The molecular genetic basis of positional information in insect segments

Canalization and control in automata networks: body segmentation in Drosophila melanogaster

We present schema redescription as a methodology to characterize canalization in automata networks used to model biochemical regulation and signalling. In our formulation, canalization becomes synonymous with redundancy present in the logic of automata. This results in straightforward measures to quantify canalization in an automaton (micro-level), which is in turn integrated into a highly scalable framework to characterize the collective dynamics of large-scale automata networks (macro-level). This way, our approach provides a method to link micro- to macro-level dynamics--a crux of complexity. Several new results ensue from this methodology: uncovering of dynamical modularity (modules in the dynamics rather than in the structure of networks), identification of minimal conditions and critical nodes to control the convergence to attractors, simulation of dynamical behaviour from incomplete information about initial conditions, and measures of macro-level canalization and robustness to perturbations. We exemplify our methodology with a well-known model of the intra- and inter cellular genetic regulation of body segmentation in Drosophila melanogaster. We use this model to show that our analysis does not contradict any previous findings. But we also obtain new knowledge about its behaviour: a better understanding of the size of its wild-type attractor basin (larger than previously thought), the identification of novel minimal conditions and critical nodes that control wild-type behaviour, and the resilience of these to stochastic interventions. Our methodology is applicable to any complex network that can be modelled using automata, but we focus on biochemical regulation and signalling, towards a better understanding of the (decentralized) control that orchestrates cellular activity--with the ultimate goal of explaining how do cells and tissues 'compute'.

Geometry and topology of parameter space: investigating measures of robustness in regulatory networks

The concept of robustness of regulatory networks has been closely related to the nature of the interactions among genes, and the capability of pattern maintenance or reproducibility. Defining this robustness property is a challenging task, but mathematical models have often associated it to the volume of the space of admissible parameters. Not only the volume of the space but also its topology and geometry contain information on essential aspects of the network, including feasible pathways, switching between two parallel pathways or distinct/disconnected active regions of parameters. A method is presented here to characterize the space of admissible parameters, by writing it as a semi-algebraic set, and then theoretically analyzing its topology and geometry, as well as volume. This method provides a more objective and complete measure of the robustness of a developmental module. As a detailed case study, the segment polarity gene network is analyzed.

Methods of robustness analysis for Boolean models of gene control networks

As a discrete approach to genetic regulatory networks, Boolean models provide an essential qualitative description of the structure of interactions among genes and proteins. Boolean models generally assume only two possible states (expressed or not expressed) for each gene or protein in the network, as well as a high level of synchronisation among the various regulatory processes. Two possible methods of adapting qualitative models to incorporate the continuous-time character of regulatory networks, are discussed and compared. The first method consists of introducing asynchronous updates in the Boolean model. In the second method, the approach introduced by Glass is adopted to obtain a set of piecewise linear differential equations that continuously describe the states of each gene or protein in the network. Both methods are applied to a Boolean model of the segment polarity gene network of Drosophila melanogaster. The dynamics of the model is analysed, and a theoretical characterisation of the model's gene pattern prediction is provided as a function of the timescales of the various processes.

Allocation and specification of the genital disc precursor cells in Drosophila

The adult structures of Drosophila melanogaster are derived from larval imaginal discs, which originate as clusters of cells within the embryonic ectoderm. The genital imaginal disc is composed of three primordia (female genital, male genital, and anal primordia) that originate from the embryonic tail segments A8, A9, and A10, respectively, and produce the sexually dimorphic genitalia and analia. We show that the genital disc precursor cells (GDPCs) are first detectable during mid-embryogenesis as a 22-cell cluster in the ventral epidermis. Analysis of mutant and double mutant phenotypes of embryonic patterning genes in the GDPCs, together with their expression patterns in these cells, revealed the following with respect to the origins and specification of the GDPCs. The allocation of the GDPCs from the ventral epidermis requires the function of ventral patterning genes, including the EGF receptor and the spitz group of genes. The ventral localization of the GDPCs is further restricted by the action of dorsal patterning genes. Along the anterior-posterior axis, several segment polarity genes (wingless, engrailed, hedgehog, and patched) are required for the proper allocation of the GDPCs. These segment polarity genes are expressed in some, but not all of the GDPCs, indicating that anterior and posterior compartments are not fully established in the GDPCs. In addition, we found that the three primordia of the larval genital disc have already been specified in the GDPCs by the coordinated actions of the homeotic (Hox) genes, abdominal-A, Abdominal-B, and caudal. By identifying how these different patterning networks regulate the allocation and primordial organization of the 22 embryonic precursors of the compound genital disc, we demonstrate that at least some of the organization of the larval disc originates as positional information in the embryo, thus providing a context for further studies on the development of the genital disc.

Specification of the meso-isthmo-cerebellar region: the Otx2/Gbx2 boundary

The midbrain/hindbrain (MH) territory containing the mesencephalic and isthmocerebellar primordial is characterized by the expression of several families of regulatory genes including transcription factors (Otx, Gbx, En, and Pax) and signaling molecules (Fgf and Wnt). At earlier stages of avian neural tube, those genes present a dynamic expression pattern and only at HH18-20 onwards, when the mesencephalic/metencephalic constriction is coincident with the Otx2/Gbx2 boundary, their expression domains become more defined. This review summarizes experimental data concerning the genetic mechanisms involved in the specification of the midbrain/hindbrain territory emphasizing the chick/quail chimeric experiments leading to the discovery of a secondary isthmic organizer. Otx2 and Gbx2 co-regulation could determine the precise location of the MH boundary and involved in the inductive events characteristic of the isthmic organizer center.

Early anterior/posterior patterning of the midbrain and cerebellum

Transplantation studies performed in chicken embryos indicated that early anterior/posterior patterning of the vertebrate midbrain and cerebellum might be regulated by an organizing center at the junction between the midbrain and hindbrain. More than a decade of molecular and genetic studies have shown that such an organizer is indeed central to development of the midbrain and anterior hindbrain. Furthermore, a complicated molecular network that includes multiple positive and negative feedback loops underlies the establishment and refinement of a mid/hindbrain organizer, as well as the subsequent function of the organizer. In this review, we first introduce the expression patterns of the genes known to be involved in this patterning process and the quail-chick transplantation experiments that have provided the foundation for understanding the genetic pathways regulating mid/hindbrain patterning. Subsequently, we discuss the molecular genetic studies that have revealed the roles for many genes in normal early patterning of this region. Finally, some of the remaining questions and future directions are discussed.

Two Pax2/5/8-binding sites in Engrailed2 are required for proper initiation of endogenous mid-hindbrain expression

During early brain development mouse Engrailed2 (En2) is expressed in a broad band across most of the mid-hindbrain region. Evidence from gene expression data, promoter analysis in transgenic mice and mutant phenotype analysis in mice and zebrafish has suggested that Pax2, 5 and 8 play a critical role in regulating En2 mid-hindbrain expression. Previously, we identified two Pax2/5/8-binding sites in a 1.0 kb En2 enhancer fragment that is sufficient to directed reporter gene expression to the early mid-hindbrain region and showed that the two Pax2/5/8-binding sites are essential for the mid-hindbrain expression in transgenic mice. In the present study we have examined the functional requirements of these two Pax2/5/8-binding sites in the context of the endogenous En2 gene for directing mid-hindbrain expression. The two Pax2/5/8-binding sites were deleted from the En2 locus and replaced with the bacterial neo gene by homologous recombination in mouse embryonic stem cells. After transmitting the mutation into mice, the neo gene was removed by breeding with transgenic mice expressing cre from a CMV promoter. Embryos homozygous for this En2 Pax2/5/8-binding site deletion mutation had a mild reduction in En2 expression in the presumptive mid-hindbrain region at the 5-7 somite stage, when En2 expression is normally initiated. However, from embryonic day 9.0 onwards, the mutant embryos showed En2 expression indistinguishable from that seen in wild type embryos. Furthermore, the mutants did not show the cerebellar defect seen in mice with a null mutation in En2. This result demonstrates that the two Pax2/5/8-binding sites that were deleted, while being required for mid-hindbrain expression in the context of a 1.0 kb En2 enhancer, are only required for proper initiation of expression of the endogenous En2 gene. Interestingly, a comparison of the lacZ RNA and protein expression patterns directed by the 1.0 kb enhancer fragment revealed that lacZ protein was acting as a lineage marker in the mid-hindbrain region by persisting longer than the mRNA. The transgene expression directed by the 1.0 kb enhancer fragment therefore does not mimic the entire broad domain of En2 expression. Taken together, these two studies demonstrate that DNA binding sites in addition to the two Pax2/5/8-binding sites must be necessary for En2 mid-hindbrain expression.

Differential requirements of the fused kinase for hedgehog signalling in the Drosophila embryo

The two signalling proteins, Wingless and Hedgehog, play fundamental roles in patterning cells within each metamere of the Drosophila embryo. Within the ventral ectoderm, Hedgehog signals both to the anterior and posterior directions: anterior flanking cells express the wingless and patched Hedgehog target genes whereas posterior flanking cells express only patched. Furthermore, Hedgehog acts as a morphogen to pattern the dorsal cuticle, on the posterior side of cells where it is produced. Thus responsive embryonic cells appear to react according to their position relative to the Hedgehog source. The molecular basis of these differences is still largely unknown. In this paper we show that one component of the Hedgehog pathway, the Fused kinase accumulates preferentially in cells that could respond to Hedgehog but that Fused concentration is not a limiting step in the Hedgehog signalling. We present direct evidence that Fused is required autonomously in anterior cells neighbouring Hedgehog in order to maintain patched and wingless expression while Wingless is in turn maintaining engrailed and hedgehog expression. By expressing different components of the Hedgehog pathway only in anterior, wingless-expressing cells we could show that the Hedgehog signalling components Smoothened and Cubitus interruptus are required in cells posterior to Hedgehog domain to maintain patched expression whereas Fused is not necessary in these cells. This result suggests that Hedgehog responsive ventral cells in embryos can be divided into two distinct types depending on their requirement for Fused activity. In addition, we show that the morphogen Hedgehog can pattern the dorsal cuticle independently of Fused. In order to account for these differences in Fused requirements, we propose the existence of position-specific modulators of the Hedgehog response.

The genetics and molecular structure of the Drosophila pair-rule gene odd Oz (odz)

Null alleles of the odz pair-rule gene have been generated as small Deficiency chromosome mutations. The true null phenotype of odz in segmentation is now seen to be very similar to that originally characterized, with each odd segment removed. No other previously isolated mutations in the genomic region proved allelic to odz. The generated Deficiencies covering the odz locus and immediately surrounding regions do not cover any other gene tested from that region. The entire odz gene has been cloned and mapped, and represents more than 120 kb of genomic DNA. The gene has been sequenced, except for two very large introns. Analysis of the upstream control region of the gene indicates the presence of a large number of putative binding sites for transcription factors that direct relevant developmentally regulated gene transcription.

Cloning, expression and properties of the alpha' subunit of casein kinase 2 from zebrafish (Danio rerio)

The protein kinase casein kinase 2 (CK2) is ubiquitous in eukaryotic cells and is apparently involved in the control of cell division. The holoenzyme is a tetramer composed of two catalytic subunits (alpha and/or alpha') and regulatory subunits (beta 2). The alpha and alpha' subunits are encoded by different genes but are very similar in amino acid sequence, except that alpha' is normally considerably shorter. There have been extensive biochemical studies with recombinant alpha and beta subunits of many species, but only one previous description of the activity of an isolated recombinant alpha' subunit from human CK2 (Bodenbach, L., Fauss, J., Robitzki, A., Krehan, A., Lorenz, P., Lozeman, F. J. & Pyerin, W. (1994) Recombinant human casein kinase II. A study with the complete set of subunits (alpha, alpha', and beta), site-directed autophosphorylation mutants and a bicistronically expressed holoenzyme, Eur. J. Biochem. 220, 263-273). In the present work, the isolation and bacterial expression of a cDNA coding for the alpha' subunit of zebrafish (Danio rerio) is reported. The clone covers the complete coding region that generates a protein of 348 amino acids that is 86% identical to the alpha' subunits of human and chicken, and 82% identical to the sequenced portion of the CK2 alpha subunit of zebrafish. The recombinant alpha' subunit has apparent K(m) values for ATP (6 microM), GTP (20 microM), casein (2.0 mg/ml) and the model peptide RRRDDDSEDD (0.3 mM) which are very similar to those of the recombinant alpha subunit of Xenopus laevis. The alpha' subunit kcat was 7.2 min-1 which is again similar to that of Xenopus laevis alpha subunit (7.5 min-1). The alpha' subunit also behaved similarly to CK2 alpha with regard to optimal concentrations for Mg+2 or Mn+2 and to the inhibition by heparin and the poly(Glu80Tyr20) peptide. However alpha' kinase activity was less sensitive to poly(U) inhibition than alpha, it was more heat stable than alpha, and alpha' was slightly more sensitive to KCl inhibition than alpha. The difference in salt sensitivity, however, was enhanced by the presence of the regulatory beta subunit which shifted the optimal salt concentration of the phosphorylating activity. The alpha' 2 beta 2 holoenzyme was inhibited by KCl concentrations above 100 mM, while the alpha 2 beta 2 enzyme was stimulated by KCl concentrations up to 150 mM and required 180 mM for inhibition. Another important difference between alpha and alpha' is seen in the degree of the stimulation of casein phosphorylation activity in the presence of the regulatory beta subunit. When assayed at 100 mM KCl stoichiometric amounts of CK2 beta produced maximal stimulation of both alpha' (D. rerio) and alpha (X. laevis), however the activity levels with alpha' were stimulated 20-fold by beta while the addition of beta stimulated alpha (X. laevis) only 7-8-fold.

Wnt4 affects morphogenesis when misexpressed in the zebrafish embryo

The Wnts are a family of secreted glycoproteins involved in cell-cell signalling and pattern formation during development, although the extent to which various Wnts are functionally equivalent remains unclear. We have cloned zebrafish (Danio rerio) wnt4, characterized its expression, and compared its activity relative to other Wnts. The wnt4 transcript is first detected early in somitogenesis, in the dorsocaudal region of the forebrain, and then appears in the dorsal and lateral regions of the caudal hindbrain and neural keel. During somitogenesis, wnt4 appears in the floor plate, and this expression is absent in cyclops mutants, which lack floor plate. wnt4 is also expressed in the developing pronephros and gill slit. To characterize the biological activity of wnt4, synthetic zebrafish wnt4 mRNA was injected into embryos of zebrafish and Xenopus laevis. The phenotypic effects of misexpression in the zebrafish include cyclopia, misfolding in the brain, and an anteriorly forking notochord. Comparison of the phenotypes arising from misexpression of wnt4 and Xwnt-5A in both organisms suggests close parallels in the response to these Wnts. Our data suggest that wnt4, like Xwnt-5A, inhibits cell movements, and that these Wnts define a functional class distinct from the class which includes Wnt-1, Xwnt-3A and Xwnt-8.

Mutations that alter the timing and pattern of cubitus interruptus gene expression in Drosophila melanogaster

The cubitus interruptus (ci) gene is a member of the Drosophila segment polarity gene family and encodes a protein with a zinc finger domain homologous to the vertebrate Gli genes and the nematode tra-1 gene. Three classes of existing mutations in the ci locus alter the regulation of ci expression and can be used to examine ci function during development. The first class of ci mutations causes interruptions in wing veins four and five due to inappropriate expression of the ci product in the posterior compartment of imaginal discs. The second class of mutations eliminates ci protein early in embryogenesis and causes the deletion of structures that are derived from the region including and adjacent to the engrailed expressing cells. The third class of mutations eliminates ci protein later in embryogenesis and blocks the formation of the ventral naked cuticle. The loss of ci expression at these two different stages in embryonic development correlates with the subsequent elimination of wingless expression. Adults heterozygous for the unique ciCe mutation have deletions between wing veins three and four. A similar wing defect is present in animals mutant for the segment polarity gene fused that encodes a putative serine/threonine kinase. In ciCe/+ and fused mutants, the deletions between wing veins three and four correlate with increased ci protein levels in the anterior compartment. Thus, proper regulation of both the ci mRNA and protein appears to be critical for normal Drosophila development.

Odd Oz: a novel Drosophila pair rule gene

We have identified a novel pair rule gene in Drosophila, odd Oz (odz). Every odd-numbered body segment is deleted in odz mutant embryos. The odz gene product is strongly expressed in the embryonic central nervous system and heart, and both of these tissues are malformed in mutant embryos. Odz represents the only known pair rule gene that does not encode a transcription factor. Instead, it encodes a protein with EGF-like repeats homologous to those of the extracellular matrix protein tenascin. The protein is also a putative transmembrane tyrosine kinase substrate. On the basis of its structure, odz must act in a cellularized embryo. This is consistent with odz expression, whose temporal appearance is indicative of a very late-acting pair rule gene.

Localized expression of sloppy paired protein maintains the polarity of Drosophila parasegments

During germ-band extension in the Drosophila embryo, intercellular communication is required to maintain gene expression patterns initiated at cellular blastoderm. For example, the wingless (wg) single-cell-wide stripe in each parasegment (PS) is dependent on a signal from the adjacent, posterior cells, which express engrailed (eN). This signal is thought to be the hedgehog (hh) gene product, which antagonizes the activity of patched (ptc), a repressor of wg expression. Genetic evidence indicates that the hh signal is bidirectional, but wg transcription is only derepressed on the anterior side of the en/hh stripes. To explain the asymmetric response of the wg promoter to the hh signal, current models predict that each PS is divided into cells that are competent to express either wg or en, but not both. The sloppy paired (slp) locus contains two transcription units, both encoding proteins containing a forkhead domain, a DNA-binding motif. Removal of slp gene function causes embryos to exhibit a severe pair-rule/segment polarity phenotype. We show that the en stripes expand anteriorly in slp mutant embryos and that slp activity is an absolute requirement for maintenance of wg expression at the same time that wg transcription is dependent on hh. The slp proteins are expressed in broad stripes just anterior of the en-positive cells, overlapping the narrow wg stripes. We propose that by virtue of their ability to activate wg and repress en expression, the distribution of the slp proteins define the wg-competent and en-competent groups. Consistent with this hypothesis, ubiquitous expression of slp protein throughout the PS abolishes en expression and, in ptc mutant embryos, results in a near ubiquitous distribution of wg transcripts. In addition to demonstrating the role of slp in maintaining segment polarity, our results suggest that slp works in, or parallel with, the ptc/hh signal transduction pathway to regulate wg transcription.

dishevelled is required during wingless signaling to establish both cell polarity and cell identity

The dishevelled gene of Drosophila is required to establish coherent arrays of polarized cells and is also required to establish segments in the embryo. Here, we show that loss of dishevelled function in clones, in double heterozygotes with wingless mutants and in flies bearing a weak dishevelled transgene leads to patterning defects which phenocopy defects observed in wingless mutants alone. Further, polarized cells in all body segments require dishevelled function to establish planar cell polarity, and some wingless alleles and dishevelled; wingless double heterozygotes exhibit bristle polarity defects identical to those seen in dishevelled alone. The requirement for dishevelled in establishing polarity in cell autonomous. The dishevelled gene encodes a novel intracellular protein that shares an amino acid motif with several other proteins that are found associated with cell junctions. Clonal analysis of dishevelled in leg discs provides a unique opportunity to test the hypothesis that the wingless dishevelled interaction species at least one of the circumferential positional values predicted by the polar coordinate model. We propose that dishevelled encodes an intracellular protein required to respond to a wingless signal and that this interaction is essential for establishing both cell polarity and cell identity.

Identification and cell lineage of individual neural precursors in the Drosophila CNS

The Drosophila CNS is complex enough to serve as a model for many of the molecular, cellular and developmental functions of the vertebrate CNS, yet simple enough for single-cell analysis. Recent advances have provided molecular markers that allow most Drosophila CNS precursors to be uniquely identified, as well as methods for determining the complete cell lineage of each precursor. A detailed understanding of wild-type neurogenesis, combined with existing molecular genetic techniques, should provide insight into the fundamental mechanisms that generate neuronal and glial diversity.

Neuroblast specification and formation regulated by wingless in the Drosophila CNS

The Drosophila central nervous system (CNS) develops from a heterogeneous population of neural stem cells (neuroblasts), yet the genes regulating neuroblast determination remain unknown. The segmentation gene wingless is regionally expressed in the neuroectoderm from which neuroblasts develop. A conditional wingless mutation is used to inactivate CNS function without affecting segmentation. The stripe of wingless-expressing neuroectoderm generates apparently normal neuroblasts after wingless inactivation; however, adjacent anterior and posterior neuroectoderm requires wingless nonautonomously for subsequent neuroblast determination and formation. Loss of wingless results in the absence or duplication of identified neuroblasts, highlighting its role in generating neuroblast diversity in the CNS.

Segmentation of the Drosophila embryo

Segmentation in Drosophila is a sequential process directed by at least 30 genes that encode various types of proteins, including: many transcription factors; a putative RNA-binding protein; a membrane-associated receptor kinase; several intracellular protein kinases; a number of secreted signaling molecules; and others of unknown function. Although the detailed molecular reactions used to generate the metameric subdivisions of the embryo are not yet understood, a general outline of the processes involved has been described. The manner in which spatial relations in the developing embryo are established can now be described.