Mitral and tufted cell axons form the lateral olfactory tract (LOT), which relays olfactory bulb output directly to pyramidal cells in the olfactory cortices. Pyramidal

neurons in olfactory cortical areas close the loop by sending axon collaterals back to the olfactory bulb (Johnson et al., 2000; Luskin and Price, 1983; Figure 1). These feedback projections are the focus of the two papers in this issue of Neuron. Two papers in this issue describe experiments in which optogenetic approaches are used to produce selective activation of feedback cortical projections to the olfactory bulb. Two divisions of primary olfactory cortex are targeted: the anterior olfactory nucleus (AON) (Markopoulos et al., 2012) and the anterior piriform cortex (APC) (Boyd et al., 2012). These two areas have similar cellular and circuit properties with pyramidal cells mediating extensive feed-forward, Trichostatin A concentration recurrent, JQ1 and feedback projections within and between brain areas. In both studies, adenoassociated viral vectors (AAVs) were used to express the light-activated ion channel, channelrhodopsin (ChR2), along with fluorescent reporter proteins in cortical neurons. Markopoulos et al. (2012) injected virus into the AON that nonspecifically infected cortical neurons;

Boyd et al. (2012) used a conjunctive approach that limited ChR2 expression to pyramidal neurons of the APC. Both approaches generated similar patterns of fluorescently labeled axons in the ipsilateral olfactory bulb. Specifically, they observed bright fluorescence in the glomerular and granule cell layers, and minimal expression in the mitral cell of and external plexiform layers, consistent with previous anatomical work on centrifugal inputs to the olfactory bulb (Luskin and Price, 1983). These data suggest that pyramidal cell axons provide strong feedback at two stages of bulbar processing; influencing circuits both in the input glomerular layer and in the deeper granule

cell layer. A second feature of this feedback is that neurons from the AON, but not the APC, provided a similar, though weaker, pattern of input to the contralateral bulb. This suggests that AON feedback plays an additional role in bilateral processing between the two olfactory bulbs (Yan et al., 2008). But what synaptic connections are made by these pathways? Optical activation of ChR2+ terminals within the olfactory bulb reveals four key features of cortical feedback. First, the dominant effect of light-activated cortical feedback is inhibition that is sufficient to suppress the firing rates of mitral cells both in vitro and during odor presentation in vivo. Both groups report that this inhibition is mediated through a disynaptic path in which axons of cortical projection neurons excite granule cells, which in turn, inhibit mitral cells.