Juvenile-Specific Burst Firing of Terminal Nerve GnRH3 Neurons Suggests Novel Functions in Addition to Neuromodulation

Juvenile-Specific Burst Firing of Terminal Nerve GnRH3 Neurons Suggests Novel Functions in... Abstract Peptidergic neurons are suggested to play a key role in neuromodulation of animal behaviors in response to sensory cues in the environment. Terminal nerve gonadotropin-releasing hormone 3 (TN-GnRH3) neurons are thought to be one of the peptidergic neurons important for such neuromodulation in adult vertebrates. On the other hand, it has been reported that TN-GnRH3 neurons are labeled by a specific GnRH3 antibody from early developmental stages to adulthood and are thus suggested to produce mature GnRH3 peptide even in the early developmental stages. However, it remains unknown when TN-GnRH3 neurons show spontaneous burst firing, which is suggested to be involved in neuropeptide release. Using a whole-brain in vitro preparation of gnrh3:enhanced green fluorescent protein (EGFP) medaka fish, we first recorded spontaneous firings of TN-GnRH3 neurons after hatching to adulthood. Contrary to what one would expect from their neuromodulatory functions—that TN-GnRH3 neurons are more active in adulthood—TN-GnRH3 neurons in juveniles showed spontaneous burst firing more frequently than in adulthood (juvenile-specific burst firing). Ca2+ imaging of TN-GnRH3 neurons in juveniles may further suggest that juvenile-specific burst firing triggers neuropeptide release. Furthermore, juvenile-specific burst firing was suggested to be induced by blocking persistent GABAergic inhibition to the glutamatergic neurons, which leads to an increase in glutamatergic synaptic inputs to TN-GnRH3 neurons. The present study reports that peptidergic neurons show juvenile-specific burst firing involved in triggering peptide release and suggests that juvenile TN-GnRH3 neurons have novel functions, in addition to neuromodulation. Animals change their behavior in response to sensory cues in the environment, as well as the physiological status of their own. For example, various behaviors are modulated by stress (1), and some social behaviors, such as schooling and sexual behaviors (2, 3), are modulated by various sensory cues in the environment, such as the temperature, day length, etc. However, the neural mechanisms of how the neural circuit integrates such information from the environment and modulates their behaviors remain unknown. It is generally accepted that such behavioral changes are induced by modulations of neural activities in the brain (generally called “neuromodulation”). Although neural activities are known to be modulated by various molecules (generally called “neuromodulators”), it has been proposed that neuropeptides play important roles in such neuromodulation (4–6). The terminal nerve (TN) gonadotropin-releasing hormone (GnRH) 3 (TN-GnRH3) neuron, which is one of the phylogenetically conserved peptidergic neurons among vertebrates (7), has been suggested to act as a neuromodulatory cell to regulate motivation for sexual behavior (8–10). This neuron projects to wide areas of the brain (11, 12), and GnRH3 peptide that is released from TN-GnRH3 neurons has been demonstrated physiologically to modulate neuronal functions involved in sensory processing (13–16). Such neuromodulation has been analyzed by experiments using only adult fish and suggested to lead finally to behavioral changes. On the other hand, it has been reported that TN-GnRH3 neurons are labeled by GnRH3 antibody from early developmental stages to adulthood and are suggested to produce a mature GnRH3 peptide (17, 18), even in the early developmental stages. Though it has been suggested that spontaneous high-frequency firing, such as bursting, is important for neuropeptide release, it remains unknown during which developmental stages TN-GnRH3 neurons show burst firing. Here, we recorded spontaneous firings of TN-GnRH3 neurons from gnrh3:enhanced green fluorescent protein (EGFP) medaka fish after hatching to adulthood. An advantage of using medaka as a model is its small and transparent brain that can be used in a whole-brain in vitro preparation. In this preparation, intact neural circuits are maintained for electrical recording and imaging. Contrary to what one would expect from their neuromodulatory functions in adults, TN-GnRH3 neurons in juvenile fish, far more frequently than in adult fish, showed burst firing (juvenile-specific burst firing), which was accompanied by significant Ca2+ signals that may be suggestive of active neuropeptide release. Our study reports that peptidergic neurons show juvenile-specific burst firing and suggests that juvenile TN-GnRH3 neurons have novel functions, in addition to neuromodulation. Materials and Methods Female and male gnrh3:EGFP medaka (Oryzias latipes) (19) were maintained at 27°C at a 24-hour dark cycle before hatching and a 14:10 light:dark cycle after hatching. The fish were fed daily with live brine shrimp and flake food. We used juvenile medaka (three; ∼9 weeks after fertilization) and adult medaka (>9 weeks after fertilization). All physiological studies were performed at ∼20°C to 27°C. All procedures were performed in accordance with the guideline principles for the care established by the Physiological Society of Japan, and the protocols were approved by the Animal Care and Use Committee of the University of Tokyo (permission number 15-3, 17-1). Electrophysiology The fish were anesthetized by 0.02% 3-aminobenzoic acid ethyl ester (MS-222; Sigma Aldrich, St. Louis, MO). They were quickly euthanized by decapitation, and the whole brain was isolated. The isolated brain was put in an artificial cerebrospinal fluid (ACSF) consisting of (in mM) 134 NaCl, 2.9 KCl, 1.2 MgCl2, 2.1 CaCl2, 10 HEPES, and 10 glucose (pH 7.4 adjusted with NaOH and ∼300 mOsm adjusted with sucrose). The whole brain of this fish can be maintained in vitro for a long period without oxygenation (20). Then, the ventral meningeal membrane was carefully removed. For the whole-cell recording, internal solution was the following (mM): K-gluconate 112.5, KCl 17.5, NaCl 4, EGTA 1, MgCl2 1, CaCl2 0.5, and HEPES 10 (pH 7.2 adjusted with KOH and ∼290 mOsm adjusted with sucrose). The junction potential was −13.8 mV, and the membrane potentials were adjusted by using this value. The tip resistance of patch electrodes (GD-1.5; Narishige, Tokyo, Japan) in ACSF was ∼10 to 15 MΩ. For the on-cell clamp recording, the pipette solution was the same as ACSF. Both recordings were performed using an Axopatch 200B patch-clamp amplifier (Molecular Devices, Sunnyvale, CA). The whole-cell current-clamp recordings were digitized (10 kHz) and stored on a computer using Digidata 1322A and pCLAMP 9.2 software (Molecular Devices). We detected action potentials by using Clampfit 10 software (Molecular Devices) and analyzed firing pattern by referring to Hasebe et al. (21). There was not any difference between firing patterns recorded by on-cell and those recorded by whole-cell patch clamp. Bath application of drugs was performed by mixing drugs in the ACSF, and puffer application (100 ms, ∼60 to 70 kPa) was performed by using the electric microinjector IM-31 (Narishige). The tip diameter of glass pipette (GD-1; Narishige) for puffer application was 1 to 5 μm, and the pipette was placed ∼10 to 30 μm from the cell bodies of TN-GnRH3 neurons. Ca2+ imaging The whole brain was isolated as described previously. Some neurons and glial cells covering the ventral region of the somata of TN-GnRH3 neurons were mechanically removed with the aid of a glass pipette under a microscope to expose the surface of the TN-GnRH3 neurons. Then, the brain was incubated in ACSF at room temperature for 30 minutes. We prepared Fura2-acetoxymethyl (AM) solution [ACSF with 10 μM Fura2-AM (Dojindo, Kumamoto, Japan) and 0.04% Cremophor-EL (Sigma Aldrich)] and vortexed it for 10 seconds. After the incubation, we put the brain into a new 1.5-mL tube containing a Fura2-AM solution. The tube was shaken mildly (30g) at 30°C for 1 hour. The brain was washed by incubation with normal ACSF at room temperature for 15 minutes and perfusion of normal ACSF at room temperature for 15 minutes. After that, we placed the brain in a recording chamber with ACSF in the ventral side-up direction. The preparation was perfused with ACSF at a continuous flow rate (2 mL/min). Bath application of drugs was performed by mixing drugs in the ACSF, and puffer application was performed as described in the previous section by using the electric microinjector IM-31 (Narishige). Fura2 and EGFP fluorescence was detected by the Chroma 74000 filter set (D480×, D340×, D380× exciter; 505DCLP dichroic mirror; HQ535/50m emitter), housed in the Lambda DG-4 Xenon light source, excitation filter exchanger (Sutter Instruments, Novato, CA), and the BX-51WI upright epifluorescence microscope (Olympus, Tokyo, Japan). The fluorescence was recorded using the Metafluor imaging software (Molecular Devices) and each camera: QuantEM 512SC electron multiplying (EM) charge-coupled device camera [exposure: 80 (340 and 380 nm) and 50 (480 nm) ms; 5 MHz EM gain; interval: 2 s; Photometrics, Tucson, AZ; see Figs. 4 and 8A–8C below] or Zyla 4.2 scientific complementary metal-oxide-semiconductor camera [exposure: 150 (340 and 380 nm) and 80 (480 nm) ms; 5 MHz EM gain; interval: 2 s; Andor Technology, Belfast, Northern Ireland; see Fig. 8D and 8E below]. For EGFP, the intensity ratio of emission (at 510 nm) from the alternating 480-nm excitation was monitored. For Fura2, the intensity ratio of emission (at 510 nm) from the alternating 340- and 380-nm excitation was monitored. The imaging data were analyzed by the following method using ImageJ software (Research Resource Identifier: SCR_003070; National Institutes of Health, Bethesda, MD) with the McMaster Biophotonics Facility ImageJ plug-ins (Tony Collins, McMaster University, Hamilton, ON, Canada). The basal ratio of fluorescence intensity (R)0 was calculated as the average of five frames from the first one. The ratio change of fluorescence intensity (Δratio) was calculated as R − R0. Peak Δratio was calculated as [(maximum Δratio for 30 seconds after puffer application) − (average Δratio for four frames from approximately eight frames before application)]. We set up the region of interest on one of the TN-GnRH3 neurons by referring to the EGFP images and averaged those Δratio of the neurons in one hemisphere of telencephalon (approximately two to eight neurons were identified in one hemisphere). Drugs γ-Aminobutyric acid (GABA) was purchased from Wako (Osaka, Japan). Glutamate (l-glutamic acid monosodium salt hydrate), muscimol, and bicuculline were purchased from Sigma Aldrich. d-(–)-2-amino-5-phosphonopentanoic acid (d-AP5) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) were purchased from Tocris (Bristol, UK). CNQX and bicuculline were dissolved with dimethyl sulfoxide as 20 mM and were diluted to ACSF. The final concentration of dimethyl sulfoxide in ACSF was 0.1%. Statistics All animal experiments were performed on fish of both sexes. Statistical analyses were performed by using Kyplot5 (Kyence, Tokyo, Japan) and Igor Pro 6 software (WaveMetrics, Lake Oswego, OR) with a Taro tool (Igor macro set, written by Dr. Taro Ishikawa, Jikei University School of Medicine, Tokyo, Japan). Analyses include the following: analysis of percentage of burst-firing neurons: Jonckheere trend test (decrease; see Fig. 1); analysis of effect of glutamatergic synapse: analysis of variance (ANOVA) with Dunnett post hoc test (see Figs. 2 and 3) or Steel test (see Fig. 4); analysis of bicuculline-induced burst firing: ANOVA with Dunnett post hoc test (see Figs. 5–7); analysis of the intracellular Ca2+ concentration ([Ca2+]i) increase during bicuculline-induced burst firing: paired Wilcoxon test (see Fig. 8). Each detailed P value is shown in figure legends. All data in the current study are as means ± standard error of the mean. Results TN-GnRH3 neurons change firing patterns during post-hatching development The spontaneous neural activities of EGFP-labeled TN-GnRH3 neurons from fish after hatching to adulthood were recorded by on-cell patch clamp (Fig. 1). The firing patterns were classified as regular, irregular, and burst firing (see Fig. 1 legend for definitions). The burst firing was defined as firing with more than three consecutive spikes at >3 Hz and with an interburst interval longer than 1 second, which was based on the regular firing frequency in adults (see later). We could not find silent neurons. In adult fish, it has been reported that TN-GnRH3 neurons most frequently show regular firing (pacemaker activity) and show irregular or burst firing on rare occasions (11). In contrast to the firing patterns in adulthood (medaka were maintained in pairs), we found that most TN-GnRH3 neurons in juvenile medaka showed irregular or burst firing (Fig. 1). As shown in Fig. 1F, nearly 60% of neurons in fish, just after hatching, were categorized as burst firing, and such “juvenile-specific spontaneous burst firing” was observed in both males and females. The intraburst firing frequency was >5 Hz, from 3 to 7 weeks after fertilization (Table 1, top row). Figure 1G shows the percentage of burst-firing neurons in a fish selected from Fig. 1F. The percentage of neurons exhibiting burst firing decreased as fish grew older (Jonckheere trend test, P < 0.001). The change in the percentage of burst-firing neurons started later than the early stages of sexual maturation. In addition, as fish aged, the percentage of regular firing neurons increased to 100% by 12.5 weeks after fertilization (Fig. 1F). Furthermore, the firing frequency of neurons with a regular firing pattern tended to be lower in juvenile medaka than in adult medaka (Table 1, bottom row). Figure 1. View largeDownload slide TN-GnRH3 neurons during post-hatching development show juvenile-specific spontaneous burst firing. (A) The scheme of medaka post-hatching development. (B–E) Representative firing patterns in medaka. The firing patterns were (B and C) recorded from 4 weeks after fertilization; (D) from 7 weeks after fertilization; and (E) from 12.5 weeks after fertilization. The scale is shown in the upper right for (B)–(E). (F) Firing patterns of TN-GnRH3 neurons in one fish for each stage. We used the average value for each pattern. The classification of the firing pattern is as follows: burst firing, consecutive spikes more than three at >3 Hz (firing frequency in adult medaka is usually lower than 2 Hz) and interburst interval longer than 1 second (longer than that of general TN-GnRH3 neurons in adults); regular firing, coefficient of variation of the interburst interval <0.65 (the coefficient of variation is defined as the ratio of the standard deviation to the mean interburst interval); irregular firing, others. Juvenile: 3 weeks after fertilization, 9 fish; 4 weeks, 9 fish; 5 weeks, 7 fish; 6 weeks, 8 fish; 7 weeks, 8 fish. Adult: >12.5 weeks, 4 fish. (G) The percentage of burst firing was selected from (F). Jonckheere trend test, P < 0.001. Figure 1. View largeDownload slide TN-GnRH3 neurons during post-hatching development show juvenile-specific spontaneous burst firing. (A) The scheme of medaka post-hatching development. (B–E) Representative firing patterns in medaka. The firing patterns were (B and C) recorded from 4 weeks after fertilization; (D) from 7 weeks after fertilization; and (E) from 12.5 weeks after fertilization. The scale is shown in the upper right for (B)–(E). (F) Firing patterns of TN-GnRH3 neurons in one fish for each stage. We used the average value for each pattern. The classification of the firing pattern is as follows: burst firing, consecutive spikes more than three at >3 Hz (firing frequency in adult medaka is usually lower than 2 Hz) and interburst interval longer than 1 second (longer than that of general TN-GnRH3 neurons in adults); regular firing, coefficient of variation of the interburst interval <0.65 (the coefficient of variation is defined as the ratio of the standard deviation to the mean interburst interval); irregular firing, others. Juvenile: 3 weeks after fertilization, 9 fish; 4 weeks, 9 fish; 5 weeks, 7 fish; 6 weeks, 8 fish; 7 weeks, 8 fish. Adult: >12.5 weeks, 4 fish. (G) The percentage of burst firing was selected from (F). Jonckheere trend test, P < 0.001. Table 1. Firing Frequency of TN-GnRH3 Neurons in Burst or Regular Firing   Juvenile   Adult   Weeks After Fertilization  3  4  5  6  7  12.5  Intraburst firing frequency, Hz  12.37 ± 1.20 (n = 7)  9.51 ± 0.04 (n = 7)  11.14 ± 0.84 (n = 6)  5.54 ± 0.84 (n = 4)  6.21 ± 0.88 (n = 2)  None  Avg. frequency in regular firing, Hz  0.38 ± 0.08 (n = 5)  0.40 ± 0.13 (n = 5)  0.28 ± 0.04 (n = 5)  0.64 ± 0.08 (n = 7)  0.62 ± 0.12 (n = 8)  0.92 ± 0.08 (n = 4)    Juvenile   Adult   Weeks After Fertilization  3  4  5  6  7  12.5  Intraburst firing frequency, Hz  12.37 ± 1.20 (n = 7)  9.51 ± 0.04 (n = 7)  11.14 ± 0.84 (n = 6)  5.54 ± 0.84 (n = 4)  6.21 ± 0.88 (n = 2)  None  Avg. frequency in regular firing, Hz  0.38 ± 0.08 (n = 5)  0.40 ± 0.13 (n = 5)  0.28 ± 0.04 (n = 5)  0.64 ± 0.08 (n = 7)  0.62 ± 0.12 (n = 8)  0.92 ± 0.08 (n = 4)  Abbreviation: Avg., average. View Large Glutamatergic synaptic inputs are involved in juvenile-specific burst firing in TN-GnRH3 neurons To examine whether the juvenile-specific burst firing mentioned in the previous section is induced by synaptic inputs, we first recorded firing activities of TN-GnRH3 neurons in juvenile medaka with an ACSF containing glutamate receptor antagonist CNQX (Fig. 2). The application of CNQX decreased the occurrence of burst firing (Fig. 2B and 2D). The median firing frequency of bursting neurons in juvenile terminal nerve was also decreased by CNQX application (Fig. 2E). On the other hand, CNQX did not significantly affect the firing frequency of TN-GnRH3 neurons in adults (Fig. 3). Thus, glutamatergic presynaptic inputs are suggested to be important for the juvenile-specific burst firing in TN-GnRH3 neurons. Figure 2. View largeDownload slide Bath application of CNQX decreases the number of burst firing in juvenile TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of juvenile medaka for 7 minutes. (A–C) Representative recordings, (D) normalized percentage of burst firing, and (E) normalized median frequency from juvenile (∼3 to 4 weeks after fertilization, n = 5). One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz). One arrow [in (A) and (C)] shows one bout of burst. Inset shows the time course of the experiment, and firing activities during the gray bars (for 2 minutes) were used for the analysis in (D). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX, P = 0.0120; Before vs Washout, P = 0.2092. (E) The vertical axis shows the median frequency normalized by that before 20 μM CNQX application (n = 5). The median frequency of each neuron was calculated by using instantaneous frequency of all spikes. ANOVA with Dunnett post hoc test, ***P < 0.001; Before vs CNQX, P = 0.0006; Before vs Washout, P = 0.0787. Figure 2. View largeDownload slide Bath application of CNQX decreases the number of burst firing in juvenile TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of juvenile medaka for 7 minutes. (A–C) Representative recordings, (D) normalized percentage of burst firing, and (E) normalized median frequency from juvenile (∼3 to 4 weeks after fertilization, n = 5). One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz). One arrow [in (A) and (C)] shows one bout of burst. Inset shows the time course of the experiment, and firing activities during the gray bars (for 2 minutes) were used for the analysis in (D). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX, P = 0.0120; Before vs Washout, P = 0.2092. (E) The vertical axis shows the median frequency normalized by that before 20 μM CNQX application (n = 5). The median frequency of each neuron was calculated by using instantaneous frequency of all spikes. ANOVA with Dunnett post hoc test, ***P < 0.001; Before vs CNQX, P = 0.0006; Before vs Washout, P = 0.0787. Figure 3. View largeDownload slide Bath application of CNQX does not affect firing frequency in adult TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of adult medaka for 4 minutes. (A–C) Representative recordings and (D) normalized frequency (n = 6) from adults (>12 weeks after fertilization). ANOVA with Dunnett post hoc test, n.s.; Before vs CNQX, P = 0.8417; Before vs Washout, P = 0.8002. n.s., not significant. Figure 3. View largeDownload slide Bath application of CNQX does not affect firing frequency in adult TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of adult medaka for 4 minutes. (A–C) Representative recordings and (D) normalized frequency (n = 6) from adults (>12 weeks after fertilization). ANOVA with Dunnett post hoc test, n.s.; Before vs CNQX, P = 0.8417; Before vs Washout, P = 0.8002. n.s., not significant. The mimicking of direct glutamatergic synaptic input can trigger juvenile-specific burst firing and induce a [Ca2+]i increase, which may suggest neuropeptide release To examine whether glutamatergic synaptic inputs can actually induce burst firing, we performed puffer application of glutamate to mimic such synaptic inputs. Puffer application of glutamate transiently increased firing frequency of TN-GnRH3 neurons in juvenile medaka in a dose-dependent manner (Fig. 4A–4C). The peak frequency in Fig. 4B was 7.27 ± 1.34 Hz. Previous literature reported that neuropeptide release requires an increase in [Ca2+]i (6, 22), and TN-GnRH3 neurons show somato-dendritic release (20, 23). To examine whether such burst firing induced by glutamate puffer application is related to neuropeptide release, we performed Ca2+ imaging from the area surrounding somata of TN-GnRH3 neurons using a Ca2+ indicator, Fura2-AM. Puffer application of glutamate mimicking synaptic input induced a significant increase in [Ca2+]i of TN-GnRH3 neurons in juvenile (Fig. 4D and 4E). The [Ca2+]i in TN-GnRH3 neurons also increased in a dose-dependent manner (Fig. 4E). With the consideration of the increase in firing frequency (Fig. 4C) and [Ca2+]i (Fig. 4D and 4E), it may be suggested that the glutamate-induced burst firing of TN-GnRH3 neurons can trigger neuropeptide release. Similarly, in adults, puffer application of glutamate to TN-GnRH3 neurons also induced burst firing and a [Ca2+]i increase, which may suggest neuropeptide release (Fig. 4F–4I). Figure 4. View largeDownload slide Puffer application of glutamate quickly induces burst-like firing and increases [Ca2+]i in a dose-dependent manner. (A and B) Glutamate (glu; 0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in juveniles for 100 ms by using a puffer pipette at the point indicated by the arrow. (A) Representative recordings and (B) frequency histograms (n = 6) from juvenile medaka. Each frequency was calculated by the number of spikes for a 0.5-second bin. (C) Peak frequency of TN-GnRH3 neurons during glutamate application. Peak frequency was calculated by the number of spikes for 1 second, just after glutamate application. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0072; 0 vs 0.5 mM, P = 0.0101; 0 vs 1 mM, P = 0.0101; 0 (n = 6), 0.1 (n = 12), 0.5 (n = 6), and 1 (n = 6). (D) The ratio (λ340/λ380) of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (E) The peak Δratio is plotted against the glutamate concentrations. Values were calculated by ImageJ. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0113; 0 vs 0.5 mM, P = 0.0050; 0 vs 1 mM, P = 0.0050. Each group, n = 7. The juvenile fish used in experiments described above were 3 weeks after fertilization. (F and G) Glutamate (0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in adults for 100 ms by using a puffer pipette at the point indicated by the arrow. (F) Representative recording and (G) frequency histograms (n = 12) from adult medaka. Bin: 0.5 seconds. (H) The ratio of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (I) The Δratio is plotted against the glutamate concentrations. Each group, n = 6. Steel test, *P < 0.05; 0 (control) vs 0.1 mM, P = 0.0956; 0 vs 0.5 mM, P = 0.01107; 0 vs 1 mM, P = 0.01107. The adult fish used in experiments described above were >6 months after fertilization. Figure 4. View largeDownload slide Puffer application of glutamate quickly induces burst-like firing and increases [Ca2+]i in a dose-dependent manner. (A and B) Glutamate (glu; 0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in juveniles for 100 ms by using a puffer pipette at the point indicated by the arrow. (A) Representative recordings and (B) frequency histograms (n = 6) from juvenile medaka. Each frequency was calculated by the number of spikes for a 0.5-second bin. (C) Peak frequency of TN-GnRH3 neurons during glutamate application. Peak frequency was calculated by the number of spikes for 1 second, just after glutamate application. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0072; 0 vs 0.5 mM, P = 0.0101; 0 vs 1 mM, P = 0.0101; 0 (n = 6), 0.1 (n = 12), 0.5 (n = 6), and 1 (n = 6). (D) The ratio (λ340/λ380) of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (E) The peak Δratio is plotted against the glutamate concentrations. Values were calculated by ImageJ. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0113; 0 vs 0.5 mM, P = 0.0050; 0 vs 1 mM, P = 0.0050. Each group, n = 7. The juvenile fish used in experiments described above were 3 weeks after fertilization. (F and G) Glutamate (0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in adults for 100 ms by using a puffer pipette at the point indicated by the arrow. (F) Representative recording and (G) frequency histograms (n = 12) from adult medaka. Bin: 0.5 seconds. (H) The ratio of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (I) The Δratio is plotted against the glutamate concentrations. Each group, n = 6. Steel test, *P < 0.05; 0 (control) vs 0.1 mM, P = 0.0956; 0 vs 0.5 mM, P = 0.01107; 0 vs 1 mM, P = 0.01107. The adult fish used in experiments described above were >6 months after fertilization. Juvenile-specific burst firing of TN-GnRH3 neurons can be triggered by the blocking of GABAergic synaptic inputs onto the glutamatergic interneurons Next, we examined the role of GABAergic synaptic inputs, which are the representative inhibitory synaptic inputs in vertebrate brains. Firing activities of TN-GnRH3 neurons were recorded in an ACSF containing GABAA receptor antagonist, bicuculline. Bath application of bicuculline induced the juvenile-specific burst firing of TN-GnRH3 neurons (Fig. 5A). Such juvenile-specific burst firing induced by bicuculline was observed in medaka from 3 weeks after fertilization onward. On the other hand, GABA directly excites TN-GnRH3 neurons in juveniles (Fig. 5B) and in adults [Fig. 5C; see also Nakane and Oka (24) and Sim et al. (25)]. In Fig. 5B and 5C, we used muscimol as a GABAA receptor agonist. Taken together, we can postulate a simple neural circuit, including TN-GnRH3 neurons (Fig. 6A). Here, the bicuculline-induced burst firing may be explained by increased glutamatergic synaptic inputs during bicuculline perfusion. To prove this possibility, we first recorded spontaneous firing activities of TN-GnRH3 neurons in an ACSF containing bicuculline after blocking glutamatergic synaptic inputs by a glutamate receptor antagonist, CNQX (Fig. 6B). Here, bicuculline neither increased firing frequency nor induced burst firing. We also used d-AP5 to block another type of glutamate receptor, N-methyl-d-aspartate (NMDA), and then added bicuculline (Fig. 6C). Bicuculline did not induce burst firing. Finally, during bath application of bicuculline, we locally blocked glutamatergic synaptic inputs by puffer application of CNQX and d-AP5 (Fig. 6D). In Fig. 6D, bicuculline first induced burst firing (Fig. 6D, b), as in Fig. 5A, but the burst firing was inhibited after local application of CNQX and d-AP5 (Fig. 6D, c) and recovered after washout (Fig. 6D, d). Figure 6D, e shows that CNQX and d-AP5 diminished bicuculline-induced burst firing. Taken together, bicuculline-induced burst firing is considered to be induced by activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)– and NMDA-type glutamate receptors during disinhibition of GABAergic synaptic inputs to glutamatergic neurons, as shown in Fig. 6A. Figure 5. View largeDownload slide Bath application of bicuculline (Bic; global blockage of GABAergic synaptic inputs) induces burst firing in juvenile TN-GnRH3 neurons. (A) A representative recording of TN-GnRH3 neuron in ∼7 weeks after fertilization. ACSF with 20 μM bicuculline was applied from 4 to 8 minutes after the start of recording. Note that TN-GnRH3 neurons from 3 weeks after fertilization showed bicuculline-induced burst firing. (B and C) GABAA receptor agonist muscimol (Mus; 20 μM dissolved in ACSF) was applied at the arrow for 100 ms. (B and C, a) Representative recordings and (B and C, b) frequency histograms (n = 7) from (B) juvenile (8 weeks after fertilization) and (C) adult (>6 months after fertilization) medaka. Inset shows the experimental design of application of muscimol, mimicking synaptic inputs. Bin, 0.5 seconds. Figure 5. View largeDownload slide Bath application of bicuculline (Bic; global blockage of GABAergic synaptic inputs) induces burst firing in juvenile TN-GnRH3 neurons. (A) A representative recording of TN-GnRH3 neuron in ∼7 weeks after fertilization. ACSF with 20 μM bicuculline was applied from 4 to 8 minutes after the start of recording. Note that TN-GnRH3 neurons from 3 weeks after fertilization showed bicuculline-induced burst firing. (B and C) GABAA receptor agonist muscimol (Mus; 20 μM dissolved in ACSF) was applied at the arrow for 100 ms. (B and C, a) Representative recordings and (B and C, b) frequency histograms (n = 7) from (B) juvenile (8 weeks after fertilization) and (C) adult (>6 months after fertilization) medaka. Inset shows the experimental design of application of muscimol, mimicking synaptic inputs. Bin, 0.5 seconds. Figure 6. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons is considered to be induced by activation of AMPA- and NMDA-type glutamate receptors. (A) Schematic neural circuit, including TN-GnRH3 neurons. GABA directly excites TN-GnRH3 neurons. Bath application of bicuculline blocks both direct and indirect GABAergic modulatory pathways. (B and C) Bicuculline-induced burst firing activity of juvenile medaka (Fig. 5A) did not occur during perfusion of (B) CNQX 20 μM (an AMPA-type glutamate receptor antagonist) or (C) d-AP5 25 μM (NMDA-type glutamate receptor antagonist). Each trace shows a representative recording. (A and B) Juvenile fish were <7 weeks after fertilization. (D, a) A representative recording of the TN-GnRH3 neuron 8 weeks after fertilization during 20 μM bicuculline perfusion. ACSF, containing 20 μM CNQX and 25 μM d-AP5, was applied for 3 minutes by using a puffer system (500-ms duration, 0.5 Hz, 90 trains) during the period indicated by the interrupted line. The puffer pipette was placed ∼20 μm from the somata of the recorded neuron. (D, b–d) Expanded trace for each gray region in (D, a). (D, e) Number of burst firing before (approximately −2 to 0 minutes), during (∼1 to 3 minutes), and after puffer application of CNQX and d-AP5 (∼4 to 6 minutes). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX and d-AP5, P = 0.0424; Before vs Washout, P = 0.9686; n = 4. GABA, GABAergic neuron; Glu, glutamatergic neuron; TN-GnRH3, TN-GnRH3 neuron. Figure 6. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons is considered to be induced by activation of AMPA- and NMDA-type glutamate receptors. (A) Schematic neural circuit, including TN-GnRH3 neurons. GABA directly excites TN-GnRH3 neurons. Bath application of bicuculline blocks both direct and indirect GABAergic modulatory pathways. (B and C) Bicuculline-induced burst firing activity of juvenile medaka (Fig. 5A) did not occur during perfusion of (B) CNQX 20 μM (an AMPA-type glutamate receptor antagonist) or (C) d-AP5 25 μM (NMDA-type glutamate receptor antagonist). Each trace shows a representative recording. (A and B) Juvenile fish were <7 weeks after fertilization. (D, a) A representative recording of the TN-GnRH3 neuron 8 weeks after fertilization during 20 μM bicuculline perfusion. ACSF, containing 20 μM CNQX and 25 μM d-AP5, was applied for 3 minutes by using a puffer system (500-ms duration, 0.5 Hz, 90 trains) during the period indicated by the interrupted line. The puffer pipette was placed ∼20 μm from the somata of the recorded neuron. (D, b–d) Expanded trace for each gray region in (D, a). (D, e) Number of burst firing before (approximately −2 to 0 minutes), during (∼1 to 3 minutes), and after puffer application of CNQX and d-AP5 (∼4 to 6 minutes). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX and d-AP5, P = 0.0424; Before vs Washout, P = 0.9686; n = 4. GABA, GABAergic neuron; Glu, glutamatergic neuron; TN-GnRH3, TN-GnRH3 neuron. On the other hand, in adults, bath application of the GABAA receptor antagonist (global inhibition as in Fig. 5A) did not induce burst firing (Fig. 7A). Therefore, we plotted the percentage of bicuculline-induced burst firing at various stages of post-hatching development (Fig. 7B). The percentage tended to decrease as fish grew older. In Fig. 7C, the number of burst firing during 6 minutes after the start of bicuculline application was plotted from 7 weeks after fertilization onward; the data before 7 weeks were omitted to avoid contamination of spontaneous burst firing as much as possible. Compared with TN-GnRH3 neurons in adults, those during juvenile periods showed bicuculline-induced burst firing more frequently (Fig. 7C). Thus, the bicuculline-induced burst firing also shows time-dependent changes during post-hatching development, similar to the change in the juvenile-specific burst firing. Figure 7. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons shows time-dependent changes during post-hatching development. (A) GABAA receptor antagonist bicuculline (Bic, 20 μM, dissolved in ACSF) was applied for 4 minutes into the recording chamber where an adult brain (>12.5 weeks after fertilization) was placed. Bicuculline is considered to block GABAergic synapse in the whole brain as in Fig. 6A. Bicuculline was applied during the period indicated by the red bar. (B) Percentage of neurons showing burst firing induced by bicuculline; 4 weeks after fertilization, n = 4; 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. (C) Number of burst firings induced by bicuculline perfusion, recorded from TN-GnRH3 neurons of medaka at various developmental stages. We used recordings for 6 minutes after bicuculline application, as the effect of bicuculline continued for ∼10 minutes after washout. ANOVA with Dunnett post hoc test, *P < 0.05, **P < 0.01; 7 vs 12 weeks, P = 0.0403; 8 vs 12 weeks, P = 0.0053; 9 vs 12 weeks, P = 0.1706; 10 vs 12 weeks, P = 0.5183. At 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz), followed by interburst hyperpolarization with an interburst interval longer than 1 second. Figure 7. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons shows time-dependent changes during post-hatching development. (A) GABAA receptor antagonist bicuculline (Bic, 20 μM, dissolved in ACSF) was applied for 4 minutes into the recording chamber where an adult brain (>12.5 weeks after fertilization) was placed. Bicuculline is considered to block GABAergic synapse in the whole brain as in Fig. 6A. Bicuculline was applied during the period indicated by the red bar. (B) Percentage of neurons showing burst firing induced by bicuculline; 4 weeks after fertilization, n = 4; 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. (C) Number of burst firings induced by bicuculline perfusion, recorded from TN-GnRH3 neurons of medaka at various developmental stages. We used recordings for 6 minutes after bicuculline application, as the effect of bicuculline continued for ∼10 minutes after washout. ANOVA with Dunnett post hoc test, *P < 0.05, **P < 0.01; 7 vs 12 weeks, P = 0.0403; 8 vs 12 weeks, P = 0.0053; 9 vs 12 weeks, P = 0.1706; 10 vs 12 weeks, P = 0.5183. At 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz), followed by interburst hyperpolarization with an interburst interval longer than 1 second. Next, we examined whether [Ca2+]i is also increased when TN-GnRH3 neurons show bicuculline-induced burst firing (Fig. 8). As shown in Fig. 8A, bath application of ACSF containing bicuculline induced multiple peaks of [Ca2+]i increase. The Fura2 Δratio during bicuculline perfusion was significantly higher than that of the vehicle control (Fig. 8A–8C). Therefore, bicuculline-induced burst firing in juvenile medaka is also suggested to induce neuropeptide release. Figure 8. View largeDownload slide TN-GnRH3 neurons in juveniles show an increase in [Ca2+]i during bicuculline (Bic) application and also during spontaneous burst firing, which suggests neuropeptide release. (A and B) Δratio (λ340/λ380) of Fura2 during application of ACSF (A) with or (B) without bicuculline. Inset shows an EGFP image of TN-GnRH3 neurons. Light-colored traces represent Δratio for the somata, circled in inset. Dark-colored trace shows the average trace for five neurons. (C) Peak Δratio during bicuculline perfusion (n = 10). The values represent peak Δratio during vehicle (control) or bicuculline application. Paired Wilcoxon test, *P < 0.05, P = 0.0273. The juvenile fish used in experiments described above were 7 weeks after fertilization. (D and E) Spontaneous increase in [Ca2+]i. (D) Arrowheads show the somata of juvenile TN-GnRH3 neurons (4 weeks after fertilization). The white circle shows the region outside of the somata, and the other colored circles show the somata of TN-GnRH3 neurons. The [Ca2+]i in the areas (white circle) surrounding somata of TN-GnRH3 neurons also appeared to increase synchronously with that in the somata area (arrowheads). (E) Each colored trace represents data for the averaged intensity of the region circled with the same color in (D). Dotted line represents the average intensity of the region circled in the white line. Original scale bars, 20 μm. A, anterior; L, lateral; M, medial; OB, olfactory bulb; P, posterior. Figure 8. View largeDownload slide TN-GnRH3 neurons in juveniles show an increase in [Ca2+]i during bicuculline (Bic) application and also during spontaneous burst firing, which suggests neuropeptide release. (A and B) Δratio (λ340/λ380) of Fura2 during application of ACSF (A) with or (B) without bicuculline. Inset shows an EGFP image of TN-GnRH3 neurons. Light-colored traces represent Δratio for the somata, circled in inset. Dark-colored trace shows the average trace for five neurons. (C) Peak Δratio during bicuculline perfusion (n = 10). The values represent peak Δratio during vehicle (control) or bicuculline application. Paired Wilcoxon test, *P < 0.05, P = 0.0273. The juvenile fish used in experiments described above were 7 weeks after fertilization. (D and E) Spontaneous increase in [Ca2+]i. (D) Arrowheads show the somata of juvenile TN-GnRH3 neurons (4 weeks after fertilization). The white circle shows the region outside of the somata, and the other colored circles show the somata of TN-GnRH3 neurons. The [Ca2+]i in the areas (white circle) surrounding somata of TN-GnRH3 neurons also appeared to increase synchronously with that in the somata area (arrowheads). (E) Each colored trace represents data for the averaged intensity of the region circled with the same color in (D). Dotted line represents the average intensity of the region circled in the white line. Original scale bars, 20 μm. A, anterior; L, lateral; M, medial; OB, olfactory bulb; P, posterior. Finally, we performed Ca2+ imaging of the medaka brain during the ages when TN-GnRH3 neurons showed juvenile-specific spontaneous burst firing (3 to 5 weeks after fertilization). Figure 8D shows representative images of a spontaneous [Ca2+]i increase. Traces in Fig. 8E show the Δratio of the regions of interest, indicated by different colors in the EGFP image in Fig. 8D. The [Ca2+]i in the areas surrounding somata of TN-GnRH3 neurons also appeared to increase synchronously with that in the somata area. From the data shown thus far, TN-GnRH3 neurons are suggested to release neuropeptides during juvenile-specific spontaneous and bicuculline-induced burst firing. Discussion The present study demonstrated that >50% of TN-GnRH3 neurons in juvenile medaka (especially up to 5 weeks after fertilization) showed spontaneous burst firing (intraburst firing frequency was >5 Hz), whereas none of them did so in adulthood. By Ca2+ imaging analysis, we showed that not only burst firing evoked by puffer application of glutamate mimicking glutamatergic synaptic input but also a juvenile-specific spontaneous one are suggested to trigger neuropeptide release in TN-GnRH3 neurons. Our results suggest that neuropeptides released from TN-GnRH3 neurons play a previously unrecognized role(s) in juveniles, which may indicate novel functions, in addition to its recognized role in neuromodulation in adulthood (see Functional significance of juvenile-specific burst firing in TN-GnRH3 neurons for functional discussion). Juvenile-specific burst firing of TN-GnRH3 neurons It has been reported by morphological analyses of GFP-labeled GnRH neurons (19) and immunohistochemical and in situ hybridization studies [also see references cited in Takahashi et al. (19)] that TN-GnRH3 neurons express GnRH3 peptide from early developmental stages. Though it has been suggested that spontaneous highfrequency firing, such as bursting, is important for neuropeptide release, it remains unknown during which developmental stages TN-GnRH3 neurons show burst firing. In the present study, we have demonstrated that TN-GnRH3 neurons in juveniles show burst firing, by far more frequently than in adulthood (Fig. 9; juvenile-specific burst firing). In addition, by combining electrophysiology and Ca2+ imaging, we could clearly demonstrate that burst firing (both evoked and juvenile-specific spontaneous ones; Figs. 4 and 8) actually induces a large increase in [Ca2+]i. It has been reported recently that in hypophysiotropic GnRH (GnRH1) the neuron typically fires spontaneously at ∼1 Hz (26) and shows a measurable increase in [Ca2+]i during electrical stimulation of 1 Hz (27). In addition, hypophysiotropic GnRH1 and kisspeptin neurons that show high-frequency firing (≥5 Hz) effectively induce neuropeptide release in the axon terminal in vivo (28, 29) and in vitro (27, 30). The release of these neuropeptides is triggered by an increase in [Ca2+]i (30, 31). In other types of neurons (dopaminergic neurons), it has also been reported that burst firing is important for a large increase in [Ca2+]i (32). In the present study, a firing frequency of >3 Hz, which was induced by puffer application of 100 μM glutamate, was strong enough to cause an increase in [Ca2+]i (Fig. 4). It should be noted that most of the spontaneous juvenile-specific burst firings were >5 Hz (Table 1, top row). A Fura2-fluorescence increase in spontaneous activity of juvenile TN-GnRH3 neurons was almost the same as that shown in hypophysiotropic GnRH1 neurons (Δratio, ∼0.1), which is large enough to trigger GnRH1 peptide release (30). Therefore, it is strongly suggested that juvenile-specific spontaneous burst firing can actually trigger release of the neuropeptide from the dendrite or axon in vivo as well (Fig. 9B). Furthermore, as shown in Fig. 8D, we observed a simultaneous [Ca2+]i increase in the areas surrounding somata of TN-GnRH3 neurons. Because the Ca2+ indicator Fura2 was loaded to the whole brain, it is possible that either surrounding non-TN-GnRH3 neurons or dendrites of TN-GnRH3 neurons were excited synchronously. TN-GnRH3 neurons form dense dendritic plexuses around the somata (11); therefore, it is more likely that the spontaneous [Ca2+]i increases in the areas surrounding somata of TN-GnRH3 neurons arise from neural activity in the dendritic plexuses. In contrast, TN-GnRH3 neurons in adulthood showed spontaneous burst firing far less frequently. However, they showed glutamate-induced burst firing, which also increased [Ca2+]i significantly (Fig. 4F and 4H). These results suggest that the TN-GnRH3 neuron in adulthood can also release neuropeptide in response to glutamatergic synaptic inputs (Fig. 9C, a). Figure 9. View largeDownload slide Summary diagram, indicating neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in juvenile and adult medaka brains. (A) Schematic illustration showing the time course of events during development. TN-GnRH3 neurons show burst firing and receive glutamatergic effects, which depend on synaptic inputs and/or receptors. In the green and magenta bars, the darker the colors, the more frequent is burst firing (green) or the stronger is glutamatergic (Glu-ergic) effect (magenta). (B and C) Hypothetical neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in (B) juveniles and (C) adults. (B and C, a) Note that GABAergic synapses are inhibitory (–) to glutamatergic neurons but are excitatory (+) to TN-GnRH3 neurons (see text for details). TN-GnRH3 neurons in juvenile medaka receive persistent glutamatergic input and show high-frequency firing and a [Ca2+]i increase. In adults, when TN-GnRH3 neurons transiently receive glutamatergic synaptic inputs, they show high-frequency firing. Both firing patterns in juveniles and adults are suggested to induce neuropeptide release. In adults, spontaneous burst firing is rarely observed, because of the decrease in expression level or the changes in the subtype of glutamate receptors in TN-GnRH3 neurons or the decrease in the number of glutamatergic synaptic inputs. For the sake of simplicity, only GnRH is illustrated as a representative neuropeptide that is released from TN-GnRH3 neurons, although neuropeptide FF (NPFF) may also be released. (B and C, b) Hypothetical functions of TN-GnRH3 neurons suggested by the current study. TN-GnRH3 neurons receive plenty of environmental information via glutamatergic synaptic inputs, show burst firing, and release neuropeptides. In juvenile medaka, this process may work more frequently than in adults. AMPAR, AMPA receptor; GABA, GABAergic neuron; Glu, glutamatergic neuron; NMDAR, NMDA receptor; TN-GnRH3, TN-GnRH3 neurons. Figure 9. View largeDownload slide Summary diagram, indicating neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in juvenile and adult medaka brains. (A) Schematic illustration showing the time course of events during development. TN-GnRH3 neurons show burst firing and receive glutamatergic effects, which depend on synaptic inputs and/or receptors. In the green and magenta bars, the darker the colors, the more frequent is burst firing (green) or the stronger is glutamatergic (Glu-ergic) effect (magenta). (B and C) Hypothetical neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in (B) juveniles and (C) adults. (B and C, a) Note that GABAergic synapses are inhibitory (–) to glutamatergic neurons but are excitatory (+) to TN-GnRH3 neurons (see text for details). TN-GnRH3 neurons in juvenile medaka receive persistent glutamatergic input and show high-frequency firing and a [Ca2+]i increase. In adults, when TN-GnRH3 neurons transiently receive glutamatergic synaptic inputs, they show high-frequency firing. Both firing patterns in juveniles and adults are suggested to induce neuropeptide release. In adults, spontaneous burst firing is rarely observed, because of the decrease in expression level or the changes in the subtype of glutamate receptors in TN-GnRH3 neurons or the decrease in the number of glutamatergic synaptic inputs. For the sake of simplicity, only GnRH is illustrated as a representative neuropeptide that is released from TN-GnRH3 neurons, although neuropeptide FF (NPFF) may also be released. (B and C, b) Hypothetical functions of TN-GnRH3 neurons suggested by the current study. TN-GnRH3 neurons receive plenty of environmental information via glutamatergic synaptic inputs, show burst firing, and release neuropeptides. In juvenile medaka, this process may work more frequently than in adults. AMPAR, AMPA receptor; GABA, GABAergic neuron; Glu, glutamatergic neuron; NMDAR, NMDA receptor; TN-GnRH3, TN-GnRH3 neurons. Mechanisms of juvenile-specific burst firing in TN-GnRH3 neurons Both juvenile-specific burst firing and burst firing by bicuculline application in TN-GnRH3 neurons were induced by glutamatergic synaptic inputs (Figs. 2, 4, and 6). These firings also showed time-dependent changes during post-hatching development (Figs. 1F and 7B). Thus, juvenile-specific burst firing is suggested to be induced by almost the same excitatory mechanism as the burst firing by bicuculline application. In other words, global bicuculline application (bath application) in juveniles is considered to block the persistent GABAergic inhibition to the glutamatergic neurons (Fig. 6A), and this disinhibitory action of bicuculline may have facilitated the glutamate receptor-mediated burst firing of TN-GnRH3 neurons in juvenile medaka (Fig. 9B, a). Juvenile-specific burst firing is suggested to be induced when such blockade of the persistent GABAergic inhibition to the glutamatergic neurons occurs. As noted in the results, our study showed that glutamatergic synaptic inputs are important for juvenile-specific burst firing. Considering the fact that bicuculline did not induce burst firing under persistent blockage of AMPA- or NMDA-type glutamate receptors (Fig. 6D, a), it is suggested that AMPA- and NMDA-type glutamatergic synaptic inputs act in concert to generate the juvenile-specific spontaneous burst firing. Interestingly, a similar mechanism for burst generation has also been proposed for the central pattern generator circuitry for locomotion in the spinal cord (33, 34). Thus, AMPA- and NMDA-type glutamatergic synaptic inputs are suggested to play an important role in juvenile-specific burst firing. Furthermore, the percentage of burst firing was changed during post-hatching development (Fig. 1). Our study also demonstrated that CNQX affected spontaneous activities of TN-GnRH3 neurons in juveniles more severely than in adulthood. With the consideration of the mechanism underlying the decrease in juvenile-specific burst firing with age, this decrease in burst firing may be due to changes in the level of glutamate receptor expression, in the subtype of glutamate receptors, or in the density of glutamatergic presynaptic inputs onto TN-GnRH3 neurons. It has been reported that subtypes of NMDA receptors in the brain changes during early development in mammals (35). It is possible that the number or subtype of NMDA receptors in TN-GnRH3 neurons is different between juvenile and adulthood (Fig. 9B, a and 9C, a), because each subtype of the NMDA receptor has specific conductance and other characteristics (35). On the other hand, the GABAA receptor antagonist did not induce burst firing in adulthood (Fig. 7A). If TN-GnRH3 neurons in juveniles receive more glutamatergic inputs than in adulthood, it is suggested that TN-GnRH3 neurons in adulthood rarely show spontaneous burst firing, because both glutamatergic synaptic inputs and persistent GABAergic inhibition decrease during post-hatching development (Fig. 9C, a). If the mechanism underlying the change in firing pattern during post-hatching development will be further clarified in the future, it should help to deepen the understanding of the nature of peptidergic neurons. Functional significance of juvenile-specific burst firing in TN-GnRH3 neurons Our final discussion will focus on the functional significance of juvenile-specific burst firing in TN-GnRH3 neurons. A histological study using a trans-synaptic tracer suggested that TN-GnRH3 neurons receive afferent input, mainly from brain regions involved in olfactory, visual, and somatosensory information processing (36). Considering our finding that juvenile-specific burst firing needs glutamatergic synaptic inputs, TN-GnRH3 neurons may receive these inputs from such brain regions. In other words, our study suggests that TN-GnRH3 neurons in juveniles receive plenty of environmental information via glutamatergic synaptic inputs, show burst firing, and release neuropeptides, far more actively than in adults (Fig. 9B, b and 9C, b). Previous studies of TN-GnRH3 neurons in adulthood suggested that the TN-GnRH3 neuron is important for motivation control for sexual behavior (8–10), which may be via neuromodulation by GnRH3 peptide (13, 14, 16). Thus, one possible function of juvenile-specific burst firing also may be neuromodulation, such as in adults. If so, the GnRH3 peptide that is released in juvenile animals is suggested to play a key role in neuromodulation of age-independent phenomena, such as energy homeostasis, mood, motivation, etc. (4, 6). Although some vertebrates (mammals, birds, eel, etc.) have lost gnrh3, it has been proposed that gnrh1 compensates for the loss of gnrh3 and is expressed in the TN neurons (7). The GnRH1 peptide expressed in the TN of such juvenile animals may serve the same functions as described previously. On the other hand, TN-GnRH3 neurons release neuropeptide FF (NPFF) in addition to the GnRH3 peptide (37). Further investigation of the role of GnRH and NPFF in neuromodulation is necessary to determine if these peptides are released separately or simultaneously to exert differential functions. Another possible novel function of TN-GnRH3 neurons may be neurotrophic and/or maturational action. Recently, it has been reported that some neuropeptides also function as neurotrophic factors during early development (38–40). For example, calcitonin gene-related peptide acts for survival and differentiation of neurons during early development and neuromodulation in adulthood (39). Spontaneous bursting activity before the onset of hearing promotes the survival and maturation of auditory neurons and the refinement of synaptic connections in auditory nuclei (41). Therefore, neuropeptide release by juvenile-specific burst firing may also have neurotrophic and/or maturational action, although other functions are also possible. It has been reported that animal behavior in adulthood can change in response to their environment in the juvenile period, as well as that surrounding the animals (42, 43). It is therefore possible that juvenile-specific burst firing in TN-GnRH3 neurons plays a key role in such environmental factors during the juvenile period to modulate behaviors in adulthood. In summary, we found that TN-GnRH3 neurons show juvenile-specific, spontaneous burst firing, which may have novel functions, in addition to neuromodulation. Abbreviations: Abbreviations: ACSF artificial cerebrospinal fluid AM acetoxymethyl AMPA α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ANOVA analysis of variance [Ca2+]i intracellular Ca2+ concentration CNQX 6-cyano-7-nitroquinoxaline-2,3-dione d-AP5 d-(–)-2-amino-5-phosphonopentanoic acid EGFP enhanced green fluorescent protein EM electron multiplying GABA γ-aminobutyric acid GnRH gonadotropin-releasing hormone NMDA N-methyl-d-aspartate NPFF neuropeptide FF TN terminal nerve TN-GnRH3 terminal nerve gonadotropin-releasing hormone 3 Δratio ratio change of fluorescence intensity Acknowledgments We thank Drs. Shinji Kanda (The University of Tokyo), Masaharu Hasebe (Osaka University), and Kazuhiko Yamaguchi (RIKEN Brain Science Institute) for valuable discussions. We also appreciate Ms. Miho Kyokuwa, Ms. Hisako Kohno, and Ms. Fumika Muguruma (The University of Tokyo) for their gentle care of the fish. Financial Support: The current study was supported by Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research (JSPS KAKENHI) Grants 26221104 and 17K15157. Author Contributions: C.U. and Y.O. designed and drafted the present work. C.U. collected, analyzed, and interpreted data and revised the paper critically for important intellectual content. Both authors approved the final version of the manuscript and agreed 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. Disclosure Summary: The authors have nothing to disclose. References 1. de Kloet ER, Joëls M, Holsboer F. Stress and the brain: from adaptation to disease. Nat Rev Neurosci . 2005; 6( 6): 463– 475. Google Scholar CrossRef Search ADS PubMed  2. Iwamatsu T. The Integrated Book for the Biology of Medaka . Okayama, Japan: University Education Press; 1997. 3. Edwards DA, Griffis KT, Tardivel C. Olfactory bulb removal: effects on sexual behavior and partner-preference in male rats. Physiol Behav . 1990; 48( 3): 447– 450. Google Scholar CrossRef Search ADS PubMed  4. Argiolas A, Melis MR. Neuropeptides and central control of sexual behaviour from the past to the present: a review. Prog Neurobiol . 2013; 108: 80– 107. Google Scholar CrossRef Search ADS PubMed  5. Dornan WA, Malsbury CW. Neuropeptides and male sexual behavior. Neurosci Biobehav Rev . 1989; 13( 1): 1– 15. Google Scholar CrossRef Search ADS PubMed  6. van den Pol AN. Neuropeptide transmission in brain circuits. Neuron . 2012; 76( 1): 98– 115. Google Scholar CrossRef Search ADS PubMed  7. Okubo K, Nagahama Y. Structural and functional evolution of gonadotropin-releasing hormone in vertebrates. Acta Physiol (Oxf) . 2008; 193( 1): 3– 15. Google Scholar CrossRef Search ADS PubMed  8. Wirsig CR, Leonard CM. Terminal nerve damage impairs the mating behavior of the male hamster. Brain Res . 1987; 417( 2): 293– 303. Google Scholar CrossRef Search ADS PubMed  9. Yamamoto N, Oka Y, Kawashima S. Lesions of gonadotropin-releasing hormone-immunoreactive terminal nerve cells: effects on the reproductive behavior of male dwarf gouramis. Neuroendocrinology . 1997; 65( 6): 403– 412. Google Scholar CrossRef Search ADS PubMed  10. Okuyama T, Yokoi S, Abe H, Isoe Y, Suehiro Y, Imada H, Tanaka M, Kawasaki T, Yuba S, Taniguchi Y, Kamei Y, Okubo K, Shimada A, Naruse K, Takeda H, Oka Y, Kubo T, Takeuchi H. A neural mechanism underlying mating preferences for familiar individuals in medaka fish. Science . 2014; 343( 6166): 91– 94. Google Scholar CrossRef Search ADS PubMed  11. Oka Y, Matsushima T. Gonadotropin-releasing hormone (GnRH)-immunoreactive terminal nerve cells have intrinsic rhythmicity and project widely in the brain. J Neurosci . 1993; 13( 5): 2161– 2176. Google Scholar PubMed  12. von Bartheld CS, Meyer DL. Tracing of single fibers of the nervus terminalis in the goldfish brain. Cell Tissue Res . 1986; 245( 1): 143– 158. Google Scholar CrossRef Search ADS PubMed  13. Eisthen HL, Delay RJ, Wirsig-Wiechmann CR, Dionne VE. Neuromodulatory effects of gonadotropin releasing hormone on olfactory receptor neurons. J Neurosci . 2000; 20( 11): 3947– 3955. Google Scholar PubMed  14. Kawai T, Abe H, Akazome Y, Oka Y. Neuromodulatory effect of GnRH on the synaptic transmission of the olfactory bulbar neural circuit in goldfish, Carassius auratus. J Neurophysiol . 2010; 104( 6): 3540– 3550. Google Scholar CrossRef Search ADS PubMed  15. Maruska KP, Tricas TC. Gonadotropin-releasing hormone (GnRH) modulates auditory processing in the fish brain. Horm Behav . 2011; 59( 4): 451– 464. Google Scholar CrossRef Search ADS PubMed  16. Umatani C, Misu R, Oishi S, Yamaguchi K, Abe H, Oka Y. GnRH suppresses excitability of visual processing neurons in the optic tectum. J Neurophysiol . 2015; 114( 5): 2775– 2784. Google Scholar CrossRef Search ADS PubMed  17. Amano M, Oka Y, Kitamura S, Ikuta K, Aida K. Ontogenic development of salmon GnRH and chicken GnRH-II systems in the brain of masu salmon (Oncorhynchus masou). Cell Tissue Res . 1998; 293( 3): 427– 434. Google Scholar CrossRef Search ADS PubMed  18. Amano M, Okubo K, Yamanome T, Oka Y, Kawaguchi N, Aida K, Yamamori K. Ontogenic development of three GnRH systems in the brain of a pleuronectiform fish, barfin flounder. Zool Sci . 2004; 21( 3): 311– 317. Google Scholar CrossRef Search ADS PubMed  19. Takahashi A, Islam MS, Abe H, Okubo K, Akazome Y, Kaneko T, Hioki H, Oka Y. Morphological analysis of the early development of telencephalic and diencephalic gonadotropin-releasing hormone neuronal systems in enhanced green fluorescent protein-expressing transgenic medaka lines. J Comp Neurol . 2016; 524( 4): 896– 913. Google Scholar CrossRef Search ADS PubMed  20. Karigo T, Oka Y. Neurobiological study of fish brains gives insights into the nature of gonadotropin-releasing hormone 1-3 neurons. Front Endocrinol (Lausanne) . 2013; 4: 177. Google Scholar PubMed  21. Hasebe M, Kanda S, Shimada H, Akazome Y, Abe H, Oka Y. Kiss1 neurons drastically change their firing activity in accordance with the reproductive state: insights from a seasonal breeder. Endocrinology . 2014; 155( 12): 4868– 4880. Google Scholar CrossRef Search ADS PubMed  22. Peng YY, Zucker RS. Release of LHRH is linearly related to the time integral of presynaptic Ca2+ elevation above a threshold level in bullfrog sympathetic ganglia. Neuron . 1993; 10( 3): 465– 473. Google Scholar CrossRef Search ADS PubMed  23. Oka Y, Ichikawa M. Ultrastructure of the ganglion cells of the terminal nerve in the dwarf gourami (Colisa lalia). J Comp Neurol . 1991; 304( 2): 161– 171. Google Scholar CrossRef Search ADS PubMed  24. Nakane R, Oka Y. Excitatory action of GABA in the terminal nerve gonadotropin-releasing hormone neurons. J Neurophysiol . 2010; 103( 3): 1375– 1384. Google Scholar CrossRef Search ADS PubMed  25. Sim JA, Skynner MJ, Herbison AE. Heterogeneity in the basic membrane properties of postnatal gonadotropin-releasing hormone neurons in the mouse. J Neurosci . 2001; 21( 3): 1067– 1075. Google Scholar PubMed  26. Constantin S, Iremonger KJ, Herbison AE. In vivo recordings of GnRH neuron firing reveal heterogeneity and dependence upon GABAA receptor signaling. J Neurosci . 2013; 33( 22): 9394– 9401. Google Scholar CrossRef Search ADS PubMed  27. Iremonger KJ, Porteous R, Herbison AE. Spike and neuropeptide-dependent mechanisms control GnRH neuron nerve terminal Ca2+ over diverse time scales. J Neurosci . 2017; 37( 12): 3342– 3351. Google Scholar CrossRef Search ADS PubMed  28. Campos P, Herbison AE. Optogenetic activation of GnRH neurons reveals minimal requirements for pulsatile luteinizing hormone secretion. Proc Natl Acad Sci USA . 2014; 111( 51): 18387– 18392. Google Scholar CrossRef Search ADS PubMed  29. Han SY, McLennan T, Czieselsky K, Herbison AE. Selective optogenetic activation of arcuate kisspeptin neurons generates pulsatile luteinizing hormone secretion. Proc Natl Acad Sci USA . 2015; 112( 42): 13109– 13114. Google Scholar CrossRef Search ADS PubMed  30. Hasebe M, Oka Y. High-frequency firing activity of GnRH1 neurons in female medaka induces the release of GnRH1 peptide from their nerve terminals in the pituitary. Endocrinology . 2017; 158( 8): 2603– 2617. Google Scholar CrossRef Search ADS PubMed  31. Moenter SM, DeFazio AR, Pitts GR, Nunemaker CS. Mechanisms underlying episodic gonadotropin-releasing hormone secretion. Front Neuroendocrinol . 2003; 24( 2): 79– 93. Google Scholar CrossRef Search ADS PubMed  32. Hage TA, Khaliq ZM. Tonic firing rate controls dendritic Ca2+ signaling and synaptic gain in substantia nigra dopamine neurons. J Neurosci . 2015; 35( 14): 5823– 5836. Google Scholar CrossRef Search ADS PubMed  33. Wallén P, Grillner S. N-Methyl-d-aspartate receptor-induced, inherent oscillatory activity in neurons active during fictive locomotion in the lamprey. J Neurosci . 1987; 7( 9): 2745– 2755. Google Scholar PubMed  34. Grillner S, Deliagina T, Ekeberg O, el Manira A, Hill RH, Lansner A, Orlovsky GN, Wallén P. Neural networks that co-ordinate locomotion and body orientation in lamprey. Trends Neurosci . 1995; 18( 6): 270– 279. Google Scholar CrossRef Search ADS PubMed  35. Cull-Candy S, Brickley S, Farrant M. NMDA receptor subunits: diversity, development and disease. Curr Opin Neurobiol . 2001; 11( 3): 327– 335. Google Scholar CrossRef Search ADS PubMed  36. Yamamoto N, Ito H. Afferent sources to the ganglion of the terminal nerve in teleosts. J Comp Neurol . 2000; 428( 2): 355– 375. Google Scholar CrossRef Search ADS PubMed  37. Saito TH, Nakane R, Akazome Y, Abe H, Oka Y. Electrophysiological analysis of the inhibitory effects of FMRFamide-like peptides on the pacemaker activity of gonadotropin-releasing hormone neurons. J Neurophysiol . 2010; 104( 6): 3518– 3529. Google Scholar CrossRef Search ADS PubMed  38. Rogge G, Jones D, Hubert GW, Lin Y, Kuhar MJ. CART peptides: regulators of body weight, reward and other functions [published corrigendum appears in Nat Rev Neurosci. 2010;11:218]. Nat Rev Neurosci . 2008; 9( 10): 747– 758. Google Scholar CrossRef Search ADS PubMed  39. Denis-Donini S. Expression of dopaminergic phenotypes in the mouse olfactory bulb induced by the calcitonin gene-related peptide. Nature . 1989; 339( 6227): 701– 703. Google Scholar CrossRef Search ADS PubMed  40. Hökfelt T. Neuropeptides in perspective: the last ten years. Neuron . 1991; 7( 6): 867– 879. Google Scholar CrossRef Search ADS PubMed  41. Tritsch NX, Bergles DE. Developmental regulation of spontaneous activity in the Mammalian cochlea. J Neurosci . 2010; 30( 4): 1539– 1550. Google Scholar CrossRef Search ADS PubMed  42. Lupien SJ, McEwen BS, Gunnar MR, Heim C. Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nat Rev Neurosci . 2009; 10( 6): 434– 445. Google Scholar CrossRef Search ADS PubMed  43. Fleming IA, Lamberg A, Jonsson B. Effects of early experience on the reproductive performance of Atlantic salmon. Behav Ecol . 1997; 8( 5): 470– 480. Google Scholar CrossRef Search ADS   Copyright © 2018 Endocrine Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Endocrinology Oxford University Press

Juvenile-Specific Burst Firing of Terminal Nerve GnRH3 Neurons Suggests Novel Functions in Addition to Neuromodulation

Loading next page...
 
/lp/ou_press/juvenile-specific-burst-firing-of-terminal-nerve-gnrh3-neurons-vavuxkrsau
Publisher
Endocrine Society
Copyright
Copyright © 2018 Endocrine Society
ISSN
0013-7227
eISSN
1945-7170
D.O.I.
10.1210/en.2017-03210
Publisher site
See Article on Publisher Site

Abstract

Abstract Peptidergic neurons are suggested to play a key role in neuromodulation of animal behaviors in response to sensory cues in the environment. Terminal nerve gonadotropin-releasing hormone 3 (TN-GnRH3) neurons are thought to be one of the peptidergic neurons important for such neuromodulation in adult vertebrates. On the other hand, it has been reported that TN-GnRH3 neurons are labeled by a specific GnRH3 antibody from early developmental stages to adulthood and are thus suggested to produce mature GnRH3 peptide even in the early developmental stages. However, it remains unknown when TN-GnRH3 neurons show spontaneous burst firing, which is suggested to be involved in neuropeptide release. Using a whole-brain in vitro preparation of gnrh3:enhanced green fluorescent protein (EGFP) medaka fish, we first recorded spontaneous firings of TN-GnRH3 neurons after hatching to adulthood. Contrary to what one would expect from their neuromodulatory functions—that TN-GnRH3 neurons are more active in adulthood—TN-GnRH3 neurons in juveniles showed spontaneous burst firing more frequently than in adulthood (juvenile-specific burst firing). Ca2+ imaging of TN-GnRH3 neurons in juveniles may further suggest that juvenile-specific burst firing triggers neuropeptide release. Furthermore, juvenile-specific burst firing was suggested to be induced by blocking persistent GABAergic inhibition to the glutamatergic neurons, which leads to an increase in glutamatergic synaptic inputs to TN-GnRH3 neurons. The present study reports that peptidergic neurons show juvenile-specific burst firing involved in triggering peptide release and suggests that juvenile TN-GnRH3 neurons have novel functions, in addition to neuromodulation. Animals change their behavior in response to sensory cues in the environment, as well as the physiological status of their own. For example, various behaviors are modulated by stress (1), and some social behaviors, such as schooling and sexual behaviors (2, 3), are modulated by various sensory cues in the environment, such as the temperature, day length, etc. However, the neural mechanisms of how the neural circuit integrates such information from the environment and modulates their behaviors remain unknown. It is generally accepted that such behavioral changes are induced by modulations of neural activities in the brain (generally called “neuromodulation”). Although neural activities are known to be modulated by various molecules (generally called “neuromodulators”), it has been proposed that neuropeptides play important roles in such neuromodulation (4–6). The terminal nerve (TN) gonadotropin-releasing hormone (GnRH) 3 (TN-GnRH3) neuron, which is one of the phylogenetically conserved peptidergic neurons among vertebrates (7), has been suggested to act as a neuromodulatory cell to regulate motivation for sexual behavior (8–10). This neuron projects to wide areas of the brain (11, 12), and GnRH3 peptide that is released from TN-GnRH3 neurons has been demonstrated physiologically to modulate neuronal functions involved in sensory processing (13–16). Such neuromodulation has been analyzed by experiments using only adult fish and suggested to lead finally to behavioral changes. On the other hand, it has been reported that TN-GnRH3 neurons are labeled by GnRH3 antibody from early developmental stages to adulthood and are suggested to produce a mature GnRH3 peptide (17, 18), even in the early developmental stages. Though it has been suggested that spontaneous high-frequency firing, such as bursting, is important for neuropeptide release, it remains unknown during which developmental stages TN-GnRH3 neurons show burst firing. Here, we recorded spontaneous firings of TN-GnRH3 neurons from gnrh3:enhanced green fluorescent protein (EGFP) medaka fish after hatching to adulthood. An advantage of using medaka as a model is its small and transparent brain that can be used in a whole-brain in vitro preparation. In this preparation, intact neural circuits are maintained for electrical recording and imaging. Contrary to what one would expect from their neuromodulatory functions in adults, TN-GnRH3 neurons in juvenile fish, far more frequently than in adult fish, showed burst firing (juvenile-specific burst firing), which was accompanied by significant Ca2+ signals that may be suggestive of active neuropeptide release. Our study reports that peptidergic neurons show juvenile-specific burst firing and suggests that juvenile TN-GnRH3 neurons have novel functions, in addition to neuromodulation. Materials and Methods Female and male gnrh3:EGFP medaka (Oryzias latipes) (19) were maintained at 27°C at a 24-hour dark cycle before hatching and a 14:10 light:dark cycle after hatching. The fish were fed daily with live brine shrimp and flake food. We used juvenile medaka (three; ∼9 weeks after fertilization) and adult medaka (>9 weeks after fertilization). All physiological studies were performed at ∼20°C to 27°C. All procedures were performed in accordance with the guideline principles for the care established by the Physiological Society of Japan, and the protocols were approved by the Animal Care and Use Committee of the University of Tokyo (permission number 15-3, 17-1). Electrophysiology The fish were anesthetized by 0.02% 3-aminobenzoic acid ethyl ester (MS-222; Sigma Aldrich, St. Louis, MO). They were quickly euthanized by decapitation, and the whole brain was isolated. The isolated brain was put in an artificial cerebrospinal fluid (ACSF) consisting of (in mM) 134 NaCl, 2.9 KCl, 1.2 MgCl2, 2.1 CaCl2, 10 HEPES, and 10 glucose (pH 7.4 adjusted with NaOH and ∼300 mOsm adjusted with sucrose). The whole brain of this fish can be maintained in vitro for a long period without oxygenation (20). Then, the ventral meningeal membrane was carefully removed. For the whole-cell recording, internal solution was the following (mM): K-gluconate 112.5, KCl 17.5, NaCl 4, EGTA 1, MgCl2 1, CaCl2 0.5, and HEPES 10 (pH 7.2 adjusted with KOH and ∼290 mOsm adjusted with sucrose). The junction potential was −13.8 mV, and the membrane potentials were adjusted by using this value. The tip resistance of patch electrodes (GD-1.5; Narishige, Tokyo, Japan) in ACSF was ∼10 to 15 MΩ. For the on-cell clamp recording, the pipette solution was the same as ACSF. Both recordings were performed using an Axopatch 200B patch-clamp amplifier (Molecular Devices, Sunnyvale, CA). The whole-cell current-clamp recordings were digitized (10 kHz) and stored on a computer using Digidata 1322A and pCLAMP 9.2 software (Molecular Devices). We detected action potentials by using Clampfit 10 software (Molecular Devices) and analyzed firing pattern by referring to Hasebe et al. (21). There was not any difference between firing patterns recorded by on-cell and those recorded by whole-cell patch clamp. Bath application of drugs was performed by mixing drugs in the ACSF, and puffer application (100 ms, ∼60 to 70 kPa) was performed by using the electric microinjector IM-31 (Narishige). The tip diameter of glass pipette (GD-1; Narishige) for puffer application was 1 to 5 μm, and the pipette was placed ∼10 to 30 μm from the cell bodies of TN-GnRH3 neurons. Ca2+ imaging The whole brain was isolated as described previously. Some neurons and glial cells covering the ventral region of the somata of TN-GnRH3 neurons were mechanically removed with the aid of a glass pipette under a microscope to expose the surface of the TN-GnRH3 neurons. Then, the brain was incubated in ACSF at room temperature for 30 minutes. We prepared Fura2-acetoxymethyl (AM) solution [ACSF with 10 μM Fura2-AM (Dojindo, Kumamoto, Japan) and 0.04% Cremophor-EL (Sigma Aldrich)] and vortexed it for 10 seconds. After the incubation, we put the brain into a new 1.5-mL tube containing a Fura2-AM solution. The tube was shaken mildly (30g) at 30°C for 1 hour. The brain was washed by incubation with normal ACSF at room temperature for 15 minutes and perfusion of normal ACSF at room temperature for 15 minutes. After that, we placed the brain in a recording chamber with ACSF in the ventral side-up direction. The preparation was perfused with ACSF at a continuous flow rate (2 mL/min). Bath application of drugs was performed by mixing drugs in the ACSF, and puffer application was performed as described in the previous section by using the electric microinjector IM-31 (Narishige). Fura2 and EGFP fluorescence was detected by the Chroma 74000 filter set (D480×, D340×, D380× exciter; 505DCLP dichroic mirror; HQ535/50m emitter), housed in the Lambda DG-4 Xenon light source, excitation filter exchanger (Sutter Instruments, Novato, CA), and the BX-51WI upright epifluorescence microscope (Olympus, Tokyo, Japan). The fluorescence was recorded using the Metafluor imaging software (Molecular Devices) and each camera: QuantEM 512SC electron multiplying (EM) charge-coupled device camera [exposure: 80 (340 and 380 nm) and 50 (480 nm) ms; 5 MHz EM gain; interval: 2 s; Photometrics, Tucson, AZ; see Figs. 4 and 8A–8C below] or Zyla 4.2 scientific complementary metal-oxide-semiconductor camera [exposure: 150 (340 and 380 nm) and 80 (480 nm) ms; 5 MHz EM gain; interval: 2 s; Andor Technology, Belfast, Northern Ireland; see Fig. 8D and 8E below]. For EGFP, the intensity ratio of emission (at 510 nm) from the alternating 480-nm excitation was monitored. For Fura2, the intensity ratio of emission (at 510 nm) from the alternating 340- and 380-nm excitation was monitored. The imaging data were analyzed by the following method using ImageJ software (Research Resource Identifier: SCR_003070; National Institutes of Health, Bethesda, MD) with the McMaster Biophotonics Facility ImageJ plug-ins (Tony Collins, McMaster University, Hamilton, ON, Canada). The basal ratio of fluorescence intensity (R)0 was calculated as the average of five frames from the first one. The ratio change of fluorescence intensity (Δratio) was calculated as R − R0. Peak Δratio was calculated as [(maximum Δratio for 30 seconds after puffer application) − (average Δratio for four frames from approximately eight frames before application)]. We set up the region of interest on one of the TN-GnRH3 neurons by referring to the EGFP images and averaged those Δratio of the neurons in one hemisphere of telencephalon (approximately two to eight neurons were identified in one hemisphere). Drugs γ-Aminobutyric acid (GABA) was purchased from Wako (Osaka, Japan). Glutamate (l-glutamic acid monosodium salt hydrate), muscimol, and bicuculline were purchased from Sigma Aldrich. d-(–)-2-amino-5-phosphonopentanoic acid (d-AP5) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) were purchased from Tocris (Bristol, UK). CNQX and bicuculline were dissolved with dimethyl sulfoxide as 20 mM and were diluted to ACSF. The final concentration of dimethyl sulfoxide in ACSF was 0.1%. Statistics All animal experiments were performed on fish of both sexes. Statistical analyses were performed by using Kyplot5 (Kyence, Tokyo, Japan) and Igor Pro 6 software (WaveMetrics, Lake Oswego, OR) with a Taro tool (Igor macro set, written by Dr. Taro Ishikawa, Jikei University School of Medicine, Tokyo, Japan). Analyses include the following: analysis of percentage of burst-firing neurons: Jonckheere trend test (decrease; see Fig. 1); analysis of effect of glutamatergic synapse: analysis of variance (ANOVA) with Dunnett post hoc test (see Figs. 2 and 3) or Steel test (see Fig. 4); analysis of bicuculline-induced burst firing: ANOVA with Dunnett post hoc test (see Figs. 5–7); analysis of the intracellular Ca2+ concentration ([Ca2+]i) increase during bicuculline-induced burst firing: paired Wilcoxon test (see Fig. 8). Each detailed P value is shown in figure legends. All data in the current study are as means ± standard error of the mean. Results TN-GnRH3 neurons change firing patterns during post-hatching development The spontaneous neural activities of EGFP-labeled TN-GnRH3 neurons from fish after hatching to adulthood were recorded by on-cell patch clamp (Fig. 1). The firing patterns were classified as regular, irregular, and burst firing (see Fig. 1 legend for definitions). The burst firing was defined as firing with more than three consecutive spikes at >3 Hz and with an interburst interval longer than 1 second, which was based on the regular firing frequency in adults (see later). We could not find silent neurons. In adult fish, it has been reported that TN-GnRH3 neurons most frequently show regular firing (pacemaker activity) and show irregular or burst firing on rare occasions (11). In contrast to the firing patterns in adulthood (medaka were maintained in pairs), we found that most TN-GnRH3 neurons in juvenile medaka showed irregular or burst firing (Fig. 1). As shown in Fig. 1F, nearly 60% of neurons in fish, just after hatching, were categorized as burst firing, and such “juvenile-specific spontaneous burst firing” was observed in both males and females. The intraburst firing frequency was >5 Hz, from 3 to 7 weeks after fertilization (Table 1, top row). Figure 1G shows the percentage of burst-firing neurons in a fish selected from Fig. 1F. The percentage of neurons exhibiting burst firing decreased as fish grew older (Jonckheere trend test, P < 0.001). The change in the percentage of burst-firing neurons started later than the early stages of sexual maturation. In addition, as fish aged, the percentage of regular firing neurons increased to 100% by 12.5 weeks after fertilization (Fig. 1F). Furthermore, the firing frequency of neurons with a regular firing pattern tended to be lower in juvenile medaka than in adult medaka (Table 1, bottom row). Figure 1. View largeDownload slide TN-GnRH3 neurons during post-hatching development show juvenile-specific spontaneous burst firing. (A) The scheme of medaka post-hatching development. (B–E) Representative firing patterns in medaka. The firing patterns were (B and C) recorded from 4 weeks after fertilization; (D) from 7 weeks after fertilization; and (E) from 12.5 weeks after fertilization. The scale is shown in the upper right for (B)–(E). (F) Firing patterns of TN-GnRH3 neurons in one fish for each stage. We used the average value for each pattern. The classification of the firing pattern is as follows: burst firing, consecutive spikes more than three at >3 Hz (firing frequency in adult medaka is usually lower than 2 Hz) and interburst interval longer than 1 second (longer than that of general TN-GnRH3 neurons in adults); regular firing, coefficient of variation of the interburst interval <0.65 (the coefficient of variation is defined as the ratio of the standard deviation to the mean interburst interval); irregular firing, others. Juvenile: 3 weeks after fertilization, 9 fish; 4 weeks, 9 fish; 5 weeks, 7 fish; 6 weeks, 8 fish; 7 weeks, 8 fish. Adult: >12.5 weeks, 4 fish. (G) The percentage of burst firing was selected from (F). Jonckheere trend test, P < 0.001. Figure 1. View largeDownload slide TN-GnRH3 neurons during post-hatching development show juvenile-specific spontaneous burst firing. (A) The scheme of medaka post-hatching development. (B–E) Representative firing patterns in medaka. The firing patterns were (B and C) recorded from 4 weeks after fertilization; (D) from 7 weeks after fertilization; and (E) from 12.5 weeks after fertilization. The scale is shown in the upper right for (B)–(E). (F) Firing patterns of TN-GnRH3 neurons in one fish for each stage. We used the average value for each pattern. The classification of the firing pattern is as follows: burst firing, consecutive spikes more than three at >3 Hz (firing frequency in adult medaka is usually lower than 2 Hz) and interburst interval longer than 1 second (longer than that of general TN-GnRH3 neurons in adults); regular firing, coefficient of variation of the interburst interval <0.65 (the coefficient of variation is defined as the ratio of the standard deviation to the mean interburst interval); irregular firing, others. Juvenile: 3 weeks after fertilization, 9 fish; 4 weeks, 9 fish; 5 weeks, 7 fish; 6 weeks, 8 fish; 7 weeks, 8 fish. Adult: >12.5 weeks, 4 fish. (G) The percentage of burst firing was selected from (F). Jonckheere trend test, P < 0.001. Table 1. Firing Frequency of TN-GnRH3 Neurons in Burst or Regular Firing   Juvenile   Adult   Weeks After Fertilization  3  4  5  6  7  12.5  Intraburst firing frequency, Hz  12.37 ± 1.20 (n = 7)  9.51 ± 0.04 (n = 7)  11.14 ± 0.84 (n = 6)  5.54 ± 0.84 (n = 4)  6.21 ± 0.88 (n = 2)  None  Avg. frequency in regular firing, Hz  0.38 ± 0.08 (n = 5)  0.40 ± 0.13 (n = 5)  0.28 ± 0.04 (n = 5)  0.64 ± 0.08 (n = 7)  0.62 ± 0.12 (n = 8)  0.92 ± 0.08 (n = 4)    Juvenile   Adult   Weeks After Fertilization  3  4  5  6  7  12.5  Intraburst firing frequency, Hz  12.37 ± 1.20 (n = 7)  9.51 ± 0.04 (n = 7)  11.14 ± 0.84 (n = 6)  5.54 ± 0.84 (n = 4)  6.21 ± 0.88 (n = 2)  None  Avg. frequency in regular firing, Hz  0.38 ± 0.08 (n = 5)  0.40 ± 0.13 (n = 5)  0.28 ± 0.04 (n = 5)  0.64 ± 0.08 (n = 7)  0.62 ± 0.12 (n = 8)  0.92 ± 0.08 (n = 4)  Abbreviation: Avg., average. View Large Glutamatergic synaptic inputs are involved in juvenile-specific burst firing in TN-GnRH3 neurons To examine whether the juvenile-specific burst firing mentioned in the previous section is induced by synaptic inputs, we first recorded firing activities of TN-GnRH3 neurons in juvenile medaka with an ACSF containing glutamate receptor antagonist CNQX (Fig. 2). The application of CNQX decreased the occurrence of burst firing (Fig. 2B and 2D). The median firing frequency of bursting neurons in juvenile terminal nerve was also decreased by CNQX application (Fig. 2E). On the other hand, CNQX did not significantly affect the firing frequency of TN-GnRH3 neurons in adults (Fig. 3). Thus, glutamatergic presynaptic inputs are suggested to be important for the juvenile-specific burst firing in TN-GnRH3 neurons. Figure 2. View largeDownload slide Bath application of CNQX decreases the number of burst firing in juvenile TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of juvenile medaka for 7 minutes. (A–C) Representative recordings, (D) normalized percentage of burst firing, and (E) normalized median frequency from juvenile (∼3 to 4 weeks after fertilization, n = 5). One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz). One arrow [in (A) and (C)] shows one bout of burst. Inset shows the time course of the experiment, and firing activities during the gray bars (for 2 minutes) were used for the analysis in (D). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX, P = 0.0120; Before vs Washout, P = 0.2092. (E) The vertical axis shows the median frequency normalized by that before 20 μM CNQX application (n = 5). The median frequency of each neuron was calculated by using instantaneous frequency of all spikes. ANOVA with Dunnett post hoc test, ***P < 0.001; Before vs CNQX, P = 0.0006; Before vs Washout, P = 0.0787. Figure 2. View largeDownload slide Bath application of CNQX decreases the number of burst firing in juvenile TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of juvenile medaka for 7 minutes. (A–C) Representative recordings, (D) normalized percentage of burst firing, and (E) normalized median frequency from juvenile (∼3 to 4 weeks after fertilization, n = 5). One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz). One arrow [in (A) and (C)] shows one bout of burst. Inset shows the time course of the experiment, and firing activities during the gray bars (for 2 minutes) were used for the analysis in (D). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX, P = 0.0120; Before vs Washout, P = 0.2092. (E) The vertical axis shows the median frequency normalized by that before 20 μM CNQX application (n = 5). The median frequency of each neuron was calculated by using instantaneous frequency of all spikes. ANOVA with Dunnett post hoc test, ***P < 0.001; Before vs CNQX, P = 0.0006; Before vs Washout, P = 0.0787. Figure 3. View largeDownload slide Bath application of CNQX does not affect firing frequency in adult TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of adult medaka for 4 minutes. (A–C) Representative recordings and (D) normalized frequency (n = 6) from adults (>12 weeks after fertilization). ANOVA with Dunnett post hoc test, n.s.; Before vs CNQX, P = 0.8417; Before vs Washout, P = 0.8002. n.s., not significant. Figure 3. View largeDownload slide Bath application of CNQX does not affect firing frequency in adult TN-GnRH3 neurons. CNQX (20 μM, dissolved in ACSF) was applied to the TN-GnRH3 neurons of adult medaka for 4 minutes. (A–C) Representative recordings and (D) normalized frequency (n = 6) from adults (>12 weeks after fertilization). ANOVA with Dunnett post hoc test, n.s.; Before vs CNQX, P = 0.8417; Before vs Washout, P = 0.8002. n.s., not significant. The mimicking of direct glutamatergic synaptic input can trigger juvenile-specific burst firing and induce a [Ca2+]i increase, which may suggest neuropeptide release To examine whether glutamatergic synaptic inputs can actually induce burst firing, we performed puffer application of glutamate to mimic such synaptic inputs. Puffer application of glutamate transiently increased firing frequency of TN-GnRH3 neurons in juvenile medaka in a dose-dependent manner (Fig. 4A–4C). The peak frequency in Fig. 4B was 7.27 ± 1.34 Hz. Previous literature reported that neuropeptide release requires an increase in [Ca2+]i (6, 22), and TN-GnRH3 neurons show somato-dendritic release (20, 23). To examine whether such burst firing induced by glutamate puffer application is related to neuropeptide release, we performed Ca2+ imaging from the area surrounding somata of TN-GnRH3 neurons using a Ca2+ indicator, Fura2-AM. Puffer application of glutamate mimicking synaptic input induced a significant increase in [Ca2+]i of TN-GnRH3 neurons in juvenile (Fig. 4D and 4E). The [Ca2+]i in TN-GnRH3 neurons also increased in a dose-dependent manner (Fig. 4E). With the consideration of the increase in firing frequency (Fig. 4C) and [Ca2+]i (Fig. 4D and 4E), it may be suggested that the glutamate-induced burst firing of TN-GnRH3 neurons can trigger neuropeptide release. Similarly, in adults, puffer application of glutamate to TN-GnRH3 neurons also induced burst firing and a [Ca2+]i increase, which may suggest neuropeptide release (Fig. 4F–4I). Figure 4. View largeDownload slide Puffer application of glutamate quickly induces burst-like firing and increases [Ca2+]i in a dose-dependent manner. (A and B) Glutamate (glu; 0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in juveniles for 100 ms by using a puffer pipette at the point indicated by the arrow. (A) Representative recordings and (B) frequency histograms (n = 6) from juvenile medaka. Each frequency was calculated by the number of spikes for a 0.5-second bin. (C) Peak frequency of TN-GnRH3 neurons during glutamate application. Peak frequency was calculated by the number of spikes for 1 second, just after glutamate application. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0072; 0 vs 0.5 mM, P = 0.0101; 0 vs 1 mM, P = 0.0101; 0 (n = 6), 0.1 (n = 12), 0.5 (n = 6), and 1 (n = 6). (D) The ratio (λ340/λ380) of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (E) The peak Δratio is plotted against the glutamate concentrations. Values were calculated by ImageJ. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0113; 0 vs 0.5 mM, P = 0.0050; 0 vs 1 mM, P = 0.0050. Each group, n = 7. The juvenile fish used in experiments described above were 3 weeks after fertilization. (F and G) Glutamate (0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in adults for 100 ms by using a puffer pipette at the point indicated by the arrow. (F) Representative recording and (G) frequency histograms (n = 12) from adult medaka. Bin: 0.5 seconds. (H) The ratio of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (I) The Δratio is plotted against the glutamate concentrations. Each group, n = 6. Steel test, *P < 0.05; 0 (control) vs 0.1 mM, P = 0.0956; 0 vs 0.5 mM, P = 0.01107; 0 vs 1 mM, P = 0.01107. The adult fish used in experiments described above were >6 months after fertilization. Figure 4. View largeDownload slide Puffer application of glutamate quickly induces burst-like firing and increases [Ca2+]i in a dose-dependent manner. (A and B) Glutamate (glu; 0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in juveniles for 100 ms by using a puffer pipette at the point indicated by the arrow. (A) Representative recordings and (B) frequency histograms (n = 6) from juvenile medaka. Each frequency was calculated by the number of spikes for a 0.5-second bin. (C) Peak frequency of TN-GnRH3 neurons during glutamate application. Peak frequency was calculated by the number of spikes for 1 second, just after glutamate application. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0072; 0 vs 0.5 mM, P = 0.0101; 0 vs 1 mM, P = 0.0101; 0 (n = 6), 0.1 (n = 12), 0.5 (n = 6), and 1 (n = 6). (D) The ratio (λ340/λ380) of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (E) The peak Δratio is plotted against the glutamate concentrations. Values were calculated by ImageJ. Steel test, *P < 0.05, **P < 0.01; 0 (control) vs 0.1 mM, P = 0.0113; 0 vs 0.5 mM, P = 0.0050; 0 vs 1 mM, P = 0.0050. Each group, n = 7. The juvenile fish used in experiments described above were 3 weeks after fertilization. (F and G) Glutamate (0.5 mM, dissolved in ACSF) was applied to the TN-GnRH3 neurons in adults for 100 ms by using a puffer pipette at the point indicated by the arrow. (F) Representative recording and (G) frequency histograms (n = 12) from adult medaka. Bin: 0.5 seconds. (H) The ratio of Fura2 fluorescence transiently increased after puffer application of glutamate (each trace represents the average of three traces). Glutamate was applied by using a puffer pipette at the point indicated by the arrow. (I) The Δratio is plotted against the glutamate concentrations. Each group, n = 6. Steel test, *P < 0.05; 0 (control) vs 0.1 mM, P = 0.0956; 0 vs 0.5 mM, P = 0.01107; 0 vs 1 mM, P = 0.01107. The adult fish used in experiments described above were >6 months after fertilization. Juvenile-specific burst firing of TN-GnRH3 neurons can be triggered by the blocking of GABAergic synaptic inputs onto the glutamatergic interneurons Next, we examined the role of GABAergic synaptic inputs, which are the representative inhibitory synaptic inputs in vertebrate brains. Firing activities of TN-GnRH3 neurons were recorded in an ACSF containing GABAA receptor antagonist, bicuculline. Bath application of bicuculline induced the juvenile-specific burst firing of TN-GnRH3 neurons (Fig. 5A). Such juvenile-specific burst firing induced by bicuculline was observed in medaka from 3 weeks after fertilization onward. On the other hand, GABA directly excites TN-GnRH3 neurons in juveniles (Fig. 5B) and in adults [Fig. 5C; see also Nakane and Oka (24) and Sim et al. (25)]. In Fig. 5B and 5C, we used muscimol as a GABAA receptor agonist. Taken together, we can postulate a simple neural circuit, including TN-GnRH3 neurons (Fig. 6A). Here, the bicuculline-induced burst firing may be explained by increased glutamatergic synaptic inputs during bicuculline perfusion. To prove this possibility, we first recorded spontaneous firing activities of TN-GnRH3 neurons in an ACSF containing bicuculline after blocking glutamatergic synaptic inputs by a glutamate receptor antagonist, CNQX (Fig. 6B). Here, bicuculline neither increased firing frequency nor induced burst firing. We also used d-AP5 to block another type of glutamate receptor, N-methyl-d-aspartate (NMDA), and then added bicuculline (Fig. 6C). Bicuculline did not induce burst firing. Finally, during bath application of bicuculline, we locally blocked glutamatergic synaptic inputs by puffer application of CNQX and d-AP5 (Fig. 6D). In Fig. 6D, bicuculline first induced burst firing (Fig. 6D, b), as in Fig. 5A, but the burst firing was inhibited after local application of CNQX and d-AP5 (Fig. 6D, c) and recovered after washout (Fig. 6D, d). Figure 6D, e shows that CNQX and d-AP5 diminished bicuculline-induced burst firing. Taken together, bicuculline-induced burst firing is considered to be induced by activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)– and NMDA-type glutamate receptors during disinhibition of GABAergic synaptic inputs to glutamatergic neurons, as shown in Fig. 6A. Figure 5. View largeDownload slide Bath application of bicuculline (Bic; global blockage of GABAergic synaptic inputs) induces burst firing in juvenile TN-GnRH3 neurons. (A) A representative recording of TN-GnRH3 neuron in ∼7 weeks after fertilization. ACSF with 20 μM bicuculline was applied from 4 to 8 minutes after the start of recording. Note that TN-GnRH3 neurons from 3 weeks after fertilization showed bicuculline-induced burst firing. (B and C) GABAA receptor agonist muscimol (Mus; 20 μM dissolved in ACSF) was applied at the arrow for 100 ms. (B and C, a) Representative recordings and (B and C, b) frequency histograms (n = 7) from (B) juvenile (8 weeks after fertilization) and (C) adult (>6 months after fertilization) medaka. Inset shows the experimental design of application of muscimol, mimicking synaptic inputs. Bin, 0.5 seconds. Figure 5. View largeDownload slide Bath application of bicuculline (Bic; global blockage of GABAergic synaptic inputs) induces burst firing in juvenile TN-GnRH3 neurons. (A) A representative recording of TN-GnRH3 neuron in ∼7 weeks after fertilization. ACSF with 20 μM bicuculline was applied from 4 to 8 minutes after the start of recording. Note that TN-GnRH3 neurons from 3 weeks after fertilization showed bicuculline-induced burst firing. (B and C) GABAA receptor agonist muscimol (Mus; 20 μM dissolved in ACSF) was applied at the arrow for 100 ms. (B and C, a) Representative recordings and (B and C, b) frequency histograms (n = 7) from (B) juvenile (8 weeks after fertilization) and (C) adult (>6 months after fertilization) medaka. Inset shows the experimental design of application of muscimol, mimicking synaptic inputs. Bin, 0.5 seconds. Figure 6. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons is considered to be induced by activation of AMPA- and NMDA-type glutamate receptors. (A) Schematic neural circuit, including TN-GnRH3 neurons. GABA directly excites TN-GnRH3 neurons. Bath application of bicuculline blocks both direct and indirect GABAergic modulatory pathways. (B and C) Bicuculline-induced burst firing activity of juvenile medaka (Fig. 5A) did not occur during perfusion of (B) CNQX 20 μM (an AMPA-type glutamate receptor antagonist) or (C) d-AP5 25 μM (NMDA-type glutamate receptor antagonist). Each trace shows a representative recording. (A and B) Juvenile fish were <7 weeks after fertilization. (D, a) A representative recording of the TN-GnRH3 neuron 8 weeks after fertilization during 20 μM bicuculline perfusion. ACSF, containing 20 μM CNQX and 25 μM d-AP5, was applied for 3 minutes by using a puffer system (500-ms duration, 0.5 Hz, 90 trains) during the period indicated by the interrupted line. The puffer pipette was placed ∼20 μm from the somata of the recorded neuron. (D, b–d) Expanded trace for each gray region in (D, a). (D, e) Number of burst firing before (approximately −2 to 0 minutes), during (∼1 to 3 minutes), and after puffer application of CNQX and d-AP5 (∼4 to 6 minutes). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX and d-AP5, P = 0.0424; Before vs Washout, P = 0.9686; n = 4. GABA, GABAergic neuron; Glu, glutamatergic neuron; TN-GnRH3, TN-GnRH3 neuron. Figure 6. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons is considered to be induced by activation of AMPA- and NMDA-type glutamate receptors. (A) Schematic neural circuit, including TN-GnRH3 neurons. GABA directly excites TN-GnRH3 neurons. Bath application of bicuculline blocks both direct and indirect GABAergic modulatory pathways. (B and C) Bicuculline-induced burst firing activity of juvenile medaka (Fig. 5A) did not occur during perfusion of (B) CNQX 20 μM (an AMPA-type glutamate receptor antagonist) or (C) d-AP5 25 μM (NMDA-type glutamate receptor antagonist). Each trace shows a representative recording. (A and B) Juvenile fish were <7 weeks after fertilization. (D, a) A representative recording of the TN-GnRH3 neuron 8 weeks after fertilization during 20 μM bicuculline perfusion. ACSF, containing 20 μM CNQX and 25 μM d-AP5, was applied for 3 minutes by using a puffer system (500-ms duration, 0.5 Hz, 90 trains) during the period indicated by the interrupted line. The puffer pipette was placed ∼20 μm from the somata of the recorded neuron. (D, b–d) Expanded trace for each gray region in (D, a). (D, e) Number of burst firing before (approximately −2 to 0 minutes), during (∼1 to 3 minutes), and after puffer application of CNQX and d-AP5 (∼4 to 6 minutes). ANOVA with Dunnett post hoc test, *P < 0.05; Before vs CNQX and d-AP5, P = 0.0424; Before vs Washout, P = 0.9686; n = 4. GABA, GABAergic neuron; Glu, glutamatergic neuron; TN-GnRH3, TN-GnRH3 neuron. On the other hand, in adults, bath application of the GABAA receptor antagonist (global inhibition as in Fig. 5A) did not induce burst firing (Fig. 7A). Therefore, we plotted the percentage of bicuculline-induced burst firing at various stages of post-hatching development (Fig. 7B). The percentage tended to decrease as fish grew older. In Fig. 7C, the number of burst firing during 6 minutes after the start of bicuculline application was plotted from 7 weeks after fertilization onward; the data before 7 weeks were omitted to avoid contamination of spontaneous burst firing as much as possible. Compared with TN-GnRH3 neurons in adults, those during juvenile periods showed bicuculline-induced burst firing more frequently (Fig. 7C). Thus, the bicuculline-induced burst firing also shows time-dependent changes during post-hatching development, similar to the change in the juvenile-specific burst firing. Figure 7. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons shows time-dependent changes during post-hatching development. (A) GABAA receptor antagonist bicuculline (Bic, 20 μM, dissolved in ACSF) was applied for 4 minutes into the recording chamber where an adult brain (>12.5 weeks after fertilization) was placed. Bicuculline is considered to block GABAergic synapse in the whole brain as in Fig. 6A. Bicuculline was applied during the period indicated by the red bar. (B) Percentage of neurons showing burst firing induced by bicuculline; 4 weeks after fertilization, n = 4; 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. (C) Number of burst firings induced by bicuculline perfusion, recorded from TN-GnRH3 neurons of medaka at various developmental stages. We used recordings for 6 minutes after bicuculline application, as the effect of bicuculline continued for ∼10 minutes after washout. ANOVA with Dunnett post hoc test, *P < 0.05, **P < 0.01; 7 vs 12 weeks, P = 0.0403; 8 vs 12 weeks, P = 0.0053; 9 vs 12 weeks, P = 0.1706; 10 vs 12 weeks, P = 0.5183. At 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz), followed by interburst hyperpolarization with an interburst interval longer than 1 second. Figure 7. View largeDownload slide Bicuculline-induced burst firing in juvenile TN-GnRH3 neurons shows time-dependent changes during post-hatching development. (A) GABAA receptor antagonist bicuculline (Bic, 20 μM, dissolved in ACSF) was applied for 4 minutes into the recording chamber where an adult brain (>12.5 weeks after fertilization) was placed. Bicuculline is considered to block GABAergic synapse in the whole brain as in Fig. 6A. Bicuculline was applied during the period indicated by the red bar. (B) Percentage of neurons showing burst firing induced by bicuculline; 4 weeks after fertilization, n = 4; 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. (C) Number of burst firings induced by bicuculline perfusion, recorded from TN-GnRH3 neurons of medaka at various developmental stages. We used recordings for 6 minutes after bicuculline application, as the effect of bicuculline continued for ∼10 minutes after washout. ANOVA with Dunnett post hoc test, *P < 0.05, **P < 0.01; 7 vs 12 weeks, P = 0.0403; 8 vs 12 weeks, P = 0.0053; 9 vs 12 weeks, P = 0.1706; 10 vs 12 weeks, P = 0.5183. At 7 weeks, n = 6; 8 weeks, n = 8; 9 weeks, n = 6; 10 weeks, n = 7; >12.5 weeks, n = 6. One bout of burst firing was defined as the time period containing more than three spikes at high frequency (>3 Hz), followed by interburst hyperpolarization with an interburst interval longer than 1 second. Next, we examined whether [Ca2+]i is also increased when TN-GnRH3 neurons show bicuculline-induced burst firing (Fig. 8). As shown in Fig. 8A, bath application of ACSF containing bicuculline induced multiple peaks of [Ca2+]i increase. The Fura2 Δratio during bicuculline perfusion was significantly higher than that of the vehicle control (Fig. 8A–8C). Therefore, bicuculline-induced burst firing in juvenile medaka is also suggested to induce neuropeptide release. Figure 8. View largeDownload slide TN-GnRH3 neurons in juveniles show an increase in [Ca2+]i during bicuculline (Bic) application and also during spontaneous burst firing, which suggests neuropeptide release. (A and B) Δratio (λ340/λ380) of Fura2 during application of ACSF (A) with or (B) without bicuculline. Inset shows an EGFP image of TN-GnRH3 neurons. Light-colored traces represent Δratio for the somata, circled in inset. Dark-colored trace shows the average trace for five neurons. (C) Peak Δratio during bicuculline perfusion (n = 10). The values represent peak Δratio during vehicle (control) or bicuculline application. Paired Wilcoxon test, *P < 0.05, P = 0.0273. The juvenile fish used in experiments described above were 7 weeks after fertilization. (D and E) Spontaneous increase in [Ca2+]i. (D) Arrowheads show the somata of juvenile TN-GnRH3 neurons (4 weeks after fertilization). The white circle shows the region outside of the somata, and the other colored circles show the somata of TN-GnRH3 neurons. The [Ca2+]i in the areas (white circle) surrounding somata of TN-GnRH3 neurons also appeared to increase synchronously with that in the somata area (arrowheads). (E) Each colored trace represents data for the averaged intensity of the region circled with the same color in (D). Dotted line represents the average intensity of the region circled in the white line. Original scale bars, 20 μm. A, anterior; L, lateral; M, medial; OB, olfactory bulb; P, posterior. Figure 8. View largeDownload slide TN-GnRH3 neurons in juveniles show an increase in [Ca2+]i during bicuculline (Bic) application and also during spontaneous burst firing, which suggests neuropeptide release. (A and B) Δratio (λ340/λ380) of Fura2 during application of ACSF (A) with or (B) without bicuculline. Inset shows an EGFP image of TN-GnRH3 neurons. Light-colored traces represent Δratio for the somata, circled in inset. Dark-colored trace shows the average trace for five neurons. (C) Peak Δratio during bicuculline perfusion (n = 10). The values represent peak Δratio during vehicle (control) or bicuculline application. Paired Wilcoxon test, *P < 0.05, P = 0.0273. The juvenile fish used in experiments described above were 7 weeks after fertilization. (D and E) Spontaneous increase in [Ca2+]i. (D) Arrowheads show the somata of juvenile TN-GnRH3 neurons (4 weeks after fertilization). The white circle shows the region outside of the somata, and the other colored circles show the somata of TN-GnRH3 neurons. The [Ca2+]i in the areas (white circle) surrounding somata of TN-GnRH3 neurons also appeared to increase synchronously with that in the somata area (arrowheads). (E) Each colored trace represents data for the averaged intensity of the region circled with the same color in (D). Dotted line represents the average intensity of the region circled in the white line. Original scale bars, 20 μm. A, anterior; L, lateral; M, medial; OB, olfactory bulb; P, posterior. Finally, we performed Ca2+ imaging of the medaka brain during the ages when TN-GnRH3 neurons showed juvenile-specific spontaneous burst firing (3 to 5 weeks after fertilization). Figure 8D shows representative images of a spontaneous [Ca2+]i increase. Traces in Fig. 8E show the Δratio of the regions of interest, indicated by different colors in the EGFP image in Fig. 8D. The [Ca2+]i in the areas surrounding somata of TN-GnRH3 neurons also appeared to increase synchronously with that in the somata area. From the data shown thus far, TN-GnRH3 neurons are suggested to release neuropeptides during juvenile-specific spontaneous and bicuculline-induced burst firing. Discussion The present study demonstrated that >50% of TN-GnRH3 neurons in juvenile medaka (especially up to 5 weeks after fertilization) showed spontaneous burst firing (intraburst firing frequency was >5 Hz), whereas none of them did so in adulthood. By Ca2+ imaging analysis, we showed that not only burst firing evoked by puffer application of glutamate mimicking glutamatergic synaptic input but also a juvenile-specific spontaneous one are suggested to trigger neuropeptide release in TN-GnRH3 neurons. Our results suggest that neuropeptides released from TN-GnRH3 neurons play a previously unrecognized role(s) in juveniles, which may indicate novel functions, in addition to its recognized role in neuromodulation in adulthood (see Functional significance of juvenile-specific burst firing in TN-GnRH3 neurons for functional discussion). Juvenile-specific burst firing of TN-GnRH3 neurons It has been reported by morphological analyses of GFP-labeled GnRH neurons (19) and immunohistochemical and in situ hybridization studies [also see references cited in Takahashi et al. (19)] that TN-GnRH3 neurons express GnRH3 peptide from early developmental stages. Though it has been suggested that spontaneous highfrequency firing, such as bursting, is important for neuropeptide release, it remains unknown during which developmental stages TN-GnRH3 neurons show burst firing. In the present study, we have demonstrated that TN-GnRH3 neurons in juveniles show burst firing, by far more frequently than in adulthood (Fig. 9; juvenile-specific burst firing). In addition, by combining electrophysiology and Ca2+ imaging, we could clearly demonstrate that burst firing (both evoked and juvenile-specific spontaneous ones; Figs. 4 and 8) actually induces a large increase in [Ca2+]i. It has been reported recently that in hypophysiotropic GnRH (GnRH1) the neuron typically fires spontaneously at ∼1 Hz (26) and shows a measurable increase in [Ca2+]i during electrical stimulation of 1 Hz (27). In addition, hypophysiotropic GnRH1 and kisspeptin neurons that show high-frequency firing (≥5 Hz) effectively induce neuropeptide release in the axon terminal in vivo (28, 29) and in vitro (27, 30). The release of these neuropeptides is triggered by an increase in [Ca2+]i (30, 31). In other types of neurons (dopaminergic neurons), it has also been reported that burst firing is important for a large increase in [Ca2+]i (32). In the present study, a firing frequency of >3 Hz, which was induced by puffer application of 100 μM glutamate, was strong enough to cause an increase in [Ca2+]i (Fig. 4). It should be noted that most of the spontaneous juvenile-specific burst firings were >5 Hz (Table 1, top row). A Fura2-fluorescence increase in spontaneous activity of juvenile TN-GnRH3 neurons was almost the same as that shown in hypophysiotropic GnRH1 neurons (Δratio, ∼0.1), which is large enough to trigger GnRH1 peptide release (30). Therefore, it is strongly suggested that juvenile-specific spontaneous burst firing can actually trigger release of the neuropeptide from the dendrite or axon in vivo as well (Fig. 9B). Furthermore, as shown in Fig. 8D, we observed a simultaneous [Ca2+]i increase in the areas surrounding somata of TN-GnRH3 neurons. Because the Ca2+ indicator Fura2 was loaded to the whole brain, it is possible that either surrounding non-TN-GnRH3 neurons or dendrites of TN-GnRH3 neurons were excited synchronously. TN-GnRH3 neurons form dense dendritic plexuses around the somata (11); therefore, it is more likely that the spontaneous [Ca2+]i increases in the areas surrounding somata of TN-GnRH3 neurons arise from neural activity in the dendritic plexuses. In contrast, TN-GnRH3 neurons in adulthood showed spontaneous burst firing far less frequently. However, they showed glutamate-induced burst firing, which also increased [Ca2+]i significantly (Fig. 4F and 4H). These results suggest that the TN-GnRH3 neuron in adulthood can also release neuropeptide in response to glutamatergic synaptic inputs (Fig. 9C, a). Figure 9. View largeDownload slide Summary diagram, indicating neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in juvenile and adult medaka brains. (A) Schematic illustration showing the time course of events during development. TN-GnRH3 neurons show burst firing and receive glutamatergic effects, which depend on synaptic inputs and/or receptors. In the green and magenta bars, the darker the colors, the more frequent is burst firing (green) or the stronger is glutamatergic (Glu-ergic) effect (magenta). (B and C) Hypothetical neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in (B) juveniles and (C) adults. (B and C, a) Note that GABAergic synapses are inhibitory (–) to glutamatergic neurons but are excitatory (+) to TN-GnRH3 neurons (see text for details). TN-GnRH3 neurons in juvenile medaka receive persistent glutamatergic input and show high-frequency firing and a [Ca2+]i increase. In adults, when TN-GnRH3 neurons transiently receive glutamatergic synaptic inputs, they show high-frequency firing. Both firing patterns in juveniles and adults are suggested to induce neuropeptide release. In adults, spontaneous burst firing is rarely observed, because of the decrease in expression level or the changes in the subtype of glutamate receptors in TN-GnRH3 neurons or the decrease in the number of glutamatergic synaptic inputs. For the sake of simplicity, only GnRH is illustrated as a representative neuropeptide that is released from TN-GnRH3 neurons, although neuropeptide FF (NPFF) may also be released. (B and C, b) Hypothetical functions of TN-GnRH3 neurons suggested by the current study. TN-GnRH3 neurons receive plenty of environmental information via glutamatergic synaptic inputs, show burst firing, and release neuropeptides. In juvenile medaka, this process may work more frequently than in adults. AMPAR, AMPA receptor; GABA, GABAergic neuron; Glu, glutamatergic neuron; NMDAR, NMDA receptor; TN-GnRH3, TN-GnRH3 neurons. Figure 9. View largeDownload slide Summary diagram, indicating neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in juvenile and adult medaka brains. (A) Schematic illustration showing the time course of events during development. TN-GnRH3 neurons show burst firing and receive glutamatergic effects, which depend on synaptic inputs and/or receptors. In the green and magenta bars, the darker the colors, the more frequent is burst firing (green) or the stronger is glutamatergic (Glu-ergic) effect (magenta). (B and C) Hypothetical neural circuitries involved in neuropeptide release from the TN-GnRH3 neurons in (B) juveniles and (C) adults. (B and C, a) Note that GABAergic synapses are inhibitory (–) to glutamatergic neurons but are excitatory (+) to TN-GnRH3 neurons (see text for details). TN-GnRH3 neurons in juvenile medaka receive persistent glutamatergic input and show high-frequency firing and a [Ca2+]i increase. In adults, when TN-GnRH3 neurons transiently receive glutamatergic synaptic inputs, they show high-frequency firing. Both firing patterns in juveniles and adults are suggested to induce neuropeptide release. In adults, spontaneous burst firing is rarely observed, because of the decrease in expression level or the changes in the subtype of glutamate receptors in TN-GnRH3 neurons or the decrease in the number of glutamatergic synaptic inputs. For the sake of simplicity, only GnRH is illustrated as a representative neuropeptide that is released from TN-GnRH3 neurons, although neuropeptide FF (NPFF) may also be released. (B and C, b) Hypothetical functions of TN-GnRH3 neurons suggested by the current study. TN-GnRH3 neurons receive plenty of environmental information via glutamatergic synaptic inputs, show burst firing, and release neuropeptides. In juvenile medaka, this process may work more frequently than in adults. AMPAR, AMPA receptor; GABA, GABAergic neuron; Glu, glutamatergic neuron; NMDAR, NMDA receptor; TN-GnRH3, TN-GnRH3 neurons. Mechanisms of juvenile-specific burst firing in TN-GnRH3 neurons Both juvenile-specific burst firing and burst firing by bicuculline application in TN-GnRH3 neurons were induced by glutamatergic synaptic inputs (Figs. 2, 4, and 6). These firings also showed time-dependent changes during post-hatching development (Figs. 1F and 7B). Thus, juvenile-specific burst firing is suggested to be induced by almost the same excitatory mechanism as the burst firing by bicuculline application. In other words, global bicuculline application (bath application) in juveniles is considered to block the persistent GABAergic inhibition to the glutamatergic neurons (Fig. 6A), and this disinhibitory action of bicuculline may have facilitated the glutamate receptor-mediated burst firing of TN-GnRH3 neurons in juvenile medaka (Fig. 9B, a). Juvenile-specific burst firing is suggested to be induced when such blockade of the persistent GABAergic inhibition to the glutamatergic neurons occurs. As noted in the results, our study showed that glutamatergic synaptic inputs are important for juvenile-specific burst firing. Considering the fact that bicuculline did not induce burst firing under persistent blockage of AMPA- or NMDA-type glutamate receptors (Fig. 6D, a), it is suggested that AMPA- and NMDA-type glutamatergic synaptic inputs act in concert to generate the juvenile-specific spontaneous burst firing. Interestingly, a similar mechanism for burst generation has also been proposed for the central pattern generator circuitry for locomotion in the spinal cord (33, 34). Thus, AMPA- and NMDA-type glutamatergic synaptic inputs are suggested to play an important role in juvenile-specific burst firing. Furthermore, the percentage of burst firing was changed during post-hatching development (Fig. 1). Our study also demonstrated that CNQX affected spontaneous activities of TN-GnRH3 neurons in juveniles more severely than in adulthood. With the consideration of the mechanism underlying the decrease in juvenile-specific burst firing with age, this decrease in burst firing may be due to changes in the level of glutamate receptor expression, in the subtype of glutamate receptors, or in the density of glutamatergic presynaptic inputs onto TN-GnRH3 neurons. It has been reported that subtypes of NMDA receptors in the brain changes during early development in mammals (35). It is possible that the number or subtype of NMDA receptors in TN-GnRH3 neurons is different between juvenile and adulthood (Fig. 9B, a and 9C, a), because each subtype of the NMDA receptor has specific conductance and other characteristics (35). On the other hand, the GABAA receptor antagonist did not induce burst firing in adulthood (Fig. 7A). If TN-GnRH3 neurons in juveniles receive more glutamatergic inputs than in adulthood, it is suggested that TN-GnRH3 neurons in adulthood rarely show spontaneous burst firing, because both glutamatergic synaptic inputs and persistent GABAergic inhibition decrease during post-hatching development (Fig. 9C, a). If the mechanism underlying the change in firing pattern during post-hatching development will be further clarified in the future, it should help to deepen the understanding of the nature of peptidergic neurons. Functional significance of juvenile-specific burst firing in TN-GnRH3 neurons Our final discussion will focus on the functional significance of juvenile-specific burst firing in TN-GnRH3 neurons. A histological study using a trans-synaptic tracer suggested that TN-GnRH3 neurons receive afferent input, mainly from brain regions involved in olfactory, visual, and somatosensory information processing (36). Considering our finding that juvenile-specific burst firing needs glutamatergic synaptic inputs, TN-GnRH3 neurons may receive these inputs from such brain regions. In other words, our study suggests that TN-GnRH3 neurons in juveniles receive plenty of environmental information via glutamatergic synaptic inputs, show burst firing, and release neuropeptides, far more actively than in adults (Fig. 9B, b and 9C, b). Previous studies of TN-GnRH3 neurons in adulthood suggested that the TN-GnRH3 neuron is important for motivation control for sexual behavior (8–10), which may be via neuromodulation by GnRH3 peptide (13, 14, 16). Thus, one possible function of juvenile-specific burst firing also may be neuromodulation, such as in adults. If so, the GnRH3 peptide that is released in juvenile animals is suggested to play a key role in neuromodulation of age-independent phenomena, such as energy homeostasis, mood, motivation, etc. (4, 6). Although some vertebrates (mammals, birds, eel, etc.) have lost gnrh3, it has been proposed that gnrh1 compensates for the loss of gnrh3 and is expressed in the TN neurons (7). The GnRH1 peptide expressed in the TN of such juvenile animals may serve the same functions as described previously. On the other hand, TN-GnRH3 neurons release neuropeptide FF (NPFF) in addition to the GnRH3 peptide (37). Further investigation of the role of GnRH and NPFF in neuromodulation is necessary to determine if these peptides are released separately or simultaneously to exert differential functions. Another possible novel function of TN-GnRH3 neurons may be neurotrophic and/or maturational action. Recently, it has been reported that some neuropeptides also function as neurotrophic factors during early development (38–40). For example, calcitonin gene-related peptide acts for survival and differentiation of neurons during early development and neuromodulation in adulthood (39). Spontaneous bursting activity before the onset of hearing promotes the survival and maturation of auditory neurons and the refinement of synaptic connections in auditory nuclei (41). Therefore, neuropeptide release by juvenile-specific burst firing may also have neurotrophic and/or maturational action, although other functions are also possible. It has been reported that animal behavior in adulthood can change in response to their environment in the juvenile period, as well as that surrounding the animals (42, 43). It is therefore possible that juvenile-specific burst firing in TN-GnRH3 neurons plays a key role in such environmental factors during the juvenile period to modulate behaviors in adulthood. In summary, we found that TN-GnRH3 neurons show juvenile-specific, spontaneous burst firing, which may have novel functions, in addition to neuromodulation. Abbreviations: Abbreviations: ACSF artificial cerebrospinal fluid AM acetoxymethyl AMPA α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ANOVA analysis of variance [Ca2+]i intracellular Ca2+ concentration CNQX 6-cyano-7-nitroquinoxaline-2,3-dione d-AP5 d-(–)-2-amino-5-phosphonopentanoic acid EGFP enhanced green fluorescent protein EM electron multiplying GABA γ-aminobutyric acid GnRH gonadotropin-releasing hormone NMDA N-methyl-d-aspartate NPFF neuropeptide FF TN terminal nerve TN-GnRH3 terminal nerve gonadotropin-releasing hormone 3 Δratio ratio change of fluorescence intensity Acknowledgments We thank Drs. Shinji Kanda (The University of Tokyo), Masaharu Hasebe (Osaka University), and Kazuhiko Yamaguchi (RIKEN Brain Science Institute) for valuable discussions. We also appreciate Ms. Miho Kyokuwa, Ms. Hisako Kohno, and Ms. Fumika Muguruma (The University of Tokyo) for their gentle care of the fish. Financial Support: The current study was supported by Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research (JSPS KAKENHI) Grants 26221104 and 17K15157. Author Contributions: C.U. and Y.O. designed and drafted the present work. C.U. collected, analyzed, and interpreted data and revised the paper critically for important intellectual content. Both authors approved the final version of the manuscript and agreed 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. Disclosure Summary: The authors have nothing to disclose. References 1. de Kloet ER, Joëls M, Holsboer F. Stress and the brain: from adaptation to disease. Nat Rev Neurosci . 2005; 6( 6): 463– 475. Google Scholar CrossRef Search ADS PubMed  2. Iwamatsu T. The Integrated Book for the Biology of Medaka . Okayama, Japan: University Education Press; 1997. 3. Edwards DA, Griffis KT, Tardivel C. Olfactory bulb removal: effects on sexual behavior and partner-preference in male rats. Physiol Behav . 1990; 48( 3): 447– 450. Google Scholar CrossRef Search ADS PubMed  4. Argiolas A, Melis MR. Neuropeptides and central control of sexual behaviour from the past to the present: a review. Prog Neurobiol . 2013; 108: 80– 107. Google Scholar CrossRef Search ADS PubMed  5. Dornan WA, Malsbury CW. Neuropeptides and male sexual behavior. Neurosci Biobehav Rev . 1989; 13( 1): 1– 15. Google Scholar CrossRef Search ADS PubMed  6. van den Pol AN. Neuropeptide transmission in brain circuits. Neuron . 2012; 76( 1): 98– 115. Google Scholar CrossRef Search ADS PubMed  7. Okubo K, Nagahama Y. Structural and functional evolution of gonadotropin-releasing hormone in vertebrates. Acta Physiol (Oxf) . 2008; 193( 1): 3– 15. Google Scholar CrossRef Search ADS PubMed  8. Wirsig CR, Leonard CM. Terminal nerve damage impairs the mating behavior of the male hamster. Brain Res . 1987; 417( 2): 293– 303. Google Scholar CrossRef Search ADS PubMed  9. Yamamoto N, Oka Y, Kawashima S. Lesions of gonadotropin-releasing hormone-immunoreactive terminal nerve cells: effects on the reproductive behavior of male dwarf gouramis. Neuroendocrinology . 1997; 65( 6): 403– 412. Google Scholar CrossRef Search ADS PubMed  10. Okuyama T, Yokoi S, Abe H, Isoe Y, Suehiro Y, Imada H, Tanaka M, Kawasaki T, Yuba S, Taniguchi Y, Kamei Y, Okubo K, Shimada A, Naruse K, Takeda H, Oka Y, Kubo T, Takeuchi H. A neural mechanism underlying mating preferences for familiar individuals in medaka fish. Science . 2014; 343( 6166): 91– 94. Google Scholar CrossRef Search ADS PubMed  11. Oka Y, Matsushima T. Gonadotropin-releasing hormone (GnRH)-immunoreactive terminal nerve cells have intrinsic rhythmicity and project widely in the brain. J Neurosci . 1993; 13( 5): 2161– 2176. Google Scholar PubMed  12. von Bartheld CS, Meyer DL. Tracing of single fibers of the nervus terminalis in the goldfish brain. Cell Tissue Res . 1986; 245( 1): 143– 158. Google Scholar CrossRef Search ADS PubMed  13. Eisthen HL, Delay RJ, Wirsig-Wiechmann CR, Dionne VE. Neuromodulatory effects of gonadotropin releasing hormone on olfactory receptor neurons. J Neurosci . 2000; 20( 11): 3947– 3955. Google Scholar PubMed  14. Kawai T, Abe H, Akazome Y, Oka Y. Neuromodulatory effect of GnRH on the synaptic transmission of the olfactory bulbar neural circuit in goldfish, Carassius auratus. J Neurophysiol . 2010; 104( 6): 3540– 3550. Google Scholar CrossRef Search ADS PubMed  15. Maruska KP, Tricas TC. Gonadotropin-releasing hormone (GnRH) modulates auditory processing in the fish brain. Horm Behav . 2011; 59( 4): 451– 464. Google Scholar CrossRef Search ADS PubMed  16. Umatani C, Misu R, Oishi S, Yamaguchi K, Abe H, Oka Y. GnRH suppresses excitability of visual processing neurons in the optic tectum. J Neurophysiol . 2015; 114( 5): 2775– 2784. Google Scholar CrossRef Search ADS PubMed  17. Amano M, Oka Y, Kitamura S, Ikuta K, Aida K. Ontogenic development of salmon GnRH and chicken GnRH-II systems in the brain of masu salmon (Oncorhynchus masou). Cell Tissue Res . 1998; 293( 3): 427– 434. Google Scholar CrossRef Search ADS PubMed  18. Amano M, Okubo K, Yamanome T, Oka Y, Kawaguchi N, Aida K, Yamamori K. Ontogenic development of three GnRH systems in the brain of a pleuronectiform fish, barfin flounder. Zool Sci . 2004; 21( 3): 311– 317. Google Scholar CrossRef Search ADS PubMed  19. Takahashi A, Islam MS, Abe H, Okubo K, Akazome Y, Kaneko T, Hioki H, Oka Y. Morphological analysis of the early development of telencephalic and diencephalic gonadotropin-releasing hormone neuronal systems in enhanced green fluorescent protein-expressing transgenic medaka lines. J Comp Neurol . 2016; 524( 4): 896– 913. Google Scholar CrossRef Search ADS PubMed  20. Karigo T, Oka Y. Neurobiological study of fish brains gives insights into the nature of gonadotropin-releasing hormone 1-3 neurons. Front Endocrinol (Lausanne) . 2013; 4: 177. Google Scholar PubMed  21. Hasebe M, Kanda S, Shimada H, Akazome Y, Abe H, Oka Y. Kiss1 neurons drastically change their firing activity in accordance with the reproductive state: insights from a seasonal breeder. Endocrinology . 2014; 155( 12): 4868– 4880. Google Scholar CrossRef Search ADS PubMed  22. Peng YY, Zucker RS. Release of LHRH is linearly related to the time integral of presynaptic Ca2+ elevation above a threshold level in bullfrog sympathetic ganglia. Neuron . 1993; 10( 3): 465– 473. Google Scholar CrossRef Search ADS PubMed  23. Oka Y, Ichikawa M. Ultrastructure of the ganglion cells of the terminal nerve in the dwarf gourami (Colisa lalia). J Comp Neurol . 1991; 304( 2): 161– 171. Google Scholar CrossRef Search ADS PubMed  24. Nakane R, Oka Y. Excitatory action of GABA in the terminal nerve gonadotropin-releasing hormone neurons. J Neurophysiol . 2010; 103( 3): 1375– 1384. Google Scholar CrossRef Search ADS PubMed  25. Sim JA, Skynner MJ, Herbison AE. Heterogeneity in the basic membrane properties of postnatal gonadotropin-releasing hormone neurons in the mouse. J Neurosci . 2001; 21( 3): 1067– 1075. Google Scholar PubMed  26. Constantin S, Iremonger KJ, Herbison AE. In vivo recordings of GnRH neuron firing reveal heterogeneity and dependence upon GABAA receptor signaling. J Neurosci . 2013; 33( 22): 9394– 9401. Google Scholar CrossRef Search ADS PubMed  27. Iremonger KJ, Porteous R, Herbison AE. Spike and neuropeptide-dependent mechanisms control GnRH neuron nerve terminal Ca2+ over diverse time scales. J Neurosci . 2017; 37( 12): 3342– 3351. Google Scholar CrossRef Search ADS PubMed  28. Campos P, Herbison AE. Optogenetic activation of GnRH neurons reveals minimal requirements for pulsatile luteinizing hormone secretion. Proc Natl Acad Sci USA . 2014; 111( 51): 18387– 18392. Google Scholar CrossRef Search ADS PubMed  29. Han SY, McLennan T, Czieselsky K, Herbison AE. Selective optogenetic activation of arcuate kisspeptin neurons generates pulsatile luteinizing hormone secretion. Proc Natl Acad Sci USA . 2015; 112( 42): 13109– 13114. Google Scholar CrossRef Search ADS PubMed  30. Hasebe M, Oka Y. High-frequency firing activity of GnRH1 neurons in female medaka induces the release of GnRH1 peptide from their nerve terminals in the pituitary. Endocrinology . 2017; 158( 8): 2603– 2617. Google Scholar CrossRef Search ADS PubMed  31. Moenter SM, DeFazio AR, Pitts GR, Nunemaker CS. Mechanisms underlying episodic gonadotropin-releasing hormone secretion. Front Neuroendocrinol . 2003; 24( 2): 79– 93. Google Scholar CrossRef Search ADS PubMed  32. Hage TA, Khaliq ZM. Tonic firing rate controls dendritic Ca2+ signaling and synaptic gain in substantia nigra dopamine neurons. J Neurosci . 2015; 35( 14): 5823– 5836. Google Scholar CrossRef Search ADS PubMed  33. Wallén P, Grillner S. N-Methyl-d-aspartate receptor-induced, inherent oscillatory activity in neurons active during fictive locomotion in the lamprey. J Neurosci . 1987; 7( 9): 2745– 2755. Google Scholar PubMed  34. Grillner S, Deliagina T, Ekeberg O, el Manira A, Hill RH, Lansner A, Orlovsky GN, Wallén P. Neural networks that co-ordinate locomotion and body orientation in lamprey. Trends Neurosci . 1995; 18( 6): 270– 279. Google Scholar CrossRef Search ADS PubMed  35. Cull-Candy S, Brickley S, Farrant M. NMDA receptor subunits: diversity, development and disease. Curr Opin Neurobiol . 2001; 11( 3): 327– 335. Google Scholar CrossRef Search ADS PubMed  36. Yamamoto N, Ito H. Afferent sources to the ganglion of the terminal nerve in teleosts. J Comp Neurol . 2000; 428( 2): 355– 375. Google Scholar CrossRef Search ADS PubMed  37. Saito TH, Nakane R, Akazome Y, Abe H, Oka Y. Electrophysiological analysis of the inhibitory effects of FMRFamide-like peptides on the pacemaker activity of gonadotropin-releasing hormone neurons. J Neurophysiol . 2010; 104( 6): 3518– 3529. Google Scholar CrossRef Search ADS PubMed  38. Rogge G, Jones D, Hubert GW, Lin Y, Kuhar MJ. CART peptides: regulators of body weight, reward and other functions [published corrigendum appears in Nat Rev Neurosci. 2010;11:218]. Nat Rev Neurosci . 2008; 9( 10): 747– 758. Google Scholar CrossRef Search ADS PubMed  39. Denis-Donini S. Expression of dopaminergic phenotypes in the mouse olfactory bulb induced by the calcitonin gene-related peptide. Nature . 1989; 339( 6227): 701– 703. Google Scholar CrossRef Search ADS PubMed  40. Hökfelt T. Neuropeptides in perspective: the last ten years. Neuron . 1991; 7( 6): 867– 879. Google Scholar CrossRef Search ADS PubMed  41. Tritsch NX, Bergles DE. Developmental regulation of spontaneous activity in the Mammalian cochlea. J Neurosci . 2010; 30( 4): 1539– 1550. Google Scholar CrossRef Search ADS PubMed  42. Lupien SJ, McEwen BS, Gunnar MR, Heim C. Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nat Rev Neurosci . 2009; 10( 6): 434– 445. Google Scholar CrossRef Search ADS PubMed  43. Fleming IA, Lamberg A, Jonsson B. Effects of early experience on the reproductive performance of Atlantic salmon. Behav Ecol . 1997; 8( 5): 470– 480. Google Scholar CrossRef Search ADS   Copyright © 2018 Endocrine Society

Journal

EndocrinologyOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial