Wiring Mechanisms for Olfaction and Vision Neuron Preview PDF

Neuron Previews Wiring Mechanisms for Olfaction and Vision—Not Completely Different after All Esther T. Stoeckli1,* 1Institute of Molecular Life Sciences, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland *Correspondence: [email protected] http://dx.doi.org/10.1016...

Neuron Previews Wiring Mechanisms for Olfaction and Vision—Not Completely Different after All Esther T. Stoeckli1,* 1Institute of Molecular Life Sciences, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland *Correspondence: [email protected] http://dx.doi.org/10.1016/j.neuron.2014.11.002 Gradients of repulsive EphrinAs in the target were thought to repel temporal retinal ganglion cell axons expressing high levels of EphA receptors. Now, in this issue of Neuron, Suetterlin and Drescher (2014) show that EphrinA expressed on nasal axons contributes to the repulsion of temporal axons. There is probably hardly any neurosci- distinct topographic pattern of RGC behavior of RGC axons in vivo mirrored ence graduate student who has not axon targeting: nasal axons innervate the observed behavior of axons in the learned about the seminal work of Roger the posterior tectum because they are Bonhoeffer stripe assay, where nasal Sperry that led to the famous chemoaffin- less sensitive to the repellents, whereas RGC axons were seen to grow on ity hypothesis by which he proposed how temporal axons are restricted to the ante- stripes containing membrane prepara- the visual system gets wired (Sperry, rior tectum (summarized by Weth et al., tions taken from the anterior and the 1963). According to his model, at least 2014; Figure 1B). Finally, some 30 years posterior tectum. In contrast, temporal two perpendicular molecular gradients after Sperry’s chemoaffinity hypothesis RGC axons only extended on stripes are necessary to identify each cell in was published, these experiments led containing membranes from the rostral a two-dimensional target. Thus, every to the discovery of Ephrins (Drescher tectum (Figure 1B). target cell in the tectum carries a specific et al., 1995; Cheng et al., 1995) and Eph However, why axons expressing EphA address label for incoming retinal gan- receptors (for a review, see Lisabeth receptors would innervate the tectum at glion cell (RGC) axons. By expressing et al., 2013). Eph receptors and Ephrins all remains an unresolved issue. Initially, the appropriate combination of receptors can be subdivided into two groups each the idea was that attractive cues ex- for these labels, RGC axons would be (Lisabeth et al., 2013). EphA receptors pressed by the tectum could explain guided exactly to their target cells, result- bind to glycosyl-phosphatidyl-inositol- the ingrowth of axons into the tectum, ing in the topographic map of the visual anchored EphrinAs, whereas EphB re- while the repulsive system set up by Eph system that truthfully maintains the spatial ceptors bind transmembrane EphrinBs. receptors and repulsive Ephrin ligands information of the sensory input. The human genome encodes nine EphA would then be sufficient to explain the Specific targeting of projection neurons and five EphB receptors, five EphrinAs, distribution of nasal and temporal axons from the olfactory epithelium to the and three EphrinBs. EphrinAs and EphA within the tectum. However, to date no glomeruli in the olfactory bulb is of course receptors were found to be responsible such driving force or attractive cue has also needed but in the olfactory system for the rostrocaudal mapping, whereas been identified. Instead, further studies encoding of spatial information is not EphrinBs and EphBs were shown to be identified countergradients of EphA re- required. Rather, olfactory sensory neu- required for lateral-medial mapping of ceptors and EphrinA ligands also in the rons (OSNs) in the olfactory epithelium RGC axons in the tectum (summarized in tectum and in the retina, respectively that respond to the same odorant con- Weth et al., 2014). (summarized by Suetterlin et al., 2012). verge in the same glomerulus of the EphrinAs are expressed in an anterior- Based on these findings, cis-interactions low olfactory bulb, forming a discrete rather -posteriorhigh gradient in the tectum between EphA receptors and EphrinA than a topographic map (Cho et al., (Figure 1C). Their receptors, in particular ligands on RGC axons were suggested 2009). Thus, the olfactory and the visual EphA3, are expressed in a nasallow-tem- to fine tune their sensitivity to the repulsive systems are wired fundamentally differ- poralhigh gradient in the retina. Thus, environment in the tectum. ently (Figure 1A). Hence, it was not sur- RGC axon targeting in the tectum was This model was also compatible with prising that the molecular mechanisms explained by increasing repulsion of studies demonstrating that absolute underlying the wiring of these two sys- axons expressing higher receptor levels levels of EphA receptors and EphrinA tems were found to be different. from more repulsive posterior areas in ligands were not important, as axons Classical in vitro experiments by the tectum due to higher ligand levels: were found to be distributed in the tectum Friedrich Bonhoeffer, the ‘‘Bonhoeffer nasal axons can extend into the posterior based on relative levels of Ephrins (Brown stripe assay,’’ supported Sperry’s che- tectum, as they express low receptor et al., 2000). These studies were in agree- moaffinity hypothesis and demonstrated levels, whereas temporal axons are ment with Sperry’s observations that a gradient of repulsive molecules along repelled more strongly from the same removal of one half of the retina would the anterior-posterior axis in the chicken areas and therefore remain confined not result in a partially innervated tectum tectum as the driving force behind the to the anterior part of the tectum. The and conversely that removal of half of Neuron 84, November 19, 2014 ª2014 Elsevier Inc. 655  Neuron Previews the tectum would still preserve the topo- graphic pattern of innervation in the re- maining tectum. Thus, the prevailing idea is that RGC axons read the relative repulsive strength of the tectum and home in on an anterior-posterior position that is compatible with their active EphA expression level (Suetterlin et al., 2012). Based on all these studies, topographic map formation was considered to be the result of RGC axon-target interaction with a competition between RGC axons for less repulsive anterior positions in the tectum. However, based on a detailed comparison of retinotectal mapping in different species during development and regeneration of the visual system and by considering observations made in vitro by the Bonhoeffer lab, Weth and colleagues recently suggested a new model for visual system wiring. Based on their observations and theoretical consid- erations, they postulated that axon-axon interactions would contribute to topo- graphic mapping (Weth et al., 2014). The paper by Suetterlin and Drescher (2014) now provides experimental sup- port for the hypothesis that axon-axon in- teractions contribute to topographic map formation. The authors used conditional knockout mice lacking EphrinA5 either in the retina or in the superior colliculus, the target for RGC axons in mammals, to revisit observations made by Friedrich Bonhoeffer and colleagues a long time ago (summarized by Weth et al., 2014; Suetterlin and Drescher, 2014). They concentrated on EphrinA5 because it is expressed in a strong gradient in the retina, in contrast to EphrinA2 and EphrinA3, which are found to be ex- pressed in a shallow gradient or uniformly. The comparative analyses of two popu- lations of RGC axons targeting in the central area of the superior colliculus demonstrated that axon-axon interac- tions are crucial for the proper targeting Figure 1. Map Formation Requires Axon-Target and Axon-Axon Interactions of axons along the anterior-posterior axis. (A) The wiring pattern of the visual and the olfactory system differ fundamentally. In the visual system, pro- They found that temporal axons inner- jection neurons from the retina, the retinal ganglion cells, maintain their spatial order in the target area, the tectum in nonmammalian vertebrates, and the superior colliculus in mammals. In the olfactory system, vating a central area of the superior colli- olfactory sensory neurons that respond to the same odorant converge onto the same glomerulus in the culus (SC) were only slightly affected in olfactory bulb. In the more recent literature, the term topographic map has unfortunately also been used for the olfactory system to indicate that the location of glomeruli in the olfactory bulb has some spatial organization. However, this should not be confused with the topographic map in the visual system, where (C) A novel study (Suetterlin and Drescher, 2014) neurons in the retina maintain their spatial organization in the target area, such that axons from neigh- demonstrates axon-axon interactions as a mecha- boring RGCs target neighboring tectal cells to maintain the visual input. nisms contributing to topographic mapping in the (B) For the Bonhoeffer stripe assay membranes of anterior (a) and posterior (p) tectum are prepared sepa- superior colliculus at a local scale. Nasal axons ex- rately and offered as substratum for retinal explants (RGC explant) in a striped pattern. Nasal axons are pressing high levels of EphrinA5 repel temporal insensitive to the repulsive activity expressed in the posterior tectum and grow on either type of stripes, axons and thus support the global response whereas temporal axons fail to grow on posterior tectal membranes. derived from axon-target interaction. 656 Neuron 84, November 19, 2014 ª2014 Elsevier Inc. Neuron Previews their targeting in mice lacking EphrinA5 system closer to the olfactory system. In actions were found to be important before either in the colliculus or in the retina. the olfactory system, axons do not need contact with the target. Furthermore, However, a strong defect in temporal to maintain any spatial information when axon-axon interactions in the olfactory RGC axon targeting was found in mice innervating their target. Rather cells re- system are required for global patterning, lacking EphrinA5 in both retina and colli- sponding to the same sensory stimulus whereas interactions between RGC culus. In this case, temporal RGC axons or odorant innervate the same glomerulus axons in the visual system are important formed ectopic termination zones at in the olfactory bulb. The molecular mech- locally to sort out axonal topography. more caudal positions, as they were no anism underlying the convergence of longer repelled by the target and by nasal axons is largely unknown. Instead of the REFERENCES axons. classical axon guidance cues identified In contrast, nasal RGC axons were not in other systems, the olfactory receptors Brown, A., Yates, P.A., Burrola, P., Ortuño, D., affected more strongly in mice deficient themselves were suggested to be respon- Vaidya, A., Jessell, T.M., Pfaff, S.L., O’Leary, in EphrinA5 in both retina and SC sible for olfactory sensory neuron axon D.D., and Lemke, G. (2000). Cell 102, 77–88. compared to mice lacking EphrinA5 only guidance to the olfactory bulb. A few Cheng, H.J., Nakamoto, M., Bergemann, A.D., and in the SC. Thus, EphrinA5 in the retina ap- years ago, axon-axon interactions before Flanagan, J.G. (1995). Cell 82, 371–381. peared not to contribute to the targeting innervation of the olfactory bulb were Cho, J.H., Prince, J.E., and Cloutier, J.F. (2009). of nasal RGC axons. Together with obser- shown to result in a global distribution of Mol. Neurobiol. 39, 1–9. vations made for nasal and temporal OSN axons in the olfactory bulb (Imai Drescher, U., Kremoser, C., Handwerker, C., RGC axons that were innervating more et al., 2009). Axons innervating different Löschinger, J., Noda, M., and Bonhoeffer, F. peripheral areas of the SC, Suetterlin areas of the olfactory bulb did not (1995). Cell 82, 359–370. and Drescher (2014) concluded that the intermingle due to the expression of a Hornberger, M.R., Dütting, D., Ciossek, T., caudal overshooting of temporal RGC secreted repulsive signal or its receptor, Yamada, T., Handwerker, C., Lang, S., Weth, F., axons in mice lacking EphrinA5 in both respectively. A contribution of axon-axon Huf, J., Wessel, R., Logan, C., et al. (1999). Neuron retina and SC was due to the absence interactions to the innervation of the 22, 731–742. of repellent axon-axon interactions. olfactory bulb is not a special feature Imai, T., Yamazaki, T., Kobayakawa, R., Kobaya- These findings not only provide experi- of the olfactory system, as axon-axon kawa, K., Abe, T., Suzuki, M., and Sakano, H. (2009). Science 325, 585–590. mental evidence for the proposed model interactions prior to target innervation by Weth et al. (2014) but also shed some were found previously also for muscle Lisabeth, E.M., Falivelli, G., and Pasquale, E.B. new light on the theory that cis-interac- innervation by sensory and motor axons, (2013). Cold Spring Harb. Perspect. Biol. 5, a009159, http://dx.doi.org/10.1101/cshperspect. tions between EphA and EphrinAs would for instance (Milner et al., 1998). a009159. fine tune the sensitivity of RGC axons With their findings of axon-axon inter- (Hornberger et al., 1999). The observed action as an important mechanism Milner, L.D., Rafuse, V.F., and Landmesser, L.T. (1998). J. Neurosci. 18, 3297–3313. axon-axon interactions would require contributing to topographic map forma- EphrinA expression on RGC axons to tion, Suetterlin and Drescher (2014) have Sperry, R.W. (1963). Proc. Natl. Acad. Sci. USA 50, repel other RGC axons expressing EphA shown that the visual system is not so 703–710. receptors by trans-interactions and, different from other systems after all. Suetterlin, P., and Drescher, U. (2014). Neuron 84, thus, compete locally for target cells in Based on these new results, the mecha- this issue, 740–752. the tectum or the superior colliculus, nisms underlying the formation of a Suetterlin, P., Marler, K.M., and Drescher, U. respectively. discrete and a topographic map are no (2012). Semin. Cell Dev. Biol. 23, 1–6. Finally, the detection of axon-axon in- longer completely different. Weth, F., Fiederling, F., Gebhardt, C., and Bast- teractions as a mechanism contributing Some differences remain, however. In meyer, M. (2014). Semin. Cell Dev. Biol. 35C, to topographic mapping brings the visual the olfactory system, axon-axon inter- 126–135. Neuron 84, November 19, 2014 ª2014 Elsevier Inc. 657