T-type channels buddy up
- 1.3k Downloads
The electrical output of neurons relies critically on voltage- and calcium-gated ion channels. The traditional view of ion channels is that they operate independently of each other in the plasma membrane in a manner that could be predicted according to biophysical characteristics of the isolated current. However, there is increasing evidence that channels interact with each other not just functionally but also physically. This is exemplified in the case of Cav3 T-type calcium channels, where new work indicates the ability to form signaling complexes with different types of calcium-gated and even voltage-gated potassium channels. The formation of a Cav3-K complex provides the calcium source required to activate KCa1.1 or KCa3.1 channels and, furthermore, to bestow a calcium-dependent regulation of Kv4 channels via associated KChIP proteins. Here, we review these interactions and discuss their significance in the context of neuronal firing properties.
KeywordsCav3 T-type KCa3.1 KCa1.1 BK Kv4 A-type
Control over the frequency and pattern of neuronal spike output defines neural coding of information in the brain. Central to this process are ion channels that conduct potassium to control excitability by hyperpolarizing the membrane potential. We know of numerous isoforms of voltage-gated potassium channels that contribute to controlling excitability [23, 47]. But few of these have as key a role in regulating the frequency and pattern of spike discharge as calcium-gated potassium channels . A great deal of work has focused on the ability for high-voltage-activated (HVA) calcium channels to activate either small conductance (SK, KCa2.x) [2, 76, 125] or big conductance (BK, KCa1.1) potassium channels to control cell excitability [10, 11, 12, 45, 139]. Calcium-dependent control of potassium channels has also been recognized to reflect interactions at the level of either a microdomain or nanodomain, a designation that signifies an interchannel distance of <50 or 50–200 nm, respectively . This is important because it reflects an entirely different degree of control that calcium influx can exert on potassium channel activation that may be necessary to effect different cellular functions. Indeed, interactions at the nanodomain level can allow the voltage dependence of specific HVA calcium channel isoforms to be conferred onto KCa1.1 channels , providing greater control over the onset voltage and time of hyperpolarizing currents.
Until recently there were only a few reports of Cav3 (T-type) calcium channels being functionally coupled to activation of either KCa2.x [26, 138] or KCa1.1 [44, 111] channels. This coupling was defined entirely on the basis of physiological interactions with little protein biochemical work to assess the nature of the link or its control of potassium channel function at the level of a microdomain or nanodomain. Recent work reveals that Cav3 calcium channels can form an association at the molecular level with calcium-gated potassium channels and even a voltage-gated potassium channel. Thus, Cav3 channels have been shown through protein biochemical and biophysical analyses to associate closely with KCa1.1 channels  as well as intermediate conductance calcium-activated potassium channels (KCa3.1, SK4, KCa3.1) . Moreover, an association at the molecular level was detected between Cav3 channels and the Kv4 family of voltage-gated potassium channels that generate transient A-type currents [4, 5]. In each case linking potassium channel activation to calcium influx through Cav3 channels allows outward current to be triggered from membrane voltages well below resting membrane potential and even over the course of a full-blown spike response. Interestingly, activation of these potassium channels by Cav3 calcium influx relies on three distinct calcium sensing mechanisms. The ability for these complexes to function at either a microdomain or nanodomain level proves to depend on the sensitivity of the calcium sensor in relation to the relatively weak conductance of Cav3 channels compared to HVA calcium channels.
Several recent reviews have been published on the properties of T-type calcium channels on topics that will not be covered here [16, 20, 27, 51, 64, 65, 100, 120, 137]. This review summarizes the current state of knowledge of how Cav3 channel associate with three distinct forms of potassium channel to form ion channel complexes that acquire functional roles that reflect the combination of biophysical properties of each partner in the complex. To fully understand the interplay between calcium and potassium channels, we briefly summarize key features of the responsible subunits and proteins involved.
Cav3 calcium channels
Voltage-gated calcium channels permit the entry of calcium ions into the cytosol in response to membrane depolarizations. Voltage-gated calcium channels can be divided into two major families: HVA calcium channels that open in response to large membrane depolarizations and low-voltage-activated (LVA) channels that open in response to smaller membrane depolarizations . The HVA channels include L-, P-, Q-, N-, and R-types which can be distinguished based on their biophysical and pharmacological properties. HVA channels share a common multimeric assembly of Cavα1, Cavα2δ, and Cavβ subunits to form a functional complex . Moreover, these channels all interact constitutively with calmodulin . In addition to their more depolarized range of activation, they are distinguished from LVA channels by their larger single channel conductance and open probability, which allow these channels to support large calcium influxes at depolarized potentials. Indeed, it has been estimated that a single HVA calcium channel will increase internal calcium from a resting value of 70–100 nM to greater than 40 μM within a millisecond 15 nm distant from the channel pore .
Calcium-gated potassium channels
KCa1.1 and KCa2.x potassium channels have been traditionally recognized as those gated by calcium entry to control membrane excitability in central neurons [1, 12, 129, 135]. The properties of KCa1.1 and KCa2.x channels differ in key respects that support fine-tuned roles in mediating spike repolarization and afterhyperpolarizations (AHPs) over relatively short time frames of activity [1, 66, 67, 85, 116, 123, 133]. All three KCa2.x channel isoforms (KCa2.1–3; SK1–3; KCNN1–3) [1, 12, 135] are purely calcium dependent due to the association of calmodulin with the C-terminal region [1, 57, 58, 63, 141]. While Cav3 calcium channels have been shown to at least functionally couple to KCa2.x channels [26, 138], we will not focus on this given the lack of protein biochemical evidence for a potential ion channel complex at this time.
KCa1.1 channels are similar to voltage-gated potassium channels in activating in response to membrane depolarization; however, the voltage dependence of activation is strongly regulated by cytosolic free calcium concentration . KCa1.1 channels are formed by the association of four identical pore forming α-subunits plus ancillary β-subunits which coassemble with the α-subunit in a 1:1 fashion. The α-subunit contains seven membrane spanning helices (S0 to S6), with an extracellular N-terminus and an intracellular C-terminus  (Fig. 1b). The C-terminus contains a “regulating conductance of potassium” (RCK) domain and a calcium bowl to confer calcium sensitivity onto KCa1.1 channel activation [32, 43, 46, 82, 102, 111]. The affinity of the calcium binding site on KCa1.1 channels is substantially lower than that of calmodulin associated with KCa2.x and KCa3.1 channels, with reported minimal intracellular calcium concentrations necessary for activation ranging from 1 to >10 μM [12, 85, 123]. As a result KCa1.1 channels are typically activated by relatively large voltage responses such as spike discharge, thereby contributing to spike repolarization and a fast AHP (fAHP) [85, 115, 139]. Previous studies have shown that KCa1.1 channels can form signaling complexes with HVA channels, thereby localizing the channels close to the source of calcium entry [10, 11, 45, 85], but the underlying channel structural determinants are unknown. Indeed, all of Cav1.2 (L), Cav2.1 (P), and Cav2.2 (N) channels can form what were proposed as supercomplexes with KCa1.1 channels, although not with the Cav2.3 calcium channel isoform [10, 12]. The functional importance of this interaction has been repeatedly established, such that blockade of HVA calcium channels (and hence, indirectly, the activity of KCa1.1 channels) in neurons results in drastic alterations of intrinsic neuronal firing properties [109, 112, 115, 117, 139]. New data summarized below now indicates that LVA Cav3 calcium current also has an important role in activating KCa1.1-mediated outward current.
KCa3.1 channels are a third class of calcium-gated potassium channel that belong to the same gene family as KCa2.x channels but share only ∼45 % protein sequence homology [52, 57, 72, 135]. KCa3.1 channels are also only calcium dependent through the association with calmodulin [37, 40, 52, 57, 72, 96, 105, 106, 113] but activate and deactivate over a much longer time frame (up to seconds) than KCa1.1 or KCa2.x channels [50, 68, 69, 126]. KCa3.1 channels are derived from a single gene (KCNN4) with virtually all KCa3.1 channels sequenced from various body tissues (i.e., pancreas, placenta, lymphocytes) sharing an equivalent sequence [52, 57, 72]. KCa3.1 channels have a six-transmembrane domain structure and intracellular N- and C-termini (Fig. 1b). The proximal C-terminus contains a constitutive binding site for calmodulin  with an IC50 for calcium from 95 to 300 nM compared to ∼300–500 nM for KCa2.x channels [1, 57], making KCa3.1 channels potentially more sensitive to changes in internal calcium concentration.
The expression pattern and properties of KCa3.1 channels have been extensively examined outside the CNS, where KCa3.1 channels are expressed in red blood cells , endo- and epithelial cells [8, 39, 48, 119, 130, 132], lymphocytes [42, 61], and glia [19, 60, 61, 71]. The function of KCa3.1 channels in controlling neuronal excitability has been best characterized in enteric neurons, where KCa3.1 channels generate a slow AHP (sAHP) of seconds duration [91, 92, 128]. However, in central regions, KCa3.1 channels were not believed to be expressed in neurons [52, 53, 72] but instead were restricted to endothelial cells and activated glia [9, 53, 124, 140]. Nevertheless, reports suggesting the expression of KCa3.1 channels in more central neurons exist, with molecular, immunocytochemical, or electrophysiological data reported in sensory cells of peripheral and autonomic nervous systems [39, 83, 90, 92, 127] as well as motor neurons [13, 83]. Most recently, KCa3.1 channel expression in a central neuron was established for the first time in cerebellar Purkinje cells , raising questions as to how widespread KCa3.1 channel expression may be in other central regions. Moreover, the first study examining KCa3.1 channels in Purkinje cells revealed the presence of a Cav3-KCa3.1 complex with important functional roles .
Voltage-gated A-type potassium channels
Among the myriad of voltage-gated potassium channels are a small subset of channels that are activated from a low membrane voltage and in a transient fashion to trigger “A-type” currents. In central regions, A-type potassium currents are often generated by members of the Kv4 channel family (Kv4.1–4.3; KCND1–3) [54, 108]. As found for Cav3 calcium channels, Kv4 channels exhibit a low voltage for activation, fast inactivation, near-complete inactivation at resting potential, and an availability that is governed by preceding membrane hyperpolarizations . Kv4 channels share the common structure of potassium channels in being comprised of four α-subunits containing six transmembrane domains, an S4 voltage-sensing domain, and intracellular C- and N-termini (Fig. 1b). All structural indices then suggest that Kv4 channels belong entirely to the class of voltage-gated channels. However, reports as early as the 1980s began to suggest that A-type potassium channels may also be regulated by calcium influx, although the mechanism was never resolved [14, 18, 74, 107, 143]. Kv4 channels were subsequently found to link to “potassium channel interacting proteins” (KChIP1–4; Fig. 1b), a class of calcium sensor molecules that affect channel translocation and kinetic properties [3, 15, 62, 94, 98, 104, 110, 131]. A second auxillary subunit termed dipeptidyl peptidase-like proteins (DPPs) was identified as membrane spanning proteins that directly interact with the Kv4 α-subunit [30, 38, 54, 56, 59, 87, 88, 99, 103]. While DPPs also exert important effects on baseline properties of Kv4 channels, there is no evidence to date for a role in mediating calcium-dependent effects. The combination of Kv4 α-subunits together with KChIP and DPP proteins together is now recognized as comprising the “Kv4 complex” [22, 24, 55, 59, 75, 86].
The presence of KChIP molecules as part of the Kv4 complex signifies the presence of a very different form of calcium sensor than those inherent to KCa3.1 or KCa1.1 channels. Here, four KChIP molecules bind to the Kv4 α-subunit N-termini to form a “cross-shaped octomer” on the cytoplasmic side to bind calcium [98, 131]. KChIPs1–4 are derived from separate genes and belong to the larger family of calcium sensor proteins that contain four EF-hand domains on the C-terminus [15, 73]. EF-1 does not bind calcium and EF-2 is bound to magnesium, leaving EF-3 and EF-4 as potential calcium binding sites. The dissociation constant for calcium binding to EF-3 and EF-4 is 5 μM  and therefore midrange in sensitivity to [Ca]i from that of calmodulin (<300 nM) or KCa1.1 RCK domains (1–10 μM). Although KChIP proteins were identified as a mechanism to provide calcium-dependent regulation of Kv4 current, the effects of calcium influx on KChIPs and Kv4 current were not subsequently reported. The potential source of calcium that might activate KChIPs also remained unknown. As elaborated below, recent work reveals that Cav3 calcium channels selectively associate with the Kv4 complex to modulate A-type current in a KChIP3-dependent manner [4, 5].
Cav3 interactions with potassium channels
The physiological role for KCa3.1 channels became apparent when simulated EPSCs (simEPSCs) were used to evoke simulated EPSPs (simEPSPs) at the level of Purkinje cell somata to test the role of postsynaptic ion channels in generating a synaptic response . Applying either mibefradil to block Cav3 calcium entry or TRAM-34 to block KCa3.1 channels produced an equivalent block of the decay phase of the simEPSP in Purkinje cells (Fig. 2g). By comparison, none of the established blockers for HVA calcium channels, KCa2.x channels, or KCa1.1 channels could reproduce these effects. When a train of subthreshold simEPSPs was applied at 25 Hz, it was found that temporal summation of EPSPs was strongly suppressed beyond the first four to five stimuli but that consistent and continual spike discharge was evoked when TRAM-34 was applied to block KCa3.1 channels (Fig. 2h). Importantly, TRAM-34 was also effective in the presence of picrotoxin used to block feedforward inhibitory projections activated by parallel fibers , indicating that both systems serve to regulate EPSP summation. KCa3.1 channel activation by Cav3 calcium influx could thus be traced to suppressing the baseline EPSP-membrane voltage response during repetitive synaptic input, acting as a high-pass filter to suppress background parallel fiber synaptic activity under normal conditions.
The link between Cav3.2 calcium influx and KCa3.1 activation was all supported by the close association indicated through coimmunoprecipitation and immunocytochemistry. To further assess the distance between the two channels in the complex, we employed the use of the two calcium chelator EGTA and BAPTA. Here, it was shown in outside-out patch recordings that TRAM-34-sensitive outward current was present when 10 mM EGTA was included in the electrode but not in the presence of 10 mM BAPTA (Fig. 2i). This was important in establishing that the Cav3-KCa 3.1 complex functions at the level of a calcium nanodomain or <50-nm distance . The exact sites that form a link between Cav3.2 and KCa3.1 have not yet been identified nor has the degree to which these findings extend to Cav3.1 or Cav3.3. Yet taken together, these data argue strongly for the first nanodomain interaction between Cav3 calcium channels and a calcium-activated potassium channel.
Protein biochemical tests revealed that Cav3.2 channels coimmunoprecipitate with KCa1.1 from lysates of either the brain or cerebellum (Fig. 3c) and from lysates of tsA-201 cells expressing only the channel α-subunits (Fig. 3d). Further tests for coimmunoprecipitation between Cav3.2 and different KCa1.1 mutant channels further narrowed down the site for interaction. Thus, Cav3.2 channels coimmunoprecipitated with a mutant construct comprised of only the KCa1.1 N-terminus and S0 transmembrane segment (N + S0), with a naturally occurring truncated N-terminal variant with SO (aN + S0), but not with a construct comprised of only the longer N-terminus (Fig. 3d). These data strongly implicate a site of interaction between Cav3.2 and the transmembrane S0 segment.
Previous work had reported at least a functional coupling between Ni2+-sensitive calcium influx and KCa1.1 in MVN cells . However, direct biophysical tests of the nature of this activation had not been reported. We used whole-cell voltage clamp of MVN cells in vitro to examine KCa1.1 current isolated by either paxilline (1 μM) or TEA (1 mM) application and compared these to outward currents isolated by perfusing the Cav3 blockers mibefradil (1 μM) or Ni2+ (300 μM) (Fig. 3e) . In each case the isolated current was fast activating and slowly inactivating over the time of a 200-ms step command to +40 mV (Fig. 3e). Using a ramp command over a −100 to −30 mV range with paxilline or mibefradil treatment further revealed LVA outward current that activated just subsequent to LVA inward calcium current, indicating that Cav3 calcium influx can activate KCa1.1 above ∼−70 mV (Fig. 3f). Finally, applying mibefradil to MVN cells under current clamp slowed spike repolarization, reduced a subsequent fAHP (Fig. 3g), and increased the gain of firing, revealing that the Cav3-KCa1.1 interaction normally reduces spike output in MVN cells. These data were entirely novel in identifying a new association between Cav3 and KCa1.1 channels in an ion channel complex, helping account for earlier reports of the role for KCa1.1 in controlling gain in MVN cells [89, 111].
A series of recent studies on stellate cell interneurons in the cerebellum led to a new understanding of how Cav3 channels can interact with even voltage-gated potassium channels. Earlier work on the expression pattern of Cav3 channel isoforms in these cells  led to the finding that membrane hyperpolarizations expected to promote a rebound increase in firing frequency instead promoted a unique voltage-dependent shift in first spike latency . As ionic control of first spike latency has traditionally been assigned to A-type potassium channels, it prompted further investigation into how Cav3 channels might influence this important aspect of spike firing dynamics. It was finally concluded that the coexpression of Cav3 and Kv4 channels led to a non-monotonic profile in the voltage-first spike latency profile, with Cav3 calcium influx reducing latency from more hyperpolarized potentials, and Kv4 current increasing spike latency from membrane potentials near rest . However, there was a curious peak in the voltage-latency profile near resting membrane potential that proved to be sensitive to Cav3 channel blockers, implying that Cav3 and Kv4 channels may establish first spike latency characteristics by interacting at a level beyond simple coexpression in the membrane.
A series of studies established that infusing either a KChIP3 or PanKChIP antibody through the electrode (1:100 dilution) during stellate cell recordings mimicked the effects of applying mibefradil in left-shifting Kv4 V h (Fig. 4d), providing the first means of interfering with Cav3-Kv4 complex function. With this tool in hand, one could test the functional role of a Cav3-Kv4 complex, with antibody infusion revealing that a Cav3 calcium-mediated rightward shift in the voltage-inactivation profile ordinarily acts to increase Kv4 window current in the region of spike threshold (Fig. 4e). As a result, the Cav3-Kv4 complex reduces cell excitability, as measured by a significant drop in firing rate gain (Fig. 4f) . A Cav3-mediated increase in Kv4 window current near rest also accounted for the increase in the voltage-first spike latency relationship that was identified in stellate cells near resting potentials (Fig. 4g) . While repetitive stimulation to increase internal calcium concentration is without apparent effect on Kv4 availability, the Cav3-Kv4 complex proves to be highly sensitive to decreases in extracellular calcium levels (Fig. 4h) . This is important because it has been established that physiological rates of excitatory afferent input to the cerebellum rapidly decrease extracellular calcium concentration in the order of 30 % from resting values [49, 93, 114]. We thus used dual recordings between a Purkinje cell and a stellate cell directly above in the molecular layer to assess the effects of climbing fiber stimulation on I A and I T . As shown in Fig. 4i, I A was isolated pharmacologically while recording from a Purkinje cell under current clamp to monitor complex spike discharge. Repetitive complex spike discharge at 10 Hz rapidly decreased I A amplitude, with full recovery within 2–3 s after the end of a stimulus train. Similarly, climbing fiber stimulation decreased I T in stellate cells during a stimulus train, with even faster kinetics of recovery (data not shown). Since the decrease in I A availability increased the firing rate gain of stellate cells, it was found that the Cav3-Kv4 complex functions to provide adaptive inhibitory control over Purkinje cells in the face of excitatory afferent-induced reductions in extracellular calcium . These findings were key in establishing that expression of the Cav3-Kv4 complex in stellate cells ultimately underlies a new form of homeostatic control over circuit function .
Microdomain versus nanodomain interactions
Given widely different calcium sensors and potassium channels involved in these ion channel complexes, it is interesting to compare the relative efficacy of Cav3 calcium channels to effect potassium channel activation. Previous work has shown that calcium influx gives rise to a domain of decreasing internal calcium surrounding the internal pore of the channel. The distance over which calcium influx can trigger a calcium-dependent event is distinguished by those functioning at the level of a calcium microdomain (50–200-nm distance) or a nanodomain (<50 nm) . It is known that interactions between the HVA calcium channels Cav2.1 or Cav2.2 and KCa1.1 in an expression system occur at the level of a nanodomain . The ability for HVA calcium influx to activate KCa1.1 is such that even the voltage dependence and kinetics of either calcium channel subtype is imparted upon KCa1.1 activation. Given that all of the complexes reviewed here share Cav3 channels as a common partner, key factors that could be predicted to control the degree of potassium channel activation would be (1) the relative proximity of calcium and potassium channels, (2) the sensitivity of the calcium sensors to changes in internal calcium, and (3) the relative conductance of Cav3 calcium channels. The latter is of particular interest given that Cav3 channels have a much lower conductance than HVA calcium channels and exhibit rapid inactivation that will substantially reduce the time for internal calcium accumulation.
In the case of KCa1.1 channels, the situation is less clear. On the one hand, coimmunoprecipitation data indicate that Cav3 and KCa1.1 channels are part of a macromolecular signaling complex that places KCa1.1 channels close to the source of Cav3 calcium entry. But Cav3-mediated activation of KCa1.1 is less reliable in being blocked by as little as 5 mM internal EGTA in both tsA-201 and MVN cells. Activating KCa1.1 in tsA-201 cells further required an initial prepulse command to −30 mV to maximally activate Cav3 current. Once sufficient activation of KCa1.1 was achieved, Cav3 calcium influx left-shifted the voltage dependence of KCa1.1 activation, with additional evidence for voltage-dependent inactivation of KCa1.1, as expected for steady-state inactivation properties of Cav3 channels. These data are perplexing in presenting a calcium-dependent activation process more reminiscent of a microdomain interaction. However, this implies a distance between the channels of 50–200 nm, which is greater than what one expects from a direct molecular interaction. In contrast, it has been shown that physical signaling complexes between N-type calcium channels and KCa1.1 channels allow a functional interaction at the level of a nanodomain (Fig. 5c). The key difference from Cav3 channels is a much larger open probability and single channel conductance of N-type channels, such that each channel is sufficient to provide the calcium levels needed for KCa1.1 activation .
Lessons learned from studying calmodulin regulation of calcium-dependent inactivation (CDI) of HVA channels can be used to explain this apparent discrepancy. For both L-type and P/Q-type channels, calmodulin is preassociated with the calcium channel. A rise in intracellular calcium results in calcium binding to calmodulin, thereby triggering a rearrangement of the channel-calmodulin complex to cause CDI . Calmodulin has two high-affinity and two low-affinity binding sites for calcium. CDI of L-type channels is dependent on the high-affinity calcium binding lobes, whereas that of P/Q-type channels relies on the low-affinity binding sites. Therefore, calcium entry via an individual L-type channel is sufficient to raise calcium concentrations near the high-affinity sites on calmodulin to trigger CDI, making this resistant to internal EGTA . The situation is different with P/Q-type channels, where the lower affinity of calmodulin for calcium requires a much larger rise in intracellular calcium for CDI to occur. Importantly, this rise in calcium cannot be supported by an individual channel but rather requires the concerted action of many channels . Thus, despite the fact that each channel is already associated with a calmodulin molecule, CDI of P/Q-type channels is EGTA sensitive.
We hypothesize that apparent incongruencies in data showing a Cav3-KCa1.1 macromolecular complex that functions at the level of a microdomain parallel what is observed with CDI of P/Q-type channels. Much like the low-affinity calcium binding site on calmodulin, the calcium binding site on KCa1.1 channels has a low affinity for calcium. Therefore, calcium entry through a single Cav3 calcium channel is insufficient to activate KCa1.1 even though the two proteins are in close proximity. However, a concerted entry of calcium through multiple Cav3-KCa1.1 complexes is proposed to support a global rise in calcium that is sufficiently high to activate KCa1.1 and, thus like CDI of P/Q-type channels, is sensitive to internal EGTA (Fig. 5d). An additional consideration concerns the stoichiometry between Cav3 calcium and an associated potassium channel. Potassium channels are tetrameric assemblies, and thus, in principle, each potassium channel could interact with four Cav3 channels, although experimental evidence for a 4:1 stoichiometry is lacking. Along these lines, each potassium channel may have four different calcium sensors (i.e., four KChiP proteins, four calmodulin molecules, or four RCK domains/calcium bowls). Hence, the calcium dynamics of potassium channel activation/modulation may be immensely complex, giving rise to unexpected EGTA sensitivities that would complicate distance estimates based on the use of intracellular calcium buffers alone.
As we have discussed here, Cav3 channels can form molecular associations with three different potassium channel classes that rely on very different mechanisms for calcium sensing. The primary determinants for functional coupling at either a microdomain or nanodomain level appear to be defined by the relative conductance of the calcium channel and sensitivity of the calcium sensor to changes in intracellular calcium concentration. As result, initial work on the Cav3-KCa1.1 complex revealed a new form of calcium-dependent activation of potassium channels not originally expected according to the conventional means of defining interactions at the micro- and nanodomain levels (i.e., the use of buffer sensitivity alone is insufficient to paint a complete picture). Another novel aspect is the observation that KCa3.1 channels are in fact expressed in CNS neurons where they regulate neuronal output, in part by forming a Cav3-KCa3.1 channel complex that can activate in the subthreshold region. The ability for Cav3 calcium current to activate KCa3.1 channels further confers voltage-dependent gating properties from Cav3 channels onto a normally voltage-insensitive KCa3.1 channel. Conversely, Cav3 channels confer calcium sensitivity onto the normally purely voltage-sensitive Kv4 channels via KChIP3 proteins. This altogether widens the traditional view of different ion channels acting independently—instead, there is intricate interplay between different ionic channels through formation of macromolecular signaling complexes. What remains to be examined is whether there is feedback regulation that allows the associated potassium channels to regulate Cav3 channels. Furthermore, it remains to be determined if the formation of signaling complexes can affect other channel features, such as trafficking to the plasma membrane and specific localization within neurons. Our observations that Cav3-K channel complexes can profoundly impact neuronal output may thus be only one of many important aspects of these molecular associations.
We gratefully acknowledge the contributions of numerous laboratory members in the experimental and analytical components of the work summarized here. This work was supported by operating grants from the Canadian Institute of Health Research. R.W.T. and G.W.Z. are AI-HS Scientists and G.W.Z. holds a Canada Research Chair.
- 5.Anderson D, Rehak R, Hameed S, Mehaffey WH, Zamponi GW, Turner RW (2010) Regulation of the K(V)4.2 complex by Ca(V)3.1 calcium channels. Channels (Austin) 4(3):163–167Google Scholar
- 30.Dougherty K, Tu L, Deutsch C, Covarrubias M (2009) The dipeptidyl-aminopeptidase-like protein 6 is an integral voltage sensor-interacting beta-subunit of neuronal K(V)4.2 channels. Channels (Austin) 3:122–128Google Scholar
- 90.Neylon CB et al (2004) Intermediate-conductance calcium-activated potassium channels in enteric neurones of the mouse: pharmacological, molecular and immunochemical evidence for their role in mediating the slow afterhyperpolarization. J Neurochem 90:1414–1422. doi: 10.1111/j.1471-4159.2004.02593.x PubMedGoogle Scholar
- 109.Shao, L.R., Halvorsrud, R., Borg-Graham, L., Storm, J.F. (1999) The role of BK-type Ca2+-dependent K+ channels in spike broadening during repetitive firing in rat hippocampal pyramidal cells. J Physiol 521 Pt 1:135-146Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.