assembly of synaptic laminae by axon guidance molecules

Assembly of synaptic laminae by axon guidance moleculesEstuardo Robles and Herwig Baier

Available online at www.sciencedirect.com

A prominent feature of nervous systems is the organization of

synapses into discrete layers, or laminae. Such laminae are

essential for the spatial segregation of synaptic connections

conveying different types of information. A prime example of

this is the inner plexiform layer (IPL) of the vertebrate retina,

which is subdivided into at least ten sublaminae. Another

example gaining prominence is the layered neuropil of the

zebrafish optic tectum. Three recent papers have shed light

on the extracellular signals that control the precise

stratification of pre- and postsynaptic neuronal processes in

these two areas. The new studies implicate well-known axon

guidance cues, including class 5 and 6 semaphorins in the

retina, as well as Slit in the optic tectum. Remarkably, the

short-range action of Slit, which is required for neurite

stratification, appears to be achieved by anchoring the

secreted guidance factor to the basement membrane at the

surface of the tectum.

Address

Max Planck Institute of Neurobiology, Department Genes – Circuits –

Behavior, Am Klopferspitz 18, 82152 Martinsried, Germany

Corresponding author: Baier, Herwig ([email protected])

Current Opinion in Neurobiology 2012, 22:799–804

This review comes from a themed issue on Neurodevelopment

and disease

Edited by Joseph Gleeson and Franck Polleux

For a complete overview see the Issue and the Editorial

Available online 23rd May 2012

0959-4388/$ – see front matter, # 2012 Elsevier Ltd. All rights

reserved.

http://dx.doi.org/10.1016/j.conb.2012.04.014

Lamina-specific targeting of neurites by cell–cell recognition and extracellular cuesThe IPL of the vertebrate retina is a premier example of a

multilayered synaptic neuropil region. Here, the den-

drites of retinal ganglion cells (RGCs) stratify and come

into contact with axons of bipolar cells, which convey

visual information from the photoreceptors to the RGCs.

RGC dendrites also contact the neurites of amacrine cells,

a highly diverse class of largely inhibitory interneurons.

Amacrine cells shape and filter the transmitted infor-

mation at the bipolar-to-RGC synapses before it is trans-

mitted via RGC axons of the to the higher visual centers

in the brain. The IPL is a stack of >10 sublaminae

defined by the synapses among these cells, the functional

properties of which are still a matter of intense investi-

gation.

www.sciencedirect.com

Recently, important inroads have been made into deci-

phering the mechanisms that ensure the proper laminar

connectivity of RGC, bipolar and amacrine neurites.

Strong evidence has been obtained that intercellular

(cell–cell) recognition helps to target pre- and postsyn-

aptic neurites to the same laminar position. This mech-

anism is carried out by cell-surface adhesion molecules

belonging to the immunoglobulin superfamily (IgSF) of

transmembrane glycoproteins [1,2]. Often these mol-

ecules act in a homophilic fashion; i.e., they function as

reciprocal receptors and ligands on the surface of both

axons and dendrites. Expression of either one specific

IgSF molecule or a unique combination of different IgSF

molecules could specify the cellular identity of each

lamina. In this model, neurites with a matching comp-

lement of cell-surface molecules should join up, forming

one of the IPL sublayers. The homophilic IgSF mol-

ecules Sidekick1 (Sdk1), Sdk2, Dscam, and DscamL are

expressed in non-overlapping subsets of bipolar cells,

amacrine cells, and ganglion cells in the retina. For

example, as predicted by this model, expression of

Sdk1 on a neurite directs its stratification to a lamina

containing high levels of Sdk1. Other classes of molecules

have been implicated in cell-cell recognition, such as

Cadherins [3,4], but evidence for their involvement in

lamina formation is still tentative. A recent study showed

that cell-type specific expression of Cadherin-6 contrib-

uted to wiring of a subset of RGC axons to Cadherin-6-

expressing nuclei in the brain, but not to laminar targeting

[5]. In any case, a small number of specific cell-surface

recognition molecules should be able to generate a suffi-

cient number of unique labels for the fifty to a hundred, or

so, neuronal cell types of the vertebrate retina.

An IgSF code, or an extended IgSF-plus-Cadherin

code, however, is unlikely to be the whole story. First,

guidance mechanisms are required to bring neurites of

distant cells (e.g., the dendrites of RGCs originating in

the innermost cell body layer and the axons of bipolar

cells from the center of the retina) into close apposition

to be able to ‘compare’ their cell-surfaces and accept, or

reject, potential synaptic partners. Time-lapse studies

in the zebrafish retina have revealed that RGC den-

drites, bipolar cell axons and amacrine neurites grow

toward the IPL in an apparently directed fashion [6]

and that RGCs are not required for the sublamination

of the interneurons presynaptic to them [7]; there is

very little trial-and-error in this process, suggesting

guidance over distances greater than the reach of an

axon arbor. Second, a model of lamina formation that

works by simply matching cell surfaces cannot explain

the stereotyped spatial order and consistent orientation


mailto:[email protected]

http://www.sciencedirect.com/science/journal/09594388/22/5



http://www.sciencedirect.com/science/journal/09594388

800 Neurodevelopment and disease

of sublaminae observed in vivo. A developing laminar

structure is likely patterned before the axonal and

dendritic processes of neurons arrive that contribute

to its mature architecture.

On these grounds, a model has been proposed in which

signaling between extracellular guidance cues and their

transmembrane receptors guides neurites to their correct

laminar positions. Localized secretion and/or deposition

of these guidance molecules onto the substrate could then

provide directionality to the outgrowth of neurites. In this

model, the stereotyped order of the sublaminae could be

achieved by the differential responses of various cell

types to the resulting gradients in their extracellular

environment. The past year has seen significant progress

in identifying possible guidance factors crucial to the

formation of synaptic laminae. Remarkably, the two

families of guidance cues had previously been shown

to be capable of steering axons to their targets, often

over distances of millimeters, or even centimeters [8,9].

Although it has been tempting to construct models in

which similar pathways control neurite stratification over

much shorter distances, direct evidence for such a model

had been lacking.

Transmembrane semaphorins conspire tospecify retinal laminationA recent study has discovered a novel role for transmem-

brane semaphorin 6A (Sema6A) signaling in the regula-

tion of retinal neurite stratification. Semaphorins are a

large family of repulsive guidance cues that include both

secreted and transmembrane proteins [10]. In addition to

a role during axon guidance, a class 3 semaphorin has

been shown to direct both the axons and dendrites of

pyramidal neurons to their appropriate cortical layers

[11,12]. Class 6 transmembrane semaphorins are repul-

sive guidance cues known to signal through class A plexin

receptors, including Plexin A4 (PlexA4), to repel sym-

pathetic neuron axons and to guide mossy fiber growth in

the hippocampus [13,14].

In a new study, Matsuoka et al. [15��] found that mice

lacking PlexA4 had specific defects in IPL lamination

involving only a few retinal cell types. Among those

neurons affected were dopaminergic amacrine cells

(DACs), which are synaptically coupled to M1 melanop-

sin-positive intrinsically photosensitive RGCs (ipRGCs)

[16,17]. In wild-type retinas, the neurites of DACs co-

stratify with M1-ipRGCs dendrites in the OFF region of

the IPL within the S1 sublamina. In PlexA4-deficient

retinas, both DAC and M1-ipRGC neurites stratify in

their normal position but also form aberrant strata within

the ON layer of the IPL. This suggests that restricted

neurite stratification in OFF layers is controlled, at least

partly, by semaphorin-mediated repulsion away from ON

layers. This is consistent with the pattern of Sema6A

protein distribution, which is concentrated within the ON


layers of the IPL, while PlexA4 expression is restricted to

a small number of neuronal subtypes in the OFF region of

the IPL, including expression in DACs. However, PlexA4

is not expressed in M1-ipRGCs, indicating that the

position of these RGC dendrites is not directly mediated

by Sema6A. A role for amacrine-derived strata as a target

for RGC dendrite stratification has previously been docu-

mented in the developing zebrafish retina [6]. It therefore

likely that this will turn out to be a general mechanism for

dendrite targeting.

Interstingly, the new findings [15��] suggest that

Sema6A/PlexA4 signaling might not be necessary for

synapse formation between DACs and M1-ipRGCs.

Although loss of Sema6A/PlexA4 signaling causes both

DAC and M1 neurites to incorrectly target ON layers of

the IPL, these mistargeted neurites do colocalize. This

implies that Sema6A/PlexA4 signaling is dispensable

for cell-type specific synapse formation, as this appar-

ently still occurs between mistargeted neurites. One

explanation for this counterintuitive finding stems from

recent evidence that DACs and M1 ipRGCs in the

OFF layers of the IPL receive en passant synaptic input

from Type 6 ON bipolar cells [18]. Since bipolar cell

morphology was apparently unaffected in Sema6Amutants, mistargeted DAC and M1-ipRGC neurites

may be drawn to the ON layers in order to form

synapses with the distal terminus of Type 6 bipolar

cells. Within this scheme, it is possible that cell

adhesion molecules, such as Sdks, are still able to direct

synapse formation between mistargeted processes of

M1-ipRGCs and DACs.

This highly specialized role for a transmembrane sema-

phorin in laminar targeting contrasts with a recently

discovered role for class 5 transmembrane semaphorins

in regulating retinal laminar organization [19�]. In

Sema5A;Sema5B double mutants (these two class 5

semaphorins are functionally redundant) there is also

IPL mistargeting, however the predominant effect is

robust and generalized aberrant neurite targeting to the

inner nuclear layer (INL) and outer plexiform layer

(OPL). Sema5A/B signaling through PlexA1 and

PlexA3 receptors (which are functionally redundant

for Sema5 signaling in the retina) controls the stratifica-

tion of many retinal cell types that stratify in the IPL,

in addition to DACs and IP-RGCs. These findings

provide an example of synaptic specificity being con-

trolled at multiple levels (see Figure 1). In the case of

DACs, Sema5 signaling repels initial neurite extension

away from the INL. Subsequently, Sema6A functions

to direct DAC terminal arborization to OFF strata of

the IPL by repelling their neurites away from ON

layers. Should these aberrantly targeted neurites actu-

ally form functional synapses, it will be interesting to

identify the molecular determinants that direct synaptic

specificity in this context.


Assembly of synaptic laminae by guidance molecules Robles and Baier 801

Figure 1

Sema5A + Sema5BONL

INL

ON

OFF

Sema6A

(a) (b) (c) (d)

THTHTHTH

IPL

GCL M1 M1 M1 M1

Current Opinion in Neurobiology

Model for semaphorin-dependent neurite stratification in the retina. (a) During early retinal development class 5 semaphorins are expressed by cells in

the ONL. Contact with Sema5-expressing tissues biases the extension of amacrine neurites, such as tyrosine hydroxylase-expressing dopaminergic

amacrine cells (DACs) labeled ‘TH’. Following this initial redirection, DAC neurites extend into the OFF layers of the IPL (b). Growth into the deeper ON

layers is prevented by Sema6A/PlexA4 signaling, resulting in DAC arborization in a superficial sublamina of the OFF layers (c). Subsequently, M1

intrinsically photosensitive RGCs, which do not express PlexA4, extend dendrites through the region of Sema6A expression in order to costratify and

form synapses with DACs (d).

Slit directs retinal axon lamination from anECM scaffoldIn the zebrafish visual system, RGC axons form discrete

laminae within the neuropil of the optic tectum, the

primary retinorecipient brain region in most vertebrates

[20]. Synaptic activity does not seem to contribute to

lamina-specific targeting in the fish tectum [21],

suggesting a role for genetic factors in axonal lamination.

Indeed, forward genetic screens for mutations that disrupt

the orderly stratification of retinotectal axons discovered a

number of molecules directly or indirectly involved in

this process [22,23]. The strongest and most specific

phenotype was observed in dragnet, a mutant deficient

in Collagen IV alpha 5 (Col4a5). Col4a5 is a component of

the basement membrane lining the surface of the tectal

neuropil [24]. Disruption of this basement membrane

caused a subset of RGC axons to stray outside of their

normal layer. However, integrin-mediated adhesion to

Col4a5 was found not to be necessary for proper RGC

axon lamination [24], suggesting that the ECM plays an

indirect role in laminar targeting in the tectum.

A new study from T. Xiao and the authors of this review

demonstrates that loss of Slit/Robo signaling phenocopies

the lamination deficits observed in the dragnet mutant

[25��]. RGC axon lamination is strongly perturbed in astraymutants lacking functional Robo2, the Slit receptor

expressed by RGCs. Moreover, similar defects in laminar

precision were observed following knockdown of Slit1

expression. Importantly, Robo2 functions cell autonom-

ously to control RGC lamination, since astray RGCs trans-

planted into wild-type hosts exhibit similar deficits in

laminar targeting. Interestingly, tectal neuron dendrite

stratification was disrupted in both astray and dragnetmutants, which is consistent with a model in which Slit/


Robo signaling serves a dual function in specifying the

laminar position of both axons and dendrites in the tectum.

These studies provide evidence that the elaborate tissue

architecture at the surface of the tectum serves an essential

role in Slit-mediated axon lamination. Acute expression of a

GFP-tagged Slit resulted in stable enrichment of this

protein at the tectal surface. Subsequent experiments

showed that this enrichment is dependent on two structural

elements unique to the basement membrane at the surface

of the tectum. Direct binding between Slit and Col4a5 invitro suggested that Slit is concentrated at the tectal surface

through direct binding to the ECM. Correspondingly, sur-

face Slit enrichment was not observed in the dragnetmutant, which lacks Col4a5. Previous findings indicated

that radial glia in the tectum attach to the basement

membrane at the surface of the tectum to form a dense

plexus of broad, flattened endfeet. Surprisingly, ablation of

radial glia in tecta with an intact ECM also disrupted Slit

enrichment at the tectal surface [25��]. Therefore, glial

endfeet and extracellularly deposited adhesion molecules

are each essential for the tectal surface to function as an

organizing center for axon lamination (Figure 2).

Not only do these findings demonstrate that a classical

axon guidance cue can specify laminar position to growing

axons, they also suggest that extracellularly deposited

adhesion molecules can modulate these signals. Xiao

et al. [25��] observed that in many cases axons devoid

of Slit/Robo signaling meandered toward the surface of

the tectum, consistent with Slit functioning as a repulsive

cue. However, there was also a subset of axons that

strayed away from the tectal surface, suggesting that

subsets of RGCs are normally attracted to regions with

high Slit levels. A potential explanation is that these axons



Figure 2

(a) Wild Type

Slit Slit

Robo2 (astray) mutant

Col4a5 (dragnet) mutantSlit overexpression

Slit morphant

Skin Cells

Collagen IV

Slit

RGC Axons

Tectal NeuronCell Body

Radial Glia

(d)

(b)

(e)

(c)

Current Opinion in Neurobiology

Role of Slit signaling in retinal axon lamination. (a) In WT skin cells at the surface of the tectum produce Collagen IVa5 (red), an ECM component of the

basement membrane at the tectal surface. Slit protein (differing shades of green) is expressed throughout the tectum and concentrated at the tectal

surface. The local concentration at the surface (solid green band) requires an intact basement membrane and attachment by the endfeet of radial glia

(RG) and functions to guide axons to precise laminar positions. The Slit gradient extending into the deeper layers of the tectum is hypothetical. RGC

axons differentially responsive to Slit (indicated by teal, magenta and blue coloring) target different laminar depths. (b) In astray mutants loss of Robo2

results in aberrant lamination of RGC axons, including formation of broad, disorganized arbors, and in some cases arborize in multiple laminae. A

subpopulation of axons that normally target a lamina near the tectal surface are misdirected to deeper neuropil layers (teal axon). (c) Similar effects are

observed when Slit expression is knocked down. However, the basement membrane remains intact. (d) Overexpression of Slit in the tectum results in

disorganized lamination and innervation of neuropil regions that are normally not retinorecipient. (e) In dragnet mutants, loss of Collagen IVa5

expression disrupts the basement membrane, which in turn prevents Slit protein from accumulating at the tectal surface. Similar lamination defects are

seen in spite of normal Slit expression levels.

could be attracted to Slit as a consequence of contact with

the ECM. A role for the ECM in the modulation of

guidance cue-receptor signaling has been demonstrated

in the guidance of RGC axons at the optic nerve head

[26], where interactions with the ECM protein Laminin

convert the guidance cue Netrin from an attractant to a

repellent. Although Robo and Laminin have been shown

to genetically interact during midline guidance of Droso-phila CNS neurons [27], it is unclear if a similar mech-

anism controls retinal axon lamination in vertebrates.

The nascent neuropil of the optic tectum ispre-patterned by specific ECM moleculesUntil last year, the strongest evidence implicating inter-

actions with the extracellular matrix (ECM) as a mech-

anism to generate layer-specific connections came from

the optic tectum of the chick embryo. Several extracel-

lularly deposited molecules are localized in bands within


or around the tectal neuropil, potentially forming borders

that guide the laminar orientation of axons and dendrites.

These large glycoproteins and proteoglycans include

Tenascin, Versican, and Nel; each have been shown to

influence neurite growth in vitro [28,29,30�]. A specific

role in vivo for these molecules has only been tested in the

case of Versican [29]. However, though Versican is nor-

mally expressed in retinorecipient layers of the chick

tectum, reduced Versican expression altered RGC arbor

size without affecting axon lamination. Nel is a glyco-

protein containing five chordin-like and six epidermal

growth factor-like domains and is localized in specific

laminae of the chick tectum that retinal axons normally do

not enter. Nel-binding activity is detected on retinal

axons, suggesting that RGCs express an unknown re-

ceptor for Nel [30�]. In vitro, Nel induces growth cone

collapse and axon retraction. It will be interesting to

conduct genetic loss-of-function and gain-of-function


Assembly of synaptic laminae by guidance molecules Robles and Baier 803

experiments for Nel and other candidate laminar organi-

zers, similar to those involving Slit/Robo and Col4a5.

Implications and conclusionThe prevalence of synaptic laminae in the nervous sys-

tem indicates that the spatial convergence of similar

inputs is essential for network function. At least two

distinct mechanisms serve to target neurites to the appro-

priate laminae: homophilic cell–cell adhesion via IgSF

molecules and short-range guidance by classical axon

guidance cues. These mechanisms are not mutually

exclusive, as it is likely that both types of signals are

integrated in the process of laminar selection and, sub-

sequently, synapse formation. For example, in the case of

DACs, it is possible that class 5 and 6 semaphorins

function to restrict arborization to a subset of developing

synaptic layers within the IPL, while homophilic

adhesion may function to specify cell type-specific

synapse formation within these layers. This would

explain why DACs and M1-ipRGCs can still co-stratify

in Sema6A and PlexA4-deficient retinas. Similarly, the

studies in the zebrafish tectum cannot exclude the reten-

tion of synaptic specificity in the absence of Slit/Robo

signaling, since the laminar stratification of tectal cells is

also disrupted. This effect on tectal cell stratification may

permit appropriate connections to be formed with aber-

rantly targeted retinal axons.

Future studies should be aimed at disentangling these two

processes, initially by determining whether functional

synaptic connections are formed in Sema6A/PlexA4-

deficient retinas and Slit1/Robo2-deficient tecta. If func-

tional connections persist independent of fine laminar

targeting, it will be of interest to determine if mistargeting

in and of itself affects circuit function. In order to properly

test this, it will be necessary to perturb a single type of

synapse required for a specific behavior. This may be

possible in the case of the DAC-M1 connection, since

M1-ipRGC output is the predominant ipRGC output to

brain regions that control the pupillary light reflex [31].

In conclusion, we propose a hierarchy in which guidance

cues, anchored to the ECM or cellular surfaces, guide

neurites to their correct laminar position within a pre-

patterned laminar scaffold. Subsequently, synapse for-

mation between co-localized neurites is specified by cell–cell recognition between future presynaptic and postsyn-

aptic partners.

References and recommended readingPapers of particular interest, published within the period of review,have been highlighted as:

� of special interest�� of outstanding interest

1. Yamagata M, Weiner JA, Sanes JR: Sidekicks: synapticadhesion molecules that promote lamina-specificconnectivity in the retina. Cell 2002, 110:649-660.


2. Yamagata M, Sanes JR: Dscam and Sidekick proteins directlamina-specific synaptic connections in vertebrate retina.Nature 2008, 451:465-469.

3. Inoue A, Sanes JR: Lamina-specific connectivity in the brain:regulation by N-cadherin, neurotrophins, andglycoconjugates. Science 1997, 276:1428-1431.

4. Poskanzer K, Needleman LA, Bozdagi O, Huntley GW: N-cadherinregulates ingrowth and laminar targeting of thalamocorticalaxons. J Neurosci 2003, 23:2294-2305.

5. Osterhout JA, Josten N, Yamada J, Pan F, Wu S, Nguyen PL,Panagiotakos G, Inoue YU, Egusa SF, Volgyi B et al.: Cadherin-6mediates axon-target matching in a non-image-forming visualcircuit. Neuron 2011, 71:632-639.

6. Mumm JS, Williams PR, Godinho L, Koerber A, Pittman AJ,Roeser T, Chien C-B, Baier H, Wong ROL: In vivo imaging revealsdendritic targeting of laminated afferents by zebrafish retinalganglion cells. Neuron 2006, 52:609-621.

7. Kay JN, Roeser T, Mumm JS, Godinho L, Mrejeru A, Wong RO,Baier H: Transient requirement for ganglion cells duringassembly of retinal synaptic layers. Development 2004,131:1331-1342.

8. Tessier-Lavigne M, Goodman CS: The molecular biology of axonguidance. Science 1996, 274:1123-1133.

9. Dickson BJ: Molecular mechanisms of axon guidance. Science2002, 298:1959-1964.

10. Raper JA: Semaphorins and their receptors in vertebrates andinvertebrates. Curr Opin Neurobiol 2000, 10:88-94.

11. Polleux F, Morrow T, Ghosh A: Semaphorin 3A is achemoattractant for cortical apical dendrites. Nature 2000,404:567-573.

12. Polleux F, Giger RJ, Ginty DD, Kolodkin AL, Ghosh A: Patterningof cortical efferent projections by semaphorin–neuropilininteractions. Science 1998, 282:1904-1906.

13. Suto F, Tsuboi M, Kamiya H, Mizuno H, Kiyama Y, Komai S,Shimizu M, Sanbo M, Yagi T, Hiromi Y et al.: Interactions betweenplexin-A2, plexin-A4, and semaphorin 6A control lamina-restricted projection of hippocampal mossy fibers. Neuron2007, 53:535-547.

14. Suto F, Ito K, Uemura M, Shimizu M, Shinkawa Y, Sanbo M,Shinoda T, Tsuboi M, Takashima S, Yagi T et al.: Plexin-A4mediates axon-repulsive activities of both secreted andtransmembrane semaphorins and plays roles in nerve fiberguidance. J Neurosci 2005, 25:3628-3637.

15.��

Matsuoka RL, Nguyen-Ba-Charvet KT, Parray A, Badea TC,Chedotal A, Kolodkin AL: Transmembrane semaphorinsignalling controls laminar stratification in the mammalianretina. Nature 2011, 470:259-263.

This work discovered a novel role for transmembrane semaphorin 6Asignaling in regulating IPL lamination of several retinal cell types, includingdopaminergic amacrine cells and M1 melanopsin-positive intrinsicallyphotosensitive RGCs.

16. Zhang D-Q, Wong KY, Sollars PJ, Berson DM, Pickard GE,McMahon DG: Intraretinal signaling by ganglion cellphotoreceptors to dopaminergic amacrine neurons. Proc NatlAcad Sci U S A 2008, 105:14181-14186.

17. Viney TJ, Balint K, Hillier D, Siegert S, Boldogkoi Z, Enquist LW,Meister M, Cepko CL, Roska B: Local retinal circuits ofmelanopsin-containing ganglion cells identified bytranssynaptic viral tracing. Curr Biol 2007, 17:981-988.

18. Dumitrescu ON, Pucci FG, Wong KY, Berson DM: Ectopicretinal ON bipolar cell synapses in the OFF inner plexiformlayer: contacts with dopaminergic amacrine cells andmelanopsin ganglion cells. J Comp Neurol 2009,517:226-244.

19.�

Matsuoka RL, Chivatakarn O, Badea TC, Samuels IS, Cahill H,Katayama K, Kumar SR, Suto F, Chedotal A, Peachey NS et al.:Class 5 transmembrane semaphorins control selectivemammalian retinal lamination and function. Neuron 2011,71:460-473.



This study implicates class 5 semaphorins in the guidance of neurites tothe inner plexiform layer by preventing growth into the inner nuclear layerand outer plexiform layers of the retina.

20. Nevin LM, Robles E, Baier H, Scott EK: Focusing on optic tectumcircuitry through the lens of genetics. BMC Biol 2010, 8:126.

21. Nevin LM, Taylor MR, Baier H: Hardwiring of fine synaptic layersin the zebrafish visual pathway. Neural Dev 2008, 3:36.

22. Xiao T, Roeser T, Staub W, Baier H: A GFP-based genetic screenreveals mutations that disrupt the architecture of the zebrafishretinotectal projection. Development 2005, 132:2955-2967.

23. Nevin LM, Xiao T, Staub W, Baier H: Topoisomerase IIb isrequired for lamina-specific targeting of retinal ganglion cellaxons and dendrites. Development 2011, 138:2457-2465.

24. Xiao T, Baier H: Lamina-specific axonal projections in thezebrafish tectum require the type IV collagen Dragnet. NatNeurosci 2007, 10:1529-1537.

25.��

Xiao T, Staub W, Robles E, Gosse NJ, Cole GJ, Baier H: Assemblyof lamina-specific neuronal connections by slit bound to typeIV collagen. Cell 2011, 146:164-176.

This study provided evidence for Slit/Robo signaling as instructive forretinal axon lamination. Local deposition and accumulation of extracel-lular cues within the basement membrane make the tectal surface anorganizing center for axon lamination.


26. Hopker VH, Shewan D, Tessier-Lavigne M, Poo M, Holt C:Growth-cone attraction to netrin-1 is converted to repulsionby laminin-1. Nature 1999, 401:69-73.

27. Stevens A, Jacobs JR: Integrins regulate responsiveness to slitrepellent signals. J Neurosci 2002, 22:4448-4455.

28. Yamagata M, Herman JP, Sanes JR: Lamina-specificexpression of adhesion molecules in developing chick optictectum. J Neurosci 1995, 15:4556-4571.

29. Yamagata M, Sanes JR: Versican in the developing brain:lamina-specific expression in interneuronal subsets and rolein presynaptic maturation. J Neurosci 2005, 25:8457-8467.

30.�

Jiang Y, Obama H, Kuan SL, Nakamura R, Nakamoto C, Ouyang Z,Nakamoto M: In vitro guidance of retinal axons by a tectallamina-specific glycoprotein Nel. Mol Cell Neurosci 2009,41:113-119.

This study identified the glycoprotein Nel as a repulsive guidance cue forRGC axons in vitro. The Nel receptor on RGCs is still unknown.Theexpression patternof Nelprotein in the chick optic tectum suggests thatit serves to demarcate a laminarboundary to ingrowing axons.

31. Ecker JL, Dumitrescu ON, Wong KY, Alam NM, Chen S-K, LeGates T,Renna JM, Prusky GT, Berson DM, Hattar S: Melanopsin-expressing retinal ganglion-cell photoreceptors: cellulardiversity and role in pattern vision. Neuron 2010, 67:49-60.


assembly of synaptic laminae by axon guidance molecules

Documents