymore when thyroid lysates of adult animalswere analyzed undernon-reducingconditions(Figs.4B,cf.3C).Moreover,Tg mul-timerizationandtheamountsofdimericTgwereobservedtooccur tocomparableextents inthyroidtissue lysatesofallgenotypes analyzedundernon-reducingconditions(Fig.4B),whileless di-andmonomericTgwasobservedinMct8-and/orMct10-deficient thyroidtissuelysatesincomparisontoWTwheninspectedunder reducingconditions(Fig.4C).
Thus, gross degradation of Tg is enhanced in Mct8- and/or Mct10-deficientmiceatallages,whilecross-linkingforTg stor-ageisaffectedprimarilyinMct8−/y/Mct10−/−miceatanageofup
to2.5months,butbyfarlessprominentlyalteredin5–8monthold mice.Theseresultsindicatethatintra-thyroidalregulationofTg turnoveraffectsenzymesinvolvedincross-linkingofTg(forreview, seeBrixetal.,2001)toalowerdegreethantheproteasesinvolved inextra-andintracellularprocessingofTgforitssolubilizationand utilization,eventuallyresultinginTHliberation.
3.3. Tg-processingproteasesaredifferentiallyaffectedbyMct8 and/orMct10deficiencies
Toinvestigatewhethertheobserved alterationsin Tgstatus mightbeduetoalteredlevelsofproteolyticenzymesweanalyzed theproteinlevelsofthecathepsinsB,D,andLinthyroidtissueof youngandadultanimals.Inyoungmice(<2.5month),proteinlevels ofcathepsinBweresignificantlyincreasedin Mct8/Mct10-double-deficientanimals,only(Fig.5).Inadultmice(5–8months),the proteinamountsofcathepsinsBandLinthyroidlysateswere sig-nificantlyincreasedinMct8-and/orMct10-deficiencies(Fig.6,top andbottompanels),whereascathepsinDproteinlevelswere sig-nificantlyincreasedinMct8/Mct10-double-deficientanimals,only (middlepanel)whencomparedtotheWT.
Sincethelocalization ofproteases,andthereforetheir possi-bleaccesstothesubstrateTg,definestheirimpactonmaintaining thyroidfunction,we usedimmunohistochemistrytoinvestigate whethertheintra-follicularandsubcellularlocalizationpatterns ofthecathepsinsB,D,andLchangedinTHtransporterdeficiency. Thyroidtissuecryo-sectionsfromWTorMct8/10-deficient ani-malswere immunolabeledwithantibodies againstcathepsin B. Inallgenotypesinspected,astainingpatternoflargelyendocytic cathepsin B, and, in few follicles,staining in thefolliclelumen aswellasattheapicalplasmamembranedomainofthyrocytes was observed (Fig. 7A–D). Moreover, the appearance of thyro-cyteswithenlargedcathepsinBpositivevesiclesinMct8−/y was
detected(Fig.7G,arrows), which waseven moreprominent in Mct8−/y/Mct10−/− mice (Fig. 7H, arrows). However, the overall
cathepsinBsignalintensitywasincreasedonlyslightlyinMct8−/y
andMct8−/y/Mct10−/−micewhencomparedtotheWT(Fig.7J).
Fig.2. AssessmentofTgstatesinthyroidfolliclesofMct8/Mct10-deficientmice.
Confocalfluorescenceimagesofcryo-sectionsofthyroidglandsfromadultWTorTHtransporter-deficientmice(5–8monthsold)oftheindicatedgenotypesimmunolabeled withantibodiesagainstTg.TgwasdetectedinmultilayeredforminthefolliclelumenofWTmicewithincreasedlabellinginproximitytotheapicalplasmamembrane domainofthyrocytes,thusindicatingsolubleformsofTg(A,asterisks).AbundanceoffollicleswithoutprominentlabellingofsolubleTg(C,a)wasdecreasedin Mct8-andMct8/Mct10-doubledeficientanimals(AandB).Likewise,theproportionoffollicleswithintenselyimmuno-labeledsolubleTg(C,b)isincreasedinMct8-deficient animalsandevenmoreincreasedinMct8/Mct10-doubledeficientanimals.Interestingly,4.3%and9.0%ofallanalyzedfolliclesofMct8orMct8/Mct10-deficientanimals, respectively,featuredhighlysolubilizedTg(C,c)intheluminalperipherybutadensemassofcovalentlycross-linkedTginthecentralregionoflargefollicles(B,diamond). A:representative,overviewimagesusedforquantification,n=250,72,190,and127follicles,respectively.B:close-upimagesdisplayingfollicleswithdifferentformsofTg. BarchartsinBandC:black(a),percentageoffollicleswithoutlabeledsolubleTg;grey(b),percentageoffollicleswithlabeledsolubleTg;lightgrey(c),percentageoffollicles withhighlysolubilizedTggivenasmeans.Scalebars:100m.
signalintensityofcathepsinDstainingwaselevatedinthyroid tis-sueofMct10−/−,Mct8−/yandMct8−/y/Mct10−/−mice(Fig.8BandF,
CandG,DandH,respectively)whencomparedtotheWT. Interest-ingly,cathepsinDpositivevesiclesinMct8-and/orMct10-deficient miceappeared largerthan those of therespectiveWT controls andinsomecases,vesicleswithprominentlyenhanceddiameters werefound(Fig.8F–H,cf.E,dashedarrows).Astrikingdifference betweenMct8−/yandMct8−/y/Mct10−/− micewastheoccurrence
ofcathepsinDstainingatthebasalsurfaceofthyrocytes,which couldeitherbedue tocathepsin Dstainingin endothelialcells (Fig.8G,asterisks),secretionofcathepsinDatthebasalpole,and itsre-associationwiththebasolateralplasmamembranedomains ofthyrocytes(arrows),orboth.
InWTanimals,cathepsinLislocalisedinvesicularstructures, butinfewfollicles,associationwiththeapicalplasmamembrane domain of thyrocytes was also observed (Fig. 9A). In
Mct10-deficientmice,thelocalizationofcathepsinLdidnotchangewhen comparedtotheWTcontrol,howeveraslightincreaseofsignal intensitywasmeasuredindicatingelevatedproteinexpression lev-els(Fig.9BandH).InMct8−/yandMct8−/y/Mct10−/−mice,cathepsin
Lwasdetectedintracellularlywithinvesicularstructuresthatwere muchlargerthanthoseinWTtissue,insomecasesreaching diam-eters comparableof thenucleioftherespectivecells, i.e. upto 6m(Fig.9CandD,EandF,E’andF’,arrows).Additionally, local-izationofcathepsinLattheapicalplasmamembranedomainof Mct8−/y/Mct10−/− thyrocyteswasincreasedandluminalstaining
nexttotheapicalplasmamembranedomainwasmuchmore abun-dantthanintheWT(Fig.9F,cf.E,arrows).
Fig.3.Thyroglobulindegradationpatternandcross-linkagein<2.5monthsoldmice.
WholethyroidlysatesofWTmiceormicedeficientinMct10,Mct8and/orMct10wereseparatedonhorizontalSDS-gels,silver-stainedortransferredtonitrocellulosefor subsequentincubationoftheblotswithantibodiesagainstTgasindicated,andcomparedtotheWT.(A)Shownisarepresentativeimmunoblotwiththeindicatedgenotypes. (B)LowmolecularmassfragmentsofTgarerepresentativeofdegradationproductswhichwererevealedbydensitometrytobesignificantlyincreasedinMct8/Mct10-deficient animals.(C)Undernon-reducingconditions,theamountsofthesolubleformofTgareincreasedinMct8/Mct10-deficientanimals,whereasthemultimericTgformswereless abundantcomparedtotheWT,Mct10-andMct8-deficientanimals.Mct8-deficientmicewereobservedtohavemoretotalTgamountsincludingthemultimericform.Upon reducingsampleswithDTTbeforeSDS-PAGE,thepersistenceofdimericTgwasreducedinMct8/Mct10-deficientanimals(E),thusindicatingalteredsolubilizationofTg, whereasMct8-deficientanimalsfeaturedincreaseddimericandmonomericTgwhencomparedtotheWT.Animalsanalyzed:n=5,3,5,and5,respectively.InB,means±SD aregiven;*,P<0.05;**,P<0.005.
levelswereenhancedinthyroidtissueofallgenotypes,andmost prominentlyinMct8−/y/Mct10−/−mice,whichwerefurther
charac-terizedbyenhancedamountsofcathepsinD.Thus,thedatapointto mostsignificantlyenhancedamountsofendo-lysosomalcathepsin LinparticularinMct8−/yandMct8−/y/Mct10−/−mousethyrocytes,
i.e.intheanimalmodelsfeaturingthemostsevereintra-thyroidal thyrotoxicstates(Mulleretal.,2014).ConsideringcathepsinL’s main function in Tg processing, the enhanced amounts of this protease in Mct8−/y/Mct10−/− thyroid tissue explain thelargely
enhanced extent of Tg degradation observed in Mct8/Mct10-deficientmice(Figs.3and4),andsuggestthatT4accumulatesin
thyrocytesoftheseanimals.Thus,theenhancedamountsof cathep-sinL,whichisalsovitalforT4liberationfromTg,inthyroidsofmice
lackingtheTH-transportingMct8,whichismainlyresponsiblefor T4releasefromthyroidfollicles,iscounterintuitive.
3.4. AlackoftheT4-liberatingcysteinecathepsinKisassociated
withenhancedamountsofthemainT4-exportingtransporter
Mct8
Thecysteine cathepsins Kand L together arecritical for TH liberation(seeTable1;Friedrichsetal.,2003).CathepsinK,in par-ticular,mediatesdirectliberationofT4fromTg(Tepeletal.,2000).
AlossofcathepsinKcanbecompensatedbyothercathepsins;this notionwasderivedfromanalysesofmiceshortlyafter establish-ingtheCtsk−/−model,i.e.whentheanimalswerestillonamixed
129Ola:C57BL/6Jbackground(Friedrichsetal.,2003).However,in thisstudy,weusedcongenicanimalsontheC57BL6/Jbackground. Immunoblottingrevealed thatthelossof cathepsinKledtoan increaseof cathepsinLexpression,whereas proteinamountsof cathepsinsBandDwerenot,oronlymarginallyelevatedinCtsk−/−
Fig.4. Thyroglobulindegradationpatternandcross-linkagein5–8monthsoldmice.
WholethyroidlysatesofmicedeficientinMct10,Mct8and/orMct10wereseparatedonverticalorhorizontalSDS-gels,silver-stainedortransferredtonitrocellulose forsubsequentincubationoftheblotswithantibodiesagainstTg,asindicated,andcomparedtotheWT.(A)Displayedisarepresentativeimmunoblotresultandthe correspondingdensitometrydatashowingincreasedamountsofsmallerTgfragments,withmolecularmassesbelow330kDa,inMct8-and/orMct10-deficientmiceas comparedtotheWTmice.(B)Whenseparatingwholethyroidlysatesinnon-reducingconditions,thebandofthedimericTgwascomparableinallgenotypesinvestigated, howeverthemultimericformsofTgwerelessabundantinMct8-andMct8/Mct10-deficientmice.(C)ReducingsampleswithDTTbeforeSDS-PAGEyieldedindecreased intensityofbandsrepresentingthemono-anddimericTginMct8-and/orMct10-deficientanimals,howevermoststrikinginMct8-andMct8/Mct10-deficientmice,which mayindicateincreasedsolubilizationandproteolyticdegradationofTgbycysteinecathepsins.Animalsanalyzed:n=2–3.InA,means±SDaregiven;*,P<0.05.
previousfindings(Friedrichsetal.,2003),andfurthervalidatedthe importanceofcathepsinL-mediatedTgprocessing.
LossofbothcathepsinsKandLleadstodecreasedT4levelsinthe
blood;however,lossofcathepsinKalonedoesnotaffectserumTH levels(Friedrichsetal.,2003).Whileendo-lysosomalTg process-ingiscompensatedforinCtsk−/−micebycathepsinL,thefunction
ofcathepsinKinextracellularT4liberationdirectlyfromTgisstill
lacking,leadingustowonderhownormalserumT4levelsare
main-tainedinCtsk−/−mice.Wereasonedthatonepossiblemechanism
explainingnormalserumTHlevelsmaybeanupregulationofTH transporters,andtothisextentweinvestigatedMct8expressionin thyroidsofCtsk−/−mice.
Thesignaldetectedafterincubationofcryo-sectionswith anti-bodiesagainstMct8(Fig.11AandB)wasquantified,anda3-fold increaseinthyroidtissueofCtsk−/−animalswasfound(Fig.11F).
Likewise,inimmunoblots,Mct8proteinamountswerefoundto beincreasedinthyroidtissuelysatesofcathepsinK-deficientover WTanimals(Fig.11GandH).However,thelocalizationofMct8at thebasolateralplasmamembranedomainofthethyrocytesdidnot changeinCtsk−/−animals(Fig.11DandE).
As described above, alterations in the levels of TH trans-porters feedback on the gross degradation of Tg mediated by endo-lysosomalproteases(seeFigs.3and4).Therefore,we fur-therinvestigatedwhetherthereversesituation,namelyalossof theTg-degradingcathepsinsBandL,isaccompaniedbydifferent expression of the TH transporter Mct8. In mice with single-deficiencies in the cathepsins B or L, or in cathepsin B- and
K-double-deficientanimals, similarobservations weremade,as such,thatMct8stainingintensitieswereincreasedoverWT,but its localization at thebasolateral plasma membrane domainof thyrocytesdidnotchange(Fig.12).However,anti-Mct8-positive punctawereadditionallyobservedinthecytoplasmofthyrocytes ofcathepsinB-,K-,orL-single-andcathepsinB/K-double-deficient mice(Fig.12,arrows).
Thus,thenotionofenhancedamountsoftheT4-exportingMct8
inconditionsofTg-processingcathepsinslackingfrommouse thy-roidepithelialcells(thisstudy)iscounterintuitive,asitmostlikely meansthatthethyroidglanditselfbecomeshypothyroid,especially intheabsenceoftheT4-liberatingcathepsinK(Tepeletal.,2000; Friedrichsetal.,2003),inordertomaintainproperserumlevelsof TH.ThebloodserummustthereforebeviewedasasinkforfreeT4
(HeuerandVisser,2013). 4. Discussion
Fig.5.ProteinlevelsofcathepsinB,D,andLinthyroidsofTHtransporter-deficientyoungmice.
Wholethyroidlysatesofyoungmice(lessthan2.5monthsold)wereseparatedonhorizontalSDS-gels,transferredtonitrocelluloseandincubatedwithantibodiesagainst therespectiveproteins(leftpanels).DensitometryoftheimmunolabeledbandsrevealedthatcathepsinB,D,andLlevelswereelevatedsignificantlyinMct8/Mct10-deficient animals,howeverthyroidlysatesofMct8-orMct10-deficientanimalsdidnotexhibitelevatedproteinamountswhencomparedtotheWT(rightpanels).Animalsanalyzed: n=6,3,6and10,respectively.HC=heavychain;SC=singlechain;pro=proform.Dataaregivenasmeans±SD;***,P<0.001.
Fig.6.ProteinlevelsofcathepsinB,D,andLinthyroidsofTHtransporter-deficientadultmice.
Wholethyroidlysatesofadultmice(5–8monthsold)wereseparatedonhorizontalSDS-gels,transferredtonitrocelluloseandincubatedwithantibodiesagainstcathepsinsB, DorL(leftpanels).DensitometryoftheindicatedbandsrevealedthatcathepsinDproteinlevelswereelevatedsignificantlyinMct8/Mct10-deficientmice,whereascathepsin BandLlevelswereincreasedinallinvestigatedgenotypeswhencomparedtoWT(rightpanels).Animalsanalyzed:n=3pergenotype.HC=heavychain;SC=singlechain; pro=proform.Dataaregivenasmeans±SD;*,P<0.05;**,P<0.005.
TH-resistantMct8-andMct8/Mct10-deficientmice(Trajkovicetal., 2007;Mulleretal.,2014).Inthisstudy,weproposethatTH trans-porters likeMct8 and Mct10 collectivelymeet the demands of facilitatingthereleaseofTHintothebloodinordertoproperly supportperipheralTHtargetorgans,whilealsoinvolvingin
cys-Fig.7. ExpressionandlocalizationofcathepsinBinthyroidglandsfromTHtransporter-deficientmice.
CathepsinB(white,red)wasmainlylocalisedinvesiclesofthyrocytesofWT(A)andMct10-(B)Mct8-(C),andMct8/Mct10-deficient(D)animalsbutalsoassociatedwith theapicalplasmamembranedomainandwithinthelumenofthyroidfollicles(EtoH,asterisks).ThyroidtissuetakenfromcathepsinB-deficientanimalsservedascontrols forantibodyspecificity,whichwasconfirmedbylackofprimaryantibodyreactivity(I).Vesicularstaining(arrows)ofcathepsinBismorepronouncedandsignalintensities (J)areslightlyincreasedinboth,Mct8-(G)orMct8/Mct10-deficient(H)mousethyroidswhencomparedtoWT(E).Representativeimagesof3animals(5–8monthsold) pergenotypearedisplayed.J:normalizedsignalintensitypercellgivenasmeans±SD;n=12,8,8,and12,respectively.Nucleiwerecounter-stainedwithDraq5TM(blue).L
denotesfolliclelumen.Scalebars:50m(A–D,I)and25m(E–H).
teinecathepsins,whilethereversesituation,cathepsindeficiency, results in up-regulated T4-exporting capabilities of thyrocytes,
sinceMct8ispresentinenhancedamounts.Although counterintu-itive,thisnotioncontributestoexplanations,whyMct8deficiency leadstointra-thyroidalhyperthyroidstatesandtotheunusual thy-roidstatesinthebloodcirculation(seebelow;MüllerandHeuer, 2014).Instarkcontrast,singledeficienciesinthecathepsinsB,K, andLfeatureeu-oronlyverymildlyhypothyroidstates(Friedrichs etal.,2003).CathepsinKandLdouble-deficiency,only,resultsin hypothyroidisminmice(Friedrichsetal.,2003).
4.1. ThyroidfunctionismaintainedbyTHreleasefromthyroid follicleswhichisenabledbyprotease-mediatedTHliberationfrom Tg
Mct8 knock-out mice exhibit an unusual pattern of altered TH serum levels with low T4 and highly increased T3 levels
(Dumitrescuetal.,2006;Trajkovicetal.,2007;Wirthetal.,2011). AnimalslackingbothMct8andMct10exhibitpartiallynormalized serumT4levels,despiteTSHlevelselevatedoverWTtoanextent
comparabletoMct8-deficientmice,buttheelevatedT3serum
lev-elstypicalforMct8-deficientmicearemaintained(Mulleretal., 2014).Thereasonfor thesehighlyunusualserumTHstatesare consideredtoresideindecreasedT4releasefromthyroidfollicles
duetodysfunctionalMct8-facilitatedexportacrossthebasolateral plasmamembranedomainofthyrocytes(Mulleretal.,2014).Inthe sameanimals,increasedconversionofT4toT3mediatedby
deiod-inase1(Dio1)inliverandkidneyexplainstheabnormallyhighT3
serumlevels.Inaddition,webelievethatthethyroidglandofMct8 knock-outmicereleasesmoreT3duetointactMct10-facilitated transport,whichcontributesfurthertotheunusuallyhighT3/T4
Fig.8.ExpressionandlocalizationofcathepsinDinthyroidglandsfromTHtransporter-deficientmice.
CathepsinDwaslocalisedinvesiclesdistributedevenlythroughoutthecytoplasminthyrocytesofWT(A),Mct10-(B),Mct8-(C),andMct8/Mct10-deficient(D)mice. Enlargedanti-cathepsinD-positivevesicleswherefoundinMct10-(F),Mct8-(G),andMct8/Mct10-deficient(H)mice(dashed,emptyarrows)whencomparedtotheWT(E). InMct8-deficientmice,cathepsinDstainingwasobservedatthebasalsurfaceofthyrocytes(G,arrows),orattheplasmamembranedomainofendothelialcells(G,asterisks). Dottedlines(E-H)representoutlinesoffollicleepithelia.Representativeimagesof3animals(5–8monthsold)pergenotypearedisplayed.Ldenotesfolliclelumen.Scale bars:50m(A–D)and10m(E–H).
theTHtransporter-deficientmice.Indeed,itislikelythat Mct8-deficient thyrocytesare stimulated more strongly than thyroid folliclecellsofMct8/Mct10-doubledeficientmice,becauseTRH lev-elsarehigherinMct8-deficientanimals(Mulleretal.,2014).Thus, intra-thyroidalTHproductionandTHliberationfromTgismost likely de-regulateddifferently in Mct8−/y and Mct8−/y/Mct10−/−
animals.Therefore,weinvestigatedtheTgstatusandthe molec-ularpathwaysofTgutilizationforTHliberationbyanalyzingthe keyplayersinvolvedinproteolyticprocessinganddegradationof Tg,thecathepsins(seeTable1andFig.1).
ThedatacollectedinthisstudybyinspectingtheTgstatusof THtransporter-deficientmice(seeFigs.2–4),suggeststhatbothTg solubilizationfromcovalentlycross-linkedstorageformsby extra-cellularTgprocessing,anditsutilizationforTHliberationby extra-andintracellularmeansofproteolyticTgdegradationareenhanced inMct8-and/orMct10-deficiencies.Therefore,wepostulatedthat Tg-processingenzymesmaybedifferentlyexpressedandlocalised inthyroidtissueofMct8-and/orMct10-deficientmice.
Secretionof cathepsin B at the apical pole of thyrocytes is increaseduponacuteTSHstimulation,i.e.1–2hafterTSH stimu-lation,whereas enhancedbiosynthesisofcathepsinBwilloccur only during long-term chronic TSH stimulation (Linke et al., 2002a,b).Accordingly,becausetheelevatedTSHlevelsin Mct8-orMct8/Mct10animalsarechronic(DiCosmoetal.,2010;Muller etal., 2014), we expectedincreasedprotein levelsof cathepsin B,which wereindeedobservedbyimmunoblotting(seeFig.6). Moreover,theintra-follicularlocalizationofcathepsinBdiffered fromWT,inparticularinMct8-andMct8/Mct10-deficientmice, suchthattheproteasewasalmostexclusivelypresentinenlarged endo-lysosomesofthethyroidepithelialcells(seeFig.7).
Anotherimportantproteinthatisinvolvedinendo-lysosomalTg degradation,butlesssoinTHliberationfromTg,istheaspartic pro-teasecathepsinD(Dunnetal.,1991;Jordansetal.,2009).Inmouse thyrocytes,cathepsinDisusuallylocalisedinvesiclesdistributed throughout thecytoplasm and accumulatingin theperinuclear region,i.e.inlysosomesofthyrocytes(Friedrichsetal.,2003).In
thisstudy,cathepsin Dwasfoundinimmenselyenlarged endo-lysosomal vesicles of Mct8- and Mct8-/Mct10-deficient mouse thyrocytes(seeFig.8),indicatingenhancedabilitiesofthesefollicle cellstocompletelydegradeTgbyintracellularmeansoflysosomal proteolysis.Moreover,cathepsinDwasfoundtobesecretedfrom thyrocytesofMct8-deficientmiceatthebasolateralpole.This find-ingisinterestingwhenconsideringtherolethatcathepsinDplays incancer(Brixetal.,2008;Brixetal.,2015).Mct8-deficientmice developpapillarythyroidcarcinoma(PTC)atolderage,possibly duetoalossofauto-regulativecontrolofthyrocyteproliferation (Wirthetal.,2011).However,basolaterallysecretedcathepsinD mayfurthercontributetotheprogressionofPTCdevelopmentas itcan,inprinciple,enhanceextracellularmatrixdegradation,and therebyfacilitatecancerprogression.
CathepsinLplaysavitalroleinthyrocytefunction,anditwas showntobeimportantforsurvivalofthyrocytes,sincethe respec-tiveknock-outmodelsshowedluminalinclusionsofdeadcellsand deadcellremnants(Friedrichsetal.,2003).Hereweshowthatthe availabilityofcathepsinLintheextracellularfolliclelumenismore pronouncedinMct8/Mct10-deficientmicethaninMct8−/y mice
(seeFig.9),suggestingenhancedTgsolubilisationforsubsequent endocyticuptake.Inaddition,thisenzymewithacriticalfunction inmaintainingproperserumlevelsofT4(Friedrichsetal.,2003)
wasenhancedinproteinamounts,inparticular,inMct8andMct10 deficientmousethyrocytes.
Hence,theproteinlevelsoftheTg-processingproteases corre-latewiththelackofTHtransporterproteinslikeMct8and/orMct10. Theseobservationsmostlikelycontributetoexplainingwhyamong theTHtransporter-lackinggenotypesinspectedinthisstudy, espe-ciallytheMct8−/y/Mct10−/−mousemodelfeaturesthemostsevere
auto-thyrotoxicstates(Mulleretal.,2014).In reverse,themain T4-exportingtransportermoleculeMct8isup-regulatedinCtsk−/−
mice,aTgprocessingenzymethatisabletodirectlyliberateT4
Fig.9. ExpressionandlocalizationofcathepsinLinthyroidglandsfromthyroidhormonetransporterdeficientanimals.
CathepsinLwasdetectedinvesicularstructuresinallgenotypes(A,B,C,andD),exceptintissuefromcathepsinL-deficientmice(G),whichservedascontrolforantibody specificity.FewfolliclesofWT(A)andMct10-deficient(B)animalsshowedapicalandluminalcathepsinLstaining,howeverintensitiesofcathepsinLlabelling(H)were increasedinMct8and/orMct10-deficientmice(BandC)whencomparedtoWTanimals(A).Notetheincreasedluminal,apicalandvesicularstainingpatternsandintensities ofcathepsinL(green)inthyrocytesofMct8/Mct10-deficientanimals(F)comparedtotheWT(E),andintherespectivehighermagnifications(E’andF’).Representative imagesof3animals(5–8monthsold)pergenotypearedisplayed.Nucleiwerecounter-stainedwithDraq5TM(blue).H:normalizedsignalintensitypercell;mean+/−SEM;
n=1193,1422,1309,and1638,respectively.Ldenotesfolliclelumen.Scalebars:50m(A–D,G)and10m(E,F).*,P<0.05.
2003),whichalsofeatureenhancedMct8atthebasolateralplasma membranedomainsofthyrocytes(thisstudy).
Such correlations between TH-liberating proteases and TH transportersaresomewhatcounterintuitive,andindicatealossof THsensingcapabilitiesofthyrocytesuponfunctionallossofeither key-playerenablingthyroidglandfunctions(seeFig.13).
4.2. Thyroidauto-regulationtoleratesminorchangesinthyroid statesbyredundancy,butaself-destructiveviciouscycleis switchedonwhenthekey-playerscathepsinLandMct8are affectedbydisease
Fig.10.ProteinlevelsofcathepsinsB,D,andLinthyroidsofcathepsinK-deficientmice.
WholethyroidlysateswereseparatedonhorizontalSDS-gels,transferredtonitrocelluloseandincubatedwithantibodiesagainsttherespectiveproteins.(A)Immunoblots ofcathepsinsB,D,andLwerequantifiedbydensitometry(B).CathepsinLlevelswereelevatedsignificantlyinCtsk-deficientanimals.Animalsanalyzed:n=3pergenotype. HC=heavychain;SC=singlechain.Dataaregivenasmeans±SD;*,P<0.05.
ofTHtransporters(asinCtsb−/−,Ctsk−/−,Ctsl−/−,orCtsb−/−/Ctsk−/−
mice),orbyenhancingcathepsin-mediatedTg-processing(asin Ctsk−/− mice), or both, without the necessity of TSH-triggered
classicalregulation(seeFig.13).Suchauto-regulatorypathways, drivenexclusivelybyintra-thyroidalmeans,mightbebasedupon thyrocyte-specificmeasuresofTH-sensing.Thus, withinthyroid follicles,thethyroidepithelialcellsdetecttheamountsofTH lib-erated,andintegratethesewiththoseTHamountsexportedfrom thefolliclesforglobalsupply(Fig.13,WT).Suchapathwayof TH-sensingwithinthyroidepithelialcellsofindividualfolliclesmost likelyrequirestheinvolvementofTHtransportersactingat differ-entcellularmembranes(Fig.13).Interestingly,althoughdifferent TH transporterproteins areexpressed in mousethyrocytes (Di Cosmoetal.,2010),compensatoryup-regulationofTHtransporters inMct8-deficientanimalshasnotbeenfound(DiCosmoetal.,2010; Wirthetal.,2011).Incontrast,enhancedTgdegradationand uti-lizationtakesplacewhenMct8andMct10arelacking(thisstudy). Eventually,thisleadstoincreasedintra-thyroidalTHlevels,which shouldfeedbacktothethyrocytetostoptheelevationofTH produc-tionatonepoint.Ironically,andagainstalllogics,sucharegulatory feedbackdoesnothappeninMct8/Mct10-deficientanimals.
Insummary,theresultsofthisstudyprovideamolecular mech-anisticexplanationforthealteredthyroidfunctionalstatesofmice withMct8-and/orMct10-deficienciesinthatenhancedTg degra-dationbycathepsinsBandL(actingextra-andintracellularlyonTg) andbyintra-lysosomallyactingcathepsinDoccurs,althoughthisis counterintuitiveandlikelyworsenstheintra-follicularthyrotoxic states.
At first glance, up-regulation of Mct8 in thyroid glands of animals lacking the Tg-processing cathepsin proteases (Figs. 11 and 12) makes more sense as the export of T4 from
the thyroid follicles would be facilitated in suchconditions of enhancedMct8atthebasolateralplasmamembraneofthyrocytes.
However,againcounterintuitively,thethyroidglanditselfmight becomehypothyroid,atleasttransiently,attheexpenseof main-tainingproperTHsupplytotheperipheralorgans,inparticular,in Ctsk−/−mice(Friedrichsetal.,2003).Withinthethyroidfolliclesof
Ctsk−/−mice,suchstatesoftoolittleTHliberationduetolacking
cathepsin K-mediated Tg-processing by extra-and intracellular meansofproteolysisareobviouslysensedandcounter-actedsuch thattheyeventuallyleadtoacompensatoryup-regulationofthe amountsofcathepsinL(Friedrichsetal.,2003).
4.3. Perspectives
In conclusion, we propose that the intra-thyroidal TH sens-ingmechanismspresentinCtsb−/−,Ctsk−/−,Ctsl−/−,Mct8−/y,and
Mct8−/y/Mct10−/−miceareenabledsuchthatdifferentTH
trans-porters are present in specific subcellular locations of thyroid epithelial cells, thereby selectively monitoring individual steps ofthyroglobulinprocessingandthus,thecellularTHstatus(see
Fig.13).Hence,futurestudiesmustelucidatethemolecularnature oftheseproposedintra-thyroidalT4-sensingmechanismsbeyond
Mct8,inordertounravel,howT4-sensingbyMct8and/orMct10at
thebasolateralplasmamembranedomainismechanistically con-nectedtoregulationoftranslationandtraffickingofTg-processing proteaseswithinindividualfollicles,andhowpreciselythe associ-ationwithendo-lysosomalfunctionsismaintained.Wepropose aninvolvement oftheTH transporterLat2, which ispresent at theendo-lysosomalmembraneofthyrocytesinWTmice(McInnes etal.,2013;Weberetal.,2015).
Fig.11.Mct8expressionlevelsandlocalizationinthyroidtissueofcathepsinK-deficientmice.
SignalintensitieswerequantifiedinmicrographsofcryosectionslabeledwithantibodiesagainstMct8(A,B).TissueofMct8-deficientanimalsservedtocontrolforstaining specificityoftheprimaryantibody(C).NotethesignificantincreaseofMct8incathepsinK-deficientmicewhencomparedtotheWT(F).However,thelocalisationofMct8 (green)atthebasolateralplasmamembranedomain(DandE)didnotchangeincathepsinK-deficiency.WholethyroidlysateswereseparatedonhorizontalSDS-gels, transferredtonitrocelluloseforsubsequentincubationoftheblotswithantibodiesagainstMct8(G)andthequantificationbydensitometryrevealedelevatedMct8levels inCtsk-deficientmice(H).F:animalsanalyzedwere8and9,respectively;dataaregivenasmeans±SD;n=41and49follicles,respectively.H:animalsanalyzedn=6,and 6,respectively.Scalebars:40m(A–C)and10mm(DandE).Nucleiwerecounter-stainedwithDraq5TM(blue).***,P<0.001.
andTgbiosynthesis(Brixetal.,2001).Thus,itwillbeimportantin futurestudiestodiscriminatebetweenTSH-dependentregulation andthyroidauto-regulationthatisindependentofthesystemic HPT-axis.Inthisregard,wearecurrentlyanalyzingknock-out mod-elsfeaturingthyroid-specificMct8-deficiency.
Fig.12.Mct8expressionlevelsandlocalizationinthyroidtissueofWTandcathepsinB,K,L,andB/K-deficientmice.
ConfocalfluorescencemicrographsofcryosectionslabeledwithantibodiesagainstMct8(green)revealedenhancedstainingintensityincathepsin-deficientmice(B–E)when comparedtotheWT(A).However,thelocalizationofMct8atthebasolateralplasmamembranedomaindidnotchangeincathepsin-deficiencies.Thyroidof cathepsin-deficientmicewerecharacterizedbyanti-Mct8-positivepuncta(arrows)whichweremostabundantinCtsl−/−(D)andCtsb−/−/Ctsk−/−(E)mice.Scalebars:10m.Nuclei
werecounter-stainedwithDraq5TM(blue).
Fig.13.TheinfluenceofcathepsinKandTHtransporterdeficiencyonliberationofTHandtheirreleasefromthyroidfollicles.
Tgissynthesisedandsecretedattheapicalplasmamembraneofthyrocytes(WT,left)tobestoredincovalentlycross-linkedforminthelumenofthyroidfollicles.Upon TSHstimulation,cysteinecathepsinsaresecreteduponretrogradetrafficking,andcathepsinsBandLareinvolvedinsolubilizationofTg,whereascathepsinKisinvolvedin directT4liberationinthefolliclelumen(WT,upperportion,andlowerpanel),whichmaybetransportedacrosstheapicalplasmamembranedomainbyanasyetunknown
T4transporter(cyan).Then,TgandpartiallydegradedTgmoleculesareendocytosedfordeliverytoearlyandlateendosomesaswellaslysosomes,whereT4andT3canbe
liberatedbytheintracellularproteolyticactionofcysteinecathepsinsbeforebeingtransportedacrosstheendo-lysosomalmembranesintothecytosol(darkpink).T4may
thenbedeiodinatedandthusactivatedtoT3,orT4isreleasedfromthethyrocytesbytransportacrossthebasolateralplasmamembranedomain(chartreuse).
MicedeficientincathepsinKexhibitnormalT4serumlevels,whileMct8deficiencyresultsinreducedT4andslightlyelevatedserumTSHconcentrations.TSHinfluencesthe
increaseofTHproductionandlikewise,expressionofcathepsinsiselevateduponlong-termTSHstimulation(Linkeetal.,2002b).MicewithdeficienciesinMct8andMct10 exhibitsimilarTSHlevelsasMct8-deficientmicebutdifferentserumT4levels.UponlossofcathepsinK(Ctsk−/−),weobservedup-regulatedcathepsinLexpressionand
activity(Friedrichsetal.,2003;redarrows)andenhancedexpressionoftheTHtransporterMct8atthebasolateralplasmamembranedomainofthethyrocytes(thisstudy; greenarrows),suggestingT4releasefromthyroidfolliclestobefavoredoverself-sustainingregulation,therebyrenderingthethyroidglandhypothyroid.Uponlossofboth
Mct8andMct10(Mct8−/y/Mct10−/−),weobservedalteredcross-linkingandmultimerizationofTgwhichmayhaveledtoenhancedutilizationofTgforTHliberation(red
arrows)andtherebyincreasingtheintra-thyroidalTHlevels(Mulleretal.,2014).WespeculatethatthelossofMct8(greencrosses)andMct10fromthebasolateralplasma membranedomainofthyrocytesledtoalossofsensingcapabilityfortheamountofliberatedTHbycathepsins,whichyieldedinincreasedprocessinganddegradationof Tgdespitetoxicintra-thyroidalTHconcentrations,therebyhoweveralsocontributingtothenormalserumT4levelsinMct8/Mct10-deficientmice.
releasefromthyrocytes,beforeTgiseventuallystoredin cross-linkedform in the thyroid folliclelumen.Both, Tgfolding and Tgcross-linkingarechaperonedandmediatedbyprotein disul-fideisomerase(PDI;Berndorferetal.,1996;MuresanandArvan,
theTgmoleculefurtherfeaturesthepresenceofdifferentinternal homologydomains(forreview,seeBrixetal.,2001),someofwhich containtheaforementionedCXXC-motifs ofthioredoxins,while othersarecharacterizedbyCWCV-motifsthatareindicativeofthe so-calledthyropins,i.e.smallproteinsorpeptidesactingascysteine peptidaseinhibitors(Lenarcicetal.,1997).Clearly,futurestudies areneededinmicetoexplorethesignificanceofTgcross-linking andthepossibleinterconnectionswithitsdegradation,sinceboth arepotentiallysubjecttointra-thyroidalauto-regulation,i.e.byTg itself(RoussetandMornex,1991;Brixetal.,2001).
Funding
ThisworkwassupportedbytheDeutsche Forschungsgemein-schaft,DFG,grantBR1308/11-1and11-2toK.B.,WI3768/1-1to E.K.W.,SCHW914/3-1toU.S.,andHE3418/7-1and7-2toH.H.in theframeworkofthepriorityprogramSPP1629/1and/2‘Thyroid TransAct’.
Disclosurestatement
Theauthorsdeclarethatnoconflictofinterestexists. Acknowledgements
WearegratefultoDrs.ChristophPetersandThomasReinheckel (both,UniversityofFreiburg)forkindlyprovidingcathepsinB-and L-deficientmousetissue,andtoDr.PaulSaftig(UniversityofKiel) forhisgeneroussharingofthecathepsinK-deficientmice. References
Arampatzidou,M.,Schutte,A.,Hansson,G.C.,Saftig,P.,Brix,K.,2012.Effectsof cathepsinKdeficiencyonintercellularjunctionproteins,luminalmucus layers,andextracellularmatrixconstituentsinthemousecolon.Biol.Chem. 393(12),1391–1403,http://dx.doi.org/10.1515/hsz-2012-0204.
Berndorfer,U.,Wilms,H.,Herzog,V.,1996.Multimerizationofthyroglobulin(TG) duringextracellularstorage:isolationofhighlycross-linkedTGfromhuman thyroids.J.Clin.Endocrinol.Metab.81(5),1918–1926,http://dx.doi.org/10. 1210/jcem.81.5.8626858.
Brix,K.,Lemansky,P.,Herzog,V.,1996.Evidenceforextracellularlyacting cathepsinsmediatingthyroidhormoneliberationinthyroidepithelialcells. Endocrinology137(5),1963–1974,http://dx.doi.org/10.1210/endo.137.5. 8612537.
Brix,K.,Summa,W.,Lottspeich,F.,Herzog,V.,1998.Extracellularlyoccurring histoneH1mediatesthebindingofthyroglobulintothecellsurfaceofmouse macrophages.J.Clin.Invest.102(2),283–293.
Brix,K.,Linke,M.,Tepel,C.,Herzog,V.,2001.Cysteineproteinasesmediate extracellularprohormoneprocessinginthethyroid.Biol.Chem.382(5), 717–725,http://dx.doi.org/10.1515/bc.2001.087.
Brix,K.,Dunkhorst,A.,Mayer,K.,Jordans,S.,2008.Cysteinecathepsins:cellular roadmaptodifferentfunctions.Biochimie90(2),194–207,http://dx.doi.org/ 10.1016/j.biochi.2007.07.024.
Brix,K.,Führer,D.,Biebermann,H.,2011.Moleculesimportantforthyroid hormonesynthesisandaction–knownfactsandfutureperspectives.Thyroid Res.4(Suppl.1),S9,http://dx.doi.org/10.1186/1756-6614-4-s1-s9. Brix,K.,McInnes,J.,Al-Hashimi,A.,Rehders,M.,Tamhane,T.,Haugen,M.H.,2015.
Proteolysismediatedbycysteinecathepsinsandlegumain-recentadvances andcellbiologicalchallenges.Protoplasma252(3),755–774,http://dx.doi.org/ 10.1007/s00709-014-0730-0.
Dauth,S.,Arampatzidou,M.,Rehders,M.,Yu,D.M.T.,Führer,D.,Brix,K.,2011. ThyroidcathepsinK:rolesinphysiologyandthyroiddisease.Clin.Rev.Bone Miner.Metab.9(2),94–106,http://dx.doi.org/10.1007/s12018-011-9093-7. DiCosmo,C.,Liao,X.H.,Dumitrescu,A.M.,Philp,N.J.,Weiss,R.E.,Refetoff,S.,2010.
MicedeficientinMCT8revealamechanismregulatingthyroidhormone secretion.J.Clin.Invest.120(9),3377–3388,http://dx.doi.org/10.1172/ jci42113.
DiJeso,B.,Arvan,P.,2016.Thyroglobulinfrommolecularandcellularbiologyto clinicalendocrinology.Endocr.Rev.37(1),2–36,http://dx.doi.org/10.1210/er. 2015-1090.
Dumitrescu,A.M.,Liao,X.H.,Weiss,R.E.,Millen,K.,Refetoff,S.,2006.
Tissue-specificthyroidhormonedeprivationandexcessinmonocarboxylate transporter(mct)8-deficientmice.Endocrinology147(9),4036–4043,http:// dx.doi.org/10.1210/en.2006-0390.
Dunn,A.D.,Crutchfield,H.E.,Dunn,J.T.,1991.Thyroglobulinprocessingby thyroidalproteasesMajorsitesofcleavagebycathepsinsB,D,andL.J.Biol. Chem.266(30),20198–20204.
Dunn,A.D.,Myers,H.E.,Dunn,J.T.,1996.Thecombinedactionoftwothyroidal proteasesreleasesT4fromthedominanthormone-formingsiteof thyroglobulin.Endocrinology137(8),3279–3285,http://dx.doi.org/10.1210/ endo.137.8.8754751.
Führer,D.,Brix,K.,Biebermann,H.,2015.Understandingthehealthythyroidstate in2015.Eur.ThyroidJ.4(Suppl.1),1–8,http://dx.doi.org/10.1159/000431318. Fliers,E.,Kalsbeek,A.,Boelen,A.,2014.Beyondthefixedsetpointofthe
hypothalamus-pituitary-thyroidaxis.Eur.J.Endocrinol./Eur.Feder.Endocr. Soc.171(5),R197–R208,http://dx.doi.org/10.1530/eje-14-0285.
Friedrichs,B.,Tepel,C.,Reinheckel,T.,Deussing,J.,vonFigura,K.,Herzog,V.,Peters, C.,Saftig,P.,Brix,K.,2003.ThyroidfunctionsofmousecathepsinsB,K,andL.J. Clin.Invest.111(11),1733–1745,http://dx.doi.org/10.1172/jci15990. Fujita,H.,1988.Functionalmorphologyofthethyroid.In:Jeon,K.W.,Friedlander,
M.(Eds.),InternationalReviewofCytology,vol.113.AcademicPress,pp. 145–185,http://dx.doi.org/10.1016/S0074-7696(08)60848-7.
Gassmann,M.,Grenacher,B.,Rohde,B.,Vogel,J.,2009.QuantifyingWesternblots: pitfallsofdensitometry.Electrophoresis30(11),1845–1855,http://dx.doi.org/ 10.1002/elps.200800720.
Herzog,V.,Berndorfer,U.,Saber,Y.,1992.Isolationofinsolublesecretoryproduct frombovinethyroid:extracellularstorageofthyroglobulinincovalently cross-linkedform.J.CellBiol.118(5),1071–1083.
Heuer,H.,Visser,T.J.,2013.Thepathophysiologicalconsequencesofthyroid hormonetransporterdeficiencies:insightsfrommousemodels.Biochim. Biophys.Acta1830(7),3974–3978,http://dx.doi.org/10.1016/j.bbagen.2012. 04.009.
Heukeshoven,J.,Dernick,R.,1988.Improvedsilverstainingprocedureforfast staininginPhastSystemDevelopmentUnit.I.Stainingofsodiumdodecyl sulfategels.Electrophoresis9(1),28–32,http://dx.doi.org/10.1002/elps. 1150090106.
Jordans,S.,Jenko-Kokalj,S.,Kuhl,N.M.,Tedelind,S.,Sendt,W.,Bromme,D.,Turk, D.,Brix,K.,2009.Monitoringcompartment-specificsubstratecleavageby cathepsinsB,K,L,andSatphysiologicalpHandredoxconditions.BMC Biochem.10,23,http://dx.doi.org/10.1186/1471-2091-10-23.
Kamentsky,L.,Jones,T.R.,Fraser,A.,Bray,M.A.,Logan,D.J.,Madden,K.L.,Ljosa,V., Rueden,C.,Eliceiri,K.W.,Carpenter,A.E.,2011.Improvedstructure,function andcompatibilityforCellProfiler:modularhigh-throughputimageanalysis software.Bioinformatics(Oxford,England)27(8),1179–1180,http://dx.doi. org/10.1093/bioinformatics/btr095.
Klein,M.,Gestmann,I.,Berndorfer,U.,Schmitz,A.,Herzog,V.,2000.The thioredoxinboxesofthyroglobulin:possibleimplicationsforintermolecular disulfidebondformationinthefolliclelumen.Biol.Chem.381(7),593–601,
http://dx.doi.org/10.1515/bc.2000.076.
Kyhse-Andersen,J.,1984.Electroblottingofmultiplegels:asimpleapparatus withoutbuffertankforrapidtransferofproteinsfrompolyacrylamideto nitrocellulose.J.Biochem.Biophys.Methods10(3–4),203–209.
Lenarcic,B.,Ritonja,A.,Strukelj,B.,Turk,B.,Turk,V.,1997.Equistatin,anew inhibitorofcysteineproteinasesfromActiniaequina,isstructurallyrelatedto thyroglobulintype-1domain.J.Biol.Chem.272(21),13899–13903.
Liao,X.-H.,DiCosmo,C.,Dumitrescu,A.M.,Hernandez,A.,vanSande,J.,St Germain,D.L.,Weiss,R.E.,Galton,V.A.,Refetoff,S.,2011.Distinctrolesof deiodinasesonthephenotypeofMct8defect:acomparisonofeightdifferent mousegenotypes.Endocrinology152,1180–1191.
Linke,M.,Herzog,V.,Brix,K.,2002a.TraffickingoflysosomalcathepsinB-green fluorescentproteintothesurfaceofthyroidepithelialcellsinvolvesthe endosomal/lysosomalcompartment.J.CellSci.115(Pt24),4877–4889.
Linke,M.,Jordans,S.,Mach,L.,Herzog,V.,Brix,K.,2002b.Thyroidstimulating hormoneupregulatessecretionofcathepsinBfromthyroidepithelialcells. Biol.Chem.383(5),773–784,http://dx.doi.org/10.1515/bc.2002.081. Müller,J.,Heuer,H.,2014.Expressionpatternofthyroidhormonetransportersin
thepostnatalmousebrain.Front.Endocrinol.5,92,http://dx.doi.org/10.3389/ fendo.2014.00092.
Mariotta,L.,Ramadan,T.,Singer,D.,Guetg,A.,Herzog,B.,Stoeger,C.,Palacin,M., Lahoutte,T.,Camargo,S.M.,Verrey,F.,2012.T-typeaminoacidtransporter TAT1(Slc16a10)isessentialforextracellulararomaticaminoacidhomeostasis control.J.Physiol.590(24),6413–6424,http://dx.doi.org/10.1113/jphysiol. 2012.239574.
McInnes,J.,Weber,J.,Rehders,M.,Saftig,P.,Peters,C.,Reinheckel,T.,Schweizer,U., Heuer,H.,Wirth,E.K.,Brix,K.,2013.Correlationoftheexpressionand localizationofthyroidhormonetransporterswiththyroglobulinprocesing cathepsinsinmousethyroidepithelialcells.In:37thAnnualMeetingofthe EuropeanThyroidAssociation,Leiden,TheNetherlands,September7–112013, pp.75–194,http://dx.doi.org/10.1159/000352096,Suppl.1.EuropeanThyroid Journal.
Muller,J.,Mayerl,S.,Visser,T.J.,Darras,V.M.,Boelen,A.,Frappart,L.,Mariotta,L., Verrey,F.,Heuer,H.,2014.Tissue-specificalterationsinthyroidhormone homeostasisincombinedMct10andMct8deficiency.Endocrinology155(1), 315–325,http://dx.doi.org/10.1210/en.2013-1800.
Muresan,Z.,Arvan,P.,1998.EnhancedbindingtothemolecularchaperoneBiP slowsthyroglobulinexportfromtheendoplasmicreticulum.Mol.Endocrinol. 12(3),458–467.
determinationwhichisindependentofotherexternalinfluences. Hoppe-Seyler’sZeitschriftfurPhysiologischeChemie360(11),1657–1670.
Nilsson,M.,Fagman,H.,2013.Mechanismsofthyroiddevelopmentand dysgenesis:ananalysisbasedondevelopmentalstagesandconcurrent embryonicanatomy.Curr.Top.Dev.Biol.106,123–170,http://dx.doi.org/10. 1016/b978-0-12-416021-7.00004-3.
Nussey,S.,Whitehead,S.,2001.Chapter3:thethyroidgland.In:Nussey,S., Whitehead,S.(Eds.),Endocrinology:AnIntegratedApproach.BIOSScientific Publishers,Oxford.
Rousset,B.,Mornex,R.,1991.Thethyroidhormonesecretorypathway-current dogmasandalternativehypotheses.Mol.Cell.Endocrinol.78(1–2),C89–C93.
Saber-Lichtenberg,Y.,Brix,K.,Schmitz,A.,Heuser,J.E.,Wilson,J.H.,Lorand,L., Herzog,V.,2000.Covalentcross-linkingofsecretedbovinethyroglobulinby transglutaminase.FASEBJ.14(7),1005–1014.
Saftig,P.,Hunziker,E.,Wehmeyer,O.,Jones,S.,Boyde,A.,Rommerskirch,W., Moritz,J.D.,Schu,P.,vonFigura,K.,1998.Impairedosteoclasticbone resorptionleadstoosteopetrosisincathepsin-K-deficientmice.Proc.Natl. Acad.Sci.U.S.A.95(23),13453–13458.
Schmitz,A.,Klein,M.,Gestmann,I.,Herzog,V.,2002.Proteincross-linkingby self-assistedintermoleculardisulfidebondformation.MethodsEnzymol.348, 306–313.
Tan,H.Y.,Ng,T.W.,2008.Accuratestepwedgecalibrationfordensitometryof electrophoresisgels.Opt.Commun.281(10),3013–3017,http://dx.doi.org/10. 1016/j.optcom.2008.01.012.
Tepel,C.,Bromme,D.,Herzog,V.,Brix,K.,2000.CathepsinKinthyroidepithelial cells:sequence,localizationandpossiblefunctioninextracellularproteolysis ofthyroglobulin.J.CellSci.113(Pt24),4487–4498.
Trajkovic,M.,Visser,T.J.,Mittag,J.,Horn,S.,Lukas,J.,Darras,V.M.,Raivich,G., Bauer,K.,Heuer,H.,2007.Abnormalthyroidhormonemetabolisminmice lackingthemonocarboxylatetransporter8.J.Clin.Invest.117(3),627–635,
http://dx.doi.org/10.1172/JCI28253.
Trajkovic-Arsic,M.,Visser,T.J.,Darras,V.M.,Friesema,E.C.,Schlott,B.,Mittag,J., Bauer,K.,Heuer,H.,2010a.Consequencesofmonocarboxylatetransporter8 deficiencyforrenaltransportandmetabolismofthyroidhormonesinmice. Endocrinology151(2),802–809,http://dx.doi.org/10.1210/en.2009-1053. Trajkovic-Arsic,M.,Muller,J.,Darras,V.M.,Groba,C.,Lee,S.,Weih,D.,Bauer,K.,
Visser,T.J.,Heuer,H.,2010b.Impactofmonocarboxylatetransporter-8 deficiencyonthehypothalamus-pituitary-thyroidaxisinmice.Endocrinology 151(10),5053–5062,http://dx.doi.org/10.1210/en.2010-0593.
Weber,J.,Rehders,M.,Saftig,P.,Verrey,F.,Schweizer,U.,Wirth,E.K.,Heuer,H., Brix,K.,2015.Functionalanalysisoftheangio-follicularunitofthemouse thyroidgland.Exp.Clin.Endocrinol.Diabetes123(03),P12–P13,http://dx.doi. org/10.1055/s-0035-1547747.
Wirth,E.K.,Roth,S.,Blechschmidt,C.,Holter,S.M.,Becker,L.,Racz,I.,Zimmer,A., Klopstock,T.,Gailus-Durner,V.,Fuchs,H.,Wurst,W.,Naumann,T.,Brauer,A., deAngelis,M.H.,Kohrle,J.,Gruters,A.,Schweizer,U.,2009.Neuronal 3’,3,5-triiodothyronine(T3)uptakeandbehavioralphenotypeofmicedeficient inMct8,theneuronalT3transportermutatedinAllan-Herndon-Dudley syndrome.J.Neurosci.29(30),9439–9449,http://dx.doi.org/10.1523/ jneurosci.6055-08.2009.