Introduction

Gonadotropin-releasing hormone (GnRH) neurons play a pivotal role in the neuroendocrine regulation of reproduction. Thus, the question arises as to how cell excitability of GnRH neurons is regulated. Both intrinsic and extrinsic mechanisms are involved in determining the membrane potential and firing pattern. We have previously investigated the intrinsic properties of rat GnRH neurons, and revealed the presence of all five types of voltage-gated Ca2+ currents [1, 2] and two types of Ca2+-activated K+ currents, namely SK and BK currents [2, 3]. In the present study, we further investigated the voltage-gated Ca2+ channels by means of their molecular biology and electrophysiology.

Voltage-gated Ca2+ channels are involved in the control of cell excitability and also mediate Ca2+ influx, thereby regulating Ca2+-dependent cellular processes such as contraction, secretion, and gene expression [4, 5]. The voltage-gated Ca2+ channels are classified into low-voltage-activated Ca2+ channels (T-type channels) and high-voltage-activated Ca2+ channels (L, N, P/Q, and R-type channels). Rat GnRH neurons possess all five types of voltage-gated Ca2+ currents [1, 2]. One characteristic is that R-type Ca2+ currents contribute more than 60% to the total Ca2+ current in neonatal rat GnRH neurons and more than 40% in adult rat GnRH neurons, suggesting that generation of Ca2+ spikes is mainly mediated through R-type Ca2+ channels in the soma and dendrites [6]. Two reports have been published on the Ca2+ currents in mouse GnRH neurons. In one report, the proportions of R-type Ca2+ currents are 8–15% in mouse GnRH neurons and the proportions of T-type Ca2+ currents are negligible [7]. However, the other report suggested the presence of T-type Ca2+ currents in mouse GnRH neurons [8]. Recently, Zhang et al. [9] reported that mouse GnRH neurons express all three α1 subunits for T-type Ca2+ channels and that their expression level is modulated by 17β-estradiol (E2). These reports led us to investigate which subtypes of T-type Ca2+ channels are expressed in rat GnRH neurons. Three subtypes have been identified, namely Cav3.1 (α1G), Cav3.2 (α1H), and Cav3.3 (α1I) channels [4, 10–12]. Each subtype exhibits distinct kinetics and Ni2+ sensitivity [4, 13]. The striking differences among these three subtypes are that the recovery time course from inactivation is rapid in α1G and slower in α1H and α1I, and that inactivation is rapid in α1G and α1H but slower in α1I [4, 14, 15]. Ni2+ sensitivity is high (IC50, 12 μM) in α1H and low (IC50 > 200 μM) in α1G and α1I [4, 13]. Consequently, expression of the different subtypes must affect cell excitability differently.

In this study, we investigated mRNA expression for the voltage-gated Ca2+ channels using multi-cell RT-PCR and performed electrophysiological analyses of the T-type Ca2+ currents in rat GnRH neurons in acute slice preparations. We found that rat GnRH neurons expressed mRNAs for all five types of voltage-gated Ca2+ channels, and that α1H was dominantly expressed for the T-type Ca2+ channel. There were no sex differences in expression of the mRNAs and no difference with regard to estrous cycle stage for the dominance of α1H expression.

A preliminary report of some of our findings has been published previously in abstract form [16].

Materials and methods

Animals

All experiments were performed after obtaining approval from the Nippon Medical School Animal Care Committee. Transgenic rats expressing enhanced green fluorescent protein (EGFP) under the control of the GnRH promoter [1] were used. The rats had free access to water and chow, and were maintained under a 14-h light/10-h dark cycle. Rats aged 2–3 months (adult) were used for electrophysiological experiments and multi-cell RT-PCR analyses.

Preparation of slices

Coronal brain slices (200-μm thick) containing the medial septum, diagonal band of Broca (DBB), organum vasculosum of the lamina terminalis (OVLT), and medial preoptic area (mPOA) were prepared from male and female rats. Rats were decapitated under ether anesthesia, and their brains were quickly removed and immersed in an ice-cold oxygenated (100% O2) cutting solution comprising (mM) 2.5 KCl, 1.25 Na2HPO4, 0.6 NaHCO3, 0.5 CaCl2, 7 MgCl2, 10 HEPES, 7 glucose, 248 sucrose, 1.3 ascorbic acid, and 3 Na pyruvate (pH 7.4, 290 mOsm). The brains were cut into blocks, glued with cyanoacrylate to the chilled stage of a Vibratome VIB3000 (Vibratome, St Louis, MO, USA) and sliced. The slices were transferred to and maintained in a chamber containing an oxygenated artificial cerebrospinal fluid (ACSF) consisting of (mM) 137.5 NaCl, 2.5 KCl, 1.25 Na2HPO4, 0.6 NaHCO3, 10 HEPES, 2 CaCl2, 2 MgCl2, and 10 glucose (pH 7.4) at room temperature for 30 min or more.

Preparation of dissociated cells

The brains were excised from rats under ether anesthesia in the afternoon (12:00). The medial septum, DBB, OVLT, and mPOA were cut out with razors and surgical blades. The sections were minced and treated with papain (21U/ml; Funakoshi, Tokyo, Japan) for 50 min at 30°C with Dulbecco’s modified Eagle’s medium (Sigma, St Louis, MO, USA). Tissues were triturated with a 5-ml plastic pipette after several washes with modified Eagle’s medium (Invitrogen, Grand Island, NY, USA). Cell suspensions were applied to a discontinuous Percoll (Amersham Biosciences, Uppsala, Sweden) density gradient composed of 1.0, 1.023, and 1.078 g/ml layers, and centrifuged. The cells were obtained from the middle layer and EGFP-positive (GnRH) cells were harvested with a glass suction pipette for RT-PCR analysis. For electrophysiological experiments, the cells were plated on poly-lysine-coated coverslips and incubated overnight in neurobasal-A medium (Invitrogen, Grand Island, NY, USA) supplemented with 0.5 mM l-glutamine, B-27 (Invitrogen), and 5 ng/ml basic FGF (Invitrogen) at 37°C. The estrous cycle stage was determined on the day of sacrifice.

Electrophysiology

A List EPC-9 patch-clamp system (Heka Electronic, Lambrecht/Pfalz, Germany) was used for the recordings and data analyses. T-type Ca2+ currents were recorded by means of the whole-cell patch clamp configuration at 32°C. Each slice was transferred to the recording chamber, kept submerged, and continuously superfused with oxygenated ACSF at a rate of 3 ml/min. The slice was viewed under an upright fluorescence microscope (BX50; Olympus, Tokyo, Japan). For whole-cell recordings, pipettes were fabricated from borosilicate glass capillaries and had a resistance of 4–5 MΩ. The pipette solution consisted of (mM) 130 Cs-gluconate, 10 EGTA, 10 HEPES, 0.3 MgCl2, and 2 Mg-ATP (pH 7.3, 280 mOsm). The T-type Ca2+ currents were recorded using TEA/Ca2+ ACSF consisting of (mM) 2 NaCl, 103 tetraethylammonium chloride (TEA-Cl), 5 CsCl, 10 CaCl2, 0.8 MgCl2, 10 glucose, 20 HEPES, 0.6 NaHCO3, and 10 4-aminopyridine (4AP) (pH 7.4, 290 mOsm). Under these recording conditions nearly all Na+ and K+ currents were eliminated. Thus, the Ca2+ currents were isolated. Positive pressure was applied to the pipette and the pipette was then targeted at the GnRH neuron identified by EGFP fluorescence. After the cell was reached, the positive pressure was removed and negative pressure was applied to seal (seal resistance, >3 GΩ) and break the patch membrane. The currents were filtered at 2.3 kHz, digitized at 10 kHz, and stored. The series resistance was electronically compensated by 70%. Data were acquired when the series resistance was stable and <30 MΩ. The cell capacitance was 17.31 ± 3.95 pF (mean ± SD, n = 190). Capacitative and leak currents were subtracted by the p/n procedure, and the liquid junction potential was not compensated. The holding potential was −50 mV. To measure the activation of T-type Ca2+ currents, 100-ms test pulses (−80 up to −20 mV in 5 or 10-mV increments), preceded by a 2-s prepulse of −100 mV to fully deactivate the T-type channels, were applied at 0.1 Hz. The inactivation time constant was determined by exponential fitting as shown in Fig. 3d. To measure the steady-state inactivation of the T-type Ca2+ currents, 2-s prepulses (−100 up to −40 mV in 5-mV increments) followed by a test pulse (−20 mV) were applied at 0.1 Hz. The data were normalized by the value obtained for the −20 mV pulse for the activation and by the value at −100 mV for the steady-state inactivation. The recovery time course of the T-type Ca2+ currents was determined using a voltage sequence consisting of 2-s depolarizing prepulses to −30 mV followed by a −100-mV pulse of variable length (0.01–2 s) and a test pulse to −20 mV at 0.1 Hz. In some experiments, the T-type channel blocker Ni2+ (100 μM) was used at the end of the recordings to confirm the T-type currents. In current-clamp experiments, ACSF and pipette solution consisting of (mM) 90 K-gluconate, 40 KCl, 10 EGTA, 10 HEPES, 0.3 MgCl2, and 2 Mg-ATP (pH 7.3, 280 mOsm) were used. The membrane potentials were kept at −55 to −60 mV. Rebound action potentials were elicited with 1-s hyperpolarizing current pulse.

Multi-cell RT-PCR

Coronal slices (200-μm thick) were prepared as described above. Each slice was transferred to the recording chamber, kept submerged, and continuously superfused with oxygenated ACSF at a rate of 3 ml/min. The slice was viewed under an upright fluorescence microscope (BX50; Olympus). Pipettes of 1–2 MΩ (inner tip diameter, 4–6 μm) were fabricated from glass capillaries baked at 200°C for 5 h. Each pipette was filled with 5 μl of an autoclaved solution comprising (mM) 150 KCl, 3 MgCl2, 5 EGTA and 10 HEPES (pH 7.2, 270 mOsm). The cytoplasmic contents were harvested from 5 GnRH neurons under visual control, and pooled in a thin-walled PCR tube containing an RNase inhibitor (RNasin Plus; Promega, Madison, WI, USA). In some experiments, GnRH neurons were harvested from acutely-dispersed preparation. The harvested contents were heated with random hexamer primers (Promega) at 95°C for 5 min and then cooled on ice for 1 min. The RT mixture (50 μl) contained cytoplasmic contents from 5 cells, 1× RT buffer, 1 mM dNTP mixture, 500 ng random hexamer primers, 40 U fresh RNasin Plus, and 200 U ReverTra Ace (M-MLV reverse transcriptase, RNase H (−); Toyobo, Osaka, Japan). Reverse transcription was carried out at 30°C for 10 min and then at 42°C for 45 min. After stopping the reaction by heating at 75°C for 15 min, the reaction mixture was treated with RNase H (Takara Bio, Shiga, Japan) at 37°C for 30 min and stored at −80°C until use. To confirm successful cDNA synthesis from the cytoplasmic contents of the GnRH neurons, a one-round PCR amplification of GnRH mRNA transcripts was performed using 2-μl aliquots of the RT mixture as a template. For voltage-gated calcium channel α1 subunits, a two-round PCR amplification was performed using 10-μl aliquots of the RT mixture as a template for the first-round PCR and 0.5-μl aliquots of the first-round PCR solution as a template for the nested second-round PCR. The PCR conditions were 94°C for 2 min, 28 cycles of 94°C for 30 s, 57°C for 20 s, and 72°C for 30 s, and, finally, 72°C for 5 min. The PCR mixture (50 μl) contained template DNA, 1× PCR buffer, 0.2 mM dNTP mixture, 0.2 μM forward and reverse primers, and 1.2 U Blend Taq polymerase (Toyobo). The primer sequences are shown in Table 1. The PCR products (5 μl) were separated by electrophoresis in 2% agarose gels, and visualized by ethidium bromide staining under UV irradiation. Gel images were captured using a FAS-III system (Toyobo). The PCR products were confirmed by DNA sequencing. Briefly, the PCR products were extracted from the agarose gels using a Wizard SV Gel and PCR clean-up system (Promega), and cloned into pGEM-T-Easy vectors (Promega). Sequencing reactions were performed using a BigDye terminator v3.1 cycle sequencing kit (Applied Biosystems, Foster City, CA, USA). Fluorescence signals were detected using an ABI Prism 310 genetic analyzer (Applied Biosystems).

Table 1 Primer sequences and amplicon lengths
Full size table

Statistical analysis

Data were obtained from at least three independent experiments, and expressed as mean ± SEM unless otherwise indicated. Fisher’s exact probability test and a t test were used for statistical analysis. The significance level was set at P < 0.05.

Chemicals

4AP and TEA-Cl were purchased from Wako Junyaku (Osaka, Japan).

Results

Transcripts for rat voltage-gated Ca2+ channels in GnRH neurons

To determine the expression patterns of the mRNAs encoding voltage-gated Ca2+ channels in rat GnRH neurons, multi-cell RT-PCR analyses were performed using specific primers for L-type (α1S, α1C, α1D and α1F), P/Q-type (α1A), N-type (α1B), R-type (α1E), and Τ-type (α1G, α1H and α1I) subunits of Ca2+ channels and GnRH. In GnRH neurons from adult rats, the R-type channel mRNA was detected in all samples examined (Figs. 1, 2a, b). The N and P/Q-type channel mRNAs were positive in ~50% of the samples. Among the L-type channel mRNAs, α1C (Cav1.2) and α1D (Cav1.3) were positive in ~20 and ~40% of the samples, respectively, and no positive bands were detected for α1S and α1F. Among the T-type channel mRNAs, α1H (Cav3.2) was dominant (40–60% of the samples). α1G (Cav3.1) and α1I (Cav3.3) were positive in ~10% of the samples. α1C and α1G were negative in the male DBB/OVLT. α1I was negative in the female mPOA. GnRH was positive in all reactions. Regarding the expression patterns of the Ca2+ channel subunit mRNAs, there were no sex differences and no differences between the DBB/OVLT and mPOA, except for α1C, α1G, and α1I (Fig. 2a, b). The dominant expression of α1H was affected by neither sex nor estrous cycle stage (Fig. 2c).

Fig. 1
figure 1

Multi-cell RT-PCR analyses of mRNAs encoding voltage-gated Ca2+ channel α1 subunits in adult rat GnRH neurons. Cytosols harvested from five GnRH neurons in slice preparation were pooled and reverse-transcribed to generate cDNAs, which were then subjected to PCR amplification with specific primers. The amount of cDNA used for each PCR corresponds to one cell for the Ca2+ channel α1 subunits and 0.2 cells for GnRH. M 100-bp marker ladder, N cytosols without reverse transcriptase, R reference tissues (hypothalamus for α1A–E, α1G–H, and GnRH; skeletal muscle for α1S; eye for αlF)

Full size image
Fig. 2
figure 2

Expression patterns of voltage-gated Ca2+ channel α1 subunits in male and female rat GnRH neurons. The appearance of positive bands in the RT-PCR analyses is shown as a percentage of the total reactions for each subunit. a, b Results were obtained with GnRH neurons harvested from DBB/OVLT slices and mPOA slices. The numbers of reactions are indicated in the figures. c Expression patterns of α1 subunits for T-type Ca2+ channel in estrous cycle stages and males. D diestrus (n = 21), P proestrus (n = 18), E estrus (n = 18), males (n = 18). Cells were harvested from an acutely dispersed preparation

Full size image

Activation and steady-state inactivation of the T-type Ca2+ currents

T-type currents are low-voltage-activated Ca2+ currents that have unique kinetics among the voltage-gated Ca2+ currents. We analyzed the T-type currents in acute slice preparations from GnRH-EGFP rats. To measure the activation of the T-type Ca2+ currents, test potentials (−80 to −20 mV) were applied and the peak current at each potential was measured (Figs. 3a, 4a). The T-type Ca2+ channel blocker Ni2+ (100 μM) blocked the inward currents (Fig. 3c), indicating the presence of T-type currents (Fig. 3d). In GnRH neurons from adult rats, the peak current densities at −30 mV were 15.0 ± 5.3 pA/pF (n = 8) in females and 15.8 ± 2.1 pA/pF (n = 10) in males. The inactivation time constants of the T-type currents at −30 mV were 22.3 ± 1.7 ms (n = 8) in females and 19.8 ± 1.4 ms (n = 9) in males. These values correspond to the currents through channels composed of α1G or α1H [4, 14, 15]. The inactivation time constants were voltage-dependent, and decreased by 70% in females and 61% in males between −45 and −25 mV (Fig. 3e). There were no statistically significant differences in the current density and inactivation time constant with regard to sex.

Fig. 3
figure 3

T-type Ca2+ currents. a Voltage sequence. Test pulses (100 ms, −80 up to −20 mV in 10-mV increments), preceded by a 2-s prepulse of −100 mV to fully deactivate the T-type channels, were applied at 0.1 Hz. b–d Representative current traces; the capacitative and leak currents were not subtracted in (b) and (c). The T-type Ca2+ channel blocker Ni2+ (100 μM) blocks the inward currents (c), suggesting the presence of T-type currents, based on the difference currents (d) between the control currents (b) and currents with Ni2+ (c). The inactivation time constant was determined by exponential fitting as shown in (d). e Variation of inactivation time constant with command voltage. Data represent means ± SEM. Upper or lower error bars are shown for clarity. (The voltages are not corrected for liquid junction potential in this figure or in Figs. 4–6)

Full size image
Fig. 4
figure 4

Activation and steady-state inactivation of T-type Ca2+ currents. a Test pulses (−80 up to −20 mV) for activation were applied at 0.1 Hz and the peak current at each potential was measured. b Two-second prepulses (−100 to −30 mV) were applied at 0.1 Hz, and then the peak current elicited by a test pulse (−20 mV) was measured. c Activation and steady-state inactivation are shown. The data were normalized by the value at −20 mV for the activation and by the value at −100 mV for the steady-state inactivation. The activation threshold was −60 mV and full steady-state inactivation was achieved at −30 mV. The half-inactivation voltage was ~−55 mV. Data represent mean ± SEM

Full size image

The T-type currents began to be activated around −60 mV (Fig. 4a, c). In steady-state inactivation experiments, the T-type currents were half-inactivated at around −55 mV and fully inactivated at −30 mV (Fig. 4b, c). Therefore, there was a small window T-type Ca2+ current between −60 and −30 mV, indicating a small proportion of the channel is active in these voltages. The voltages are not corrected for the liquid junction potential. There were no sex differences in the voltage-dependent activation and steady-state inactivation.

Recovery time constants of the T-type Ca2+ currents

The recovery time constants ranged from 0.45 to 0.75 s with a median of 0.61 s in males and from 0.25 to 1.32 s with a median of 0.75 s in females. All the values except one among the females were slower than 0.45 s. Exponential regression analyses of the pooled data produced recovery time constants of 0.59 s (n = 5) in males and 0.65 s (n = 6) in females (Fig. 5). These values correspond to the currents through channels composed of α1H or α1I [4, 14, 15]. There were no sex differences in the recovery time courses from the inactivation produced by a 2-s pulse of −30 mV.

Fig. 5
figure 5

Recovery time courses of the T-type Ca2+ currents in adult rats. The voltage sequence is shown (b, inset). A 2-s pulse of −30 mV was applied to fully inactivate the T-type Ca2+ channels, followed by −100 mV prepulses of variable lengths to deactivate the channels, and finally a 100-ms test pulse of −20 mV was applied. a Representative current traces are shown. The number above each trace indicates the length of the prepulse. b Recovery time courses of the T-type Ca2+ currents. The time constants of the recovery were determined by exponential regression analysis of the pooled data for males and females. The time constants were 0.61 s (n = 6) for females and 0.54 s (n = 5) for males. Data represent mean ± SEM

Full size image

Rebound action potential

In current-clamp recordings rebound action potential was generated by a hyperpolarizing current pulse both in dissociated GnRH neurons and in the neurons in slice preparation (Fig. 6). The rebound action potential was blocked by 100 μM Ni2+, indicating an involvement of T-type Ca2+ channels. As shown in Fig. 6, only a single rebound action potential but not a burst of action potentials was observed.

Fig. 6
figure 6

In current-clamp recordings rebound action potential was generated by 1-s hyperpolarizing current pulse both in GnRH neurons in slice preparation (a) and in dispersed GnRH neurons (b). The rebound action potential was blocked by 100 μM Ni2+

Full size image

Discussion

These results demonstrated that GnRH neurons from adult rats express α1 subunit mRNAs for all types of voltage-gated Ca2+ channels, namely L, N, P/Q, R, and T-type channels. These results are consistent with our previous report on the voltage-gated Ca2+ currents in rat GnRH neurons [1, 2]. Qualitatively similar Ca2+ currents have also been reported in mouse GnRH neurons [7, 8].

This study also revealed that rat GnRH neurons express α1D mRNA and, to a lesser extent, α1C mRNA, but not α1S and α1F mRNAs, for the L-type Ca2+ channels. Therefore, the L-type Ca2+ channels in rat GnRH neurons are mainly composed of the neuroendocrine subunit type (α1D) and are devoid of the skeletal muscle subunit type (α1S) and retina subunit type (α1F) [5]. For T-type Ca2+ channels, α1H was dominant in GnRH neurons of adult rats. The dominant expression of α1H was not affected by either sex or estrous cycle stage. In contrast with our results, both α1H and α1I are reported to be dominant in GnRH neurons from ovariectomized E2-treated mice [9]. A high-dose of E2 elevates expression of α1H and α1I in the morning, and suppresses expression of α1H without affecting that of α1I in the afternoon [9]. Thus, there is a clear difference in the expression of α1I between mouse and rat GnRH neurons. This may be because of a species difference. Concerning the effect of E2, expression of α1H was still dominant in the afternoon of proestrus, during which E2 level is high, in rat GnRH neurons, whereas that is suppressed in E2-treated mice. The precise cause of this discrepancy is not known, although it may be because of a species difference and/or the different experimental design. The subtype specificity of the T-type Ca2+ channels in rat GnRH neurons was also confirmed by kinetic analyses of the T-type Ca2+ currents. In the GnRH neurons from adult rats, the inactivation time constant was ~20 ms at −30 mV, suggesting the expression of α1G and/or α1H but not α1I [10, 15]. The recovery time constant was ~0.6 s, indicating the expression of α1H and/or α1I [14]. Taken together, these results indicate that the T-type Ca2+ channels in GnRH neurons from adult rats are mainly composed of α1H both in males and females.

Regarding the physiological implications, T-type Ca2+ channels are most likely to be involved in the regulation of the intracellular Ca2+ concentration and the generation of Ca2+ spikes. Our results reveal that the window current of the T-type channels in rat GnRH neurons ranged from −60 to −30 mV. These voltages are not corrected for the liquid junction potential of ~10 mV [17, 18]. It is also to be noted that the extracellular Ca2+ concentration was 10 mM in our experiments, except current-clamp recordings. High concentration (10 mM) of Ca2+ exerts the surface charge shielding effect. Consequently, a certain proportion of the T-type Ca2+ channels must be active at the resting potential of ~−60 mV, thereby contributing to cellular Ca2+ homeostasis. No other voltage-gated Ca2+ channels are active at this voltage [1]. For example, the activation threshold of R-type Ca2+ currents is −40 mV (without correction for the liquid junction potential) [1]. The resting Ca2+ influx through the T-type Ca2+ channels may activate the Ca2+-dependent cellular machineries, for example the SK channels [2]. Indeed, a small proportion of SK channels may be active at the resting membrane potential, because application of the SK channel blocker apamin was reported to depolarize mouse GnRH neurons [19].

T-type Ca2+ channels are also expressed in somatodendritic membranes [6]. In some neurons, T-type Ca2+ channels are thought to participate in the generation of low-frequency spiking and high-frequency bursting [20–22]. In either case, rebound action potentials, which are generated by relatively strong hyperpolarization, play an important role [23]. However, GnRH neurons do not show typical bursting activity, in contrast with that observed in magnocellular hypothalamic neurons [24–26]. Indeed, only a single rebound action potential was observed in rat GnRH neurons. During spiking, GnRH neurons hyperpolarize after each action potential or after weak bursting to around −70 mV at most [24–26], which may not deactivate T-type Ca2+ channels sufficiently to generate rebound Ca2+ spikes. However, the T-type Ca2+ channels may be involved in the generation of rhythmic membrane potential oscillations that underlie a tonic or burst firing [27]. Therefore, the T-type Ca2+ channels seem to function together with SK channels to generate continuous spiking in rat GnRH neurons [2]. On the other hand, dendritic T-type Ca2+ channels, together with R-type Ca2+ channels, may be involved in the generation of Ca2+ spikes [6]. Dendritic Ca2+ spikes mainly function for the back propagation of somatic spikes to the entire dendrite [28], which reduces the excitability of the dendritic membrane [29]. Dendritic spikes produced by T and R-type Ca2+ channels may also be conveyed orthodromically to the soma when strong depolarization occurs at the distal dendritic membrane [28]. Anti-dromic and ortho-dromic action potentials have been demonstrated in mouse GnRH neurons, in which Na+ channels reside in the dendritic membrane [30]. Therefore, rat GnRH neurons are likely to generate dendritic spikes by mainly activating T and R- type Ca2+channels, and possibly Na+ channels. In addition, Ca2+ influx through the somatodendritic membrane may facilitate autocrine and/or paracrine actions of GnRH that affect the functions of GnRH neurons, as shown in hypothalamic magnocellular peptidergic neurons [31]. In fact, we have recently shown that ambient GnRH downregulates the expression of melatonin receptors in GT1-7 cells [32].

In conclusion, rat GnRH neurons express all five types of voltage-gated Ca2+ channels. For T-type Ca2+ channels, α1H is dominant in adults. There are no sex differences in this expression and no difference with regard to estrous cycle stage for the dominancy of α1H expression.