Male 6- to 8-week-old C57BL/6 mice (Vital River, China) were used for the studies. All experiments were performed in accordance with the guidelines of the Animal Care and Use Committee of Hebei Medical University and approved by the Animal Ethics Committee of Hebei Medical University.

All experiments were performed in accordance with the guidelines of Animal Care and Use Committee of Hebei Medical University.

The details for preparation of coronal brain sections containing DRN were the same as described in our previous work (Zhao et al., 2017). Briefly, mice were anesthetized with chloral hydrate [200 mg/kg, intraperitoneally (i.p.)]. After intracardial perfusion with an ice-cold sucrose solution (260 mM sucrose, 25 mM NaHCO₃, 2.5 mM KCl, 1.25 mM NaH₂PO₄, 2 mM CaCl₂, 2 mM MgCl₂, and 10 mM D-glucose; osmolarity, 295–305 mOsm; saturated with 95% O₂ and 5% CO₂), the brains of the mice were removed quickly and placed into the slicing solution. Coronal midbrain slices (200 μm thick) containing DRN (AP −3.8 to −4.8 mm; LM 0 mm; and DV −2.8 to −3.8 mm) were sectioned with a vibratome (VT1200S; Leica, Germany). The sections were incubated for 30 min at 36°C in oxygenated artificial cerebrospinal fluid (ACSF) (in mM: 124 NaCl, 3 KCl, 1.25 NaH₂PO₄, 2 CaCl₂, 2 MgCl₂, 25 NaHCO₃, 10 D-glucose; osmolarity, 280–300 mOsm), and stored at room temperature for 90 min (23–25°C) before use.

5-HT neurons located in the midline of the ventromedial subdivisions of the DRN were used. DRN 5-HT neurons were identified by single-cell PCR for the presence of tryptophan hydroxylase (TPH). Recordings in the slices were performed in whole-cell voltage-clamp configurations on a Multiclamp 700B amplifier coupled with a Digidata 1440A AD converter (Molecular Devices, United States) using borosilicate patch electrodes (1–3 MΩ) wrapped with parafilm to reduce pipette capacitance. Pipette series resistance (typically 4–8 MΩ) was compensated by 70–85% during voltage-clamp experiments and was checked frequently throughout the experiment; data were not used if series resistance changed by >15%. Voltage signals were filtered at 10 kHz and sampled at 20 μs using a Digidata 1440A data acquisition interface (Molecular Devices) and pClamp 9 software (Molecular Devices). For recording K⁺ currents, glass electrodes (3–5 MΩ) were filled with the following internal solutions, namely, 115 mM K-methylsulfate, 20 mM KCl, 1 mM MgCl₂, 10 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid (HEPES), 0.1 mM EGTA, 2 mM MgATP, and 0.3 mM Na₂GTP, pH adjusted to 7.4 with KOH. For recording I_M and SK currents, ACSF was used as extracellular solution. For recording I_A current, HEPES-buffered ACSF (130 mM NaCl, 4 mM KCl, 2 mM CaCl₂, 2 mM MgCl₂, 10 mM HEPES, 10 mM D-glucose; 280–300 mOsm) was used as extracellular solution. To optimally isolate the outward SK currents from other K⁺ currents and the Na⁺ currents, 5 mM tetraethylammonium (TEA) and 1 μM tetrodotoxin were added to the extracellular solution in voltage-clamp experiments (Sailer et al., 2002; Pedarzani et al., 2005). For isolating A-type currents (I_A), 1 μM tetrodotoxin was used to block fast voltage-activated Na⁺ channels, and 0.3 μM CdCl₂ was used to block voltage-activated Ca²⁺ channels (Itri et al., 2010; Hu et al., 2019).

For recording spontaneous firing of the neurons, cell-attached “loose-patch” (100–300 MΩ) recordings were used (Burlet et al., 2002). In this case, patch pipettes (2–4 MΩ) were filled with ACSF, and the spontaneous activity was recorded in the current-clamp mode (I = 0). All of the experiments were performed at room temperature (25 ± 2°C). In our recordings, the majority of the recorded neurons, ∼90%, were silent without added PE, and indeed a small number of recorded cells (∼10%, 10 out of 100) had activity of spontaneous firing. However, these neurons with spontaneous firing were found mostly to be either dopaminergic neurons or glutamatergic neurons as verified by the single-cell PCR post the electrophysiological recordings (n = 8, 80% dopaminergic neurons; n = 1, 10% glutamatergic neurons). The dopamine and glutamatergic neurons in DRN were also reported in the literature (Soiza-Reilly and Commons, 2011; Matthews et al., 2016; Commons, 2020). Therefore, neurons exhibiting spontaneous activity were not included for data analysis.

After intracardial perfusion with 4% paraformaldehyde (PFA) in 0.01 M phosphate-buffered saline (PBS) (pH 7.4), followed by 0.01 M PBS, mice brains were post-fixed in 4% PFA at 4°C for 48 h and coronal midbrain slices were prepared. Sections (50 μm thick) were blocked with 0.3% Triton X-100 and 10% donkey serum (Biological Industries, Israel) in PBS and incubated overnight at 4°C with a mixture of primary antibodies. Sections were then washed with PBS three times (5 min) and then incubated in a mixture of secondary antibodies at room temperature for 2 h. Sections were then washed three times (7 min) with PBS. Finally, slices were mounted with Prolong Gold antifade reagent (Life Technologies, United States). Images were obtained on a Leica TCS SP5 confocal laser microscope (Leica, Germany) equipped with laser lines for FITC (Argon 488) and cy3 (HeNe 543). Images were analyzed with LAS-AF-Lite software (Leica, Germany).

PrimeScript™ II 1st Strand cDNA Synthesis Kit (Takara-Clontech, Japan) was used to perform reverse transcription. At the end of electrophysiological recordings, the recorded cell was aspirated into a pipette and then expelled into a sterile PCR tube containing 1 μl oligo-dT primer and 1 μl dNTP mixture. The mixture was heated to 65°C for 5 min and then cooled on ice for 2 min. Synthesis of the first single-strand cDNA from the cellular mRNA was performed with PrimeScript II reverse transcriptase (Takara) at 50°C for 50 min and then 85°C for 5 min. cDNA was stored at −20°C. Then, single-strand cDNA was amplified using GoTaq Green Master Mix (Promega, United States). Two rounds of conventional PCR with pairs of gene-specific outer (first round) and inner primers (second round) for GAPDH (positive control), TPH, Kv4.1–4.3, 3.3, 3.4, Kv1.4, SK1–3, and KCNQ1–5 were performed. After adding the specific outer primer pairs into each PCR tube (final volume 25 μl), first-round synthesis was performed as follows, namely, 95°C (5 min); 30 cycles of 95°C (50 s), 60°C (50 s), 72°C (50 s); 72°C (5 min). Then, 2 μl of the first PCR product were used for the second amplification with specific inner primers (final volume 25 μl). The second-round amplification was performed as follows, namely, 95°C (5 min); 35 cycles of 95°C (50 s), 58–63°C (45 s), 72°C (50 s) and 5 min elongation at 72°C. The final PCR products were separated by electrophoresis on 2% agarose gels. Negative control reactions with no added template were included in each experiment.

The “outer” primers (from 5′ to 3′) were as follows:

The “inner” primers (from 5′ to 3′) were as follows:

All drugs were bath applied at the following concentrations, namely, PE (10 μM; Tocris), prazosin (5 μM; Sigma), apamin (500 nM; Sigma), 4-aminopyridine (4-AP; 4 mM; Sigma), XE991 (3 μM; Tocris), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10 μM; Sigma), DL-2-amino-5-phosphonopentanoic acid (APV; 50 μM; Sigma), strychnine (2 μM; Sigma), and gabazine (10 μM; Sigma).

Commercial antibodies used were anti-TPH (1:400, mouse, sigma, T0678, RRID:AB_261587), anti-Kv4.2 (1:400, rabbit, AlomoneLabs, APC-023, RRID:AB_2040176), anti-Kv4.3 (1:400, rabbit, AlomoneLabs, APC-017, RRID:AB_2040178), anti-SK2 (1:200, rabbit, Bioss, DF13499, RRID:AB_2846518), anti-SK3 (1:200, rabbit, Proteintech, 17188-1-AP), anti-KCNQ2 (1:200, goat, Santa Cruz, sc-7793, RRID:AB_2296585), anti-KCNQ3 (1:200, goat, Santa Cruz, sc-7794, RRID:AB_2131714), and anti-KCNQ4 (1:200, rabbit, AlomoneLabs, APC-164, RRID:AB_2341042). Secondary antibodies used were donkey anti-mouse IgG (H + L) highly cross-adsorbed secondary antibody (Alexa Fluor 568, Thermo Fisher Scientific, A10037, RRID:AB_2534013, 1:1,000), donkey anti-rabbit IgG (H + L) highly cross-adsorbed secondary antibody (Alexa Fluor 488, Thermo Fisher Scientific, A-21206, RRID:AB_2535792, 1:1,000), and donkey anti-goat IgG (H + L) cross-adsorbed secondary antibody (Alexa Fluor 488, Thermo Fisher Scientific, A-11055, RRID:AB_2534102, 1:1,000).

All data are expressed as mean ± SEM. Group size (n) indicates the number of independent, non-technical replicates. For electrophysiological data, the discharge rate and the current amplitudes were compared using the paired t-test, when data were normally distributed and there was no significant variance inhomogeneity. When normality or equal variance of samples was not present, the Wilcoxon matched-pairs signed-rank test was used. p-Values ≤ 0.05 were accepted as statistically significant. Data analysis was carried out using GraphPad Prism 6.0 (RRID:SCR_002798).

In a previous study, A-type currents (I_A) were found to be inhibited by PE in DRN 5-HT neurons. However, it was not tested whether this inhibition contributes to the PE-induced firing of these neurons (Aghajanian, 1985). Moreover, the expression profiles of A-type channel in DRN 5-HT neurons have not been investigated. Therefore, we first examined the subtypes of A-type K⁺ channels expressed in DRN 5-HT neurons using single-cell PCR and immunofluorescence analysis. Single-cell PCR results revealed a strong expression of Kv4.2 and 4.3, a weak expression of Kv4.1, and no detectable expression of Kv1.4, Kv3.3, and Kv3.4 (Figures 2Ai,ii). Expression of Kv4.2 and Kv4.3 proteins in DRN 5-HT neurons was also confirmed by immunofluorescence, which showed strong signals for these channel proteins (Figure 2B), consistent with the results of single-cell PCR. These results suggest that Kv4.2 and 4.3 are the dominant A-type K⁺ channels in DRN 5-HT neurons and mediate the majority of I_A. Next, we investigated the role of these A-type K⁺ channels in PE-induced firing of 5-HT neurons. Synaptic blockers (CNQX, APV, and gabazine) were added to isolate the intrinsic firing properties and I_A were recorded using the protocol shown in Figure 2Ci; the cells were voltage-clamped at −70 mV, followed by a hyperpolarizing step to −100 mV (200 ms), and then a step depolarization to −20 mV (300 ms). I_A were isolated as characteristic transient currents with rapid activation and inactivation, measured as instantaneous currents at the beginning of the −20 mV step. Bath application of PE (10 μM) significantly reduced peak I_A currents (from 0.77 ± 0.05 to 0.37 ± 0.04 nA, n = 6, p < 0.0001, paired t-test), and this reduction was significantly reversed by prazosin (5 μM; Figure 2C), the antagonist of α₁-adrenoceptors. This result, in combination with the results shown in Figure 1, suggests that inhibition of I_A currents contributes to the PE-induced spontaneous firing of DRN 5-HT neurons. To confirm this, we tested the effect of A-type channel blocker, 4-AP. 4-AP at maximal I_A inhibiting concentration (4 mM) (Yao and Tseng, 1994; Serôdio et al., 1996; Song et al., 1998) also induced spontaneous firing in DRN 5-HT neurons, suggesting that inhibition of I_A indeed triggers spontaneous firing. However, in the continued presence of 4-AP, PE (10 μM) further increased the firing frequency from 0.51 ± 0.06 to 1.23 ± 0.07 Hz (n = 13, p < 0.0001, paired t-test, Figure 2D) in a statistically significant manner, indicating that another mechanism besides I_A inhibition was involved in the PE-induced spontaneous firing of DRN 5-HT neurons.

Our previous studies have shown that the KCNQ4 channel is the predominant Kv7/KCNQ isoform expressed in DRN 5-HT neurons (Zhao et al., 2017), although other neuronal KCNQ members (KCNQ2, KCNQ3, and KCNQ5) in these neurons have not been studied. In this study, the results of single-cell PCR analysis revealed robust expression of KCNQ2, KCNQ3, and KCNQ4 mRNA in DRN 5-HT neurons (Figures 3Ai,ii). Consistently, the result of immunofluorescence analysis showed high expression levels of the KCNQ2, KCNQ3, and KCNQ4 proteins in DRN 5-HT neurons (Figure 3B). To correlate the PE-induced spontaneous firing with its modulation of KCNQ/M-type currents, we first examined whether PE could inhibit KCNQ/M-type currents in DRN 5-HT neurons. M-type currents were measured using the protocol shown in Figure 3Ci as characteristic slow deactivating tail currents at a −50 mV step from a depolarized potential of −20 mV (Zhao et al., 2017). As shown in Figure 3C, PE (10 μM) significantly inhibited the M-type currents from 74.67 ± 9.46 to 40.63 ± 5.69 pA (n = 7, p < 0.01, paired t-test). Subsequently, it was found that XE991, a selective KCNQ blocker, failed to further inhibit the M-type currents, indicating complete inhibition of this K⁺ conductance by PE. Moreover, inhibition of M-type currents by XE991 resulted in depolarization of the resting membrane potential (from −63.91 ± 2.21 to −57.87 ± 1.83 mV, n = 6, p < 0.05, paired t-test, Figure 3E). With continued presence of XE991, PE (10 μM) further depolarized in a significant manner the resting membrane potential to −52.27 ± 2.47 mV. These results suggest that PE-induced inhibition of M-type current might trigger the neuronal firing. Next, we showed that blocking M-type current by addition of XE991 (3 μM) produced spontaneous firing of DRN 5-HT neurons (0.42 ± 0.06 Hz, n = 14, p < 0.001, Wilcoxon matched-pairs signed-rank test, Figure 3D). These results indicate that inhibition of KCNQ/M-type currents contributes to the PE-induced spontaneous firing of the DRN 5-HT neurons. Consistent with the involvement of multiple K⁺ channels in the PE-induced spontaneous firing of the DRN 5-HT neurons, the XE991-induced firing rate was further increased when PE was applied (1.30 ± 0.12 Hz, n = 14, p < 0.0001, Wilcoxon matched-pairs signed-rank test, Figure 3D).

Ca²⁺-activated K⁺ (SK) channels have been shown to regulate the firing pattern of central neurons, including DRN 5-HT neurons (Pan et al., 1994; Wagner et al., 2001; Adelman et al., 2012; Gocho et al., 2013; Matschke et al., 2018). However, conflicting results have been reported regarding the role of SK channels in PE-induced firing of DRN 5-HT neurons (Pan et al., 1994; Maingret et al., 2008). Moreover, the molecular identities of the SK currents in DRN 5-HT neurons are not clear. Three isoforms of SK channels (SK1, SK2, and SK3) have been described (Adelman et al., 2012). We first examined the expression profiles of these SK channels in DRN 5-HT neurons using single-cell PCR (Figure 4A) and immunofluorescence approaches (Figure 4B). We observed strong expression of SK2 and SK3 channels in DRN 5-HT neurons at both mRNA and protein levels, suggesting that SK currents in DRN 5-HT neurons are mediated by these SK channels. It has been suggested that SK currents are primarily involved in slow afterhyperpolarization (sAHP) during action potential firing (Pan et al., 1994; Wagner et al., 2001; Adelman et al., 2012; Gocho et al., 2013; Matschke et al., 2018). We isolated the AHP outward currents (I_AHP) encoded by SK channels using a one-step voltage-clamp protocol (Matschke et al., 2018), the tail currents measured at the beginning of −60 mV following a depolarizing potential of 0 mV (Figure 4Ci). PE (10 μM) significantly inhibited the I_AHP in DRN 5-HT neurons, from initial current amplitudes of 44.04 ± 6.64 to 25.18 ± 6.28 pA (n = 6, p < 0.01, paired t-test, Figure 4C). It appears that PE only partially inhibited SK currents because apamin, a selective SK channel blocker, had a stronger inhibition on SK currents when applied either after (Figures 4Ci,ii,iii) or before (Figures 4Di,ii,iii) of PE. However, even with maximal inhibition of SK currents, apamin did not induce significant, sustained spontaneous firing of DRN 5-HT neurons (only a transient increase was occasionally observed, e.g., Figure 4Eii), although subsequent application of PE elicited firing of these neurons (Figure 4E). These results suggest that inhibition of SK channels does not trigger spontaneous firing of DRN 5-HT neurons.

While blocking SK channels with apamin did not elicit as strong firing activity as blocking A-type and KCNQ/M channels, apamin triggered transient, sparse firing activity in some DRN 5-HT neurons (see, e.g., Figure 4Eii). This prompted us to further test the effect of apamin. We first induced firing of DRN 5-HT neurons using both XE991 and 4-AP to block A-type and KCNQ/M K⁺ currents, and then additionally applied apamin. As shown in Figure 5A, the firing rate was further increased after apamin addition (from 0.87 ± 0.05 to 1.30 ± 0.07 Hz, n = 12, p < 0.0001, paired t-test, Figure 5A). Interestingly, PE also induced further firing activity when administered in addition to XE991 and 4-AP (from 0.96 ± 0.13 to 1.54 ± 0.20 Hz, n = 12, p < 0.001, paired t-test, Figure 5B), likely due to inhibition of SK channels. Taken together, these results imply that inhibition of SK currents, although not directly triggering spontaneous firing of DRN 5-HT neurons, contributed to the PE-induced firing activity of DRN 5-HT neurons.

Finally, to prove that the K⁺ conductance of A-type, KCNQ/M, and SK channels are sufficient components for the PE-induced spontaneous firing of DRN 5-HT neurons, a cocktail of the blockers for these K⁺ channels (4-AP, XE991, and apamin) was tested. The blocker cocktail evoked spontaneous firing of DRN 5-HT neurons (1.34 ± 0.39 Hz, n = 9, Figure 5C), which was not further enhanced by subsequent addition of PE (1.47 ± 0.32 Hz, n = 9, p > 0.05, paired t-test). These results suggest that blocking K⁺ channels (A-type, KCNQ/M, and SK currents) is a sufficient mechanism to trigger PE-induced spontaneous firing of DRN 5-HT neurons.