Functional mapping of single spines in cortical neurons in vivo

Mapping neural connectivity in the brain It is notoriously difficult to determine how individual neurons integrate the inputs from the many incoming dendritic signals in live animals. Arthur Konnerth and colleagues use a new variant of the two-photon imaging technique to visualize sound-evoked activity in the spines of cortical neurons at single-synapse resolution in vivo . They find that individual spines are highly tuned for specific tones and that even neighbouring spines on the same dendrite can be tuned to different frequencies. This work establishes a new method for the mapping of functionally defined single synapses in the living brain. The individual functional properties and spatial arrangement of afferent synaptic inputs on dendrites have a critical role in the processing of information by neurons in the mammalian brain 1 , 2 , 3 , 4 . Although recent work has identified visually-evoked local dendritic calcium signals in the rodent visual cortex 5 , sensory-evoked signalling on the level of dendritic spines, corresponding to individual afferent excitatory synapses, remains unexplored 6 . Here we used a new variant of high-resolution two-photon imaging 7 to detect sensory-evoked calcium transients in single dendritic spines of mouse cortical neurons in vivo . Calcium signals evoked by sound stimulation required the activation of NMDA ( N -methyl- D -aspartate) receptors. Active spines are widely distributed on basal and apical dendrites and pure-tone stimulation at different frequencies revealed both narrowly and widely tuned spines. Notably, spines tuned for different frequencies were highly interspersed on the same dendrites: even neighbouring spines were mostly tuned to different frequencies. Thus, our results demonstrate that NMDA-receptor-dependent single-spine synaptic inputs to the same dendrite are highly heterogeneous. Furthermore, our study opens the way for in vivo mapping of functionally defined afferent sensory inputs with single-synapse resolution.