View Full Text | Abstract |
Article as PDF | Print this Article |
Pubmed | PMC |
PubReader | Export to Citation |
Email Alerts | Open Access |
Exp Neurobiol 2023; 32(2): 83-90
Published online April 30, 2023
https://doi.org/10.5607/en22040
© The Korean Society for Brain and Neural Sciences
Dong Cheol Jang1,2,3, Changhyeon Ryu1†, Geehoon Chung3, Sun Kwang Kim3 and Sang Jeong Kim1*
1Department of Physiology, Neuroscience Research Center, Wide River Institute of Immunology, Seoul National University College of Medicine, Seoul 03080, 2Department of Brain and Cognitive Science, College of Natural Science, Seoul National University, Seoul 08826, 3Department of Physiology, College of Korean Medicine, Kyung Hee University, Seoul 02447, Korea
Correspondence to: *To whom correspondence should be addressed.
TEL: 82-2-740-8229, FAX: 82-2-763-9667
e-mail: sangjkim@snu.ac.kr
†Current address: The Picower Institute for Learning and Memory, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Ca2++ transients can be observed in the distal dendrites of Purkinje cells (PCs) despite their lack of action potential backpropagation. These Ca2++ events in distal dendrites require specific patterns of PC firing, such as complex spikes (CS) or simple spikes (SS) of burst mode. Unlike CS, which can act directly on voltage-gated calcium channels in the dendrites through climbing fiber inputs, the condition that can produce the Ca2++ events in distal dendrites with burst mode SS is poorly understood. Here, we propose the interspike interval threshold (ISIT) for Ca2++ transients in the distal dendrites of PC. We found that to induce the Ca2++ transients in distal dendrites the frequency of spike firing of PC should reach 250 Hz (3 ms ISI). Metabotropic glutamate receptor 1 (mGluR1) activation significantly relieved the ISIT and established cellular conditions in which spike firing with 50 Hz (19 ms ISI) could induce Ca2++ transients in the distal dendrites. In contrast, blocking T-type Ca2++ channels or depleting the endoplasmic reticulum Ca2++ store resulted in a stricter condition in which spike firing with 333 Hz (2 ms ISI) was required. Our findings demonstrate that the PC has strict ISIT for dendritic Ca2++ transients, and this ISIT can be relieved by mGluR1 activation. This strict restriction of ISIT could contribute to the reduction of the signal-to-noise ratio in terms of collecting information by preventing excessive dendritic Ca2++ transients through the spontaneous activity of PC.
Keywords: Cerebellum, Purkinje cells, Dendrites, Calcium signaling
The cerebellar Purkinje cell (PC) has pace-making activity through the following two different firing types: simple spikes (SS) that occur spontaneously and complex spikes (CS) that are induced by the climbing fiber (CF) inputs. Although both types of firing originate in the axon [1, 2], depolarization can lead to the elevation of Ca2+ concentration in the other loci of the PC, such as cell soma or dendrites. The distinct effects on the Ca2+ transients between the SS and CS have been reported in previous studies. That is, while somatic Ca2+ transients were observed in the SS and CS, dendritic Ca2+ transients were observed only in the CS [3, 4]. Studies in the hippocampal neurons suggested that backpropagation of action potential is required to induce Ca2+ transients in the dendrites beyond soma [5]. However, PC lacks action potential backpropagation owing to the distribution of Na+ and K+ channels in the soma and dendrites [1, 5-8]. Nonetheless, for the CS, CF inputs can cause significant membrane depolarization, which directly activates voltage-gated Ca2+ channels (VGCCs) in both the soma and dendrites, as CF innervates them. The probability of opening VGCCs in distal dendrites is low for SS, resulting in a lack of Ca2+ transients in that region. Moreover, a study reported that dendritic Ca2+ transients could be induced when the SS firing satisfies a specific pattern [3]. The PC has different types of SS (burst or tonic firing); however, dendritic Ca2+ transients could be observed only with the burst firing pattern. This suggests that PCs may have an interspike interval threshold (ISIT) to induce dendritic Ca2+ transients because the burst firing pattern has a much shorter interspike interval (ISI) than that of the tonic firing pattern. However, no studies have reported ISIT or the correlation between PC firing rates and dendritic Ca2+ transients.
Metabotropic glutamate receptor 1 (mGluR1) is predominantly expressed in cerebellar PCs and plays an important role in dendritic Ca2+ signaling [9]. The mGluR1 can generate Ca2+ transients by interacting with the T-type Ca2+ channel after activation by parallel fiber (PF) inputs [10]. Additionally, it can activate the inositol 1,4,5-trisphosphate receptor (IP3R), triggering Ca2+ release from the endoplasmic reticulum (ER) stores, leading to a dendritic Ca2+ transient [11]. Ca2+ released from the ER also contributes significantly to dendritic Ca2+ transients because the ER extends from the soma to the dendritic spine [12-14]. ER Ca2+ release is mediated by either ryanodine receptors or IP3Rs [15]. In PCs, the IP3Rs play a more critical role in dendritic Ca2+ transients [13, 14], and are necessary for cerebellar long-term depression (LTD) induction [15]. Moreover, mGluR1-dependent LTD is accompanied by decreased intrinsic excitability in PCs [11], indicating that mGluR1 activation can regulate PC excitability. Although mGluR1 can regulate both Ca2+ transient and PC excitability, the involvement of mGluR1 activity in ISIT for dendritic Ca2+ transients remain unclear.
Thus, we used whole-cell patch clamp recording in combination with two-photon microscopy to investigate ISIT to induce dendritic Ca2+ transients in PC. We mimicked the SS train by using 200 repeats of a 1-ms depolarization pulse and varied the intervals between the depolarizations. Using this voltage command, we recorded Ca2+ transients in the primary, secondary, and distal dendrites and soma. Our findings indicated that the ISIT to induce dendritic Ca2+ transients were extremely low, which requires an extremely high firing frequency to overcome the threshold. Activation of mGluR1 increased ISIT, while blocking the T-type Ca2+ channel or depleting ER stores decreased ISIT. To the best of our knowledge, this is the first study to describe ISIT in PCs. Our results suggest that PCs have strict ISIT to limit unnecessary Ca2+ transients induced by their spontaneous activity, and that mGluR1 plays a critical role in regulating ISIT.
All procedures for slice preparation were similar to our previous paper [16]. Sagittal slices of the cerebellar vermis (250 µm thick) were obtained from 5- to 9-week-old C57BL/6 mice using a vibratome (VT1200, Leica) after isoflurane anesthesia and decapitation. The slices were cut with ice-cold sucrose cutting solution containing (in mM): 75 sucrose, 75 NaCl, 2.5 KCl, 7 MgCl2, 0.5 CaCl2, 1.25 NaH2PO4, 26 NaHCO3, 25 Glucose. The slices were immediately moved to artificial cerebrospinal fluid (aCSF) containing (in mM): 125 NaCl, 2.5 KCl, 1 MgCl2, 2 CaCl2, 1.25 NaH2PO4, 26 NaHCO3, 10 glucose. For recovery, slices were incubated at 32℃ for 30 minutes and further 1 hour at room temperature. All procedures were approved by the Institutional Animal Care and Use Committee of Seoul National University College of Medicine.
The procedure for two-photon Ca2+ imaging was similar to our previous paper [16]. We used a multi-photon microscope (LSM 7 MP, Zeiss) equipped with a Ti:Sapphire laser (Chameleon Vision II, Coherent) and water-immersion objectives (W.Plan-Apochromat 20x/1.0 DIC M27 75 mm, W.Plan-Apochromat 63x/1.0 M27; Zeiss). Wide-field images were taken using a CCD camera (UM-300, Uniq vision). Depolarization stimuli were evoked by an EPC-8amplifier (HEKA Elektronik). For the stimuli, we set 200 ms of 0 mV depolarization as control stimulation (square depolarization, Fig. 1A). Then, we split this 200 ms square pulse into two hundreds of 1 ms depolarization (depolarization train, Fig. 1A). The 1 ms of depolarization was considered as the simple spike of the Purkinje cell, and we controlled the interval between the depolarizations (Fig. 1A). Cs-containing pipette solution was used for the experiment, in mM: 150 Cs-Methanesulfonate, 2 MgCl2*6H2O, 10 HEPES, 2 Na2ATP, 0.4 Na3GTP with 200 µM Oregon BAPTA 488 Green-1 (OGB-1, Invitrogen) and 25 µM Alexa Fluor 594 hydrazide (Sigma Aldrich). Purkinje cells were dialyzed for at least 30 minutes after membrane rupture for sufficient dye diffusion. We simultaneously obtained both Ca2+-sensitive (by 200 µM OGB-1) and Ca2+-insensitive images (by 25 µM Alexa-594) via two different optical filters (filter 1: 500~550 nm band-pass filter, filter 2: 575~610 nm band-pass filter). For drug treatments, all drugs were purchased from Tocris Bioscience.
Although the Alexa-594 is not a Ca2+-sensitive dye, we could observe some transients. To compensate these Ca2+-insensitive transients, we calculated the ratio (
All statistical analyses were performed using Graphpad Prism 9. Unpaired t-test, One-way ANOVA with post-hoc Dunnett test were used. All graphs except box and whisker plots are shown as mean±SEM. The whiskers of box plots indicate minimum and maximum values, and the midline indicates the median value. Asterisks *, ** and *** indicates p<0.05, p<0.01 and p<0.001, respectively. Detailed statistical methods and n for each experiment are written in the figure legends.
We first investigated the intracellular Ca2+ propagation under natural SS firing. We obtained the ratio of the OGB-1 to Alexa-594 signal (
To investigate the ISIT for Ca2+ transients at distal dendrites, we applied depolarization trains with various intervals and compared the area under the curve (AUC) before and during the stimulation (Fig. 2A). In the normal aCSF group, the Ca2+ transients at distal dendrites were significantly increased until 3 ms ISI and abolished with 4 ms ISI, corresponding to 250 Hz and 200 Hz, respectively (Fig. 2B). These results indicated that the Ca2+ transients by SS could occur under strictly limited conditions, and only ISIs under 4 ms could unlock this restriction.
A previous study reported that Ca2+ transients at distal dendrites are potentiated by mGluR1 activation [4]. Thus, we questioned whether mGluR1 activation could modulate the ISIT. To address the question, we tested a series of ISIs under bath application of 10 µM 3,5-Dihydroxyphenylglycine (DHPG), a mGluR1 agonist. Interestingly, the Ca2+ transients were significantly increased even with an ISI of 9 or 19 ms when DHPG was treated (Fig. 2C). Considering that these conditions satisfy the physiological range of SS firing in the PC (50~100 Hz frequency), these results indicate that mGluR1 activation could establish the cellular condition in which Ca2+ transients are promoted at distal dendrites via SS. Notably, DHPG treatment resulted in much higher levels of calcium elevation when an ISI of at least 3 ms was applied. The Ca2+ transients with an ISI of 3, 4, 9, or 19 ms were significantly higher in the DHPG-treated group than in the group without DHPG (Fig. 2C). The modulation of ISIT and the potentiation of Ca2+ transients by the DHPG treatment suggest that the Ca2+ elevation at the distal dendrites could be achieved by regular SS firing if mGluR1 is activated. Considering that mGluR1 is activated by PF inputs to dendrites, the mGluR1-mediated modulation of ISIT implies an important relationship between synaptic activity and PC plasticity, including synaptic and intrinsic plasticity.
In the PC, the elevation of intracellular Ca2+ concentration is achieved via several pathways, including the opening of T-type Ca2+ channels [10, 19] and IP3Rs [13, 14]. Thus, we investigated whether T-type Ca2+ channels and ER Ca2+ stores contribute to the modulation of ISIT. We found that bath application of mibefradil (1 µM), a T-type Ca2+ channels blocker, reduced the ISIT to 1 ms (Fig. 2D). The amount of Ca2+ elevation was comparable to that in the aCSF group stimulated using a square depolarization command; however, it was smaller when the depolarization train with 1 ms ISI was applied (Fig. 2D). To test the contribution of ER Ca2+ stores to the ISIT, we depleted ER stores by applying cyclopiazonic acid (CPA), a sarco/endoplasmic reticulum Ca2+ ATPase (SERCA) blocker. Since SERCA is a final executor of the ER refilling process [20], blocking SERCA causes ER store depletion. Before we experimented, the brain slices were treated for CPA (30 µM) for at least 30 mins [16]. Similar to T-type Ca2+ channel blocker treatment, CPA application also reduced the ISIT to 1 ms (Fig. 2D). Even when the ISIT was exceeded, the peak Ca2+ transients under CPA were significantly lower than those in the aCSF group (Fig. 2D, right). These results indicate that both T-type Ca2+ channels and ER Ca2+ stores contribute to the determination of ISIT, and blocking these Ca2+ sources increases the minimum firing frequency required for the induction of Ca2+ transients in distal dendrites.
Subsequently, we compared the AUC of the Ca2+ transients to estimate the amounts of Ca2+ elevation in each condition. When 200 ms of square depolarization pulse was injected, the AUC was not significantly different from untreated conditions regardless of mGluR1 activation, T-type Ca2+ channel block, or depletion of ER Ca2+ stores (Fig. 3, Square). In contrast, when the depolarization train protocols were used, depending on the ISI used, a marked change in AUC was observed. At 1 ms intervals, both mibefradil- or CPA-treated groups showed significantly decreased levels of AUC compared to the normal aCSF group, while the DHPG-treated group showed significantly increased levels of AUC (Fig. 3, 1 ms). The results suggest that Ca2+ influx via the T-type Ca2+ channel or Ca2+ release from ER stores contribute to Ca2+ transients in distal dendrites. As the interval threshold in the mibefradil and CPA treatment groups was 1 ms, the AUC of these groups at 2 ms intervals was approximately null (Fig. 3C, D). Similar to the 1 ms interval, the DHPG treatment group showed significantly higher levels of AUC than the untreated group under 2 ms interval stimulation (Fig. 3B). When the 4 ms interval, which was above the ISIT of the untreated group, was applied, only the DHPG-treated group showed significant levels of Ca2+ transients. The DHPG-treatment group showed a significant level of Ca2+ transients until the 19 ms interval stimulation (Fig. 3B). Notably, the AUC value at the 4 ms interval was greater than that at the 2 ms interval stimulation under DHPG conditions (Fig. 3D). This occurred presumably because the stimulation length of the 4 ms interval (1 s) was much longer than that of the 2 ms interval (600 ms).
In this study, we suggest that PCs have an ISIT for induction of Ca2+ transients in distal dendrites, that is, the ISI should be under than 3 ms (Fig. 4). In a normal, tonic firing mode, the PC fires at constant frequencies of 50~100 Hz [3, 17, 18] and the SS firings cannot reach the ISIT. Therefore, dendritic Ca2+ transients would not be introduced in normally firing PCs, as previously reported [3]. However, approximately 30% of PCs in the anterior lobule (lobule III to V) can show burst firing mode [21]. In burst mode, SS firings can reach the ISIT, and Ca2+ transients can be observed in distal dendrites [3]. In this case, Ca2+ transients would occur regularly in distal dendrites, increasing the probability of PF-PC LTD induction. In contrast, the remaining 70% of tonic firing PCs would not induce PF-PC LTD by its normal firing frequency, unless Ca2+ elevation is promoted by mGluR1 activation.
Further, we demonstrated that mGluR1 plays a critical role in ISIT regulation. Upon mGluR1 activation, the ISIT for dendritic Ca2+ transients is extremely relieved, allowing even 50 Hz SS to reach the threshold. We hypothesized that both tonic- and burst-firing SS would be capable of introducing dendritic Ca2+ transients under this condition. Since PF inputs activate mGluR1 to dendrites, mGluR1-mediated modulation of the ISIT implies an essential relationship between synaptic activity and plastic change in the PC. Therefore, PF bursts are a crucial source of acquired novel information [22], and these inputs activate mGluR1, which loosens the ISIT. In this case, tonic, as well as burst, SS firing can generate Ca2+ transients in the distal dendrite, increasing the probability of synaptic LTD. Presumably, repeated PF burst inputs in the early stage of learning could be a factor inducing LTD. Additionally, the continuous inputs in the late stage of learning may act as a positive feedback loop to maintain plasticity by loosening the ISIT and keeping the Ca2+ elevated. Thus, the ISIT might play an important role as a shutter, opening and closing the information window and enabling the induction and maintenance of synaptic plasticity at PF-PC synapses in an activity-dependent manner.
The dendritic Ca2+ transients generated by mGluR1 activation plays a critical role in the induction and maintenance of synaptic plasticity. A previous study demonstrated that DHPG treatment increases CS-induced dendritic Ca2+ transients, which are mediated by P/Q-type VGCCs [4]. However, whether the high-frequency SS firing, which is above the ISIT threshold under mGluR1 activation, induces dendritic depolarization as CS remains unclear. In this case, the additional Ca2+ transients following DHPG treatment would be mediated by P/Q-type VGCCs. If it fails to induce dendritic depolarization, as previously observed, the extra Ca2+ transients may be attributed to an additional release of Ca2+ from the ER store. The source of these Ca2+ transients should be identified in further studies.
This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (NRF-2021R1C1C1013840 to D.C.J.; NRF-2020R1C1C1009162 to G.C.; NRF-2019R1A2C2086052 to S.K.K; NRF-2018R1A5A2025964 to S.J.K).