The Journal of Neuroscience, March 1, 2006, ():
Neuronal Basis of the Slow (<1 Hz) Oscillation in Neurons of the Nucleus Reticularis Thalami In Vitro
J. Neurosci. Blethyn et al.
26: 2474
Supplemental data
Files in this Data Supplement:
- supplemental material
-
Supplementary Figure 1. Extracellular recordings demonstrate the normal occurrence of the slow (<1 Hz) oscillation.
A. Extracellular recording in the PGN showing a lack of spontaneous action potential output in control conditions (1). Application of trans-ACPD (100 μM) induced rhythmic LTCP-mediated burst firing at δ frequencies (1-4 Hz) (2). This then developed into a slow oscillation comprising episodes of intense single action potential output separated by periods of rhythmic LTCP-mediated bursts (3). After continued exposure to trans-ACPD, the slow oscillation was replaced by sustained tonic action potential firing (4). After 3 minutes of washing off trans-ACPD the slow oscillation was reinstated (5). Further washout of trans-ACPD initially increased the frequency of the slow oscillation (6), presumably due to a further hyperpolarization of the neuron (see Figs. 2 and 5), but eventually led to a quiescent state being restored (7). The sections marked by continuous bars are expanded immediately to the right. B. Interspike interval plots for the enlarged LTCP bursts shown on the far right in A illustrating the accelerating-decelerating pattern which is characteristic of NRT neurons. CNQX (10 μM), APV (50 μM), SR95531 (20 μM) and CGP54626 (20 μM) were present in the recording medium during this experiment.
- supplemental material
-
Supplementary Figure 2. Extracellular recordings showing the normal occurrence of the slow (<1 Hz) oscillation following application of the Group I selective mGluR agonist, DHPG.
A. Extracellularly-recorded slow oscillation from a neuron in the peri-VB sector after 30 minutes exposure to 100 μM DHPG (1). The slow oscillation is replaced by continuous tonic firing after a further 5 mins exposure (2). Upon washout of DHPG, the neuron initially exhibited a slow oscillation (3) which gradually increased in frequency (4 and 5) before being replaced by a state of inactivity (6). A brief (10 minutes) reintroduction of DHPG (100 μM) reinstated the slow oscillation (7). The sections marked by continuous bars are expanded immediately to the right. B. Interspike interval plots for the enlarged LTCP bursts shown on the far right in A illustrating the accelerating-decelerating pattern which is characteristic of NRT neurons. CNQX (10 μM), APV (50 μM), SR95531 (20 μM) and CGP54626 (20 μM) were present in the recording medium during this experiment.
- supplemental material
-
Supplementary Figure 3. Effect of artificial ILeak reduction using a dynamic clamp system in a non-oscillating NRT neuron.
In an NRT neuron (located in the PGN) which does not exhibit a slow oscillation in the presence of trans-ACPD (100 μM) and TTX (1 μM) (A), artificial reduction of ILeak (in addition to that already induced by mGluR1a activation) using a dynamic clamp system (gLeak(artificial) = -6 nS) brings about a slow oscillation (B). The underlined section is expanded immediately to the right.
- supplemental material
-
Supplementary Figure 4. Effect of ZD7288 on an NRT neuron that does not exhibit an overt slow depolarization during the ‘down’ state during the slow oscillation.
Example of a “grouped LTCP” slow oscillation recorded in the peri-VB sector in the presence of 100 μM trans-ACPD which exhibits a more subtle slow depolariziation during the ‘down’ state than that illustrated in Fig. 7A (1). Despite the lack of such a clear depolarization this oscillation was still abolished by ZD7288 (30 μM) (2). Underlined sections in 2 are expanded below. Note the similarity in the activity of this neuron to that shown in Fig. 6C which initially does not exhibit an obvious slow depolarization, but which does so clearly after apamin application. Response of the neuron to hyperpolarizing current pulses reveals the presence of a subtle depolarization (left traces) which is blocked by ZD7288 (right traces) (3).
- supplemental material
-
Supplementary figure 5. Summary of the cellular mechanism of the slow (< 1 Hz) oscillation in NRT neurons.
The upper diagram illustrates the different ionic currents which are active at various points during the slow oscillation whereas the lower diagram depicts the proposed sequence of ionic events that shape the slow oscillation (cf. Fig. 3 in Crunelli et al., 2005; see also Fig. 10 in Beurrier et al., 1999). The presence of ‘up’ and ‘down’ states is primarily due to the presence and absence of ITwindow, respectively. During the ‘up’ state, IK(Ca) is rapidly activated and deactivated during action potential output which limits firing frequency, the level of depolarization reached and, consequently, the duration of the ‘up’ state. Upon transition to the ‘down’ state, any residual IK(Ca) quickly deactivates. In contrast, IK(Na) builds up gradually during the ‘up’ state and decays slowly during the ‘down’ state.
Note that the properties and mechanism of the slow oscillation in NRT neurons are similar to those previously described for TC neurons (Hughes et al., 2002b; Hughes et al., 2004; Crunelli et al. 2005). However, a number of important distinctions exist. First, the difference between ‘up’ and ‘down’ states is usually much less in NRT than in TC neurons (10-15 mV vs 20-30 mV). This is mainly due to the influence of IK(Ca) because apamin substantially increases the difference between ‘up’ and ‘down’ states in NRT neurons to ~30 mV. Second, whilst application of ZD7288 to slow oscillating TC neurons always induces membrane potential bistability (Williams et al., 1997; Hughes et al., 2002b), this is not the case in NRT neurons. These neurons are able to produce self-sustained equivalent ‘up’ states in response to small current steps in this condition but these are not maintained indefinitely. Again, this can be explained by the activation of various K+ currents that occurs during the ‘up’ state in response to Na+ and Ca2+ entry. Indeed, in the absence of IK(Na), IK(Ca) and Ih, our NRT model exhibits full membrane potential bistability (see also McCormick et al., 1997). Third, the NRT slow oscillation is much slower than that observed in TC neurons where the frequency is seldom below 0.1 Hz. In particular, whilst the ‘down’ state in TC neurons rarely lasts longer than 1-2 seconds, the ‘down’ state in NRT neurons is commonly in the range 5-15 seconds. The difference between the two cases is less pronounced when Na+ channels are blocked in NRT neurons suggesting that IK(Na), which is not present in TC neurons, determines the especially slow oscillation frequency and long ‘down’ state in NRT neurons. Indeed, specific removal of IK(Na) from our NRT model causes a similar decrease in ‘down’ state duration to that brought about by removing INa. Thus, it appears that the major differences in the slow oscillation between TC and NRT neurons can be largely attributed to the presence of certain K+ channels which are specific to cells in the NRT.
