3.1. Characterize of Dl-evoked field potentials in the Dm 3.1.1. Excitatory postsynaptic potentials in Dm division of
the telencephalon recorded by multi-electrode arrays (MED64)
In Fig. 2-1 an outline of a telencephalon slice border structures is sketched. To recorded evoked field potential, we placed 300 μm thick telencephalon brain over an 8 x 8 MED64 probe with an inter-electrode distance of 150μm (Fig. 2-2A). In 2-2B a bipolar stimulation of Dl division was performed through an electrode located on the border between sulcus ypsiloniformis (Y) and posterior (Dp) zones. Biphasic positive-negative current applied to the stimulation electrode evoked a
negative peak in Dm. For the first time, the present results reveal that electrical stimulation of the Dl can evoke a negative-going potential in the Dm of zebrafish. Neuronal response properties can be characterized by means of the input-output response function, such as input-output (I-O) curve. We characterized the input–output relationship between stimulation intensity in the Dl division and the amplitude of population spikes in Dm division. As shown in Fig. 2-3, the amplitude of the population spike gradually rises from threshold values to an approximately linear form, and then smoothly saturates. Our results demonstrating increased excitatory synaptic transmission in response to increased stimulation intensity of Dl afferents.
3.1.2. Electrical stimulation evoked a field potential in Dm result from direct activation of axons and neurons
Electrical stimulation within the dorsal telencephalon evoked a complex field potential, always consisting of an initial positive deflection (P1) followed by a larger negative peak (N2) (Fig. 2-4A). In order to determine the field potential is mediated by synaptic processes, several tests were performed. As illustrated in Fig. 2-4A-C, superfusion of aCSF containing 0.5 mM Ca2+ and 8.0 mM Mg2+ reversibly abolished the N2 component, while P1 was remains intact. From Fig. 2-4C-D it is evident that 0.5μM TTX abolished all components of the evoked potential.
Moreover, almost all synaptic transmission exhibits either paired-pulse facilitation (PPF), depending mainly on the release probability of terminals. In Fig. 2-5 a typical example of PPF is shown, paired pulse
stimulation produced a facilitation of the field potential in response to a second stimulus at 20 to 200 ms interpulse intervals (20 ms, 184.5 ± 5%, p<0.01; 50 ms, 176.9 ± 4%, p<0.01; 100 ms, 172.0 ± 7%, p<0.01; 150 ms, 159.8 ± 8%, p<0.01; 200 ms, 146.0 ± 9%, p<0.01). Clearly, the findings indicate that the P1 component of the evoked potential in Dm is a non-synaptic component, N2 a monosynaptic population spike (PS).
3.1.3 . The transmitter system mediating the synaptic response
To investigate whether Dl stimulation is able to activate glutamatergic fibers, slices were exposed to the AMPA/kainate receptor antagonist CNQX. In Fig. 2-6A – C an example of the potent antagonizing action of 5 μM CNQX is shown. It reversibly abolished the amplitudes of population spike, but did not affect the P1 (n=2). Our results suggest that the evoked potential in the Dm of the pallium is mediated through the AMPA/kainate receptor. To test whether NMDA receptors participate in mediating the synaptic response, telencephalon slices were superfused with Mg2+ free aCSF. After perfusion with Mg2+
free aCSF the evoked response was enhanced in amplitude, and a prolonged bursting activity with multiple spike was shown in Fig 2-7B-b.
On the other hand, perfusion with the GABAA receptor ( GABAA R) antagonist bicuculline (BIC), which has been shown to be a potent antagonist of inhibitory activity in the telencephalon (Kim et al., 2004).
BIC (3 μM) had the same effects, but the duration of prolonged bursting activity with multiple spike was shorter than Mg2+ free aCSF treatment.
3.2. Long-term potentiation at Dl-Dm synapses 3.2.1. High frequency stimulation (HFS)-induced LTP
Long-term potentiation (LTP) has been proposed as a candidate cellular mechanism for learning and memory (Harris et al., 1984). In a previous study, LTP was described in the adult telencephalon brain of the zebrafish (Nam et al., 2004a). However, LTP in subdivisions of dorsal telencephalon remain unclear. Thus, we investigated the expression of LTP in the Dm division induced by high frequency stimulation (HFS) of the Dl division. We showed that the application of three trains HFS sufficient to elicit LTP at the Dl–Dm pathway (Fig. 2-8A), which lasted for at least 1 h. The amplitude of population spike at 1 h after the HFS was 223.4 ± 12% (n = 6, p < 0.01) of baseline (Fig. 2-8B). Previous studies demonstrated that the LTP are NMDA receptor-dependent (Volianskis and Jensen, 2003). This raises a possibility that the glutamatergic NMDA receptor may also be closely related to LTP in the Dl and Dm of the zebrafish. In order to investigate whether NMDA receptors are involved in LTP formation, we examined the effects of an NMDA receptor antagonist, DL-AP5, on the induction phase of LTP. As shown in Fig. 2-9, application of 40 μM of DL-AP5 alone did not produce any significant changes; however, there was a slight reduction in the baseline amplitude of the population spike when compared with control (91.2 ± 6% of the baseline, n = 6, p = 0.168). When LTP-inducing HFS was delivered after incubation with DL-AP5 for 30 min, no significant differences were found in the amplitude of the population spike for up to 30 min after HFS (96.5 ± 5% of the baseline, p = 0.535).
When the same LTP-inducing HFS was delivered after the washout of DL-AP5, the amplitude of population spike 10 min after HFS were significantly increased (154.8 ± 8% of baseline, p < 0.01), and remained stable for more than 60 min (141.8 ± 8% of baseline, p = 0.019). Thus, our data suggests that NMDA receptor activation is necessary for the induction of LTP by HFS (Fig. 2-9).
3.2.2. Forskolin (FSK)-induced LTP
The activation of adenylyl cyclase results in increased cytosolic levels of cyclic adenosine monophosphate (cAMP) and subsequent action of protein kinase A (PKA). cAMP/PKA transduction cascade has been demonstrated to play a critical role in synaptic plasticity and learning and memory in Drosophila (Davis et al., 1995) and rodents (Vazquez et al., 2000). In present study, we used forskolin, an adenylyl cyclase activator, to examine the effects of cAMP accumulation on synaptic plasticity at the Dl-Dm synapse. As in Figure 2-10A, brief application of forskolin in 50μM for 15 min induced a long-lasting LTP at the Dl–Dm pathway, which lasted for at least 1 h. The amplitude of population spike at 1 h after application of forskolin was 164.0 ± 4% (n = 3, p < 0.01) of baseline.
In vehicle experiments, we found that 15 min application of 0.2% DMSO had no effect on basal synaptic transmission, thus the long-lasting effect was elicited by forskolin. In addition, our results demonstrated that the activation of ERK in the telencephalon was induced at 15 min after forskolin (50 μM) bath applied. Only a single protein of 44kDa was clearly detected by phosphor-ERK 1/2 antibodies (Fig 2-10B).
3.3 . Long-term depression at Dl-Dm synapses 3.3.1. DHPG-induced LTD
To verify that mGluR is important for synaptic plasticity in telencephalon of zebrafish, we examined the effects of application of an mGluR agonist, DHPG, on LTD. We showed that the application of DHPG (25 μM for 10 min) induced LTD at Dl-Dm synapse, which lasted for at least 1 h. The amplitude of population spike at 1 h after the brief application of DHPG was 71.6 ± 13% (n = 6, p < 0.05) of baseline (Fig.
2-11). After 10 min application of 25 μM DHPG, it transient reversibly abolished the amplitudes of population spike, but did not affect the fiber volley. This finding supports our hypothesis that the mGluRs play important roles in synaptic plasticity.
4. Discussion