Plasticity in the representation of odor mixtures in the antennal lobe of honeybees

LOCATELLI F; GALIZIA G; SMITH B

San Diego, EEUU.

Congreso; XXXVII Annual Meeting of American Society for Neurosciences (SFN); 2007

American Society for Neurosciences

Program#/Poster#: 612.12/LL13 Title: Plasticity in the representation of odor mixtures in the antennal lobe of honeybees Location: San Diego Convention Center: Halls B-H Presentation Start/End Time: Tuesday, Nov 06, 2007, 11:00 AM -12:00 PM Authors: *F. F. LOCATELLI1, G. GALIZIA2, B. H. SMITH1; 1Sch. Life Sci., Arizona State Univ., Tempe, AZ; 2Lehrstuhl für Neurobiologie, Universität Konstanz, Germany Odors are normally present in nature as mixtures of several compounds. This constitutes a challenge for animals with odor driven behavior since they have to extract and recognize meaningful stimuli from a background of irrelevant components. To solve this task becomes even more difficult when the meaning or irrelevance is not fixed, but instead depends on experience. This is the case for honeybees, which use floral scents to find and exploit resources that the colony needs. Foragers continuously learn the contingency between odors and rewards, but equally important, they also learn to ignore stimuli that have no predictive value. This last form of learning, called “latent inhibition” or “CS pre-exposure effect” is induced by repeated unrewarded exposure to an odor. We studied if unrewarded exposure to one odor affects the neural representation of that odor when it is later present in a mixture. We performed calcium imaging on output neurons of the antennal lobe in a subpopulation of projection neurons that convey information from the AL to other areas of the brain. Neurons were labeled by backfilling with the calcium sensor dye Fura2. Odor-evoked activity was measured for two pure odors (1-hexanol and 2-octanone) and the corresponding binary mixture. After that, animals were subjected to a training protocol consisting of 40 unrewarded exposures that induces odor-specific latent inhibition for one of the odors. The representation for the pure odors and the mixtures was measured again after training. Analysis based on intensity and spatial distribution of the evoked activity shows odor specific activity patterns conserved across animals and characterized by partially overlapping groups of glomeruli. The mixture evoked a different activity pattern that involves the glomeruli activated by the pure components. The relative weight of each component in the activity elicited by the mixture, measured as the correlation coefficient between the component and the mixture, was variable across animals. Analysis of correlation and of euclidean distance between the activity patterns elicited by the components and the mixture after training shows that the activity elicited by the mixture is less similar to the pre-exposed odor and shifts toward the pattern elicited by the non-exposed odor. No change was observed for the correlation between pure odors, before and after training. These results suggest that unrewarded exposure induces changes in the antennal network that determines the interaction between odors. Disclosures:  F.F. Locatelli, None; G. Galizia, None; B.H. Smith, None. Support: NIH-NIDCD (DC007997 to BHS) NIH-NCRR (RR14166 to BHS)