Spine Ca2+ signaling in spike-timing-dependent plasticity - PubMed (original) (raw)

Comparative Study

Spine Ca2+ signaling in spike-timing-dependent plasticity

Thomas Nevian et al. J Neurosci. 2006.

Abstract

Calcium is a second messenger, which can trigger the modification of synaptic efficacy. We investigated the question of whether a differential rise in postsynaptic Ca2+ ([Ca2+]i) alone is sufficient to account for the induction of long-term potentiation (LTP) and long-term depression (LTD) of EPSPs in the basal dendrites of layer 2/3 pyramidal neurons of the somatosensory cortex. Volume-averaged [Ca2+]i transients were measured in spines of the basal dendritic arbor for spike-timing-dependent plasticity induction protocols. The rise in [Ca2+]i was uncorrelated to the direction of the change in synaptic efficacy, because several pairing protocols evoked similar spine [Ca2+]i transients but resulted in either LTP or LTD. The sequence dependence of near-coincident presynaptic and postsynaptic activity on the direction of changes in synaptic strength suggested that LTP and LTD were induced by two processes, which were controlled separately by postsynaptic [Ca2+]i levels. Activation of voltage-dependent Ca2+ channels before metabotropic glutamate receptors (mGluRs) resulted in the phospholipase C-dependent (PLC-dependent) synthesis of endocannabinoids, which acted as a retrograde messenger to induce LTD. LTP required a large [Ca2+]i transient evoked by NMDA receptor activation. Blocking mGluRs abolished the induction of LTD and uncovered the Ca2+-dependent induction of LTP. We conclude that the volume-averaged peak elevation of [Ca2+]i in spines of layer 2/3 pyramids determines the magnitude of long-term changes in synaptic efficacy. The direction of the change is controlled, however, via a mGluR-coupled signaling cascade. mGluRs act in conjunction with PLC as sequence-sensitive coincidence detectors when postsynaptic precede presynaptic action potentials to induce LTD. Thus presumably two different Ca2+ sensors in spines control the induction of spike-timing-dependent synaptic plasticity.

PubMed Disclaimer

Figures

Figure 1.

Timing-dependent induction of LTP and LTD by pairing an EPSP and a short burst of APs. A, Illustration of the recording configuration for the synaptic plasticity experiments. The extracellular stimulation pipette was placed close to the basal dendrites, ∼50 μm from the soma of the L2/3 pyramidal neuron. B, The induction protocol for LTP is depicted on the left. An EPSP evoked by extracellular stimulation (pre) was paired with a short burst of three APs at 50 Hz elicited by current injections into the postsynaptic cell (post). The first AP followed the onset of the EPSP by Δ_t_ = 10 ms. The time interval Δ_t_ is defined as the time between the EPSP and the AP closest in time to the EPSP. For bursts that follow the EPSP, Δ_t_ is equivalent to Δ_t_′, which is defined as the time interval between the EPSP and the first AP in the burst. Pairing was repeated 60 times, every 10 s. To the right, the EPSP amplitude, membrane potential, and input resistance are plotted over time during the experiment. The dashed line indicates the average EPSP amplitude before the pairing. The pairing protocol depicted to the left resulted in potentiation of the EPSP amplitude. The EPSPs averaged over the times indicated by the red bars are shown on the bottom. The average EPSP amplitude increased from 2 to 3.5 mV at 20–40 min after the pairing period. C, The protocol for the induction of LTD is depicted to the left. A short burst of three APs at 50 Hz was paired with a following EPSP. Here the first AP in the burst preceded the onset of the EPSP by Δ_t_′ = −50 ms, which is equivalent to the definition that the last AP in the burst preceded the onset of the EPSP by Δ_t_ = −10 ms. To the right, the EPSP amplitude, membrane potential, and input resistance are plotted over time for the experiment in which the burst preceded the EPSP. This pairing protocol resulted in depression of the EPSP amplitude. The voltage recordings averaged over the times indicated by the red bars are shown on the bottom. The average EPSP amplitude decreased from 2 to 1 mV at 20–40 min after the pairing period.

Figure 2.

Burst-timing-dependent plasticity curve. A, Pooled and normalized EPSP amplitudes for different time intervals Δ_t_′ between the onset of the EPSP and the first AP in the burst. The burst consisted of three APs (50 Hz). Insets show schematically the timing between the EPSP and the three APs. B, Summary of the change in the EPSP amplitude for different EPSP and AP burst-timing intervals Δ_t_′. If the AP burst preceded the EPSP by more than −90 ms or followed by more than +50 ms, no change in EPSP amplitude was found (p > 0.5; n = 4–6). A timing interval of Δ_t_′ = −50 ms resulted in depression of the EPSP amplitude by a factor of 0.67 ± 0.06 (mean ± SEM; **p < 0.01; _n_ = 9), whereas a timing interval of Δ_t_′ = +10 ms resulted in potentiation by a factor of 2.08 ± 0.25 (**_p_ < 0.01; _n_ = 11). If the EPSP was evoked within the AP burst, either no change (Δ_t_′ = −30 ms; _p_ > 0.5; n = 8) or potentiation by a factor of 1.5 ± 0.3 (Δ_t_′ = −10 ms; *p < 0.05; n = 6) was observed. The horizontal dashed line represents no change in EPSP amplitude, and the vertical dashed line represents the onset of the EPSP.

Figure 3.

LTP and LTD dependence on the number of APs and AP frequency in the burst. A, Pooled and normalized EPSP amplitudes for different numbers of APs in the burst and a time interval between the AP closest to the EPSP and the onset of the EPSP of Δ_t_ = −10 ms and Δ_t_ = +10 ms. Insets depict the pairing protocol. Pairing an EPSP with a single AP at Δ_t_ = +10 ms had no effect on EPSP amplitude. The sequence of AP and EPSP at Δ_t_ = −10 ms resulted in depression. An EPSP paired with a burst of two APs (50 Hz) at Δ_t_ = +10 ms resulted in potentiation. Two APs (50 Hz) preceding the EPSP resulted in depression of the EPSP amplitude. B, LTP induction at Δ_t_ = +10 ms (open circles) depended on the number of APs that followed the EPSP, whereas LTD induction (open boxes) was correlated only weakly with the number of APs preceding the EPSP by Δ_t_ = −10 ms. The open diamond to the left corresponds to EPSP-only stimulation during the pairing period. Significant differences, *p < 0.05 and **p < 0.01. C, Pooled and normalized EPSP amplitudes for different AP frequencies in the burst and a time interval between the AP closest to the EPSP and the onset of the EPSP of Δ_t_ = −10 ms and Δ_t_ = +10 ms. Pairing an EPSP with a burst of three APs at 20 Hz did not result in significant potentiation. A burst of three APs at 20 Hz preceding an EPSP resulted in depression. Pairing an EPSP with a burst of three APs at 100 Hz that followed at Δ_t_ = +10 ms resulted in strong potentiation. A burst of three APs at 100 Hz preceding the EPSP resulted in depression. D, Summary plot of the change in EPSP amplitude versus the AP frequency in a burst of three APs. The induction of LTP is frequency-dependent (open circles), whereas LTD induction did not depend on the burst frequency. The dashed lines indicate no change in EPSP amplitude.

Figure 4.

LTP and LTD are equally sensitive to fast and slow Ca2+ buffers. A, Buffering of postsynaptic Ca2+ with 2 m

EGTA (open circles) or 2 m

BAPTA (filled circles) blocked the induction of LTP by the sequence of an EPSP and three APs (50 Hz) at Δ_t_ = +10 ms. A lower concentration of EGTA or BAPTA of 0.5 m

blocked 40% of LTP. B, Titration curve for different concentrations of EGTA (open circles) and BAPTA (filled circles) and their effects on the induction of LTP. Data were normalized to the change in EPSP amplitude for no buffer added. The slow Ca2+ buffer EGTA blocked LTP with the same concentration dependence as the fast Ca2+ buffer BAPTA. A sigmoidal fit yielded a half-concentration of 0.6 and 0.5 m

for EGTA (gray line) and BAPTA (black line), respectively. C, A concentration of 1 m

EGTA (open squares) or BAPTA (filled squares) was sufficient to blocked the induction of LTD by the sequence of three APs (50 Hz) and an EPSP at Δ_t_ = −10 ms. A lower concentration of EGTA or BAPTA of 0.25 m

had no significant effect on the induction of LTD. D, Titration curve for different concentrations of EGTA (open boxes) and BAPTA (filled boxes) and their effect on the induction of LTD. Data were normalized to the change in EPSP amplitude for no buffer added. A sigmoidal fit yielded a half-concentration of 0.39 and 0.36 m

for EGTA (gray line) and BAPTA (black line), respectively. The dashed lines indicate no change in EPSP amplitude.

Figure 5.

[Ca2+]i transients in single spines evoked by pairing protocols. A, Two-photon fluorescence image of a L2/3 pyramidal neuron overlaid with the simultaneously acquired IR image. The region indicated by the dashed box is shown on an expanded scale in B. B, Fluorescence image of a spine responding with a [Ca2+]i transient to synaptic stimulation. The dashed line indicates the position of the line scan. An example line scan during a synaptically evoked EPSP paired with three APs (50 Hz) at Δ_t_ = +10 ms is shown in the bottom image. The green fluorescence of the Ca2+-sensitive indicator Oregon Green BAPTA-6F (500 μ

) is overlaid with the red fluorescence from the Ca2+-insensitive dye Alexa 594 (50 μ

). Stimulation begins 50 ms after the beginning of the recording. The bar at the top of the line scan indicates the region of interest over which fluorescence was averaged for each time point. C, Somatic voltage recordings (top traces) and the corresponding [Ca2+]i transients (bottom traces) for an EPSP, three APs, three APs (50 Hz) and an EPSP at Δ_t_′ = −50 ms, and an EPSP and three APs (50 Hz) at Δ_t_′ = +10 ms. The solid gray lines are exponential fits to the decay phase of the [Ca2+]i transient yielding the peak amplitude (Δ_G_/R)max. D, Peak amplitude (top graph) and nonlinearity factor (bottom graph) plotted for the time intervals Δ_t_′. The dashed line indicates linear summation of the [Ca2+]i transients. Significant differences, *p < 0.05 and **_p_ < 0.01. **_E_**, Peak amplitude (top graph) increases linearly with the number of APs in the burst for APs only (triangles), an EPSP and APs (50 Hz) at Δ_t_ = +10 ms (circles), and the sequence of APs (50 Hz) and an EPSP at Δ_t_ = −10 ms (boxes). The diamond to the left indicates the peak amplitude for the EPSP-evoked [Ca2+]i transient. The nonlinearity factor (bottom graph) does not depend on the number of APs in the burst, but it depends on the relative timing between the EPSP and APs. APs that follow the EPSP by Δ_t_ = +10 ms result in supralinear summation of the [Ca2+]i transients, whereas APs preceding the EPSP by Δ_t_ = −10 ms result in linear summation. **_F_**, Peak amplitude (top graph) increases linearly with the frequency in a burst of three APs for APs only (triangles), an EPSP and APs at Δ_t_ = +10 ms (circles), and the sequence of APs and an EPSP at Δ_t_ = −10 ms (boxes). The nonlinearity factor (bottom graph) indicates linear summation of the [Ca2+]i transients for APs preceding the EPSP, independent of AP frequency and supralinear summation of [Ca2+]i transients for APs that follow the EPSP for frequencies >20 Hz.

Figure 6.

The contribution of NMDARs to Ca2+ signaling and synaptic plasticity. A, Fluorescence image of a spine responding with a [Ca2+]i transient to synaptic stimulation. The dashed line indicates the position of the line scan. B, [Ca2+]i transients for the sequence of an EPSP and three APs (50 Hz) for Δ_t_ = +10 ms (top trace) and Δ_t_ = −10 ms (bottom trace) for control (gray trace) and after the bath application of the NMDAR blocker

-APV (50 μ

; black trace). C, Average peak amplitudes for pairing an EPSP with three APs (50 Hz) at Δ_t_ = −10 ms and Δ_t_ = +10 ms for control (gray circles) and after the bath application of

-APV (black circles).

-APV significantly reduced the peak amplitude for both time intervals (*p < 0.05; n = 3). D, Normalized EPSP amplitude over time for the pairing protocol of an EPSP and three APs (50 Hz) at Δ_t_ = +10 ms in the presence of

-APV (bath application). LTP induction was abolished for Δ_t_ = +10 ms. Dashed lines indicate no change in EPSP amplitude. E, Bath application of

-APV also abolished the induction of LTD for Δ_t_ = −10 ms. F, Intracellular application of the open NMDAR channel blocker MK-801 (1 m

) blocked the induction of LTP. G, In contrast, intracellular application of MK-801 had no effect on the induction of LTD. H, Either bath application of

-APV or intracellular application of MK-801 blocked the induction of LTP by the pairing protocol of an EPSP and three APs (50 Hz) at Δ_t_ = +10 ms (p > 0.3; n = 3–6). I, Bath application of

-APV blocked the induction of LTD (p > 0.5; n = 6) by the pairing protocol of three APs (50 Hz) and an EPSP at Δ_t_ = −10 ms, whereas intracellular application of MK-801 had no effect on LTD (*p < 0.05; n = 7).

Figure 7.

The contribution of VDCCs to Ca2+ signaling and synaptic plasticity. A, Fluorescence image of a spine responding with a [Ca2+]i transient to synaptic stimulation. The dashed line indicates the position of the line scan. B, [Ca2+]i transients for the sequence of an EPSP and three APs (50 Hz) for Δ_t_ = +10 ms (top trace) and Δ_t_ = −10 ms (bottom trace) for control (gray trace) and after the bath application of the L-VDCC blocker nimodipine (10 μ

; black trace). C, Average peak amplitudes for pairing an EPSP with three APs (50 Hz) at Δ_t_ = −10 ms and Δ_t_ = +10 ms for control (gray circles) and after the bath application of nimodipine (black circles). Nimodipine significantly reduced the peak amplitude for Δ_t_ = −10 ms (*p < 0.05; n = 3) but had no effect on the peak amplitude for Δ_t_ = +10 ms. D, The L-VDCC blocker nimodipine had no effect on the induction of LTP. E, Nimodipine also had no effect on the induction of LTD. F, Blocking T-VDCCs with Ni2+ (50 μ

) abolished the induction of LTD for the pairing protocol of one AP and an EPSP at Δ_t_ = −10 ms. G, In contrast, Ni2+ had no effect on the induction of LTD for the pairing protocol of three APs (50 Hz) and an EPSP at Δ_t_ = −10 ms. H, The sequence of three APs (50 Hz) and an EPSP for Δ_t_ = −10 ms evoked [Ca2+]i transients in the spine, indicated in the fluorescence image by the dashed line under control conditions (gray trace) and a corresponding smaller [Ca2+]i transient after the bath application of Ni2+ (black trace). I, Summary of LTD induction with pharmacological block of VDCCs. Ni2+ blocked the induction of LTD if one AP preceded the EPSP during pairing (p > 0.5; n = 3). Neither Ni2+ nor nimodipine alone had an effect on LTD induction if a burst of three APs (50 Hz) preceded the EPSP (*p < 0.05; _n_ = 4–8). Bath application of nimodipine and Ni2+ together blocked LTD (_p_ > 0.5; n = 3).

Figure 8.

The peak [Ca2+]i amplitude does not predict LTP or LTD. Summary plot of the average change in EPSP amplitude versus the average peak [Ca2+]i amplitude expressed as (Δ_G_/R)max. The change in EPSP amplitude is not a unique function of the peak [Ca2+]i amplitude. Similar peak levels of [Ca2+]i can result in either LTP or LTD (data points in shaded area). The clustering of induction protocols in which the APs either precede the EPSP (open boxes) or follow the EPSP (open circles) indicate that the inductions of LTP and LTD are separate processes. Fitting sigmoid functions to the data sets (solid blue line, APs following; solid green line, APs preceding) shows that LTP and LTD both depend on the peak [Ca2+]i amplitude. The protocols in which APs preceded and followed the EPSP (red circles) fall close to the curve for the Ca2+ dependence of LTP.

Figure 9.

Effect of the mGluR blocker MCPG on [Ca2+]i transients and synaptic plasticity. A, Two-photon image of an active synaptic spine. The dashed line indicates the position of the line scan. B, [Ca2+]i transients for the sequences of an EPSP and three APs (50 Hz) at Δ_t_ = +10 ms (top traces), three APs (50 Hz) and an EPSP at Δ_t_ = −10 ms (middle traces), and three APs (100 Hz) and an EPSP at Δ_t_ = −10 ms (bottom traces) for control (gray traces) and after bath application of the mGluR blocker MCPG (500 μ

; black traces). The fluorescence traces in the presence of MCPG do not differ from control. C, Summary plot of the peak [Ca2+]i amplitudes for control versus peak [Ca2+]i amplitude after the bath application of MCPG. The points for the stimulation protocols depicted in B fall close to the unity line (dashed line), indicating no significant effect of MCPG on the peak amplitude of the [Ca2+]i transients. D, Summary of normalized EPSP amplitudes for pairing an EPSP with three APs (50 Hz) at Δ_t_ = +10 ms under control conditions (gray trace) and for experiments in MCPG (black trace). LTP induction in the presence of MCPG was not different from control. Dashed lines indicate no change in EPSP amplitude. E, Summary of normalized EPSP amplitudes for pairing three APs (50 Hz) with an EPSP at Δ_t_ = −10 ms under control conditions (gray trace) and for experiments in MCPG (black trace). LTD was abolished in this case. F, Summary of normalized EPSP amplitudes for pairing three APs (100 Hz) with an EPSP at Δ_t_ = −10 ms under control conditions (gray trace) and for experiments in MCPG (black trace). The LTD under control conditions was reversed to LTP by MCPG. G, Summary plot of change in EPSP amplitude for the pairing protocols depicted in B as a function of peak [Ca2+]i amplitude under control conditions (gray) and in the presence of MCPG (black). In the presence of MCPG no LTD is induced, and the change in EPSP amplitude becomes a monotonically increasing function of the peak [Ca2+]i amplitude.

Figure 10.

Signaling pathway for the induction of LTD. A, Summary of normalized EPSP amplitudes for pairing three APs (50 Hz) with an EPSP at Δ_t_ = −10 ms in the presence of the PLC blocker U73122 (5 μ

). U73122 blocked the induction of LTD. B, Two-photon image of an active spine. The dashed line indicates the position of the line scan. U73122 had no effect on the [Ca2+]i transients evoked by the sequence of three APs (50 Hz) and an EPSP at Δ_t_ = −10 ms (black trace) as compared with control (gray trace). C, Heparin (400 U/ml), a blocker of Ca2+ release from internal stores mediated by IP3 receptors, had no effect on the induction of LTD. D, Amplitude of averaged peak [Ca2+]i transients measured during the induction of LTD with the protocol of three APs (50 Hz) and an EPSP at Δ_t_ = −10 ms under control conditions. Every sixth transient was recorded during the 10 min (0.1 Hz) induction period. No correlation between the peak amplitude and the number of stimuli was found (linear correlation, _r_2 = 0.2; n = 7). E, Bath application of the CB1 receptor antagonist AM251 (2 μ

) blocked the induction of LTD. F, Summary of the signaling pathway for the induction of LTD. Block of PLC and CB1 receptors abolished the induction of LTD (p > 0.1; n = 4–5), whereas the block of Ca2+ release from internal stores had no effect on the induction of LTD (*p < 0.05; n = 9). Dashed lines represent no change in EPSP amplitude.

Cited by

Competitive processes shape multi-synapse plasticity along dendritic segments.
Chater TE, Eggl MF, Goda Y, Tchumatchenko T. Chater TE, et al. Nat Commun. 2024 Aug 31;15(1):7572. doi: 10.1038/s41467-024-51919-0. Nat Commun. 2024. PMID: 39217140 Free PMC article.
Synaptopodin is required for long-term depression at Schaffer collateral-CA1 synapses.
Inglebert Y, Wu PY, Tourbina-Kolomiets J, Dang CL, McKinney RA. Inglebert Y, et al. Mol Brain. 2024 Apr 2;17(1):17. doi: 10.1186/s13041-024-01089-3. Mol Brain. 2024. PMID: 38566234 Free PMC article.
Ion-concentration gradients induced by synaptic input increase the voltage depolarization in dendritic spines.
Eberhardt F. Eberhardt F. J Comput Neurosci. 2024 Feb;52(1):1-19. doi: 10.1007/s10827-024-00864-4. Epub 2024 Feb 13. J Comput Neurosci. 2024. PMID: 38349479 Free PMC article.
Contiguity in perception: origins in cellular associative computations.
Hansel C. Hansel C. Trends Neurosci. 2024 Mar;47(3):170-180. doi: 10.1016/j.tins.2024.01.001. Epub 2024 Feb 2. Trends Neurosci. 2024. PMID: 38310022 Review.
Synaptotagmin 7 Sculpts Short-Term Plasticity at a High Probability Synapse.
Chiu DN, Carter BC. Chiu DN, et al. J Neurosci. 2024 Feb 28;44(9):e1756232023. doi: 10.1523/JNEUROSCI.1756-23.2023. J Neurosci. 2024. PMID: 38262726 Free PMC article.

References

1. Anwyl R. Metabotropic glutamate receptors: electrophysiological properties and role in plasticity. Brain Res Brain Res Rev. 1999;29:83–120. - PubMed
1. Anwyl R. Induction and expression mechanisms of postsynaptic NMDA receptor-independent homosynaptic long-term depression. Prog Neurobiol. 2006;78:17–37. - PubMed
1. Artola A, Singer W. Long-term depression of excitatory synaptic transmission and its relationship to long-term potentiation. Trends Neurosci. 1993;16:480–487. - PubMed
1. Augustine GJ, Santamaria F, Tanaka K. Local calcium signaling in neurons. Neuron. 2003;40:331–346. - PubMed
1. Barria A, Muller D, Derkach V, Griffith LC, Soderling TR. Regulatory phosphorylation of AMPA-type glutamate receptors by CaM-KII during long-term potentiation. Science. 1997;276:2042–2045. - PubMed

Spine Ca2+ signaling in spike-timing-dependent plasticity - PubMed (original) (raw)