spacer gif spacer gif spacer gif spacer gif spacer gif
 QUICK SEARCH:   [advanced]


spacer gif
     Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents    


This Article
Right arrow Summary Freely available
Right arrow Full Text
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Altun-Gultekin, Z.
Right arrow Articles by Hobert, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Altun-Gultekin, Z.
Right arrow Articles by Hobert, O.

A regulatory cascade of three homeobox genes, ceh-10, ttx-3 and ceh-23, controls cell fate specification of a defined interneuron class in C. elegans

Zeynep Altun-Gultekin1, Yoshiki Andachi2, Ephraim L. Tsalik1, David Pilgrim3, Yuji Kohara2 and Oliver Hobert1,*

1 Department of Biochemistry and Molecular Biophysics, Center for Neurobiology and Behavior, Columbia University, College of Physicians and Surgeons, New York, NY 10032, USA
2 Genome Biology Lab, Center for Genetic Resource Information, National Institute of Genetics, Mishima, Shizuoka, 411-8540, Japan
3 Department of Biological Sciences, University of Alberta, Edmonton, Alberta T6G 2E9, Canada



View larger version (74K):

[in a new window]
 
Fig. 1. Expression of homeobox genes in the AIY interneuron class. (A) Overlap of a DIC and a fluorescent image showing expression of a ttx-3full::gfp reporter gene construct (schematically shown in part B) in the nucleus of several head neurons of an early L2 stage animal. The fluorescent images were taken in several optical sections and layered upon each other. In this animal (genotype: otEx57) comparably strong expression can be seen in AIY, AIA, ADL and ASI. Within a population, 41/41 otEx57 animals show strong expression in AIY and 39/41 show expression in AIA, 28/41 in ADL and 25/41 in ASI. Similar patterns of expression can be seen with a total of 12 independently obtained extrachromosomal arrays (otEx45-otEx49, otEx57 and otEx97-otEx101). Pharyn. n., pharyngeal neuron. (B) Schematic representation of the expression constructs used. ttx-3resc rescues the thermotaxis phenotype of ttx-3 mutant animals (Hobert et al., 1997), ttx-3prom::gfp reveals postembryonic expression exclusively in the AIY interneuron class (Hobert et al., 1997), ttx-3full::gfp expression is shown in A. (C) Schematic representation of the expression of ceh-10::lacZ (Svendsen and McGhee, 1995), ttx-3full::gfp (this study) and ceh-23::gfp (Forrester et al., 1998) reporter constructs in the C. elegans nervous system. Expression in neurons of the tail ganglia is not shown. The name of neurons in which expression of any of the three genes overlaps is bold and underlined.

 


View larger version (25K):

[in a new window]
 
Fig. 2. ceh-23 and ttx-3 sequences and alleles. (A) Relationship of the CEH-23 homeodomain to its most closely related homeodomains. Although the CEH-23 homeodomain bears a strong affinity to the distal-less and empty spiracles protein families, other C. elegans proteins represent the true Dll and ems orthologs. CEH-23 appears to have no ortholog in other species, suggesting that CEH-23 may have arisen by a gene duplication event in the nematode lineage. C. elegans proteins are bold and underlined. Homeodomains that show the highest similarity to the CEH-23 homeodomain were identified by BLAST searches (including searches of the 11/23/00 release of the C. elegans genome, the 11/6/2000 release of the human genome sequence and the 9/16/2000 release of the Drosophila genome sequence) and assembled into a phylogenetic tree using the pileup/distances/growtree algorithms of the GCG package with default parameters. The tree is rooted with LIM and POU homeodomain proteins (not shown). (B) ceh-23(ms23) deletion allele. (C) ttx-3 mutant alleles. The upper panel shows the schematic localization of the nucleotide changes in ttx-3 alleles. The lower panel shows the corresponding amino acid changes in the homeodomain. mg158 contains a missense mutation in one of the most highly conserved residues of the homeodomain, F49 (homeodomain numbering), which appears to be indispensable for the structural integrity of homeodomains (Gehring et al., 1994). nj14 was provided by Ikue Mori; due to its molecular similarity to ot22, it was not further characterized.

 


View larger version (57K):

[in a new window]
 
Fig. 3. Phenotypic characterization of ttx-3 and ceh-23 alleles. (A) Thermotaxis assays (see Materials and Methods). Animals were grown at 20°C. Number of assays per genotype: wild type N2=11; ceh-23(ms23)=8; ttx-3(ks5)=6; ttx-3(mg158)=6; ttx-3(ot22)=7; ttx-3(ot23)=8. For each genotype, an average of 284-417 animals were tested per assay. The error bars represent the standard error of the mean. The difference in cryophilic behavior between wild type and ceh-23(ms23) is not statistically significantly different (P>0.3; paired and unpaired Student’s t-test). Our observation of a certain fraction of wild-type animals showing cryophilic behavior is consistent with previous reports (Hedgecock and Russell, 1975; Mori and Ohshima, 1995). (B) Effect of ttx-3 alleles on the level of ttx-3prom::gfp expression in AIY (white arrows point to AIYL/R in different animals). All animals shown contain the integrated ttx-3prom::gfp array mgIs18, which was crossed into the respective mutant backgrounds. In ks5 mutant animals signal strength of mgIs18 is clearly reduced, yet enough freely diffusible GFP protein is made in the cell to allow visualization of the axons. In contrast, mgIs18 expression is mostly undetectable in mg158, ot22, and ot23 mutants. (C) Quantification of the defects shown in B. ‘Dim’ refers to clearly reduced gfp expression with the axons being barely visible, ‘very dim’ to gfp expression that is insufficient to visualize the axons. (D) Dauer assays (see Materials and Methods). Results are from four experiments.

 


View larger version (56K):

[in a new window]
 
Fig. 4. Regulation of homeobox gene expression in AIY. (A) ttx-3prom::gfp expression is absent in ceh-10(gm58) animals. Homozygous ceh-10(gm58) offspring of ceh-10(gm58)/+; mgIs18 hermaphrodites arrest as L1 larvae and were identified based on their clr phenotype. Expression was never observed (n>50). (B) ceh-23::gfp expression is dependent on ttx-3. Adult ttx-3(mg158); kyIs5 were scored. Quantification is shown in Table 1. (C) ttx-3prom::gfp expression is unaffected in ceh-23(ms23) null mutant animals (n>50). Note the morphological integrity of the AIY neuron in terms of its cell position and axon morphology.

 


View larger version (35K):

[in a new window]
 
Fig. 5. Expression of the AIY cell fate marker sra-11::gfp in ttx-3 and ceh-23 mutant animals. sra-11::gfp expression was monitored by mating otIs62 animals that harbor an integrated sra-11::gfp construct with animals of the respective mutant backgrounds. The animals carry a rol-6 injection marker, leading to a slight distortion in the spatial arrangement of neurons. Quantification of the expression data in the AIY interneurons is shown in Table 1. While expression of sra-11::gfp in AVB is strong and highly penetrant in L1 larvae, it is observed less consistently and more weakly in adult animals (compare upper and lower panels; this observation was made with three independently created integrants, otIs62, otIs122 and otIs123). Expression of sra-11::gfp in AVB came to lie out of the plane of focus in the ceh-23(ms23); otIs62 adult animal shown in the lower right panel. Expression of sra-11::gfp in AIA is difficult to consistently detect in larvae, yet clearly visible in adult animals. 64% of wild-type adults show expression (n=45) in AIA and 38% (n=70) of ttx-3(mg158); otIs62 adults show expression in AIA (see lower middle panel).

 


View larger version (81K):

[in a new window]
 
Fig. 6. Examination of pan-neuronal cell fate in ttx-3 mutants. (A) Expression of three pan-neuronal markers, unc-119::gfp (otIs45), unc-33::gfp (otEx75) and F25B3.3::gfp (evIs111), in wild-type L1 larvae (see also Materials and Methods). (B) F25B3.3::GFP is expressed in postmitotic neurons. Three evIs111 embryos are shown in a fluorescence micrograph (left panel) and a corresponding DIC micrograph (right panel); the embryos are at the threefold stage (>500 minutes; upper left), the 1.5-fold stage (>400 minutes; upper right) and approximately the 350 minute stage (lower right). Most embryonically generated neurons have been born by the 350 minute stage (Sulston et al., 1983). (C) Expression of pan-neuronal markers in ttx-3(mg158) animals. Fluorescence micrographs and the corresponding DIC images are shown side by side. The three white arrows point to the characteristic row of three neurons behind the excretory cell, AIM, AIY and AVK. Animals are late larvae/young adults. 100% ttx-3(mg158); evIs111 animals showed expression in AIY (n=25; large white arrow). Expression of otIs45 and otEx75 is mosaic in wild-type animals. 90% of otIs45 animals show gfp expression in AIY (n=39); in ttx-3(mg158);otIs45, 88% of animals show gfp expression in AIY (n=48). In otEx75, 62% of animals show expression in AIY (n=24); in ttx-3(mg158); otEx75, 53% of animals show expression in AIY (n=28).

 


View larger version (49K):

[in a new window]
 
Fig. 7. Effects of ectopic ttx-3 expression and dependence on ceh-10 activity. (A,B) RID and CAN can adopt AIY-like features upon ectopic expression of ttx-3. (A) A transgenic animal is shown that expresses ttx-3prom::gfp (normally only expressed in AIY; see Fig. 3B) and unc-119::ttx-3 from the independently integrated arrays mgIs18 and otIs97IV, respectively, in a ttx-3(ks5) background. Note that the autoregulatory defect of ttx-3prom::gfp in AIY is completely rescued and that ectopic expression is induced in RID and CAN. The animal is slightly twisted owing to the presence of a rol-6 injection marker. Ectopic RID and CAN expression could be observed with several independent transgenic lines (see Materials and Methods). (B) Ectopic expression of ceh-23::gfp in RID upon pan-neuronal ttx-3 expression. The genotype of this strain is kyIs5IV; otEx65. Eleven out of 14 examined adults show expression in RID. kyIs5 animals never show GFP expression in RID (n>20). (C) ceh-10 dependence of ectopic ttx-3prom::gfp expression. Homozygous ceh-10(gm58) offspring of ceh-10(gm58)/+; otIs97mgIs18IV hermaphrodites arrest as L1 larvae and were identified based on their clr phenotype. In these animals, no GFP signal can be detected in RID, CAN, or AIY (the approximate position of the AIY interneurons is indicated with an arrow), showing that expression of ttx-3 from an exogenous promoter can not rescue the ceh-10(gm58) phenotype. White dots derive from autofluorescent gut granules. (D) Summary of expression of several AIY cell fate markers in RID and CAN either in wild-type animals or in animals that pan-neuronally express ttx-3. Ectopic sra-11::gfp expression in RID was observed in 19/19 animals (data not shown) and was never observed in wild-type animals (n>20).

 


View larger version (82K):

[in a new window]
 
Fig. 8. Characterization of the neuroanatomical defects in ttx-3 mutant animals. (A) Axonal defects observed in various mutants. The wild-type AIY axon is shown in black. In wild-type animals, a short process (shown in dark gray) can occasionally be observed in the region where the main axon turns dorsally to enter the nerve ring; those are not scored as a mutant phenotype, unless they are longer than 10 µm (shown as light gray addition to the dark gray line). Schematic examples of aberrant axon sprouts are in light gray. We set the minimum length for scoring aberrant sprouts that emanate from the axon arbitrarily at 2 µm. As cell bodies occasionally narrow into a thin ending, we arbitrarily set the criteria to score sprouts from the cell body more stringently at 4 µm. Occasional ‘splits’ of the axon in the nerve ring region are counted as sprouts. Small enlargements along the axon, rarely observed in wild-type animals, often observed in pathfinding mutants, are not scored as sprouts. Every case in which the main axon does not meet the axon of its contralateral homolog at the dorsal midline is classified as ‘short stop’; the main axon stalls either at the turning point into the nerve ring (but never before that point), or at various points below or above the dorsoventral midline (black arrows). (B) Characteristic examples of axonal defects observed in ttx-3 mutants. AIY is visualized with mgIs18. Quantitative data are shown in Table 3. Similar results have also been obtained with several extrachromosomal ttx-3prom::gfp lines (Hobert et al., 1997). For this analysis we could make use only of the ttx-3(ks5) allele because, owing to the more severe reduction of ttx-3prom::gfp expression in the stronger ttx-3 alleles, the axon of the AIY interneurons cannot be visualized. It is thus possible that in the stronger alleles AIY neuroanatomy is even more strongly affected. (C) Representative examples of AIY neuroanatomy in unc-33(e120), unc-73(e936) and unc-119(e2498) mutant animals, visualized with mgIs18. White triangles point to short stops, white arrows to sprouts. Animals shown in this figure are late larvae and young adults. Notably, besides the short-stop defect of the main axon of the AIY neurons, these axon pathfinding mutants also cause axon sprouting defects. As neural activity defects do not cause axon sprouting in AIY (Table 3), the axon sprouting observed in pathfinding mutants are unlikely to be a secondary consequence of aberrant connectivity caused by aberrant pathfinding. We rather hypothesize that the axon sprouting defect observed in pathfinding mutants as well as in ttx-3 mutants reflects an inability of the neuron to initiate and maintain the proper outgrowth of a single, monopolar axon.

 


View larger version (17K):

[in a new window]
 
Fig. 9. Model for transcriptional regulatory events in the AIY interneurons. We propose that CEH-10 activates the expression of the TTX-3 protein, which then – either in conjunction with CEH-10 or other as yet unknown transcription factors – activates the expression of ceh-23 and other target genes. TTX-3 and CEH-23 presumably both contribute to the regulation of sra-11 expression. Pan-neuronal features are determined either upstream or in parallel to the ceh-10->ttx-3->ceh-23 cascade. Regulatory events that determine pan-neuronal and cell-type specific features of AIY (via ceh-10/ttx-3/ceh-23) are unknown and are denoted by a question mark. Based on their prominent roles in neuronal patterning in several systems, we tested whether the C. elegans Pax-6/Eyeless homolog vab-3 (Chisholm and Horvitz, 1995), the C. elegans homolog of MATH1/Atonal, lin-32 (Zhao and Emmons, 1995) and the C. elegans NeuroD homolog cnd-1 (Hallam et al., 2000) couple early embryonic patterning events to the transcriptional regulatory cascade of ceh-10, ttx-3 and ceh-23 in AIY and found this not to be the case (data not shown). It is conceivable that some of the TTX-3 target genes shown here (ser-2, sra-11, C36B7.7, kal-1) are directly involved in axonal defects observed in ttx-3(ks5) mutants; this possibility is denoted by a broken arrow. Isolation of mutations in these genes will address this issue. Although unc-17 hypomorphic mutants do not show axonal defects, it is possible that cholinergic transmission contributes together with other genes to axogenesis of AIY (Table 3). At this point it is not clear whether any of the regulatory events that involve CEH-10, TTX-3 and CEH-23 depend on direct interaction of the respective homeodomain protein with the promoter of the presumptive target gene. Within the ttx-3 promoter we have defined two motifs of 5bp and 7bp length, spaced by 20 nucleotides, which are both necessary for the activity of the promoter; both elements contain a variant of the core homeodomain binding site (A. Wenick and O. H., unpublished). Similar clustered motifs can be found in other presumptive TTX-3 target gene promoters. We speculate that these motifs may represent binding sites for TTX-3 and possible co-factor(s). ceh-10 autoregulation has been demonstrated by Forrester et al., 1998.

 





© The Company of Biologists Ltd 2001