Potassium Channel Types in Arterial Chemoreceptor Cells and Their Selective Modulation by Oxygen

Po assium Channel Types in A e ial Chemo ecep o Cells and Thei Selec i e Modula ion by Oxygen MAR A DOLORES GANFORNINA and Josg LOPEZ-BARNEO F om he Depa amen o de Fisiolog a y Bio lsica, Facuhad de Medicina, Uni e sidad de Se illa, 41009 Se illa, Spain ABSTRACT Single K + channel cu en s we e eco ded in excised memb ane pa ches om dispe sed chemo ecep o cells o he abbi ca o id body unde condi ions ha abolish cu en low h ough Na + and Ca 2+ channels. We ha e ound h ee classes o ol age-ga ed K + channels ha di e in hei single-channel conduc ance (~/), dependence on in e nal Ca 2+ (Cai2+), and sensi i i y o changes in 02 ension (Po2). Ca2+-ac i a ed K + channels (Kca channels) wi h ~/ ~ 210 pS in symme ical K + solu ions we e obse ed when [Ca2+]i was > 0.1 IzM. Small conduc- ance channels wi h ~/= 16 pS we e no a ec ed by [Ca2+]i and hey exhibi ed slow ac i a ion and inac i a ion ime cou ses. In hese wo channel ypes open p obabili y (Pop,n) was una ec ed when exposed o no moxic (Po2 = 140 mmHg) o hypoxic (P02 - 5-10 mmHg) ex e nal solu ions. A hi d channel ype ( e e ed o as Ko 2 channel), ha ing an in e media e ,/( ~ 40 pS), was he mos equen ly eco ded. Ko~ channels a e s eeply ol age dependen and no a ec ed by [Ca2+]i, hey inac i a e almos comple ely in < 500 ms, and hei Pope, e e sibly dec eases upon exposu e o low P02. The e ec o low Po2 is ol age dependen , being mo e p onounced a mode a ely depola ized ol ages. A 0 mV, o example, Popen diminishes o ~ 40% o he con ol alue. The ime cou se o ensemble cu en a e ages o Ko 2 channels is ema kably simila o ha o he O2-sensi i e K + cu en . In addi ion, ensemble a e age and mac oscopic K + cu en s a e a ec ed simila ly by low P02. These obse a ions s ongly sugges ha Ko 2 channels a e he main con ibu o s o he mac oscopic K + cu en o glomus cells. The e e sible inhibi ion o Ko 2 channel ac i i y by low P02 does no desensi ize and is no ela ed o he p esence o F-, ATP, and GTP--¢-S a he in e nal ace o he memb ane. These esul s indica e ha Ko 2 channels con e upon glomus cells hei unique chemo ecep o p ope ies and ha he O2-K + channel in e ac ion occu s ei he di ec ly o h ough an O2 senso in insic o he plasma memb ane closely associa ed wi h he channel molecule. INTRODUCTION Al hough i has been known o decades ha he mammalian ca o id bodies pa icipa e in he egula ion o b ea hing by adjus ing he en ila o ), a e o he le el Add ess ep in eques s o D . J. L6pez-Ba neo, Depa amen o de Fisiolog a y Bio isica, Facul ad de Medicina, A enida S~ nchez Pizjm n, 4, 41009 Se illa, Spain. j. GEN. PHYSIOL. C) The Rocke elle Uni e si y P ess • 0022-1295/92/09/0401/26 $2.00 Volume 100 Sep embe 1992 401-426 401 on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 402 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 • 1992 o oxygen ension (Po2) in a e ial blood (De Cas o, 1926; Heymans, Bouckae , and Dau ebande, 1930; Fidone and Gonz ilez, 1986; Fi zge ald and Lahi i, 1986) he mechanisms in ol ed in he p ocess o 02 sensing ha e emained unknown. The e has been a gene al consensus ha ype I (o glomus) cells, he mos nume ous in he ca o id body, a e he elemen s esponsible o chemo ansduc ion since hey make synapses wi h a e en ne e endings, ha e cy osolic g anules con aining ca echola- mines, and sec e e dopamine in esponse o hypoxia and high ex e nal K ÷ (Fidone, Gonz ilez, and Yoshizald, 1982; Fishman, G eene, and Pla ika, 1985; Alma az, Gonz~ilez, and Obeso, 1986; Rigual, Gonz ilez, Gonz ilez, and Fidone, 1986; Obeso, Fidone, and Gonz ilez, 1987). Di ec p oo o he chemo ecep i e p ope ies o ype I ceils has come, howe e , om ecen elec ophysiological expe imen s. I has been shown ha ype I cells can i e ac ion po en ials epe i i ely and, as in o he elec ically exci able cells, hey gene a e ol age-dependen Na +, Ca z+, and K + cu en s (Duchen, Caddy, Ki by, Pa e son, Pon e, and Biscoe, 1988; L6pez-Ba neo, L6pez-L6pez, U e ia, and Gonz ilez, 1988; U e ia, L6pez-L6pez, Gonz ilez, and L6pez-Ba neo, 1989a). Fu he mo e, i has also been ound ha he ol age-ga ed K ÷ cu en o ype I cells is e e sibly a enua ed by lowe ing en i onmen al Po2, whe eas Na s and Ca e+ cu en s emain unal e ed (L6pez-Ba neo e al., 1988). These indings, con i med by he pa allel wo k o o he in es iga o s on se e al mammalian species (Delpiano and Heschele , 1989; Heschele , Delpiano, Acke , and Pie uschka, 1989; Pee s, 1990; S ea and Nu se, 1991), ha e p o ided a amewo k o unde - s anding he basic mechanisms unde lying senso y ansduc ion in he ca o id body. Inhibi ion o he O2-sensi i e K ÷ cu en unde hypoxic condi ions p oduces an inc ease in he i ing equency o glomus cells (L6pez-L6pez, Gonz ilez, U e ia, and L6pez-Ba neo, 1989), which could lead o Ca 2+ in lux, enhanced ansmi e elease, and ac i a ion o he a e en ibe s o he sinus ne e. This basic scheme is also suppo ed by wo k on ype I cells loaded wi h luo escen Ca 2s indica o s showing an inc ease in cy osolic Ca 2s in esponse o low Po2 (Biscoe and Duchen, 1990a; Beno , A., J. U e ia, and J. L6pez-Ba neo, unpublished obse a ions). The mac oscopic K ÷ cu en eco ded in ype I cells has a small Ca2S-dependen componen ha disappea s a e wash-ou o Ca 2+ channels and when in e nal solu ions wi h high Ca2+-bu e ing capaci y a e used. In addi ion, he Ca2S-indepen - den componen o he cu en inac i a es almos en i ely in 200 ms bu he deg ee o inac i a ion a ies among di e en cells (U e ia e al., 1989a). The e o e, i can be expec ed ha , as in o he exci able cells (Ma y and Nehe , 1985; Hoshi and Ald ich, 1988), glomus cells possess se e al classes o K s channels wi h speci ic biophysical p ope ies ( o e iew, see Rudy, 1988). The p esen esea ch was unde aken o es ablish a i s classi ica ion o K + channels in ype I cells and o asce ain whe he a speci ic K s channel class is esponsible o he O2 sensi i i y o he cells' elec ical p ope ies. These ques ions a e o c i ical impo ance o elucida ing he molecula mechanisms unde lying 02 sensing. The iden i ica ion o he p ima y si e in ol ed in 02 de ec ion is also o in e es because i has been a gued ha he a enua ion o he mac oscopic K + cu en on lowe ing Po2 could be a seconda y phenomenon a he han an ini ial s ep in he p ocess o chemo ansduc ion (Biscoe and Duchen, 1989, 1990a, b). In his pape we p esen a sys ema ic analysis o he single K s channel ypes ound in glomus cells. We show ha in excised memb ane pa ches he e a e h ee di e en on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND L PEZ-BARNEO Single K + Channel Modula ed by 02 in Glomus Cells 403 K + channel ypes and ha only one o hem ( e e ed o as "Ko~ channel") is selec i ely and e e sibly modula ed by hypoxia. The cha ac e is ics o hese channels ully accoun o he p ope ies o he mac oscopic O2-sensi i e K + cu en . Ou expe imen al esul s indica e ha Ko 2 channels a e di ec ly egula ed by 02 and s ongly sugges ha he 02 senso o chemo ecep o cells is in, o closely associa ed wi h, he plasma memb ane. In he ollowing a icle we ocus on a mo e de ailed desc ip ion o he kine ic p ope ies o he Ko 2 channel and p opose a minimal model ha .explains he e ec s o hypoxia on channel ga ing. A b ie epo o pa o he wo k p esen ed in his pape has been published (Gan o nina and L6pez-Ba neo, 1991). METHODS Cell P epa a ion Expe imen s we e pe o med on ype I cells isola ed om abbi ca o id bodies. The p ocedu es ollowed o enzyma ic cell dispe sion and cul u e we e he same as desc ibed TABLE I Composi ion o Solu ions Ex e nal NaCl KCI CaCle MgCI2 HEPES Glucose S anda d Na 140 2.7 2.5-5 2 10 5 140 K -- 140 2.5-5 2 10 5 80 K 60 80 2.5-5 2 10 5 In e nal KCI K-glu ama e KF HEPES EGTA MgCI~ S anda d K 30 80 20 10 10 2.58 130 K 130 -- -- 10 10 2.58 All alues a e gi en in millimola . Te odo oxin a a concen a ion o 0.6 IxM was added o he s anda d Na and 80 K solu ions. Solu ions wi h a iable [Ca 2+] we e made by mixing CaCI2, MgClz, and EGTA. The inal ee [Ca 2÷] was calcula ed wi h a compu e p og am ha akes in o accoun he a ini y cons an s o EGTA o Ca 2+ and Mg 2+ a di e en pH (Taba es, U e ia, and L6pez-Ba neo, 1989). In hose solu ions wi h 10 mM EGTA and no Ca 2+ added he es ima ed [Ca ~+] was < 10 -9 M. p e iously (L6pez-Ba neo e al., 1988; U e ia e al., 1989a). Cells we e pla ed on sli e s o glass co e slips ea ed wi h poly-l-lysine and used o eco ding be ween 12 h and 2 d a e dissocia ion. Du ing he expe imen s a co e slip was ans e ed o a small chambe o ~ 0.2 ml olume wi h con inuous low o solu ions ha could be eplaced in 10-15 s. Solu ions The composi ion o solu ions used in he expe imen s is shown in Table I. Solu ions we e adjus ed o a pH o 7.3 (in e nal) o be ween 7.35 and 7.4 (ex e nal) and had an osmola i y o 290-300 mosmol/kg. In he ex and in he igu e legends solu ions a e gi en as ex e nal// in e nal wi h speci ica ion o he inal [Ca z÷] o [EGTA] used. Expe imen s we e pe o med a on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 404 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 • 1992 oom empe a u e (22-25°C). Du ing he expe imen s he ex e nal solu ions we e equilib a ed wi h ei he ai , N2, o a mix u e o bo h, in o de o ob ain he desi ed Po2 in he eco ding chambe (see below). Reco ding Techniques The da a p esen ed in his a icle a e mainly based on single K + channel cu en s eco ded om memb ane pa ches o glomus cells using he pa ch clamp echnique (Hamill, Ma y, Nehe , Sakmann, and Sigwo h, 1981). In some expe imen s whole-cell K + cu en s we e s udied be o e es ablishing he ou side-ou pa ch con igu a ion ollowing he me hodology p e iously epo ed (U e ia e al., 1989a). Fo single-channel eco ding we used i e-polished glass pipe es ab ica ed om bo osilica e glass (Kimax 51) ha once illed wi h solu ion had a esis ance o 4-10 MI'~. The eco ding bandwid h o he ampli ie was 10 kHz, howe e i s ou pu signal was low-pass il e ed by an 8-pole Bessel il e (model 902; F equency De ices Inc., Ha e hill, MA) wi h cu o equencies be ween 1 and 2 kHz, gi ing an e ec i e cu o equency o 0.95-1.29 kHz (Colquhoun and Sigwo h, 1983). The ime esolu ion o ou eco ding sys em and i s in luence on he measu emen o he ampli ude and du a ion o single-channel e en s is u he explained in he accompanying pape . Da a Acquisi ion and Analysis An IBM-PC-AT compu e in e aced o he analog elec onics was used o da a acquisi ion, display, and analysis. In mos expe imen s we eco ded ionic cu en s in esponse o ol age s eps. In hese cases he cu en signal was digi ized on-line by an inpu -ou pu in e ace buil in ou labo a o y (U e ia, Ma eos, and L pez-Ba neo, 1989b). A sweep was de ined by ei he 500 o 1,000 digi al poin s. Single-channel cu en s gene a ed in esponse o long-las ing (> 220 ms) o s a iona y depola iza ions we e ini ially s o ed on ideo ape. The segmen s o he signal equi ed o igu es o analysis we e eplayed on a cha eco de o con e ed in o digi al o m using a GPIB-PC ca d plugged in o he compu e expansion slo s. The sample equency a ied acco ding o he expe imen al p o ocol and is gi en in he igu e legends. Leakage and uncompensa ed capaci y cu en s we e digi ally sub ac ed using scaled empla es cons uc ed by i ing smoo h unc ions o ei he eco ds wi h no openings o o he a e age o 20 consecu i e cu en sweeps gene a ed by 20-mV hype pola iza ions om a holding po en ial o -80 mV. Ensemble a e ages we e ob ained om o iginal aces a e capaci y and leakage sub ac ion. Single-channel cu en ampli ude was measu ed by a e aging alues ob ained om 20-40 well- esol ed single e en s. We used a 50% ampli ude c i e ion o de ec opening and closing ansi ions (Colquhoun and Sigwo h, 1983). Unless o he wise no ed, no co ec ion o un esol ed e en s was pe o med. The numbe o ac i e channels in a pa ch (N) was de e mined by obse ing o long ime pe iods he maximum numbe o simul aneous cu en s eps ha appea ed a s ongly depola ized ol ages. In pa ches whe e one o wo simul aneous openings we e obse ed, he p obabili y ha his numbe could be smalle han he ac ual alue o N was s a is ically es ed by he binomial dis ibu ion me hod as indica ed by Pa lak and Ho n (1982). In hose eco dings whe e he es ima ed alue o N was one, a e age channel open p obabili y (Pop,n) was calcula ed by di iding he ime spen in he open s a e by he o al du a ion o he eco ding. In pa ches wi h mo e han one channel, Pop~n was es ima ed by he o mula: Popen = (N'i'T)- ' :l( ) "d (1) whe e i is he single-channel cu en ampli de, T is he du a ion o he pulse o o he obse a ion pe iod in s a iona y condi ions, and I( ) is he ne cu en du ing he eco ding on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND L6PEZ-BARNEO Single K + Channel Modula ed by 02 in Glomus Cells 405 pe iod. In channels ac i a ed du ing depola izing pulses, we ob ained an ensemble cu en a e age (l(a)) gi en by: I (a) = N'i'Po( ) (2) whe e Po( ) is he open p obabili y as a unc ion o ime. Signi icance o di e ences be ween mean alues ob ained (e.g., o di e en Po2) was de e mined wi h a S uden 's es o pai ed samples. Unless o he wise indica ed, he le el o signi icance (cx) o he es was se a 0.05. A > 0 > 2.0 1.5 1.0 0.5 0 B 1.0 '~ I 0.8 i i > 0.6 o 0.4 I I > I I I 0.2 0 ~- 0 0.3 0.6 0.9 1.2 0 V bo e y (V) I I I 30 60 90 % 02 i 120 C D N2 N2 N 2 PO2(mmHg) lOO 50 o Ai Ai Ai 1 min FIGURE 1. Pe o mance and calib a ion o he O2-sensing elec ode. (A) Ou pu ol age o he cu en - o- ol age con e e as a unc ion o he nega i e pola izing po en ial and 02 ension. 150 mmHg ( illed ci cles), 87 mmHg (open ci cles), and 0--5 mmHg ( illed squa es). A a pola izing ol age be ween -0.6 and -0.8 V he ou pu ol age is linea ly ela ed o he Po 2 in he solu ion ( ange be ween dashed lines). (B) Ou pu ol age s. %02 ela i e o ai a a pola izing ol age o -0.7 V. (C and D) Responses o he 02 elec ode in he chambe (C) and du ing ins an aneous imme sions in solu ions equilib a ed wi h N2 and ai (D). Measu emen o Oxygen Tension Because some expe imen al p o ocols equi ed epe i i e exposu e o a memb ane pa ch o ex e nal solu ions wi h a ep oducible Po2 alue, we buil an O2-measu ing elec ode o es ima e Po2 alues in he icini y o he cu en - eco ding pipe e. We used a nega i ely pola ized 100- .m- hick pla inum wi e insula ed by an O2-impennean enamel excep a he end sec ion (Tsacopoulos and Lehmenkiihle , 1977; Tsacopoulos, Poi y, and Bo sellino, 1981). Cu en gene a ed in he wi e in esponse o a iable PO 2 alues was eco ded by an on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 406 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 " 1992 ope a ional ampli ie wi ed as an 1/V con e e . The nega i e pole o a d.c. ba e y was connec ed o he nonin e ing inpu o he I/V con e e o main ain he pola izing ol age a a cons an alue. The majo cha ac e is ics as well as he pe o mance o he O2-measu ing elec ode a e illus a ed in Fig. 1. Plo A shows he changes in he ou pu ol age o he 1/V con e e (Vou0 as a unc ion o he nega i e po en ial applied o he pla inum elec ode (Vba e~y) o h ee di e en Po~ alues. In he Vba e~y ange be ween -0.6 and -0.8 V, Vou was p opo ional o he 02 concen a ion; hus we used -0.7 V as he mos app op ia e alue o pola ize he pla inum elec ode. A his pola izing ol age, which was used in all he expe imen s, he ou pu ol age o he eco ding elec ode is linea ly ela ed o he O2 concen a ion and he e o e a calib a ion cu e could be done using solu ions equilib a ed wi h known concen a ions o 02 (Fig. 1 B ). Du ing he expe imen al p o ocol Vou was con inuously moni o ed and s o ed on ape. The esponse o he elec ode is illus a ed in Fig. 1, C and D. When imme sed in solu ions equilib a ed wi h N2 o ai , he change in ol age was almos ins an aneous (D), swi ching o simila solu ions when placed in he expe imen al chambe p oduced a as change (~ 80% o he maximum) in a ew seconds bu comple e equilib a ion equi ed >40 s (C). The ol age signal om he Oz-sensing elec ode was una ec ed by changes in pH o by modi ica ions in he ionic composi ion o he solu ions. Mos o he expe imen s epo ed he e and in he accompanying epo we e based on epe i i e exposu e o he K ÷ channels o a ep oducible Poz alue. Gi en ha ou eco ding chambe is in con ac wi h he ai , we ound ha he mos easily ep oducible Poz le el in he icini y o he cells (Po2 = ~ 5-10 mmHg) was ob ained by bubbling he es solu ion wi h N 2. RESULTS The e A e Th ee Majo K + Channel Types in Glomus Cells Single K + channel cu en s eco ded om memb ane pa ches wi h well- esol ed single-channel e en s allowed he classi ica ion o he K + channels o ype I cells in o h ee majo classes. To acili a e compa ison, single-channel cu en s ep esen a i e o he a ious K + channel ypes a e shown in isola ion in Fig. 2. Table II summa izes he majo p ope ies o each channel popula ion. The aces o Fig. 2 a e om h ee di e en inside-ou pa ches wi h wo unc ional channels. In all cases he memb ane was exposed o asymme ical K + solu ions and he cu en was eco ded a a ious memb ane po en ials (Vm). The h ee se s o eco dings display openings and closu es o he channels ha appea as cu en s eps o ixed ampli ude. Fas ansi ions a e pa ially il e ed due o he limi ed eco ding bandwid h. Opening o each channel ype p oduced cu en s eps o clea ly di e en ampli ude, bu in all cases channel opening was a o ed by memb ane depola iza ion. I will be shown below ha besides dis inc single-channel conduc ance alues he h ee channel popula ions also di e in hei kine ic p ope ies, Ca z+ dependence, and sensi i i y o changes in Po2 (see Table II). Fo he sake o cla i y abb e ia ions a e used o each channel ype. Kc, deno es he la ge Ca2+-ac i a ed K + channels, SK a e channels o small conduc ance, and Ko 2 e e s o he O2-sensi i e K + channels. Ca2+-dependen K + Channels (Kca Channels) Kca channel ac i i y was clea ly obse ed in inside-ou memb ane pa ches (a o al o 24 pa ches) when [Ca 2+] in he solu ion acing he cy osolic side o he memb ane was >0.1 ~M. These channels we e ac i e du ing main ained depola iza ions and on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND LOPEZ-BARNEO Kca V::oV Single K + Channel Modula ed by 02 in Glomus Cells SK Ko2 0 2 ~ -- 02 01 02 01 __01 C --C C 407 20 ~15p A ....... ~ 2pA 50 ms FIGURE 2. Rep esen a i e eco dings o he ac i i y o he h ee majo K ÷ channel ypes ound in ype I cells. The da a we e ob ained om di e en inside-ou excised pa ches, con aining wo channels each, depola ized a he indica ed memb ane po en ials. Kca = Ca~+-ac i a ed channels; SK = small conduc ance channels; K% = O~-sensi i e channels. In all igu es upwa d de lec ions om he ze o cu en le el (c) indica e ou wa d cu en . K~ and SK channels we e eco ded unde s eady depola iza ions and K% channels upon 200-ms s ep depola iza ions om -80 mV. E ec i e cu o equency = 0.95 kHz and sampling in e al = 500 Ixs. Solu ions: s anda d Na, TIX//130 K, 10 EGTA. K~ channels we e eco ded wi h an in e nal solu ion con aining 1 IxM ee Ca 2+. he e o e hei single-channel cu en - ol age (i-Vm) ela ion and Popen we e s udied a e s eady-s a e changes in he memb ane po en ial. Single Kca channel ac i i y as a unc ion o he memb ane po en ial is illus a ed in Fig. 3 A. The aces a e om a pa ch ba hed in symme ical K + solu ions ha con ained a leas h ee ac i e channels. Single-channel cu en ampli ude a ied in pa allel wi h he elec ochemi- cal d i ing o ce o K + ions, and he numbe o ac i e channels as well as he ime TABLE II Classi ica ion o K ÷ Channels in Glo aus Cells Channel Conduc ance in 130 K//130 K Conduc ance in 2.7 K//130 K Ca 02 ype Inside-ou Ou side-ou Inside-ou Ou side-ou dependence sensi i i y K~ 206.7 (10) -- 83.7 (3) -- Yes No SK 16 (6) -- 6 (2) -- No No K% 41.5 (1) 41.5 (2) 17.6 (18) 20.1 (14) No Yes 21 (10)* 19.5 (5)* A e age slope conduc ance alues a e gi en in picosiemens. The numbe o pa ches is in pa en heses. Kc~ = Ca-ac i a ed channel; SK = small conduc ance channel; K% - oxygen-sensi i e channel. Fo he K% channel alues a e gi en in con ol and hypoxic ( ) condi ions. Po assium concen a ions a e in millimola and indica ed as ex e nal//in e nal. on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 408 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 • 1992 spen in he open s a e inc eased wi h depola iza ion. The a e age i-Vm ela ions ob ained om da a pooled om se e al pa ches exposed o symme ical and asymme ical K ÷ concen a ions ([K÷]) a e plo ed in Fig. :3 B. In symme ical [K ÷] ( illed symbols) he i-Vm plo is linea be ween -50 and +60 mV, and he e e sal po en ial was 0 mV, as expec ed o K+-selec i e channels. A linea eg ession i o he da a poin s yields a alue o 206.7 pS (n = 10 pa ches) as he a e age single- A Vm ~- o2 --01 (mV) +30 --C ~ --03 --0 2 --01 +20 --C -10 ........ C --0 2 -20 -~--C --01 --0 2 1 O0 ms B I -90 15 9 -30 • i(pA) I I 30 90 FIGURE 3. Single-channel cu en - ol age ela ion o he Kca channels. (A) Uni a y e en s eco ded in an inside-ou excised pa ch wi h a leas h ee simul aneously open channels a a ious s a iona y memb ane po en ials. E ec i e cu o equency = 0.95 kHz and sampling in e al = 500 I~s. Solu ions: 130 K, 0.01 ~M Ca2+//130 K, 1 I~M Ca 2+. (B) Plo o single-channel cu en (i) as a unc ion o he memb ane po en ial (Vm) measu ed in symme i- cal ( illed symbols, n = 10 inside-ou pa ches) and asymme ical (open symbols, n = 3 inside-ou pa ches) K + concen a ions. The poin s indica e he mean ± SD alues and he s aigh lines he linea eg ession i s o he da a. Slope conduc ances a e: 206.7 pS ( illed symbols, = 0.99) and 83.7 pS (open symbols, = 0.98). Solu ions: 130 K, 0.01 I~M Ca2+//130 K, 1 I~M Ca z+ (symme ical [K+]); s anda d Na, TFX//130 K, 1 I~M Ca 2+ (asymme ical [K+]). channel conduc ance. In asymme ical [K ÷] (open symbols) he a e age slope conduc ance, measu ed be ween -30 and +50 mV, is 83.7 pS (n = 3 pa ches). The possible modula o y e ec o Po2 on he ac i i y o Kca channels was in es iga ed in inside-ou excised pa ches ha we e ini ially exposed o a iable in e nal Ca ~+ concen a ions ([Ca2+]i) o es he Ca 2+ dependence o channel ac i a ion, and he ea e o a ious Po~ le els keeping [Ca2+]i unal e ed. An example on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND L PEZ-BARNEO Single K + Channel Modula ed by 02 in Glomus Cells 409 o his expe imen al p o ocol is illus a ed in Fig. 4. The memb ane was ba hed in symme ical high K + solu ions wi h a [Ca2+]i o 1 ~M and held a a po en ial o +20 inV. Unde hese condi ions opening o he wo ac i e channels included in he pa ch p oduced an ou wa d cu en ha disappea ed comple ely a e swi ching o a solu ion wi h 0.01 o,M Ca ~+. This e ec was pe ec ly e e sible on ein oduc ion o 1 ~M Ca 2+ in he chambe (Fig. 4 A ). When a a ixed [CaZ+]i o 1 ,M he same pa ch was exposed o low Po2, no app eciable changes in single-channel ac i i y o uni a y A 02 -- 01 -- C -- 0.01 #.M Ca 2+ 1 /, ,M Ca 2+ i l B 02-- l 0 -- C -- 10 pA 140 mmHg N 2 Ai FIGURE 4. Dependence o Kca channels on in e nal Ca ~+ and lack o e ec o changes in Po2. (A) Ac i i y a +20 mV o wo Kca channels in an inside-ou pa ch exposed o 1 I~M in e nal Ca 2+ and e e sible inhibi ion when [Ca2+]i is dec eased o 0.01 IxM. (B) A he same memb ane po en ials and [Ca2+]i = 1 I~M channel ac i i y was una ec ed by lowe ing Po2. The signal om he O~-measu ing elec ode is shown in he lowe panel. E ec i e cu o e- quency = 0.95 kHz. Solu ions: 130 K, 0.01 I~M Ca~÷//130 K, 1 IJ.M o 0.01 I~M Ca ~+. cu en ampli ude we e obse ed (Fig. 4 B). The signal om he Po2-measu ing elec ode is shown in he lowe panel. Following his same expe imen al p ocedu e, bu using a ious [Ca2+]i (be ween 0.04 and 1 ~M) and V~ alues (be ween -30 and +30 mV), exposu e o hypoxia did no al e Kca channel open p obabili y (Popen) in he 16 pa ches es ed (pai ed es ). Fig. 5 summa izes he e ec s o [Ca2+]i and low Po 2 on Kca channel ac i i y. In Fig. 5 A, single-channel Po~, as a unc ion o Vm is plo ed a wo [Ca~+]i. In bo h cases he da a poin s a e he mean +_ SE alues om on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 416 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 • 1992 eco dings (panel B) show a peak Pope, a he beginning o he depola iza ion and a p og essi e dec ease du ing he pulse. Single-channel Pope, in eg a ed h oughou he pulse du a ion (see Eq. 1) is 0.31 in he con ol solu ion bu only 0.18 du ing exposu e o low Po2. Thus, hypoxia p oduces a e e sible dec ease in channel open p obabili y bu , as shown in Fig. 8 B ( iangles), i does no modi y ei he he i-Vm ela ion o he single-channel conduc ance (see Table II). A summa y o he e ec o low Po2 on he Popen o Ko 2 channels a di e en memb ane po en ials is shown in Fig. 11. The a e age Popen alue du ing 200-ms pulses was measu ed in pa ches wi h one o wo unc ional Ko~ channels. Low Po2 p oduced a dec ease in single-channel Pope. ha was s a is ically signi ican a all memb ane ol ages (pai ed es , x < 0.01). A +30 mV, o example, Popen is ~0.43 in he con ol solu ion and ~0.31 in low Po2, bu a 0 mV hese alues a e, espec i ely, ~0.2 and ~0.07. This indica es ha he inhibi ion o K + channel ac i i y by hypoxia is mo e ma ked a less depola ized memb ane po en ials (see also GL 0 O- 0.6 0.4 0.2 [~] con ol low PO2 0 +10 +20 +30 w (my) FIGURE 11. Dec ease o Ko 2 channel Pop, n by hypoxia. A e age open p ob- abili y du ing 200-ms pulses (o dina e) was measu ed om ei he inside-ou o ou side-ou pa ches a a ious memb ane po en ials. Pop,n alues measu ed in he con ol and in he low (<5 mmHg) Po2 solu ions a e ep esen ed by he mean +-- SE. Num- be o expe imen s we e: n = 5 (0 mV); n = 3 (+10 mV); n = 23 (+20 mV); and n = 2 (+30 mV). A all ol ages he di e ences we e s a is i- cally signi ican (pai ed es , c~ < 0.01). Solu ions: s anda d Na +, TIX//130 KCI, 10 EGTA o s anda d K, 10 EGTA. accompanying pape ). Fo simplici y, in his se o expe imen s solu ions we e bubbled wi h ei he N~ o ai and hus he low Po 2 alues in he icini y o memb ane pa ches we e ~ 5-10 mmHg (see Me hods). Wi h his p o ocol he dec ease o Pope. by hypoxia is unde es ima ed since i is known ha on exposu e o ex emely low Po2 he inhibi ion o K + channel ac i i y is ela i ely less p onounced han when he channels a e exposed o mode a ely low Po2 alues (be ween 60 and 80 mmHg; L6pez-L6pez e al., 1989; Gan omina and L6pez-Ba neo, 1991). Di ec modula ion o Ko 2 channels accoun s o he p ope ies o he mac oscopic O2-sensi i e K + cu en . The kine ic and pha macological p ope ies o he Ko 2 channels indica e ha hey a e he main channels esponsible o he mac oscopic O~-sensi i e K + cu en o ype I cells. The ensemble a e age cu en s shown ea lie (see Fig. 10 B) illus a e ha he ime cou se exhibi ed by Ko 2 channels du ing a depola iza ion closely esembles he kine ics o he mac oscopic K + cu en which u ns on in a ew milliseconds and inac i a es almos comple ely in 200--300 ms on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND L6PEZ-BARNEO Single K + Channel Modula ed by 02 m Glomus Cells 417 (L pez-L6pez e al., 1989; see also Fig. 12). In addi ion, TEA + e e sibly blocks he whole-cell K + cu en (U e ia e al., 1989a) as well as he Koz channels (Gan o nina and L pez-Ba neo, 1991). The pa allel ime cou ses o Ko~ channel Pope, and he mac oscopic K + cu en we e clea ly e iden when eco dings in he whole-cell mode and in ou side-ou mul ichannel pa ches we e ob ained ollowing he same expe i- men al p o ocol. Fig. 12 illus a es he e e sible inhibi ion o he mac oscopic K + cu en ( aces in C) and he dec ease in K + channel Popen in an ou side-ou pa ch wi h a leas i e channels ( aces in A and B ) du ing a ansien exposu e o hypoxia. The ensemble a e ages o Fig. 12 B, which ep esen he beha io o a ew channels, ha e a ime cou se compa able o ha o he whole-cell cu en s. Con ol Hypoxia Reco e y B C 50 ms pA • ~A FIGURE 12. Compa ison o he e ec o lowe ing Po2 on single Ko 2 channel and mac oscopic K + cu en s elici ed by depola iza ion o 0 mV om a holding po en ial o -80 mV. (A and B ) Rep esen a i e single sweeps and ensemble a e ages o cu en eco ded om an ou side-ou pa ch, con aining a leas i e channels, in he con ol solu ion (Po~ = 150 mmHg; n = 13 consecu i e sweeps; Pope, = 0.29), du ing a 9-min exposu e o hypoxia (Po~ < 5 mmHg; n = 18 consecu i e pulses; Pop~, = 0.15), and a e eco e y in he no mal Po2 solu ion (n = 18 consecu i e pulses; Po~, = 0.27). E ec i e cu o equency = 0.95 kHz and sampling in e - al = 500 p.s. (C) Whole-cell cu en s eco ded wi h he same expe imen al p o ocol. E ec i e cu o equency = 10 kHz and sampling in e al = 500 Ws. Cu en calib a ion ba is 2 pA o A and B, and 0.6 nA o C. Solu ions: s anda d Na, TI'X//130 K, l0 EGTA. Wi h a high Ca 2+ bu e capaci y a he in e nal solu ion, mos o he mac oscopic K + cu en is due o he ac i i y o Ko~ channels since Kca channels canno be ac i a ed and, in addi ion, he densi y o SK channels is low and hei uni a y conduc ance is small. The e o e, he numbe o Ko~ channels can be es ima ed by di iding he peak K ÷ cu en by he alue o he single Ko~ channel cu en ampli ude a he same ol age and co ec ing o he peak channel Popen (~ 0.8 a +20 mV). Ou es ima e gi es alues o 720 -+ 80 (mean -+ SD, n = 18) channels pe cell, which co esponds o wo o ou channels pe squa e mic ome e . This ela i ely on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 418 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 - 1992 high densi y may explain why we ob ained mul ichannel pa ches in ~ 50% o he expe imen s (87 o 168) e en hough we used ela i ely high esis ance pipe es (> 8 MI~). I was shown in a p e ious epo ha he modula ion by 02 o he mac oscopic K + cu en o ype I cells is independen o in e nal Ca 2÷ o he p esence o exogenous nucleo ides (L6pez-Ba neo e al., 1988; L6pez-L6pez e al., 1989). Ou esul s a he single-channel le el con i m and ex end hese obse a ions since e e sible inhibi ion A Con ol B Hypoxia _J I C ~, con ol ~. I1 pA 2 pA 100 ms FIGURE 13. Lack o desensi iza ion a e main ained exposu e o low Po2. (A and B) Single-channel cu en s eco ded om an inside-ou pa ch wi h wo Ko~ channels du ing 350-ms depola iza ions o +20 mV om -80 mV. The cell was exposed o 30 min o low Po2 ( < 5 mmHg) be o e excising he pa ch om which he sweeps we e ob ained du ing al e na ing exposu es o con ol and hypoxic solu ion as explained in Fig. 10. (C) Supe imposed ensemble a e ages o aces eco ded in con ol (n = 25 sweeps, Pop~n = 0.52) and hypoxic (n = 16 sweeps, Po~n = 0.28) condi ions. E ec i e cu o equency = 0.95 kHz and sampling in e - al = 500 ~s. Solu ions: s anda d Na, TIX//130 K, 10 EGTA. o Ko~ channel ac i i y was obse ed in excised pa ches wi hou Ca 2+ o nucleo ides added o he in e nal solu ion. We also es ed whe he 02 could ac h ough he ac i a ion o a memb ane-bound G p o ein, a amily o p o eins ha a e i e e sibly ac i a ed by GTP-,/-S (Gilman, 1987) o by A1F 4 o med om luo ide (a no mal componen o some o ou in e nal solu ions) and aluminum (which could be eleased om he mic opipe e glass) (S e nweis and Gilman, 1982; Bigay, De e e, P is e , and Chab e, 1985). The e e sible modula ion o Ko 2 channels by changes in Poe was on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND LOPEZ-BARNEO Single K + Channel Modula ed by 02 in Glomus Cells 419 unal e ed in solu ions ee o F- (n = 51) o when up o 200 ,m GTP-',/-S was added o he in e nal solu ion (n = 22). Thus, he esul s sugges ha soluble cy osolic media o s o memb ane-bound G p o eins do no pa icipa e in he e ec o 02 on he Ko 2 channel, and ha Oz may in e ac wi h an in insic senso closely associa ed wi h he channel p o ein. Repe i i e exposu e o low Po2 does no p oduce desensi iza ion. In many examples o ligand- ecep o in e ac ion, epea ed o pe manen exposu e o he agonis p oduces an a enua ion o he physiological esponse. This phenomenon, called "desensi iza ion," has been ypically s udied in some ligand-ac i a ed channels and i is well known ha , a e wi hd awal o he agonis , eco e y o he es ing condi ions is slow (see Hille, 1984). Al hough desensi iza ion is a e m applicable o ligand- ecep o in e ac ions, i is a phenomenon ha could play a pa in he physiological adap a ion obse ed in some senso y ecep o s (S ebbens, B own, and Pe e son, 1984). I was desc ibed in ou p e ious wo k ha chemosenso y ansduc ion in he ype I cell is a nonadap ing, o slowly adap ing, p ocess since e e sible a enua ion o he mac oscopic K + cu en can be epea edly obse ed in a gi en cell (L6pez-Ba neo e al., 1988; see also Gan o nina, 1991). Fig. 13 shows single-channel cu en sweeps eco ded du ing al e na ing exposu e o con ol (A) and low Po2 (B) solu ions in a pa ch wi h wo Ko 2 channels excised om a cell ha had been p eincuba ed in ex eme hypoxia (Po2 = 5 mmHg) o 30 min. To acili a e compa ison, ensemble a e ages in he wo expe imen al condi ions a e shown supe imposed in Fig. 13 C. Re e sible inhibi ion o Ko 2 channel ac i i y by low Po2 can be obse ed epea edly a e long-las ing exposu e o ex eme hypoxia, u he sugges ing ha he O2-Ko~ channel in e ac ion does no desensi ize. DISCUSSION In his a icle we desc ibe he p ope ies o h ee ypes o K + channels in chemo e- cep o cells o he ca o id body ha can be dis inguished by hei biophysical cha ac e is ics. We also demons a e ha in excised memb ane pa ches only he ac i i y o a speci ic K + channel class, he Ko~ channel, is e e sibly inhibi ed by lowe ing en i onmen al Po~. Ou indings explain he modula ion by O~ o he mac oscopic K + cu en o glomus cells and s ongly sugges ha he O2-sensing mechanism esides in he plasma memb ane. K + Channel Types in Glomus Cells Bo h cell-a ached and excised memb ane pa ches o abbi glomus cells con ain h ee majo K+-selec i e channels (Kca, SK, and Ko~ channels). A Ca~+-independen and high-conduc ance CI- channel encoun e ed in a ype I cells (S ea and Nu se, 1989) was no s udied. The h ee classes o K ÷ channels di e in hei single-channel conduc ance and kine ics as well as in hei dependence on in e nal Ca 2+ and 02 sensi i i y. K + channels a e ex ao dina ily di e se (see o e iews Rudy, 1988; Adams and Nonne , 1989) and i is well known ha di e en subpopula ions coexis in a gi en memb ane (Dubois, 1983; Con i, Hille, and Nonne , 1984; Ma y and Nehe , 1985; Hoshi and Ald ich, 1988; Llano, Webb, and Bezanilla, 1988). In his espec , he single K + channels iden i ied in ype I cells sha e mos o hei p ope ies on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 420 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 • 1992 wi h hose classi ied in bo ine ch oma in (Ma y and Nehe , 1985) and mouse neu oblas oma (Quand , 1988) cells. Equi alen single-channel cu en s a c also p esen in pheoch omocy oma cells (Hoshi and Ald ich, 1988). In e es ingly, all hese cell ypes ha e a close emb yological o igin. The Kca channels ha e, in symme ical high K + solu ions, an a e age conduc ance o 206.7 pS. These channels a e simila o he maxi-K + Ca2+-dependen channels o o he p epa a ions (Ma y, 1981; Ba e e al., 1982; Quand , 1988), and, in excised pa ches, hey a c una ec ed by changes in Po~. These obse a ions con i m ou p e ious expe imen s showing ha pa o he mac oscopic K + cu en , p esumably a Ca2+-dcpenden componen , disappea s a e wash-ou o Ca 2+ channels (U e ia e al., 1989a) and ha unde hese condi ions, and wi h l0 mM EGTA added o he in e nal solu ion, he K + cu en is s ill e e sibly a enua ed by lowe ing Po2 (L6pez-Ba neo e al., 1988; L6pez-L6pez e al., 1989). I has been epo ed ha hypoxia speci ically inhibi s he Ca2+-dependen componen o he mac oscopic K + cu en eco ded in dialyzed ca o id body cells om newbo n a s (Pee s, 1990). The disc epancy be ween hese da a and ou wholc-ccU and single-channel esul s may e lec a di e ence be ween animal species; howe e , wc also belie e ha he conclusion eached by Pee s (1990) may ha e been biased by he expe imen al p o ocol used in he isola ion o he Ca2+-ac i a ed K + cu en . We know, o example, ha millimola concen a ions o Cd ~+ and Co 2+, which could p oduce a dec ease o he Ca2+-ac i a cd K + cu en due o blockade o Ca 2+ channels, can also p oduce a la ge and e e sible inhibi ion o he Ca2+-independen and O~-sensi i e componen o he K + cu en (Gan o nina, M.D., and J. L6pez-Ba neo, unpublished esul s). We ollowed he e minology o Ma y and Nehe (1985) o deno e a second popula ion o K+-selec i e channels ha ha e a small conduc ance (SK channels). In asymme ical K + solu ions he uni a y conduc ance o ca o id body SK channels (~ 6-7 pS) is simila o he alues epo ed in ch oma in and neu oblas oma cells (Ma y and Nche , 1985; Quand , 1988). Hoshi and Ald ich (1988) ha e also ound in pheoch omocy oma cells wo popula ions o K + channels (Ky and Kx) wi h he same uni a y conduc ance alue. Al hough we did no s udy in de ail he kine ic p ope ies o SK channels, in acco d wi h p e ious wo k (Ma y and Nchc , 1985; Quand , 1988), hey beha ed as Ca2+-independen and slowly ac i a ing channels. In glomus cells inac i a ion o SK channels, i any, mus be also e y slow since channel ac i i y could be eco ded o minu es a depola ized memb ane po en ials. SK channels a e only mode a ely ol age dependen (Popen = 0.2 a +20 and 0.05 a -70 mV) and hei Popen a nega i e ol ages sugges s ha hey may con ibu e o he gcnc a ion o he cs ing po en ial o he cells. In excised pa ches SK channels wc c unal c cd by changcs in Po2, bu we canno discoun ha in si u hey could be subjec ed o modula ion. This idea is based on he ac ha in some inside-ou pa ches he ac i i y o SK channels appea ed ab up ly se e al minu cs a e excision o he memb ane, which could be explained by he dilu ion o some soluble media o ha blocks he channels o ha a o s hei closed con o ma ion. Wc ha e coined he e m Ko 2 o designa e he K+-sclec i c and O2-scnsi i e channels o ype I cells. These channels we e he mos equen ly obse ed, p obably because hey a e densely packed in he glomus cell memb ane. Ou es ima e is wo o on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND L PEZ-BARNEO Single K + Channel Modula ed by 02 in Glomus Cells 421 ou channels pe squa e mic ome e . In asymme ical K + solu ions he uni a y conduc ance o he Ko 2 channel is ~ 20 pS. This alue is in excellen ag eemen wi h he conduc ance o as ac i a ing (FK) channels in ch oma in, neu oblas oma, and pheoch omocy oma cells (Ma y and Nehe , 1985; Hoshi and Ald ich, 1988; Quand , 1988) as well as o delayed ec i ie and A- ype K + channels desc ibed in a numbe o p epa a ions (Coope and Sh ie , 1985, 1989; Kasai, Kameyama, Yamaguchi, and Fukuda, 1986; Llano e al., 1988). K% channels a e no in luenced by changes in in e nal Ca 2+ bu hey a e s eeply dependen on memb ane ol age; he ac i a ion h eshold is a ~-50 o -40 mV, and a +20 mV he peak/)open is 0.8. This las pa ame e is simila o alues epo ed o o he mammalian inac i a ing K ÷ channels (Ma y and Nehe , 1985; Coope and Sh ie , 1989). Al hough he ac i a ion and inac i a ion kine ics a e s udied in mo e de ail in he accompanying a icle, he e we show ha K% channels ha e a as ac i a ion. As he memb ane is mo e depola ized he numbe o ac i e channels inc eases and he la ency o he i s opening dec eases. Du ing main ained depola iza ions Ko 2 channels inac i a e comple ely in a ew hund ed milliseconds. These p ope ies a e pe ec ly compa ible wi h he cha ac e is ics o he mac oscopic K + cu en o ype I cells (U e ia e al., 1989a) and s ongly sugges ha he Ko~ channels a e he main con ibu o s o his cu en . This is also suppo ed by he close pa allelism exis ing be ween he ime cou ses o he whole-cell K + cu en and he ensemble a e ages om pa ches con aining only Ko~ channels. Modula ion o Ko 2 Channels by 02 Tension A dis inc p ope y o Ko~ channels is ha hei Popen dec eases on exposu e o low Po2. The inhibi ion o channel ac i i y by hypoxia is e e sible and concen a ion dependen (Gan o nina and L6pez-Ba neo, 1991), and is a p ocess ha does no unde go desensi iza ion. These p ope ies i pe ec ly wi h hose encoun e ed in he modula ion o he mac oscopic K + cu en by Po2 (L6pez-Ba neo e al., 1988; L6pez-L6pez e al., 1989). The co espondence be ween he 02 modula ion o whole-cell and single-channel cu en s is qui e ema kable and can be clea ly seen when ensemble a e ages om pa ches con aining Ko 2 channels in isola ion a e compa ed wi h he O2-sensi i e K ÷ cu en (see, o example, Fig. 12). The majo e ec o lowe ing Po2 is a dec ease in he Pope, o he channels lea ing unal e ed uni a y conduc ance. In e es ingly, he magni ude o he dec ease o Pope, in hypoxic condi ions in he ol age ange be ween +10 and +30 mV (~25% o he con ol alue) is he same as he inhibi ion o he mac oscopic K + cu en by low Po2 (L6pez-Ba neo e al., 1988). The ac ion o low Po2 is mo e p onounced a less depola ized memb ane ol ages, which is an obse a ion ha , as discussed in he accompanying pape , could be expec ed i he lack o Oz a o s closed o inac i a ed con o ma ions o he channels. Delpiano and Heschele (1989) ha e epo ed in cell-a ached pa ches o glomus cells om abbi emb yos he exis ence o a K ÷ channel e e sibly inhibi ed by lowe ing Po2. This channel was eco ded du ing s a iona y memb ane depola iza- ions and he uni a y conduc ance alue was 137 pS wi h high K ÷ in he pipe e solu ion. These esul s a e di icul o compa e wi h ou own since in Delpiano and Heschele 's wo k he O2-sensi i e channel was no cha ac e ized and i does no on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 422 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 • 1992 include da a abou o he possible channel ypes. The kine ics and conduc ance o Delpiano and Heschele 's channel a e compa able o hose o he Kca channel, which, as shown be o e, is insensi i e o changes in Po2 in excised pa ches. Al hough we ha e obse ed in si u he same ypes o K + channels as in excised pa ches, we did no a emp an ini ial classi ica ion o K + channels in he cell-a ached con igu a ion because he memb ane po en ial, cy osolic Ca 2+ concen a ion, and o he a iables a e unknown. In addi ion, we ha e also obse ed ha ype I cells a e elec ically e y compac and he e o e cu en lowing h ough a cell-a ached pa ch can induce modi ica ions in he memb ane po en ial o he cell (Gan o nina, 1991). Ne e he- less, we canno elimina e he possibili y ha he e could be a change in he O2-sensing mechanisms o ca o id body cells du ing de elopmen (He zbe g, Hell- s 6m, Lage c an z, and Pequigno , 1990). In his espec i is in e es ing o no e ha whe eas emb yonic glomus cells seem o lack Na + channels (Heschele e al., 1989), in he adul issue la ge Na + cu en s can be eco ded (U e ia e al., 1989a). The ac ha he modula ion o Ko 2 channels by O2 is main ained o long pe iods o ime in excised pa ches s ongly sugges s ha he O2-Ko 2 channel in e ac ion occu s h ough an in insic senso o he plasma memb ane which may be pa o he channels o a molecule closely associa ed wi h hem. This in e ac ion seems o be di ec wi hou he pa icipa ion o soluble cy osolic media o s. We ha e sough o he possible in ol emen o memb ane-di usible G p o eins, which a e being ac i a ed can di ec ly egula e ionic channel ac i i y (Logo he is, Ku achi, Galpe , Nee , and Claphan, 1987; B own and Bi nbaume , 1988), wi h nega i e esul s. Bo h GTP-~/-S and F-, agen s ha i e e sibly ac i a e G p o eins (Gilman, 1987), we e ine ec i e in p e en ing he e e sibili y o he inhibi ion o Ko 2 channels by lowe ing Po2. None heless, we canno discoun ha he Ko 2 o o he channels o glomus cells migh be modula ed by cy osolic media o s in si u. Exposu e o hypoxia al e s he con en o cGMP and cAMP in he ca o id body (Wa y, Cheng, Dinge , and Fidone, 1989; P6 ez-Ga cia, Alma az, and Gonz~lez, 1990) and hese agen s a e known o egula e a b oad numbe o ionic channels. The Ko 2 channel may belong o a amily o 02 senso s b oadly dis ibu ed in na u e. Apa om he well-known 02 anspo unc ions o heme p o eins, he e a e, om bac e ia o mammalian cells, examples o heme-linked enzymes, he ac i i y o which is egula ed by en i onmen al 02 (Goldbe g, Dunning, and Bunn, 1988; Gilles-Gonz,41ez, Di a, and Helinski, 1991). C oss, Hende son, Jones, Delpiano, Hen schel, and Acke (1990) ha e ecen ly p oposed ha he ac i i y o a NADPH- oxidase in glomus cells (con aining a b- ype cy och ome) could be egula ed by 02, and ha his enzyme could de e mine he edox s a e o hiol g oups o p o eins and in luence he p ope ies o ionic channels. Physiological Signi icance o he Ko 2 Channel The Ko 2 channel ep esen s he i s known example o an ionic channel egula ed by 02. A simila ype o egula ion has been sough , bu no ound, in sep al neu ons (L6pez-L6pez e al., 1989) and in he small dopamine gic in e neu ons o he sympa he ic ganglia (S ea and Nu se, 1991), which a e de elopmen ally ela ed o glomus cells. Ou esul s demons a e he speci ici y o he Ko 2 channels loca ed in an O2- esponsi e cell and s ongly sugges ha hey ep esen he ini ial s ep in on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANVORNINA AND L6PEZ-BARNEO Single K + Channel Modula ed by Oe in Glomus Cells 423 chemo ansduc ion and hus con e upon glomus cells hei unique chemo ecep o p ope ies. Due o ou expe imen al equi emen s (see Me hods), he Poz alues o he hypoxic solu ions used in his and he accompanying pape a e much lowe han he ones ha can be a ained unde physiological condi ions; howe e , we ha e shown be o e ha he Popen o Ko~ channels is e e sibly modi ied by changes o Poz in a physiological ange (Gan o nina and L6pez-Ba neo, 1991). The lack o app e- ciable desensi iza ion in he O2-Ko 2 channel in e ac ion may ensu e ha changes o he physico-chemical a iable (02 ension) would be ansla ed in o a main ained elec ophysiological esponse. This cha ac e is ic could be ela ed o he ac ha single- ibe chemo ecep o a e en discha ges can be main ained o long pe iods du ing sus ained hypoxia (Nielsen, Bisga d, and Vid uk, 1988). The signi icance o Ko 2 channels o espi a o y physiology is ob ious bu hey may ha e a b oade unc ional, and pe haps pa hophysiological, ele ance. Simila ypes o channels may exis in lung al eolus and in he ine b anches o he pulmona y a e y and pa icipa e in he egula ion o egional pulmona y pe usion, o in small essels o b ain and hea issues whe e hey may con ibu e o he au o egula ion o blood low. The au ho s wish o hank D s. R. Ald ich, T. Hoshi, and W. Zago a (S an o d Uni e si y) o commen s on he manusc ip . This esea ch was suppo ed by a g an om he Di ecci6n Gene al de In es igaci6n Cien i ica y T~cnica (PB86-0250). O iginal e sion ecei ed 18 Decembe 1991 and accep ed e sion ecei ed 16 Ap il 1992. REFERENCES Acke , H., and F. Pie uschka. 1977. Meaning o he ype I cell o he chemo ecep i e p ocess: an elec ophysiological s udy on cul u ed ype I cells o he ca o id body. In Chemo ecep ion in he Ca o id Body. H. Acke , S. Fidone, D. Pallo , C. Eyzagui e, D. W. Liibbe s, and R. W. To ance, edi o s. Sp inge -Ve lag, Be lin. 92-98. Adams, D. J., and W. Nonne . 1989. Vol age-dependen po assium channels: ga ing, ion pe mea ion and block. In Po assium Channels: S uc u e, Classi ica ion, Func ion and The apeu ic Po en ial. D. L. Cook, edi o . Ellis Ho wood Limi ed, New Yo k. 40-69. Alma az, L., C. Gonz~ilez, and A. Obeso. 1986. E ec s o high po assium on he elease o [3H] dopamine om he ca ca o id body in i o. Jou nal o Physiology. 379:293-307. Ba e , J. N., K. L. Magleby, and B. S. Pallo a. 1982. P ope ies o single calcium-ac i a ed po assium channels in cul u ed a muscle. Jou nal o Physiology. 331:211-230. Bigay, J., P. De e e, C. P is e , and M. Chab e. 1985. Fluo oalumina es ac i a e ansducin-GDP by mimicking he gamma-phospha e o GTP in i s binding si e. FEBS Le e s. 191:181-185. Biscoe, T.J., and M. R. Duchen. 1989. Elec ophysiological esponses o dissocia ed ype I cells o he abbi ca o id body o cyanide.Jou nal o Physiology. 413:447-468. Biscoe, T. J., and M. R. Duchen. 1990a. Responses o ype I cells dissocia ed om he abbi ca o id body o hypoxia. Jou nal o Physiology. 428:39--59. Biscoe, T. J., and M. R. Duchen. 1990b. Cellula basis o ansduc ion in ca o id chemo ecep o s. Ame ican Jou nal o Physiology. 258:L271-L278. Bla z, A. L., and K. L. Magleby. 1987. Calcium-ac i a ed po assium channels. T ends in Neu osciences. 10:463-467. on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 424 THE JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 100 • 1992 B own, A. M., and L. Bi nbaume . 1988. Di ec G p o ein ga ing o ion channels. Ame ican Jou nal o Physiology. 254:H401-H410. Colquhoun, D., and F. L. Sigwo h. 1983. Fi ing and s a is ical analysis o single channel eco ds. In Single Channel Reco dings. B. Sakmann and E. Nehe , edi o s. Plenum Publishing Co p., New Yo k. 191-263. Con i, F., B. Hille, and W. Nonne . 1984. Non-s a iona y luc ua ions o he po assium conduc ance a he node o Ran ie o he og.Jou nal o Physiology. 353:199-230. Coope , E., and A. Sh ie . 1985. Single-channel analysis o as ansien po assium cu en s om a nodose neu ones. Jou nal o Physiology. 369:199-208. Coope , E., and A. Sh ie . 1989. Inac i a ion o A cu en s and A channels on a nodose neu ons in cul u e.Jou nal o Gene al Physiology. 94:881-910. C oss, A. R., L. Hende son, O. T. G. Jones, M. A. Delpiano, J. Hen schel, and H. Acke . 1990. In ol emen o an NAD(P)H oxidase as a Po2 senso p o ein in he a ca o id body. Biochemical Jou nal. 272:743-747. De Cas o, F. 1926. Su la s uc u e e l'inne a ion de la glande in e ca o idienne (glomus ca o icum) de l'homme e des mami ~ es, e su un nou eau sys ~me d'inne a ion au onome du ne glossopha yngien. T abajos del Labo a o io de ln es igaciones Biol6gicas de la Uni e sidad de Mad /d. 24:365--432. Delpiano, M. A., andJ. Heschele . 1989. E idence o a Po~-sensi i e K + channel in he yped cell o he abbi ca o id body. FEBS Le e s. 249:195-198. Dubois, J. M. 1983. Po assium cu en s in he og node o Ran ie . P og ess in Biophysics and Molecula Biology. 42:1-20. Duchen, M. R., K. W. T. Caddy, G. C. Ki by, D. L. Pa e son, J. Pon e, and T. J. Biscoe. 1988. Biophysical s udies o he cellula elemen s o he abbi ca o id body. Neu oscience. 26:291-311. Eyzagui e, C., R. S. Fi zge ald, S. Lahi i, and P. Zapa a. 1983. A e ial chemo ecep o s. In Handbook o Physiology: The Ca dio ascula Sys em. J. T. Sheppe d, and F. Abboud, edi o s. Ame ican Physiological Socie y, Be hesda, MD. 557-621. Fidone, S. J., and C. Gonz~ilez. 1986. Ini ia ion and con ol o chemo ecep o ac i i y in he ca o id body. In Handbook o Physiology: The Respi a o y Sys em II. A. P. Fishman, edi o . Ame ican Physiological Socie y, Be hesda, MD. 247-312. Fidone, S.J., C. Gonz~ilez, and K. Yoshizaki. 1982. E ec s o low oxygen on he elease o dopamine om he abbi ca o id body in i o. Jou nal o Physiology. 333:93-110. Fishman, M. C., W. L. G eene, and D. Pla ika. 1985. Oxygen chemo ecep ion by ca o id body cells in cul u e. P oceedings o he Na ional Academy o Sciences, USA. 82:1448-1450. Fi zge ald, R. S., and S. Lahi i. 1986. Re lex esponses o chemo ecep o s imula ion. In Handbook o Physiology: The Respi a o y Sys em II. A. P. Fishman, edi o . Ame ican Physiological Socie y, Be hesda, MD. 313-362. Gan o nina, M. D. 1991. Es udio elec o isiol6gico del canal Ko 2, un canal de K ÷ modulado po 02 en las c ilulas quimio ecep o as del cue po ca o ideo. Doc o al hesis. Uni e sidad de Se illa, Se illa, Spain. 183 pp. Gan o nina, M. D., and J. L6pez-Ba neo. 1991. Single K + channels in memb ane pa ches o a e ial chemo ecep o cells a e modula ed by 02 ension. P oceedings o he Na ional Academy o Sciences, USA. 88:2927-2930. Gilles-Gonz~ilez, M. A., G. S. Di a, and D. R. Helinski. 1991. A haemop o ein wi h kinase ac i i y encoded by he oxygen senso o Rhizobium melilo i. Na u e. 350:170-172. Gilman, A. G. 1987. G p o eins: ansduce s o ecep o -gene a ed signals. Annual Re iew o Biochemis y. 56:615-649. on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992 GANFORNINA AND L6PEZ-BARNEO Single K + Channel Modula ed by 02 m Glomus Cells 425 Goldbe g, M. A., S. P. Dunning, and H. F. Bunn. 1988. Regula ion o he e y h opoie in gene: e idence ha he oxygen senso is a heme p o ein. Science. 242:1412-1415. HamiU, O. P., A. Ma y, E. Nehe , B. Sakmann, and F. J. Sigwo h. 1981. Imp o ed pa ch-clamp echniques o high- esolu ion cu en eco ding om cells and cell- ee memb ane pa ches. P l ige s A chi . 391:85-100. He zbe g, T., S. Hells 6m, H. Lage c an z, and J. M. Pequiguo . 1990. De elopmen o he a e ial chemo e lex and u no e o ca o id body ca echolamines in he newbo n a . Jou nal o Physiology. 425:211-225. Heschele , J., M. A. Delpiano, H. Acke , and F. Pie nschka. 1989. Ionic cu en s on ype-I cells o he abbi ca o id body measu ed by ol age-clamp expe imen s and he e ec o hypoxia. B ain Resea ch. 486:79--88. Heymans, C., J. J. Bouckae , and L. Dau ebande. 1930. Sinus ca o idien e 6 lexes espi a oi es. II. In luences espi a oi es ~ lexes de l'acid6se, de l'alcal6se, de l'anhyd ide ca bonique, de I'ion hyd og6ne e de I'anox6mie: sinus ca o idiens e 6changes espi a oi es dans les poumons e au dei~ des poumons. A chi es In e na ionales de Pha macodynamie e de Thi apie. 39:400--408. HiUe, B. 1984. Ionic channels o exci able memb anes. Sinaue Associa es, Inc., Sunde land, MA. 426 pP. Hoshi, T., and R. W. Ald ich. 1988. Vol age-dependen K + cu en s and unde lying single K + channels in pheoch omocy oma cells.Jou nal o Gene al Physiology. 91:73-106. Kasai, H., M. Kameyama, K. Yamaguchi, and J. Fukuda. 1986. Single ansien K channels in mammalian senso y neu ons. Biophysical Jou nal. 49:1243-1247. Llano, I., C. K. Webb, and F. BezaniUa. 1988. Po assium conduc ance o he squid gian axon.Jou nal o Gene al Physiology. 92:179-196. Logo he is, D. E., Y. Ku achi, J. Galpe , E. J. Nee , and D. E. Claphan. 1987. The [3 subuni s o GTP-binding p o eins ac i a e he musca inic K ÷ channel in hea . Na u e. 325:321-326. L6pez-Ba neo, J., J. R. L6pez-L6pez, J. U e ia, and C. Gonz~ilez. 1988. Chemo ansduc ion in he ca o id body: K ÷ cu en modula ed by Po~ in ype I chemo ecep o cells. Science. 241:580-582. L6pez-L6pez, J., C. Gonz~ilez, J. U e ia, and J. L6pez-Ba neo. 1989. Low Po~ selec i ely inhibi s K channel ac i i y in chemo ecep o cells o he mammalian ca o id body. Jou nal o Gene al Physiology. 93:1001-1015. Ma y, A. 1981. Ca-dependen K channels wi h la ge uni a y conduc ance in ch oma in cell memb anes. Na u e. 291:497-500. Ma y, A., and E. Nehe . 1985. Po assium channels in cul u ed bo ine ad enal ch oma in cells. Jou nal o Physiology. 367:117-141. Nielsen, A. M., G. E. Bisga d, and E. H. Vid uk. 1988. Ca o id chemo ecep o ac i i y du ing acu e and sus ained hypoxia in goa s.Jou nal o Applied Physiology. 65:1796-1802. Obeso, A., S. Fidone, and C. Gonz~lez. 1987. Pa hways o calcium en y in o ype I cells: signi icance o he sec e o y esponse. In Chemo ecep o s in Respi a o y Con ol. J. A. Ribei o and D. J. Pallo , edi o s. C oom Helm, London. 91-97. Pa lak, J., and R. Ho n. 1982. E ec o N-b omoace amide on single sodium channel cu en s in excised memb ane pa ches.Jou nal o Gene al Physiology. 79:333-351. Pee s, C. 1990. E ec s o D600 on hypoxic supp ession o K + cu en s in isola ed ype I ca o id body cells o he neona al a . FEBS Le e s. 271:37-40. P6 ez-Ga c a, M. T., L. Alma az, and C. Gonz~ lez. 990. E ec s o di e en ypes o s imula ion on cyclic AMP con en in he abbi ca o id body: unc ional signi icance. Jou nal o Neu ochemis y. 55:1287-1293. Quand , F. N. 1988. Th ee kine ically dis inc po assium channels in mouse neu oblas oma cells. Jou nal o Physiology. 395:401-418. on Sep embe 22, 2014jgp. up ess.o gDownloaded om Published Sep embe 1, 1992