ContentslistsavailableatScienceDirect
Behavioural Brain Research
j o ur na l h o me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / b b r
Research report
The effects of CRF and urocortins on the preference for social novelty of mice
Zsolt Bagosi
∗, András Czébely-Lénárt, Gergely Karasz, Krisztina Csabafi, Miklós Jászberényi, Gyula Telegdy
DepartmentofPathophysiology,FacultyofMedicine,UniversityofSzeged,Hungary
h i g h l i g h t s
•MaleandfemaleCFLPmicewereinvestigatedinaCrawleysocialinteractiontest.
•CRFandUCN1decreasedthepreferenceforsocialnoveltyofmalemice.
•CRF1receptormediatestheeffectsofCRFandUCN1onmale-femaleinteraction.
•UCN2andUCN3didnotinfluencethepreferenceforsocialnoveltyofmalemice.
•CRF2receptordoesnotparticipatetomale-femaleinteraction.
a r t i c l e i n f o
Articlehistory:
Received7December2016
Receivedinrevisedform1February2017 Accepted6February2017
Availableonline9February2017
Keywords:
CRF Urocortins Socialnovelty Mice
a b s t r a c t
Theaimofthepresentstudywastodeterminetheroleofcorticotropin-releasingfactor(CRF),theuro- cortins(UCN1,UCN2andUCN3)andtheirreceptors(CRF1andCRF2)inthepreferenceforsocialnovelty ofmice.MaleCFLPmicewereadministeredintracerebroventricularly(ICV)withCRF,UCN1,UCN2 orUCN3and/orantalarminorastressin2B,selectiveantagonistsofCRF1receptorandCRF2receptor, respectively.ThemicewereinvestigatedinaCrawleysocialinteractiontestarenaconsistingofthree chambers:anunknownfemalewassetinthefirstchamberandaknownfemale,withwhichthemale wasfamiliarizedpreviouslyfor24h,wassetinthethirdchamber.Firstthetestedmalewashabituated withthemiddlechamberfor5minandthenallowedtoexploretheremainingchambersfor5min,during whichthenumberofentriesandthetimeofinteractionweremeasured.CRFdecreasedsignificantlythe numberofentriesandthetimeofinteractionwiththeunknownfemale,butnottheknownfemale.UCN 1decreasedsignificantlythenumberofentriesintothechamberoftheunknownfemale,butnotthe knownfemale,withoutchangingthetimeofinteraction.Alldecreasingeffectswerereversedbyanta- larmin,butnotastressin2B.UCN2andUCN3didn’tinfluencesignificantlyanyoftheparameters.The presentstudysuggeststhatCRFandUCN1decreasethepreferenceforsocialnoveltybyactivatingCRF1
receptor,whileUCN2andUCN3,activatingselectivelyCRF2receptor,donotparticipatetomale-female interaction.
©2017ElsevierB.V.Allrightsreserved.
1. Introduction
Corticotropin-releasingfactor(CRF)isa hypothalamicneuro- hormone and also an extrahypothalamic neurotransmitter that regulatestheendocrine,autonomicandbehavioral responsesto stress[1].DuringstressCRF,alongwiththesynergisticarginine-
∗Correspondingauthorat:DepartmentofPathophysiology,UniversityofSzeged, 6725,Szeged,Semmelweisstr.1,Hungary.
E-mailaddress:bagosi.zsolt@med.u-szeged.hu(Z.Bagosi).
vasopressin(AVP),isreleasedfromtheparaventricularnucleusof thehypothalamus(PVN)and,gettingintothecirculationthrough themedianeminence,stimulatesthereleaseoftheadrenocorti- cotropichormone(ACTH)intheanteriorpituitary[1].ACTH,inturn stimulatesthesynthesisofglucocorticoidsintheadrenalcortex resultinginelevationoftheconcentrationoftheplasmaglucocorti- coids[2].Theelevationoftheplasmaglucocorticoidconcentration not only reflects the activation of the hypothalamic-pituitary- adrenal (HPA)axis,but also exerts negativefeedback effect on thehypothalamicCRFandthepituitaryACTHrelease,inhibiting theHPAaxis[3].ICVinjectionofCRFand UCNImaystimulate http://dx.doi.org/10.1016/j.bbr.2017.02.009
0166-4328/©2017ElsevierB.V.Allrightsreserved.
ACTHsecretionintheanteriorpituitarythatpeaksin15minand maydecreasetheactivesocialinteraction.ACTH,inturnstimulates thesynthesisofglucocorticoidsintheadrenalcortexresultingin elevationoftheconcentrationoftheplasmaglucocorticoids[2].
Theelevationoftheplasmaglucocorticoidconcentrationnotonly reflectstheactivationofthehypothalamic-pituitary-adrenal(HPA) axis,butalsoexertsnegativefeedbackeffectonthehypothalamic CRFandthepituitaryACTHrelease,inhibitingtheHPAaxis.
SinceCRFwasfirstlyisolated[1],agrowingfamilyofCRF-like peptideshavebeendiscovered.Todaythemammalianmembers ofthisfamilyincludefourligands:CRF,urocortin1(UCN1)[4], urocortin2 (UCN2),alsoknown asstresscopin-relatedpeptide (SRP)[5],andurocortin3(UCN3),alsoknownasstresscopin(SCP) [6],two receptors(CRF1 and CRF2)[7]and onebindingprotein (CRF-BP)[8].Thenameurocortinderivesfromthefishhomologue urotensin(63%sequenceidentity)andthemammaliananalogue corticotropin(45%sequenceidentity)[9],astheurocortinsshare commonaminoacidicelements(45–33%),buthavedifferentphar- macologicalpropertiescomparedtoCRF.CRFbindspreferentially toCRF1receptor,whileUCN1attachesequipotentlytobothCRF receptors(CRF1andCRF2),andUCN2andUCN3bindselectively toCRF2receptor[10].ThesereceptorsbelongtotheclassBsubtype ofGprotein-coupledreceptors(GPCRs)and,likeallGPCRs,con- sistofanamino-terminalextracellularregion,acarboxyl-terminal intracellulartailandseven,transmembranesegments,connected byalternatingintracellularandextracellularloops[11].Thereis nearly70%identitybetweenCRF1andCRF2receptorsattheamino acidlevelwiththetransmembraneandintracellulardomainsofthe CRFreceptorspresentingthehighesthomology(over80%identity) [11].Inaddition,CRFandUCN1canbefoundattachedtoCRF-BP [10],whichwasfoundinthebrainandthepituitaryandisthought toinhibittheactionsofCRFandUCN1[8].
Asregardstheanatomicaldistributionoftheligands,theyare represented both in the brain and the peripheral tissues [12].
CRFissynthesizedmainlyinthehypothalamus(PVN),thecentral nucleusoftheamygdalaandthehindbrainregionsintheCNS,and expressedinthegut,skin,andadrenalglandintheperiphery.UCN 1expressionhasbeendescribedpredominantly intheEdinger- Westphalnucleusinthebrainandexpressedinthegastrointestinal tract,testis, cardiac myocytes, thymus, skin, and spleen in the periphery.UCN2expressionhasbeendetectedinthehypotha- lamus(e.g.arcuatenucleus),thebrainstemandthespinalcordin thecentralnervoussystem(CNS),andintheheart,thebloodcells andtheadrenalglandintheperiphery.UCN3expressionhasbeen observedintheamygdala(e.g.medialnucleus)intheCNS,andin thegastrointestinaltractandthepancreasintheperiphery[13].
Asregardstheanatomical localizationofthereceptors,CRF1
receptorisrepresentedmoreanundantlyintheCNS,whereasCRF2 receptorisdistributedpredominantlyintheperiphery[12].Inthe CNS,CRF1receptorisexpressedinthecerebralcortex,thecerebel- lumandtheanteriorpituitary,butalsofoundintheamygdala,the striatumandthehypothalamus[14].CRF2receptorislimitedcen- trallytosubcorticalregions:theamygdala,thehippocampusand thehypothalamusandscatteredallovertheperiphery:theheart, thegastro-intestinaltract,thelung,theskeletalmusclesandthe vessels[14].
Besidestheirprincipalroleinthemodulationofstressresponses [15–18],regulationoffoodintakeandsatiety[19],gastrointesti- nalmotility[20],vasodilationandcardioprotection[21],CRFand theCRF-relatedpeptideshavebeenimplicatedinsocialinteraction [22–24],althoughtheirintimaterolewasnotinvestigatedthor- oughly[25,26].Hence,theaimofourstudywastodeterminethe roleofCRF,theurocortins(UCN1,UCN2andUCN3)andtheir receptors(CRF1 andCRF2)inthepreferenceforsocialnoveltyof mice.
2. Materialsandmethods 2.1. Animals
MaleandfemaleCFLPmiceweighing24–30gwereused.The totalnumberofmicewas504,ofwhich168malesand336females (dividedin168partnerfemalesand168strangerfemales).CFLP micewereusedbecauseweintendedtotestanoutbredstrainof mice,ratherthananinbredstrain,suchasC57/BL6mice,which, accordingtopreviousstudies,failtoexhibitpreferenceforsocial noveltyinthethree-chamberapparatus[27].Femalemicewere usedaspartnersorstrangersinsteadofthemalemice,becausewe aimedtoinvestigatethepreferenceforsocialnoveltyfollowingpair bondformation(male-femaleinteraction),andnotsocialaffiliation (male–maleinteraction).Four-fivemaleswerehousedtogetherin theirhomecagesandseparatedfromthestrangerfemales,butthen kepttogetherwiththepartnerfemalesfor24hbeforetheexperi- mentsstarted.Duringtheexperimentstheanimalswerekeptand handledinaccordancewiththeinstructionsoftheUniversityof SzegedEthicalComitteefortheProtectionofAnimalsinResearch, atconstantroomtemperature(23◦C)onastandardillumination schedule,with12-hlightand12-hdarkperiods(lightsonfrom6:00 a.m.).Commercialfoodandtapwaterwereavailableadlibitum.All experimentswereperformedinaccordancewiththeARRIVEguide- linesandtheU.K.Animals(ScientificProcedures)Act,1986and associatedguidelines,EUDirective2010/63/EUforanimalexperi- ments.
2.2. Surgery
The mice were implanted with a stainless steel Luer can- nula of 10mm lengthand 0.4mm diameteraimed atthe right lateralcerebralventricleunderanesthesiawith60mg/kgEuthana- sol (CEVA-Phylaxia,Hungary).The stereotaxiccoordinates were 0.5mmposteriorand0.5mmlateraltothebregma,and3mmdeep fromtheduralsurface,adaptedfromtheatlasdescribingthemouse braininstereotaxiccoordinates[28].Cannulasweresecuredtothe skullwithdentalcementandacrylate.Thepolyethylenetubeofthe injectorhaving0.8mmoutsidediameterand0.4mminsidediam- eterwasfittedcloselyintothecannulawithasharppointingtip projectingbeyondthecannula.Themicewereallowedfor5days torecoverafterthesurgeryandthepermeabilityofthecanullawas testedwithmethylene-blueaftertheexperiments.Thehitandmiss ratiowas168–143,whichmeans85%forrightpositioningofthe cannula.
2.3. Treatment
Fourexperimentswereperformedwithfourdifferentgroups.In experimentI,thefirstgroupwastreatedwith2lsalinesolution, thesecondonewith5g/2lCRF,thethirdonewith0.1g/2l of the selective CRF1 receptor antagonist antalarmin and the fourthonewith1g/2loftheselectiveCRF2receptorantagonist astressin2B,30minbeforethesocialinteractiontest.Inexperi- mentII,thefirstgroupwastreatedwith2lsaline,theremaining groups weretreated with5g/2l CRF,30minbeforethetest and pretreatedin orderwith2lsaline, 0.1g/2lantalarmin or1g/2lastressin2B,60minbeforethetest.InexperimentIII, thefirstgroupwastreatedwith2lsaline,thesecond,thethird andthefourthoneswith5g/2lUCN1,5g/2lUCN2and 5g/2lUCN3,respectively,30minbeforethetest.Inexperiment IV,thefirstgroupwastreatedwith2lsaline,theremaininggroups weretreatedwith5g/2lUCN1,30minbeforethetestandpre- treatedinorderwith2lsaline,0.1g/2lantalarminor1g/2l astressin2B60minbeforethetest.Thesaline(0.9%NaCl)solution wasprovidedbyBiogalLtd.,Hungary,CRF,UCN1,UCN2andUCN
3werepurchasedfromBachemLtd.,Switzerland,andantalarmin andastressin2BwerepurchasedfromSigmaAldrichInc.,USA.The ICVinfusionofthesubstanceswasperformedwithahand-held microinjector(HormuthLtd.,Germany)andthemice,beingprevi- ouslyhandleddailytominimizetheeffectsofnonspecificstress, werealsohand-heldduringinfusion.
2.4. Socialinteractiontest
ThirtyminutesaftertheICVinjectionofthepeptidesmicewere investigatedinasocialinteractiontestarenadescribedoriginallyby Crawleyandcolleaguesandusedtotestthesociabilityandthepref- erenceforsocialnoveltyofmice.[29].Inthepresentexperiments weusedamodifiedversionofthistestthatwasmeanttoinvesti- gatethepreferenceforsocialnoveltyfollowingpairbondformation (male-femaleinteraction), but not social affiliation (male–male interaction).Theprincipleofthetestisbasedontheassessment thatawildtypemousewouldvisitandspendmoretimewiththe strangerfemaleoverthepartnerfemale,indicativeforanintact socialmemoryandanaturalpreferenceforsocialnovelty.
Theapparatusisarectangular,three-chamberboxmadefrom clearPlexiglass.Eachchamberisof19×45×25cm,withanopen middlesection,whichallowsfreeaccesstoeachchamber.Theright andleftchamberscouldbeisolatedfromthemiddleonebyusing twodividingPlexiglasswalls.Twoidentical,wirecup-likecellsof 10×17cmwithremovablelidsthatlargeenoughtoholdasin- glemousewereplacedverticallyinsidetheapparatus,oneineach sidechamber.Eachcelliscomprisedofmetalwirestoallowfor airexchangebetweentheinteriorandexteriorofthecylinderbut smallenoughtopreventdirectphysicalinteractionsbetweenan animalontheinsidewithoneontheoutside.
Inthepresentexperimentsanunknownfemalemouse(stranger female)wassetinthefirstchamberandaknownfemalemouse (partner female), withwhich themale was familiarizedprevi- ouslyfor24h,wassetinthethirdchamber.Firstthetestedmale mousewashabituatedwiththemiddlechamberfor5minandthen allowedtoexploretheremainingchambersfor5min,duringwhich thenumberofentriesandthetimeofinteractionweremeasured.
Thetestedmalemousewashabituatedforthemiddlecompart- mentfor5min,butnottheothercompartments.Thepartnerand thestrangerfemalemiceweretransferredintheirhomecagesinto thebehavioralroom30minbeforethefirsttrial,buttheywerenot habituatedpreviouslytothecompartments.Thepartnerandthe strangermiceweresystematicallyalternatedbetweentheleftor rightcompartmentacrossthetrialstopreventsidepreference.All trialswereperformedbetween9:00a.m.and13:00p.m.General roomlightingwas650lx.Thepersonwhomadetheobservation wasatleast2mawayfromtheapparatus.Aftereachtrial,allcham- berswerecleanedwith70%ethanolandthenwithClidox1:5:1to preventolfactorycuebiasandtoensureproperdisinfection[30].
2.5. Statisticalanalysis
Statisticalanalysis of theresultswas performedby analysis ofvariance(ANOVA, GraphPadPrismSoftware).Thedifferences betweengroups were tested by two-way ANOVA, followed by Tukey post-hoc comparison test. A probability level of 0.05 or lesswasacceptedasindicatingastatisticallysignificantdifference (Tables1–4).
3. Results
CRF decreased significantly the number of entries (F(3,28)=5.489; p<0.001) and the time spent in interaction (F(3,28)=4.641; p<0.001) with the stranger female, but not that with thepartner female (F(3,28)=0.03286; p=0.9918 and
Table1
TheeffectsofCRF,antalarminandastressin2Bonthepreferenceforsocialnovelty ofmice.
Numberofentries
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 92.05 6 15.34 F(6,108):3.582 p=0.0028 Novelty 182.2 2 91.08 F(2,108):21.27 p<0.0001 Treatment 175.4 3 58.48 F(3,108):13.65 p<0.0001
Residual 462.5 108 4.282
Timeofinteraction
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 61947 6 10325 F(6,108):2.711 p=0.0172 Novelty 189754 2 94877 F(2,108):24.91 p<0.0001 Treatment 134586 3 44862 F(3,108):11.78 p<0.0001
Residual 411337 108 3809
Abbreviations:SSsumofsquares;DFtotaldegreesoffreedom;MSmeansquare;F (DFnDFd)Fdistribution(degreesoffreedomnumeratordegreesoffreedomdenom- inator);Pvalueprobabilityvalue.
Summaryofthestatisticaldatafromthesocialinteractiontest(two-wayANOVA withfactorA=noveltyandfactorB=treatmentwasperformed).
Table2
TheeffectsofCRF,CRFwithantalarminandCRFwithastressin2Bonthepreference forsocialnoveltyofmice.
Numberofentries
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 58.89 6 9.816 F(6,84)=3.905 p=0.0017
Novelty 50.29 2 25.14 F(2,84):10 p=0.0001
Treatment 98.72 3 32.91 F(3,84):13.09 p<0.0001
Residual 211.1 84 2.513
Timeofinteraction
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 43738 6 7290 F(6,84)=1.915 p=0.0877 Novelty 117742 2 58871 F(2,84):15.47 p<0.0001 Treatment 123125 3 41402 F(3,84):10.78 p<0.0001
Residual 319728 84 3806
Abbreviations:SSsumofsquares;DFtotaldegreesoffreedom;MSmeansquare;F (DFnDFd)Fdistribution(degreesoffreedomnumeratordegreesoffreedomdenom- inator);Pvalueprobabilityvalue.
Summaryofthestatisticaldatafromthesocialinteractiontest(two-wayANOVA withfactorA=noveltyandfactorB=treatmentwasperformed).
Table3
TheeffectsofUCN1,UCN2andUCN3onthepreferenceforsocialnoveltyofmice.
Numberofentries
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 18.2 6 3.034 F(6,105):0.789 p=0.5801 Novelty 235.9 2 118 F(2,105):30.7 p<0.0001 Treatment 129.8 3 43.28 F(3,105):11.26 p<0.0001
Residual 403.4 105 3.842
Timeofinteraction
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 2509 6 418.2 F(6,105):0.1446 p=0.9897 Novelty 268522 2 134261 F(2,105):46.43 p<0.0001 Treatment 4298 3 1433 F(3,105):0.495 p<0.6862
Residual 303621 105 2892
Abbreviations:SSsumofsquares;DFtotaldegreesoffreedom;MSmeansquare;F (DFnDFd)Fdistribution(degreesoffreedomnumeratordegreesoffreedomdenom- inator);Pvalueprobabilityvalue.
Summaryofthestatisticaldatafromthesocialinteractiontest(two-wayANOVA withfactorA=noveltyandfactorB=treatmentwasperformed).
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
into the partner's area into the stranger's area total entries
nu m ber of en tr ie s
control (10) CRF (10) antalarmin (10) astressin 2B (10)
*
*
0 50 100 150 200 250 300
with the partner with the stranger total time
tim e s pe nt (s )
control (10) CRF (10) antalarmin (10) astressin 2B (10)
*
*
Fig.1.TheeffectsofCRF,antalarminandastressin2Bonthepreferenceforsocialnoveltyofmice.Thenumberofanimalsusedineachgrouphasbeenindicatedinbrackets.
Valuesarepresentedasmeans±SEM;statisticallysignificantdifferencewasacceptedforp<0.05andindicatedwith*forCRForantalarminvs.control.
Table4
TheeffectsofUCN1,UCN1withantalarminandUCN1withastressin2Bonthe preferenceforsocialnoveltyofmice.
Numberofentries
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 22.99 6 3.832 F(6,84):1.965 p=0.0799 Novelty 157.1 2 78.54 F(2,84):40.26 p<0.0001 Treatment 156.5 3 52.17 F(3,84):26.74 p<0.0001
Residual 163.9 84 1.951
Table4(Continued) Timeofinteraction
ANOVAtable SS DF MS F(DFn,DFd) pvalue
Interaction 35465 6 5911 F(6,84):2.187 p:0.0521 Novelty 322153 2 161077 F(2,84):59.59 p<0.0001 Treatment 25498 3 8499 F(3,84):3.144 p=0.0294
Residual 227046 84 2703
Abbreviations:SSsumofsquares;DFtotaldegreesoffreedom;MSmeansquare;F (DFnDFd)Fdistribution(degreesoffreedomnumeratordegreesoffreedomdenom- inator);Pvalueprobabilityvalue.
Summaryofthestatisticaldatafromthesocialinteractiontest(two-wayANOVA withfactorA=noveltyandfactorB=treatmentwasperformed).
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
into the partner's area into the stran ger's area total entries
nu m ber of en tr ie s
control (10) CRF (10) CRF + antalarmin (6) CRF + astressin 2B (6)
*
*
#
#
0 50 100 150 200 250 300
with the partner with the stranger total time
tim e s pe nt (s )
control (10) CRF (10) CRF + antalarmin (6) CRF + astressin 2B (6)
*
*
#
#
Fig.2. TheeffectsofCRF,CRFwithantalarminandCRFwithastressin2Bonthepreferenceforsocialnoveltyofmice.Thenumberofanimalsusedineachgrouphasbeen indicatedinbrackets.Valuesarepresentedasmeans±SEM;statisticallysignificantdifferencewasacceptedforp<0.05andindicatedwith*forCRFvs.controland#forCRF withantalarminvs.CRF.
F(3,28)=0.4039,p=0.7513),comparedtothecontrols(Fig.1).In addition,thetotalnumberofentries(F(3.28)=16.04;p<0.001)and thetotal timeof interaction(F(3.28)=8.259;p<0.05)decreased significantly(Fig.1).Antalarmin and astressin2Badministered alonewereineffective(p<0.05)(Fig.1).Thedecreasingeffectsof CRFwereblockedbyantalarmin(p<0.05),but notastressin2B (p<0.05)(Fig.2).
UCN1decreasedsignificantlythenumberofentriesintothe chamberoftheunknownfemale(F(3,35)=6.277;p<0.001), but notintothatwiththepartnerfemale(F(3,35)=1.614;p=0.2036),
withoutchanging thetimesof interactionspentwiththepart- ner female (F(3,35)=0.144; p=0.09328) or the strangerfemale (F(3,35)=0.2763; p=0.8421), compared to the control (Fig. 3).
Thetotalnumberofentries(F(3,35)=4.827;p<0.001)decreased significantly, in contrast with the total time of interaction (F(3,35)=0.5033;p=0.6825)thatwasnotinfluencedconsiderably byUCN1(Fig.3).UCN2andUCN3didnotinfluenceremarkably anyoftheparametersmeasured(p>0.05)(Fig.3).Thedecreasing effectsofUCN1werereversedbyantalarmin(p<0.05),butnot astressin2B(p>0.05)(Fig.4).
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
into the partner's area into the strang er's area total entries
nu m ber of en tr ie s
control (10) urocortin 1 (10) urocortin 2 (10) urocortin 3 (9)
*
*
0 50 100 150 200 250 300
with the partner with the stranger total time
tim e s pe nt ( s)
control (10) urocortin 1 (10) urocortin 2 (10) urocortin 3 (9)
Fig.3. TheeffectsofUCN1,UCN2andUCN3onthepreferenceforsocialnoveltyofmice.Thenumberofanimalsusedineachgrouphasbeenindicatedinbrackets.Values arepresentedasmeans±SEM;statisticallysignificantdifferencewasacceptedforp<0.05andindicatedwith*forUCNvs.control.
4. Discussion
PreviousstudiesonthepossibleroleofCRFandCRF-relatedpep- tidesinsocialbehaviorofdifferentspecieshavebeenreviewedin tworecentstudies[25,26].Despitethattheprimaryfocusinthese studieshasbeenontheeffectsofsocialstressors,suchassocial defeatandsocialisolationontheCRFsystem[26],therehavebeen alsoinsightsontheroleofCRFsysteminprosocialandaffiliative behaviors,suchasparentalcare,maternaldefense,sexualbehav- iorandpairbonding [25].Theaimofthepresent studywasto investigatetheeffectsofthesepeptidesonthepreferenceforsocial noveltyfollowingpairbondformationinmice.
OurresultsdemonstratethatcentraladministrationofCRFand UCN1inducesadecreaseofthepreferenceforsocialnoveltyvia CRF1receptor,whichmayreflecttheanxiogenicactionoftheCRF1 receptor agonists[31,32].ButCRF1 receptoragonists mayexert theiranxiogenicactionthroughbothcentralandperipheraleffects [31,32].Thecentraleffectcanbemediatedbydifferentextrahy- pothalamicbrainregionsandneuronpopulationsinvolvedinstress reaction,suchasthebasolateralnucleusof theamygdala(BLA) [33].PreviousstudiesreferredthatinjectionofCRFandUCN1into theBLAreducedthetimesofsocialinteractioninmaleWistarrats andthatthiseffectwascompletelyabolishedbyadministrationof selectiveCRF1andnon-selectiveCRFreceptorantagonists[34–36].
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
into the partner's area into the stranger's area total entries
number of entries
control (10) urocortin 1 (10) urocortin 1 +antalarmin (6) urocortin 1 + astressin 2B (6)
*
*
#
#
0 50 100 150 200 250 300
with the partner with the stranger total time
time spent (s)
control (10) urocortin 1 (10) urocortin 1 + antalarmin (6) urocortin 1 + astressin 2B (6)
Fig.4.TheeffectsofUCN1,UCN1withantalarminandUCN1withastressin2Bonthepreferenceforsocialnoveltyofmice.Thenumberofanimalsusedineachgrouphas beenindicatedinbrackets.Valuesarepresentedasmeans±SEM;statisticallysignificantdifferencewasacceptedforp<0.05andindicatedwith*forUCN1vs.controland
#forUCN1withantalarminvs.UCN1.
UCN1provedevenmorepotentthanCRFinreducingthetimeof interaction,whichisconsistentwithitshigheraffinityfortheCRF1 receptor[34–36].TheperipheraleffectoftheCRF1 agonistscan bemediatedbyactivationoftheHPAaxis.Actually,ICVinjection ofthesamedose(5g/2l)ofCRForUCNIthatwasusedinthe presentexperimentsledtoelevationofplasmaACTHandcorticos- teroneconcentrationswithin30mininourpreviousexperiments [15].ThereforetheimpactoftheHPAaxisactivationonthebehav- iorofthemiceobservedinthethree-chambersocialinteraction testcannotbeexcluded.Inaddition,a previousstudyusingICV injectionofCRFinmaleListerrats,suggestedthattheanxiogenic effectofCRFobservedinaclassicalsocialinteractiontest[31],can becausedbythereleaseofpituitaryACTH,sincethishormonealso decreasedsocialinteractionwithoutreducinglocomotoractivity [31,32],butisunlikelytobeduebythereleaseofadrenalcorticos-
terone,sincethishormonedidnotdecreasesocialinteraction,and insomedosesincreasedit[31,32].
Our resultsalso demonstrate that central administration of UCN2and UCN3doesnotalter thepreference forsocial nov- elty,andgenerallythesocialinteractionofmalemicewithfemale mice.However,previousstudiesclaimedthatmale,butnotfemale, UCN2knock-outmiceexhibitedmorepassivesocialinteractions and reduced aggressivenesstonovel conspecifics[22] and that bothmaleandfemaleUCN3andCRF2receptorknock-outmice expressedanenhancedsocialmemoryandincreasedpreference forsocialnovelty,whencomparedtowild-typemice[23].These studiessuggestthatCRF2agonistswouldmodulatesomeaspects ofsocialbehavior.Incontrast,amorerecentstudystatedthatmice deficientinUCN3orCRF2 receptor localizedspecificallyinthe medialnucleusoftheamygdala(MEA)showeddecreasedprefer- enceforsocialnovelty[37].Moreover,pharmacologicalactivation
oftheCRF2receptorsoroptogeneticactivationofUCN3neurons intheMEAprovedtheopposite,anincreasedpreferenceforsocial noveltyofmice.ThisstudysuggeststhatUCN3fromtheMEAwould modulatetheabilityofmicetocopewithsocialchallengesviaCRF2 receptors[37].
Ourexperimentswereoriginallyinspiredbythreestudiesusing fourdifferentvolespecies (prairie,pine,montane and meadow voles),which,based ontheircloserelationship,butremarkably differentsocialbehavior(prairieandpinevolesaremonogamous, whilemontaneandmeadowvolesarepromiscuous),areconsid- eredemergingmodelorganismsforunderstandingthesocialbrain [38].Inthefirststudyusingreceptorautoradiography,amarked differencein thebrain distributionof CRF1 and CRF2 receptors between monogamous and promiscuous species (e.g., a higher expressionofCRF2receptorandalowerexpressionofCRF1recep- torinthenucleusaccumbens(NACC)oftheprairieandpinevoles), andmaleandfemalesex(e.g.,higherlevelsofCRF2receptorbind- ingintheencapsulatedbednucleusofstriaterminalis(BNST)in malevoles)wasdemonstrated.Inthesecondstudyusinginsitu hybridizationandimmunocytochemistry,nodifferencebetween thebrain distribution of CRF (the NACC, the BNST, thecentral nucleusoftheamygdalaetc.)andUCN1(theEdinger-Westphal nucleus)wasdemonstratedbetweenmonogamousandpromiscu- ousvoles[39].Inthethirdstudy,microinjectionofCRFandUCN1 intotheNACCinducedanaccelerationofthepartnerpreferencefor- mationinmonogamousprairievoles,butnotinnon-monogamous meadowvoles,thatwaseffectivelyblockedbyintraaccumbalinjec- tionofbothselectiveCRF1andselectiveCRF2receptorantagonists [40].Takingintoconsiderationthaturocortinergicfiberswerenot presentintheNACC,thisstudysuggeststhatthereleaseofCRFin theNACCfacilitatespartnerpreferenceinthemonogamousspecies activatingbothCRFreceptors[40].Thelackofthisfacilitationeffect inthenon-monogamousspeciescanbeexplainedbythelackof thecorrespondentreceptorsintheirbrain,despiteofthesimilar distributionoftheirligands[39,41].
Inconclusion,thepresentstudysuggeststhatCRFandUCN1 decreasethepreferenceforsocialnoveltybyactivatingCRF1recep- tors,whileUCN2andUCN3,activatingselectivelyCRF2receptors, donotparticipatetomale-femaleinteractioninthepromiscuous CFLPmice.Thisconclusioncanbeunderlinedbythehigherexpres- sionofCRF1 receptorsandlowerexpressionofCRF2receptorsin thebrainsofnon-monogamousspecieswhichhavebeenpreviously demonstrated.
Acknowledgements
ThisstudywassupportedbytheNeuroscienceResearchGroup oftheHungarianAcademyof SciencesandtheHungarianBrain ResearchProgram.
References
[1]W.Vale,J.Spiess,C.Rivier,J.Rivier,Characterizationofa41-residueovine hypothalamicpeptidethatstimulatessecretionofcorticotropinand beta-endorphin,Science213(1981)1394–1397.
[2]C.Tsigos,G.P.Chrousos,Hypothalamic-pituitary-adrenalaxis,
neuroendocrinefactorsandstress,J.Psychosom.Res.53(2002)865–871.
[3]G.A.Carrasco,L.D.VandeKar,Neuroendocrinepharmacologyofstress,Eur.J.
Pharmacol.463(2003)235–272.
[4]J.Vaughan,C.Donaldson,J.Bittencourt,M.H.Perrin,K.Lewis,S.Sutton,etal., Urocortin,amammalianneuropeptiderelatedtofishurotensinIandto corticotropin-releasingfactor,Nature378(1995)287–292.
[5]T.M.Reyes,K.Lewis,M.H.Perrin,K.S.Kunitake,J.Vaughan,C.A.Arias,etal., UrocortinII:amemberofthecorticotropin-releasingfactor(CRF) neuropeptidefamilythatisselectivelyboundbytype2CRFreceptors,Proc.
Natl.Acad.Sci.U.S.A.98(2001)2843–2848.
[6]K.Lewis,C.Li,M.H.Perrin,A.Blount,K.Kunitake,C.Donaldson,etal., IdentificationofurocortinIII,anadditionalmemberofthe
corticotropin-releasingfactor(CRF)familywithhighaffinityfortheCRF2 receptor,Proc.Natl.Acad.Sci.U.S.A.98(2001)7570–7575.
[7]C.P.Chang,R.V.Pearse,2nd,O’ConnellS,RosenfeldMG.Identificationofa seventransmembranehelixreceptorforcorticotropin-releasingfactorand sauvagineinmammalianbrain,Neuron11(1993)1187–1195.
[8]D.P.Behan,E.B.DeSouza,P.J.Lowry,E.Potter,P.Sawchenko,W.W.Vale, Corticotropinreleasingfactor(CRF)bindingprotein:anovelregulatorofCRF andrelatedpeptides,Front.Neuroendocrinol.16(1995)362–382.
[9]K.H.Skelton,M.J.Owens,C.B.Nemeroff,Theneurobiologyofurocortin,Regul.
Pept.93(2000)85–92.
[10]E.M.Fekete,E.P.Zorrilla,Physiology,pharmacology,andtherapeutic relevanceofurocortinsinmammals:ancientCRFparalogs,Front.
Neuroendocrinol.28(2007)1–27.
[11]D.K.Grammatopoulos,H.S.Randeva,M.A.Levine,K.A.Kanellopoulou,E.W.
Hillhouse,Ratcerebralcortexcorticotropin-releasinghormonereceptors:
evidenceforreceptorcouplingtomultipleG-proteins,J.Neurochem.76 (2001)509–519.
[12]F.M.Dautzenberg,R.L.Hauger,C.R.F.The,peptidefamilyandtheirreceptors:
yetmorepartnersdiscovered,TrendsPharmacol.Sci.23(2002)71–77.
[13]J.M.Reul,F.Holsboer,Corticotropin-releasingfactorreceptors1and2in anxietyanddepression,Curr.Opin.Pharmacol.2(2002)23–33.
[14]K.VanPett,V.Viau,J.C.Bittencourt,R.K.Chan,H.Y.Li,C.Arias,etal., DistributionofmRNAsencodingCRFreceptorsinbrainandpituitaryofrat andmouse,J.Comp.Neurol.428(2000)191–212.
[15]Z.Bagosi,K.Csabafi,M.Palotai,M.Jaszberenyi,I.Foldesi,J.Gardi,etal.,The effectofurocortinIonthehypothalamicACTHsecretagoguesanditsimpact onthehypothalamic-pituitary-adrenalaxis,Neuropeptides48(2014)15–20.
[16]T.Suda,K.Kageyama,S.Sakihara,T.Nigawara,Physiologicalrolesof urocortins,humanhomologuesoffishurotensinI,andtheirreceptors, Peptides25(2004)1689–1701.
[17]Z.Bagosi,K.Csabafi,M.Jaszberenyi,G.Telegdy,Theeffectsof corticotropin-releasingfactorandtheurocortinsonhypothalamic gamma-aminobutyricacidrelease–theimpactsonthe
hypothalamic-pituitary-adrenalaxis,Neurochem.Int.60(2012)350–354.
[18]Z.Bagosi,K.Csabafi,M.Palotai,M.Jaszberenyi,I.Foldesi,J.Gardi,etal.,The interactionofUrocortinIIandUrocortinIIIwithamygdalarandhypothalamic cotricotropin-releasingfactor(CRF)–reflectionsontheregulationofthe hypothalamic-pituitary-adrenal(HPA)axis,Neuropeptides47(2013) 333–338.
[19]D.Richard,Q.Lin,E.Timofeeva,Thecorticotropin-releasingfactorfamilyof peptidesandCRFreceptors:theirrolesintheregulationofenergybalance, Eur.J.Pharmacol.440(2002)189–197.
[20]V.Martinez,L.Wang,J.E.Rivier,W.Vale,Y.Tache,Differentialactionsof peripheralcorticotropin-releasingfactor(CRF),urocortinII,andurocortinIII ongastricemptyingandcolonictransitinmice:roleofCRFreceptorsubtypes 1and2,J.Pharmacol.Exp.Ther.301(2002)611–617.
[21]K.Takahashi,K.Totsune,O.Murakami,S.Shibahara,Urocortinsas cardiovascularpeptides,Peptides25(2004)1723–1731.
[22]J.Breu,C.Touma,S.M.Holter,A.Knapman,W.Wurst,J.M.Deussing,Urocortin 2modulatesaspectsofsocialbehaviourinmice,Behav.BrainRes.233(2012) 331–336.
[23]J.M.Deussing,J.Breu,C.Kuhne,M.Kallnik,M.Bunck,L.Glasl,etal.,Urocortin 3modulatessocialdiscriminationabilitiesviacorticotropin-releasing hormonereceptortype2,J.Neurosci.30(2010)9103–9116.
[24]D.R.Gehlert,A.Shekhar,S.M.Morin,P.A.Hipskind,C.Zink,S.L.Gackenheimer, etal.,StressandcentralUrocortinincreaseanxiety-likebehaviorinthesocial interactiontestviatheCRF1receptor,Eur.J.Pharmacol.509(2005)145–153.
[25]C.M.Hostetler,A.E.Ryabinin,C.R.F.The,systemandsocialbehavior:areview, Front.Neurosci.7(2013)92.
[26]T.Backstrom,S.Winberg,Centralcorticotropinreleasingfactorandsocial stress,Front.Neurosci.7(2013)117.
[27]B.L.Pearson,E.B.Defensor,D.C.Blanchard,R.J.Blanchard,C57BL/6Jmicefailto exhibitpreferenceforsocialnoveltyinthethree-chamberapparatus,Behav.
BrainRes.213(2010)189–194.
[28]G.Paxinos,K.B.J.Franklin,TheMouseBraininStereotaxicCoordinates.
Compact,2nded.,ElsevierAcademicPress,Amsterdam;Boston,2004.
[29]J.N.Crawley,What’sWrongwithMyMouse?:BehavioralPhenotypingof TransgenicandKnockoutMice,2nded.,Wiley-Interscience,N.J.Hoboken, 2007.
[30]O.Kaidanovich-Beilin,T.Lipina,I.Vukobradovic,J.Roder,J.R.Woodgett, Assessmentofsocialinteractionbehaviors,J.Vis.Exp.(2011).
[31]A.J.Dunn,S.E.File,Corticotropin-releasingfactorhasananxiogenicactionin thesocialinteractiontest,Horm.Behav.21(1987)193–202.
[32]S.E.File,P.Seth,Areviewof25yearsofthesocialinteractiontest,Eur.J.
Pharmacol.463(2003)35–53.
[33]D.Janssen,T.Kozicz,Isitreallyamatterofsimpledualism?
Corticotropin-releasingfactorreceptorsinbodyandmentalhealth,Front Endocrinol(Lausanne)4(2013)28.
[34]T.J.Sajdyk,D.A.Schober,D.R.Gehlert,A.Shekhar,Roleof
corticotropin-releasingfactorandurocortinwithinthebasolateralamygdala ofratsinanxietyandpanicresponses,Behav.BrainRes.100(1999)207–215.
[35]B.M.Campbell,J.L.Morrison,E.L.Walker,K.M.Merchant,Differential regulationofbehavioral,genomic,andneuroendocrineresponsesbyCRF infusionsinrats,Pharmacol.Biochem.Behav.77(2004)447–455.
[36]F.Spiga,S.L.Lightman,A.Shekhar,C.A.Lowry,Injectionsofurocortin1into thebasolateralamygdalainduceanxiety-likebehaviorandc-Fosexpression inbrainstemserotonergicneurons,Neuroscience138(2006)1265–1276.
[37]Y.Shemesh,O.Forkosh,M.Mahn,S.Anpilov,Y.Sztainberg,S.Manashirov, etal.,Ucn3andCRF-R2inthemedialamygdalaregulatecomplexsocial dynamics,Nat.Neurosci.19(2016)1489–1496.
[38]L.A.McGraw,L.J.Young,Theprairievole:anemergingmodelorganismfor understandingthesocialbrain,TrendsNeurosci.33(2010)103–109.
[39]M.M.Lim,N.O.Tsivkovskaia,Y.Bai,L.J.Young,A.E.Ryabinin,Distributionof corticotropin-releasingfactorandurocortin1inthevolebrain,BrainBehav.
Evol.68(2006)229–240.
[40]M.M.Lim,Y.Liu,A.E.Ryabinin,Y.Bai,Z.Wang,L.J.Young,CRFreceptorsinthe nucleusaccumbensmodulatepartnerpreferenceinprairievoles,Horm.
Behav.51(2007)508–515.
[41]M.M.Lim,H.P.Nair,L.J.Young,Speciesandsexdifferencesinbrain distributionofcorticotropin-releasingfactorreceptorsubtypes1and2in monogamousandpromiscuousvolespecies,J.Comp.Neurol.487(2005) 75–92.