Acidic regulation of junction channels between glomus cells in the rat carotid body. Possible role of [Ca(2+)](i).

The purpose of this work was to characterize the gap junctions between cultured glomus cells of the rat carotid body and to assess the effects of acidity and accompanying changes in [Ca(2+)](i) on electric coupling. Dual voltage clamping of coupled glomus cells showed a mean macrojunctional conductance (G(j)) of 1.16 nS+/-0.6 (S.E.), range 0.15-4.86 nS. At normal pH(o) (7.43), a steady transjunctional voltage (DeltaV(j)=100.1+/-10.9 mV) showed multiple junction channel activity with a mean microconductance (g(j)) of 93.98+/-0.6 pS, range 0.3-324.5 pS. Single-channel conductances, calculated as variance/mean g(j), gave a mean value of 16.7+/-0.2 pS, range 5.13-39.38 pS. Manual measurements of single-channel activity showed a mean g(j) of 22.03+/-0.2 pS, range 1.3-160 pS. Computer analysis of the noise spectral density distribution gave a channel mean open time of 12.7+/-1.5 ms, range 6.37-23.42 ms. The number of junction channels, estimated in each experiment from G(j)/single-channel g(j), showed a range of 7 to 258 channels (mean, 107.2). Optical measurements of [Ca(2+)](i) gave a mean value of 80.2+/-4.27 nM at pH(o) of 7.43. Acidification of the medium with lactic acid (1 mM, pH 6.3) induced: 1) Variable changes in G(j) (decreases and increases); 2) A significant decrease in mean g(j) (to 80.36+/-0.34 pS) and in single-channel conductance (g(j)=12.8+/-0.2 pS in computer analyses and 17.23+/-0.2 pS when measured by hand); 3) Variable changes in open times, resulting in a similar mean (12.8+/-1.5 ms) and 4) No change in the number of junction channels. When pH(o) was lowered to 6.3 [Ca(2+)](i) did not change significantly (there were increases and decreases). However, when pH(o) was lowered to 4.4, [Ca(2+)](i) increased significantly to 157.1+/-8.1 nM. It is concluded that saline acidification to pH 6.3 depresses the conductance of junction channels and this effect may be either a direct effect on channel proteins or synergistically enhanced by increases in [Ca(2+)](i). However, there are no studies correlating changes of [Ca(2+)](i) and intercellular coupling in glomus cells. Stronger acidification (pH(o) 4.4), producing much larger changes in [Ca(2+)](i), may enhance this synergism. But, again, there are no studies correlating these effects.

Short- and long-term regulation of rat carotid body gap junctions by cAMP. Identification of connexin43, a gap junction subunit.

Intact and cultured carotid bodies (CBs) of the rat were used in this study. Applications of membrane-permeant db-cAMP to cultured carotid bodies increased electric coupling between most glomus cells (increasing junctional conductance) probably by opening preformed intercellular channels. This a short-term effect of the nucleotide, increasing gating between glomus cells. When cultures and intact carotid bodies were treated with membrane-permeant 8Br-cAMP for 3 h or more (to increase cytosolic cAMP), there was enhanced gap junction formation and better dye spread between carotid body cells. Connexin43 (CX43) was identified by immunocytochemical methods as forming part of the intercellular channels between carotid body cells, and the expression of Cx43 increased by cAMP. This is a long-term effect, inducing the formation of gap junctions. Thus, cAMP had short and long-term effects on the intercellular junctions of the carotid body. Long-term formation of gap junctions may be important in modulating carotid body functions during stimulation by chronic hypoxia.

Cells of the carotid body express connexin43 which is up-regulated by cAMP.

We identified a gap junction protein subunit, connexin43 (Cx43) by immunofluorescence and immunoblotting, in cultured rat carotid body cells and in whole organs. In 1-week-old cultures, all cells were flat but after 3 h exposure to 8Br-cAMP (1 mM), tyrosine hydroxylase (TH) positive cells (chemoreceptors), but not TH negative cells, adopted a round body with multiple thin arborization processes. The incidence of dye coupling between cultured cells of the same type increased from 26% in controls to 73% after treatment with 8Br-cAMP. In control cultures, Cx43 immunoreactivity showed a diffuse perinuclear distribution and after 8Br-cAMP treatment, it was also found at cell-cell contacts. Both 8Br-cAMP-induced dye coupling and cellular redistribution of Cx43 were blocked by pretreatment with actinomycin D (5 microM), a mRNA transcription blocker. Moreover, 3 h exposure to 8Br-cAMP increased the levels of Cx43 in entire organs. We suggest that conditions that promote a sustained increase in cytosolic cAMP up-regulate coupling between carotid body cells in a transcription-dependent manner. The possible functional significance of these findings is discussed.

Carotid body glomus cells: chemical secretion and transmission (modulation?) across cell-nerve ending junctions.

Glomus cells of the carotid body contain and secrete chemicals during 'natural' stimulation (hypoxia, hypercapnia and acidity), thus, the birth of the 'transmitter hypothesis of chemoreception'. Released chemicals would cross the synaptic cleft between glomus cells and carotid nerve terminals to depolarize the nerve ending membrane during excitation and hyperpolarize the membrane during inhibition. The main problem with this hypothesis is that specific synaptic blockers modify but do not block the effects of natural stimulation, while blocking the effects of the putative transmitters. It is proposed in this review that the secretion of chemicals is modulated by changes in electric coupling between glomus cells and that glomus cell-nerve ending transmission is not blocked by specific blockers for two reasons. One is that multiple transmitters are released. The other, and more the likely explanation, is that there are electric connections between these elements allowing the flow of currents that are unaffected by the blockers.

Modulation of junctional conductance between rat carotid body glomus cells by hypoxia, cAMP and acidity.

Short-term cultures of glomus cells (up to seven days), were employed to study intercellular electrical communications. Bidirectional electric coupling was established under current clamping after impaling two adjacent glomus cells with microelectrodes, and alternate stimulation and recording. Their resting potential (Vm) and input resistance (Ro) were thus measured. Both coupled cells were then voltage clamped at a level between their Vms. Current pulses applied to either cell elicited a transjunctional voltage (Vj) and current (Ij), used to calculate the junctional conductance (Gj). Gj was 1.52+/-0.29 nS (mean+/-S.E.; n=147). Vj linearly influenced Gj, suggesting ohmic junctions. Gj was not affected by Vm in 50% of the cases. However, there was Vm-dependence in the others, but voltage changes had to be large (>+/-40 mV from the Vm). Therefore, physiologically or pharmacologically induced glomus cell depolarization or hyperpolarization may not significantly affect intercellular coupling unless there are large variations in Vm. Hypoxia (induced by Na2S2O4 1 mM or 100% N2) decreased Gj in 60-80% of the pairs while producing tighter coupling in the rest. Similar effects were obtained when the medium was acidified with lactic acid 1-10 mM. Cobalt chloride (3 mM) prevented, diminished or reversed the changes in Gj observed during low PO2, suggesting that [Ca2+]i changes are important in hypoxic uncoupling. However, non-specific cationic effects of Co2+ have not been ruled out. Applications of the membrane-permeant dB-cAMP 1 mM tightened coupling in almost all cell pairs. This is important because endogenous cAMP increases during hypoxia. Our results suggest that multiple factors modulate junctional conductance between glomus cells. Changes in Gj by 'natural' stimuli and/or cAMP may play an important role in chemoreception, especially in titrating the release of transmitters toward the carotid nerve terminals.

Effects of calcium on the electric coupling of carotid body glomus cells.

Pairs of electrically coupled glomus cells from rat carotid bodies were impaled with microelectrodes. In the current clamp mode, intracellular stimulation and recording established the coupling coefficient (KC), across the intercellular junctions. About 80% of 26 pairs uncoupled during exposure to 9.45 mM [Ca2+]o, and about 72% of 18 pairs showed the same effect during applications of ionophore A23187. During superfusion with zero [Ca2+]o and EGTA, about 73% of 40 pairs of cells became more tightly coupled. Similar results (71%) were obtained during exposure of 42 cell pairs to BAPTA/AM, a membrane-permeant calcium chelator. Thus, [Ca2+]i seemed to play a significant a role in glomus cell intercellular communication. A23187 and BAPTA/AM, dissolved in DMSO, tended to reduce intercellular coupling during prolonged exposures of the preparations to this solvent. Consequently, the effects elicited by A23187 and BAPTA/AM were superimposed on a coupling effect produced by DMSO.

Effects of hypoxia on the intercellular channel activity of cultured glomus cells.

Dual voltage clamp experiments have shown that hypoxia induced by Na-dithionite or N2 reduced junctional macroconductance (Gj) in about 70% of cultured and coupled glomus cell pairs while increasing it in the rest. To explore possible mechanisms for these effects, we studied the activity of gap junction channels under similar conditions. The calculated single channel conductances (gj) fell into two categories. A low-conductance group, which was most frequently observed, had a mean gj of 27.8 +/- 0.29 pS (mean +/- SEM; n = 968 events). The other group had higher conductances (47.6 +/- 0.35 pS; n = 528). When PO2 was reduced (hypoxia), the low conductances did not change significantly in any of the junctions. The high-conductance units appeared less frequently in some junctions whereas in others they remained unaltered. Thus, rapid channel flickering during hypoxia may not be the only mechanism determining Gj during coupling or uncoupling. It is possible that slow (seconds) opening and closing of the channels could play an important role in this phenomenon.

Possible role of coupling between glomus cells in carotid body chemoreception.

Glomus cells of the carotid body are dye and electrically coupled due to the presence of gap junctions between them. During stimulation by hypoxia, hypercapnia and acidity, about 70% of the cells uncouple to various degrees, whereas the rest either develop tighter coupling or are unaffected. Similar results have been obtained with exogenous administrations of naturally present transmitters such as dopamine and cholinergic substances. Uncoupling is associated with a decrease in junctional conductance and closing of intercellular channels. Tighter coupling is accompanied by opposite effects on these parameters. We think that cell isolation uncoupling leads to release of larger amounts of transmitters toward the carotid nerve sensory terminals. Tighter coupling would reduce the quantities of released chemicals. We may have a delicate titration process modulating the sensory discharge frequency, since a single sensory fiber divides to innervate up to 20 glomus cells. Thus, the discharge frequency of this fiber (the sensory unit) will result from the contributions of its many branches, each impinging on variously active glomus cells.

Electrical coupling between cultured glomus cells of the rat carotid body: observations with current and voltage clamping.

Electrically coupled pairs of cultured rat glomus cells were used. In one group of experiments, both cells were current-clamped. Delivery of positive or negative pulses to Cell 1 elicited appreciable voltage noise in this cell and large action potentials (probably Ca2+ spikes) in about 10% of them. Both passive and active electrical events spread to Cell 2, presumably through the gap junctions between them. The coupling coefficient (Kc) was larger for the spikes than for non-regenerative voltage noise. In another group of experiments, Cell 1 was current-clamped and Cell 2 was voltage-clamped at Cell 1 EM. Pulses of either polarity, delivered to Cell 1, produced current flow through the intercellular junction and allowed direct measurements of junctional currents (Ij) and total conductances (Gj). Ij had a mean value of about 12.5 pA and Gj of 391 pS. Unitary (presumably single channel) conductance (gj) was about 78 pS.

Electrical communication between glomus cells of the rat carotid body.

Glomus cells of rat carotid bodies can be electrotonically coupled. This was determined by simultaneous intracellular recording and stimulation of two neighboring cells. Voltage applied into one cell (V1), was detected in the other cell as E2. The ratio E2/V1 or coupling coefficient (KC), varied from 0.003 to 1. R0 or input resistance (24.1-3,500 M omega), was calculated from the voltage elicited in the injected cell by current injection (V1/I1). The coupling resistance (RC) was estimated by using Bennett's model and was inversely related to KC. It ranged from 8.5 to 46,112 M omega. Values for KC are provisional since we may not have always recorded from immediately adjacent cells. Similarly, calculations of R0 and RC may not be accurate since, in all probability, there is a multicellular network. Stimulation by hypoxia (100% N2 or Na2S2O4), acidity (lactic acid or 100% CO2), dopamine, ACh, nicotine and bethanechol depolarized the majority of glomus cells, their input resistance decreased and cells became uncoupled. Fewer cells were either unaffected or coupling increased. There was a significant and negative correlation between changes in coupling coefficient and in coupling resistance.

Effects of dopamine on type I chemoreceptor cells of the rat carotid body.

The effects of dopamine (DA) were studied on type I cells of the rat carotid body. DA (0.35 to 8.4 x 10(-4) M) elicited a long-lasting depolarization and decreased input resistance (Ro) in 75% of the cells, in a dose-dependent manner. Hyperpolarization and increased Ro were observed in 8% of cells. Membrane potential (Em) and Ro changes were highly correlated (-0.62, P < 0.001). The reversal potential (Erev) of the DA effect was +2.16 +/- 4.75 (mV +/- S.E.M.).