All experiments were performed in accordance with the NIH guideline to the care and use of laboratory animals (Publication NO.85-23, revised 1985) and approved by the Animal Research Advisory Committee of Beijing Institute of Biological Science. The procedures were similar to those we described previously (Zhao et al., 2016). In brief, male Sprague Dawley rats (14–16 days old) were decapitated. The brain was rapidly removed and submerged in oxygenated sucrose solution at 4°C containing (in mM): 2.5 KCl, 1.25 NaH₂PO₄, 24 NaHCO₃, 10 glucose, 10 MgSO₄, 0.5 CaCl₂, 2 sodium pyruvate, 230 sucrose (295–305 mOsm/l). The value of pH was adjusted to 7.4 with NaOH. Coronal striatal slices (300 μm) were prepared with a vibratome (MA 752, Campden instruments). Slices were then transferred into a chamber (BSC-PC, Warner instruments) continuously bubbled with 95% O₂ and 5% CO₂ gas mixture for 30 min at 32°C. The solution incubating the slices was the standard NaHCO₃-buffered saline solution containing (in mM): 126 NaCl, 2.5 KCl, 1.25 NaH₂PO₄, 26 NaHCO₃, 10 glucose, 2 sodium pyruvate, 2 CaCl₂, 2 MgCl₂ (295–305 mOsm/l). pH was adjusted to 7.4 with HCl. The chamber was then maintained at room temperature bubbled with O₂/CO₂ gas mixture. Under this condition, the slices could be stored for several hours.

Electrophysiological recording in the striatal ChIs was conducted as described previously (Zhao et al., 2016). Briefly, the pipettes had a resistance of 3–5 MΩ when filled with the internal solution consisted of (in mM): 80 CsOH, 80 gluconate acid, 30 CsCl, 40 HEPES, 10 tetraethylammonium chloride (TEA-Cl), 5 EGTA, 12 Na₂phosphoceatine, 1 MgCl₂, 2 Mg-ATP, 0.5 Na-GTP (265–270 mOsm/l), which was adjusted with CsOH to pH 7.3 (Miki et al., 2013). Slices were bathed in an external solution of (in mM): 105 NaCl, 20 TEA-Cl, 2 CaCl₂, 6 MgCl₂, 6 KCl, 26 NaHCO₃, 10 glucose, 3 myo-inositol, 2 sodium pyruvate, 0.5 ascorbic acid, 1.25 NaH₂PO4, 0.0005 tetrodotoxin (TTX, pH = 7.2 with TEA-OH, Miki et al., 2013). The slice in the recording chamber was visualized with a 40 × water-immersion objective (NIR Apo, Nikon, Japan) using standard infrared and differential interference contrast (IR-DIC) microscopy and a CCD camera. Cells in the striatum approximately 50 μm beneath the slice surface were patched. Electrophysiology was performed using an Axon 200B amplifier (Molecular devices, Foster city, CA, United States) and Clampex 10.1 software (Molecular devices) at room temperature 23 ∼ 25°C (Hawkins et al., 2015). Data were filtered at 2 kHz and digitized at 10 kHz online. Only those recordings with stable holding currents and access resistance were accepted.

The rats of the same age to those used for electrophysiological experiments were used for histochemical experiments. The rats were anesthetized with Nembutal. After perfused with phosphate buffered saline (PBS) and fixed in 4% paraformaldehyde, the brain was removed carefully, cut sagittally at the midline and post-fixed in the same fixative for 3 h at 4°C (Zhao et al., 2016). The brain was sectioned at 5 μm to form the sequential brain slices. The slices containing the striatum were mounted on polylysine-coated slides. The staining was performed by using commercially available antibodies: a rabbit anti-sigma-1 receptor antibody (Abcam, Cat#ab53852, RRID: AB_881796 using at 1:200) and a goat anti-ChAT antibody (Millipore, Cat#AB144P, RRID: AB_2079751 using at 1:500).

Sections were washed with PBS for 3 × 5 min and blocked with normal serum consistent with the host of secondary antibody for 1 h at RT. After that, the serum was removed and primary antibody (diluted in PBS containing 1% BSA) was then applied for overnight at 4°C. Thereafter, sections were rinsed with PBS for 3 min × 5 min and stained with the secondary antibody in blocking serum for 1 h at room temperature in dark. At last, after another three washes, the sections were mounted in medium fluoroshield^TM with DAPI (Sigma–Aldrich F6057, Saint Louis, MO, United States) for visualization. Images were acquired using an Olympus microscope under 40× lens.

ScRT-PCR was performed using protocols previously described (Yan and Surmeier, 1996; Day et al., 2002; Zhao et al., 2016). Individual neuron was aspirated into micropipettes by applying negative pressure while being continuously perfused. After aspiration, the tip of the electrode was broken off and the content was expelled into a 0.5 ml Eppendorf (EP) tube containing 0.5 μl of random primer, 0.25 μl of RNasin (40 U/μl) and 4 μl of DEPC-treated water.

Single-stranded cDNA was acquired by reverse transcription. The neuron-containing mixture was heated to 70°C for 5 min to denature the nucleic acids, and then immediately chilled on ice for at least 5 min. The reverse transcription reaction mixture was added to the reaction tube. The reaction mixture contained primer mixture, GoScript^TM 5× Reaction buffer (4 μl), MgCl₂ (2.4 μl), dNTP (1.0 μl), RNasin (0.25 μl), GoScript^TM Reverse Transcriptase (1.0 μl), nuclease-free water (6.35 μl), and was processed according to the following steps to synthesize the single-strand cDNA: annealing at 25°C for 5 min, extending at 42°C for 60 min, inactivating reverse transcriptase at 70°C for 15 min, and then icing. In the end, to eliminate any residual RNA, 1 μl of RNase H (2 U/μl) was added to the mixture that was incubated at 37°C for 20 min. All the reagents were from Promega Inc. (Madison, WI, United States).

The resulting cDNA from RT was amplified by using a modified protocol from Zhao’s report (Zhao et al., 2016). The amplification was carried out in a thermal cycler (Applied Biosystems). The PCR reaction mixture contained: 10 μl green GoTaq flexi buffer, 2.5 mM MgCl₂, 0.2 mM each dNTP, 0.8 μM primers, 2 μl cDNA template, 125 U GoTaq G2 flexi DNA polymerase, and Nuclease-free water for a final volume of 50 μl. The thermal cycling protocol was set as: 94°C for 3min; and then 94°C for 1 min, 58°C for 1 min, 72°C for 1 min for 43 cycles; and followed by 72°C for 5 min.

The PCR primer sequences for β-actin, GAPDH, and ChAT are described in Yan and Surmeier (1996). The primer sequences of sigma-1 receptor and Ca²⁺ channels in scRT-PCR were reported by Zamanillo et al. (2000) and D’Ascenzo et al. (2004), respectively. The primers were synthesized by Beijing Biomed Inc. All the procedures were performed to minimize the cross-contamination (Mavournin et al., 1990). In each set of single-neuron reactions, negative controls were performed by omitting reverse transcriptase to ensure extraneous and genomic DNA did not contribute to the PCR product. To control for extraneous cDNA, the cellular template was replaced with Nuclease-free water. PCR products were then analyzed by the electrophoresis in 2% agarose gels with ethidium bromide staining. Both controls were consistently negative in these experiments.

Full-length cDNA for human sigma-1 receptor variant 1 was obtained from OriGene (Rockville, MD, United States). The sequence is available under GenBank accession number NM_005866.2. For oocyte expression, sigma-1 receptor was subcloned into the pGH-19 vector modified by BamH I and EcoR I restriction enzymes sites. The orientation of sigma-1 receptor-pGH19 was T7 and the linearizing site for RNA was Xhol I. The Ca_v2.2 (α1b) cDNA used in oocyte study was a gift from Dr. Diane Lipscombe (Lin et al., 1997). GenBank accession number of the α1b, β1b, and α2δ1 cDNA sequence is AF055477, L06110, and AF296488, respectively. The sigma-1 receptor cDNA was also subcloned into the CMV promotor-driven eukaryotic expression vector Dsred-N1 (Clontech) for expression in HEK-293T cells. The construct encoding the full-length GFP-tagged Ca_v2.2 subunit was a generous gift from Dr. Gerald W. Zamponi (Kisilevsky et al., 2008).

Oocytes were harvested from anesthetized Xenopus laevis through a small abdominal incision. The follicular membranes were removed by digesting in OR2 solution (82.0 mM NaCl, 2.5 mM KCL, 1.0 mM MgCl₂, 5.0 mM HEPES, pH 7.6 adjusted with NaOH) containing 1.5 mg/ml collagenase type I (Sigma–Aldrich, St. Louis, MO, United States) for total of 1 h at room temperature. Oocytes were then incubated in ND-96 solution (96.0 mM NaCl, 2.0 mM KCl, 1.0 mM MgCl₂, 1.8 mM CaCl₂, 5.0 mM HEPES, and 2.0 mM sodium pyruvate. pH was adjusted to 7.4 with NaOH) at 18°C for at least 12 h before injection. With a microinjection pipette, 46 nl cRNA was injected into oocytes. The injection was carried out using Drummond nano-injector (Drummond Scientific Co., Broomall, PA, United States). Oocytes were then cultured for 48–96 h before recording. cRNA was synthesized from a T7 promotor using the Ribo MAX^TM Large Scale RNA Production System (Promega, Gaithersburg, MD, United States), which was diluted in nuclease-free water to different concentrations.

After 48 h, currents were recorded using an Axon 2B amplifier (Molecular devices, Foster city, CA, United States) and Clampex 10.1 software (Molecular devices) at room temperature. We recorded the N-type Ca²⁺ current using the following voltage protocol: the holding potential was held at -80 mV and the testing potential was stepped from -80 to +50 mV for 500 ms with an increment of 10 mV. Oocytes were impaled with two electrodes pulled from borosilicate glass pipettes and filled with 3 M KCl. Electrode resistance was 0.5–2.5 MΩ. Recording was performed with oocytes bathing in the external solution containing (in mM): 5 BaCl₂, 50 N-methyl-d-glucamine, 5 KCl, 5 HEPES, and pH was adjusted with methanesulfonic acid to 7.4 (Zhang et al., 2015). Leak and capacitive currents were subtracted using the P/4 procedure.

HEK-293T cells were cultured in Dulecco’s modified Eagle’s medium (DMEM, Invitrogen) containing10% fetal bovine serum (v/v), 50 U/ml penicillin, and 50 μg/ml streptomycin at 37°C supplied with 5% CO₂. Poly-l-lysine coated coverslips were put at the bottom of 35 mm dish. The cells were planted on the coverslips for transfection. The cells were then transfected with the transfection reagent PolyJet (SignaGen) in accordance with the manufacture’s protocol at a 3:1 reagent to DNA ratio (Cassidy et al., 2014). The DNA mixtures (1 μg) used in the transfection consisted of pEGFP-N1-Cacna1b, pcDNA3.1_Cacnb1b, and pcDNA3.1_Cacna2d1 at a ratio of 2:1:1. The medium was replaced 16 h after transfection. Forty-eight hours after transfection, the cells were washed three times in PBS for 5 min each time and mounted them with DAPI. Then fluorescence was observed using confocal microscopy (Olympus).

FRET was measured in HEK-293T cells that were co-transfected with sigma-1 receptor-Dsred and pEGFP-Ca_v2.2 (α1b + β1b + α2δ1) using PerkinElmer Ultra IEWVox. The pEGFP-Ca_v2.2 was the donor and sigma-1 receptor-Dsred was the acceptor. Band limited excitation (420–495 nm) was provided by a laser light and three filters. Cells were imaged using an inverted microscope (Zeiss LSM-880) and an oil-immersion 40× objective lens. Data analysis was done with the Volocity 6.0. The method of nFRET was as shown in Xia and Liu (2001). The equation was as following: NFRET = (FRET-(Acceptor × A)-(Donor × B))/sqrt (Acceptor × Donor).

All processes described below were performed on ice except noted specifically. Sigma-1 receptor-Dsred and pEGFP-Ca_v2.2 expressing cells were washed three times with cold PBS, then lysed with 500 μl RIPA Lysis buffer (CW Biotech) and incubated on ice for 30 min. Lysates were centrifuged at 12000 g for 20 min at 4°C, and the supernatants were removed to another EP tube. Protein concentration of cellular extracts was measured using a BCA assay kit (Thermo Scientific). After that, 500 μg of supernatants were mixed with 1 μg of the homolog antibody: the anti-sigma-1 receptor antibody (Abcam, Cat#ab53852, RRID: AB_881796), the anti-Ca_v2.2 antibody (Millipore, Cat#AB5154, RRID: AB_2069093), or 1 μg IgG (Sigma–Aldrich, Cat#I8140, RRID: AB_1163661), and rotated at 4°C for 4 h. Then, 20 μg agrose A beads (Vigorous Biotechnology) were added to each sample that was rotated continuously at 4°C for overnight. Thereafter, samples were centrifuged at 3000 rpm for 5 min at 4°C, and then the supernatants were removed. Immunoprecipitants were washed twice with 1 ml of 1× lysis buffer, then once with 1 ml of PBS for 5 min at 3000 rpm each time. Both washing buffers contained 10 μl of protease inhibitor. After wash, samples were boiled in 60 μl of 2 × SDS loading buffer at 95°C for 10 min. Proteins were probed with anti-sigma-1 receptor antibody (Proteintech, Cat#15168-1-AP, RRID: AB_2301712 using at 1:500) and anti-Ca_v2.2 antibody (Proteintech, Cat#19681-1-AP, RRID: AB_10638918 using at 1:500) separately. Bands were visualized with an ECL (Gene Co. Ltd) (Kourrich et al., 2013; Yan et al., 2014).

Chemicals were obtained from Sigma–Aldrich (St. Louis, MO, United States), TCI (Shanghai, China), TOCRIS bioscience (Bristol, United Kingdom). SKF-10047, BD-1063, PRE-084, and NE-100 were dissolved in nuclease-free water to form a 10 mM stock solution and further diluted in the bathing solution for the final concentration that were perfused at a rate of 2 ml/min by a peristaltic pump. Data analysis was performed with softwares including Clampfit Version 10.2, Prism Version 6.0 and Origin Version 9.0. The significance of the fit parameters (Mean ± SEM) was tested using student’s t-test, one-way ANOVA, and two-way ANOVA. Difference of p < 0.05 was considered statistically significant.

To observe a possible interaction between sigma-1 receptors and N-type Ca²⁺ channels, we had to use a specific cell that expressed both of these proteins. It was reported that N-type Ca²⁺ channel was the dominant one among all the voltage-gated Ca²⁺ channels expressed in ChIs (Yan and Surmeier, 1996). However, the expression of sigma-1 receptor in ChIs is not reported until now. Therefore, the expression of sigma-1 receptors and N-type Ca²⁺ channels in ChIs should be clarified firstly.

Cholinergic interneurons were picked out by their large and specific shape in rat striatal slices (Figure 1A). scRT-PCR was performed by sucking the cellular content without nucleus into the recording pipette. PCR products were separated by the electrophoreses in 2% agarose gels, stained with ethidium bromide and visualized by UV light. In Figure 1B, the positive ChAT lane revealed that the tested cell was ChI neuron, and sigma-1 receptors, Ca_vα1A and α1B were observed in the same neuron. For analysis of calcium channel expression, the same experiment was performed on 38 ChIs and the summarized result was plotted in Figure 1C. Briefly, three types of voltage-gated Ca²⁺ channels were found with different detection rates of 78.9% (α1A, P/Q-type), 92.1% (α1B, N-type), and 28.9% (α1C, L-type), respectively. It was obvious that N-type Ca²⁺ channel was the prominent isoform in all the high-voltage gated Ca²⁺ channels in ChIs.

Meanwhile, we detected the expression of sigma-1 receptor on 30 neurons which are included in the 38 neurons. The mRNA of sigma-1 receptor was found in every tested ChI neuron (Figure 1C). There were about 90% (27/30) ChIs that co-expressed the sigma-1 receptors and N-type Ca²⁺ channels. The protein of sigma-1 receptor in ChIs was further verified by using immunofluorescence staining on sequential brain slices (5 μm). Two adjacent slices were separately incubated with goat anti-ChAT and rabbit anti-Sigma-1 receptor antibodies. Red immunofluorescence of ChAT staining, as illustrated in Figure 1D, showed the morphology and the localization of ChIs, while the green staining represented sigma-1 receptor. When two slices were merged, colocalization of both ChAT and sigma-1 receptor was clearly observed in several ChIs (as indicated by the arrows). In combination with scRT-PCR, these results proved that sigma-1 receptors were also expressed in ChIs.

To investigate whether the sigma-1 receptor can regulate the N-type Ca²⁺ channel, the Ca²⁺ current of ChIs in the striatal slice was first observed using the whole-cell recording technology. Ca²⁺ currents were evoked by a stimulating protocol as following: the holding potential was -80 mV and the testing potential was stepped from -80 mV to +20 mV for 500 ms with an increment of 10 mV. Recorded Ca²⁺ current was shown in Figure 2A and its related I–V curve was plotted in Figure 2B. When a specific N-type Ca²⁺ channel blocker, ω–conotoxin-GVIA (CTX, 0.5 μM), was applied, the current was inhibited by 72.3% (n = 3, p < 0.05; Figures 2C,F), which further confirmed that N-type Ca²⁺ channels were predominant voltage-gated Ca²⁺ channels in rat striatal ChIs.

The modulation of sigma-1 receptors on N-type Ca²⁺ channels was observed by using sigma-1 receptor agonists of SKF-10047 and Pre-084. When SKF-10047 was added to the bath solution, the amplitude of Ca²⁺ current was inhibited by 28.4% ± 4.9% (n = 9, p < 0.05) and 38.7% ± 7.1% (n = 11, p < 0.05) at the concentration of 25 and 50 μM, respectively (Figures 2D,F). Pre-084 (50 μM) also depressed the currents by 27.1% ± 6.4% (n = 7, p < 0.05) (Figures 2E,F). BD-1063, a selective antagonist of sigma-1 receptor, did not display any visible effect on Ca²⁺ currents. However, the inhibition of SKF-10047 on Ca²⁺ currents was abolished in the presence of BD-1063 (Figure 2G).

In order to investigate the possible constitutive interaction between sigma-1 receptors and N-type Ca²⁺ channels, the two proteins were co-expressed in Xenopus oocytes. N-type Ca²⁺ channel cRNA contained α1b, β1b, and α2δ1 in a 2:1:1 ratio. No visible currents were evoked in native oocytes. However, big inward currents were recorded after cRNA of N-type Ca²⁺ channel was injected into Xenopus oocytes after 48 h and CTX (0.2 μM) blocked the currents completely (Figures 3A,B), which indicated that functional N-type Ca²⁺ channels were successfully expressed in oocytes. After sigma-1 receptor cRNA was injected into oocytes for 48 h, Western Blot was performed. The schematic Western blot strap in the left lane showed the sigma-1 receptor expressed successfully in oocytes (Figure 3C).

To observe the possible interaction between sigma-1 receptors and N-type Ca²⁺ channels, a mixture of their cRNA was injected into oocytes with a ratio of 1:1. Surprisingly, no visible currents were induced by a depolarized stimulation. To clarify a possible chaperone-mediated interaction between sigma-1 receptors and N-type Ca²⁺ channels, we decreased the concentration of sigma-1 receptor cRNA and mix it with the constant concentration of N-type Ca²⁺ channel cRNA at the ratio of 0.5:1, 0.25:1, 0.125:1, 0.0625:1, and 0.03125:1, respectively. The results showed that the amplitude of N-type Ca²⁺ channel current was increased obviously with decreased sigma-1 receptor cRNA (Figures 4A,B). Compared to the currents recorded in oocytes with only expression of N-type Ca²⁺ channels, the current amplitudes were 29.9% ± 3.9% (n = 18, p < 0.001), 52.6% ± 3.8% (n = 17, p < 0.001), 72.1% ± 3.7% (n = 26, p < 0.001), 85.0% ± 5.8% (n = 17, p < 0.05), and 89.9% ± 7.5% (n = 10, p > 0.05) with the cRNA ratio of 0.5:1, 0.25:1, 0.125:1, 0.0625:1, and 0.03125:1, respectively. However, the expression levels of the two proteins was positively correlated with the amount of cRNA injected, indicating that the expression of each protein was not affected by each other (Figure 4C). These results suggested that sigma-1 receptors probably regulated N-type Ca²⁺ channels constitutively by direct interaction. As a negative control, we studied sigma-1 receptor and another ion channel, HCN1, in the same way as with N-type Ca²⁺ channel in oocytes. We found HCN1 current was not affected by sigma-1 receptor. This experiment indicated us the effect on N-type Ca²⁺ channels was specific (Data was not shown).

To prove if sigma-1 receptor agonists could modulate the Ca²⁺ channel in oocytes the same way as in ChIs, we observed the effect of SKF-10047 on the currents recorded in oocytes in which sigma-1 receptor to N-type Ca²⁺ channel was injected at 0.25:1 ratio (Figure 5A). When SKF-10047 (50 μM) was added into the bath solution, the current was inhibited by 29.7% ± 6.7% comparing with control (Figures 5A–C, n = 6, p < 0.05). BD-1063 (50 μM) alone did not present remarkable inhibition on N-type Ca²⁺ currents (93.2 ± 2.1% of control, Figure 5C, n = 6, p > 0.05), but could block the inhibition of SFK-10047 on N-type Ca²⁺ current (87.2 ± 2.2% of control, Figure 5C, n = 6, p > 0.05). This result indicated that SKF-10047 inhibited N-type Ca²⁺ channel through sigma-1 receptor activation.

As reviewed in Su et al. (2016), we speculated that the inhibition of sigma-1 receptor on N-type Ca²⁺ channels may be mediated by protein–protein interaction. To investigate if there is a protein–protein interaction between receptor and channel, we took HEK-293T cells as an exogenous expressing model. Cultured HEK-293T cells were transiently transfected with the GFP-Ca_v2.2 channels (α1b+β1b+α2δ1subunits) in the absence and presence of sigma-1-Dsred receptor. The cells were observed using confocal microscopy. As illustrated in Figures 6A,B, the red and green fluorescence indicated sigma-1 receptors and N-type Ca²⁺ channels were successfully expressed, respectively. When cells were co-transfected with sigma-1 receptor and N-type Ca²⁺ channel, there is yellow fluorescence in merged channel. This indicated that receptors and channels were co-expressed in the same cell, as shown in Figure 6C. These results allowed us to further study the interaction of the receptors and channels.

To examine a direct interaction between sigma-1 receptors and N-type Ca²⁺ channels, the technology of FRET was performed. Using the co-transfected HEK-293T cells, we chose GFP-Ca_v2.2 as the FRET donor and sigma-1 receptor-Dsred as the acceptor. In the experimental group, we used 488 nm laser-light to stimulate the donor (GFP), which then excites the red fluorescence from FRET channel when both are in close proximity as shown in Figure 7A. Using the Volocity 6.0, we measured the intensity of the signals from nFRET channel. We set the area of signals 0.02 μm² and the intensity ≥0.5 as valid standards. Then we counted the number of signals under this condition. In experimental group, we obtained 203 signals from 11 fields subjected to the condition (Figure 7C). An example including the signals in this field was shown in Figure 7B, where the nFRET value of each signal was 0.81 ± 0.06 (Mean ± SD). However, in the control experiment, where pEGFP vector alone was co-transfected with sigma-1 receptor-Dsred, the number of signals matched the condition was 0 (zero) (also see Figure 7C). This indicated the existence of the receptor-channel complex.

To investigate the receptor-channel physical interaction further, we performed the co-immunoprecipitation. Samples were processed with anti-Ca_v2.2 antibody or anti-sigma-1 receptor antibody. Probing the gel with antibody against N-type Ca²⁺ channel in the sample immunoprecipitated with anti-sigma-1 antibody revealed a band at 260 kDa shown in Figure 7D left panel. On the other hand, probing with antibodies against sigma-1 receptor in the sample immunoprecipitated with anti-N-type Ca²⁺ channel antibody revealed a band at 52 kDa, as shown in Figure 7D right panel. The input of the cell lysates indicated that the receptor and the channel were co-expressed, IgG was used as negative control, as shown in Figure 7D. These results demonstrated the physical association of sigma-1 receptors with N-type Ca²⁺ channels.