Transient axonal glycoprotein-1 (TAG-1) and laminin-α1 regulate dynamic growth cone behaviors and initial axon direction in vivo
How axon guidance signals regulate growth cone behavior and guidance decisions in the complex in vivo environment of the central nervous system is not well understood. We have taken advantage of the unique features of the zebrafish embryo to visualize dynamic growth cone behaviors and analyze guidance mechanisms of axons emerging from a central brain nucleus in vivo.
We investigated axons of the nucleus of the medial longitudinal fascicle (nucMLF), which are the first axons to extend in the zebrafish midbrain. Using in vivo time-lapse imaging, we show that both positive axon-axon interactions and guidance by surrounding tissue control initial nucMLF axon guidance. We further show that two guidance molecules, transient axonal glycoprotein-1 (TAG-1) and laminin-α1, are essential for the initial directional extension of nucMLF axons and their subsequent convergence into a tight fascicle. Fixed tissue analysis shows that TAG-1 knockdown causes errors in nucMLF axon pathfinding similar to those seen in a laminin-α1 mutant. However, in vivo time-lapse imaging reveals that while some defects in dynamic growth cone behavior are similar, there are also defects unique to the loss of each gene. Loss of either TAG-1 or laminin-α1 causes nucMLF axons to extend into surrounding tissue in incorrect directions and reduces axonal growth rate, resulting in stunted nucMLF axons that fail to extend beyond the hindbrain. However, defects in axon-axon interactions were found only after TAG-1 knockdown, while defects in initial nucMLF axon polarity and excessive branching of nucMLF axons occurred only in laminin-α1 mutants.
These results demonstrate how two guidance cues, TAG-1 and laminin-α1, influence the behavior of growth cones during axon pathfinding in vivo. Our data suggest that TAG-1 functions to allow growth cones to sense environmental cues and mediates positive axon-axon interactions. Laminin-α1 does not regulate axon-axon interactions, but does influence neuronal polarity and directional guidance.
KeywordsGrowth Cone Convergence Point Zebrafish Brain Neuronal Polarity Medial Longitudinal Fascicle
In the developing nervous system, axon pathway formation is governed by signals that steer the axonal growth cone in the appropriate direction. Although multiple guidance signals have been identified, our understanding of how specific growth cone behaviors and pathfinding steps are regulated in vivo is far from complete. We are investigating mechanisms controlling the pathfinding of axons extending from the nucleus of the medial longitudinal fascicle (nucMLF) in the zebrafish brain. The nucMLF neurons are the first neurons to differentiate and extend axons in the midbrain. They navigate uncharted territory and establish a major axon tract upon which multiple later-growing axons extend. We have focused particularly on the initial pathfinding steps that nucMLF axons must make: emergence from the cell body with correct polarity, extension in the correct direction, and convergence into an axon tract or fascicle. These crucial first steps are required by axons of many developing brain nuclei, although little is known about how they are regulated. We identify two guidance signals critical for initial nucMLF axon guidance: laminin, a component of the extracellular matrix (ECM) through which axons grow, and transient axonal glycoprotein-1 (TAG-1), a cell adhesion molecule (CAM) present on axons.
The first axons to extend in the brain use the ECM and undifferentiated neuroepithelial cells as growth substrata and guidance sources. Laminins are heterotrimeric glycoproteins, consisting of α, β, and γ polypeptide chains, that are secreted into the ECM and form a major component of the ECM in the CNS . In vitro studies have shown that laminin provides a permissive substratum for neuronal growth cones by binding and signaling through integrin receptors [2, 3, 4, 5, 6]. Moreover, laminin can modulate growth cone responsiveness to other extracellular guidance cues, such as netrins and ephrins [7, 8]. Interestingly, contact with laminin stimulates axon formation in cultured hippocampal neurons , and netrin controls axon initiation in Caenorhabditis elegans , suggesting laminin or its modulation of netrin could contribute to the initial polarity of axon emergence. Mutations in particular laminin subunits in C. elegans, Drosophila, and zebrafish embryos [11, 12, 13, 14, 15], as well as antibody perturbation of laminin in grasshopper , cause axon guidance defects, indicating that laminins are important for axon guidance in vivo. However, our understanding of laminin function and its influence on growth cone behaviors in vivo is incomplete.
CAMs of the immunoglobulin (Ig) superfamily play crucial roles in mediating interaxonal relationships and communication between axons and their surrounding cells [17, 18]. TAG-1 (or its chick homolog axonin-1) is a glycosyl-phosphatidylinositol (GPI)-linked IgCAM present on the surface of growing axons that promotes neurite outgrowth and interacts homophilically and heterophilically with a host of other cell surface molecules [19, 20, 21]. In vitro assays have shown that interactions with other IgCAMs, including L1 and NrCAM (NrCAM = neuron-glia related cell adhesion molecule), mediate axon outgrowth and cell-cell adhesion [22, 23, 24, 25, 26]. Furthermore, interactions of TAG-1 with L1 or NrCAM appear to be required in vivo for midline crossing by spinal commissural axons and for target selection by sensory neurons in the chick spinal cord [27, 28, 29]. Axon guidance defects in TAG-1 knockout mice have not been shown; however, neuronal migration in the caudal medulla is defective in these animals . We have found previously that TAG-1 knockdown in zebrafish inhibits extension of sensory axons in the spinal cord, although whether this is mediated by a homophilic or heterophilic interaction is unknown [27, 28, 29, 31]. Despite these studies demonstrating the importance of TAG-1, we have yet to fully understand whether or how TAG-1 controls directional axon guidance decisions or how it influences dynamic growth cone behaviors in vivo.
Here we investigate how TAG-1 and laminin-α1 influence growth cone behavior and axon outgrowth during the initial period of nucMLF axon development in vivo. Wild-type growth cone behaviors suggest that axon-axon interactions and possibly repulsion by surrounding tissue direct the initial pattern of nucMLF axon growth. Interestingly, TAG-1 knockdown causes defects in the directional extension and initial convergence of MLF axons that, in fixed tissue, appear identical to pathfinding errors previously observed in a laminin-α1 mutant, bashful . However, in vivo time-lapse imaging of growth cone behaviors reveals that although there are some similar effects of TAG-1 or laminin-α1 loss of function, there are also distinct and significant differences. Our results suggest that TAG-1 functions to guide the direction of axon extension by interpreting environmental cues and also mediates positive axon-axon contacts. Laminin-α1 does not appear to affect axon-axon interactions, but does regulate neuronal polarity and influence growth cone responsiveness to surrounding cues.
Characterization of normal MLF axon convergence
We first characterized the pattern of nucMLF axon outgrowth by examining fixed embryos at stages between 16 and 24 hpf. To specifically visualize nucMLF neurons, we used a transgenic line in which a pitx2 promoter construct drives green fluorescent protein (GFP) expression (Tg(pitx2c:gfp)) in nucMLF neurons (Figure 1b–e; see Materials and methods). We verified that all nucMLF neurons are labeled by double labeling embryos with antibodies against ZN-12, a marker of nucMLF neurons, and GFP (Figure 1b). At the onset of nucMLF axonogenesis, cell bodies each extended one axon in the direction of the caudal-most nucMLF cell body (Figure 1c–e, asterisk), which extended its own axon into the hindbrain, parallel to the ventral midline. In their initial trajectory, the earliest nucMLF axons typically did not contact neighboring axons since their cell bodies were positioned at some distance from one another, but later as these axons approached the caudal-most nucMLF cell and as the number of nucMLF cells increased, axon-axon contact was prevalent. Interestingly, the axons did not appear to extend into the tissue surrounding the nucMLF. Rather, they appeared to extend directly toward the caudal-most nucMLF cell body and eventually converged on a point just rostral to this cell (Figure 1c–e, arrow). In our subsequent analysis, we will refer to this point as the convergence point. Beyond the convergence point, MLF axons extend into the hindbrain as a tight fascicle.
Axon-axon interactions and the surrounding tissue contribute to the convergence of nucMLF axons
TAG-1 and laminin-α1 are required for the initial convergence of nucMLF axons
Given the significant axon-axon interactions we observed in the live imaging experiments, we asked whether TAG-1, a GPI-linked Ig superfamily CAM known to mediate axon-axon interactions , was critical for the initial growth and convergence of MLF axons. tag-1 mRNA is expressed by a subset of neurons in the zebrafish brain, including the nucMLF (Figure 3a) . We examined the expression of TAG-1 protein during initial stages of MLF development (16–24 hpf), and found that it is expressed on the nucMLF axons and weakly on the nucMLF cell bodies throughout these stages (Figure 3b–d, and data not shown). In addition, TAG-1 is expressed on axons from the nucleus of the tract of the postoptic commissure (nucTPOC) that begin to extend past the lateral side of the nucMLF around 19–20 hpf, and on a couple of cells lateral to the nucMLF, but not otherwise in the surrounding neuroepithelium. We injected Tg(pitx2c:gfp) embryos at the 1–4 cell stage with a translation blocking TAG-1 morpholino antisense oligonucleotide (TAG1MO) , allowed them to develop to 24 hpf, and examined the pattern of nucMLF axon outgrowth in fixed embryos by immunolabeling with an anti-GFP antibody. We have previously shown that injection of TAG1MO, but not a standard control morpholino (STDCONMO), eliminates TAG-1 immunolabeling, indicating the loss of TAG-1 protein . TAG-1 knockdown caused several defects in nucMLF axon initial growth and convergence, but did not appear to alter the number of nucMLF neurons (Figure 3i–k, m). Normally, nucMLF axons initially extend only within the area containing the nucMLF cell bodies (called the 'nucMLF zone'; outlined in Figure 3i–l) prior to reaching the convergence point. However, after TAG-1 knockdown, nucMLF axons aberrantly extended into the tissue medial, lateral, and rostral to the nucMLF zone (Figure 3k, open arrowhead). This phenotype was observed in 67.6% (n = 173) of TAG1MO injected embryos, compared with only 8.4% (n = 107) of STDCONMO injected embryos (Figure 3i–k, m). In addition, the axons converged at a point more caudal than the typical convergence point in 73.2% (n = 168) of embryos injected with TAG1MO, while injection of STDCONMO caused this defect in only 8.4% (n = 107) of embryos (Figure 3i–k, arrowheads, and Figure 3m).
Interestingly, the nucMLF axon defects induced by TAG-1 knockdown appeared very similar to defects we previously observed in bashful (bal) embryos, which have a mutation in the laminin-α1 subunit (Figure 3l) [15, 40]. Laminin is expressed broadly in the basal lamina along the ventral surface of the neuroepithelium, directly apposed to the nucMLF and the MLF axons during all stages examined (16–24 hpf; Figure 3e–h, and data not shown). In bal mutants, the nucMLF axons also appear to extend aberrantly outside the nucMLF zone before eventually converging in the anterior hindbrain. Moreover, these axons appear defasciculated and stalled in the hindbrain. The similar defects suggest that TAG-1 and laminin-α1 may function together to guide the initial convergence of nucMLF axons or that they may have similar influences on growth cone behavior to produce comparable pathfinding defects. However, the fixed tissue analysis of nucMLF axon defects was insufficient to address these questions.
NucMLF growth cones exhibit behavioral defects after TAG-1 or laminin-α1 loss of function
Additional file 3: Time-lapse sequence of MLF axon outgrowth in a TAG1MO injected Tg(pitx2c:gfp) embryo. Ventral view, anterior to the left, ventral midline is up. The excerpt begins at 17 hpf. The movie shows MLF axons aberrantly growing into the tissue surrounding the nucMLF. Images were captured every minute, and the movie runs at ten frames per second. (MOV 7 MB)
Additional file 4: Time-lapse sequence of MLF axon outgrowth in a bal;Tg(pitx2c:gfp) embryo. Ventral view, anterior to the left, ventral midline is up. The excerpt begins at 18 hpf. The movie shows MLF axons aberrantly innervating surrounding tissue and growing towards the ventral midline. Images were captured every minute, and the movie runs at ten frames per second. (MOV 8 MB)
TAG-1, but not laminin-α1, affects nucMLF axon-axon interactions
nucMLF axon branching increases in bal, but not after TAG-1 knockdown
nucMLF axons have reduced growth rates in TAG-1 knockdown and bal embryos
The stunted MLF outgrowth phenotype we observed in the fixed tissue analysis of TAG1MO injected and bal embryos (Figure 3i, j, m)  may be due, in part, to their initial extension in incorrect directions, but a reduced growth rate may also contribute to this phenotype. Therefore, we quantified the growth rate of axons in uninjected, TAG1MO injected, and bal embryos from their initial extension until they reached the convergence point or an equivalent position in the rostral-caudal axis (Figure 6c). Axons in uninjected embryos grew at an average of 14.11 ± 1.08 μm/h (Figure 6c). However, axons in TAG1MO injected or bal embryos grew significantly slower at an average of 10.63 ± 1.13 μm/h and 8.27 ± 0.96 μm/h, respectively (Figure 6c). Thus, in addition to their inefficient pathfinding towards the convergence point, reduced growth rates likely contribute to the retarded MLF axons observed after TAG-1 or laminin-α1 loss of function.
In this study, we use in vivo time-lapse imaging to show how TAG-1 and laminin-α1 affect dynamic growth cone behaviors, initial axon direction, and interactions among nucMLF axons in the zebrafish brain. Normally, nucMLF axons converge and fasciculate soon after extending from their cell bodies. Our results suggest this growth pattern is driven by a combination of guidance by surrounding tissue and axon-axon fasciculation. TAG-1 or laminin-α1 loss of function disrupted this stereotypical outgrowth pattern. First, TAG-1 or laminin-α1 loss of function caused MLF axons to aberrantly extend into the surrounding tissue. TAG-1 knockdown also caused MLF growth cones to retract from neighboring MLF axons, which led to aberrant directional extension of MLF axons and possibly influenced changes in neuronal polarity. Axon-axon interactions appeared normal in bal embryos; however, neuronal polarity was affected, since MLF axons often emerged from their cell bodies in the incorrect direction. MLF axons were also excessively branched in bal embryos. Finally, MLF axons in TAG1MO injected or bal embryos exhibited reduced growth rates and were stalled in the hindbrain. Taken together, these results reveal critical roles for TAG-1 and laminin-α1 in controlling specific growth cone behaviors, neuronal polarity, and directional outgrowth to guide proper formation of a major CNS pathway.
Interestingly, TAG-1 may serve dual functions in mediating nucMLF axon pathfinding as it appears to be required for sensing a signal in the surrounding tissue and for mediating axon-axon interactions. Several Ig superfamily CAMs, including TAG-1, have been shown to mediate axon fasciculation via homophilic and heterophilic binding partners [17, 18, 19]. Therefore, the defective nucMLF axon-axon interactions caused by TAG-1 knockdown were not surprising. It is highly possible that TAG-1 promotes adhesion among nucMLF axons either homophilically or heterophilically, although a heterophilic partner has not been determined in this case. The apparent consequences of these failed interactions were more intriguing. Often, growth cone retraction upon contacting a neighboring axon was followed by a change in direction and extension into the surrounding tissue, suggesting that these defective interactions may have contributed to further pathfinding errors. Furthermore, in a few cases, defective axon-axon interactions led to the complete retraction of the original axon and the emergence of a new axon on the opposite side of the neuron, thus altering the polarity and directional extension of nucMLF axons. Collectively, these results exemplify the significance of TAG-1 mediated axon-axon interactions not only to fasciculation, but also to directional guidance and neuronal polarity. Finally, it is possible that TAG-1 mediates interactions between nucMLF and nucTPOC axons, which also extend through the midbrain and join the MLF. However, nucTPOC axons grow into this region 2–3 hours after the initial outgrowth of nucMLF axons and appeared unaffected by TAG-1 knockdown.
In addition to this expected role for TAG-1, our results also suggest TAG-1 is required for nucMLF axons to sense a guidance signal from the environment because TAG-1 knockdown also caused nucMLF axons to grow in aberrant directions without first contacting other axons. This signal potentially could be an inhibitory cue located in tissue surrounding the nucMLF and/or an attractive cue emanating from the convergence point. This finding is interesting because TAG-1 is not known to function as a receptor for repulsive or attractive ligands. However, related Ig superfamily CAMs, such as L1 and NrCAM, have recently been shown to be required receptor components for repulsive semaphorin signaling [42, 43, 44, 45]. Thus, these observations raise the intriguing possibility that TAG-1 may also serve as a receptor component for repulsive and/or attractive ligands in vivo.
Laminin also appears to be involved in preventing MLF axon growth into inappropriate surrounding tissues and orienting outgrowth direction. Laminin is expressed throughout the basal lamina upon which MLF axons extend, and thus it could function directly to guide nucMLF axons by signaling via integrins or other receptor molecules on nucMLF growth cones. Alternatively, laminin-α1 may affect axons indirectly through more global effects on brain organization or by binding and presenting other guidance signals. An interesting possibility is that laminin-α1 may regulate growth cone responsiveness to other cues expressed by the surrounding tissue, such as netrin or ephrin, which have been shown to elicit repulsion in the presence of laminin [7, 8]. Without functional laminin-α1, nucMLF axons may be attracted or non-responsive to such cues. Interestingly, in bal, misguided axons appeared biased to extend towards the ventral midline rather than wandering in all directions equally. This observation suggests that laminin-α1 indirectly regulates axon repulsion by another cue in the midline region because laminin-α1 is expressed broadly, not only at the midline .
Loss of laminin-α1 also caused defects in polarity and branching. Establishing appropriate neuronal polarity is arguably the first step towards properly guiding an axon because it determines the initial direction of axon outgrowth. Normally, almost all nucMLF neurons immediately extended their axon in the direction of the convergence point. However, in bal, we observed a significant number of neurons that initially extended axons in aberrant directions, suggesting that laminin-α1 is critical for establishing the proper polarity of these neurons. Again, whether nucMLF neuronal polarity is directly or indirectly regulated by laminin-α1 remains to be determined. Recently, netrin was shown to regulate neuronal asymmetry and define the site of axon formation . It is possible that these roles may also require the presence of laminin, similar to netrin-mediated repulsion. Functionally, the nucMLF polarity defect may also have contributed to the increased percentage of time that axons extended off-track. In addition to polarity defects, axons were excessively branched in bal. It might be surprising that loss of laminin-α1, which usually functions as a permissive growth substratum, would directly stimulate increased branching. However, it is possible that the loss of laminin-α1 created an inhibitory substratum causing growth cone stalling or retraction, which has been correlated with back branching [47, 48]. These observations reveal novel roles for laminin-α1 in establishing neuronal polarity and regulating axon branching.
TAG-1 or laminin-α1 loss of function ultimately caused nucMLF axons to prematurely stall in the hindbrain. Interestingly, the stalled axons were almost always observed in the anterior hindbrain, instead of being positioned along a continuum from the anterior hindbrain to the anterior spinal cord, where they should have reached by 24 hpf. This observation suggests that an additional cue may be required to promote axon extension through this region and that late arriving axons may miss the temporal expression window of this factor. Alternatively, nucMLF axons may become non-responsive to the potential factor in the absence of TAG-1 and/or laminin-α1 function. In any case, it is interesting that the position of stalled nucMLF axons is the same in embryos deficient for TAG-1 or laminin-α1.
Collectively, our live imaging data reveal that TAG-1 and laminin-α1 loss of function cause some similar defects in nucMLF growth cone behavior, while also causing defects unique to the loss of each gene. The fact that both contributed to restricting nucMLF axon outgrowth to within the nucMLF zone might suggest that TAG-1 and laminin-α1 may function cooperatively to mediate inhibition of growth cones by surrounding tissue. One possibility is that TAG-1 and laminin-α1 act in separate signaling pathways, with common and distinct downstream effectors. Alternatively, laminin-α1 in the environment may directly bind TAG-1 on nucMLF growth cones, although no such interaction has been demonstrated. Instead, laminin-α1 may indirectly regulate the presentation of a binding partner of TAG-1. Future experiments will be required to address these alternative mechanisms.
Materials and Methods
Zebrafish (Danio rerio) were maintained in a laboratory breeding colony on a 14/10 h light/dark cycle. Embryos were maintained at 28.5°C and staged as described previously . bashful mutant embryos were obtained from fish carrying the baluw-1 allele , which contains a mutation that results in a predicted protein truncation at amino acid 1,424 of laminin-α1 . Animals were handled in accordance with guidelines set forth by NIH and IACUC, and the animal use protocol was approved by the University of Wisconsin Animal Care and Use Committee (assurance number A3368-01).
Generation of Tg(pitx2c:gfp) transgenics
The Tg(pitx2c:gfp) transgenic line was generated using an internal promoter of the pitx2 gene that specifically produces the pitx2c isoform . This promoter was obtained by amplifying a region from exon 2 to exon 4 of the pitx2 gene from a PAC clone. This fragment was fused to the enhanced GFP (EGFP) coding sequence followed by the SV40p(A) signal sequence from pEGFP-N1 (Clontech, Mountain View, CA, USA) and cloned into pBSIIKS- (Stratagene, La Jolla, CA, USA). The primers used for PCR amplification of the pitx2c promoter were: 5'-AGGGTACCGGACTCCCACTGCCGCAAACT-3' and 5'-TCTAGAGCTCTGTAATGCAAAAGGAAACAC-3'. The primers used to amplify the EGFP-SV40p(A) fragment were: 5'-TAACGAGCTCATGGTGAGCAAGGGCGAGGA and 5'-CAGAATTCTGAGTTTGGACAAACCACAAC-3'. To generate transgenic zebrafish, the vector sequence was removed from the pitx2c:egfp construct prior to injection into embryos at the one-cell stage. Three independent lines were isolated that all showed expression of EGFP in the trigeminal ganglia and the nucMLF.
For whole-mount immunohistochemistry, embryos were fixed in 4% paraformaldehyde overnight, blocked in 5% sheep serum and 2 mg/ml bovine serum albumin in phosphate buffered saline with 0.1% Tween, and incubated overnight (4°C) in anti-GFP (1:1,000; Sigma, St Louis, MO, USA), ZN-12 (1:250; Zebrafish International Zebrafish Center, Eugene, OR, USA) or anti-TAG1 (1:500; ). Antibody labeling was performed with the Vectastain IgG ABC immunoperoxidase labeling kit (Vector Laboratories, Burlingame, CA, USA). For fluorescent double labeling, Alexa-conjugated secondary antibodies (4 μg/ml; Invitrogen-Molecular Probes, Carlsbad, CA, USA) were used to bind primary antibodies.
Morpholino oligonucleotides were synthesized by Gene Tools (Corvallis, OR, USA). To knockdown TAG-1, we used a morpholino sequence (TAG1MO; 5'-CCACACCCA-GACCAGACACTTATTT-3') that has been previously reported and shown to block TAG-1 protein . As a control, we injected a standard control morpholino (STDCONMO; 5'-CCTCTTACCTCAGTTACAATTTATA-3'). Morpholino oligos were injected into newly fertilized embryos at the one- to four-cell stage as described previously . Optimal concentrations for the TAG1MO were determined by titrating the dose from 1–10 ng per embryo; 10 ng injections appeared to cause non-specific toxicity, 2.5–5 ng injections appeared to cause robust defects without non-specific cell death, and 1 ng injections caused partial, but insignificant, defects. Therefore, we injected TAG1MO at approximately 2.5–5 ng to elicit full knockdown and equivalent doses of STDCONMO were injected for each experiment.
TAG-1 mRNA rescue construct overexpression
To create a TAG-1 construct potentially capable of rescuing TAG-1 knockdown, we subcloned full-length TAG-1 into the pCS2+ vector and then removed the first 20 bases of the sequence complementary to the 25-mer TAG1MO sequence, located in the 5' untranslated region. The mMessage mMachine Kit (Ambion, Austin, TX, USA) was used to transcribe 5' capped mRNA from pCS2+-TAG1. Approximately 100 pg of RNA was injected into one-cell stage embryos. Overexpression was verified by immunolabeling for TAG-1 at 24 hpf.
All brightfield images were captured on a Nikon TE300 inverted microscope equipped with a Spot RT camera (Diagnostic Instruments, Sterling Heights, MI, USA) and processed with Metamorph software (Universal Imaging Corp., West Chester, PA, USA). Fluorescent images of fixed preparations are confocal images captured on an Olympus Fluoview 1000 Laser Scanning Confocal Microscope.
In vivo time-lapse imaging
Preparation of embryos for imaging was adapted from Langenberg and colleagues . Briefly, the yolk cell was paralyzed by adenosine 5' (β,γ-imido)triphosphate (Calbiochem, San Diego, CA, USA) injection at 15.5 hpf, and then removed. Each embryo was mounted between two coverslips separated by high vacuum silicon grease in 67% L-15. Embryos ranged in age from 16–18 hpf at the start of imaging and were imaged for 2–12 hours in a temperature controlled environment set to 29°C. Images were captured using a 60× dipping objective (NA = 1.00) on a Nikon (Tokyo, Japan) E-600FN equipped with standard epifluorescence, a filter wheel, and a CoolSnap HQ camera (PhotoMetrics, Tucson, AZ, USA). Images were captured every one minute, and exposure times were typically 300–600 ms. Images and movies were captured, processed, and analyzed with MetaMorph software (Molecular Devices, Sunnyvale, CA, USA.
We thank Ms Ann Regnery for technical assistance, Dr Nina Hasen for statistics consultation, Dr Claudia Stuermer for providing the TAG-1 antibody, and Dr Timothy Gomez for discussion and comments on the manuscript. This work was supported by NINDS grants NS042228 to MCH, NS040449 to AC, NHLBI grant NL66292 to HJY, and NIGMS T32 GM07507 and NINDS NRSA 1F31NS054430 to MAW. The confocal microscope was acquired with an NIH Shared Instrumentation Grant (S10RR023717, PI Bill Bement, Department of Zoology, University of Wisconsin).
- 25.Pavlou O, Theodorakis K, Falk J, Kutsche M, Schachner M, Faivre-Sarrailh C, Karagogeos D: Analysis of interactions of the adhesion molecule TAG-1 and its domains with other immunoglobulin superfamily members. Mol Cell Neurosci. 2002, 20 (3): 367-381. 10.1006/mcne.2002.1105.CrossRefPubMedGoogle Scholar
- 30.Denaxa M, Kyriakopoulou K, Theodorakis K, Trichas G, Vidaki M, Takeda Y, Watanabe K, Karagogeos D: The adhesion molecule TAG-1 is required for proper migration of the superficial migratory stream in the medulla but not of cortical interneurons. Dev Biol. 2005, 288: 87-99. 10.1016/j.ydbio.2005.09.021.CrossRefPubMedGoogle Scholar
- 40.Pollard SM, Parsons MJ, Kamei M, Kettleborough RN, Thomas KA, Pham VN, Bae MK, Scott A, Weinstein BM, Stemple DL: Essential and overlapping roles for laminin alpha chains in notochord and blood vessel formation. Dev Biol. 2006, 289 (1): 64-76. 10.1016/j.ydbio.2005.10.006.CrossRefPubMedGoogle Scholar
- 42.Falk J, Bechara A, Fiore R, Nawabi H, Zhou H, Hoyo-Becerra C, Bozon M, Rougon G, Grumet M, Püschel AW, Sanes JR, Castellani V: Dual functional activity of semaphorin 3B is required for positioning the anterior commissure. Neuron. 2005, 48 (1): 63-75. 10.1016/j.neuron.2005.10.024.CrossRefPubMedGoogle Scholar
- 50.Semina EV, Bosenko DV, Zinkevich NC, Soules KA, Hyde DR, Vihtelic TS, Willer GB, Gregg RG, Link BA: Mutations in laminin alpha 1 result in complex, lens-independent ocular phenotypes in zebrafish. Dev Biol. 2006, 299 (1): 63-77. 10.1016/j.ydbio.2006.07.005. Epub 2006 Jul 12.CrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.