The cortico-hippocampal circuit is implicated in the formation, storage, and retrieval of spatial and episodic memories (Lisman, 1999). The dentate gyrus (DG), the first stage in the hippocampal circuitry, receives abundant excitatory projections from the entorhinal cortex via the perforant-path (PP) synapses. Theoretical models of hippocampal function propose that the DG is critically involved in pattern separation and that synaptic transmission and plasticity at PP-granule cell (GC) synapses in the DG is required to remove redundant memory representations (Marr, 1971; McNaughton and Morris, 1987; Treves and Rolls, 1994). In agreement with theoretical predictions, knockout of N-methyl-D-aspartate (NMDA) receptors in GCs impairs long-term potentiation (LTP) and the ability to rapidly form a contextual representation and discriminate it from previous similar memories in a contextual fear conditioning task (McHugh et al., 2007). Moreover, GCs that were activated by contextual fear conditioning, referred to as memory engram cells, present clear signatures of synaptic potentiation such as a larger AMPA-NMDA ratio and a greater density of dendritic spines (Ryan et al., 2015). Thus, knowledge of plasticity at PP-GC synapses is essential for understanding the hippocampal function.

Attempts to understand synaptic plasticity rules at PP-GC synapses date to studies of Bliss and Lomo (1973), who first demonstrated LTP in the hippocampus by high-frequency stimulation of the PP fibers in vivo. More recently, theta-burst high-frequency stimulation (TBS) of the PP has been used to induce LTP at PP-GC synapses (Schmidt-Hieber et al., 2004; McHugh et al., 2007; Ge et al., 2007). However, the underlying mechanisms for the induction of LTP with TBS at these synapses remain unclear. Associative forms of synaptic plasticity depend on a presynaptic activity (e.g. excitatory postsynaptic potential, EPSP) and a postsynaptic signal (e.g. action potential, AP). Classically, a backpropagating AP (bAP) provides the associative postsynaptic signal at the synaptic site for the induction of LTP (Hebb, 1949; Magee and Johnston, 1997; Dan and Poo, 2006; Feldman, 2012). However, axosomatic APs are poorly propagated back into the dendrites of the GCs (Krueppel et al., 2011) and are unlikely to contribute to the induction of LTP. In addition, mature GCs are relatively silent during exploration (Schmidt-Hieber et al., 2014; Diamantaki et al., 2016) and fire with a low number of APs. Thus, spike-timing-dependent plasticity (STDP) that depends on an axosomatic postsynaptic AP will rarely occur under natural conditions (Feldman, 2012). The pronounced attenuation of AP backpropagation, the low occurrence of APs, along with the distinct intrinsic features of mature GCs, such as a hyperpolarized resting membrane potential and reduced excitability (Scharfman and Schwartzkroin, 1990; Mongiat et al., 2009; Pernía-Andrade and Jonas, 2014), cannot explain how distal synaptic PP inputs can be potentiated. Resolving this question has critical implications for understanding both the mechanism and function of memory encoding and retrieval (Ryan et al., 2015).

An alternative postsynaptic signal that may contribute to LTP induction at PP synapses innervating distal dendrites of GCs is a locally generated dendritic spike (Sjöström et al., 2008). Synaptic plasticity at distal synapses may occur in the absence of axosomatic APs via dendritic sodium spikes, which provide an alternative source of postsynaptic depolarization necessary for LTP induction (Golding et al., 2002; Kim et al., 2015). In GCs, linear somatic responses can be seen when glutamate is applied in the dendrites (Krueppel et al., 2011). However, without directly accessing the electrical properties of the distal GC dendrites, it is still not known whether synaptic stimulation can generate local dendritic spikes in GCs.

To address this question, we performed subcellular patch-clamp recordings on the thin dendrites of GCs. We found that inhibition of dendritic Na⁺ channels prevented LTP induction via TBS at PP-GC synapses. Because dendritic Na⁺ channels could generate local spikes that were independent of axosomatic AP generation and were not affected by voltage attenuation, we suggest that Na⁺ spikes in the dendrites provide the postsynaptic signal necessary for the induction of LTP at PP-GC synapses.

We examined GCs with an input resistance (R_in) <200 MΩ (R_in = 108.4 ± 2.6 MΩ; resting membrane potential: –81.3 ± 0.2 mV; see ‘Materials and methods’) which corresponds to the mature GC population in the acute hippocampal slices from rats (Schmidt-Hieber et al., 2004).

TBS-induced LTP at the PP-GC synapses does not require postsynaptic bAPs

We first induced long-term potentiation (LTP) in GCs by theta-burst stimulation (TBS) of the PP synapses in the outer third of the molecular layer (Figure 1A). To activate the PP synapses, the tip of a stimulation electrode was placed in close proximity to the dendrite (<50 µm) using fluorescence microscopy (‘Materials and methods’). When bursts of postsynaptic APs were evoked during TBS, TBS-induced LTP caused an increase in the amplitude of the excitatory postsynaptic potentials (EPSPs; from 6.72 ± 0.71 mV to 12.69 ± 1.44 mV, n = 6, p<0.05; Figure 1B). However, axosomatic APs in GCs are substantially attenuated with distance from the soma (Krueppel et al., 2011), and therefore are not likely to provide the necessary depolarization at distal dendrites. To test whether axosomatic APs are required for the induction of LTP at these distal synapses, we either locally applied tetrodotoxin (TTX) to the proximal axon, soma, and proximal dendrites in a subset of experiments (6 of 13 experiments) or adjusted the stimulus intensity to prevent axosomatic AP initiation and backpropagation occur during TBS. The absence of axosomatic spikes did not block LTP induction, indicating that bAPs are not critical for this form of LTP (from 6.96 ± 0.40 mV to 9.96 ± 0.86 mV, n = 13, p<0.01; Figure 1C). Intriguingly, activation of the PP synapses during TBS often produced voltage responses with a fast depolarizing phase similar to somatic events observed during dendritic spike generation in other types of neurons (Figure 1C,D; Golding and Spruston, 1998; Golding et al., 2002; Jarsky et al., 2005; Losonczy and Magee, 2006; Kim et al., 2012). Consistent with previous results (Golding and Spruston, 1998; Golding et al., 2002; Losonczy and Magee, 2006; Kim et al., 2015), various shapes of putative dendritic spikes were observed during TBS (Figure 1D). These voltage responses were first distinguished from EPSPs without dendritic spikes by the spikelet waveform (Figure 1D) and then identified as putative dendritic spikes when the peak of the temporal derivative (dV/dt) of somatic voltage responses was larger than 2.5 mV/ms. Typically, weak (2.5 mV/ms ≤ dV/dt < 10 mV/ms) and strong (dV/dt > 10 mV/ms) putative dendritic spikes were identified (weak dendritic spikes: orange, 4.2 ± 0.4 mV/ms, n = 17; strong dendritic spikes: red, 14.5 ± 0.8 mV/ms, n = 11; p<0.0001; Figure 1D). The dV/dt of the EPSPs without any putative dendritic spikes was 1.3 ± 0.01 mV/ms (n = 2498), which was significantly smaller than that of weak putative dendritic spikes (p<0.0001; Figure 1D). These putative dendritic spikes were accompanied by a sustained plateau potential (Figure 1D; see also Figure 2B). Remarkably, the presence of these weak and strong putative dendritic spikes during TBS was correlated with LTP induction (in the presence of putative dendritic spikes, 173.2 ± 12.1%, n = 7; in the absence of putative dendritic spikes 114.1 ± 12.8%, n = 6; p<0.005; Figure 1E,G). Indeed, we found a strong correlation between the number of putative dendritic spikes observed during TBS and the magnitude of LTP (r = 0.77; p<0.005; n = 13; Figure 1F). These results suggest that dendritic spikes but not axosomatic APs contribute to LTP induction at PP-GC synapses.

Figure 1

Download asset Open asset

Figure 1—source data 1

Source data for Figure 1.

https://doi.org/10.7554/eLife.35269.003

Download elife-35269-fig1-data1-v2.xlsx

Figure 2 with 1 supplement see all

Download asset Open asset

Figure 2—source data 1

Source data for Figure 2.

https://doi.org/10.7554/eLife.35269.006

Download elife-35269-fig2-data1-v2.xlsx

To further test whether axosomatic spikes contribute to the induction of LTP at the PP-GC synapse, we used associative pairing protocols (Figure 2). Presynaptic activity was either followed (pre-postsynaptic sequence) or preceded (post-presynaptic sequence) at a 10 ms interval by single or AP bursts (2 APs at 100 Hz). Both protocols failed to induce significant changes in EPSP amplitude (pre-postsynaptic sequence, control: 4.97 ± 0.38 mV, after induction: 5.83 ± 0.86 mV, n = 15, p=0.23; post-presynaptic sequence, control: 4.78 ± 0.54 mV, after induction: 4.91 ± 0.99 mV, n = 9, p=0.54; Figure 2A,B), suggesting that APs of axonal origin are not important for potentiation. We further performed similar experiments at the medial PP–GC synapses by stimulating axons in the middle third of the molecular layer (Figure 2—figure supplement 1A). Pairing EPSPs and APs in both pre-postsynaptic and post-presynaptic sequences did not induce LTP of EPSPs (pre-postsynaptic sequence, control: 4.86 ± 0.40 mV, after induction: 5.09 ± 0.99 mV, n = 7, p=0.9015; post-presynaptic sequence, control: 3.86 ± 0.26 mV, after induction: 4.27 ± 0.57 mV, n = 6, p=0.56; Figure 2—figure supplement 1B–E), consistent with previous studies that reported no synaptic potentiation after similar pairing protocols (Yang and Dani, 2014; Lopez-Rojas et al., 2016). Together, these results imply that dendritic spikes, rather than axosomatic APs, are the essential signal for LTP induction at distal GC synapses.

LTP by TBS at PP–GC synapses requires NMDARs and Na⁺ channels

We next examined the receptors involved in LTP expression at the PP-GC synapses. Bath application of the NMDAR antagonist DL-AP5 (50 µM) abolished LTP (control: 8.89 ± 0.95 mV; after induction: 9.88 ± 1.21 mV, n = 9, p=0.50; Figure 3A,E,F). While sustained plateau potentials during LTP induction were inhibited by DL-AP5 in the external solution, fast rising events remained unchanged (Figure 3B). Thus, we next tested whether dendritic voltage-gated Na⁺ channels are involved in LTP. To block dendritic Na⁺ channels without affecting synaptic transmission, we included the intracellular sodium channel blocker QX-314 (5 mM) in the whole-cell patch pipette. Stimulus intensity was set to elicit baseline EPSP of similar or slightly larger amplitude than that required to produce putative dendritic spikes during TBS (Figure 3C,D). This manipulation abolished both putative dendritic spikes (0 of 7 cells; Figure 3D) and TBS-induced LTP (control: 9.24 ± 1.21 mV, after induction: 8.77 ± 1.67 mV, n = 7, p=0.69; Figure 3C–E). These results indicate that TBS-induced LTP in GCs depends on the activation of NMDARs and voltage-gated Na⁺ channels on the postsynaptic dendritic membrane, reinforcing the idea that dendritic Na⁺ spikes may play a pivotal role in TBS induction at distal synapses.

Figure 3

Download asset Open asset

Figure 3—source data 1

Source data for Figure 3.

https://doi.org/10.7554/eLife.35269.009

Download elife-35269-fig3-data1-v2.xlsx

Backpropagation of axosomatic APs in the dendrites of GCs

To directly dissect the dendritic mechanism that determines the induction rules of synaptic plasticity in GCs, we employed subcellular patch-clamp techniques to analyze AP backpropagation and initiation in GC dendrites (Figure 4A). First, we characterized backpropagation in GCs with somatic current injection evoking trains of APs at the soma while simultaneously recording in the dendrite (Figure 4B). The AP always appeared first in the somatic recording and then in the dendrites (Figure 4C). Similar to the previous report (Krueppel et al., 2011), the peak amplitude of the bAPs attenuated as a function of distance from the soma (Figure 4D,E). At dendritic distances beyond 150 µm from the soma, the peak amplitude of bAPs was reduced to ~36% of the amplitude of somatic APs (soma: 93.6 ± 1.7 mV; dendrite: 33.7 ± 3.1 mV; n = 10, p<0.005; Figure 4D). The attenuation per dendritic length is much more pronounced in GCs compared to the other neocortical (Stuart et al., 1997; Nevian et al., 2007) and hippocampal pyramidal neurons (Spruston et al., 1995; Kim et al., 2012). It is interesting to note that the extent of bAP attenuation when normalized to the overall length of GC dendrites (mean dendritic length: 278 ± 7.4 μm; n = 11) is quite similar to layer five pyramidal neuron dendrites (cf. Supplementary Figure 3B in Nevian et al. (2007); see also Krueppel et al. (2011); Figure 4F), suggesting that the failure of LTP induction during pairing protocols at distal GC synapses (Figure 2) might be explained by the voltage attenuation of bAPs, as was reported in layer five pyramidal neurons (Letzkus et al., 2006; Sjöström et al., 2008); Feldman; 2012). We also estimated the conduction velocity of bAPs by analyzing the AP latencies measured at the half-maximal amplitude of the rising phase. APs were initiated axonally and propagated back into the dendrites with a velocity of 226 µm/ms (Figure 4G; n = 56; corresponding to 0.2–0.3 m/s, Senzai and Buzsáki, 2017). This conduction velocity is slower than those measured in the apical and basal dendrites of layer five pyramidal neurons (apical dendrites, 508 µm/ms; basal dendrites, 341 µm/ms; Nevian et al. (2007) and Stuart et al., 1997) and is similar or lower than the velocity estimated in the apical dendrite of other hippocampal principal neurons (Spruston et al., 1995; Kim et al., 2012). In summary, these experiments show that bAPs in GCs propagate into the dendrites with substantial voltage attenuation and moderate conduction velocity.

Figure 4

Download asset Open asset

Figure 4—source data 1

Source data for Figure 4.

https://doi.org/10.7554/eLife.35269.011

Download elife-35269-fig4-data1-v2.xlsx

Ionic mechanisms of AP backpropagation

To determine the ionic mechanisms underlying the strong voltage attenuation and the moderate conduction velocity of bAPs, we assayed the somatic and dendritic distribution of voltage-gated Na⁺ and K⁺ currents in outside-out patches isolated at various locations using pipettes of similar open tip resistance (soma: 17.9 ± 0.5 MΩ, n = 24; dendrite: 19.4 ± 0.5 MΩ, n = 36; Figure 5). Depolarizing voltage pulses from –120 mV to 0 mV evoked TTX-sensitive inward Na⁺ currents in the majority of outside-out patches excised from both soma and dendrites (Figure 5A and Figure 5—figure supplements 1 and 2). Pooled data demonstrated that the current amplitude of Na⁺ channels is uniformly distributed over the dendritic membrane (Figure 5D). On average, the peak amplitude of Na⁺ currents was not significantly different between somatic and dendritic patches (soma: –6.81 ± 1.36 pA, n = 19; proximal dendrite (within 100 µm, PD): –5.02 ± 1.55 pA, n = 8; distal dendrite (beyond 100 µm, DD): –7.17 ± 1.36 pA, n = 21; p>0.99 for all cases, Kruskal-Wallis test with Dunn’s multiple comparison; Figure 5D). Surprisingly, very large transient K⁺ currents that appeared to activate and inactivate rapidly in response to voltage pulses from –120 mV to +70 mV was found in patches obtained from the dendrites (Figure 5B). These inactivating components of K⁺ channel activity were reduced by the A-type K⁺ channel blocker 4-aminopyridine (4-AP, Figure 5B and Figure 5—figure supplements 1 and 2). Plotting the current amplitude along the dendrite demonstrated that GC dendrites show significantly larger A-type K⁺ currents than the soma (soma: 30.4 ± 4.2 pA, n = 21; PD: 61.5 ± 8.7 pA, n = 7; DD: 102.3 ± 15.3 pA, n = 21; soma vs. PD, p=0.07; soma vs. DD, p<0.0001; PD vs. DD, p=0.96, Kruskal-Wallis test with Dunn’s multiple comparisons; Figure 5E). Consistent with this finding, we found that bath application of 5 mM 4-AP in simultaneous somatic and dendritic voltage recordings results in a significant enhancement of the peak amplitude and duration of bAPs in the dendrites, indicating that dendritic A-type K⁺ current (I_A) contributes to voltage attenuation of bAPs in GCs (Figure 5—figure supplement 3). Finally, the delayed rectifier K⁺ current components, which was largely inhibited by 20 mM extracellular tetraethylammonium (TEA, Figure 5C and Figure 5—figure supplement 1), showed a low and uniform expression over the entire somatodendritic axis (soma: 19.6 ± 2.2 pA, n = 21; PD: 18.3 ± 3.8 pA, n = 7; DD: 21.7 ± 3.8 pA, n = 22; p>0.99 for all cases, Kruskal-Wallis test with Dunn’s multiple comparisons; Figure 5C,F). Taken together, these voltage-clamp data reveal that a markedly high density of K⁺ channels and a uniform density of Na⁺ channels are present in the dendrites of GCs, which are likely to account for both the strong dendritic AP attenuation and the moderate AP propagation velocity.

Figure 5 with 3 supplements see all

Download asset Open asset

Figure 5—source data 1

Source data for Figure 5.

https://doi.org/10.7554/eLife.35269.016

Download elife-35269-fig5-data1-v2.xlsx

A burst of PP stimuli can evoke dendritically initiated Na⁺ spikes in GCs

Voltage-gated Na⁺ channels may contribute to the generation of dendritically initiated local spikes in regions where the dendrites are very small (Holmes, 1989). Indeed, depolarizing dendrites with brief current pulse injection evoked local spikes in distal dendrites of GCs (55 of 63 recordings; Figure 6A–C). In the proximal domain (within 100 µm from the soma), dendritic spikes were not detected and depolarizing dendrites resulted in an axosomatic AP that propagated back to the dendritic recording site (Figure 6D). At dendritic recording locations 70 to 150 µm from the soma, dendritic spikes were associated with axosomatic spikes. For distances larger than 150 µm from the soma (approximately corresponding to the outer molecular layer), current injections robustly initiated isolated dendritic spikes (Figure 6D). Pharmacological analysis revealed that dendritic spikes were resistant to 200 μM CdCl₂ and 50 μM NiCl₂ but were inhibited by 0.5 μM TTX, indicating that they were mediated by dendritic voltage-gated Na⁺ channels rather than by Ca²⁺ channels (TTX: 11.3 ± 3.3%, n = 3; CdCl₂: 93.9 ± 7.8%, n = 4; NiCl₂: 97.9 ± 10.2%, n = 4; Figure 6—figure supplement 1). Because our results indicate that GC dendrites contain a high density of transient, A-type K⁺ channels (Figure 5B,E), we further explored how these channels influence dendritic spike initiation. We combined dual somatodendritic recordings and focal application of 4-AP (10 mM) to the dendritic patch (Figure 6—figure supplement 2A). Local block of I_A significantly decreased the current threshold for dendritic spike initiation, suggesting that dendritic I_A affect the generation of dendritic spikes (Figure 6—figure supplement 2B,C).

Figure 6 with 2 supplements see all

Download asset Open asset

We next examined whether Na⁺ spikes in GC dendrites can be evoked by synaptic stimulation of PP inputs. Extracellular synaptic stimulation was combined with simultaneous double patch-clamp recordings from the soma and dendrites (Figure 7A). Dendritic spikes could be triggered by stimulating the PP inputs with TBS protocol (Figure 7A,B). Repeated trials of a high-frequency PP stimulation using the same stimulus intensity produced variable dendritic responses with subthreshold EPSP, weak dendritic spikes, or strong dendritic spikes that appeared either in isolation or associated with axosomatic APs (Figure 7B). When dendritic spikes were present, the dV/dt of the corresponding somatic voltages showed variable peak amplitudes that were similar to those of the putative dendritic spikes during TBS induction shown in Figure 1D (EPSPs with a dendritic spike: dV/dt = 7.1 ± 1.1 mV/ms, n = 26; p=0.58 compared to EPSPs with putative dendritic spikes: dV/dt = 8.2 ± 1.0 mV/ms, n = 28 in Figure 1D; Figure 7C,D). Overall, dendritic recordings reveals that a high-frequency burst activation of PP synapses can produce dendritic Na⁺ spikes that are manifested as rapidly rising voltage responses at the soma (Figure 1D).

Figure 7

Download asset Open asset

Figure 7—source data 1

Source data for Figure 7.

https://doi.org/10.7554/eLife.35269.026

Download elife-35269-fig7-data1-v2.xlsx

Dendritically initiated Na⁺ spikes are required for TBS-induced LTP at PP–GC synapses

Finally, we examined whether dendritic Na⁺ spikes are necessary for TBS-induced LTP at PP synapses. To this end, we applied a low concentration of TTX (10 nM; Kim et al., 2015) to block dendritic Na⁺ spikes without interfering synaptic transmission (Figure 8A,B). Inhibition of dendritic Na⁺ spikes prevented TBS-induced LTP (control: 6.58 ± 0.51 mV; after induction: 7.04 ± 0.88 mV, n = 8; p=0.64; Figure 8C–E), without affecting the baseline EPSP amplitude (control: 5.21 ± 0.78 mV; after TTX: 5.06 ± 0.68 mV, n = 11; p=0.83; Figure 8—figure supplement 1). All together, these results suggest that dendritically generated local Na⁺ spikes in response to TBS of the PP are necessary for LTP induction in GCs.

Figure 8 with 1 supplement see all

Download asset Open asset

Figure 8—source data 1

Source data for Figure 8.

https://doi.org/10.7554/eLife.35269.029

Download elife-35269-fig8-data1-v2.xlsx

In summary, the present study demonstrates several major findings. First, our results show that conventional STDP protocols (Dan and Poo, 2006; Feldman, 2012) do not trigger synaptic potentiation at PP-GC synapses. Second, a physiologically relevant TBS paradigm can efficiently induce LTP in GCs without axosomatic APs. Finally, direct dendritic recordings revealed that LTP induction requires dendritic spikes. To our knowledge, these studies show the first direct evidence for dendritic spike generation in GCs and the role of these spikes during induction of LTP in the dentate gyrus network. Whether our findings also hold for GCs at early stages of development remains to be determined.

Considerable evidence has shown that the electrical properties of neurons at the cellular and subcellular levels are brain region-specific and cell-type-specific and endow unique rules and characteristics on circuit function (Sjöström et al., 2008; Stuart and Spruston, 2015). Hyperpolarized resting potential, strong dendritic voltage attenuation, and low occurrence of APs are the known electrophysiological characteristics of mature GCs (Scharfman and Schwartzkroin, 1990; Schmidt-Hieber et al., 2004; Mongiat et al., 2009; Krueppel et al., 2011; Pernía-Andrade and Jonas, 2014), which are distinct from other types of hippocampal principal neurons (Spruston et al., 1995; Kim et al., 2012; Stuart and Spruston, 2015). While backpropagated APs are an important associative signal for triggering plasticity at the synaptic site via the voltage-dependent relief of Mg²⁺ block of NMDARs (Magee and Johnston, 1997; Dan and Poo, 2006; Feldman, 2012; Mishra et al., 2016), the above features of mature GCs are highly unfavorable for LTP induction at distal synaptic contacts. Accordingly, we found that GCs do not show LTP at PP-GC synapses during standard STDP (Figure 2 and Figure 2—figure supplement 1), which is consistent with recent reports (Yang and Dani, 2014; Lopez-Rojas et al., 2016).

However, several studies have demonstrated EPSP-AP pairing protocol-induced synaptic potentiation at the same synapses (Levy and Steward, 1983; Lin et al., 2006). Given that Levy and Steward (1983) and Lin et al. (2006) employed in vivo and in vitro field recording configurations, respectively, the discrepant results could be attributed to GCs at different stages of maturation as immature GCs exhibit a lower threshold for LTP induction (Schmidt-Hieber et al., 2004; Ge et al., 2007) or under different recording circumstances that were exposed to various neuromodulators. For example, Yang and Dani (2014) reported that the pairing protocol that showed no synaptic potentiation could induce reliable LTP at PP-GC synapses after D1-type dopamine receptor activation by which dendritic A-type K⁺ currents are suppressed. Because I_A is known to limit the backpropagation of APs (Hoffman et al., 1997), suppression of I_A can boost AP backpropagation, allowing sufficient dendritic depolarization for LTP induction at distal synapses. Our observations of a high density of dendritic I_A and their effects on AP backpropagation in GCs directly support the finding of Yang and Dani (2014). Therefore, under physiological conditions, long-lasting dendritic depolarization during theta oscillation (Buzsáki, 2002) or activation of neuromodulatory systems (Hamilton et al., 2010; Yang and Dani, 2014) may cause inactivation of dendritic I_A and trigger pairing-induced LTP (Lin et al., 2006; Brunner and Szabadics, 2016).

Although the presence of A-type K⁺ channels in GC dendrites had been shown in earlier immunocytochemical studies (Birnbaum et al., 2004; Monaghan et al., 2008; Menegola et al., 2008), it has been proposed that dendritic A-type K⁺ channels have no significant impact on bAP-induced Ca²⁺ signals (Krueppel et al., 2011). Krueppel et al. (2011) results appear to be inconsistent with our present data, which show the strong expression of functional A-type K⁺ channels in GC dendrites. In contrast to that study, we directly tested the effect of 4-AP in simultaneous soma-dendrite recordings by using both global and local application methods. Our results show that the effect of local puff application of 4-AP to the dendritic recording site is much smaller than that of global application. As Krueppel et al. (2011) used local application halfway between the soma and the Ca²⁺ imaging site, the impact of 4-AP on bAPs in their study could be underestimated.

We further found that high-frequency stimulation of PP synapses is efficient in eliciting dendritic spikes required for LTP induction. As reported in a recent in vivo study, mature GCs are exposed to abundant functional glutamatergic inputs from the entorhinal cortex during theta rhythm (Pernía-Andrade and Jonas, 2014; Schmidt-Hieber et al., 2014). Moreover, several lines of evidence demonstrated that mature GCs are highly innervated by PP synaptic connections while receiving a powerful perisomatic inhibition (Dieni et al., 2013; 2016; Temprana et al., 2015). Under these in vivo network conditions, strong dendritic excitation by high-frequency PP inputs and strong perisomatic inhibition may promote amplification of dendritic responses without axosomatic APs. Therefore, dendritic spikes are likely physiologically relevant signals for the induction of LTP. Support for the idea of cooperative LTP at PP-GC synapses has also come from McNaughton et al. (1978), who demonstrated that high-frequency stimulation of PP synapses could induce synaptic enhancement in the absence of GC discharges.

This specific high-frequency stimulation-dependent synaptic potentiation in GCs presumably stems from distinct functional and geometrical features of GC dendrites. A moderate density of dendritic Na⁺ channels together with higher input impedance of the distal dendrites (Hama et al., 1989; Schmidt-Hieber et al., 2007; Holmes, 1989) suffices for initiating spikes locally in distal dendrites that have small capacitive load but not for supporting active backpropagation of axosomatic APs. Although GCs have relatively short dendrites, a high K⁺ to Na⁺ current ratio in these thin-caliber dendrites imposes a strong distance-dependent attenuation of axosomatic APs, leading to a lack of pairing-induced LTP at distal synapses. Thus, Na⁺ spikes in the dendrites contribute the postsynaptic depolarization necessary for the induction of associative plasticity (Golding et al., 2002; Kim et al., 2015). However, it should be noted that dendritic Na⁺ spikes were accompanied by NMDAR-mediated plateau potentials and therefore these two dendritic events could act in concert to trigger LTP in GCs (Schiller et al., 2000).

Consequently, our findings suggest that in the absence of axonal firing (Alme et al., 2010; Diamantaki et al., 2016), dendritic spikes would allow a silent GC to participate in the storage of memories via LTP induction. It would further permit a functional separation between a storage phase mediated by Na⁺ channels in GC dendrites, and a recall phase (O'Neill et al., 2008) that effectively activates the CA3 neurons (Vyleta et al., 2016) via Na⁺ channels in GC axons.

1. 1. Alme CB
   2. Buzzetti RA
   3. Marrone DF
   4. Leutgeb JK
   5. Chawla MK
   6. Schaner MJ
   7. Bohanick JD
   8. Khoboko T
   9. Leutgeb S
   10. Moser EI
   11. Moser MB
   12. McNaughton BL
   13. Barnes CA

   (2010) Hippocampal granule cells opt for early retirement

   Hippocampus 20:1109–1123.

   https://doi.org/10.1002/hipo.20810

   • PubMed
   • Google Scholar
2. 1. Birnbaum SG
   2. Varga AW
   3. Yuan LL
   4. Anderson AE
   5. Sweatt JD
   6. Schrader LA

   (2004) Structure and function of Kv4-family transient potassium channels

   Physiological Reviews 84:803–833.

   https://doi.org/10.1152/physrev.00039.2003

   • PubMed
   • Google Scholar
3. 1. Bliss TV
   2. Lomo T

   (1973) Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path

   The Journal of Physiology 232:331–356.

   https://doi.org/10.1113/jphysiol.1973.sp010273

   • PubMed
   • Google Scholar
4. 1. Brunner J
   2. Szabadics J

   (2016) Analogue modulation of back-propagating action potentials enables dendritic hybrid signalling

   Nature Communications 7:13033.

   https://doi.org/10.1038/ncomms13033

   • PubMed
   • Google Scholar
5. 1. Buzsáki G

   (2002) Theta oscillations in the hippocampus

   Neuron 33:325–340.

   https://doi.org/10.1016/S0896-6273(02)00586-X

   • PubMed
   • Google Scholar
6. 1. Dan Y
   2. Poo MM

   (2006) Spike timing-dependent plasticity: from synapse to perception

   Physiological Reviews 86:1033–1048.

   https://doi.org/10.1152/physrev.00030.2005

   • PubMed
   • Google Scholar
7. 1. Diamantaki M
   2. Frey M
   3. Berens P
   4. Preston-Ferrer P
   5. Burgalossi A

   (2016) Sparse activity of identified dentate granule cells during spatial exploration

   eLife 5:e20252.

   https://doi.org/10.7554/eLife.20252

   • PubMed
   • Google Scholar
8. 1. Dieni CV
   2. Nietz AK
   3. Panichi R
   4. Wadiche JI
   5. Overstreet-Wadiche L

   (2013) Distinct determinants of sparse activation during granule cell maturation

   Journal of Neuroscience 33:19131–19142.

   https://doi.org/10.1523/JNEUROSCI.2289-13.2013

   • PubMed
   • Google Scholar
9. 1. Dieni CV
   2. Panichi R
   3. Aimone JB
   4. Kuo CT
   5. Wadiche JI
   6. Overstreet-Wadiche L

   (2016) Low excitatory innervation balances high intrinsic excitability of immature dentate neurons

   Nature Communications 7:11313.

   https://doi.org/10.1038/ncomms11313

   • PubMed
   • Google Scholar
10. 1. Feldman DE

    (2012) The spike-timing dependence of plasticity

    Neuron 75:556–571.

    https://doi.org/10.1016/j.neuron.2012.08.001

    • PubMed
    • Google Scholar
11. 1. Ge S
    2. Yang CH
    3. Hsu KS
    4. Ming GL
    5. Song H

    (2007) A critical period for enhanced synaptic plasticity in newly generated neurons of the adult brain

    Neuron 54:559–566.

    https://doi.org/10.1016/j.neuron.2007.05.002

    • PubMed
    • Google Scholar
12. 1. Golding NL
    2. Spruston N

    (1998) Dendritic sodium spikes are variable triggers of axonal action potentials in hippocampal CA1 pyramidal neurons

    Neuron 21:1189–1200.

    https://doi.org/10.1016/S0896-6273(00)80635-2

    • PubMed
    • Google Scholar
13. 1. Golding NL
    2. Staff NP
    3. Spruston N

    (2002) Dendritic spikes as a mechanism for cooperative long-term potentiation

    Nature 418:326–331.

    https://doi.org/10.1038/nature00854

    • PubMed
    • Google Scholar
14. 1. Guzman SJ
    2. Schlögl A
    3. Schmidt-Hieber C

    (2014) Stimfit: quantifying electrophysiological data with Python

    Frontiers in Neuroinformatics 8:16.

    https://doi.org/10.3389/fninf.2014.00016

    • PubMed
    • Google Scholar
15. 1. Hama K
    2. Arii T
    3. Kosaka T

    (1989) Three-dimensional morphometrical study of dendritic spines of the granule cell in the rat dentate gyrus with HVEM stereo images

    Journal of Electron Microscopy Technique 12:80–87.

    https://doi.org/10.1002/jemt.1060120203

    • PubMed
    • Google Scholar
16. 1. Hamilton TJ
    2. Wheatley BM
    3. Sinclair DB
    4. Bachmann M
    5. Larkum ME
    6. Colmers WF

    (2010) Dopamine modulates synaptic plasticity in dendrites of rat and human dentate granule cells

    PNAS 107:18185–18190.

    https://doi.org/10.1073/pnas.1011558107

    • PubMed
    • Google Scholar
17. Book

    1. Hebb DO

    (1949)

    The Organization of Behavior

    John Wiley & Sons.

    • Google Scholar
18. 1. Hoffman DA
    2. Magee JC
    3. Colbert CM
    4. Johnston D

    (1997) K⁺ channel regulation of signal propagation in dendrites of hippocampal pyramidal neurons

    Nature 387:869–875.

    https://doi.org/10.1038/43119

    • PubMed
    • Google Scholar
19. 1. Holmes WR

    (1989) The role of dendritic diameters in maximizing the effectiveness of synaptic inputs

    Brain Research 478:127–137.

    https://doi.org/10.1016/0006-8993(89)91484-4

    • PubMed
    • Google Scholar
20. 1. Jarsky T
    2. Roxin A
    3. Kath WL
    4. Spruston N

    (2005) Conditional dendritic spike propagation following distal synaptic activation of hippocampal CA1 pyramidal neurons

    Nature Neuroscience 8:1667–1676.

    https://doi.org/10.1038/nn1599

    • PubMed
    • Google Scholar
21. 1. Kim S
    2. Guzman SJ
    3. Hu H
    4. Jonas P

    (2012) Active dendrites support efficient initiation of dendritic spikes in hippocampal CA3 pyramidal neurons

    Nature Neuroscience 15:600–606.

    https://doi.org/10.1038/nn.3060

    • PubMed
    • Google Scholar
22. 1. Kim S

    (2014) Action potential modulation in CA1 pyramidal neuron axons facilitates OLM interneuron activation in recurrent inhibitory microcircuits of rat hippocampus

    PLoS One 9:e113124.

    https://doi.org/10.1371/journal.pone.0113124

    • PubMed
    • Google Scholar
23. 1. Kim Y
    2. Hsu CL
    3. Cembrowski MS
    4. Mensh BD
    5. Spruston N

    (2015) Dendritic sodium spikes are required for long-term potentiation at distal synapses on hippocampal pyramidal neurons

    eLife 4:e06414.

    https://doi.org/10.7554/eLife.06414

    • PubMed
    • Google Scholar
24. 1. Krueppel R
    2. Remy S
    3. Beck H

    (2011) Dendritic integration in hippocampal dentate granule cells

    Neuron 71:512–528.

    https://doi.org/10.1016/j.neuron.2011.05.043

    • PubMed
    • Google Scholar
25. 1. Letzkus JJ
    2. Kampa BM
    3. Stuart GJ

    (2006) Learning rules for spike timing-dependent plasticity depend on dendritic synapse location

    Journal of Neuroscience 26:10420–10429.

    https://doi.org/10.1523/JNEUROSCI.2650-06.2006

    • PubMed
    • Google Scholar
26. 1. Levy WB
    2. Steward O

    (1983) Temporal contiguity requirements for long-term associative potentiation/depression in the hippocampus

    Neuroscience 8:791–797.

    https://doi.org/10.1016/0306-4522(83)90010-6

    • PubMed
    • Google Scholar
27. 1. Lin YW
    2. Yang HW
    3. Wang HJ
    4. Gong CL
    5. Chiu TH
    6. Min MY

    (2006) Spike-timing-dependent plasticity at resting and conditioned lateral perforant path synapses on granule cells in the dentate gyrus: different roles of N-methyl-D-aspartate and group I metabotropic glutamate receptors

    European Journal of Neuroscience 23:2362–2374.

    https://doi.org/10.1111/j.1460-9568.2006.04730.x

    • PubMed
    • Google Scholar
28. 1. Lisman JE

    (1999) Relating hippocampal circuitry to function: recall of memory sequences by reciprocal dentate-CA3 interactions

    Neuron 22:233–242.

    https://doi.org/10.1016/S0896-6273(00)81085-5

    • PubMed
    • Google Scholar
29. 1. Lopez-Rojas J
    2. Heine M
    3. Kreutz MR

    (2016) Plasticity of intrinsic excitability in mature granule cells of the dentate gyrus

    Scientific Reports 6:21615.

    https://doi.org/10.1038/srep21615

    • PubMed
    • Google Scholar
30. 1. Losonczy A
    2. Magee JC

    (2006) Integrative properties of radial oblique dendrites in hippocampal CA1 pyramidal neurons

    Neuron 50:291–307.

    https://doi.org/10.1016/j.neuron.2006.03.016

    • PubMed
    • Google Scholar
31. 1. Magee JC
    2. Johnston D

    (1997) A synaptically controlled, associative signal for Hebbian plasticity in hippocampal neurons

    Science 275:209–213.

    https://doi.org/10.1126/science.275.5297.209

    • PubMed
    • Google Scholar
32. 1. Marr D

    (1971) Simple memory: a theory for archicortex

    Philosophical Transactions of the Royal Society B: Biological Sciences 262:23–81.

    https://doi.org/10.1098/rstb.1971.0078

    • PubMed
    • Google Scholar
33. 1. McHugh TJ
    2. Jones MW
    3. Quinn JJ
    4. Balthasar N
    5. Coppari R
    6. Elmquist JK
    7. Lowell BB
    8. Fanselow MS
    9. Wilson MA
    10. Tonegawa S

    (2007) Dentate gyrus NMDA receptors mediate rapid pattern separation in the hippocampal network

    Science 317:94–99.

    https://doi.org/10.1126/science.1140263

    • PubMed
    • Google Scholar
34. 1. McNaughton BL
    2. Douglas RM
    3. Goddard GV

    (1978) Synaptic enhancement in fascia dentata: cooperativity among coactive afferents

    Brain Research 157:277–293.

    https://doi.org/10.1016/0006-8993(78)90030-6

    • PubMed
    • Google Scholar
35. 1. McNaughton BL
    2. Morris RGM

    (1987) Hippocampal synaptic enhancement and information storage within a distributed memory system

    Trends in Neurosciences 10:408–415.

    https://doi.org/10.1016/0166-2236(87)90011-7

    • Google Scholar
36. 1. Menegola M
    2. Misonou H
    3. Vacher H
    4. Trimmer JS

    (2008) Dendritic A-type potassium channel subunit expression in CA1 hippocampal interneurons

    Neuroscience 154:953–964.

    https://doi.org/10.1016/j.neuroscience.2008.04.022

    • PubMed
    • Google Scholar
37. 1. Mishra RK
    2. Kim S
    3. Guzman SJ
    4. Jonas P

    (2016) Symmetric spike timing-dependent plasticity at CA3-CA3 synapses optimizes storage and recall in autoassociative networks

    Nature Communications 7:11552.

    https://doi.org/10.1038/ncomms11552

    • PubMed
    • Google Scholar
38. 1. Monaghan MM
    2. Menegola M
    3. Vacher H
    4. Rhodes KJ
    5. Trimmer JS

    (2008) Altered expression and localization of hippocampal A-type potassium channel subunits in the pilocarpine-induced model of temporal lobe epilepsy

    Neuroscience 156:550–562.

    https://doi.org/10.1016/j.neuroscience.2008.07.057

    • PubMed
    • Google Scholar
39. 1. Mongiat LA
    2. Espósito MS
    3. Lombardi G
    4. Schinder AF

    (2009) Reliable activation of immature neurons in the adult hippocampus

    PLoS One 4:e5320.

    https://doi.org/10.1371/journal.pone.0005320

    • PubMed
    • Google Scholar
40. 1. Nevian T
    2. Larkum ME
    3. Polsky A
    4. Schiller J

    (2007) Properties of basal dendrites of layer 5 pyramidal neurons: a direct patch-clamp recording study

    Nature Neuroscience 10:206–214.

    https://doi.org/10.1038/nn1826

    • PubMed
    • Google Scholar
41. 1. O'Neill J
    2. Senior TJ
    3. Allen K
    4. Huxter JR
    5. Csicsvari J

    (2008) Reactivation of experience-dependent cell assembly patterns in the hippocampus

    Nature Neuroscience 11:209–215.

    https://doi.org/10.1038/nn2037

    • PubMed
    • Google Scholar
42. 1. Pernía-Andrade AJ
    2. Jonas P

    (2014) Theta-gamma-modulated synaptic currents in hippocampal granule cells in vivo define a mechanism for network oscillations

    Neuron 81:140–152.

    https://doi.org/10.1016/j.neuron.2013.09.046

    • PubMed
    • Google Scholar
43. 1. Ryan TJ
    2. Roy DS
    3. Pignatelli M
    4. Arons A
    5. Tonegawa S

    (2015) Memory. engram cells retain memory under retrograde amnesia

    Science 348:1007–1013.

    https://doi.org/10.1126/science.aaa5542

    • PubMed
    • Google Scholar
44. 1. Scharfman HE
    2. Schwartzkroin PA

    (1990) Responses of cells of the rat fascia dentata to prolonged stimulation of the perforant path: sensitivity of hilar cells and changes in granule cell excitability

    Neuroscience 35:491–504.

    https://doi.org/10.1016/0306-4522(90)90324-W

    • PubMed
    • Google Scholar
45. 1. Schiller J
    2. Major G
    3. Koester HJ
    4. Schiller Y

    (2000) NMDA spikes in basal dendrites of cortical pyramidal neurons

    Nature 404:285–289.

    https://doi.org/10.1038/35005094

    • PubMed
    • Google Scholar
46. 1. Schmidt-Hieber C
    2. Jonas P
    3. Bischofberger J

    (2004) Enhanced synaptic plasticity in newly generated granule cells of the adult hippocampus

    Nature 429:184–187.

    https://doi.org/10.1038/nature02553

    • PubMed
    • Google Scholar
47. 1. Schmidt-Hieber C
    2. Jonas P
    3. Bischofberger J

    (2007) Subthreshold dendritic signal processing and coincidence detection in dentate gyrus granule cells

    Journal of Neuroscience 27:8430–8441.

    https://doi.org/10.1523/JNEUROSCI.1787-07.2007

    • PubMed
    • Google Scholar
48. 1. Schmidt-Hieber C
    2. Wei H
    3. Häusser M

    (2014)

    Synaptic mechanisms of sparse activity in hippocampal granule cells during mouse navigation

    Society for Neuroscience Abstract.

    • Google Scholar
49. 1. Senzai Y
    2. Buzsáki G

    (2017) Physiological properties and behavioral correlates of hippocampal granule cells and mossy cells

    Neuron 93:691–704.

    https://doi.org/10.1016/j.neuron.2016.12.011

    • PubMed
    • Google Scholar
50. 1. Sjöström PJ
    2. Rancz EA
    3. Roth A
    4. Häusser M

    (2008) Dendritic excitability and synaptic plasticity

    Physiological Reviews 88:769–840.

    https://doi.org/10.1152/physrev.00016.2007

    • PubMed
    • Google Scholar
51. 1. Sjöström PJ
    2. Turrigiano GG
    3. Nelson SB

    (2001) Rate, timing, and cooperativity jointly determine cortical synaptic plasticity

    Neuron 32:1149–1164.

    https://doi.org/10.1016/S0896-6273(01)00542-6

    • PubMed
    • Google Scholar
52. 1. Spruston N
    2. Schiller Y
    3. Stuart G
    4. Sakmann B

    (1995) Activity-dependent action potential invasion and calcium influx into hippocampal CA1 dendrites

    Science 268:297–300.

    https://doi.org/10.1126/science.7716524

    • PubMed
    • Google Scholar
53. 1. Stuart G
    2. Schiller J
    3. Sakmann B

    (1997) Action potential initiation and propagation in rat neocortical pyramidal neurons

    The Journal of Physiology 505:617–632.

    https://doi.org/10.1111/j.1469-7793.1997.617ba.x

    • PubMed
    • Google Scholar
54. 1. Stuart GJ
    2. Häusser M

    (2001) Dendritic coincidence detection of EPSPs and action potentials

    Nature Neuroscience 4:63–71.

    https://doi.org/10.1038/82910

    • PubMed
    • Google Scholar
55. 1. Stuart GJ
    2. Spruston N

    (2015) Dendritic integration: 60 years of progress

    Nature Neuroscience 18:1713–1721.

    https://doi.org/10.1038/nn.4157

    • PubMed
    • Google Scholar
56. 1. Temprana SG
    2. Mongiat LA
    3. Yang SM
    4. Trinchero MF
    5. Alvarez DD
    6. Kropff E
    7. Giacomini D
    8. Beltramone N
    9. Lanuza GM
    10. Schinder AF

    (2015) Delayed coupling to feedback inhibition during a critical period for the integration of adult-born granule cells

    Neuron 85:116–130.

    https://doi.org/10.1016/j.neuron.2014.11.023

    • PubMed
    • Google Scholar
57. 1. Treves A
    2. Rolls ET

    (1994) Computational analysis of the role of the hippocampus in memory

    Hippocampus 4:374–391.

    https://doi.org/10.1002/hipo.450040319

    • PubMed
    • Google Scholar
58. 1. Vyleta NP
    2. Borges-Merjane C
    3. Jonas P

    (2016) Plasticity-dependent, full detonation at hippocampal mossy fiber-CA3 pyramidal neuron synapses

    eLife 5:e17977.

    https://doi.org/10.7554/eLife.17977

    • PubMed
    • Google Scholar
59. 1. Yang K
    2. Dani JA

    (2014) Dopamine D1 and D5 receptors modulate spike timing-dependent plasticity at medial perforant path to dentate granule cell synapses

    Journal of Neuroscience 34:15888–15897.

    https://doi.org/10.1523/JNEUROSCI.2400-14.2014

    • PubMed
    • Google Scholar

Author details

1. Sooyun Kim

   1. Department of Physiology, Seoul National University College of Medicine, Seoul, Korea
   2. Neuroscience Research Institute, Seoul National University College of Medicine, Seoul, Korea

   Contribution

   Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   sooyun.kim@snu.ac.kr

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2035-3247

2. Yoonsub Kim

   Department of Physiology, Seoul National University College of Medicine, Seoul, Korea

   Contribution

   Formal analysis, Investigation, Writing—review and editing

   Competing interests

   No competing interests declared

3. Suk-Ho Lee

   1. Department of Physiology, Seoul National University College of Medicine, Seoul, Korea
   2. Neuroscience Research Institute, Seoul National University College of Medicine, Seoul, Korea

   Contribution

   Resources, Writing—review and editing

   Competing interests

   No competing interests declared

4. Won-Kyung Ho

   1. Department of Physiology, Seoul National University College of Medicine, Seoul, Korea
   2. Neuroscience Research Institute, Seoul National University College of Medicine, Seoul, Korea

   Contribution

   Resources, Supervision, Funding acquisition, Writing—review and editing

   For correspondence

   wonkyung@snu.ac.kr

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-1568-1710

• 4,201

  views

• 664

  downloads

• 16

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.