Developmental change of T-type Ca²⁺ channel expression and its role in rat chromaffin cell responsiveness to acute hypoxia

Abstract

Neonatal chromaffin cells of the adrenal medulla (AM) are intrinsic chemoreceptors that secrete catecholamines in response to hypoxia, thus contributing to fetal adaptation to extrauterine life. In most mammals studied, oxygen sensitivity of AM cells disappears a few days after birth, possibly due to innervation of the adrenal gland by the cholinergic fibres of the splanchnic nerve (∼postnatal day 7 in the rat). The mechanisms underlying these homeostatic changes in chromaffin cells are unknown. Low voltage-activated, T-type, Ca²⁺ channels regulate cell excitability and their expression is up-regulated by hypoxia. Hence, we hypothesized that these channels contribute to the developmental changes in the chemoreceptive properties of AM chromaffin cells. Using electrophysiological, immunocytochemical and molecular biology methodologies we show here that neonatal AM chromaffin cells express T-type Ca²⁺ channels (of α1H or Ca_v3.2 sub-type) and that the function of these channels is necessary for catecholamine release in response to acute hypoxia. T-type Ca²⁺ channel expression, as well as chromaffin cell responsiveness to hypoxia, decrease with postnatal maturation. Adult chromaffin cell sensitivity to hypoxia reappears after AM denervation in parallel with the recruitment of T-type Ca²⁺ channels. These observations indicate that T-type Ca²⁺ channels are essential for the acute response of chromaffin cells to hypoxia and help explain the disappearance of O₂ sensitivity in adult AM chromaffin cells. Our results may also be relevant for understanding the pathogenesis of disorders associated with chronic hypoxia or maternal nicotine consumption.

Catecholamine release from chromaffin cells of the adrenal medulla (AM) is a fundamental physiological reaction to stress situations, such as hypoxia and hypercapnia. This homeostatic response of AM cells is particularly critical in the newborn to trigger the metabolic, respiratory and cardiovascular changes necessary for adaptation to extrauterine life (Comline & Silver, 1966; Lagercrantz & Bistoletti, 1977; Jones, 1980; Faxelius et al. 1984). In adult mammals, catecholamines, especially adrenaline, are produced by AM chromaffin cells and secreted to the circulation after neurogenic stimulation through the splanchnic nerve. As innervation of the AM in some mammals is only completed during postnatal life (approximately after the first week of age in the rat; Seidler & Slotkin, 1985), maintenance of the adrenergic activity in the neonatal adrenal gland is supported by specialized, non-neurogenic, mechanisms capable of eliciting catecholamine release in the absence of neural input. It was known that a non-neurogenic secretory response to hypoxia exists in adrenal glands of newborn rats (Seidler & Slotkin, 1985, 1986), and, more recently, direct activation of neonatal AM chromaffin cells by CO₂ has also been reported (Muñoz-Cabello et al. 2005).

Regarding the neonatal sensitivity to hypoxia, it has been shown that AM chromaffin cells are O₂ sensors that, similar to glomus cells of the carotid body (see for review López-Barneo, 2003), release catecholamines in response to local decreases of O₂ tension (Mochizuki-Oda et al. 1997; Mojet et al. 1997; Thompson et al. 1997; García-Fernández et al. 2007). It is generally believed that hypoxia induces AM chromaffin cell depolarization and catecholamine release primarily due to inhibition of K⁺ conductances (see for review Nurse et al. 2006). O₂-sensitive K⁺ currents have been reported in chromaffin cells from various species (Mochizuki-Oda et al. 1997; Thompson et al. 1997, 2002; Rychkov et al. 1998; Lee et al. 2000; Keating et al. 2001, 2004). On the other hand, activation by hypoxia of a depolarizing cationic conductance has also been proposed (Inoue et al. 1998). However, there is no consensus on whether the intrinsic newborn chromaffin cell sensitivity to hypoxia is maintained in the adult. Some studies have described a developmental loss of hypoxia responsiveness in AM cells (Mojet et al. 1997; Thompson et al. 1997, 2002; Keating et al. 2004; Rico et al. 2005; García-Fernández et al. 2007), although chemosensitivity can re-emerge if the adult AM gland is deprived of neural input (Seidler & Slotkin, 1986; Cheung, 1990). In contrast, others have suggested that the ‘non-neurogenic’ intrinsic O₂ sensitivity is maintained in adult chromaffin cells (Mochizuki-Oda et al. 1997; Inoue et al. 1998; Lee et al. 2000; Takeuchi et al. 2001).

The current study was designed to investigate the role of T-type Ca²⁺ channels in the responsiveness of chromaffin cells to acute hypoxia. T-type Ca²⁺ channels regulate cell excitability in numerous tissues (see, for review, Perez-Reyes, 2003) and, although poorly expressed in adult chromaffin cells, they can be recruited in stressful conditions and, thus, contribute to regulating cell resting membrane potential and catecholamine exocytosis (Carbone et al. 2006; Giancippoli et al. 2006). The gene encoding the T-type Ca²⁺ channel α1H subunit is up-regulated in chronic hypoxia both in PC12 (del Toro et al. 2003) and adult chromaffin (Carabelli et al. 2007) cells. In this last preparation, the increase of T-type Ca²⁺ channel availability results in a reduction of the resting membrane potential and enhanced responsiveness of the cells to mild depolarizations. Hence, it could be that modifications in the level of α1H expression contribute to the ontogenic changes in the hypoxic chemosensitivity of AM chromaffin cells. Herein we show that in contrast with adult cells, neonatal chromaffin cells express relatively high levels of α1H T-type Ca²⁺ channels and that the function of these channels is required for a proper secretory response to acute hypoxia. T-type Ca²⁺ channel expression in AM cells is almost entirely lost in young adults, as it occurs with the sensitivity to hypoxia. However, in adult animals T-type Ca²⁺ channel expression and O₂ chemosensitivity reappear upon chromaffin cell denervation.

Methods

Isolation of chromaffin cells

Neonatal (P0–P5) and adult (4–12 months) Wistar rats were used for this work. All animals were obtained from the central animal facilities of the University of Seville, housed in a temperature-and humidity-controlled room with a 12 h light–dark cycle and allowed free access to food and water. Procedures followed in animal care and experimentation were approved by the Experimental Animal Committee of our institution. Animals were anaesthetized by i.p. injection of 350 mg kg⁻¹ chloral hydrate (Ferosa, Spain), using a stock solution of 70 mg ml⁻¹ in 0.9% (w/v) sodium chloride. After rat decapitation, adrenal glands were quickly excised and placed on ice-cooled and O₂-saturated modified Tyrode solution (in mm: 148 NaCl, 2 KCl, 3 MgCl₂, 10 Hepes, 10 glucose; pH 7.4). Chromaffin cells were dispersed as previously described (Muñoz-Cabello et al. 2005). The pellet was resuspended in 100–150 μl of Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mm l-glutamine, 100 units ml⁻¹ penicillin and 100 mg ml⁻¹ streptomycin, and plated on slivers of glass coverslips previously treated with poly-l-lysine (1 mg ml⁻¹). Freshly dispersed cells were used for the experiments.

Adrenal gland slice preparation

Adrenal glands from rats were removed and transferred to ice-cooled Tyrode solution. Slices were prepared as previously described (García-Fernández et al. 2007). Briefly, after cleaning from adrenal cortex and inclusion into low-melting temperature agarose, the adrenal gland was glued on a vibratome stage (Leica Mycrosystems, Germany). After cutting, adult or neonatal slices (120 μm thick) were incubated for 15 min in a solution containing (in mm): 120 NaCl, 2 KCl, 23 NaHCO₃, 1 MgCl₂, 2.5 CaCl₂, 10 glucose (recording solution in amperometric experiments) bubbled with 95% O₂ and 5% CO₂ at 37°C. Thereafter slices were maintained at room temperature in the recording solution and used for the experiments not longer than 6 h after slicing. A clear secretory burst in response to high extracellular K⁺ was considered a good and reproducible indication of cell viability in all experiments.

Electrophysiological recordings

Macroscopic Ca²⁺ and Na⁺ currents were recorded using the whole-cell configuration of the patch-clamp technique as adapted in our laboratory (del Toro et al. 2003; Muñoz-Cabello et al. 2005). Patch electrodes (1–2 MΩ) were pulled from capillary glass tubes (Kimax; 1.5–1.6 mm OD, Kimble Products), fire polished on a microforge MF-830 (Narishige, Japan) and coated with silicone elastomer (Sylgard 184; Dow Corning, USA) to decrease capacitance. Voltage-clamp recordings were obtained with an EPC-8 patch-clamp amplifier (Heka Elektronik, Germany) using standard voltage-clamp protocols designed with Pulse software (Heka Elektronik). Unless otherwise noted, holding potential was −80 mV. Data were filtered at 10 kHz, digitized at a sampling interval of 20 μs with an ITC-16 A/D converter (Instrutech, USA), and stored on a Macintosh computer. Off-line analysis of data was performed using custom software and Pulse Fit (Heka Elektronik). All experiments were conducted at room temperature, 22–25°C. For whole-cell patch recordings, the internal solution contained (in mm): 110 CsCl, 30 CsF, 10 EGTA, 10 Hepes and 4 ATP-Mg; pH was adjusted with CsOH to 7.2, and osmolality was 285 mosmol kg⁻¹. The standard bath solution contained (in mm): 140 NaCl, 9 BaCl₂, 1 CaCl₂, 10 Hepes and 10 glucose; pH was adjusted with NaOH to 7.4 and osmolality was 300 mosmol kg⁻¹.

In some experiments we monitored the action potential firing frequency in intact chromaffin cells using the cell-attached configuration of the patch-clamp technique. In this condition the transmembrane current flow during an action potential is manifested as a bi-/tri-phasic action current signal (Fenwick et al. 1982; Montoro et al. 1996). Generation of action potentials was favoured by application of hyperpolarizing (−40 mV) voltage (Fenwick et al. 1982). In these experiments we used a pipette-filling solution similar to that in the bath (in mm: 140 NaCl, 2.5 KCl, 10 Hepes, 10 glucose, 2.5 CaCl₂ and 4 MgCl₂).

Experimental setup and amperometric monitoring of secretion

In each experiment, a slice was transferred to a recording chamber (∼200 μl volume) where it was continuously superfused with the recording external solution. A lyre-like device made of silver wire with glued fine nylon threads was used to hold the slice to the bottom of the chamber. The standard (normoxic) solution was bubbled with a gas mixture of 5% CO₂, 20% O₂ and 75% N₂. Hypoxia was obtained by continuously bubbling the solution in one of the reservoirs with 5% CO₂ and 95% N₂. The pH of the normoxic and hypoxic solutions was 7.4. In the high K⁺ solutions, 40 mm KCl replaced NaCl equimolarly. In all the experiments, the temperature of the solutions and the chamber was maintained between 22 and 25°C. Secretory events were recorded with an 8 μm carbon fibre electrode connected to the input of a high-gain current-to-voltage converter (Ortega-Sáenz et al. 2003; García-Fernández et al. 2007). The electrode was polarized to +800 mV, a value more positive than the oxidation potential of catecholamines in AM chromaffin cells. Amperometric currents were recorded with an EPC-8 patch-clamp amplifier (Heka Electronics, Germany). The currents were filtered at 100 Hz and digitized at 250 Hz before storage on the computer. Analysis of the amperometric data was made using Igor Pro (WaveMetrics, USA) software.

RNA extraction, reverse transcription and polymerase chain reaction

PolyA⁺ RNA was extracted from chromaffin cells using the Dynabeads mRNA Direct micro kit (Dynal, Norway) as indicated by the manufacturer. The reverse transcription (RT) reaction was performed immediately after mRNA isolation. First-strand cDNA was synthesized from total RNA extraction using the Superscript first-strand synthesis system for RT-PCR (Invitrogen, Life Technologies) with random primers according to manufacturer's directions.

Real-time PCR

Real-time PCR reactions were run on an Abiprism 7000 (Applied Biosystems) device following the thermocycler conditions recommended by the manufacturer. Quantitative PCR reactions were performed with the SYBR Green PCR Master mix (Applied Biosystems) by triplicates in a 25 μl reaction mixture containing 1× SYBR Green PCR Master mix (included ROX as a passive reference dye) and 1 or 2 μl of the RT reaction. Primers were designed using the computer program Primer Express (Applied Biosystems). The primers used for rat α1H (Genebank accession number AF290213) were: forward 5′-GGCGTGGTGGTGGAGAACTT-3′ and reverse 5′-GATGATGGTGGGATTGAT-3′. Each sample was analysed for cyclophilin A to normalize for RNA input amounts and to perform relative quantifications. Melting curve analysis showed a single sharp peak with the expected melting temperature for all samples.

Immunocytochemistry and confocal microscopy

Adrenal glands were removed from anaesthetized animals, immediately fixed with 4% paraformaldehyde in PBS in mm: 137 NaCl, 2.7 KCl, 10 Na₂HPO₄; pH 7.4 overnight at 4°C, and embedded in paraffin. Slices, 5–10 μm thick, were cut with a microtome (Leica RM2125, Germany). Tyrosine hydroxylase (TH) immunohistological detection was done as described previously (Pardal et al. 2000). Ca_v3.2 expression was analysed with an anti-Ca_v3.2 antibody (1: 100; Alomone). In both cases a biotin-conjugated anti-rabbit antibody (1: 200; Pierce, Rockport, IL) was used for detection.

For fluorescent analysis, slices were rinsed in PBS and incubated 30 min in a non-specific binding blocking solution (3% fetal calf serum (FCS) in PBS, without Triton). We used an anti-Ca_v3.2 antibody (1: 100; Alomone) and anti-TH (1: 200; T-1299, Sigma) as primary antibodies, and a polyclonal donkey anti-rabbit antibody conjugated to Alexa 488 (1: 700; Molecular Probes) and CY3-conjugated donkey anti-mouse (1: 400; Jackson Immunoresearch) as secondary antibodies, respectively. Fluorescent image acquisition by sequential scan was performed using a confocal microscope (TCS SP2, Leica, Germany).

Adrenal denervation

Surgical procedures were performed under chloral hydrate anaesthesia. Bilateral adrenal denervations were made in adult male rats (4–6 weeks) through a ventral midline incision. After dissection of the suprarenal region a smaIl fragment of the splanchnic nerve was excised (Seidler & Slotkin, 1986). Care was taken to ensure that the adrenal blood supply was not disturbed. When post-mortem inspection revealed adrenal atrophy or other signs of impaired circulation to the tissue, the affected animal was eliminated from the study. Bilaterally denervated animals were killed 4–6 weeks after surgery. Each adrenal gland was divided in two halves: one of them was fixed with paraformaldehyde (4%) for subsequent immunohistochemical studies; the other half was used either for slicing (to carry out amperometric recordings) or for the preparation of dispersed chromaffin cells (to do patch-clamp experiments). Sham-operated animals were subjected to similar abdominal surgery but without resection of the splanchnic nerve.

Statistical analysis

Data were analysed using a Student's t test for unpaired observations with Excel. Values are given as mean ±s.e.m.P values <0.05 were considered as statistically significant.

Results

T-type Ca²⁺ channels in neonatal rat chromaffin cells

The properties of the macroscopic voltage-dependent inward currents recorded in neonatal (<P5) rat chromaffin cells are shown in Fig. 1. Upon step depolarization to +20 mV a large rapidly inactivating inward Na⁺ current (I_Na) was generated followed by a smaller sustained Ca²⁺ current (I_Ca). At the end of the pulse a large Ca²⁺ tail was recorded (Fig. 1A). This tail current had fast and slowly deactivating components, suggesting the presence of two kinetically distinct populations of Ca²⁺ channels (Fig. 1B) (Matteson & Armstrong, 1986; Castellano & López-Barneo, 1991; del Toro et al. 2003). At −70 mV the macroscopic tail currents were fitted by exponential functions with time constants of 0.12 ± 0.01 ms and 1.75 ± 0.08 ms (n= 31 cells) for the fast and slow components, respectively. Therefore, besides the high-voltage-activated (HVA, fast deactivating) Ca²⁺ channels characteristic of adult rat chromaffin cells (see Cesetti et al. 2003), most neonatal cells seemed to have also slowly closing, low-voltage-activated (LVA) or T-type Ca²⁺ channels (see, for review, Huguenard, 1996). When Ca²⁺ currents were recorded in isolation (after replacement of external Na⁺ with N-methyl-d-glucamine) the T-type channels generated a transient component of the inward Ca²⁺ current that became appreciable during small depolarizations (from −100 to −40 mV) (Fig. 1C). The Ca²⁺ current–voltage relationship exhibited an activation threshold at ∼−50 mV, a maximal current amplitude at about +10 mV, and a shoulder in the negative voltage range (arrows in Fig. 1D), further indicating the existence of low- and high-voltage-activated Ca²⁺ channels (Carabelli et al. 2007). The low-threshold component of the current was reversibly inhibited by application to the external solution of 50 μm Ni²⁺, which at this concentration selectively inhibits T-type Ca²⁺ channels (Fig. 1D) (Kang et al. 2006). In fair agreement with this experimental observation, the slow component of the tails, representing the deactivation of the T-type channels, was reversibly reduced in amplitude by application of Ni²⁺, whereas that amplitude of the fast component, representing the activity of rapidly deactivating, HVA channels was unaltered (Fig. 1E and F). The partial reversibility of Ni²⁺ effects on the low-threshold component of the current–voltage curve may be due to slow washout of the metal, although it could also result from cumulative inactivation of the channels repeatedly subjected to the relatively slow depolarizing ramps. Altogether, these data indicate that neonatal chromaffin cells possess a significant population of T-type Ca²⁺ channels.

Figure 1. Expression of T-type Ca²⁺ channels in neonatal chromaffin cells.

A, representative macroscopic sodium (I_Na) and calcium (I_Ca) currents recorded in neonatal chromaffin cells subjected to whole-cell patch clamp during a depolarization to +20 mV from a holding potential of −80 mV. The decay of the tail current generated on repolarization to −70 mV reflects the closing time course of the channels open during the pulse. B, single (red (fast) or green (slow)) exponential functions fitted to the Ca²⁺ tail current. The fast and slow time constant values in this example are, respectively, 0.12 and 2.08 ms. C, family of Ca²⁺ current traces recorded in isolation after replacement of external Na⁺ with N-methyl-d-glucamine. The holding potential was −80 mV and the depolarization voltages (mV) are indicated near each trace. The arrows indicate the duration of the pulse (100 ms). Note that a measurable inward current was already appreciable at −40 mV. D, Ca²⁺ current–voltage relations obtained during depolarization ramps from −100 to +40 mV lasting 500 ms. The activation threshold of the current is at ∼−50 mV and a typical ‘shoulder’, due to the activation of low voltage-activated Ca²⁺ channels, appears preceding activation of a larger population of high voltage-activated channels. This ‘shoulder’ is abolished by application of Ni²⁺ (50 μm) to the external solution. E, selective inhibition of the slow component of the tail current by Ni²⁺. F, quantitative analysis of the selective effect of Ni²⁺ on the amplitude (current density) of fast and slowly-deactivating components of tail currents. *Statistically significant difference, P < 0.05; n= 8 cells.

The deactivation parameters of the T-type current in neonatal chromaffin cells suggested that it is mediated by the α1H (Ca_v3.2) subunit (McRory et al. 2001), the main contributor to the LVA Ca²⁺ current in PC12 cells (del Toro et al. 2003). This same channel subunit is the one recruited in adult rat chromaffin cells after exposure to hypoxia (Carabelli et al. 2007) or β-adrenergic stimulation (Novara et al. 2004). The nature of the T-type Ca²⁺ channel in neonatal chromaffin cells was confirmed by immunohistochemical experiments using specific antibodies against the α1H subunit. These experiments showed the appearance of selective α1H staining in the adrenal medulla (but not in the cortex) and, within the medulla, in chromaffin (tyrosine hydroxylase-positive) cells (Fig. 2).

Figure 2. Immunohistochemical identification of the α1H T-type calcium channel subunit in neonatal adrenal glands.

A, staining of adrenal medulla cells with antibodies against tyrosine hydroxylase (TH, left panel) and α1H (right panel). B, confocal images of a cluster of chromaffin cells illustrating the co-localization of TH and α1H.

Responsiveness of neonatal chromaffin cells to acute hypoxia is abolished after blockade of T-type Ca²⁺ channels

As shown before (García-Fernández et al. 2007), neonatal chromaffin cells in slices responded to hypoxia with a surge of catecholamine release that was easily monitored by amperometry. It was noticed earlier that basal secretory activity greatly varies among chromaffin cells in slices (Ortega-Sáenz et al. 2006), and in the current study ∼90% of the cells showed a high level of basal secretion. Although not studied in detail, basal secretion could be related to the firing of spontaneous action potentials by the cells (∼60% of dispersed neonatal cells; see also Fig. 5 below). The exocytotic response to hypoxia was completely abolished by application of Ni²⁺ at concentrations (25 μm) at which the cation selectively blocks T-type Ca²⁺ channels (Fig. 3A and C; see Fig. 1 above). The secretory response to a pulse of high extracellular K⁺, which presumably produces a strong depolarization and opening of HVA channels, was maintained even in the presence of 50 μm Ni²⁺ (Fig. 3A). The inhibitory effect of Ni²⁺ on the response of chromaffin cells to hypoxia was independent of the order of application (Fig. 3A and B). Pimozide (1 μm), another selective blocker of the T-type Ca²⁺ channels (Santi et al. 2002) also inhibited the hypoxia-elicited catecholamine release (Fig. 3D). These results indicate that functional α1H T-type Ca²⁺ channels appear to be necessary for neonatal rat chromaffin cell responsiveness to hypoxia.

Figure 5. Spontaneous electrical activity in neonatal and adult chromaffin cells.

A, spontaneous firing frequency in representative neonatal and adult chromaffin cells. Cells from neonatal rats had tonic firing, but in the adult they presented a typical burst-firing pattern. B, example of action current recorded from a chromaffin cell with a cell-attached patch electrode. C, summary of spontaneous average firing frequency (action potentials/min) of dispersed neonatal (n= 14) and adult (n= 32) cells. *Statistically significant, P < 0.05.

Figure 3. Amperometric recordings of catecholamine secretion in slices of neonatal adrenal glands.

A and B, secretory response to hypoxia and reversible blockade by application of Ni²⁺ (25 μm). The effect of Ni²⁺ is independent of the order of application. Note in A that even a larger concentration of Ni²⁺ (50 μm) does not abolish the strong secretory burst induced by 40 mm K⁺. C, quantitative summary of the effect of Ni²⁺ on secretion rate (in picocoulombs/min) in neonatal chomaffin cells. n= 10 measurements for control vs hypoxia and n= 6 measurements for hypoxia vs Ni²⁺ plus hypoxia. *Statistically significant, P < 0.05. D, reversible inhibition of the response to hypoxia by pimozide (1 μm).

Developmental loss of T-type Ca²⁺ channel expression in chromaffin cells

The participation of T-type Ca²⁺ channels in the secretory response to low O₂ tension in neonatal chromaffin cells suggested that the developmental loss of this channel type could help explain the lack of hypoxia responsiveness of cells in the adult adrenal gland. To test this hypothesis we carried out electrophysiological experiments in which the expression of functional LVA and HVA Ca²⁺ channels in chromaffin cells from neonatal (<P5) and young adult (6–15 weeks) rats was quantified. These experiments were performed within 4 h of cell dispersion to avoid changes in channel synthesis during cell culture. Ca²⁺ channel current density was estimated by measuring the amplitude of the fast (HVA channels) and slow (LVA channels) components of tail currents recorded after depolarizing pulses to 5 and 50 ms. After 5 ms depolarizations, fitting the Ca²⁺ tail currents of adult cells at −70 mV yielded two components with similar kinetics as those measured in neonatal cells (fast time constant 0.14 ± 0.01 ms and slow time constant 1.81 ± 0.13 ms; n= 30 cells). However, the relative amplitudes of the two tail components were clearly different in neonatal versus adult cells. The slow component was well identifiable at the end of 5 ms pulses in neonatal cells and, as expected, was not recorded at the end of 50 ms pulses due to inactivation of the T-type current (Fig. 4A). In contrast, adult cells almost lacked the slow component and, as a result, the tail currents recorded at the end of 5 and 50 ms pulses had time courses that were superimposable (Fig. 4B). The plots in Fig. 4C and D are quantitative summaries of the data illustrating that whereas current density of the HVA (fast deactivating) channels was similar in neonatal and adult chromaffin cells, neonatal cells had a much higher density of LVA (slowly deactivating) channels. In fact, LVA channels were practically absent in our population of young adult chromaffin cells. The nature of a small, non-inactivating, slow component of the tail current recorded at the end of 50 ms pulses with similar amplitude in neonatal and adult cells is unknown. These data indicate that functional T-type channels are lost during postnatal ontogenic development in chromaffin cells.

Figure 4. Loss of T-type Ca²⁺ channels in adult chromaffin cells.

A, sodium and calcium currents recorded from neonatal chromaffin cells during short (5 ms) and long (50 ms) depolarizing pulses. Superposition of the tail currents on the right side of the panel illustrates the disappearance of the slow component of the tail generated at the end of 50 ms pulses (dashed line) due to inactivation of the T-type channels. B, similar experimental protocol applied to adult cells demonstrates the absence of slowly deactivating channels. C and D, quantitative summary of the amplitude (current density) of fast (C) and slowly (D) deactivating components of the tail currents in neonatal (n= 31) and adult (n= 30) chromaffin cells. Note the marked decrease of the slow component of the current in adult cells. The additional slow component of the tail currents recorded at the end of 50 ms pulses (with similar amplitude in neonatal and adult cells) cannot be ascribed to inactivating T-type channels. The nature of this current component is unknown. *Statistically significant, P < 0.05.

The electrophysiological observations described above were complemented with molecular data comparing the expression of both α1H mRNA and protein in neonatal and adult cells. Quantitative RT-PCR experiments designed to amplify the α1H subunit mRNA (del Toro et al. 2003), indicated that the level of T-type channel mRNA is 72 ± 4 times (n= 3) higher in the neonatal versus adult adrenal tissue, where the α1H subunit is practically absent. Immunocytochemical experiments with anti-α1H antibodies also demonstrated the lack of specific immunostaining of slices from adult adrenal glands (see below Fig. 7A, left).

Figure 7. Re-expression of α1H subunit and sensitivity to hypoxia in denervated adult chromaffin cells.

A, immunohistochemical analysis demonstrating the absence of the α1H subunit in adult adrenal medulla (left panel) and its reappearance after denervation (right panel). B, amperometric recording illustrating the secretory response to hypoxia of chromaffin cells in adrenal slices from denervated animals. As in neonatal glands (Fig. 3), the cells exhibit a secretory response to hypoxia that is reversibly inhibited by application of Ni²⁺. C, quantitative summary of hypoxia responsiveness in chromaffin cells from denervated animals and blockade by external application of Ni²⁺. n= 4 measurements for control vs hypoxia and n= 4 measurements for hypoxia vs Ni²⁺ plus hypoxia *Statistically significant, P < 0.05.

Decrease of adrenal chromaffin cell excitability with postnatal maturation

Expression of T-type Ca²⁺ channels is known to regulate cell excitability (Perez-Reyes, 2003). Moreover, chromaffin cells in which T-type channels were induced by exposure to chronic hypoxia show lower resting membrane potential values and respond with exocytotic bursts to smaller depolarizations than controls (Carabelli et al. 2007). To evaluate whether our neonatal and adult cell preparations actually had different excitability in resting conditions, we measured ‘action currents’ in the cell-attached configuration of the patch-clamp technique. In these conditions, the cells are undialysed and action potential firing frequency can be monitored without altering the cell's cytosolic constituents (Fenwick et al. 1982; Montoro et al. 1996). Representative spontaneous ‘action current’ recordings of neonatal and adult chromaffin cells are shown in Fig. 5A and B and a summary of the changes of average firing frequency (in action potentials/min) in the two cell types is given in Fig. 5C. The data show that in our experimental conditions the average spontaneous firing frequency of neonatal cells is more than twice the value of adult cells, thus indicating that, consistent with the decrease of T-type Ca²⁺ channel expression, adult chromaffin cells are less excitable than their neonatal counterparts.

Re-expression of T-type Ca²⁺ channels after adult adrenal medulla denervation and recovery of sensitivity to hypoxia

It was previously reported (Seidler & Slotkin, 1986) that responsiveness of chromaffin cells to hypoxia is recovered in adult rats after adrenal sympathectomy. Hence, we hypothesized that this functional ‘rejuvenation’ of the adrenal gland could be associated with the recovery of T-type Ca²⁺ channel expression in chromaffin cells. In accord with this idea, Ca²⁺ currents recorded from chronically denervated cells had a clear slowly deactivating component suggesting the reappearance of a population of T-type Ca²⁺ channels (Fig. 6A). This slow component, with an average time constant of 2.21 ± 0.13 ms appeared in 90% of the cells (n= 20 cells studied from 3 denervated animals). Overall, Ca²⁺ current amplitude decreased in chromaffin cells from denervated animals due to a ∼50% reduction of HVA Ca²⁺ channel density. In contrast, LVA channel density in the same cells was significantly increased (Fig. 6B). In parallel with the electrophysiological experiments, we performed immunohistochemical studies of the α1H subunit in adult and denervated adrenal glands. In fair agreement with the functional data, the α1H subunit was absent in the adult AM but clearly present in denervated glands (Fig. 7A). Therefore, the electrophysiological and immunological analyses indicate that the T-type Ca²⁺ channels reappear in adult adrenal chromaffin cells after denervation.

Figure 6. Reappearance of T-type Ca²⁺ channels in chromaffin cells from denervated adult rats.

A, inward currents recorded from a denervated cell following the same experimental protocol as in Fig. 4A. Note the appearance of a slow component of the tail current at the end of 5 ms pulses. B, quantitative summary of the amplitude (current density) of fast (left) and slowly (right) deactivating components of the tail currents in chromaffin cells from adult (n= 30), denervated (n= 20) and sham-operated (n= 25) animals. *Statistically significant, P < 0.05.

As shown in previous studies (García-Fernández et al. 2007) the secretory response of chromaffin cells to acute hypoxia was lost in adult animals. However, responsiveness to hypoxia was re-established in slices of adrenal glands from denervated animals despite that in this condition HVA channel density decreased to half the control value. Catecholamine release from denervated cells in acute hypoxia was abolished, as in cells from neonatal glands (see Fig. 3), by the application of extracellular Ni²⁺ (Fig. 7B and C).

Discussion

The main findings in this report are: (i) neonatal AM chromaffin cells express T-type Ca²⁺ channels (of α1H or Ca_v3.2 sub-type) and the activity of these channels is necessary for catecholamine release in response to acute hypoxia; (ii) T-type Ca²⁺ channel density decreases with postnatal maturation in parallel with responsiveness to hypoxia; and (iii) in adult chromaffin cells, T-type channels and hypoxia sensitivity reappear after AM denervation. These observations indicate that T-type Ca²⁺ channels are essential for the acute response of chromaffin cells to hypoxia and help explain the disappearance of O₂ sensitivity in adult AM chromaffin cells.

Immunocytochemical, molecular biology and electrophysiological experiments demonstrate the presence of T-type Ca²⁺ channels in neonatal chromaffin cells, and selective blockade of these channels with Ni²⁺ (<50 μm) abolishes responsiveness of the cells to acute hypoxia. The most plausible explanation for these findings is that postnatal down-regulation of T-type Ca²⁺ current decreases chromaffin cell excitability and responsiveness to stimuli. In fact, we have also observed that maturation of chromaffin cells is parallelled by a decrease in the resting firing frequency of the cells and that recruitment of T-type channels after denervation restores the response to hypoxia. These results are in fair agreement with recent data showing that up-regulation of T-type Ca²⁺ channels diminishes the resting membrane potential of adult chromaffin cells and increases their excitability (Carabelli et al. 2007). They also fit well with a previous report showing that exposure of PC12 cells to chronic hypoxia, which induces the expression of T-type Ca²⁺ channels (del Toro et al. 2003), increased the acute responsiveness of the cells (Taylor & Peers, 1999). Besides a generalized increase in excitability, another important consequence of T-type channel recruitment in neonatal and denervated adult chromaffin cells is the increase of ‘low-threshold’ exocytosis. As described by Carbone and co-workers (Carbone et al. 2006; Giancippoli et al. 2006) this form of catecholamine release is mediated by Ca²⁺ influx through LVA, T-type, channels thus favouring secretory responses to mild stimulations. This could be the reason why Ni²⁺ blocks responsiveness to hypoxia, a stimulus that induces a subtle activation of the cells, without abolishing the robust secretory burst induced by 40 mm K⁺ (a stimulus that elicits a large depolarization and the opening of HVA Ca²⁺ channels). A similar rationale explains why T-type channel up-regulation re-establishes O₂ sensitivity in adult denervated cells despite the fact that they have a smaller density of HVA Ca²⁺ channels. In addition to the important role of T-type channels in hypoxia-evoked exocytosis, it is well established that a full response to hypoxia requires the presence of the various types of HVA Ca²⁺ channels existing in chromaffin cells (Adams et al. 1996; Mochizuki-Oda et al. 1997; García-Fernández et al. 2007).

The observations in our study also have implications for the mechanisms of O₂ sensing in AM chromaffin cells. It is well established that in neonatal chromaffin cells low O₂ tension produces a decrease of membrane conductance and that the transduction of the hypoxic stimulus is mediated by inhibition of one or several classes of K⁺ channels (see, for review, Nurse et al. 2006). However, modulation of T-type channels by hypoxia, or hypoxia-induced redox intermediates, could also facilitate O₂ sensitivity of chromaffin cells. Indeed, T-type currents in pacemaker neurons of ventrolateral medulla are enhanced upon exposure to acute hypoxia (Sun & Reis, 1994). On the other hand, although hypoxia slightly inhibits recombinant α1H channels stably expressed in HEK293 cells (Fearon et al. 2000) both native neuronal and recombinant α1H channels are selectively potentiated by reducing agents such as GSH, DTT or l-cysteine (Fearon et al. 2000; Joksovic et al. 2006). Therefore, hypoxia sensitivity might be an intrinsic property of AM chromaffin cells present throughout life that is only manifested in those cells with the proper density of T-type channels that can open near the resting membrane potential and thus are able to convert small depolarizations into the Ca²⁺ influx required for secretion (see Carbone et al. 2006). A decrease of cell excitability due to T-type channel down-regulation might also explain the diminished responsiveness of adult chromaffin cells to hypercapnia (Muñoz-Cabello et al. 2005).

The mechanisms underlying the postnatal changes in chromaffin cell T-type channel expression or the appearance of the channels after denervation are surely related with the well-known modulation of these channels in stress conditions. T-channel recruitment occurs in smooth muscle after vascular damage (Schmitt et al. 1995), in cardiac myocytes during hyperthrophy and heart failure (Nuss & Houser, 1993; Sen & Smith, 1994) and in lesioned neurons (Chung et al. 1993; Jagodic et al. 2008) or epilepsy (Tsakiridou et al. 1995), as well as in proliferating tumour cells (Mariot et al. 2002), and during the terminal differentiation of myoblasts (Bijlenga et al. 2000). In the context of the current paper it is particularly interesting that the α1H subunit is up-regulated by chronic hypoxia both in PC12 (del Toro et al. 2003) and adult chromaffin (Carabelli et al. 2007) cells in a hypoxia inducible factor 2α-dependent manner. In agreement with this observation, the 5′ flanking region of the α1H gene contains several sequences compatible with hypoxia-responsive elements that could regulate transcription (del Toro et al. 2003). Therefore, it is highly plausible that the relatively low O₂ tension of fetal blood induces T-type channel expression, with the level of functional T-type channels persisting in the neonate until a new homeostatic equilibrium is reached. T-type currents are expressed in more than 50% of fetal rat chromaffin cells (Bournaud et al. 2001) and these cells are also hypoxia sensitive (Bournaud et al. 2007; see also Rychkov et al. 1998). Besides higher blood oxygenation in the newborn, which inhibits the hypoxia-dependent α1H gene induction, adrenal innervation by cholinergic fibres of the splanchnic nerve seems to critically contribute to T-channel down-regulation since the channels disappear when the gland is innervated (∼P7 in the rat) and reappear after denervation. T-channel down-regulation in innervated cells could be a consequence of increased cytosolic Ca²⁺ concentration after cholinergic receptor(s) activation, as it is known that Ca²⁺ influx decreases α1H gene expression (see del Toro et al. 2003). It is also plausible that cholinergic activation of the chromaffin cells alters signal transduction cascades resulting in α1H gene repression. In fact, T-type channels are, as stated above, highly susceptible to modulation in numerous stressful conditions and in chromaffin cells they are up-regulated by β-adrenergic activation and intracellular cAMP (Novara et al. 2004; Giancippoli et al. 2006).

The observations in this paper have not only physiological relevance but they might have also pathophysiological interest. Abnormal T-type channel down-regulation in AM chromaffin cells might underlie some diseases in the newborn, such as sudden infant death syndrome (SIDS), a disorder associated with decreased chemoreceptor excitability (see references in Slotkin, 1998; López-Barneo et al. 2008). In this regard, epidemiological studies have pointed to cigarette smoking as one of the major risk factors for SIDS (Haglund & Cnattingius, 1990). Maternal exposure to nicotine favours fetal death (Slotkin et al. 1995; Slotkin, 1998) and decreases the sensitivity of AM cells to hypoxia (Buttigieg et al. 2008). Nicotine is a cholinergic agonist that once in fetal blood might act as a false stimulus of early AM innervation. Therefore, nicotine-induced abolishment of AM responsiveness to hypoxia could be explained, among other factors, by a decrease of T-type channels expression in neonatal chromaffin cells. In contrast, excessive T-type channel up-regulation could contribute to the high sympathetic tone and cardiovascular alterations seen in subjects exposed to chronic intermittent hypoxia (see Fletcher, 2001). These hypotheses concerning the pathophysiological implications of α1H T-type channels should be tested in future experimental work.