Clustering of IP₃ receptors by IP₃ retunes their regulation by IP₃ and Ca²⁺

Abstract

The versatility of Ca²⁺ signals derives from their spatio-temporal organization^1,2. For Ca²⁺ signals initiated by inositol trisphosphate (IP₃) this requires local interactions between IP₃ receptors (IP₃R)^3,4 mediated by their rapid stimulation and slower inhibition⁴ by cytosolic Ca²⁺. This allows hierarchical recruitment of Ca²⁺ release events as the IP₃ concentration increases⁵. Single IP₃R respond first, then clustered IP₃R open together giving a local Ca²⁺ puff, and as puffs become more frequent they ignite regenerative Ca²⁺ waves^1,5-9. We demonstrate, using nuclear patch-clamp recording¹⁰, that IP₃R are initially randomly distributed with an estimated separation of ~1 μm. Low concentrations of IP₃ cause IP₃R to aggregate rapidly and reversibly into small clusters of ~4 closely associated IP₃R. At resting cytosolic [Ca²⁺], clustered IP₃R open independently, but with lower open probability (P_o), shorter open time, and lesser IP₃ sensitivity than lone IP₃R. Increasing cytosolic [Ca²⁺] reverses the inhibition caused by clustering, IP₃R gating becomes coupled, and the duration of multiple openings is prolonged. Clustering both exposes IP₃R to local Ca²⁺ rises and increases the effects of Ca²⁺. Dynamic regulation of clustering by IP₃ tunes IP₃R sensitivity to IP₃ and Ca²⁺, facilitating hierarchical recruitment of the elementary events that underlie all IP₃-evoked Ca²⁺ signals^3,5.

IP₃-activated currents recorded from patches excised from the outer nuclear envelope of DT40 cells¹⁰ expressing rat IP₃R3 are entirely due to IP₃R3 (Fig. 1). With 10 μM IP₃ in the pipette solution (PS) the single channel open probability (P_o) was 0.44 ± 0.05 (n = 6) and the mean open time (τ_o) was 11.9 ± 1.6 ms. The distribution of closed times (τ_c) had two components (Fig. 1d). Recordings in the on-nucleus configuration confirmed these results (not shown). The results are consistent with the gating scheme shown in Figure 1d (see Supplementary Methods).

Figure 1. IP₃R are randomly distributed.

a, IP₃-evoked Ca²⁺ release from permeabilized DT40-IP₃R3 (EC₅₀ = 281 ± 46 nM) and DT40-KO cells (means ± SEM, n ≥ 3). Immunoblot with IP₃R3-specific antiserum (10 μg membrane protein/lane, 220-kDa marker shown). b, Currents recorded from excised patches with 10 μM IP₃ in PS. No currents were detected without IP₃ (n = 20), with heparin (100 μg/ml) and IP₃ (n = 15), or with IP₃ in DT40-KO cells (n > 30). C denotes closed state. c, i-V relationship for IP₃-evoked current (γ_K = 121 ± 2.8 pS, n = 7). d, Dwell time distribution of single IP₃R3 stimulated with 10 μM IP₃. Open time distribution of this typical recording is fitted with a single probability density function (pdf) with τ_o = 10.4 ms (mean = 11.9 ± 1.6 ms, n = 6). The pdf for the τ_c distribution has two components (1.07 ms, 88% and 109 ms, 12%). Dwell time distributions are consistent with the gating scheme (Supplementary Methods, Supplementary Figs 5, 6). e, Typical all-points current amplitude histogram of an excised patch containing 3 IP₃R stimulated with 10 μM IP₃; C and O denote closed and open states. f, Observed and predicted numbers of IP₃R/patch from 109 patches (mean = 1.34) stimulated with 10-100 μM IP₃.

The number of channels within a patch (1.34 ± 0.13, n = 109) can be estimated reliably from the largest multiple of simultaneous openings to the unitary current level (Fig. 1e, Methods). The distribution of IP₃R in a patch is random: it is not significantly different from a Poisson distribution (χ², p>0.05; Fig. 1f, Supplementary Table 1). Others suggested that IP₃R are clustered in the nuclear envelope^11,12, but it seems likely that in making repeated recordings from the same nucleus they stimulated nuclei with IP₃ before recording and thereby caused IP₃R clustering (see below).

Channel activity (P_o, Fig. 2a-c), but not the number of active IP₃R (Fig. 2d), increased with IP₃ concentration (EC₅₀ = 1.38 ± 0.03 μM for patches with one IP₃R). There was more than one IP₃R in 57% of active patches, and each opened to the same γ (Fig. 1e, 2a), but NP_o (the overall channel activity) was less than expected from the summed behaviour of lone IP₃R (Fig. 2e). For multi-IP₃R patches, the sensitivity to IP₃ of NP_o was also significantly reduced (EC₅₀ = 2.47 ± 0.25 μM for patches with 3 IP₃R, Fig. 2c, Supplementary Table 2). Do IP₃R behave independently in such multi-IP₃R patches or do they interact, like some ryanodine receptors^13,14? For each of the four states in patches with three IP₃R (closed and 1, 2 or 3 simultaneously open IP₃R), P_o predicted from the binomial distribution matched the observed P_o (Fig. 2f, Methods). Similar results were obtained for patches with different numbers of IP₃R and for type 1 IP₃R (Supplementary Figs 1, 2). At resting cytosolic [Ca²⁺], therefore, each IP₃R within a multi-IP₃R patch behaves identically and opens independently.

Figure 2. Lone IP₃R are more active than clustered IP₃R at resting cytosolic Ca²⁺.

a, Typical records from patches (2 IP₃R/patch) stimulated with IP₃ (μM). b, c, Effect of IP₃ on P_o of patches containing a single IP₃R (b) or on NP_o of patches with 3 IP₃R (c) (n ≥ 4). d, Numbers of IP₃R detected in each patch for each IP₃ concentration (n = 9-25). e, Predicted (ie NP_lone) and observed NP_o for patches containing 1-5 IP₃R (n ≥ 3; n = 2 for 5-IP₃R patch). f, For patches with 3 IP₃R, observed/predicted values are shown for the indicated numbers of simultaneous openings (Supplementary equation 4). g, P_o as a function of the number of IP₃R within a patch after stimulation with 10 μM IP₃ (Supplementary equation 5). h, Effect of IP₃ on P_o for lone IP₃R and IP₃R within multi-IP₃R patches (n ≥ 4).

How can randomly distributed IP₃R that open independently behave with such uniformity, and yet so differently from lone IP₃R, when a patch fortuitously contains several IP₃R? Recordings from Xenopus nuclei also suggest that heterogenous behaviour of lone IP₃R becomes more uniform when patches contain several IP₃R¹⁵. We suggest that IP₃ causes IP₃R to cluster¹⁶ and that clustered IP₃R are less active. To test this hypothesis, nuclei were bathed in IP₃ (10 μM, 2 min) before forming seals for patch-clamp recording. In these paired experiments, the mean number of IP₃R per patch was unaffected by IP₃-pre-treatment (Supplementary Table 1), confirming that IP₃ neither inactivated IP₃R nor affected the area of membrane trapped beneath the patch. But the distributions of IP₃R were very different before and after IP₃ treatment (Fig. 3a). In naïve nuclei IP₃R were randomly distributed (Fig. 3b), but their distribution after IP₃-pre-treatment differed significantly from the Poisson distribution (p<0.05): many patches had no IP₃R, single IP₃R were under-represented, and several patches had unusually large numbers of IP₃R (Fig. 3c). This clustering of IP₃R fully reversed within 8-10 min of removing IP₃ (Fig. 3a, d). P_o of lone IP₃R from naïve nuclei (0.44 ± 0.05, n = 6) was indistinguishable from P_o of the only lone IP₃R caught within a patch after IP₃ pre-treatment (0.41). P_o for each IP₃R within a cluster was also indistinguishable for recordings from naïve (0.24 ± 0.01, n = 18) and IP₃-pre-treated nuclei (0.25 ± 0.01, n = 18). Furthermore, there was no decrease in P_o during recordings that outlasted the IP₃ pre-treatment (Supplementary Fig. 3). We conclude that clustering, rather than IP₃ per se, decreases P_o.

Figure 3. Reversible clustering of IP₃R by IP₃.

a, Numbers of IP₃R detected in patches from naive nuclei (n = 63), after pre-treatment with bath-applied IP₃ (10 μM, ~2 min; n = 88), or the latter after recovery for 8-10 min without IP₃ (n = 40). b-d, Observed and predicted numbers of IP₃R/patch. e, Effects of IP₃ on IP₃R clustering and gating. Clustering is reported by P_o/P_lone for patches with 2 or 3 IP₃R, and gating by NP_o for patches with 2 IP₃R (EC₅₀ = 2.02 ± 0.20 μM). f, τ_o for patches with 2 or 3 IP₃R measured from the duration of single channel openings (blue line, τ_single) or calculated from the duration of openings to the N^th level (red line, τ_calculated = N.τ_o,N). These are compared with τ_o for lone IP₃R (τ_lone). Typical trace is from a patch with 2 IP₃R. g, IP₃ drives IP₃R into small clusters consistent with arrays (grey) formed by IP₃R at high density¹⁹. Within a cluster, each IP₃R opens independently, but closes more rapidly than a lone IP₃R. h, Typical recording from a patch containing 4 IP₃R with IP₃ released from caged IP₃ in PS by flash photolysis (electrical noise caused by the flash is shown). i, From records similar to h (Supplementary Fig. 8), P_o (from NP_o/N) and τ_o were measured during each 0.5 s interval after the flash (1.5 s for first interval). The ratio (multi-IP₃R patch/lone IP₃R) is shown for both τ_o and P_o. Results (means ± SEM) are from 4 (single) and 7 (multiple, with 2-4 IP₃R/patch) patches.

The decrease in P_o as IP₃R cluster is identical whether clustering is evoked by application of IP₃ to an isolated patch (Fig. 2e, h) or the entire nucleus (Fig. 3e). Both reduce P_o to ~54% that of lone IP₃R. The latter condition better replicates the situation in vivo, confirming that results with isolated patches (Figs 1, 2) faithfully report the behaviour of IP₃R roaming freely within the nuclear envelope. The effect of cluster size on P_o indicates that pairing of IP₃R is sufficient to cause the maximal decrease in P_o. Additional IP₃R can join a cluster, and their activity is attenuated, but IP₃R within larger clusters are no more inhibited than pairs of IP₃R (Figs 2g, h, Supplementary Table 2). IP₃R associate with actin⁴ and microtubules¹⁷, but neither is required for clustering-evoked changes in P_o (Supplementary Fig. 4).

To examine the effects of clustering on IP₃R gating, we compared mean open time (τ_o, Supplementary Information) of lone IP₃R with τ_o for single channel openings from patches with several (N) IP₃R (blue line in Fig. 3f). These τ_o should be similar if lone and grouped IP₃R behave identically. For multi-IP₃R patches, we also measured the duration of events in which all IP₃R were simultaneously open (τ_o,N, red line in Fig. 3f), and from that calculated τ_o for individual, independently gated IP₃R (=Nτ_o,N). Both analyses gave the same result: τ_o for IP₃R within a cluster was reduced to 47% of that for lone IP₃R (Fig. 3f). A similar analysis of closed states confirmed that neither was affected by clustering (Supplementary Fig. 5, Supplementary Table 3). IP₃-evoked clustering almost doubles the rate of channel closure (1/τ_o) and this is alone sufficient (Supplementary Fig. 6, Supplementary Table 4) to account for the decreased P_o of clustered IP₃R (Fig. 2g). Clustered IP₃R open for half as long as lone IP₃R (5.4 vs 11.9 ms), and pairing of IP₃R is enough to cause the full effect (Fig. 2g). Other regulators of IP₃R usually influence τ_c and so rates of channel opening⁴. The difference is important because τ_o will affect the time course of the initial Ca²⁺ release within elementary events⁷ and thereby Ca²⁺-mediated interplay between clustered IP₃R. This is confirmed by simulations of intracellular Ca²⁺ spikes, where the ~50% decrease in τ_o of clustered IP₃R causes the frequency of Ca²⁺ spiking to decrease by 4-fold (Supplementary Fig. 7).

Within a patch, cluster size is limited to the number of IP₃R fortuitously caught beneath the patch-pipette, but for nuclei pre-treated with bath-applied IP₃ the clusters are larger (Fig. 3c). This demonstrates that a maximal concentration of IP₃ causes >93% of IP₃R to cluster (85/91 IP₃R from 88 nuclei pre-treated with IP₃) and the average cluster contains 4.25 ± 0.38 IP₃R (Methods). Inhibition of IP₃R within a cluster is not caused by feedback inhibition⁴ from Ca²⁺ passing through neighbouring IP₃R. Both BS and PS have the same [Ca²⁺] and are buffered with BAPTA, the inhibition occurs at positive (Fig. 2) and negative holding potentials (Supplementary Discussion), and clustered IP₃R open independently (Fig. 2f). Because permeating ions cannot regulate neighbouring IP₃R under our recording conditions, inhibition must be mediated by contacts between IP₃R. From this, we estimate that the average separation of IP₃R falls from ~1 μm to ~20 nm after clustering, and that clusters are ~2 μm apart (Supplementary Discussion). These spacings concur with confocal measurements suggesting that a Ca²⁺ puff originates from a cluster ~50 nm wide and that clusters are ~3 μm apart¹⁸. When expressed at high densities, IP₃R¹⁹ and ryanodine receptors²⁰ form arrays with each tetrameric receptor contacting four others. We speculate that IP₃-evoked clusters (of 4.25 ± 0.38 IP₃R) exploit similar contacts and so, with single IP₃R, form the fundamental units of Ca²⁺ signalling (Fig. 3g).

IP₃-evoked clustering is complete within seconds of stimulation with a maximal concentration of IP₃ (Supplementary Fig. 3). To resolve the time course, we used photolysis of caged IP₃ rapidly to increase the IP₃ concentration bathing IP₃R trapped beneath the patch-pipette. IP₃R were initially quiescent and then rapidly activated when IP₃ was photo-released (Fig. 3h). Irrespective of the number of IP₃R caught within a patch, τ_o was initially similar for all IP₃R (~10 ms). It then remained stable for many minutes for lone IP₃R (11.4 ± 0.5 ms), but τ_o fell within 2.5 s to 5.8 ± 0.3 ms for patches containing more than one IP₃R (Fig. 3i, Supplementary Fig. 8). Using τ_o to report IP₃R clustering suggests that clustering is complete within 2.5 s of IP₃ addition. A similar analysis of P_o suggests a half-time for clustering of ~1.5-2 s (Fig. 3i). Our evidence that clustering does not require the cytoskeleton and measurements of IP₃R3 mobility^21,22 suggest that diffusion alone may be sufficient to allow IP₃R3 clustering within a few seconds (Supplementary Discussion).

We can define the IP₃ sensitivity of clustering by measuring the extent to which P_o of each IP₃R within a multi-IP₃R patch (P_o = NP_o/N, Supplementary Abbreviations) falls below P_o of an identically stimulated lone IP₃R (P_lone). This demonstrates that IP₃R clustering (EC₅₀ < 300 nM) is ~10-times more sensitive to IP₃ than channel opening (EC₅₀ = 2.02 μM, Fig. 3e). Steady-state exposure to low IP₃ concentrations that evoke Ca²⁺ puffs^5,7 would, by assembling IP₃R clusters, allow both generation of puffs and loss of Ca²⁺ blips²³.

Clustering moves IP₃R (~1 μm apart) from being insulated from their neighbours by Ca²⁺-buffering to domains (~20 nm apart) where they will instantly experience high local [Ca²⁺] whenever a neighbour opens²⁴ (Supplementary Fig. 7). Hitherto (Figs 1​-3), we prevented such interactions by using K⁺ as charge-carrier and recording at a free [Ca²⁺] (200 nM) that mimics a resting cell. Subsequent experiments include 1 μM free [Ca²⁺] with IP₃ in PS to simulate the [Ca²⁺] near open IP₃R. For simplicity we use K⁺ as charge-carrier. With 1 μM [Ca²⁺] in PS, IP₃R activity was increased: P_o for lone IP₃R almost doubled, as τ_c decreased (Fig. 4a)⁴. Neither the number of IP₃R/patch (1.12 ± 0.24) nor their random distribution (Fig. 4b) was affected by Ca²⁺, but the interplay between IP₃R was altered. Whereas clustering reduced the overall activity of IP₃R (NP_o) at resting [Ca²⁺] (Fig. 2e), the inhibition was reversed by increased [Ca²⁺], such that the collective activity of a pair of IP₃R (NP_o) was the same as that predicted from the summed activity of two lone IP₃R (Fig. 4c). This did not result from disaggregation of clusters because at increased [Ca²⁺], IP₃R no longer opened independently. In patches with two IP₃R (open-channel noise prevented analysis of larger clusters), open probabilities did not fit the binomial distribution (Fig. 4e): double open and closed events were over-represented (Supplementary Fig. 9). Furthermore, there were many examples of IP₃R opening and closing directly to and from states with both IP₃R open (Fig. 4d). For paired IP₃R, the double openings were prolonged by 50% (Fig. 4f), but 47% less frequent than expected (Fig. 4g). The overall increase in P_o for double openings was therefore small (12%) and counteracted by a 39% decrease in the probability of only one IP₃R being open and a 116% increase in the probability of both being closed (Fig. 4e). Clustered IP₃R exposed to increased [Ca²⁺] do not therefore behave independently. Their gating is coupled^13,14: they are more likely to open and close together, and their simultaneous openings are prolonged (Supplementary Fig. 9). Coupled gating is not caused by local increases in cytosolic [Ca²⁺], and must instead result from physical coupling of IP₃R. Under physiological conditions, clustered IP₃R are more likely to experience increased [Ca²⁺] (because their neighbours may release it), and they are also tuned to respond most to it. By suppressing IP₃R activity at resting [Ca²⁺], clustering increases the impact of a subsequent local increase in [Ca²⁺] (Supplementary Fig. 7). Within a cluster, increased Ca²⁺ increases P_o (as it does for lone IP₃R), but it also reverses the inhibition evoked by clustering and it causes coupled gating. These interactions exaggerate the effect of Ca²⁺ within a cluster (Fig. 4h). We conclude that IP₃ dynamically regulates the assembly and behaviour of Ca²⁺ puff sites. IP₃ rapidly drives IP₃R into small clusters, wherein their IP₃ and Ca²⁺ sensitivities are re-tuned to exaggerate Ca²⁺-mediated recruitment of IP₃R and allow hierarchical recruitment of Ca²⁺ release events (Fig. 4h​, Supplementary Fig. 7)^5,7.

Figure 4. Clustering retunes Ca²⁺ regulation of IP₃R.

a-e, Patches were stimulated with PS containing 10 μM IP₃ and (unless otherwise stated) 1 μM Ca²⁺. a, Typical recording and summary data (n = 5-6) from lone IP₃R show that increasing Ca²⁺ increases P_o by reducing τ_c. b, Observed and expected numbers of IP₃R/patch. c, Observed and predicted NP_o for patches containing 1 or 2 IP₃R and stimulated with 10 μM IP₃ in PS containing 200 nM or 1 μM Ca²⁺ (n = 5-6). d, Typical recording from a patch with 2 IP₃R, enlarged (red) to highlight transitions directly between closed (C) and double open (O2) states. e, Observed and predicted P_o for closed (C) and single (O1) or double openings (O2) for patches with 2 IP₃R (n = 6, Supplementary equations 4, 5). f, Observed and expected durations of events when both IP₃R are simultaneously open (τ_o,2) or closed (τ_c,2) for patches with 2 IP₃R (n = 6, Supplementary equations 6, 7). g, Observed and predicted numbers of transitions to each of the 3 states in a patch with 2 IP₃R (n = 6)²⁶. h, At resting [Ca²⁺], IP₃ drives IP₃R into small clusters wherein IP₃R gate independently, but with reduced P_o and IP₃ sensitivity. Ca²⁺ reverses the inhibition imposed by clustering, openings within a cluster are more synchronized, and simultaneous openings are prolonged. Clustering primes IP₃R to respond by repressing their activity, and then allowing Ca²⁺ to unleash the coordinated gating of clustered IP₃R (Supplementary Fig. 7).

METHODS SUMMARY

Nuclei from DT40-IP₃R3 cells²⁵ were used for patch-clamp recording from excised patches¹⁰.