Effects of cellular interactions on calcium dynamics in prolactin-secreting cells

E ec s o Cellula In e ac ions on Calcium Dynamics in P olac in-Sec e ing Cells* ELIZABETH J. ABRAHAM, CARLOS VILLALOBOS, AND L. STEPHEN FRAWLEY Labo a o y o Molecula Dynamics, Depa men o Cell Biology and Ana omy, Medical Uni e si y o Sou h Ca olina, Cha les on, Sou h Ca olina 29425 ABSTRACT Signals de i ed om o he pi ui a y cells can ha e a d ama ic e ec on PRL gene exp ession and sec e ion by mammo opes. How- e e , he in acellula mechanisms by which hese e ec s a e man- i es ed on he a ge cell emain unexplo ed. Inasmuch as calcium is a key modula o o bo h gene exp ession and ho mone expo in mammo opes, we e alua ed he e ec s o cell o cell con ac s. speci ic cellula in e ac ions on calcium dynamics wi hin hese cells. This was accomplished by digi al-imaging luo escence mic oscopy o u a-2 in pi ui a y cells ha we e isola ed in cul u e (singles) o adjoining one o he cell (double s). A e calcium imaging, we hen subjec ed cells o immunocy ochemis y o PRL. Double s we e u - he ca ego ized in o mammo opes a ached o ano he mammo ope (M-M) o o a nonmammo ope (M-nonM). We hen calcula ed and compa ed Mean [Ca 21 ] i alues as well as Oscilla ion Indices (which e lec he oscilla o y beha io o cells) in singles and double s and ound ha hey we e no di e en (P.0.05). Howe e , he pheno ype o he adjoining cell had a p o ound in luence on bo h o hese calcium pa ame e s, such ha he p esence o one mammo ope could con- sis en ly dec ease (P,0.05) he Mean [Ca 21 ] i alue (39.17 63.83 s. 56.24 65.56 in M-nonM) and Oscilla ion Index (10.19 61.76 s. 21.21 63.73 in M-nonM) o i s neighbo ing coun e pa . A mo e de ailed analysis o oscilla o y pa e ns in hese cells e ealed ha nonoscilla o s we e mo e abundan in M-M (23%) han in M-nonM (12%) double s. Taken oge he , ou esul s indica e ha PRL- sec e ing cells con ey a signal ha dampens he oscilla o y beha io o neighbo ing mammo opes. Thus, i appea s ha i is he pheno- ype a he han he physical p esence o a neighbo ha con ols in e cellula egula ion o calcium dynamics among mammo opes. (Endoc inology 139: 2988–2993, 1998) THERE IS NOW o e whelming expe imen al e idence o suppo he exis ence and impo ance o in e cellula communica ion among cells o he an e io pi ui a y gland, and he PRL-sec e ing mammo ope has been a model o choice o s udies o his phenomenon. E idence in suppo o his iew de i es om s udies demons a ing ha sec e- ion o a ho mone, such as PRL, is in luenced by he addi ion o elimina ion o o he sec e o y cell ypes o he pi ui a y (see Re s. 1–3 o e iews). Mo eo e , cell densi y-dependen modula ion o ho mone sec e ion wi hin pi ui a y cul u es also sugges s he exis ence o cell-cell in e ac ions (4, 5). In- deed, mammo opes exis in a h ee-dimensional con igu a- ion o a issue. He e, hey a e in in ima e con ac wi h o he cells om which hey ecei e a a ie y o signals ha ha e he po en ial o egula e bo h gene exp ession and ho mone elease. Ye , despi e he weal h o in o ma ion a o ing in- e cellula con ol o mammo ope unc ion, su p isingly li - le is known abou how hese e ec s a e ansduced wi hin he a ge cell. One pa icula ly s ong candida e o unc ion as an in e media y in his p ocess is calcium. Clea ly, calcium plays a pi o al ole in cell o cell adhesion o many cell ypes, including hose o he pi ui a y, and cell-cell o cell-ma ix adhesions ha e been shown o d ama ically in luence he exp ession o a numbe o ho monal genes (6–9). Like mos pi ui a y cells, mammo opes exhibi spon aneous oscilla- ions o in acellula ee calcium ([Ca 21 ] i ) ha a e d i en by elec ical ac i i y (10–12) and a e posi i ely co ela ed wi h basal ho mone elease (13, 14). In addi ion, pha macological manipula ion o [Ca 21 ] i wi h agonis s o an agonis s e okes p edic able changes o PRL elease and messenge RNA ac- cumula ion (15, 16). Thus, calcium is a pi o al egula o o he PRL biosyn he ic pa hway om he poin o gene ansc ip- ion o ha o ho mone expo , bu he e ec s o cell o cell signaling on [Ca 21 ] i dynamics emain unexplo ed. In a ecen s udy on in e cellula communica ion and gene exp ession, we a emp ed o pa i ion he e ec s upon a mammo ope o ha ing a neighbo (cell o cell con ac ) as opposed o he na u e (pheno ype) o ha neighbo , and ound ha he la e conside a ion was a mo e impo an o PRL gene exp ession han he o me (17). Ou s a egy o ha in es iga ion was o make “ eal- ime” measu emen s o PRL gene exp ession in li ing, p ima y pi ui a y cells and o ollow his wi h immunocy ochemis y ( o pos ac o iden- i ica ion o he pheno ype o a gi en cell o i s neighbo ). The pa adigm o he p esen s udy was concep ually iden ical o his, wi h he only majo depa u e being ha digi al-imag- ing luo escence mic oscopy o a calcium-sensi i e dye was subs i u ed o analysis o gene exp ession. In his manne , we could a emp o de e mine whe he he in e nal calcium dynamics o a mammo ope we e also modula ed by he p esence and/o sec e o y na u e o a neighbo . Ma e ials and Me hods Animals and cell cul u e An e io pi ui a y glands om p imipa ous lac a ing a s (days 5–12 pos pa um) we e collec ed a e decapi a ion and enzyma ically dis- pe sed in o single cells as desc ibed p e iously (18). Cells we e allowed Recei ed Decembe 22, 1997. Add ess all co espondence and eques s o ep in s o: D . L. S ephen F awley, Labo a o y o Molecula Dynamics, Depa men o Cell Biology and Ana omy, Medical Uni e si y o Sou h Ca olina, Cha les on, Sou h Ca olina 29425. E-mail: [email p o ec ed]. * This s udy was suppo ed by unds om NIH G an DK-38215 and a USDA-Compe i i e Resea ch G an (95-37206-2438). 0013-7227/98/$03.00/0 Vol. 139, No. 6 Endoc inology P in ed in U.S.A. Copy igh © 1998 by The Endoc ine Socie y 2988 Downloaded om h ps://academic.oup.com/endo/a icle/139/6/2988/2987337 by UNIVERSIDAD DE VALLADOLID. BIBLIOTECA use on 19 Janua y 2021 o a ach on o poly-lysine coa ed, g idded, glass co e slips a a densi y o 250,000 cells/90 m l de ined medium [equal olumes o phenol- ee M-199 wi h Nu ien mix F-12 (Li e Technologies, G and Island, NY), supplemen ed wi h 0.1% BSA, insulin- ans e in-selenium P emix, and an ibio ics] and hen placed in o 35-mm plas ic Pe i dishes. A e a 1-h a achmen pe iod, de ined medium con aining 5% FBS was added, and he cells we e cul u ed a 37 C in 95% ai : 5% CO 2 o 2 days un il hey we e used o [Ca 21 ] i imaging o gap-junc ion coupling expe imen s. Digi al-imaging luo escence mic oscopy o u a-2 Loading o cells. On day 2 o cul u e, cells we e washed and loaded wi h 2 m mo he calcium-sensi i e dye u a-2/AM (Molecula P obes, Eugene, OR) in phenol- ee DMEM (Li e Technologies) supplemen ed wi h 5 mmd-glucose, 2 mml-glu amine, 1 mmsodium py u a e, 5 mm sodium bica bona e, 10 mmHEPES, and 34 mmsodium chlo ide (pH 7.4). A e 1.5 h, cells we e washed 3 imes wi h he same medium and placed on he hea ed s age (a 37 C) o an in e ed mic oscope equipped wi h a 403objec i e. Cells we e al e na ely epi-illumina ed wi h a UV-ligh a 340 and 380 nm exci a ion wa eleng h. Ligh emi ed abo e 520 nm was eco ded o 20 min and analyzed using an A o luo Ra io Vision Sys em (A o Ins umen s, Rock ille, MD). Fo e e y aw image, a o al o 4 ideo ames (2 ames pe wa eleng h) we e acqui ed. Then, a a io o he mean luo escence in ensi y a 340/380 was calcula ed wi h an o e all ime esolu ion o 4 sec. Calib a ion o luo escence a ios [R] was pe o med acco ding o he o mula desc ibed by G ynkiewicz e al. (19). Fu a-2 s anda ds we e used o cons uc a calib a ion cu e. This cu e was hen used o con e a ios [R] in o [Ca 21 ] i (20, 21). Analysis o [Ca 21 ] i .In he cu en s udy, we calcula ed wo unc ional pa- ame e s o Ca 21 dynamics: he Mean [Ca 21 ] i alue and he Oscilla ion Index. As desc ibed elsewhe e, hese pa ame e s e lec dis inc ly di e en ways o e alua ing he same da a se (22, 23). The Mean [Ca 21 ] i was ob ained by in eg a ing alues collec ed a 4-sec in e als and di iding hem by ime o measu emen . The Oscilla ion Index, on he o he hand, deno es he ela i e change in [Ca 21 ] i du ing he collec ion pe iod. We achie ed his by calcula ing he di e ence in [Ca 21 ] i alues be ween con- secu i e in e als o measu emen o e he du a ion o sampling. Thus, i is possible ha a gi en cell could ha e a high Oscilla ion Index (because o high equency oscilla ions) e en hough i s Mean [Ca 21 ] i alue is low, o ice- e sa. Immunocy ochemis y. A e [Ca 21 ] i imaging, cells we e washed and ixed wi h B-5 bu e ed o malin o 45 min. They we e hen subjec ed o immunocy ochemical de ec ion o PRL as con olled and desc ibed p e iously (18). PRL-posi i e cells we e e ealed a e exposu e o dia- minobenzidine (Sigma, S . Louis, MO) solu ion o 6 min, and he same ield p e iously used o [Ca 21 ] i imaging was eiden i ied by i s posi ion on he co e slip, which was pho oeng a ed wi h a numbe ed/le e ed g id pa e n. Gap-junc ions The possible exis ence o unc ional gap-junc ions be ween an e io pi ui a y cells was de e mined by comic oinjec ing Luci e yellow (4%) and dex an- hodamine (0.1%) in o one o a pai o adjoining cells. Mic oinjec ion was achie ed h ough inely pulled, glass capilla ies in se ies wi h an Eppendo semiau oma ed mic oinjec ion sys em (17). Ten minu es pos injec ion, co e slips we e assembled in o Sykes-Moo e chambe s ha we e illed wi h se um- ee DMEM. This assembly was hen ans e ed o he s age o an Axiopho mic oscope equipped wi h a UV ligh sou ce and app op ia e exci a ion-emission il e s o luo- escen isualiza ion o Luci e yellow and hodamine. As a posi i e con ol, we also coinjec ed con luen cul u es o GT1–1 neu ons (a clonal GnRH cell line) wi h he same a io o luo escen dyes used o pi ui a y cells. P e ious epo s by o he in es iga o s con i med he p esence o unc ional gap-junc ions in he majo i y o hese neu ons (24). S a is ical analysis Compa isons be ween any wo ea men g oups we e made wi h a wo- ailed, S uden ’s es and esul s we e exp essed as mean 1sem. Da a epo ed in Table 1 we e analyzed using a wo-way ANOVA ollowed by a mul iple compa isons es (Fishe ’s leas signi ican di - e ence). Di e ences we e conside ed signi ican a P,0.05. Resul s Ou ini ial aim was o e alua e he e ec o cell o cell con ac on [Ca 21 ] i oscilla ions in mammo opes. To his end, FIG. 1. Expe imen al p o ocol employed in his s udy. A, An e io pi- ui a y cells (one double shown) we e subjec ed o [Ca 21 ] i measu e- men s by digi al-imaging luo escence mic oscopy o u a-2. The pe im- e e o each box dema ca es he a ea in which all luo escence measu emen s we e eco ded in ha cell. No e ha each box co esponds o indi idual cell and does no o e lap wi h he adjacen one. Rela i e [Ca 21 ] i is indica ed in panel A by he wa me colo s on he ascending pseudocolo scale o he igh . B, A e [Ca 21 ] i measu emen s, he same cells we e ixed and subjec ed o immunocy ochemis y o PRL. In his pa icula ins ance, he double shown is comp ised o wo mammo- opes. C, [Ca 21 ] i ansien s o he same cells a e illus a ed. No e ha he spon aneous oscilla ions we e asynch onous. PARACRINE COMMUNICATION AND CALCIUM DYNAMICS IN MAMMOTROPES 2989 Downloaded om h ps://academic.oup.com/endo/a icle/139/6/2988/2987337 by UNIVERSIDAD DE VALLADOLID. BIBLIOTECA use on 19 Janua y 2021 we combined digi al-imaging luo escence mic oscopy o u a-2 wi h immunocy ochemis y o PRL (Fig. 1, A and B). Iden i ied mammo opes ha we e cul u ed in isola ion (sin- gles) o in physical con ac wi h jus one o he cell (double s) we e hen compa ed in e ms o he a o emen ioned [Ca 21 ] i pa ame e s. As shown in Fig. 2, Oscilla ion Index alues o single mammo opes and hose in he double con igu a ion we e i ually iden ical (P.0.05). Likewise, he Mean [Ca 21 ] i alues o singles and double s we e also indis in- guishable (P.0.05). These esul s demons a e ha cell o cell con ac pe se does no in luence Ca 21 dynamics in mam- mo opes cul u ed unde basal condi ions. A e de e mining ha he me e p esence o ano he cell had no measu able e ec on [Ca 21 ] i in mammo opes, we nex assessed whe he he na u e (pheno ype) o he neighbo migh ha e an in luence. Acco dingly, we pe o med [Ca 21 ] i imaging on only hose cells in he double con igu a ion and ollowed his wi h immunocy ochemical de ec ion o PRL as be o e. Iden i ied mammo opes in he double con igu a ion we e subsequen ly ca ego ized on he basis o whe he hey we e a ached o ano he mammo ope (M-M) o o a nonmammo- ope (M-nonM; in hese cells [Ca 21 ] i alues o mammo opes and no hose o nonmammo opes we e analyzed). In e es - ingly, we ound ha bo h Oscilla ion Index and he Mean [Ca 21 ] i alues we e signi ican ly lowe (P,0.05) o he M-M double s when compa ed wi h hei M-nonM coun e pa s (Fig. 3). Inasmuch as “a e aged” esponses do no e lec he oscil- la o y beha io o indi idual mammo opes in he M-M o M-nonM ca ego ies, we plo ed he Mean [Ca 21 ] i alue o each mammo ope agains i s co esponding Oscilla ion Index alue. As shown in Fig. 4, double s in he M-nonM con igu a- FIG. 2. Cell- o-cell con ac has no e ec on spon aneous [Ca 21 ] i os- cilla ions. Ba s (mean 1SEM) in panels A and B ep esen Oscilla ion Index and Mean [Ca 21 ] i o mammo opes ha we e isola ed om o he cells (singles; n 5160 cells) o in con ac wi h one o he cell (double s; n 5141 cells). Cells ob ained om nine di e en dispe - sions we e used. No e ha he [Ca 21 ] i alue in he aces shown in Figs. 1C and 5 a e highe han hose depic ed in Figs. 2 and 3 because he ace ep esen s [Ca 21 ] i alues eco ded a 4-sec in e als, whe eas Mean [Ca 21 ] i ep esen s he sum o all eco ded alues di- ided by o al ime (i.e. nM/sec). FIG. 3. Spon aneous [Ca 21 ] i oscilla ions a e in luenced by he phe- no ype o a neighbo ing cell. Ba s (mean 1SEM) in panels A and B ep esen Oscilla ion Indices and Mean [Ca 21 ] i alues, espec i ely, o mammo opes in con ac wi h ano he mammo ope (M-M; n 552 cells) o a nonmammo ope (M-nonM; n 589 cells). Cells ob ained om nine di e en dispe sions we e used. *, P,0.05 s. M-nonM. 2990 PARACRINE COMMUNICATION AND CALCIUM DYNAMICS IN MAMMOTROPES Endo •1998 Vol 139 •No 6 Downloaded om h ps://academic.oup.com/endo/a icle/139/6/2988/2987337 by UNIVERSIDAD DE VALLADOLID. BIBLIOTECA use on 19 Janua y 2021 ion and singles displayed a disc e e subpopula ion wi h a ela i ely high Oscilla ion Index and Mean [Ca 21 ] i alue. A co esponding g oup o highly ac i e mammo opes was i - ually absen o double s in he M-M ca ego y. Thus, he in- hibi o y e ec o one mammo ope upon ano he appea s o be mani es ed p e e en ially upon hose cells ha a e mos ac i e in e ms o Ca 21 dynamics. Ha ing es ablished ha he a e age oscilla o y ac i i y o M-M cells is lowe han o M-nonM cells, we hen se ou o de e mine he basis o hese di e ences. This was accom- plished by analyzing he pa e n o spon aneous [Ca 21 ] i os- cilla ions in M-M and M-nonM double s. P e ious s udies conduc ed in ou labo a o y success ully cha ac e ized spon- aneous [Ca 21 ] i oscilla ions o mammo opes in o he ol- lowing majo pa e ns: (A) nonoscilla ing o “silen ”; (B) high equency, low ampli ude oscilla ions; and (C) high e- quency, high ampli ude oscilla ions (20). In he p esen s udy, we ound ha al hough M-M and M-nonM cells dis- played simila ypes o oscilla o y p o iles (Fig. 5), he el- a i e p opo ions exhibi ing pa e ns A and C as well as he co esponding Oscilla ion Indices and Mean [Ca 21 ] i alues o he la e p o ile we e s ikingly di e en (Table 1). Spe- ci ically, nonoscilla o s we e 2- old mo e abundan o M-M (23%) han o M-nonM (12%) cells. Mo e impo an ly, he Oscilla ion Index and Mean [Ca 21 ] i o “C” ype oscilla o s we e 2- old highe in M-nonM han in M-M cells. Thus, he highe equency and ampli ude o spon aneous [Ca 21 ] i os- cilla ions seen in hose mammo opes associa ed wi h non- mammo opes appea o con ibu e, a leas in pa , o he inc eased oscilla o y ac i i y obse ed in his subse o PRL sec e o s. Con e sely, dec eased oscilla o y ac i i y in neigh- bo ing mammo opes may in pa be due o an inc ease in he p opo ion o nonoscilla o s in his con igu a ion. Al hough ou esul s showed clea ly ha Ca 21 dynamics o mammo opes we e in luenced by he neighbo ing cell, hey did no p o ide any insigh s as o whe he he com- munica ion was pa ac ine o jux ac ine. In an a emp o dis inguish be ween hese possibili ies, we coinjec ed Luci e yellow and dex an- hodamine in o cells main ained as dou- ble s. The o me luo escen dye passes eadily h ough gap-junc ions, whe eas he cell memb ane is impe meable o he la e . Thus, he hodamine complex se ed bo h as a means o iden i ying he injec ed cell and as a con ol o memb ane in eg i y. O 137 double s s udied in his manne , only h ee exhibi ed unc ional gap-junc ions, whe eas he emainde showed no dye- ans e be ween adjoining cells. Inasmuch as 60–65% o he cells in ou pi ui a y cul u es we e PRL-sec e o s, hese da a demons a e clea ly ha jux- ac ine signaling media ed by gap-junc ion coupling canno accoun o he in e ac ions we obse ed be ween mammo- opes and hei neighbo s. In con as , gap-junc ion cou- pling was obse ed in he majo i y o GT1–1 neu ons coin- jec ed wi h he same mix o dyes. A inal poin wo hy o men ion is ha we ound no e idence o sugges ha adjacen cells we e elec ically coupled. As illus a ed by he ep e- sen a i e example in Fig. 1C, cells in he M-M con igu a ion (o M-nonM, o ha ma e ) we e jus as likely o exhibi dissimila as simila pa e ns o [Ca 21 ] i oscilla ions. Discussion Ou esul s p o ide compelling e idence ha cell o cell communica ion has a d ama ic in luence on [Ca 21 ] i dynam- ics wi hin li ing mammo opes. To be mo e speci ic, we ound ha he physical p esence o a neighbo ing cell pe se had no in luence on he Ca 21 ac i i y o a gi en mammo ope bu ha he sec e o y na u e o he neighbo was o pa a- moun impo ance. Indeed, when one mammo ope was in close physical apposi ion wi h ano he , he e was a s iking FIG. 4. Sca e plo s o mammo opes main ained as singles, M-nonM, and M-M double s a e depic ed he e. No e ha subpopu- la ions o mammo opes ha exhibi high Oscilla ion Indices and Mean [Ca 21 ] i alues a e es ic ed o single mammo opes and hose in he M-nonM con igu a ion. The aw da a om expe imen s shown in Figs. 2 and 3 we e used o his plo . PARACRINE COMMUNICATION AND CALCIUM DYNAMICS IN MAMMOTROPES 2991 Downloaded om h ps://academic.oup.com/endo/a icle/139/6/2988/2987337 by UNIVERSIDAD DE VALLADOLID. BIBLIOTECA use on 19 Janua y 2021 educ ion o he Oscilla ion Index (51.9%), and, o a lesse ex en , an a enua ion o he Mean [Ca 21 ] i alue (30.3%). Inasmuch as he o me pa ame e is a mo e di ec and e- liable indica o o equency/ampli ude changes han he la e , i seems easonable o p opose ha mos o his in- hibi o y e ec was a ibu able o a diminu ion o oscilla o y ac i i y as opposed o modula ion o he a e age amoun o Ca 21 p esen . Mo eo e , analysis o spon aneous [Ca 21 ] i os- cilla ions indica e ha inhibi ion o such ac i i y in M-M cells is no a ibu able o he p esence o absence o disc e e oscilla o y pheno ypes exhibi ed by M-nonM double s. In- s ead, i is due o a di e en dis ibu ion o mammo opes exhibi ing each pa e n. To be mo e speci ic, he e was a much highe p opo ion o nonoscilla o y cells in he M-M g oup han in he M-nonM ca ego y, and his inc emen was achie ed la gely a he expense o he high equency, high ampli ude oscilla o s. I appea s, hen, ha PRL-sec e ing cells con ey a signal ha dampens he oscilla o y beha io o neighbo ing mammo opes. O cou se, an al e na i e in e - p e a ion ha dese es conside a ion is ha nonmammo- opes emi a signal ha s imula es [Ca 21 ] i oscilla ions in adjoining mammo opes. I his we e he case, one would expec o see a p eponde ance o cells wi h bo h high Os- cilla ion Indices and [Ca 21 ] i alues when mammo opes we e a ached o nonmammo opes. In eali y, his was no he case a all: he sca e plo s o M-nonM double s and singles (Fig. 4) we e ex emely simila . Thus, al hough we canno comple ely exclude he possibili y o a s imula o y FIG. 5. Pa e ns o spon aneous [Ca 21 ] i oscilla ions in M-M and M-nonM double s. All iden i ied mammo opes we e subg ouped based on hei oscilla o y pa e ns in o (A) nonoscilla ing, (B) high equency, low ampli ude oscilla ions, and (C) high equency, high ampli ude oscilla ions. Rep esen a i e aces o each p o ile a e depic ed o M-M and M-nonM cells. No e ha he e we e no di e ences be ween A and B ype oscilla o s, bu ha he ampli ude o “C” ype oscilla o s ended o be highe in M-nonM han in M-M cells. 2992 PARACRINE COMMUNICATION AND CALCIUM DYNAMICS IN MAMMOTROPES Endo •1998 Vol 139 •No 6 Downloaded om h ps://academic.oup.com/endo/a icle/139/6/2988/2987337 by UNIVERSIDAD DE VALLADOLID. BIBLIOTECA use on 19 Janua y 2021 signaling mechanism o explain his phenomenon, ou da a a e mo e consonan wi h an inhibi o y one. Resul s concep ually simila o hese we e ob ained in ou p e ious s udy in which we ound ha M-M double s dis- played a g ea ly educed le el o PRL gene exp ession com- pa ed wi h hei M-nonM coun e pa s (17). The s iking sim- ila i ies o hese esponses aise he possibili y ha inhibi o y in e ac ions be ween mammo opes on PRL gene exp ession a e media ed by changes o [Ca 21 ] i oscilla ions. In suppo o his possibili y a e obse a ions ha pha macologic diminu ion o [Ca 21 ] i leads o a dec ease o PRL gene exp ession, whe eas an inc ease o [Ca 21 ] i has an opposi e e ec (25–27). How migh such signals be ansmi ed om one mammo- ope o ano he ? One can en ision a leas h ee possible mech- anisms. The i s o hese, gap-junc ions, is ende ed implau- sible by ou di ec obse a ion ha his o m o communica ion was exceedingly a e unde he p esen expe imen al condi- ions. A second possibili y is elec ical coupling o adjacen cells. This scena io also seems highly unlikely because elec ically coupled cells ypically display synch onized oscilla ions, and we did no ind any e idence o his in he p esen s udy, e en in M-M double s. A inal possibili y en i ely consis en wi h ou esul s is pa ac ine communica ion. Indeed, he e is a weal h o in o ma ion suppo i e o pa ac ine ac ions on mammo opes (1–3, 28), and ou e o s a e cu en ly ocused on isola ing and iden i ying his pu a i e signal. Al hough he physiologic ele ance o inhibi o y in e ac- ions among mammo opes emain o be unequi ocally es- ablished, hey may se e o hold PRL gene exp ession and sec e ion in check un il he mammo ope is p esen ed wi h a challenge om o he egula o y (hypophysio opic) sig- nals. This may be pa icula ly ele an in a eas whe e he e is a concen a ion o mammo opes such as nea he poles o he an e io pi ui a y (29). Acknowledgmen s The au ho s hank D . R. I. Weine o p o iding he GT1–1 cells and D . L. Nun˜ez o sha ing he cul u es wi h us. We would also like o hank W. J. Faugh o assis ance wi h mic oinjec ion s udies and J. Nicholson o access o he A o luo Vision sys em. We also hank D s. S. T. Willa d, M. D. Ams u z, and L. Nun˜ez and Ms. A. Go e o expe c i ique o he manusc ip . Re e ences 1. Schwa z J, Che ny R 1992 In e cellula communica ion wi hin he an e io pi ui a y in luencing he sec e ion o hypophysial ho mones. Endoc Re 13:453–475 2. Dene C, Baes M, Sch amme C 1986 Pa ac ine in e ac ions in he an e io pi ui a y: ole in he egula ion o p olac in and g ow h ho mone sec e ion. F on Neu oendoc inol 9:115–148 3. Schwa z J 1992 The o es , he ees and he an e io pi ui a y. Mol Cell Endoc inol 85:C45–C49 4. Hoe e MT, Heiman ML, Ben-Jona han N 1984 P olac in sec e ion by cul u ed an e io pi ui a y cells: in luence o cul u e condi ions and endoc ine s a us o he pi ui a y dono . Mol Cell Endoc inol 35:229–235 5. Pe ez FM, Rose JC, Schwa z J 1995 An e io pi ui a y cells: ge ing o know hei neighbo s. Mol Cell Endoc inol 111:C1–C6 6. Lail- ecke MR, Han ahan CJ, Whi e BA 1996 Role o cell-cell adhesion in he egula ion o p olac in gene exp ession by ex acellula CaCl 2 . Endoc ine 5:341–349 7. Pa pu a V, Basa sky TA, Liu F, Je inija K, S dija Je inija, Haydon PG 1994 Glu ama e-media ed as ocy e-neu on signaling. Na u e 369:744–747 8. Wein aub WH, Cle eland-Wol e P, Few ell C 1994 Pa ac ine Ca 21 signaling in i o: se o onin-media ed cell-cell communica ion in mas cell/smoo h mus- cle cocul u es. J Cell Physiol 160:389–399 9. Kobayashi H, Mizuki T, Wada A, Izumi F 1992 Cell-cell con ac modula es exp ession o cell-adhesion molecule L1 in PC-12 cells. Neu oscience 49:437–441 10. Molla d P, Schlegel W 1996 Why a e endoc ine pi ui a y cells exci able? T ends Endoc inol Me ab 7:361–365 11. Ashwo h R, Hinkle PM 1996 Thy o opin- eleasing ho mone-induced in a- cellula calcium esponses in indi idual a lac o ophs and hy o ophs. En- doc inology 137:5205–5212 12. S ojilko ic SS 1996 Calcium signaling in pi ui a y cells. T ends Endoc inol Me ab 7:357–360 13. Lledo PM, Gue ineau N, Molla d P, Vincen JD, Is ael JM 1991 Physiological cha ac e iza ion o wo unc ional s a es in sub-popula ions o p olac in cells om lac a ing a s. J Physiol (Lond) 437:477–494 14. Hinkle PM, Jackson AE, Thompson TM, Za acki AM, Coppola DA, Banc o C1988 Calcium channel agonis s and an agonis s: e ec s o ch onic ea men on pi ui a y p olac in syn hesis and in acellula calcium. Mol Endoc inol 2:1132–1138 15. Whi e BA, Baue le LR, Banc o FC 1981 Calcium speci ically s imula es p olac in syn hesis and messenge RNA sequences in GH 3 cells. J Biol Chem 256:5942–5945 16. Day RN, Maue RA 1990 Pi ui a y calcium channel modula ion and egula ion o p olac in gene exp ession. Mol Endoc inol 4:736–742 17. Ab aham EJ, Faugh WJ, F awley LS 1996 In e cellula communica ion: el- a i e impo ance o cellula adhesion and pa ac ine signaling o ho monal gene exp ession. Endoc inology 137:4050–4053 18. Boock o FR, Hoe le JP, F awley LS 1986 Analysis by plaque assays o GH and p olac in elease om indi idual cells in cul u e o male pi ui a ies. Neu- oendoc inology 42:64–70 19. G ynkiewicz G, Poenie M, Tsien RY 1985 A new gene a ion o Ca 21 indi- ca o s wi h g ea ly imp o ed luo escence p ope ies. J Biol Chem 260:3440–3450 20. Villalobos C, Faugh WJ, F awley LS 1998 Dynamic changes in spon aneous in acellula ee [Ca 21 ] i oscilla ions and hei ela ionship o p olac in gene exp ession in single, p ima y mammo opes. Mol Endoc inol 12:87–95 21. Zheng T, Villalobos C, Nusse KD, Ge ys TW, Faugh WJ, Cas an˜o JP, F awley LS 1997 Pheno ypic cha ac e iza ion and unc ional co ela ion o a -MSH binding o pi ui a y cells. Am J Physiol 272:E282–E287 22. Villalobos C, Ga cı´a-Sancho J 1995 Capaci a i e Ca 21 en y con ibu es o he Ca 21 in lux induced by TRH in GH 3 pi ui a y cells. P lu¨ge s A ch 430:923–935 23. Villalobos C, Ga cı´a-Sancho J 1996 Ca eine-induced oscilla ions o cy osolic Ca 21 in GH 3 pi ui a y cells a e no due o Ca 21 elease om in acellula s o es bu o enhance Ca 21 in lux h ough ol age-ga ed Ca 21 channels. P lu¨ge s A ch 4:371–378 24. Cha les AC, Kodali SK, Tyndale RF 1996 In e cellula calcium wa es in neu ons. Mol Cell Neu osci 7:337–353 25. Gick GG, Banc o C 1985 Regula ion by calcium o p olac in and g ow h ho mone mRNA sequences in p ima y cul u es o a pi ui a y cells. J Biol Chem 260:7614–7618 26. Da is JRE, Hogga d N, Wilson EM, Vidal ME, Sheppa d MC 1991 Calcium/ calmodulin egula ion o he a p olac in gene is con e ed by he p oximal enhance egion. Mol Endoc inol 5:8–12 27. La e ie e J-N, Tixie -Vidal A, Buisson N, Mo in A, Ma ial JA, Gou dji D 1988 P e e en ial ole o calcium in he egula ion o p olac in gene ansc ip ion by hy o opin- eleasing ho mone in GH 3 pi ui a y cells. Endoc inology 122:333–340 28. Vankelecom H, Dene C 1997 Pa ac ine communica ion in he an e io pi ui a y as s udied in eagg ega e cell cul u es. Mic osc Res Tech 39:150–156 29. Bu ns G, Sa ka DK 1993 T ans o ming g ow h ac o - b 1 -like immuno eac- i i y in he pi ui a y gland o he a : e ec o es ogen. Endoc inology 133: 1444–1449 TABLE 1. Pa ame e s o calcium dynamics in M-M and M-nonM double s AB C M-M % 23 50 19.2 Oscilla ion Index 1.97 60.22 6.52 60.79 29.93 65.26 a Mean [Ca 21 ] i 21.75 64.27 34.19 64.04 67.63 611.2 c M-nonM % 12.3 52 25.8 Oscilla ion Index 2.14 60.09 5.62 60.34 65.18 69.76 b Mean [Ca 21 ] i 13.6 62.31 34.81 63.23 119.29 612.5 d The p opo ions and alues o cells exhibi ing each o h ee majo oscilla o y pa e ns a e p esen ed: A, nonoscilla o y; B, high equency, low-ampli ude oscilla ions; C, high equency, high ampli ude oscilla- ions. A mino i y o mammo opes (8–10%) changed om one pa e n o ano he du ing he collec ion pe iod and he e o e we e no classi ied in o any speci ic pa e n. They we e, howe e , included in he o al cell numbe o he pu pose o calcula ing pe cen ages. Values wi h di e en le e s a e signi ican ly (P,0.05) di e en om each o he . PARACRINE COMMUNICATION AND CALCIUM DYNAMICS IN MAMMOTROPES 2993 Downloaded om h ps://academic.oup.com/endo/a icle/139/6/2988/2987337 by UNIVERSIDAD DE VALLADOLID. BIBLIOTECA use on 19 Janua y 2021