Olfactory map formation in the Drosophila brain: genetic specificity and neuronal variability
Introduction
Olfaction depends on the differential activation of olfactory sensory neurons (ORNs) and proper transmission of their activities to the brain. From insects to mammals, olfactory input is spatially organized into distinct sensory channels, which form a discrete neural map in the brain [1, 2, 3, 4]. Based on the specific OR type they express, ORNs in the sensory epithelium fall into distinct functional classes and all ORNs of the same class converge their axons onto a common synaptic target in the brain. The primary synaptic center of olfactory processing, the insect antennal lobe (AL) and the olfactory bulb in vertebrates, is subdivided into synaptic glomeruli and monospecificity of ORN sensory input defines an evolutionary conserved glomerular feature [5].
The Drosophila AL consists of ∼50 glomeruli of stereotypical shape and position, which fits the estimated 50 ORN classes localized in the two olfactory sense organs, antenna and maxillary palp [1, 4, 6, 7, 8] (Figure 1a). Inside the glomerulus, ORNs make synapses with two major types of second-order interneurons, the projection neurons (PNs) and the local interneurons (LNs). Most PNs elaborate dendrites in only one of the 50 AL glomeruli, thereby transmitting ORN input with point-to-point correspondence to the high brain centers [9]. PNs extend axons through distinct tracts and acquire class-specific arborization patterns in the mushroom bodies and the lateral horns [8, 9]. By contrast, LNs form an extensive network of inhibitory and excitatory synaptic connections with both PNs and ORNs, and these interconnections play central roles in the processing of olfactory information [10, 11, 12].
Although the molecular and cellular organization of the insects and mammalian olfactory system is remarkably similar [5], the developmental control mechanisms underlying olfactory circuit formation are quite different. In mammals, sensory neurons stochastically select a single OR gene, which subsequently blocks the expression of a second receptor in a negative feedback mechanism [13, 14, 15]. In addition, the functional OR type localizes in growing ORN axons to determine their synaptic specificity in the olfactory brain center [16]. By contrast, receptor-to-neuron map in Drosophila is highly stereotyped and depends on the lineage-related genetic control mechanisms with no indication of a negative feedback control [17]. Furthermore, OR expression in developing sensory neurons starts after their axons have established class-specific connections in the brain, excluding a role of Drosophila ORs in neuronal wiring [4, 18, 19]. This raises the question of how a precise olfactory map can be established in an OR-independent fashion.
Here we discuss recent findings on the genetic control of sensory and synaptic specificity in the Drosophila olfactory system. Mutant studies have identified critical components in sensory neuron differentiation that ensures unique OR gene expression. A comprehensive cell lineage analysis showed that different developmental strategies are employed to specify the neuronal identity of ORN classes and PN/LN classes.
Section snippets
Specification of olfactory receptor neurons
The Drosophila antennae and the maxillary palps are covered by more than 450 olfactory sensilla, each of which contains a stereotypic combination of up to four different ORNs [8]. Three main morphological types of sensilla, concentrated at distinct antennal regions, can be distinguished [1, 8, 20]. Whereas basiconic and trichoid sensilla contain ORNs that express one of the 60 OR genes encoding seven-transmembrane G protein-coupled receptors (GPCRs) [20, 21, 22], the coeloconic sensilla contain
Development of olfactory receptor neuron connectivity
During Drosophila pupal development, newly specified ORNs send their axons in separate fascicles from the antennal epithelium towards the AL, in which PN dendrites have already established a coarse positional map [30]. ORN axons first project in distinct medial-lateral pathways along the periphery of the AL, before they converge into initial protoglomeruli [18]. Glomerulus maturation involves the class-specific assembly of ORN axons and PN dendrites into functional microcircuits, as well as the
Olfactory receptor gene choice
The mechanism underlying the specific expression of OR genes is not well understood. As the premature expression of an ectopic OR does not effect the endogenous receptor selection, a negative feedback mechanism to ensure the ‘One neuron–One receptor’ rule is absent in Drosophila [17]. Specification of OR identity rather seems to be linked to segregation of distinct cell fates in a lineage-related mechanism as described above. Studies on the organization of OR gene regulatory elements have
Specification of second-order interneurons
The AL contains synaptic connections between axons of 1300 ORNs and dendrites of roughly 200 PNs and 100 LNs [46•, 47]. In contrast to the ‘single class-single glomerulus’ segregation of sensory axons, the innervation of PN and LN dendrites in the AL is more heterogeneous (Figure 2a). Although the majority of PN classes limit their dendritic arborizations to a single glomerulus, recent systematic analysis of neuronal diversity in the AL revealed additional multiglomerular PNs [46•] (Figure 2a).
Cellular interactions in PN–LN patterning
The initial targeting of PN dendrites precedes the arrival of pioneering ORN axons, indicating that the coarse glomerular map derives from complex cellular interactions among PN dendrites [30]. PN cell-body position in the three cell clusters does not predict the final locations of their dendrite projections [9, 50]. PN dendrite targeting to distinct glomeruli seems to occur in a step-wise fashion, in which outgrowing dendritic processes are initially positioned in a coarse map in the AL
Conclusions
Although recent studies have provided an almost complete morphological description of the hundreds of different neuronal cell types that constitute an olfactory map in the Drosophila brain, we are just beginning to understand the molecular mechanisms that provide each of these neurons with a unique cellular identity. Initially, sensory neurons and second-order target neurons in the brain become specified independently by cell-intrinsic determinants, followed by self-organizing patterning
References and recommended reading
Papers of particular interest, published within the period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Acknowledgements
We thank C. Klämbt, P. Kain and members of the Hummel lab for helpful comments on the manuscript and apologize to all whose work we were not able to include. T.H. is supported by the DFG Heisenberg Program, the EMBO Young Investigator Program and the Schram Foundation.
References (66)
- et al.
Molecular, anatomical, and functional organization of the Drosophila olfactory system
Curr Biol
(2005) - et al.
Visualizing an olfactory sensory map
Cell
(1996) - et al.
Topographic organization of sensory projections to the olfactory bulb
Cell
(1994) - et al.
An olfactory sensory map in the fly brain
Cell
(2000) - et al.
Genetic and functional subdivision of the Drosophila antennal lobe
Curr Biol
(2005) - et al.
Transmission of olfactory information between three populations of neurons in the antennal lobe of the fly
Neuron
(2002) - et al.
Two-photon calcium imaging reveals an odor-evoked map of activity in the fly brain
Cell
(2003) - et al.
Interchromosomal interactions and olfactory receptor choice
Cell
(2006) - et al.
Gene switching and the stability of odorant receptor gene choice
Cell
(2004) - et al.
Axon guidance of mouse olfactory sensory neurons by odorant receptors and the beta2 adrenergic receptor
Cell
(2004)