This Title All WIREs
How to cite this WIREs title:
WIREs Dev Biol
Impact Factor: 5.814

Brn3/POU‐IV‐type POU homeobox genes—Paradigmatic regulators of neuronal identity across phylogeny

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract One approach to understand the construction of complex systems is to investigate whether there are simple design principles that are commonly used in building such a system. In the context of nervous system development, one may ask whether the generation of its highly diverse sets of constituents, that is, distinct neuronal cell types, relies on genetic mechanisms that share specific common features. Specifically, are there common patterns in the function of regulatory genes across different neuron types and are those regulatory mechanisms not only used in different parts of one nervous system, but are they conserved across animal phylogeny? We address these questions here by focusing on one specific, highly conserved and well‐studied regulatory factor, the POU homeodomain transcription factor UNC‐86. Work over the last 30 years has revealed a common and paradigmatic theme of unc‐86 function throughout most of the neuron types in which Caenorhabditis elegans unc‐86 is expressed. Apart from its role in preventing lineage reiterations during development, UNC‐86 operates in combination with distinct partner proteins to initiate and maintain terminal differentiation programs, by coregulating a vast array of functionally distinct identity determinants of specific neuron types. Mouse orthologs of unc‐86, the Brn3 genes, have been shown to fulfill a similar function in initiating and maintaining neuronal identity in specific parts of the mouse brain and similar functions appear to be carried out by the sole Drosophila ortholog, Acj6. The terminal selector function of UNC‐86 in many different neuron types provides a paradigm for neuronal identity regulation across phylogeny. This article is categorized under: Gene Expression and Transcriptional Hierarchies > Regulatory Mechanisms Invertebrate Organogenesis > Worms Nervous System Development > Vertebrates: Regional Development
Lineage transformation phenotypes of unc‐86 mutants. Reprinted with permission from Chalfie et al. (). Copyright 1981 Cell Press/Elsevier; Chalfie and Sulston (). Copyright 1981 Elsevier; Finney (). Copyright 1987 Thesis at MIT
[ Normal View | Magnified View ]
Brn3/POU‐IV‐type binding motifs. (a) C. elegans UNC‐86 binding motif defined by PBM (Weirauch et al., ). UNC‐86 also binds the POU4F1 motif (see (d)) which is considered the best inferred dimeric binding motif (Narasimhan et al., ). (b) Drosophila melanogaster ACJ6 binding motif defined by PBM (Weirauch et al., ). (c) Mus musculus BRN3C binding motif defined by PBM (Berger et al., ). BRN3A and BRN3B have no direct experimental binding motifs available but the best inferred motifs are the human POU4F1 and POU4F2 motifs (see 5D) having identical DNA binding domain identity scores (see (e)). (d) Homo sapiens POU4F1, POU4F2, POU4F3 binding motif defined by SELEX (Jolma et al., ). (e) Heatmap of DNA binding domain identities among POU‐IV family homologs. DNA binding domain (DBD) identity scores (Weirauch et al., ) suggest POU‐IV family proteins have similar binding specificities. All DBD identity scores shown are above the POU‐IV family DBD confidence threshold suggesting common sequence preferences. Above this threshold, binding motifs can be assumed to be identical between two proteins. A DBD identity score of 1.0 reflects identical DBD amino acid sequence between homologs. Positions with no score shown in the heatmap did not have DBD identity scores above the family confidence threshold and hence have diverged DNA binding specificities. The mouse BRN3A, BRN3B, and BRN3C proteins have nearly identical DNA binding domains (DBD identity score = 1.0) with the human POU4F1, POU4F2, and POU4F3 proteins respectively. The UNC‐86 POU‐IV DBD is highly similar to fly (ACJ6), mouse (BR3A/B), and human (POU4F1/2/3) DBDs
[ Normal View | Magnified View ]
Different modes of UNC‐86 function in different neuron types and under different conditions. (a) This schematic encapsulates a number of core principles of UNC‐86 function. First, UNC‐86 directly controls the expression of distinct batteries of effector genes in distinct cell types; these effectors genes encode for nuts and bolts proteins, that define terminal identity features of a given neuron type (ion channels, neurotransmitters, neuropeptides, etc.). Second, the ability of unc‐86 to turn on distinct effector batteries in distinct cell types is enabled by UNC‐86's pairing with different partner TFs (X, Y) in different neuron types. These distinct pairings endow the respective proteins to be directed to distinct cis‐regulatory elements (colored semi‐circles imposed on effector gene loci). In one case (UNC‐86/MEC‐3) this pairing is based on a direct physical interaction between the proteins (Rohrig, Rockelein, Donhauser, & Baumeister, ; Xue et al., ), in other cases the nature of the interaction is not presently clear. Third, UNC‐86‐dependent effector genes that are shared between distinct neuron types—for example, the VGLUT gene eat‐4, are independently regulated by distinct UNC‐86 complexes. This is inferred from a cis‐regulatory dissection not only of the eat‐4/VGLUT (Serrano‐Saiz et al., ), but also the unc‐17/VAChT and cho‐1/CHT locus (which define cholinergic identity). In each of these cases distinct UNC‐86 responsive elements were identified for different cell types in these loci. However, these binding sites do not necessarily have to be distinct. There is also evidence that deletion of a single UNC‐86 binding site can abrogate unc‐86‐dependent effector gene expression in multiple distinct neuron types, suggesting that UNC‐86, bound to a specific site may collaborate with distinct partners. Panneuronally expressed genes are regulated independently of unc‐86, demonstrating that like many other terminal selectors unc‐86 “individuates” neurons, but does not determine whether a cell becomes a neuron or not. (b,c) Conceptually similar mechanisms may be used to modulate UNC‐86‐mediated differentiation events. (b) Closely related neuron classes that share the expression of many genes can be diversified through coactivators or corepressors that modulate the ability of UNC‐86 to turn on specific subsets of genes. For example, the embryonically generated ALM and PLM neurons, which share a large amount of unc‐86/mec‐3‐dependent genes employ HOX genes, for example, the PLM‐expressed EGL‐5 Hox gene to help unc‐86/mec‐3 to activate PLM‐specific genes and antagonize unc‐86/mec‐3 dependent genes. Conversely, harsh and light touch receptor neurons, which both require UNC‐86 and MEC‐3 for their differentiation, differentially express repressor proteins that antagonize the ability of UNC‐86/MEC‐3 to turn on specific target genes. See text for references. Whether these coactivating and corepressing mechanisms are indeed integrated on the level of cis‐regulatory control elements, as indicated here, is speculative. (c) unc‐86 is not only required for genetically hardwired expression of effector genes, but is also required for the expression of genes whose expression is dynamically controlled. This includes genes that are turn on or turned off in a sex‐specific manner during sexual maturation; in these cases unc‐86 function is either promoted by male‐specific factors (e.g., male specific DMD‐3 TF in PHC or LIN‐29 in AIM) or antagonized by the hermaphrodite‐specific TRA‐1 master regulator of sexual identity. See text for references. A number of innexin genes are also turned on or off in unc‐86‐dependent neuron classes upon entry into the dauer stage (Bhattacharya, Aghayeva, Berghoff, & Hobert, ). It is not presently clear if and how unc‐86 and dauer‐specific TFs, such as DAF‐16/FoxO are involved in controlling dynamic innexin expression, but one possibility is that in analogy to the sex‐specific cases, DAF‐16/FoxO may promote or antagonize the activity of UNC‐86 on innexin target genes. Whether these coactivating and corepressing mechanisms are indeed integrated on the level of cis‐regulatory control elements, as indicated here, is speculative
[ Normal View | Magnified View ]
unc‐86 expression and phenotype. (a) UNC‐86 expression matched onto lineage diagram. Reprinted with permission from Finney and Ruvkun (). Copyright 1990 CellPress/Elsevier using expression data from Finney and Ruvkun () and Serrano‐Saiz et al. (). Thickened line indicates UNC‐86 expression and its onset. (b) UNC‐86 expression matched onto worm. Color coding in (c,d) indicates sensory (red), inter‐ (blue) and motorneuron (green). Reprinted with permission from (Baumeister et al. (). Copyright 1996 Cold Spring Harbor Laboratory Press. (c) UNC‐86 expression matched onto a representation of the connectome that reflects information flow from sensory neurons (top) to motor neurons (bottom) (Varshney, Chen, Paniagua, Hall, & Chklovskii, ). (d) Schematized summary of lineage defects observed in unc‐86 mutants. See text for references. Thickened line indicates UNC‐86 expression and its onset. (e) Examples of expression of several UNC‐86 collaborators of the LIM homeodomain class and their expression pattern. See text for references
[ Normal View | Magnified View ]
POU homeodomain protein families with representative in C. elegans. (a) POU homeodomain proteins are composed of two domains, the POUS and POUHD domain, connected by a flexible linker sequence, as initially determined by (Klemm, Rould, Aurora, Herr, & Pabo, ). Reprinted with permission from Malik, Zimmer, and Jauch (). Copyright 2018 Springer. (b) There are six POU domain subfamilies, the most divergent of which (POU‐VI; no C. elegans representative) is not shown in this tree. Reprinted with permission from Serrano‐Saiz, Leyva‐Diaz, De La Cruz, and Hobert (). Copyright 2018 CellPress/Elsevier. (c) Phylogenetic distribution of POU genes. Reprinted with permission from Gold, Gates, and Jacobs (). Copyright 2014 Oxford University Press
[ Normal View | Magnified View ]

Browse by Topic

Nervous System Development > Vertebrates: Regional Development
Nervous System Development > Worms
Gene Expression and Transcriptional Hierarchies > Regulatory Mechanisms