Rapid spine stabilization and synaptic enhancement at the onset of behavioural learning

Abstract

Behavioural learning depends on the brain's capacity to respond to instructive experience and is often enhanced during a juvenile sensitive period. How instructive experience acts on the juvenile brain to trigger behavioural learning remains unknown. In vitro studies show that forms of synaptic strengthening thought to underlie learning are accompanied by an increase in the stability, number and size of dendritic spines, which are the major sites of excitatory synaptic transmission in the vertebrate brain. In vivo imaging studies in sensory cortical regions reveal that these structural features can be affected by disrupting sensory experience and that spine turnover increases during sensitive periods for sensory map formation. These observations support two hypotheses: first, the increased capacity for behavioural learning during a sensitive period is associated with enhanced spine dynamics on sensorimotor neurons important for the learned behaviour; second, instructive experience rapidly stabilizes and strengthens these dynamic spines. Here we report a test of these hypotheses using two-photon in vivo imaging to measure spine dynamics in zebra finches, which learn to sing by imitating a tutor song during a juvenile sensitive period. Spine dynamics were measured in the forebrain nucleus HVC, the proximal site where auditory information merges with an explicit song motor representation, immediately before and after juvenile finches first experienced tutor song. Higher levels of spine turnover before tutoring correlated with a greater capacity for subsequent song imitation. In juveniles with high levels of spine turnover, hearing a tutor song led to the rapid ( approximately 24-h) stabilization, accumulation and enlargement of dendritic spines in HVC. Moreover, in vivo intracellular recordings made immediately before and after the first day of tutoring revealed robust enhancement of synaptic activity in HVC. These findings suggest that behavioural learning results when instructive experience is able to rapidly stabilize and strengthen synapses on sensorimotor neurons important for the control of the learned behaviour.

Figure 1. Tutor song experience affects spine turnover in juvenile zebra finches

a, Top: schematic of the zebra finch song system, experimental protocol and timeline of the experiments. Inset: Nissl stained image of HVC in parasagittal section showing its location on the floor of the lateral telencephalic ventricle, ~100µm below the pial surface. Scale bar, 200µm. Right: low magnification in vivo two-photon image of GFP labelled spinous HVC neurons amidst retrogradely labelled HVC_RA (red) and HVC_X (blue) projection neurons. Scale bar, 20µm. b, Repeated in vivo imaging of dendritic branches from an HVC neuron of a 130d finch over 4 days (yellow box, left image; scale bar, 20 µm on left image and 10 µm on right). c, High magnification view of dendritic segment shown in b (yellow box, scale bar, 2 µm), imaged two hours apart. Arrows point to stable (blue), lost (yellow) and gained (green) spines. d, High magnification view of another dendritic segment of an HVC neuron showing the gain and loss of dendritic spines across a two hour imaging interval (scale bar, 2 µm).

Figure 2. Levels of HVC dendritic spine turnover correlate with song imitation

a, Mean two-hour HVC spine turnover levels are elevated in 45d (P = 0.009) and 60d (P = 0.04), but not 90d, untutored birds (circles) compared to age matched control birds (triangles)(untutored 45d: 180 spines from 6 cells in 4 birds; 45d control: 621 spines from 11 cells in 3 birds; untutored 60d: 1,419 spines from 17 cells in 14 birds; 60d control: 1,579 spines from 23 cells in 16 normally reared birds; untutored 90d: 188 spines from 4 cells in 2 birds; 90d control: 1,544 spines from 23 cells in 11 birds). b, Levels of turnover in HVC prior to tutoring correlate with subsequent song imitation. Scatter plot showing the relationship between levels of HVC dendritic spine turnover (2hr) measured the night prior to the first exposure to a song model, and the total increase in similarity to the tutor song over song development (P = 0.02, r = 0.63, 1,419 spines from 17 cells in 14 birds). Each circle represents a single bird. The eight open circles correspond to birds in which post-tutoring dendritic spine turnover measurements were taken.

Figure 3. Tutoring can trigger rapid stabilization and accumulation of dendritic spines on HVC neurons

a, Tutoring triggers a rapid decrease in the level of HVC dendritic spine turnover in HT (P < 0.01), but not LT (P = 0.3) birds. Levels of HVC dendritic spine turnover measured the night before and the night after the first day of either live (N = 5) or operant (N = 3) tutoring in HT and LT birds (HT birds: P < 0.01, 12.2 ± 0.8% → 7.0 ± 1.2% pre and post-tutoring turnover, respectively, 468 spines from 5 birds; LT birds: P = 0.3, spine turnover = 3.6 ± 0.5% → 3.0 ± 0.2% pre and post-tutoring, respectively, 449 spines from 3 birds). Further, spine turnover levels measured in HVC prior to tutoring correlated positively with the post-tutoring decrease in spine turnover (P = 0.02, r = 0.79, N = 8 birds, not shown). b, The bulk of the decrease in turnover levels (67%) observed during sensorimotor learning occur during the first 48hs following hearing a tutor for the first time. Levels of HVC dendritic spine turnover in a “high turnover” bird measured over 30 days following initial exposure to a tutor. The tutor did not sing during the first two days with the HT juvenile, at which point turnover levels markedly and persistently decreased. c, Example of stable spines (green arrows) that formed during the first day of tutoring in a bird with high pre-tutoring levels of spine turnover (scale bar, 2 µm). d, Tutoring triggers a rapid increase in HVC dendritic spine density in HT (P = 0.03), but not LT (P = 0.5) birds. Percent change in dendritic spine density by the first night following tutoring (post-tutoring density / pre-tutoring density; HT birds: P = 0.03, 14.0 ± 4.4% increase in spine density; LT birds: P = 0.5, −6.4 ± 7.0% increase in spine density; from 4 HT and 3 LT birds, 1,198 spines). Further, spine turnover levels measured in HVC prior to tutoring correlated positively with the post-tutoring increase in spine density (P = 0.04, r = 0.71, N = 8 birds, not shown). e, Mean afternoon entropy variance scores increase by the afternoon of the first day of tutoring in HT birds (blue), but not LT birds (Post-hoc Tukey test, α = 0.05 in 3 of 4 HT birds, data shows mean afternoon EV value for 4 HT birds and 3 LT birds). Error bars denote s.e.m.

Figure 4. Tutoring triggers enlargement of stable dendritic spines in HVC

a, Example of two stable spines that exhibited increased fluorescence intensity following tutoring, indicating an increase in dendritic spine volume. Scale bar, 2 µm. b, The size of stable spines increased in HT but not LT birds by the first night following tutoring, as revealed by measurements of relative integrated fluorescence intensity (81 spines, 8 birds; HT bird: P = 0.001, 1.08 ± 0.14 → 1.39 ± 0.16 (relative fluorescence intensity ± s.e.m.), n = 47 spines, Wilcoxon-Signed rank test for paired samples; LT birds: P = 0.4, 1.68 ± 0.20 → 1.61 ± 0.20, n = 34 spines, Wilcoxon-Signed rank test for paired samples). Moreover, HT birds had smaller stable dendritic spines prior to tutor exposure (P = 0.02; HT spines: 1.08 ± 0.14, n = 47 spines; LT birds: 1.67 ± 0.20, n = 34 spines). c, The first night following tutor exposure, the mean size of stable spines increased by 28% in HT birds, but did not change in LT birds (HT birds, P = 0.001; LT birds, P = 0.4).

Figure 5. Tutoring triggers enhancement of spontaneous synaptic activity in HVC

a, In vivo intracellular membrane potential recordings made from six different HVC neurons in a juvenile bird one night before (left) and one night after (right) initial tutor exposure show that tutoring drives a rapid increase in the amplitude of spontaneous depolarizing synaptic activity. b, (top) Cumulative frequency distribution of spontaneous synaptic activity showing the amplitude of depolarizing synaptic events recorded intracellularly in HVC immediately before (blue) and after (red) tutoring (P < 0.00001). Data collected from 24 cells in 3 birds. (bottom) Cumulative frequency distribution of spontaneous synaptic activity showing the amplitude of depolarizing synaptic events recorded intracellularly in HVC on two consecutive days from a 45d untutored bird (green, 5 cells on each night) and a normally reared 45d bird (black, 2 cells on each night).