Lack of a sensory input not only alters the cortical circuitry subserving the deprived sense, but also produces compensatory changes in the functionality of other sensory modalities. Here we report that visual deprivation produces opposite changes in synaptic function in primary visual and somatosensory cortices in rats, which are rapidly reversed by visual experience. This type of bidirectional cross-modal plasticity is associated with changes in synaptic AMPA receptor subunit composition.Loss of vision is usually accompanied by the increased functionality of other sensory modalities 1,2 . Systems-level analyses of cross-modal plasticity have revealed anatomical and functional rewiring of cortical circuits 3 . However, little is known about the cellular and molecular mechanisms underlying this type of plasticity. Here we examined whether manipulation of visual experience can induce bidirectional cross-modal plasticity of synaptic function in primary sensory cortices, and investigated the molecular mechanisms underlying this form of plasticity.To study cross-modal changes in synaptic function by visual deprivation, we dark-reared 4-week-old Long-Evans rats for a period of 1 week and then measured AMPA receptor (AMPAR)-mediated miniature excitatory postsynaptic currents (mEPSCs) in layer 2/3 pyramidal neurons in slices from primary visual, somatosensory and auditory cortex (Supplementary Methods online). In visual cortex, dark rearing produced an increase in mESPC amplitude that was reversed by re-exposing the rats to lighted conditions for 2 d Correspondence should be addressed to H.-K.L. (hlee21@umd.edu). 4 Current address: Brain Science Institute, Riken, Wako City, Saitama, Japan. 5 These authors contributed equally to this work.Note: Supplementary information is available on the Nature Neuroscience website. AUTHOR CONTRIBUTIONSA.G. and B.J. conducted the electrophysiology experiments (mEPSC recordings and rectification measurements, respectively) and assisted in writing the manuscript; L.W.X. and L.S. performed the biochemistry experiments; A.K. oversaw the electrophysiology (rectification measurements), contributed to discussions on experimental designs and collaborated on manuscript writing; H.-K.L. designed the studies, oversaw experiments, contributed to the electrophysiology (mEPSC recordings) and biochemistry and wrote the manuscript. COMPETING INTERESTS STATEMENTThe authors declare that they have no competing financial interests. (normal-reared (NR): 10.7 ± 0.6 pA, n = 8; dark-reared (DR): 12.4 ± 0.4 pA, n = 16; re-exposure to light (L): 10.7 ± 0.4 pA, n = 13; analysis of variance (ANOVA): F 2,34 = 5.968, P < 0.01; Fig. 1a). Notably, we observed the opposite changes in somatosensory cortex, where 1 week of dark rearing decreased the amplitude of mEPSCs and 2 d of light exposure reversed this effect (NR: 13.8 ± 0.8 pA, n = 12; DR: 11.3 ± 0.7 pA, n = 16; L: 14.1 ± 0.9 pA, n = 16; ANOVA: F 2,40 = 3.830, P < 0.04; Fig. 1b). Changes in synaptic transmission by dark rearing seems to be general for pr...

SUMMARY Endocannabinoids are widely regarded as negative modulators of presynaptic release. Here we present evidence that in visual cortex endocannabinoids are crucial for the maturation of GABAergic release. We found that between eye opening and puberty, release changes from an immature state with high release probability, short-term depression (STD) and high release variability during irregular patterned activity, to a mature state with reduced release probability, STD and variability. This transition requires visual experience and stimulation of CB1 cannabinoid receptors as it is mimicked by administration of CB1 agonists, blocked by antagonists and is absent in CB1R KO mice. In immature slices, activation of CB1 receptors induces long-term depression of inhibitory responses (iLTD), and a reduction in STD and response variability. Based on these findings, we propose that visually induced endocannabinoid-dependent iLTD mediates the developmental decrease in release probability, STD and response variability, which are characteristic of maturation of cortical GABAergic inhibition.