A common theme for LIM homeobox gene function across phylogeny?
The genome of the nematode Caenorhabditis elegans is almost completely sequenced, thus allowing the analysis of complete gene families in a metazoan organism. C. elegans contains seven LIM homeobox genes. Almost all C. elegans LIM homeobox genes fall into subclasses that are defined by the presence of similar genes from arthropods and vertebrates, suggesting a common origin for different subclasses of LIM homeobox genes ([ILLUSTRATION FOR FIGURE 1 OMITTED]; C. elegans proteins are underlined).
Function of the C. elegans ttx-3 and lin-11 homeobox genes
We recently described the function of two C. elegans LIM homeobox genes, ttx-3 and lin-11, in a neural circuit subserving thermoregulatory behavior (5, 6, 7). The neural pathway for thermotaxis includes the sensory neuron AFD and the connected interneurons AIY and AIZ (Ref. 5; [ILLUSTRATION FOR FIGURE 2 OMITTED]). The ttx-3 null mutation causes the same behavioral defect as laser ablation of AIY, implying that AIY does not signal in this mutant (5). A ttx-3-GFP reporter construct shows that ttx-3 is expressed exclusively in the AIY interneuron pair (6). AIY is generated in ttx-3 mutants, arguing that no fundamental changes in cell fate have taken place. However, AIY exhibits abnormal axonal projections, manifested mainly by the outgrowth of additional small neurites. These defects could be due to misregulation of ttx-3 downstream target genes involved directly in axonal pathfinding, or they could be due to misregulation of ttx-3 downstream target genes involved in synaptic signaling, which could, as a secondary consequence, cause axonal sprouting defects.
ttx-3 is continuously expressed in AIY from mid-embryogenesis throughout adulthood and is required to maintain its own expression, suggesting that ttx-3 may also act in a neural maintenance pathway for AIY. Considering that thermotactic behavior manifests a simple learning and memory task, AIY represents a prime candidate for an interneuron that integrates and memorizes sensory inputs, for example by variable patterns of synaptic connections. We consider the possibility that ttx-3 is part of an autoregulatory loop that regulates the initial expression of downstream target genes involved in neural signaling and that may also modulate downstream gene expression in behavioral plasticity (6).
We have identified a second LIM homeobox gene, lin-11, that is expressed and functions in the opposing interneuron of the thermoregulatory circuit, AIZ (7). lin-11 null mutant animals display cryophilic defects that phenocopy laser ablation of the AIZ interneuron. Although the lin-11 expressing neurons, including AIZ, are formed in lin-11 null mutant animals, they display neuroanatomical defects, comparable to those neural defects observed in ttx-3 mutant animals. Like ttx-3, lin-11 expression is also maintained in postmitotic neurons throughout adulthood. Thus, distinct LIM homeobox genes specify two functionally related antagonistic interneurons within a neural network dedicated for thermoregulatory processes [ILLUSTRATION FOR FIGURE 2 OMITTED].
How are thermoregulatory neural centers organized in more complex organisms? And is there any evidence for a conserved role for ttx-3 and lin-11 in the control of these neural centers? In fact, the organization of the C. elegans thermoregulatory network into two parallel, warm- and cold-processing pathways is remarkably similar to thermocontrol in vertebrates. The major thermoregulatory organ of vertebrates, the hypothalamus, contains distinguishable warm- and cold-sensing temperature processing units (8) that may be homologous to the antagonistic high and low temperature sensing pathways of the C. elegans thermotactic response pathway [ILLUSTRATION FOR FIGURE 2 OMITTED]. The vertebrate ttx-3 homolog Lhx2 and the lin-11 homolog Lhx1 are indeed expressed in the diencephalon, which gives rise to the thermoregulatory hypothalamus (9, 10). lin-11 and ttx-3 in C. elegans, and their homologs in mammals, may thus play a similar role in the development of two components of these related thermal processing networks.
Apart from their suggested role in the hypothalamus, the vertebrate lin-11 and ttx-3 homologs Lhx1 and Lhx2 are expressed in several additional places in the nervous system (9, 10). The additional roles of the vertebrate genes might parallel the function of the nematode homologs, making additional cases for a conservation of function throughout evolution. For example, lin-11 is expressed and functions in the ventral nerve cord of C. elegans, where it is required for correct axon bundle fasciculation (7); vertebrate Lhx1 is similarly expressed in motor neurons of the spinal cord. Additionally, Lhx1 expression can be observed in sensory structures in the brain (9), which correlates with lin-11 expression in C. elegans head sensory neurons (7). In contrast, the comparison of expression and functions of nematode lin-11 and vertebrate Lhx1 also makes a very strong point for the acquisition of additional functions for a regulatory gene (or, alternatively, the loss of a function): While Lhx1 is involved early
in embryogenesis in neural induction (11), no such embryonic role exists for lin-11 (7). Similarly, as C. elegans has no appendages, the function of apterous, the Drosophila homolog of C. elegans ttx-3 in wing patterning, represents a clear case of co-option of a regulatory gene to a new developmental process.
Is there a common theme for LIM homeobox gene function in C. elegans? The functional analysis of the LIM homeodomain-encoding ttx-3, lin-11, and mec-3 genes, all of which act late in neural development, demonstrated their role in determining the differentiated neural phenotype (6, 7, 12). To learn whether the other C. elegans LIM homeobox genes might share a similar role, we examined their expression pattern using GFP reporter gene fusion. We found lim-4, lim-6, and lim-7 to be expressed in a non-overlapping subset of neuronal cells. While the expression of the isl-homolog lim-7 is very dynamic and not confined to the nervous system, we found lim-4 and lim-6 to be exclusively expressed in a non-overlapping set of head sensory-, inter- and motorneurons. Note that, like mec-3, ttx-3 and lin-11, lim-4 and lim-6 are also expressed in neurons after their final division and continue to be expressed throughout adulthood, suggesting that they might be involved in neuronal maintenance. We speculate that a common theme of C. elegans LIM homeobox genes is to determine a specific neural phenotype, as manifested perhaps by a specific neural connectivity or neurotransmitter choice. Our findings suggest that this is the phylogenically conserved function of LIM homeobox genes, and that some of the functions of the genes in C. elegans - such as the role of lin-11 in vulval development - represent a later recruitment of these genes into additional cellular processes.
A comparison of expression characteristics of the C. elegans LIM homeobox genes leads to another interesting point: the expression of most, if not all of these genes is maintained in neural tissues throughout adulthood. This suggests a nontransient, but constitutive requirement for these genes throughout the life of the neuron, e.g., in the maintenance of specific neural features.
We further propose that LIM homeobox gene function in neural development represents a function of these genes that has been conserved across phylogeny. This hypothesis is based on the functioning of Drosophila LIM homeobox genes in axon pathfinding and determination of neurotransmitter identity (3, 4), as well as the maintained neural expression of vertebrate LIM homeobox genes in postmitotic neurons (1). To our knowledge, LIM homeobox genes have so far been found exclusively in organisms that contain a nervous system, which provides some circumstantial evidence that LIM homeobox genes might have co-evolved with neural structures, whose complexity obviously requires the use of new classes of regulatory genes.
LIM homeobox genes presumably arose in evolution by a recombination event of homeodomain and LIM domain coding exons. This event probably happened only once, since (1) all LIM homeobox genes contain a very similar architecture, with two LIM domains at the N-terminus and one homeodomain at the C-terminus, and since (2) the first LIM domain of LIM homeodomain proteins is usually more similar to the first LIM domain of other LIM homeodomain protein than to their own second LIM domain (1). Gene duplications of a single common ancestor conceivably created the different subclasses of LIM homeodomain proteins; these duplication must have happened before the divergence of nematodes, arthropods, and chordates. This common ancestor, which contained multiple LIM homeobox genes, might have already contained a simple nervous system in which LIM homeodomain protein were employed to define specific neural features.
As mentioned above, LIM homeobox genes have obviously been recruited to function in additional non-neural processes, such as vulval patterning, limb development, and neural induction during gastrulation. These additional and relatively specialized functions of LIM homeobox genes in organs and processes specific for distinct phylogenetic branches presumably have been co-opted by specific phyla at later stages of evolution.
This work was supported in part by Hoechst AG to G. R. and by a postdoctoral fellowship from the Human Frontiers Science Program to O. H.
1. Dawid, I.B., R. Toyama, and M. Taira. 1995. LIM domain proteins. C. R. Acad. Sci. Paris/Life Sci. 318: 295-306.
2. Tsuchida, T., M. Ensini, S. B. Morton, M. Baldassare, T. Edlund, T. M. Jessell, and S. L. Pfaff. 1994. Topographic organization of embryonic motor neurons defined by expression of LIM homeobox genes. Cell 79: 957-970.
3. Lundgren, S.E., C.A. Callahan, S. Thor, and J.B. Thomas. 1995. Control of neuronal pathway selection by the Drosophila LIM homeodomain gene apterous. Development 121: 1769-1773.
4. Thor, S., and J.B. Thomas. 1997. The Drosophila islet gene governs axon pathfinding and neuotransmitter identity. Neuron 18: 397-409.
5. Mori, I., and Y. Ohshima. 1995. Neural regulation of thermotaxis in Caenorhabditis elegans. Nature 376: 344-348.
6. Hobert, O., I. Mori, Y. Yamashita, H. Honda, Y. Ohshima, Y. Liu, and G. Ruvkun. 1997. Regulation of interneuron function in the C. elegans thermoregulatory pathway by the ttx-3 LIM homeobox gene. Neuron 19: 345-357.
7. Hobert, O., T. d'Albert, Y. Liu, and G. Ruvkun. 1998. Control of neural development and function in a thermoregulatory network by the LIM homeobox gene lin-11. J. Neurosci. 18: 2084-2096.
8. Boulant, J. A., and J. B. Dean. 1986. Temperature receptors in the central nervous system. Annu. Rev. Physiol. 48: 639-654.
9. Fuji, T., J. G. Pichel, M. Taira, R. Toyama, I. B. Dawid, and H. Westphal. 1994. Expressions patterns of the murine LIM class homeobox gene lim1 in the developing brain and excretory system. Der. Dynam. 199: 73-83.
10. Porter, F. D., J. Drago, Y. Xu, S. Cheema, S. P. Huang, E. Lee, A. Grinberg, J. S. Massalas, D. Bodine, F. W. Alt, and H. Westphal. 1997. Lhx2, a LIM homeobox gene, is required for eye, forebrain and definitive erythrocyte development. Development 124: 2935-2944.
11. Shawlot, W., and R. R. Behringer. 1995. Requirement for Lim1 in head-organizer function. Nature 374: 425-430.
12. Way, J.C., and M. Chalfie. 1988. mec-3, a homeobox-containing gene that specifies differentiation of the touch receptor neurons in C. elegans. Cell 54: 5-16.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Genetic Regulatory Networks in Embryogenesis and Evolution|
|Author:||Hobert, Oliver; Ruvkun, Gary|
|Publication:||The Biological Bulletin|
|Date:||Dec 1, 1998|
|Previous Article:||Heterochronic genes in development and evolution.|
|Next Article:||Mechanisms of specification in ascidian embryos.|