Impaired chromaffin cell excitability and exocytosis in autistic Timothy syndrome TS2‐neo mouse rescued by L‐type calcium channel blockers

Abstract

Key points

• Tymothy syndrome (TS) is a multisystem disorder featuring cardiac arrhythmias, autism and adrenal gland dysfunction that originates from a de novo point mutation in the gene encoding the Cav1.2 (CACNA1C) L‐type channel.

• To study the role of Cav1.2 channel signals in autism, the autistic TS2‐neo mouse has been generated bearing the G406R point‐mutation associated with TS type‐2.

• Using heterozygous TS2‐neo mice, we report that the G406R mutation reduces the rate of inactivation and shifts leftward the activation and inactivation of L‐type channels, causing marked increase of resting Ca²⁺ influx (‘window’ Ca²⁺ current).

• The increased ‘window current’ causes marked reduction of Na_V channel density, switches normal tonic firing to abnormal burst firing, reduces mitochondrial metabolism, induces cell swelling and decreases catecholamine release.

• Overnight incubations with nifedipine rescue Na_V channel density, normal firing and the quantity of catecholamine released. We provide evidence that chromaffin cell malfunction derives from altered Cav1.2 channel gating.

Abstract

L‐type voltage‐gated calcium (Cav1) channels have a key role in long‐term synaptic plasticity, sensory transduction, muscle contraction and hormone release. A point mutation in the gene encoding Cav1.2 (CACNA1C) causes Tymothy syndrome (TS), a multisystem disorder featuring cardiac arrhythmias, autism spectrum disorder (ASD) and adrenal gland dysfunction. In the more severe type‐2 form (TS2), the missense mutation G406R is on exon 8 coding for the IS6‐helix of the Cav1.2 channel. The mutation causes reduced inactivation and induces autism. How this occurs and how Cav1.2 gating‐changes alter cell excitability, neuronal firing and hormone release on a molecular basis is still largely unknown. Here, using the TS2‐neo mouse model of TS we show that the G406R mutation altered excitability and reduced secretory activity in adrenal chromaffin cells (CCs). Specifically, the TS2 mutation reduced the rate of voltage‐dependent inactivation and shifted leftward the activation and steady‐state inactivation of L‐type channels. This markedly increased the resting ‘window’ Ca²⁺ current that caused an increased percentage of CCs undergoing abnormal action potential (AP) burst firing, cell swelling, reduced mitochondrial metabolism and decreased catecholamine release. The increased ‘window’ Ca²⁺ current caused also decreased Na_V channel density and increased steady‐state inactivation, which contributed to the increased abnormal burst firing. Overnight incubation with the L‐type channel blocker nifedipine rescued the normal AP firing of CCs, the density of functioning Na_V channels and their steady‐state inactivation. We provide evidence that CC malfunction derives from the altered Cav1.2 channel gating and that dihydropyridines are potential therapeutics for ASD.

Key points

• Tymothy syndrome (TS) is a multisystem disorder featuring cardiac arrhythmias, autism and adrenal gland dysfunction that originates from a de novo point mutation in the gene encoding the Cav1.2 (CACNA1C) L‐type channel.

• To study the role of Cav1.2 channel signals in autism, the autistic TS2‐neo mouse has been generated bearing the G406R point‐mutation associated with TS type‐2.

• Using heterozygous TS2‐neo mice, we report that the G406R mutation reduces the rate of inactivation and shifts leftward the activation and inactivation of L‐type channels, causing marked increase of resting Ca²⁺ influx (‘window’ Ca²⁺ current).

• The increased ‘window current’ causes marked reduction of Na_V channel density, switches normal tonic firing to abnormal burst firing, reduces mitochondrial metabolism, induces cell swelling and decreases catecholamine release.

• Overnight incubations with nifedipine rescue Na_V channel density, normal firing and the quantity of catecholamine released. We provide evidence that chromaffin cell malfunction derives from altered Cav1.2 channel gating.

Introduction

Timothy syndrome (TS) is a rare multiorgan channelopathy characterized by cardiac arrhythmias, long QTs, immune deficiencies and autism (Splawski et al. 2004, 2005). It is associated with a de novo single point mutation in the pore‐forming subunit of Ca_V1.2 L‐type Ca²⁺ channels (CACNA1C) and occurs in two major forms (TS1 and TS2), depending on whether the point mutation appears on exon 8a (Splawski et al. 2004) or exon 8 (Splawski et al. 2005). The two exons code for the IS6 helix of the Ca_V1.2 channel controlling the voltage dependence of activation and inactivation. In the TS2 type, the mutation occurs either at Gly406 (G406R) or at Gly402 (G402S) within exon 8, which is highly expressed in brain and heart (80%) and to a lesser extent in the adrenal glands (Splawski et al. 2005). Both mutations cause reduced channel inactivation and increased Ca²⁺ influx during cell activity. Curiously, TS2 patients present autistic forms of behaviour if they carry the G406R mutation but are neurologically intact if they exhibit the G402S mutation (Splawski et al. 2005; Frohler et al. 2014; Hiippala et al. 2015). Recent findings suggest that the ability of the G406R vs. G402S mutation in inducing neurological abnormalities (Li et al. 2016) is linked to the opposing effects of G406R and G402S mutations on Ca_V1.2 channel activation. The mutation G406R shifts the voltage dependence of activation toward more negative potentials, whereas G402R shifts the activation in the opposite direction (Raybaud et al. 2006; Dick et al. 2016). A marked negative shift of voltage‐dependent activation is also characteristic of the point mutations of the Ca_V1.3 α‐subunit (A749G, V401L) identified in patients with autism spectrum disorder (ASD) (Pinggera et al. 2015; Pinggera et al. 2017).

TS is one of the most penetrant monogenic forms of autism and is widely studied using either cell transfection systems (Splawski et al. 2004; Erxleben et al. 2006; Barrett & Tsien, 2008; Yarotskyy et al. 2008; Bidaud & Lory, 2011; Dick et al. 2016) or somatic cell reprogramming to generate induced pluripotent stem cells (iPSCs) from patients with TS (Pasca et al. 2011; Krey et al. 2013). An alternative approach is now provided by the availability of the TS2‐neo mouse, which exhibits behavioural traits reminiscent of ASD (Bader et al. 2011; Bett et al. 2012; Barnabei et al. 2014). The TS2‐neo mice have largely normal brain size and structure (Bader et al. 2011). Their cortical neurons, however, exhibit the same activity‐dependent dendritic retraction observed in rat cortical neurons transfected with Cav1.2 G406R‐mutated channels and in human iPSC‐derived neurons from individuals with TS (Krey et al. 2013). The TS2‐neo mouse also displays altered brain activities that are common across several ASD mouse models (Ehlinger & Commons, 2017), and thus appears to be an interesting animal model for studying the origins of neuronal firing and Ca²⁺ signal mistuning that generate ASD.

Adrenal chromaffin cells (CCs) express high densities of Cav1.2 and Cav1.3 channels (Baldelli et al. 2004; Marcantoni et al. 2007, 2010) and are thus an excellent cell model to study the role of Cav1.2 channels on action potential (AP) firing and catecholamine secretion (Carabelli et al. 2003; Lingle et al. 2018). Here, we studied how the TS2‐neo mutation alters mouse chromaffin cell (MCC) function. Our interest was further boosted by a recent case report of a 2‐month‐old TS patient whose post‐mortem autopsy revealed severe bilateral adrenal gland dystrophy, most likely caused by increased intracellular Ca²⁺ associated with the Cav1.2 channel mutation, which occurred mainly in the medulla (Kawaida et al. 2016).

We found that TS2‐neo MCCs possess L‐type currents similar to the G406R‐mutated phenotype described in transfected HEK cells, having slower inactivation, leftward shifted voltage‐dependent activation and inactivation and large ‘window’ current at rest. The latter is likely the cause of the increased resting Ca²⁺ that switches AP firing from tonic to burst, alters cell morphology and reduces mitochondrial metabolism and catecholamine secretion. High‐resolution electron microscopy in intact adrenal glands of TS2‐neo mice reveals extended cytoplasmic vacuolization and cell swelling, which partly explains the reduced density of Na_V channels responsible for the altered AP firing. Overnight incubation of TS2‐neo MCCs with submicromolar concentrations of the L‐type channel blocker nifedipine (0.3 μM) effectively rescues the Nav channel density, the spontaneous firing and the secretory activity of TS2‐mutated MCCs. Control tests indicated that chronic applications of 0.3 μM nifedipine are not per se beneficial on WT‐MCCs. On the contrary, the dihydropyridine (DHP) suppresses the resting spontaneous firing of WT cells. Our data provide evidence that the TS2‐neo mouse is a valid murine model for studying the effects of the Cav1.2 G406R mutation on AP firing and Ca²⁺ signals associated with ASD.

Methods

Ethical approval

Ethical approval was obtained for all experimental protocols from the University of Torino Animal Care and Use Committee, Torino, Italy. All experiments were conducted in accordance with the National Guide for the Care and Use of Laboratory Animals adopted by the Italian Ministry of Health. All animals had free access from the shelter to water and food. Every effort was made to minimize animal suffering and the number of animals used. For removal of tissues, animals were deeply anaesthetized with CO₂ inhalation at fixed concentration and rapidly killed by cervical dislocation. TS2‐neo mice were designed and produced by G. C. Bett and R. L. Rasmusson at the Department of Physiology & Biophysics, State University of New York, Buffalo (NY, USA) as detailed in Bader et al. (2011) and Bett et al. (2012).

Cell culture

Chromaffin cells were obtained from male C57BL/6J mice (Envigo, San Pietro al Natisone, Italy) of 2 months. Under sterile conditions the abdomen was opened, and the adrenal glands were isolated and transferred to an ice‐cold Ca²⁺‐ and Mg²⁺‐free Locke's buffer containing (in mM) 154 NaCl, 3.6 KCl, 5.6 NaHCO₃, 5.6 glucose and 10 Hepes, pH 7.4 (Marcantoni et al. 2009; Vandael et al. 2012). Under a dissecting microscope the adrenal glands were decapsulated and subsequently subjected to an enzymatic dissociation with 20–25 units/ml papain (Worthington Biochemical Corp., Segrate, Italy) dissolved in Dulbecco's modified Eagle's medium (DMEM; Gibco, Invitrogen/Life Technologies, Monza, Italy) supplemented with 1.5 mM of l‐cysteine, 1 mM of CaCl₂ and 0.5 mM of EDTA (Sigma‐Aldrich, Munich, Germany) for 25–30 min at 37°C in a water‐saturated atmosphere with 5% CO₂. Afterwards, two washing steps were performed with DMEM supplemented with 1 mM CaCl₂ and 10 mg/ml of BSA (Sigma‐Aldrich). Adrenal medullas were resuspended in DMEM containing 1% penicillin/streptomycin and 15% fetal bovine serum (both from Sigma‐Aldrich) and were mechanically dissociated with a fire polished Pasteur pipette. A drop (100 μl) of this concentrated cell suspension was plated on poly‐ornithine (1 mg/ml)‐ and laminin (5 μg/ml)‐coated Petri dishes and subsequently (30 min later) 1.9 ml of DMEM containing 1% penicillin/streptomycin and 15% fetal bovine serum (all from Sigma‐Aldrich) was added. The primary chromaffin cell cultures were kept in an incubator at 37°C at water‐saturated atmosphere with 5% CO₂. Measurements were performed on cultured MCCs 2–5 days after plating.

Mouse genotyping

Genomic DNA extraction from mouse tail tissue was performed by incubating samples in 0.1 mg/ml proteinase K at 65°C for 3 h. Samples were placed at 95°C for 10 min, centrifuged for 5 min at maximum speed, and 1 μl of the supernatant was used for PCR amplification. Primers specific for Cav1.2 (sense: 5′‐CCT CCA CTT TGC TTG TTC‐3′; anti‐sense: 5′‐GGC TCC TGA GTG ACC CT‐3′) were used to amplify a 667 bp fragment. Primer specific for the neomycin gene (5′‐GCT CCA GAC TGC CTT GGG AA‐3′) was included in the reaction to amplify 400 bp specific to the transgene. PCR conditions were as follows: denaturation at 94°C for 5 min followed by 35 cycles of 94°C for 30 s, annealing at 60°C for 30 s and extension at 72°C for 1 min, followed by a final extension at 72°C for 5 min. PCR products were separated by agarose gel electrophoresis using 1% agarose and then visualized by CYBR Safe DNA gel staining for typing analysis; 1 kb Plus DNA Ladder (Invitrogen) was used.

RNA isolation from tissue and cDNA synthesis

Adrenal glands were removed and transferred to an ice‐cold Ca²⁺‐ and Mg²⁺‐free Locke's buffer containing (in mM): 154 NaCl, 3.6 KCl, 5.6 NaHCO₃, 5.6 glucose and 10 Hepes, pH 7.4. The glands were decapsulated and medullae were separated from the cortical tissue. Isolated medullae were immediately transferred into RNAlater RNA Stabilization Reagent (Qiagen, GmbH, Hilden, Germany). Purification of total RNA from adrenal medullae was implemented using Qiagen RNeasy mini kit according to manufacturer's instructions. The RNA concentration was measured photometrically yielding approximately 40 ng/μl RNA; 13 μl of total RNA was reverse transcribed using Maxima H Minus First Strand cDNA Synthesis kit with random hexamer primers (Thermo Fisher Scientific, Waltham, MA, USA).

Quantitative real‐time PCR

Custom TaqMan gene expression assays (Table 1) (Thermo Fisher Scientific, Waltham, MA, USA) were designed to span exon‐exon boundaries. The expression of Cav1.2 exon 8 and exon 8a was assessed using a standard curve method based on PCR fragments of known concentration (Schlick et al. 2010). Cav1.2 cDNA fragments were amplified from mouse whole brain or adrenal medulla using primers spanning exons 7–9 and specific for either exon 8 or exon 8a (Table 2).

Table 1.

Custom TaqMan gene expression assay for Cav1.2 exons 8 and 8a splice variants

Subunit variant	GenBank number	Assay ID	Exon boundary	Sequences for custom assay
Cav1.2 ex8	NM_001255999.2	APU63XM	7–8	Fwd: 5′‐GGCTGGACAGACGTGCTGTA‐3′ Rev: 5′‐CACCCAGGGCAACTCATAGC‐3′ Probe: 5′‐TGGATGCAAGACGCTAT‐3′
Cav1.2 ex8a	NM_001256000.2	APWCXHJ	7–8	Fwd: 5′GGCTGGACAGACGTGCTGTA‐3′ Rev: 5′‐ CAGGGCCAGTCCCTTCCT ‐3 Probe: 5′‐TGGGTCAATGATGCCG‐3′

Table 2.

cDNA specific primer sequences for standard template cloning

Primer name	Primer sequence
1.2 ex8/8a std fwd_SalI	5′‐ATATGTCGACAGTGTCAGAACGGGACCGTG‐3′
1.2 ex8/8a std rev_BglII	5′‐CTTTGAGATCTTCTTCTAGTTGCTGC‐3′
1.2 ex8 fusion fwd	5′‐GTCAGTCTGGTCATCTTTGGATCCT‐3′
1.2 ex8 fusion rev	5′‐AGGATCCAAAGATGACCAGACTGAC‐3′
1.2 ex8a fusion fwd	5′‐GTAACACTAATCATCATAGGGTCAT‐3′
1.2 ex8a fusion rev	5′‐ATGACCCTATGATGATTAGTGTTAC‐3′

The fragments were subsequently cloned into the pGFP37 vector. In order to generate DNA templates of known concentrations for quantitative real‐time PCR (qRT‐PCR) standard curves, the concentration of the digested fragments was determined using the Quant‐IT PicoGreen dsDNA Assay Kit (Invitrogen, Carlsbad, CA, USA). Subsequently, standard curves were generated using a serial dilution ranging from 10⁷ to 10¹ DNA molecules in water containing 1 μg/ml of poly‐dC‐DNA (Midland Certified Reagent Company Inc., Midland, TX, USA). qRT‐PCRs of standard curves and samples were performed as described previously (Schlick et al. 2010). Samples for qRT‐PCR quantification (50 cycles) contained 20 ng of total RNA equivalent of cDNA, the respective TaqMan gene expression assay, and TaqMan Universal PCR Master Mix (Thermo Fisher Scientific). Specificity of the custom designed assays recognizing either exon 8 or 8a was confirmed using different DNA ratios of corresponding and mismatched Cav1.2 DNA fragments. Importantly, both assays recognized only the corresponding fragment even in the presence of a 10‐fold higher concentration of the other splice variant fragment (Fig. 1 A). qRT‐PCR was performed in duplicates from at least four independent RNA preparations from two (WT) and three (TS2) biological replicates. Samples without template and RNA samples without reverse transcription served as negative controls. The expression of seven different endogenous control genes (Actb, B2m, Gapdh, Hprt1, Tbp, Tfrc and Sdha; Schlick et al. 2010) was routinely measured and used for data normalization as previously described (Vandesompele et al. 2002) (Fig. 1 B). Briefly, data were normalized to the two most stable endogenous control genes (Gapdh and Actb) and normalized molecule numbers were calculated for each assay based on their respective standard curve. Analyses were performed using the 7500 Fast System (Thermo Fisher Scientific). Statistical analysis was performed using one‐way ANOVA followed by the Holm–Sidak post hoc test.

Figure 1.

Specificity of Cav1.2 Taqman assays and expression stability of endogenous control genes

A, binding specificity of Cav1.2 exon 8 and exon 8a assays. Exon 8 and exon 8a were recognized with a high specificity (low C _T value) by the corresponding assay even in the presence of a 10‐fold higher concentration of the mismatching DNA fragments from exon 8a and 8, respectively. Furthermore, both assays did not recognize the similar exons 8a and 8b from Cav1.3 L‐type channels. High binding specificity was confirmed by inefficient detection (high C _T value) of all non‐matching DNA fragments. B, average expression stability of endogenous control genes in MCCs. All reference genes maintained comparable cDNA concentrations throughout the experiments. Data were normalized to the most stable endogenous control genes listed at the top (Gapdh and Actb).

Action potential and ion current recordings

Macroscopic whole‐cell currents and APs were recorded in perforated‐patch conditions using a multiclamp 700‐B amplifier and pCLAMP 10.0 software (Molecular Devices, Sunnyvale, CA, USA) (Marcantoni et al. 2010; Vandael et al. 2012). Traces were sampled at 10 kHz using a Digidata 1440A acquisition interface (Molecular Devices) and filtered using a low‐pass Bessel filter set at 1–2 kHz. Borosilicate glass pipettes (Kimble Chase Life Science, Vineland, NJ, USA) with a resistance of 2–3 MΩ were dipped in an Eppendorf tube containing intracellular solution before being back‐filled with the same solution containing 500 μg/ml of amphotericin B (Sigma‐Aldrich), dissolved in dimethyl sulfoxide (Sigma‐Aldrich) (Cesetti et al. 2003). Recordings were initiated after amphotericin B lowered the access resistance below 15 MΩ (5–10 min). Series resistance was compensated by 60–80% and monitored throughout the experiment. Fast capacitive transients during step‐wise depolarizations (in voltage‐clamp mode) were minimized online by the use of the patch clamp analog compensation. Uncompensated capacitive currents were further reduced by subtracting the averaged currents in response to P/4 hyperpolarizing pulses. Off‐line data analysis was performed with pCLAMP 10.0 software.

The normalized voltage‐dependent conductance of Cav channels (g _Ca), was calculated with the equation: g _Ca  = I _Capeak/(V − E _Ca), with E _Ca equal to the reversal potential for Ca²⁺, and fitted with a Boltzmann function with variable V _1/2 (in mV) and k slope (in mV) (Guarina et al. 2017). The same was done for g _Na and g _BK.

Solutions

Intracellular solution for current‐clamp and Na⁺ and K⁺ current measurements in voltage clamp or AP‐clamp mode was composed of (in mM) 135 potassium aspartate, 8 NaCl, 2 MgCl₂, 5 EGTA, 20 Hepes, pH 7.4 (with KOH; Sigma‐Aldrich). For Ca²⁺ current recordings the intracellular solution contained (in mM) 135 Cs‐MeSO₃, 8 NaCl, 2 MgCl₂, 5 EGTA and 20 Hepes, pH_o 7,4 (with CsOH; Sigma‐Aldrich). The extracellular solution used for current‐clamp measurements is a physiological Tyrode solution containing (in mM): 130 NaCl, 4 KCl, 2 CaCl₂, 2 MgCl₂, 10 glucose and 10 Hepes; pH_o 7.4 (with NaOH; Sigma‐Aldrich). The same solution was used to measure K⁺ currents. K_V currents were obtained by adding 1 μM paxilline to the external solution while Ca²⁺‐activated BK currents were estimated by subtracting K_V from the total K⁺ currents. The extracellular solution used for Na⁺ current measurements was composed of (in mM): 104 NaCl, 30 TEACl, 4 KCl, 2 CaCl₂, 2 MgCl₂, 10 glucose and 10 Hepes, pH_o 7.4 (with NaOH). The extracellular solution used for Ca²⁺ and Ba²⁺ current measurements in voltage‐clamp configuration contained (in mM): 135 TEACl, 10 CaCl₂ (or 10 BaCl₂), 2 MgCl₂, 10 glucose and 10 Hepes, pH_o 7.4 (with TEA‐OH; Sigma‐Aldrich). Nifedipine and verapamil were obtained from Sigma‐Aldrich. The two drugs were dissolved, stored and used as previously described (Marcantoni et al. 2010).

Membrane capacitance changes to monitor vesicle secretion

Ca²⁺ currents and the corresponding depolarization‐evoked capacitance changes were measured in the perforated‐patch configuration using an EPC‐10 double patch amplifier (HEKA Elektronic, Lambrecht, Germany) (Carabelli et al. 2003, 2007; Marcantoni et al. 2009). Mouse chromaffin cells were kept in saline solution, containing (in mM): 4 TEACl, 126 NaCl, 10 CaCl₂, 4 KCl, 2 MgCl₂, 10 glucose, 10 Hepes (pH 7.4 with NaOH). The internal solution contained (in mM): 135 CsMeSO₃, 8 NaCl, 2 MgCl₂, 20 Hepes (pH 7.3 with CsOH) plus amphotericin B (Sigma‐Aldrich). Ca²⁺ currents were evoked by applying step depolarizations (100 ms duration), from a holding potential of −70 mV to +10 mV. Superimposed on the square pulse, a sinusoidal wave function (1000 Hz, ±25 mV) was applied in order to monitor membrane capacitance increases following the depolarizing step using the Lockin extension of the Patchmaster software (HEKA Elektronic, Lambrecht, Germany). Fast capacitive transients due to depolarizing pulses were minimized online by the patch‐clamp analog compensation. Series resistance was compensated by 80% and monitored during the experiment. The amount of Ca²⁺ entering the cells during a depolarization (quantity of charge in pC) was calculated as the time integral of the Ca²⁺ current. All the experiments were performed at room temperature. Ca²⁺ currents were sampled at 10 kHz and low‐pass filtered at 2 kHz.

Transition electron microscopy imaging

Adrenal glands were freshly dissected from male C57BL/6J (n = 3) and TS2‐neo mice (n = 3) and fixed by immersion in 4% formaldehyde + 2% glutaraldehyde in phosphate buffer (PB, 0.1 M, pH 7.4) overnight at 4°C. After washing in PB, the glands were post‐fixed in osmium ferrocyanide (1 volume of 2% aqueous osmium tetroxide: 1 volume of 3% potassium ferrocyanide) for 1 h at 4°C, dehydrated for 15 min in increasing concentrations of acetone (30%, 60%, 90%, 100%), and then embedded in Spurr resin in 0.5 ml Eppendorf tubes. Ultrathin sections were cut with an ultramicrotome (EM UC6, Leica Mikrosysteme Vertrieb GmbH, Wetzlar, Germany), collected on uncoated nickel grids (200 mesh) and counterstained with lead citrate (10 min) (Salio et al. 2017). Grids were observed with a JEM‐1010 transmission electron microscope (JEOL, Tokyo, Japan) equipped with a side‐mounted CCD camera (Mega View III, Olympus Soft Imaging System, GmbH, Muenster, Germany). To determine the area of the cell profiles visible in single sections, 50 MCCs (×6000 magnification) from both C57BL/6J and TS2‐neo mice were randomly selected. The surface of sectioned cells was measured by using the analySIS software (Olympus Soft Imaging System). The number of mitochondria/area (n/μm²) and large dense‐core granules/area (n/μm²) were calculated in the same MCCs by using the count particle function of ImageJ software (NIH, Bethesda, MD, USA) and results were expressed as the mean ± SEM.

Mitochondrial respiration and ATP synthesis

The electron transport between complexes I and III was measured in mitochondrial extracts as described earlier (Campia et al. 2009). The results were expressed as nmol reduced cytochrome c/min/mg mitochondrial proteins. The amount of ATP in mitochondria was measured with the ATP Bioluminescent Assay Kit (FL‐AA, Sigma‐Aldrich), using a Synergy HT Multi‐Mode Microplate Reader (BioTek, Bad Friedrichshall, Germany). ATP was quantified as relative light units (RLU); data were converted into nmol ATP/mg mitochondrial proteins.

Mitochondrial electric potential measurement

The mitochondrial electric potential (Δψ) was measured using the JC‐1 staining method (Riganti et al. 2015); 1 × 10⁶ cells, resuspended in 0.5 mL phosphate‐buffered saline (PBS), were incubated for 30 min at 37°C with 2 μmol/l of the fluorescent probe JC‐1 (Biotium Inc., Hayward, CA, USA), then centrifuged at 13,000 g for 5 min and resuspended in 0.5 mL PBS. The fluorescence of each sample was read using a Synergy HT Multi‐Mode Microplate Reader (BioTek): the red fluorescence, an index of polarized mitochondria, was detected at 550 nm (λ excitation) and 600 nm (λ emission); the green fluorescence, an index of depolarized and damaged mitochondria, was detected at 485 nm (λ excitation) and 535 nm (λ emission). The fluorescence units were used to calculate the percentage of green‐fluorescent (i.e. depolarized/damaged mitochondria) vs. red‐fluorescent mitochondria (i.e. polarized mitochondria).

DAPI and TUNEL immunofluorescence assays

Adrenal medulla glands were dissected and fixed for 12–16 h at 4°C, equilibrated in 30% sucrose in PBS for 12–24 h at 4°C, embedded with Tissue‐TEK (O.C.T, Sakura Finetek, Alphen aan den Rijn, The Netherlands), frozen in liquid nitrogen and stored at −20°C as already detailed (Sgro et al. 2016; Bianchi et al. 2017). Sectioning was then performed with a cryostat (20 μm). For staining, cryo‐sections were subjected to antigen retrieval using 10 mM sodium citrate pH 6.5, 1% glycerol in PBS. Sections were then permeabilized with 0.3% Triton X‐100 in PBS for 30 min (min) and quenched with 0.1 M glycine for 30 min. Apoptosis was analysed by terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) assay using ‘In Situ cell death detection kit, TMD red’ (Roche, Basel; Switzerland) according to manufacturer's protocol. Following TUNEL staining, sections were counterstained with 4′,6‐diamidino‐2‐phenylindole (DAPI) for 30 min at room temperature and sections were mounted using ProLong anti‐fade reagent (Thermo Fisher Scientific).

Microscopy and image analysis

Imaging was performed using a Leica TCS SP5‐AOBS 5‐channel confocal system (Leica Microsystems) equipped with a 405 nm diode, an argon ion, a 561 nm diode‐pumped solid‐state and a He–Ne laser. Fixed slices were imaged using a PL APO x40/1.2 NA oil immersion objective (Bianchi et al. 2017). All the images were analysed by using Fiji software.

Statistics

Data are given as mean ± SEM for n number of cells. Statistical significance was estimated using either paired or unpaired Student's t test or one‐way ANOVA followed by a Holm–Sidak post hoc test when two or multiple groups of measurements had to be compared. Data were found statistically significant when P ≤ 0.05. Unpaired Student's t test was used also to estimate the statistical significance of the slope of the regression lines calculated with Origin 8.5 software (OriginLab Corp., Northampton, MA, USA) as best fit through data points. Origin 8.5 was also used for statistical and data analysis.

Results

qPCR analysis of Cav1.2 exon 8 and exon 8a expression in MCCs

Cav1.2 and Cav1.3 are the only L‐type channels effectively expressed in MCCs (Baldelli et al. 2004) with Cav1.2 contributing to nearly half of the L‐type current (Marcantoni et al. 2010). Nevertheless, it has never been tested how much of the Cav1.2 exon 8 or exon 8a splice variant is expressed in the MCCs of the adrenal medulla. Cav1.2 exons 8 and 8a are mutually exclusive exons, with high sequence similarity, encoding for the IS6 transmembrane domain (Fig. 2 A and B). To identify the amount of exon 8 and 8a splice variants expressed in the adrenal medulla of WT and heterozygote TS2‐neo mice, we performed a standard curve‐based absolute qRT‐PCR analysis (Schlick et al. 2010). Figure 2 C shows that WT‐MCCs express exon 8 and exon 8a in a ratio of 1:5 (P < 0.001), while the heterozygote TS2‐neo MCCs express the two exons in a ratio of 1:1. The ∼50% reduced expression of exon 8a in TS2‐neo compared to WT‐MCCs (P < 0.001) is attributed to the neo‐cassette that introduces a stop codon in exon 8a of the mutated allele, thus likely rendering the truncated mRNA sensitive to nonsense‐mediated degradation. Interestingly, the total amount of Cav1.2 mRNA was similar in heterozygote TS2 and WT‐MCCs, confirming a previously suggested upregulation of exon 8 as a consequence of the reduced expression of exon 8a (P < 0.001) (Bader et al. 2011). Our exon 8 qRT‐PCR assay does not discriminate between WT and TS2‐mutated allele and therefore we cannot exactly quantify the total amounts of WT and TS2‐mutated exon 8. Given the unpredicted upregulation of exon 8 in TS2‐neo adrenal medulla, it seemed reasonable to expect that the expression ratio of mutated vs. WT allele may vary between 1:1 and 1:5.

Figure 2.

Cav1.2 exon 8 and exon 8a expression in mouse adrenal medulla

A, topology of Cav1.2 α₁‐subunit showing the domain encoded by the alternative exon 8 or 8a (blue) and the position of G406R mutation on the intracellular side of IS6 segment. B, amino acid sequence alignment of Cav1.2 exon 8 and 8a showing a high level of sequence homology (Genbank accession numbers: NM_001255999.2, NM_001256000.2). C, expression levels of Cav1.2 exon 8 and 8a in adrenal medullae of WT (n = 2 mice) and TS2‐neo (n = 3 mice) mice using quantitative RT‐PCR. One‐way ANOVA followed by Holm–Sidak post hoc test; ^*** P < 0.001. [Color figure can be viewed at wileyonlinelibrary.com]

Altered L‐type channel gating in TS2‐neo MCCs

L‐type currents were recorded in isolation by incubating MCCs in solutions containing 2 μM ω‐CTx‐MVIIC, 60 nM SNX‐482 and 3 μM TTA‐P2 to selectively block P/Q, N, R and T‐type Ca²⁺ channels, which are effectively expressed in MCCs (Mahapatra et al. 2012). Pharmacologically isolated L‐type currents were recorded in 10 mM Ca²⁺. They activated around −40 mV, peaked at nearly +15 mV, reversed at around +60 mV (Fig. 3 A and C) and were fully blocked by 15 μM nifedipine in WT‐MCCs (Fig. 3 B, n = 6) (Marcantoni et al. 2010). The corresponding normalized L‐type channel conductance (g _Ca(V)) had half‐maximum values (V _1/2) at −0.7 mV and steep voltage dependence (k = 10.7 mV for an e‐fold change; n = 17) (Fig. 3 D). In TS2‐neo MCCs, Ca²⁺ currents activated at slightly more negative potentials (Fig. 3 A and C) and reached similar current amplitudes at +20 mV compared to WT‐MCCs (n = 17; Fig. 3 E). The corresponding current densities (pA/pF) were significantly smaller (mean current densities were 15.2 and 9.9 pA/pF WT vs. TS2‐neo cells; Fig. 3 E) due to the increased cell capacitance (Fig. 3 E). The membrane capacitance increased by 31% in the cells used for Fig. 3 C (n = 19 WT and n = 20 TS2‐neo; P < 0.05), but the increase was consistently observed over a wider pool of MCCs (n = 65 and n = 59 cells; 14.5%, P < 0.05; Fig. 3 E, middle). The negative voltage shift of channel activation was quantified by comparing the g _Ca(V) curve to WT cells. TS2‐neo data were best fitted with a Boltzmann function shifted by ∼−5 mV with respect to WT cells (Fig. 3 D).

Figure 3.

Voltage dependence of L‐type currents and cell capacitance are altered in TS2‐neo mutated MCCs

A, two sets of pharmacologically isolated L‐type Ca²⁺ currents evoked by depolarizing pulses of 20 ms from −40 to +20 mV in step of 10 mV from a WT and TS2‐neo MCC. B, overlapped L‐type Ca²⁺ currents recorded from a WT‐MCC at +20 mV in control condition (black trace) and after acute application of 15 μM nifedipine (red trace), which fully blocks the Ca²⁺ current. C, normalized I–V relationship of peak L‐type Ca²⁺ currents recorded during step depolarizations of 20 ms at the indicated potentials from V _h = −70 mV for WT (black squares) and TS2‐neo MCCs (red circles) (n = 17 WT cells in 6 cultures and n = 17 TS2‐neo cells in 6 cultures; ^* P < 0.05). TS‐2 neo L‐type current activation (red curve) is leftward shifted with respect to WT (black curve) (^** P < 0.01; ^* P < 0.05, unpaired t test). D, normalized L‐type Ca²⁺ channel conductance, g _Ca(V), for WT (black line) and TS2‐neo MCCs (red line) calculated from g _Ca(V) = I _Ca/(V − E _Ca) with E _Ca = +60 mV in which is evident the same negative shift of the I–V relationship. The two continuous curves are the results of the fit with two Boltzmann equations with half‐maximal values V _1/2 (in mV) and slope factors k (in mV) obtained from the fit: −0.7 and 10.7 mV (WT; black curve) and −5.8 and 10.5 mV (TS2‐neo; red curve) (n = 17 WT cells in 6 cultures and n = 17 TS2‐neo cells in 6 cultures). E, mean peak L‐type currents, membrane capacitance and current density from WT (black bars) and TS2‐neo MCCs (red bars). TS2‐neo mutation increases the membrane capacitance either determined from the pool of cells used to obtain g _Ca(V) (n = 19 WT cells from 6 cultures and n = 20 TS2‐neo cells from 6 cultures) or from a larger pool of cells used for other experiments (n = 65 WT cells in 12 cultures and n = 59 TS2‐neo cells in 12 cultures). The capacitance increase reduces the L‐type Ca²⁺ current density of TS2‐neo MCCs.

In 2 mM Ca²⁺ L‐type Ca²⁺ currents in MCCs inactivate with fast and slow decaying components during long depolarizations, while in 2 mM Ba²⁺ only slow inactivation is observed. The fast component represents the Ca²⁺‐dependent inactivation (CDI) and slow component the voltage‐dependent inactivation (VDI) (Scharinger et al. 2015). In TS2‐neo mice L‐type channel inactivation is expected to be slower and less complete (Splawski et al. 2005), but it is unclear whether the effect is preferentially on CDI or VDI (Barrett & Tsien, 2008; Dick et al. 2016). Therefore, we compared the time course of L‐type Ca²⁺ currents during pulses of 1 s to +10 mV in 10 mM Ca²⁺ in WT and TS2‐mutated cells. In both cell types, L‐type channel inactivation occurred to a variable degree but in all cells tested inactivation exhibited a fast and a slow component. The percentage of Ca²⁺ current inactivation after 1 s was significantly attenuated in TS2‐neo MCCs (Fig. 4 A). The remaining Ca²⁺ current after 1 s increased from 44 ± 3% (n = 11) in WT to 59 ± 3% (n = 11) in TS2‐neo MCCs (P < 0.01) preserving the characteristic double exponential decaying time course. Best fit of the mean normalized Ca²⁺ currents in WT and TS2‐neo cells gave the following values for the amplitudes (A _fast, A _slow), time constants (τ_fast, τ_slow) and baseline value (C) of the two exponentials: A _fast −0.15, τ_fast 37.9 ms, A _slow −0.55, τ_slow 851 ms, C −0.29 in WT (n = 11); and A _fast −0.14, τ_fast 34.5 ms, A _slow −0.38, τ_slow 727 ms, C −0.48 in TS2‐neo (n = 11). The mutation mainly reduced the size of A _slow (−0.55 vs. −0.38) associated with the VDI while preserving the size of A _fast and time constants of both components. This was confirmed by the changes of Ba²⁺ current inactivation. Ba²⁺ currents lack the fast CDI component in TS2‐neo MCCs (Fig. 4 B). After 1 s, the persisting Ba²⁺ current is ∼15% higher in TS2‐mutated cells (P < 0.05; Fig. 4 B). Curve fitting furnishes the following amplitudes and time constants: A _slow −0.67, τ_slow 1230 ms, C −0.23 in WT, and A _slow −0.59, τ_slow 1280 ms, C −0.41 in TS2‐neo cells, suggesting that the mutation mainly decreased the size of A _slow by ∼14% by an increased percentage of non‐inactivated current. We also quantified the voltage dependence of CDI using a double pulse protocol (Fig. 4 C). Inactivation revealed the typical U‐shaped curve and was significantly different at more positive voltages (Fig. 4 C) (Scharinger et al. 2015).

Figure 4.

L‐type channel inactivation is slower and less complete in TS2‐neo mutated MCCs

A, mean normalized L‐type Ca²⁺ currents evoked by 1 s depolarizing pulses to +10 mV (left) for WT (n = 11 cells in 4 cultures; black trace) and TS2‐neo MCCs (n = 11 cells in 4 cultures; red trace). In the two mean traces are evident the fast (CDI) and slowly inactivating (VDI) components of L‐type Ca²⁺ currents. Right: averages of the percentage of L‐type Ca²⁺ currents persisting after 1 s depolarization to +10 mV from WT (black) and TS2‐neo (red) MCCs (^** P < 0.01). The two continuous curves of different colour drawn on top of the traces are the result of a curve fit with a double exponential function (see text). B, same normalized mean traces as described in A, with Ba²⁺ replacing Ca²⁺ as charge carrier (n = 11 WT cells in 4 cultures and n = 9 for TS2‐neo cells in 4 cultures). The TS2‐neo mutation slows the VDI of L‐type currents. C, top: pulse protocol used to determine the voltage dependence of CDI of isolated L‐type currents for WT (black traces) and TS2‐neo MCCs (red traces). Bottom: mean normalized values for CDI vs. voltage for WT (black line) and TS2‐neo MCCs (red line). Data represents the fraction of the peak current remaining after 100 ms upon depolarization with Ca²⁺. TS2‐neo‐isolated L‐type Ca²⁺ currents are less inactivating, especially at potentials >+40 mV (n = 7 WT cells in 2 cultures and n = 7 TS2‐neo cells in 2 cultures; ^* P < 0.05).

Resting ‘window’ Ca²⁺ current increases in TS2‐neo MCCs

To quantify the resting window Ca²⁺ current we determined the voltage dependence of steady‐state inactivation (SSI) to estimate the availability of L‐type channels at various potentials. Figure 5 A and B illustrates the pulse protocol used (Pinggera et al. 2015, 2017) and shows examples of Ca²⁺ currents recorded during test pulses P₁ and P₂ at +20 mV separated by 5 s depolarization to a pre‐determined potential in WT (black traces) and TS2‐neo MCCs (red traces). The ratio I _P2/I _P1 gives the percentage of available L‐type channels at the fixed potential. As for other slowly inactivating voltage‐gated Ca²⁺ channels, the I _P2/I _P1 ratio is maximal at very negative voltages and decreases steeply in a voltage‐dependent manner with increasing voltages to reach a minimum at positive potentials in WT‐MCCs (Fig. 5 C). In TS2‐neo MCCs, channel availability is shifted toward negative potentials in a broad range of voltages and exhibits an increased minimum (from 0.15 to 0.23) above 0 mV (Fig. 5 C). Best fit analysis using the Boltzmann functions reveals a −9 mV shift of V _1/2 in TS2‐neo cells.

Figure 5.

Steady‐state inactivation of L‐type current is leftward shifted and ‘window current’ is markedly increased in TS2‐neo mutated MCCs

A, pulse protocol used to determine the steady state inactivation (SSI) of isolated L‐type Ca²⁺ currents in MCCs. B, representative L‐type Ca²⁺ currents recorded during test protocols P₁ and P₂ at +20 mV, separated by 5 s depolarization to a pre‐determined potential in WT (black traces) and TS2‐neo MCCs (red traces). C, steady‐state inactivation curves for WT (black squares) and TS2‐neo (red circles) MCCs. The two continuous curves are the results of a curve fit with two Boltzmann equations with V _1/2 and slope factors k: −20.8 mV and 7.4 mV (n = 11 WT cells in 4 cultures; black curve) and −29.5 mV and 8.2 mV (n = 12 TS2‐neo cells in 4 cultures; red curve). D, window Ca²⁺ current (I _w) calculated from SSI(V) and g _Ca(V). I _w is obtained by multiplying the probability of channel opening p _o(V) by the driving force (V − E _Ca) with E _Ca = +60 mV (continuous curves in bottom panel). The two p _o(V) curves are obtained by multiplying SSI(V) by g _Ca(V) (dotted curves in top panel). Notice the increased amplitude of I _w between −70 and −40 mV (shaded area).

The negative shifts of both SSI and activation predicted an increased window Ca²⁺ current (I _W) near resting potential. We calculated I _W by first multiplying the two curves to give the probability of open L‐type channels (p _o(V); Fig. 5 D, dotted curves) and then multiplying for the driving force (V – E _Ca) to estimate the amount of normalized I _W in WT and TS2‐neo cells (continuous curves; Fig. 5 D bottom). A drastic increase of inward Ca²⁺ currents was evident in mutated MCCs, particularly around resting voltages in 10 mM Ca²⁺ (−60 to −40 mV; shadowed area).

Spontaneous AP firing is altered in TS2‐neo MCCs

A major consequence of the altered Cav1.2 activation/inactivation kinetics and increased window Ca²⁺ current in TS2‐mutated MCCs is a drastic change in spontaneous AP firing at rest. The mutation had no effects on resting potential (−47.9 vs. −48.5 mV; Fig. 6 A) but increased the percentage of cells that underwent burst firing (Fig. 6 B). Mutated and WT‐MCCs had similar percentage of silent cells (20% WT n = 20 total cells vs. 15% TS2‐neo n = 20 total cells) but differing percentages of burst and tonic firing cells at rest. Half of TS2‐neo cells underwent burst firing, while only 35% had tonic firing (Fig. 6 B). In contrast, WT cells had mainly tonic firing (65%) while only 15% produced bursts (P < 0.05; chi‐square test).

Figure 6.

TS2‐neo mutation increases the percentage of cells undergoing burst firing and alters the waveform of spontaneous repetitive action potentials

A, mean V _rest in WT (n = 14 cells in 6 cultures, black bar) and TS2‐neo MCCs (n = 18 cells in 6 cultures, red bar). B, proportions of silent, tonically or burst firing MCCs in WT (n = 20 cells in 6 cultures; black bars) and TS2‐neo MCCs (n = 20 cells in 6 cultures, red bars). The TS2‐neo mutation switches MCCs’ spontaneous activity into bursts (^* P < 0.05; chi‐square test). C, mean frequency, peak amplitude, half‐width and antipeak amplitude during spontaneous tonic firing in WT (n = 11 cells in 6 cultures, black bars) and TS2‐neo MCCs (n = 5 cells in 6 cultures, red bars). Mean values were obtained by averaging single values from 6 consecutive spikes during the spontaneous activity (^* P < 0.05; unpaired t test). D and E, representative traces of spontaneously repetitive AP firing in a WT (D) and a TS2‐neo MCC (E). Details of the firing activity are shown in an expanded time scale in the grey inset below. F, overlap of mean WT (black) and TS2‐neo APs (red). Averaged traces were obtained from six consecutive APs during the trains. G, phase‐plane plot (dV/dt vs. V) of six consecutive APs from WT (black traces) and TS2‐neo MCCs (red traces). Inset: mean maximal dV/dt as a measure of maximal rate of rise of APs from n = 5 WT cells in 5 cultures and n = 7 TS2‐neo cells in 5 cultures (^** P < 0.01, unpaired t test).

TS2‐neo mutation changed also the waveform of single APs and bursts. Tonic firing (Fig. 6 C and D) occurred at about the same frequency (0.9 vs. 1.1 Hz; Fig. 6 C) but had significantly smaller amplitude (36.0 vs. 57.5 mV; P < 0.01), larger half‐width (10.2 vs. 3.9 ms; P < 0.01), reduced after‐hyperpolarization (−4.3 vs. −7.2 mV; P < 0.05) and slower rising phases (Fig. 6 F). The rising phase velocity was estimated from the mean maximum of the dV/dt vs. V phase‐plane plot of six consecutive APs (14.6 vs. 48.5 V/s; P < 0.01; Fig. 6 G), suggesting a markedly reduced density of Nav channels that would be available to support the AP upstroke.

The TS2‐neo mutation altered also the waveform of bursts (Fig. 7 A and B). Bursts in mutated cells had undershoot peaks (below zero). The amplitudes of the first and last AP were significantly smaller than in WT‐MCCs (−15.6 and −24.9 mV in TS2‐neo n = 7 vs. 5.0 and −16.3 mV in WT n = 4; P < 0.01). We also noticed that burst firing occurred either at low (n = 4 out of 7 cells; Fig. 7 B; slow burst) or at high frequency in which single spikes appeared intermixed with burst (n = 3 out of 7 cells; Fig. 7 B; fast burst). The two types of firing were pooled together to determine the mean bursts duration, the mean plateau amplitudes and number of bursts per minute. These parameters were estimated as shown in the inset of Fig. 7 A (red trace; see figure legend for details) and as previously described in Fig. 9 of Guarina et al. (2017). They were comparable to WT‐MCCs (Fig. 7 D), while the mean number of events decreased significantly from 4.0 to 2.7 (P < 0.01; Fig. 7 C), suggesting rapid loss of APs during the plateau depolarization of the burst. In conclusion, the TS2‐neo mutation induces an increased percentage of MCCs undergoing bursts with APs of lower amplitudes that quickly decay during the plateau potential.

Figure 7.

The waveform of spontaneous burst firing is markedly altered in TS2‐neo mutated cells

A–C, representative traces of spontaneous burst firing in WT (A) and TS2‐neo MCCs (B and C). TS2‐neo burst activity at low‐ (B) and high‐frequency (C) markedly undershoots with respect to WT. Grey windows show expanded intervals of bursts at the same time scale. The expanded burst (red trace in the right inset) shows how the number of events, mean plateau phase and bursts duration were estimated. The mean plateau amplitude was determined as the half‐amplitude of the slow repolarization that was assumed to increase linearly from the first to the last spike. Burst duration was calculated from the fast rising of the first AP to the end of the fast repolarization of the last AP, just before the onset of the following slower depolarization. D, means of number of events during burst (left), first and last peak amplitude (right) for WT (n = 4 cells in 3 cultures, black bars) and TS2‐neo MCCs (n = 7 cells in 5 cultures, red bars) (^** P < 0.01, ^* P < 0.05; unpaired t test). TS2‐neo mutation drastically reduces the number of events during burst and shifts to negative or more negative values the first and the last peak amplitude. E, comparison of mean plateau amplitude, burst duration and number of bursts per minute between WT (n = 4 cells in 3 cultures, black bars) and TS2‐neo MCCs (n = 7 cells in 5 cultures, red bars). The burst parameters are comparable between the two groups.

The findings of Figs 6 and 7 suggest that the increased percentage of burst firing in mutated cells may derive from a reduced density of functional Nav channels. This may attenuate the activation of BK and K_V channels with consequent decrease of the repolarization phase and subsequent insufficient recovery of inactivated Nav channels that sustain the following APs (see Vandael et al. 2015). In order to test this, we next examined the properties of Nav channels in WT and TS2‐mutated cells.

Nav1 channel density is markedly reduced in TS2‐neo MCCs

The increased ‘window’ Ca²⁺ current (Fig. 5 D) and the drastic changes of AP firing in TS2‐mutated MCCs (Figs 6 and 7) suggest that the Cav1.2 TS2‐mutation may produce drastic changes to the ionic conductances responsible for MCCs excitability. We therefore tested whether the TS2 mutation altered the voltage dependence and size of Nav currents responsible for the AP upstroke (Vandael et al. 2015 b; Lingle et al. 2018). We found that peak Nav currents during step depolarization to −10 or 0 mV in 2 mM Ca²⁺ had variable amplitudes with a clear trend to decrease in TS2‐mutated MCCs (mean values 1.25 vs. 0.94 nA in WT vs. TS2‐neo cells) (Fig. 8 A and B). The trend was converted into a significant decrease of Nav current density when normalized for the increased membrane capacitance (Fig. 8 B middle), in good agreement with the capacitance values of Fig. 3 E. Nav current densities were ∼43% smaller in TS2‐neo MCCs (P < 0.05; Fig. 8 B right).

Figure 8.

Nav channel density is markedly attenuated and voltage‐dependent steady‐state inactivation of Nav channels is leftward shifted in TS2‐neo mutated MCCs

A, isolated Na⁺ currents evoked by depolarizing pulses of 10 ms from −50 to +50 mV in steps of 10 mV from WT (black traces) and TS2‐neo MCCs (red traces). B, mean values of peak Nav currents at 0 mV, membrane capacitance and current density from WT (n = 21 cells in 4 cultures, black bars) and TS2‐neo MCCs (n = 21 cells in 4 cultures, red bars; ^* P < 0.05, unpaired t test). The TS2‐neo mutation significantly increases the membrane capacitance thus reducing Nav current density. C and D, I–V relationship and normalized conductance for Nav current densities measured at V _h = −70 mV for WT (n = 11 cells in 3 cultures; black traces) and TS2‐neo MCCs (n = 13 cells in 3 cultures; red traces). E and F, mean normalized values for fast and slow Nav channel inactivation from WT (n = 11 cells in 4 cultures; black traces) and TS2‐neo MCCs (n = 14 cells in 4 cultures; red traces).

The shape of the normalized I–V curves and the voltage dependence of Nav channel conductance, g _Na(V), were not significantly affected (Fig. 8 C and D) and no changes were observed in the voltage dependence of fast inactivation (Fig. 8 E), suggesting that the Cav1.2 mutation has no major effects on the activation–inactivation kinetics of functionally expressed Nav channels. The 43% decrease of Nav current density resulted from a decreased number of functional Nav channels and an increased size of cell area (swelling), most likely associated with the altered resting Ca²⁺ level induced by the increased ‘window’ Ca²⁺ current.

Since increases of intracellular Ca²⁺ are likely to alter a variety of intracellular pathways that regulate the SSI of Nav channels (Carr et al. 2003; Jo & Bean, 2017), we also tested whether the voltage dependence of the slow recovery from inactivation of these channels was altered. Figure 8 F shows that indeed Nav channels of mutated MCCs recover less during 5 s‐long pulses more positive than −40 mV. The SSI curve had more or less the same V _1/2 voltage (−30 mV, WT n = 11 vs. −28 mV TS2‐neo n = 14) but was significantly lower at potentials >−40 mV. Minimum values for SSI at +30 mV were 0.57 for WT and 0.43 for TS2‐neo cells, suggesting a reduced number of available channels at potentials more positive than −40 mV. This explains why during prolonged depolarization between −40 and −20 mV (Fig. 7 B and C) a considerable fraction of Na_V channels remain inactivated and do not participate to the burst firing in mutated MCCs.

BK channel currents are preserved in TS2‐neo MCCs

Given the importance of BK channels in the control of AP repolarization and setting of burst firing in MCCs (Marcantoni et al. 2010; Martinez‐Espinosa et al. 2014; Lingle et al. 2018), we checked whether BK currents were altered by the TS2‐neo mutation. The voltage dependence of BK channels was tested using the pulse protocol shown in Fig. 9 A (Gavello et al. 2015; Guarina et al. 2017). BK currents were obtained by first determining the contribution of K_V currents to the total by adding 1 μM paxilline to block BK channels (Zhou & Lingle, 2014). Subtraction of Kv currents from the total gave the BK currents (bottom panel). Despite a sizeable V _1/2 leftward shift, the voltage dependence of BK channels in TS2‐mutated cells (n = 10) was not significantly different from WT‐MCCs (n = 8) (Fig. 9 B).

Figure 9.

Voltage and Ca²⁺ dependence of BK channels is not affected in TS2‐neo mutated MCCs

A, top, pulse protocol used to evaluate the voltage dependence of BK currents. Bottom, representative total K⁺ currents (BK + K_V) (top), during application of the BK channel blocker paxilline (1 μM, middle) and subtraction of total from paxilline‐resistant currents (bottom). WT currents are shown to the left and TS2‐neo currents to the right. B, conductance of BK channels obtained by a sigmoidal fit for WT (n = 8 cells in 3 cultures black trace) and TS2‐neo currents (n = 10 cells in 3 cultures,red traces). The mutation leads to a left shift of the Boltzmann curve. C, top, pulse protocol used to evaluate the Ca²⁺ dependence of BK currents. Bottom, BK and Kv currents measured in WT (left) and TS2‐neo MCCs (right). The BK current amplitude (black arrow, I _BK) is obtained by subtraction of the K_V current (measured at +120 mV; red arrow, I _Kv) from the total K⁺ current. D, normalized I _BK currents vs. voltage for WT (n = 8 cells in 3 cultures, black trace) and TS2‐neo MCCs (n = 10 cells in 3 cultures, red trace). Notice the ‘bell‐shaped’ curve corresponding to the Ca²⁺ dependence of BK currents. Right, mean BK current amplitude measured at +120 mV, for WT (black) and TS2‐neo cells (red). E, left, mean Kv current amplitudes in WT (black) and TS2‐neo cells (red). Right, mean Kv current amplitude measured at +120 mV, for WT (black n = 8 cells in 4 cultures) and TS2‐neo cells (n = 10 cells in 4 cultures red).

We also tested the Ca²⁺ dependence of BK channels by preloading the cell with Ca²⁺ using pre‐steps of increasing voltage (−60 to +120 mV) to activate BK and K_V channels and measure their peak current amplitude at +120 mV. Since at +120 mV there is no inward Ca²⁺ entry, the outward current is the K_V current (I _Kv). This, subtracted from the total current gives I _BK (arrows in Fig. 9 C right). In mutated MCCs (n = 10) both the maximal peak amplitude at +20 mV (2.3 ± 0.8 nA) and the voltage dependence of the ‘bell shaped’ I–V relationship of BK channels were not significantly different from WT‐MCCs (n = 8) (Fig. 9 D and E). Similar maximal amplitudes in WT and TS2‐neo cells were also found for I _Kv at +120 mV (Fig. 9 E). We also found no statistically significant changes in the time course of BK current inactivation (whether inactivating or non‐inactivating; Solaro et al. 1995) between WT and TS2‐neo MCCs. In n = 10 TS2‐neo mutated cells, six were non‐inactivating and four inactivating while in n = 8 WT cells, four were inactivating and four non‐inactivating (chi‐square test, P = 0.67). The same ratio of activating vs. non‐inactivating currents was evident in a larger pool of WT cells used for other experiments. In conclusion, the TS2‐neo mutation does not affect the functional expression of activating and non‐inactivating BK channels.

TS2 mutation induces cell swelling and impairs mitochondria

The drastic decrease of Nav current density (Fig. 8), the increased membrane capacitance (Figs 3 and 8) and the altered AP firing (Figs 6 and 7) cannot be simply attributed to changes of Cav1.2 channel gating. Some secondary structural changes may occur in mutated MCCs, triggered by the increased Ca²⁺ entering the cell through Cav1.2‐mutated channels (Fig. 5 D). Conventional electron microscopy was used to view the ultrastructure of CCs in intact adrenal medulla glands of WT and TS2‐neo mice (Salio et al. 2017). Transition electron microscopy images exhibited remarkable structural changes at both the nuclear and cytoplasmic levels (Fig. 10 A). Nuclei of TS2‐neo MCCs had a more heterochromatic appearance than WT‐MCCs. TS2‐neo MCCs showed also large areas of vacuolization, conferring a more electron‐lucent appearance, as well as an apparent reduction of intracellular organelles (Fig. 10 A). Mitochondria were abundant and densely distributed at the perinuclear and cortical regions in WT‐MCCs (Villanueva et al. 2014), but their density was strongly reduced in CCs of TS2‐neo (0.05 ± 0.003/μm², n = 50 vs. WT‐MCCs (0.22 ± 0.01/μm², n = 50; Mann–Whitney test, P < 0.0001). Mitochondria of mutant mice had also altered ultrastructure, exhibiting a reduced number of cristae when compared with WT‐MCCs (Fig. 10 A). In addition, the density of large dense‐core granules (LDCGs) was significantly lower in the cytoplasm of TS2‐neo MCCs (1.81 ± 0.09/μm², n = 50) compared to WT‐MCCs (3.00 ± 0.17/μm², n = 50; Mann–Whitney test, P < 0.0001). In mutant mice, LDCGs were visibly separated from their perigranular membrane by a swollen electron‐lucent space.

Figure 10.

Transition electron microscopy images, biochemical assays of mitochondrial function, TUNEL and DAPI tests in WT and TS2‐neo MCCs

A, top, electron micrographs showing WT (left) and TS2‐neo MCCs (right) at low magnification. TS2‐neo MCCs show an altered nuclear (Nu) and cytoplasmic morphology, with extensive vacuolization (vac) and fewer intracellular organelles. Bottom, higher magnification corresponding to the boxed regions, showing the abnormal morphology of mitochondria (mit) and large dense core granules (LDCG, white arrows) in TS2‐neo MCCs. Specifically, in mutated cells mitochondria have fewer cristae whereas LDCGs show an evident perigranular swelling. Scale bars: top images 2 μm; bottom images 500 nm. B, top, histogram illustrating the cell area distribution in WT (grey and black bars) and TS2‐neo MCCs (light‐ and dark‐red bars). Note the clear separation with significantly different means. Bottom, mean cell area (μm²) in TS2‐neo (red bars; n = 50 cells) and WT‐MCCs (black bars; n = 50 cells; Mann–Whitney test; ^**** P < 0.0001). C, the rate of electron flux from complex I to complex III was evaluated in isolated MCCs (left panel). The mitochondrial membrane potential was assessed by the JC‐1 staining method. The percentage of green vs. red mitochondria was considered a marker of mitochondrial depolarization and permeability transition (middle panel). ATP was measured in duplicates in the mitochondrial extracts by a chemiluminescence‐based assay (right panel). All measurements were performed in duplicate and data are presented as means ± SD (n = 6 animals/group; ^*** P < 0.001). D, low‐magnification micrographs of adrenal medulla gland sections obtained from WT and TS2‐neo mice. All sections were decorated by TUNEL (red) and DAPI (grey) to reveal apoptotic cells. Scale bar: 50 μm. The yellow arrows in the right panel indicate two rare TUNEL positive mutated cells.

The abundant vacuolization observed in nearly all TS2‐neo mutated cells suggested a marked cell swelling, in agreement with the cell capacitance increases. We calculated the area of the cell profiles visible in single sections and found that the histograms of WT (grey bars) and TS2‐neo cell areas (light–dark red bars) had different means: 113.3 μm² (WT; n = 50) and 140.2 μm² (TS2‐neo; n = 50) (Mann–Whitney test, P < 0.0001) (Fig. 10 B). Thus, TS2‐neo cells undergo marked swelling that increases the cell surface area by ∼24%, which is comparable to the estimate increase of membrane capacitance (∼19%).

Mitochondrial function was further assayed by measuring (1) the electron flux between complex I and III of the mitochondrial respiratory chain as a measure of mitochondrial metabolic activity (Fig. 10 C left), (2) the mitochondrial electrochemical potential to assay the percentage of damaged mitochondria (Fig. 10 C middle), and (3) the levels of mitochondrial ATP produced by the electron transport chain (Fig. 10 C right) (Riganti et al. 2015). All three parameters were markedly altered in the TS2‐mutated MCCs. The electron flux between complex I and III and the synthesis of mitochondrial ATP were drastically reduced (P < 0.001), while the percentage of damaged cells increased (P < 0.001) due to a sustained mitochondrial depolarization in TS2‐neo cells. Thus, mitochondrial metabolism appears significantly impaired in TS2‐neo cells, justifying the severe ultrastructural changes of mutated MCCs. We also tested whether the reduced mitochondrial metabolism could drive any mitochondria‐dependent apoptotic machinery. To test this, we prepared sections of adrenal glands derived from WT and TS2‐neo mice for TUNEL and DAPI assays to reveal apoptotic cells (Fig. 10 D). No significant differences were observed, either in the cortical or in the medulla areas. The amounts of TUNEL positive cells and/or picnotic nuclei were similar. This excludes that the altered mitochondrial metabolism may induce massive cell death.

Ca²⁺‐dependent secretion is attenuated in TS2‐mutated MCCs

The altered morphology of secretory granules also suggested that the Ca²⁺‐dependent secretion of catecholamines may be altered in TS2‐mutated MCCs. We therefore tested how the mutation affects the depolarization‐evoked capacitance increases (∆C) and the size of the readily releasable pool (RRP) of vesicles (Carabelli et al. 2003; Marcantoni et al. 2009; Gosso et al. 2011). In WT‐MCCs, step depolarization of 100 ms to +10 mV induced a mean ∆C of 26.5 fF (n = 12) that decreased to 13.0 fF (n = 16) in TS2‐neo MCCs (P < 0.01), suggesting a significant decrease of vesicle secretion despite the maintained average quantity of Ca²⁺ charge (Q) at comparable levels in WT and TS2‐neo cells (Fig. 11 A and B). A similar ∆C reduction was observed at 0 mV where the mean ∆C decreased from 24.4 (n = 11) to 10.8 fF (n = 13) (Fig. 11 A and B). These data were obtained by repeating the 100 ms depolarizations to +10 and 0 mV either one or two times on the same cell, separated by 60 s to allow full replenishment of the RRP between stimulations (Carabelli et al. 2003; Gavello et al. 2013). Mean values of ∆C for each cell were used to calculate the mean ∆C at +10 and 0 mV in Fig. 11 B. To further study the origin of the decreased ∆C induced by the mutation, we also estimated the Ca²⁺ dependence of the secretory response by plotting the mean ∆C vs. the mean Ca²⁺ charge density (pC/pF) obtained from each cell, pooling together the values at +10 and 0 mV (Fig. 11 C). The slope of the linear Ca²⁺ dependence of secretion was estimated by linear regression and resulted 8.4 ± 1.8 fF/(pC/pF) in WT‐MCCs (black line, n = 12 cells) and decreased to 3.0 ± 0.8 fF/(pC/pF) in TS2‐neo mutated cells (red line, n = 16 cells). Although a tendency to decrease of the Ca²⁺ dependence of secretion in TS2‐mutated cells is evident, the two slopes were not significantly different due to the large scattering of data (P > 0.05), suggesting caution in drawing quantitative conclusions on this issue.

Figure 11.

Ca²⁺‐dependent secretion is attenuated in TS2‐neo mutated MCCs

A, representative traces of depolarization‐evoked secretion in WT (black traces, left) and TS2‐neo MCCs (red traces, right). The capacitance increase (ΔC, top) and the Ca²⁺ current (bottom) recorded during a step depolarization to +10 mV (upper panel) or 0 mV (lower panel) are shown. B, mean capacitance changes ΔC (left) and mean quantity of charge Q (right) associated to the depolarizing pulse measured in WT (n = 12 cells in 6 cultures, black bars) and TS2‐neo MCCs (n = 16 cells in 7 cultures, red bars) both at +10 mV and measured in WT (n = 11 cells in 6 cultures, black bars) and TS2‐neo MCCs (n = 13 cells in 7 cultures, red bars) at 0 mV. ΔC is reduced at both voltages in mutated MCCs with respect to WT (^** P < 0.01, unpaired t test). C, Ca²⁺ dependence of the exocytosis determined by plotting ΔC (fF) vs. the corresponding Ca²⁺‐charge density Q (pC/pF). The slope of the linear regression for WT (filled squares, black line, n = 12 cells, R ² = 0.67) is not statistically different from the TS2‐neo one (see text) (open circles, red line, n = 16 cells; R ² = 0.51; P > 0.05, unpaired t test; see Methods). D, representative traces of the double pulse protocol used to estimate the maximal size of the RRP in WT (left, black traces) and TS2‐neo MCCs (right, red traces). Upper limit of the RRP estimated from B _max = (ΔC ₁ + ΔC ₂)/(1 − (ΔC ₂/ΔC ₁)²), where ΔC ₁ and ΔC ₂ are the capacitance increases associated to the consecutive depolarizing pulses. E, mean B _max (RRP, left plot) and probability of release (P _r; right plot) for WT (n = 8 cells in 3 cultures, black bar) and mutated MCCs (n = 8 in 5 cultures, red bar). P _r is statistically smaller in TS2‐neo MCCs (^* P < 0.05, unpaired t test).

The size of the RRP and the probability of vesicle release (P _r) were determined using a double‐pulse protocol consisting of two consecutive pulses of 100 ms of equal amounts of Ca²⁺ charges, quickly delivered to avoid vesicle replenishment (Gillis et al. 1996; Carabelli et al. 2003). From the ∆C measured at the end of the two pulses (∆C ₁, ∆C ₂; Fig. 11 D) we calculated B _max = (ΔC ₁ + ΔC ₂)/(1 − (ΔC ₂/ΔC ₁)² corresponding to the size of the RRP and P _r = 1 − (∆C ₂/∆C ₁). We found that TS2‐neo and WT‐MCCs possess similar RRPs of ∼33 fF but distinct P _r (Fig. 11 E). P _r was ∼25% smaller in TS2‐neo MCCs, suggesting that the reduced catecholamine secretion in TS2‐mutated cells is likely due to mistuned molecular mechanisms associated to vesicle docking and fusion that are altered by the TS2‐neo mutation.

The Cav1.2 blocker verapamil rescues AP burst firing in TS2‐neo MCCs

To demonstrate that the altered excitability of TS2‐mutated MCCs derives from an increased Ca²⁺ entry through less inactivating Cav1.2 channels, we tested whether common L‐type channel blockers could rescue the normal firing of mutated MCCs by reducing Ca²⁺ influx at rest and during cell activity. We first tested if chronically applied verapamil (5–15 μM) could help rescuing the normal firing of mutated MCCs. Verapamil is a benzothiazepine that blocks Cav1 channels with high affinity when open (Lee & Tsien, 1983). First, however, we tested whether verapamil effectively blocked AP firing when applied acutely on WT‐MCCs. In n = 5 cells, we found that acute application of 5–15 μM verapamil hyperpolarized the cells by ∼8 mV and reversibly blocked AP firing (Fig. 12).

Figure 12.

Spontaneous repetitive firing recorded in current‐clamp from two representative WT‐MCCs

WT‐MCCs were bathed in Tyrode solution (2 mM Ca²⁺) and recorded in current‐clamp mode with 0 current injection. Verapamil at two different concentrations (15 and 5 μM) was applied during the period indicated by the horizontal bars. Notice the marked resting potential hyperpolarization and firing block during the application of the Ca²⁺ channel blocker.

Overnight application (18 h) of 15 μM verapamil was sufficient to rescue the normal burst firing of mutated MCCs (Fig. 13 A). The rescued firing was characterized by an increased first peak amplitude overshoot (+17.7 mV, n = 4 vs. −15.6 mV, n = 7; verapamil‐treated vs. TS2‐neo MCCs; P < 0.001) and by an increased last peak amplitude (−14.0 vs. −24.9 mV; P < 0.01) (Fig. 13 B). In agreement with these observations, the mean plateau amplitude was also more positive (−24.2 vs. −37.9 mV; P < 0.01) and the number of events within a burst returned to normal (4.5 vs. 2.7; P < 0.01) (Fig. 13 B). Interestingly, verapamil worked at relatively high doses (15 μM) and was unable to rescue the tonic repetitive firing observed in ∼65% of WT‐MCCs (Fig. 6). Lower doses of verapamil (1–5 μM) failed to rescue normal AP firing (n = 8).

Figure 13.

Overnight incubations of the L‐type channel blockers verapamil and nifedipine rescues the normal spontaneous firing in TS2‐neo mutated cells

A, representative burst firing from TS2‐neo MCCs overnight incubated with 15 μM verapamil. The pharmacological treatment completely recovers a normal burst activity. B, top: mean values for first and last peak amplitude during burst in WT (n = 4 cells in 3 cultures, black bars), TS2‐neo (n = 7 cells in 5 cultures, red bars) and TS2‐neo + verapamil 15 μM overnight‐treated MCCs (n = 4 cells in 3 cultures, grey bars). Verapamil incubation rescues the overshoot of first peak amplitude and makes less negative the last peak amplitude (^** P < 0.01, ^* P < 0.05, one‐way ANOVA followed by Holm–Sidak post hoc test). Bottom: mean number of events during burst (left) and mean plateau amplitude (right) in WT (black bars), TS2‐neo (red bars) and TS2‐neo + verapamil 15 μM (grey bars). Verapamil treatment rescues values comparable to WT. C, repetitive firing recorded from a representative TS2‐neo MCC overnight incubated with 0.3 μM nifedipine. Tonic firing is more evident in the last part of the recording, after the manual hyperpolarization induced by passing −5 pA to hold V _rest to −70 mV. This procedure was not used routinely and did not bias the estimate of maximal AP amplitude, which was always estimated soon after the initial track of the recording. D, proportions of silent, tonically or burst firing MCCs in WT (n = 20 cells in 6 cultures; black bars), TS2‐neo MCCs (n = 20 cells in 6 cultures, red bars) and TS2‐neo + 0.3 μM nifedipine‐treated MCCs (n = 23 cells in 6 cultures, grey bars) (^* P <0.05, ^** P < 0.01; chi‐square test). E, top left: single APs obtained from WT (black), TS2‐neo (red) and TS2‐neo + 0.3 μM nifedipine‐treated MCCs (grey). Top right and bottom: mean peak amplitude, antipeak amplitude and half‐width for WT (n = 14 cells in 6 cultures, black bars), TS2‐neo (n = 10 cells in 6 cultures, red bars) and TS2‐neo + 0.3 μM nifedipine‐treated MCC (n = 16 cells in 7 cultures, grey bars; ^** P < 0.01, ^* P < 0.05, one‐way ANOVA followed by Holm–Sidak post hoc test).

The Cav1.2 blocker nifedipine rescues the normal AP firing in TS2‐neo MCCs

The inability of verapamil to rescue normal tonic firing suggested that Cav1 blockers like nifedipine, which blocks Cav1.2 channels in the closed and inactivated state (Bean, 1984), might be more effective. Preliminary attempts using high doses of nifedipine (from 15 to 1 μM) were ineffective, presumably due to their toxic effects on MCC survival. We thus reduced the DHP concentration to 0.3 μM to block 40% of the L‐type current in MCCs (Marcantoni et al. 2010; Mahapatra et al. 2011). Overnight incubation of TS2‐mutated cells with 0.3 μM nifedipine (18–42 h) restored the normal repetitive firing. After washout of the DHP, 70% of n = 23 pretreated cells exhibited single repetitive APs with clear overshoot (Fig. 13 C), 13% underwent burst firing and 17% had no activity. These percentages were similar to those observed in WT cells and significantly different from TS2‐neo (Fig. 13 D; P < 0.05 and P < 0.01; chi‐square test). Concerning the tonic firing, overnight pretreatment with 0.3 μM nifedipine caused an increased peak amplitude that was comparable to WT‐MCCs (55.2 ± 2.8 mV, n = 16; Fig. 13 E), while the half‐width (4.0 ± 0.4 ms) and antipeak amplitude (−8.3 ± 1.2 mV) remarkably recovered compared with TS2‐mutated cells (P < 0.01; Fig. 13 E). In conclusion, 18–42 h incubation with nifedipine was able to rescue the normal repetitive firing of mutated MCCs.

To assess whether chronic treatment with nifedipine was beneficial only on TS2‐mutated cells with altered resting Ca²⁺ influx, and not a general positive consequence of the DHP on normal MCCs, we tested the effects of chronic nifedipine (18–42 h) in WT‐MCCs. Visual inspection under the microscope revealed that most nifedipine‐treated WT cells were swollen, vacuolated and had unclear membrane borders. In current‐clamp recordings after DHP washout, we found that of a total of n = 14 nifedipine‐treated MCCs, eight cells did not seal and six cells were depolarized (mean V _rest = −43.2 mV) and had no AP activity at rest. Of the latter group of cells, only two of them fired a single AP after 5–10 s hyperpolarization to −65, −70 mV and subsequent depolarization to their resting potential (Fig. 14 A). As a control, in 4 out of 4 nifedipine‐untreated WT‐MCCs we could record the normal spontaneous repetitive firing shown in Fig. 6 D, with overshoot APs (not shown). We conclude that partial block of L‐type channels for 18–42 h with 0.3 μM nifedipine completely abolishes the spontaneous repetitive firing and is therefore toxic for WT‐MCCs. This may occur since a 40% block of L‐type channels could drastically reduce the fraction of SK and BK channels steadily activated at rest in WT‐MCCs (Marcantoni et al. 2010; Vandael et al. 2012). Partial SK and BK channel block causes cell depolarization and continuous Ca²⁺ entry through unblocked Cav1 and Cav2 channels that over long periods could lead to loss of excitability and cell deterioration.

Figure 14.

Overnight incubations and acute application of L‐type channel blockers nifedipine abolishes the spontaneous firing of WT‐MCCs

A, representative trace with no evident resting activity recorded from a WT‐MCC overnight incubated with 0.3 μM nifedipine (n = 14 cells in 2 cultures). The DHP treatment completely abolishes the spontaneous firing activity. Single APs were evident following pre‐hyperpolarizations to ∼−70 mV passing negative current. Ba, representative trace recorded from a TS2‐mutated MCC in which it is evident that bath application of 3 μM nifedipine fully blocks the spontaneous firing activity (n = 5 cells in 2 cultures). Bb, representative trace recorded before, during and after acute application of 0.3 μM nifedipine on a TS2‐mutated MCC. Also at this low concentration, the drug completely blocks the spontaneous firing activity (n = total of 6 cells in 3 cultures).

All the above experiments were carried out after chronic treatment with nifedipine and in the absence of the DHP during current‐clamp recordings. Cell dishes were washed with standard Tyrode solution, which was the extracellular solution for these recordings. To ensure that the effects of nifedipine were due to the chronic treatment and not to an acute action of any residual nifedipine, in a series of experiments we tested the effects of acute application of the DHP on TS2‐mutated cells. Acute application of 3 μM nifedipine fully blocked the activity of all the TS2‐mutated cells tested (n = 5), regardless of their firing mode (tonic or bursts) (Fig. 14 Ba). The blocking action was immediate (2–3 s) and persisted for the entire duration of the application (>60 s) as previously reported for WT‐MCCs (Marcantoni et al. 2010; Vandael et al. 2010; Guarina et al. 2017). Application of 0.3 μM nifedipine in other TS2‐mutated cells was variable. In 3 out of 6 cells the DHP fully blocked the activity (Fig. 14 Bb), while in the remaining three cells nifedipine caused a small depolarization of about 5 mV followed by a reduced frequency of AP firing. In conclusion, acute application of nifedipine (3–0.3 μM) caused a depressive action on AP firing similar to that observed in WT‐MCCs.

Overnight incubation with nifedipine rescues the density and steady‐state inactivation of Nav channels and Ca²⁺‐dependent secretion of TS2‐neo MCCs

Recovery of the AP firing after chronic incubation with nifedipine suggests that the density of functional Nav channels recovers during the DHP treatment. Figure 15 A clearly shows that the mean maximum of the dV/dt vs. V phase‐plane plot calculated on the first AP (50.1 V/s; n = 16) is comparable to that of WT‐MCCs on the same panel (41 V/s; n = 10) and in Fig. 6 G (48.5 V/s; n = 7). Although this is already highly indicative of an increased density of functional Nav channels, we also checked whether the size, time course and voltage‐dependent properties of Nav currents recovered during the DHP treatment. Figure 15 B shows that Nav current amplitudes recovered partially following the overnight incubation with nifedipine (1.1 ± 0.06 nA, n = 11, TS2+nife vs. 0.87 ± 0.07 nA; n = 13; TS2‐neo; P < 0.05) while membrane capacitance was not significantly different from TS2‐neo mutated cells. The mean density of Nav currents increased by ∼30%: from 129.9 to 187.8 pA/pF (P < 0.05). Activation and inactivation time course was not altered (Fig. 15 C) and the same occurred to the voltage dependence of I–V relationships (blue triangles in Fig. 15 D), which was similar to that of WT (dashed black curve) and TS2‐neo mutated cells (red circles). The voltage dependence of fast inactivation was unchanged (not shown) while the voltage dependence of steady‐state inactivation recovered partially (blue triangles vs. red circles in Fig. 15 E), to become nearly indistinguishable from the SSI of WT cells (dashed black curve). The red and blue curves had nearly the same V _1/2 (−28 mV TS2+nife vs. −30 mV TS2‐neo cells) but well separated minimum values (0.52 TS2+nife and 0.43 TS2‐neo cells at +30 mV). In conclusion, it appears that the recovery of AP firing after overnight nifedipine application was paralleled by a robust recovery of functional Nav channels.

Figure 15.

Overnight incubations with the L‐type channel blocker nifedipine rescues the density and SSI of Nav channels as well as Ca²⁺‐dependent secretion in TS2‐neo MCCs

A, phase‐plane plot (dV/dt vs. V) of six consecutive APs from WT (black traces), TS2‐neo MCCs (red traces) and TS2‐neo + 0.3 μM nifedipine (blue traces). Right: maximum values of dV/dt attained in the phase‐plane plot calculated on the first AP. TS2‐neo nifedipine‐treated MCCs (n = 16 in 7 cultures) have values comparable to WT cells (n = 10 cells in 5 cultures) and significantly higher than TS2‐neo untreated cells (n = 7 cells in 5 cultures, ^** P < 0.001; one‐way ANOVA followed by Holm–Sidak post hoc test). B, mean values of Na_V peak currents (at −10 or 0 mV), membrane capacitance and density of currents in TS2‐neo and TS2‐neo + 0.3 μM nifedipine MCCs. The DHP treatment partially recovers the Na_V channel current amplitude and increases by ∼30% the current density (n = 13 TS2‐neo cells in 5 cultures and n = 11 TS2‐neo + nife cells in 5 cultures). C, normalized mean Na_V currents evoked by depolarizing pulses of 10 ms to −10 mV from −70 mF. Na_V currents were recorded and averaged from n = 4 from TS2‐neo cells (red trace) and n = 4 TS2‐neo + 0.3 μM nifedipine cells (blue trace), and fully blocked by 300 nM TTX. D, normalized I–V relationship of Na_V current measured at V _h = −70 mV for TS2‐neo (n = 13 cells in 5 cultures; red traces) and TS2‐neo + 0.3 μM nifedipine MCCs (n = 13 cells in 5 cultures; blue traces). E, mean normalized values for steady‐state Na_V channel inactivation for TS2‐neo (n = 13 cells in 5 cultures; red traces) and TS2‐neo + 0.3 μM nifedipine MCCs (n = 13 cells in 5 cultures; blue traces). F, representative traces of depolarization‐evoked secretion in TS2‐neo (red traces, left) and TS2‐neo + 0.3 μM nifedipine MCCs (blue traces, right). The capacitance increase (∆C, top) and the Ca²⁺ current (bottom) recorded during a step depolarization to +10 mV (upper panel) or 0 mV (lower panel) are visible. G, mean capacitance changes ∆C (left) and mean quantity of charge Q (right) associated to the depolarizing pulse measured in TS2‐neo (n = 16 cells in 7 cultures, red bars) and TS2‐neo + 0.3 μM nifedipine MCCs (n = 9 cells in 6 cultures, blue bars) at +10 mV and measured in TS2‐neo (n = 13 cells in 7 cultures, red bars) and TS2‐neo + 0.3 μM nifedipine MCCs (n = 9 cells in 6 cultures, blue bars) at 0 mV. ∆C is recovered at both voltages in mutated MCCs treated with 0.3 μM nifedipine with respect to untreated MCCs (^** P < 0.01, one‐way ANOVA followed by Holm–Sidak post hoc test). H, Ca²⁺ dependence of the exocytosis determined by plotting ∆C (fF) vs. the corresponding Ca²⁺‐charge density Q (pC/pF). The slope of the linear regression for TS2‐neo (open circles, red line, n = 16 cells; R ² = 0.51) is not statistically different from that of the TS2‐neo + 0.3 μM nifedipine (see text) (filled triangles, blue line, n = 9 cells, R ² = 0.88; P > 0.05, unpaired t test).

Chronic treatment with nifedipine recovers also the secretory activity of MCCs measured through membrane capacitance changes (∆C). In Fig. 15 F are shown typical recordings of ∆C following 100 ms step depolarization to +10 and 0 mV, highlighting how overnight incubation with 0.3 μM nifedipine rescues the secretory activity of TS2‐neo mutated cells. After chronic treatment and washout of the DHP, on average ∆C increased from 13.0 (n = 16) to 28.9 fF (n = 9; P < 0.01) at + 10 mV and from 10.8 fF (n = 13) to 20.2 mV at 0 mV (n = 9; P < 0.01) (Fig. 15 G) while the quantity of Ca²⁺ charges (Q) remained nearly unchanged and comparable to those in WT cells. Thus, it seems evident that chronic nifedipine application helps recruit catecholamine secretion of TS2‐mutated cells induced by the altered Cav1.2 channel gating. As for WT and TS2‐neo mutated cells, we plotted ∆C vs. the Ca²⁺ charge density to estimate the Ca²⁺ dependence of the secretory response (Fig. 15 H). The slope of the linear Ca²⁺ dependence of secretion was 3.0 ± 0.8 fF/(pC/pF) in TS2‐neo MCCs (red line, n = 16 cells) and increased to 9.2 ± 1.2 fF/(pC/pF) in nifedipine‐treated TS2‐neo cells (blue line, n = 9 cells). As for the data of Fig. 11 C, the two slope values were not significantly different due to the large scattering of data around the linear regressions (P > 0.05), suggesting caution in drawing quantitative conclusions concerning effective changes of the Ca²⁺ dependence of secretion.

Discussion

We provided evidence that the TS2‐neo mouse (Bader et al. 2011) is a suitable murine model for studying the molecular bases of altered MCC function associated with the G406R point mutation of the CACNA1C gene. The mutation reduces L‐type channel inactivation and shifts leftward the voltage dependence of channel activation and inactivation, thus increasing the resting window Ca²⁺ current. These gating changes reduce mitochondrial function, alter cell firing and decrease catecholamine secretion. Chronic application of L‐type channel blockers is sufficient to rescue Na_V channel density, normal spontaneous firing and Ca²⁺‐dependent secretion of chromaffin cells, thus anticipating possible therapeutic treatments for the forms of ASD associated with TS.

The decrease of L‐type channel inactivation is surprisingly small in mutated MCCs: 10–25% of the total Ca²⁺ current after 1 s. This is in good agreement with what is expected, considering that Cav1.2 contributes to about half of the total L‐type Ca²⁺ current (Marcantoni et al. 2010; Vandael et al. 2010) and that exon 8 contributes to 20% of the Cav1.2 channel expression in WT and 50% in heterozygous mutated mice. In heterozygous mice, the expression ratio of mutated vs. WT channels varies between 1:1 and 1:5 (Fig. 2), and thus we expected inactivation changes varying between 10 and 25%. Our data are in relatively good agreement with previous findings showing that in heterologous systems, full expression of the G406R Cav1.2 mutation causes nearly complete loss of VDI (Fig. 5G and H in Splawski et al 2004). Expression of 50% G406R and 50% WT Cav1.2 channels, as occurs in heterozygous mice, would predict only a 50% loss of VDI, and a further 50% reduction due to an equal contribution of Cav1.3 and Cav1.2 channels to total current decreases the change further to 25%. Given the great variability and functional differences between heterologously expressed and native Cav1.2 channels, the reported small change in VDI observed in TS2‐mutated MCCs appears consistent. Concerning the ∼−5 mV leftward shift of L‐type current activation in TS2 mutated MCCs, the shift appears significantly smaller and in line with the ∼−10 mV shift reported for the human G406R‐mutated Cav1.2 channel fully expressed in HEK293 cells (Dick et al. 2016) and the ∼−14 mV shift for the rabbit G436R‐mutated Cav1.2 channel fully expressed in Xenopus laevis oocytes (Raybaud et al. 2006).

Given this, our goal was to find the mechanism of how small changes of L‐type channel inactivation impair MCC firing and secretion. Our conclusion is that L‐type channels contribute to ∼50% of the total Ca²⁺ current and play a critical role in the regulation of MCC firing. Cav1.3 and Cav1.2 regulate the inward pacemaking current that generates spontaneous (Marcantoni et al. 2010; Vandael et al. 2010) or evoked repetitive firing (Vandael et al. 2012). Thus, a small continuous increase (over days) of the resting pacemaking current induced by the mistuned Cav1.2 channel inactivation is sufficient to trigger a number of Ca²⁺‐dependent pathways that affect cell function (Voronina et al. 2015). This includes a decreased density of Na_V channels and an increase of their SSI that resembles the protein kinase‐dependent reduction of Na_V1.2 channel availability induced by neurotransmitters (Carr et al. 2003).

Treatment with nifedipine recovers the altered firing of TS2‐mutated MCCs by rescuing Na_V channel density and steady‐state inactivation

We have shown that the TS2 mutation drastically alters the firing properties of MCCs. Most TS2‐mutated cells fire in bursts while a minority undergo repetitive firing below overshoot (Figs 6 and 7). This is markedly different from the firing modes of most WT cells that exhibit normal tonic firing and a small fraction that undergoes burst firing (Vandael et al. 2015 b). Burst firing in WT‐MCCs occurs during altered physiological conditions (Martinez‐Espinosa et al. 2014; Vandael et al. 2015 a, b ; Guarina et al. 2017; for a review see Lingle et al. 2018), specifically when (1) Na_V channels are steadily inactivated, (2) K⁺‐permeable channels (BK, TASK1‐3) are blocked, (3) inward L‐type currents are increased, or (4) following deletion of the BK‐β₂ subunit that completely abolishes BK channel inactivation. Some of these conditions occur in TS2‐mutated cells. The density of Na_V1 channel is largely reduced and SSI is increased. Consequently, APs are smaller and broader, decreasing the activation of BK and K_V channels. This favours a less complete repolarization and the formation of a plateau phase that sustain AP bursts. TS2‐mutated Cav1.2 channels contribute further to stabilize burst firing. They inactivate less completely and stay more open during the plateau phase because of the increased ‘window’ current. This contributes significantly to sustain AP bursts of 300–400 ms duration because of the Cav1.2 reduced inactivation. A possible cause for the increased burst firing in TS2‐mutated MCCs could derive also from the strong attenuation of BK channel inactivation observed in 6 out of 10 mutated cells (Fig. 9). Removal of BK channel inactivation by deletion of the BK‐β₂ subunit shifts also BK channel activation toward more positive voltages (Martinez‐Espinosa et al. 2014). This attenuates the activation of K_V channels that in turn slows the rising phase, lowers the amplitude and reduces the repolarization phase of APs. All this facilitates burst firing in MCCs. In conclusion, removal of BK channel inactivation could also account for the increased percentage of burst firing in TS2‐neo cells. However, the ratio of TS2‐mutated cells exhibiting non‐inactivating vs. inactivating channels (6:4) does not appear significantly different from WT cells (4:4), making this possibility less likely for TS2‐neo MCCs.

A key observation of our work is that overnight incubations of TS2‐mutated cells with either verapamil or nifedipine are able to rescue the normal firing of WT‐MCCs. Nifedipine at 300 nM concentration rescued the normal tonic firing of MCCs (Fig. 13 C). At lower doses (<100 nM) the DHP application was ineffective while at higher doses (>3 μM) it was toxic for cell survival, most likely due to sustained cell depolarizations induced by either the full block of L‐type channels that may drastically reduce the SK currents contributing to resting potential (Vandael et al. 2012) or the direct unspecific block of other K⁺ channels by the DHP (Liu et al. 2007). The binding of low doses of nifedipine to the inactivated state of L‐type channels (Bean, 1984; Sanguinetti & Kass, 1984; Uehara & Hume, 1985) is favoured by the spontaneous burst firing of mutated MCCs, during which Cav1.2‐mutated channels are maintained at less negative potentials for longer periods. Partial block of L‐type channels by chronic exposure to nifedipine is expected to attenuate for several hours the abnormal ‘window’ current of TS2‐mutated cells and to re‐establish more physiological conditions that favour Na_V channel trafficking (Di Biase et al. 2011), recovery of SSI and return to normal AP firing. These rescuing effects of the brain‐permeant blocker nifedipine on the AP firing of TS2‐mutated cells are of great potential significance. They provide a rationale for experimental therapeutic attempts using clinically available Ca²⁺ channel blockers in autistic individuals harbouring Cav1.2 or Cav1.3 channel mutations.

The attenuated Ca²⁺‐dependent secretion in TS2‐mutated MCCs is rescued by overnight exposure to nifedipine

Catecholamine secretion in MCCs is strictly Ca²⁺ dependent and regulated by voltage‐gated Ca²⁺ channels (Garcia et al. 2006; Mahapatra et al. 2012). All Cav channel types expressed in CCs (L, N, P/Q, R, T) contribute to vesicle fusion and secretion proportionally to their expression density (Marcantoni et al. 2008; Mahapatra et al. 2012). Thus, in MCCs L‐type channels contribute to half of the Ca²⁺ current and are responsible for half of the secreted catecholamines (Carabelli et al. 2003; Marcantoni et al. 2009). The TS2 mutation decreases markedly the amount of Ca²⁺‐induced secretion (∆C), suggesting a reduced number of secretory granules that release their content in mutated MCCs. The reduction is associated with a lower rate of catecholamine release (P _r) rather than with a reduction of the RRP. A rationale for this is that part of the reduced ∆C is due to the swelling induced by the TS2 mutation that reduces the P _r by either reducing the density of Ca²⁺ channels around the secretory sites or by modifying some of the many structural elements that form the complex secretory apparatus regulating the priming, docking and fusion of LDCGs (Jahn & Sudhof, 1999; Dhara et al. 2018). An interesting finding of our study is that overnight treatment with nifedipine helps rescue the reduced vesicle secretion in TS2‐mutated cells, confirming once more that the dysfunction induced by the TS2 mutation originates from the resting influx of Ca²⁺ through Cav1.2 mutated channels. Future works will clarify how this occurs at the molecular level.

The altered neurotransmitter release induced by TS2‐mutated Cav1.2 channels suggests the intriguing possibility that similar changes could occur on brain synapses. Cav1.2 and Cav1.3 do not contribute to the synchronous neurotransmitter release in central synapses (Simms & Zamponi, 2014; Zamponi, 2016) but regulate the Ca²⁺ required for vesicle replenishment (Okita et al. 2000; Fourcaudot et al. 2009; Zorrilla de San Martin et al. 2010; Zhang et al. 2015). Compelling evidence suggests that synaptic neurotransmission shares common molecular elements with the secretory apparatus of CCs (Jahn & Fasshauer, 2012; Neher, 2018). Thus, presynaptic mistuning at brain synapses could be a further cause of ASD associated with Cav1.2 mutations.

Cav1.2 G406R and Cav1.3 missense mutations associated with ASD share similar gating mechanisms and cell signalling

We have shown that TS2‐mutated MCCs exhibit reduced inactivation and leftward shifts of L‐type channel activation and inactivation. Similar effects are shown for the V401L Cav1.3 mutation associated with ASD (Pinggera et al. 2017), suggesting common pathogenic mechanisms of Cav1.2 and Cav1.3 mutations. Mutations G406R and V401L occur at the IS6 helix and reduce channel inactivation and shift leftward the voltage dependence of activation and inactivation. A second Cav1.3 mutation (A749G) also causes autism (Pinggera et al. 2015) but has only partial analogy with G406R. It occurs on the IIS6 helix, exhibits similar leftward shifts of activation and inactivation, but in contrast to G406R enhances Cav1.3 channel inactivation. It is therefore likely that it is not the rate of channel inactivation but the ‘leftward’ shifts of activation and inactivation that is the triggering cause of ASD. This conclusion is also supported by the observation that, despite the G406R and G402S mutations reducing Cav1.2 channel inactivation, only G406R produces ASD (Splawski et al. 2005). Indeed, the G402S mutation causes a ‘positive’ shift of activation and G402S patients are neurologically intact (Frohler et al. 2014; Tallila et al. 2014; Hiippala et al. 2015).

How the negative voltage shift of Cav1 channel activation may alter the excitation/inhibition balance, synaptic plasticity and neuronal development remains to be clarified. One possibility is that Cav1 gain‐of‐function mutations directly alter synapse and dendritic spine stability (Stanika et al. 2016). Alternatively, autistic symptoms may derive from a mistuning of the Cav1.2‐mediated transcription signals associated with Ca²⁺ signals followed by voltage‐dependent conformational changes (V∆C) of Cav1.2 channel opening (Li et al. 2016). In this view, it is of key importance to understand how the leftward shift of activation alters AP firing in central neurons during neurogenesis, when Cav1.2 and Cav1.3 channel expression progressively changes with time (Fabel & Kempermann, 2008; Marschallinger et al. 2015). The negatively shifted activation likely affects the percentage of excitable neurons, the shape and frequency of APs and the type of firing (tonic vs. burst, as in MCCs), with consequent altered synaptogenesis and excitation/inhibition balance of neuronal microcircuits (Ben‐Ari & Spitzer, 2010). If this were the case, a search for a new class of Ca²⁺ channel modulators that shift ‘rightward’ the activation of Cav1.2 L‐type channels would be even more beneficial as therapeutics for ASD than most common DHP antagonists that block unspecifically all Cav1 channels. Further studies will clarify these issues.

Additional information

Competing interests

All authors declare no competing interests.

Author contributions

C.C., D.G. and L.G. performed and analysed electrophysiological experiments. C.S. and M.S. performed electron microscopy. C.R. designed and performed mitochondrial metabolism assays. N.T.H., P.T. and G.H. designed the primers and performed RNA isolation, cDNA synthesis and qRT‐PCR. F.T.B. performed the TUNEL and DAPI assays. P.D., F.B. and E.T. designed the primers and performed mouse genotyping. R.L.R. and G.C.B. conceived and generated the TS2‐neo mouse. E.C. designed the study, analysed electrophysiological experiments and wrote the manuscript with input from all co‐authors. All authors have read and approved the final version of this manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

This work was supported by Telethon Foundation (grant no. GGP15110) to E.C., Austrian Science Fund (grant no. SFB F4415) to G.O. and (grant no. FWF P27809‐B26) to N.H., AIRC (grant no. IG‐20107) to P.D. and (grant no. IG15232) to C.R. and MIUR project ‘Departments of Excellence 2018–2022’ to M.S.

Biographies

Chiara Calorio received her Master Degree in Pharmacy in 2010 from the University of Turin. In January 2016 she obtained a PhD in Neuroscience at the University of Torino under the supervision of Prof. Emilio Carbone. Her thesis was focused on T‐type calcium channel expression and function in dorsal root ganglion central neurons and in chromaffin cells under hypoxic conditions, highlighting their crucial role in neuropathic pain and Ca²⁺ evoked catecholamine secretion. She is currently a post‐doctoral fellow at the University of Torino focused on the role of L‐type calcium channels in the development of several intellectual disabilities and autistic spectrum disorders.

Daniela Gavello received her Master Degree in Medical Biotechnology from the University of Torino in 2007. After graduation, she obtained a PhD in Neuroscience at the University of Torino under the supervision of Profs V. Carabelli and E. Carbone. The dissertation focused on the effects of leptin on hippocampal neurons and mouse chromaffin cell function. She is currently an electrophysiologist working as a post‐doctoral fellow at the University of Torino. Her present main research goals are the role of calcium channels in excitability and secretory processes in central neurons and mouse chromaffin cells during neurodegenerative diseases (autism and intellectual disabilities).