, 2004 and Yu et al., 2005). The functional optical imaging experiments that revealed an intermediate-term memory trace in the DPM neurons were initially designed to challenge the now outdated hypothesis that the DPM neurons represent US input into the MBs, by testing the prediction that these neurons selleck chemicals llc would respond with calcium influx and synaptic release to electric shock delivered to the body of the fly but not to odor stimuli delivered to the antennae (Yu et al., 2005). Although the neurons do respond to electric shock pulses as predicted, they also respond to odors, and they show little discrimination in

their response between odors. Indeed, they responded to all 17 odors that were tested (Yu et al., 2005), making them “odor generalists.” These observations offered the possibility that the DPM neurons might form a memory trace, given their response to both CS and US stimuli. To probe this possibility,

flies were trained with odors and electric shock and then the responses of DPM neurons to the trained odors were assayed at different times after training. Remarkably, the coincidence of electric shock with odor caused a significant increase in the subsequent response of the DPM neurons to the trained odor (Figure 7), but not to an odor unpaired with shock (Yu et al., 2005). Furthermore, this training-induced plasticity forms only after a delay of ∼30 min. In other words, no increased calcium influx or synaptic release in response to the CS+ is detectable immediately AT13387 in vitro after conditioning; rather, this increase is detectable only 30 min later, indicating that this memory trace is “delayed” in its formation. The time course for the DPM memory trace coincides with intermediate-term behavioral memory. Initial experiments (Yu et al., 2005) indicated that the memory trace persists for at least 60 min after training with detectability becoming unreliable by 2 hr. More recent data show that the aversive memory trace persists to 70 min

after conditioning and is undetectable at 90 min after conditioning (I. Cevantes-Sandoval Adenylyl cyclase and R.L.D., unpublished data). The DPM memory trace is dependent on the expression of a wild-type copy of the amn gene in the DPM neurons: amn mutants fail to exhibit the memory trace while expressing a wild-type version specifically in the DPM neurons rescues the formation of the memory trace ( Yu et al., 2005). Most remarkably, the DPM memory trace is observed only in the DPM processes that innervate the vertical lobe of the MBs; the memory trace does not form in the processes that innervate the horizontal lobes. The role that this branch specificity plays in aversive olfactory memory remains unknown.