Extended Data Fig. 7: Details about dendritic spine calcium and glutamate imaging.
From: Emergence of a brainstem somatosensory tonotopic map for substrate vibration
a, Convergence between different end-organ types in the dendrites of single DCN neurons. In vivo two-photon calcium imaging of a DCN neuron with its dendritic field (top left). An example field of view with the soma of the targeted cell (top right). Nice example fields of view covering a part of the dendritic arbour (bottom right). Functional organisation of various submodalities in the dendritic field of the targeted cell (bottom left). b, Left, the percentage of the seven classified groups of dendritic spines in the DCN, defined by three mechanosensitive end-organ-specific stimuli. Right, the percentage of DCN neurons receiving synaptic inputs from one, two, or all three of the unimodal afferent input categories (that is MERK, MEIS, or PACI). c, Top, the distribution of different frequency preferences (tonotopy organisation) on a dendritic segment of the DCN neuron labelled with GCaMP8s. Bottom, postsynaptic activity from a single example spine was isolated by subtracting the component of the signal from the dendritic shaft. Red line: Duration of light stimulus. d, The relationship of distance between spines in a pair and frequency preference was calculated and it revealed that spines with a similar orientation preference were more likely to cluster. Error bars represent S.E.M. ∗p < 0.001, permutation test (n/N = 205 spines and 5 cells from 5 mice). e, Top, the distribution of the tuning preference of spines measured from the GCaMP8s calcium imaging or iGluSnFR3 glutamate imaging. bottom, the distribution of the tuning width measured from the calcium or glutamate imaging. No significant difference was found in any comparison (permutation test). f, The probability to find a responsive spine in iGluSnFR3 (n/N = 102 spines and 3 cells) is three times more than in GCaMP8s imaging (n/N = 246 spines and 4 cells). Error bars represent S.E.M. g, After alignment of the soma across multiple cells (N = 21), the centroid of the distribution of spines receiving the inputs from the information likely supplied by different mechanosensitive end-organs reveals the similar rostral-caudal bias of the end-organ types.
