This page has been archived and is no longer updated
Arthropod Segmentation: beyond the Drosophila paradigm
Author: Andrew D. Peel
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"� 2005 Nature Publishing Group University Museum of Zoology, Department of Zoology, Downing Street, Cambridge CB2 3EJ, UK. e-mail: m.akam@zoo.cam. ac.uk doi:10.1038/nrg1724 Published online 10 November 2005 MACROEVOLUTION Evolutionary processes that lead to significant morphological change. This usually refers to processes that occur above the species level. LIFE HISTORY The sum of the morphological stages and the ecological environments that an organism goes through during its life. BILATERIANS Members of the animal kingdom that have bilateral symmetry ? the property of having two similar sides, with definite upper and lower surfaces, and anterior and posterior ends. The gene networks that generate segments in arthro- pods provide an excellent model for understanding MACROEVOLUTION at the molecular level. Arthropods share a basic body plan, but elaborate a wide diver- sity of morphologies that affect segment form and number. They also have different modes of embryo- genesis that are adaptations to different LIFEuniF6BAHISTORY strategies, and these differences often affect how and when segments are generated in relation to other processes of embryogenesis. Segmentation is also of interest because it has either evolved repeatedly in several animal phyla, or has been lost entirely in many others. Molecular phylogenies identify three distinct lineages of BILATERIAN animals 1 . Each of these lineages contains one of the three princi- pal segmented phyla ? the vertebrate chordates within the DEUTEROSTOMES, the annelid worms within the LOPHOuniF6BA TROCHOZOANS and the arthropods within the ECDYSOZOANS. This raises developmental and evolutionary questions. To what extent are the mechanisms that underlie the development of segmented body plans similar across phylogenetically diverse bilaterian phyla? And, if there are similarities, do they reflect a common evolutionary origin of segmentation or the convergent recruitment of similar developmental mechanisms during evolution? Put simply, and perhaps naively, was the bilaterian common ancestor a segmented animal, or have segmented body plans evolved multiple times? Interest in these questions has been stirred recently by the discovery that some arthropods seem to pattern their posterior segments using genetic mechanisms that are similar to those that operate in vertebrates during SOMITOGENESIS 2,3 . It is therefore timely to review our understanding of segment patterning across the arthropods. In the 1980s large-scale genetic screens in the fruit- fly Drosophila melanogaster identified about 40 genes that are necessary to generate a normal segmentation pattern. Subsequent studies showed that these genes function in a hierarchy ? they encode a cascade of interacting transcription factors that generate pro- gressively finer patterns of gene expression in the BLASTODERM-stage embryo 4 uniF6AEBOX 1uniF6AF. Until recently most studies of segment patterning in arthropods other than D. melanogaster have been limited to descriptions of gene-expression patterns. Tests for function have been limited to a few small-scale genetic screens 5?7 and one or two attempts at gene knockdown. The advent of effective RNAi methods 8 has meant that the function, as well as the expression, of segmentation genes can be examined in a wide range of arthropods, and successful transgenesis in some new ?model? species indicates that ARTHROPOD SEGMENTATION: BEYOND THE DROSOPHILA PARADIGM Andrew D. Peel, Ariel D. Chipman and Michael Akam Abstract | Most of our knowledge about the mechanisms of segmentation in arthropods comes from work on Drosophila melanogaster. In recent years it has become clear that this mechanism is far from universal, and different arthropod groups have distinct modes of segmentation that operate through divergent genetic mechanisms. We review recent data from a range of arthropods, identifying which features of the D. melanogaster segmentation cascade are present in the different groups, and discuss the evolutionary implications of their conserved and divergent aspects. A model is emerging, although slowly, for the way that arthropod segmentation mechanisms have evolved. NATURE REVIEWS | GENETICS VOLUME 6 | DECEMBER 2005 | 905 FOCUS ON THE BODY PLAN � 2005 Nature Publishing Group HB mat hb BCD CAD NOS gt Kr kni gt eve tll tll Mater nal gradients Gap domains Pair -rule patter n Segment polarity patter n Parasegment boundaries ftz en wg Step 1 Step 2 Step 3 Step 4 Anterior Posterior DEUTEROSTOMES One of the three main branches of the bilaterian animals. The deuterostomes include chordates, hemichordates and echinoderms. LOPHOTROCHOZOANS One of the three main branches of the bilaterian animals. Lophotrochozoans include annelids, molluscs, flatworms and several other smaller phyla. ECDYSOZOANS One of the three main branches of the bilaterian animals. Ecdysozoans are characterized by an unciliated integument, and grow by ecdysis, or moulting. They include nematodes, arthropods and many other smaller phyla. SOMITOGENESIS The process of progressive formation, during embryogenesis, of metameric mesodermal units (somites) that represent the precursor structures of dermis, skeletal muscles and the axial skeleton. BLASTODERM The layer of cells that completely surrounds an internal mass of yolk in an arthropod embryo. FOLLICLE CELLS The somatic cells in Drosophila melanogaster that surround the oocyte; they provide patterning signals to the oocyte and secrete the egg-shell. DERIVED Having undergone significant evolutionary change relative to the ancestral state. misexpression and reporter-gene studies will soon be widely applicable 9?11 . Understanding the genetic basis of segmentation in a wider range of arthropods is important for two reasons. First, it might help us to understand how the fly genetic model for segmentation evolved, and sec- ond, it might reveal similarities between arthropod and vertebrate genetic mechanisms of segmentation that are obscured, or no longer exist, in the DERIVED mode of segmentation that is seen in D. melanogaster. In this review we examine what recent func- tional studies have taught us about the conserva- tion, or otherwise, of genes and gene networks that operate in each tier of the segmentation cascade of D. melanogaster. We then highlight the limitations of the ?candidate-gene? approach, and consider the lessons that might be learned from the segmentation mechanisms that operate in vertebrates. We conclude by outlining some important questions for future research in this field. Box 1 | Segmentation in Drosophila melanogaster Steps 1 and 2 From maternal signals to gap domains. Oogenesis provides the Drosophila melanogaster egg with a ?ready mix? cytoplasm, so that segmentation can proceed rapidly after fertilization. Maternal transcripts of the segmentation genes caudal (cad) and hunchback (hb) are ubiquitously distributed. bicoid (bcd) is localized at the anterior pole of the egg, and a complex of proteins and RNAs (the pole plasm) is localized at the posterior. After fertilization, translation of localized maternal transcripts, and the diffusion of their protein products, generates protein gradients along the egg (step 1 in the figure). BCD is both a transcriptional activator of hb and a translational repressor of cad mRNA. Nanos (NOS), a key component of germ plasm, represses translation of maternal hb RNA. The resulting egg contains the HB protein in the anterior half (maternal HB, HB mat), and long-range gradients of both BCD (high at the anterior) and CAD (high at the posterior). At the same time, signals that are embedded in the egg-shell by localized populations of FOLLICLE CELLS activate a transmembrane receptor, Torso, at the poles of the egg. Together, the signals from these maternal proteins activate a small set of zygotic ?gap? genes (tailless (tll), giant (gt), Kr�ppel (Kr) and knirps (kni)) at specific positions along the anterior?posterior axis of the egg (step 2). Interactions between the gap genes refine their expression further, but because these proteins are free to diffuse in the egg syncytium, the distribution of each gap protein is graded around its peak site of synthesis. Step 3 From gap domains to pair-rule pattern. The pattern of gap-gene expression is aperiodic. The concentrations of their transcription factor products, together with inputs from maternal proteins, give nuclei a unique axial identity. At least three of the pair-rule genes (the primary pair-rule genes hairy (h), runt (run) and even skipped (eve)) interpret these identities to generate periodic stripes of gene expression. Again, transcriptional interactions between these primary pair-rule genes and the other genes that they regulate (for example, fushi tarazu (ftz) and paired (prd)) refine the domains of expression until the edges of each stripe are defined to single-cell resolution (step 3 in the figure). The boundaries of the pair-rule stripes predict the boundaries of parasegments. Step 4 From pair-rule pattern to stable boundaries. The pattern of pair-rule gene expression is transient. However, as cell boundaries form, the activities of the pair-rule proteins (again, mostly transcription factors) result in the activation of the segment polarity genes. Odd- and even-numbered parasegments express different combinations of pair-rule genes, but produce the same output of segment polarity gene expression in every segment. The segment polarity pattern is stable, and at least for some genes (notably engrailed (en)) it will persist into the adult. The boundary between en-expressing cells and their anterior neighbours (which express wingless (wg)) becomes the parasegment boundary. Segment boundaries are established slightly later, posterior to en-expressing cells. 906 | DECEMBER 2005 | VOLUME 6 www.nature.com/reviews/genetics REVIEWS � 2005 Nature Publishing Group Segment polarity genes (phylotypic stage) Hox genes Pair-rule genes Gap genes Maternal determinants GERM BAND In an arthropod embryo, this is the differentiated portion, which has a distinct anterior? posterior axis, and is where the segmentation process takes place. SEGMENT POLARITY GENES A group of genes that define different parts of each segmental repeat. When segment polarity genes are mutated the normal number of segments is formed, but these show internal pattern replication and polarity reversals. PARASEGMENT The initial segmental unit that is formed during the segmentation process. The final segment boundaries lie in the middle of the parasegments. HOLOMETABOLOUS Insects for which the life cycle includes distinct larvae, pupae and (usually) winged adults. SYNCYTIUM A population of nuclei that are not separated by cell membranes. It is typical of the developing blastoderm in Drosophila melanogaster. SHORT GERM A mode of insect development in which anterior segments are patterned in the blastoderm, with posterior segments forming sequentially from a cellularized growth zone after gastrulation. HEMIMETABOLOUS Insects for which the life cycle includes several larval stages, ending in a sexually mature, winged adult, without going through a pupal stage. A segmented body plan A segmented body plan is a defining characteristic of the arthropods, and is almost certainly a trait that was ancestral to the whole phylum. All arthropod embryos pass through a segmented GERMuniF6BABAND stage that, at the morphological level at least, seems to be remarkably conserved, and has been referred to as the ?phylotypic stage? 12 . Embryonic events either side of this stage are much less conserved, presumably as a result of direc- tional selection for divergent life histories. Interestingly, the segmentation genes that function at the bottom of the D. melanogaster segmentation cascade, just before and during the phylotypic stage, seem to be conserved across the arthropods (FIG. 1). These genes ? which include homologues of the D. melanogaster SEGMENT POLARITY GENES engrailed (en), wingless (wg) and hedgehog (hh), and encode proteins that have a range of functions ? establish definitive segment (or, to be precise, PARASEGMENT 13 uniF6AEBOX 1uniF6AF) boundaries. They show similar patterns of expression in diverse arthropods 14?21 and constitute an evolutionarily conserved regulatory cassette 22,23 . Exactly why the segment polarity genes are so conserved has been the subject of some debate 15,21 . Divergent embryology before the phylotypic stage indicates that the genetic networks that func- tion towards the top of the segmentation cascade in D. melanogaster might be much less conserved. In particular, the syncytial context of segmentation in D. melanogaster is a derived characteristic that is shared with only some other groups among the higher (HOLOMETABOLOUS) insects. Most arthropods do not pat- tern all of their segments while the embryo is still a SYNCYTIUM uniF6AEBOXES 1,2uniF6AF, which raises the question as to whether local transcription factor gradients could be operating as they do in D. melanogaster. Most arthropods pattern their posterior segments sequentially from a cellularized growth zone uniF6AEBOX 2uniF6AF, a trait that is thought to be primitive to the arthropods 24 . In this review, species in which segmentation occurs by sequential addition from a posterior growth zone are referred to as ?sequentially segmenting? arthropods, but it is important to note that these species show signifi- cant embryonic differences in relation to one another, as well as to D. melanogaster uniF6AEBOX 2uniF6AF. The term ?SHORT GERM?, which is often used to describe such arthropods, is misleading when referring to those arthropods ? such as the centipede Strigamia maritima ? in which segments are generated from a large pool of cells, rather than a small embryonic primordium. Therefore, we prefer not to use it as a generalized term. Much of the available data on segmentation mecha- nisms relates to insects, but even among these, some groups are poorly represented. The more derived, holo- metabolous insect orders are relatively well-sampled (data exist for 4 out of 11 orders), particularly with the devel- opment of Tribolium castaneum as a powerful model for functional studies in Coleoptera (beetles) 8,25,26 . However, it is clear that even within this group there is a wide diversity of segmentation mechanisms, perhaps reach- ing an extreme in the polyembryonic wasp Copidosoma floridanum 27 . The more basal, HEMIMETABOLOUS insects are represented by descriptive studies in just a handful of species, and by functional studies in just two ? the orthopteran Gryllus bimaculatus 28?30 (a cricket) and the hemipteran Oncopeltus fasciatus 31?34 (the milkweed bug) (FIG. 2). The diversity of Crustacea has barely been sampled, and functional data are so far available only for the brine shrimp Artemia franciscana 35 . The insects and crustaceans together ? referred to as the Pancrustacea ? are thought to constitute one of three principal monophyletic lineages within the arthro- pods (the traditional view that insects and crustaceans form two closely allied but distinct monophyletic clades is not supported by molecular data) 36,37 . The other two main lineages are the chelicerates (represented here by studies of spiders 2,3,14,38,39 and mites 40 ) and the myriapods. Myriapods are no longer thought to be closely related to the insects, but instead are an ancient lineage in their own right 37 . They are represented here by studies on segmentation in centipedes 17,41,42 and millipedes 20 . How pancrustaceans, myriapods and chelicerates are related remains unclear, but the hope is that by encompassing appropriate representatives of all three of these clades (FIG. 2), comparative studies might reveal which aspects of the segmentation machinery represent ancestral characteristics of the arthropods. Figure 1 | Conservation of the segmentation cascade in arthropods. The well-studied segmentation cascade of Drosophila melanogaster represents a derived mechanism compared with that of other arthropods. The degree of conservation of genes that function in successive steps of the segmentation cascade are represented by the width of the hourglass. The earliest stage of the cascade, axis determination by maternal gradients, has diverged significantly between arthropod groups. Gap-gene homologues can be found in all arthropods, but their function in segmentation is variable. Pair-rule patterning has been described in several arthropods, but it is not clear whether this is an ancestral feature or one that has evolved convergently. The best-conserved stage is the one in which segmental boundaries are defined by the interaction of segment polarity genes. The expression of these genes coincides with the so-called ?phylotypic stage? of arthropods ? the segmented germ band 12,21 . Later on in the cascade, when axial identity is conferred by Hox genes, arthropod groups diverge again, with genes of the Hox family being expressed at different axial levels in different species. NATURE REVIEWS | GENETICS VOLUME 6 | DECEMBER 2005 | 907 FOCUS ON THE BODY PLAN � 2005 Nature Publishing Group Head a Drosophila melanogaster b Strigamia maritima c Malacostracan crustacean Thorax Abdomen Forming segments Terminal disk Undifferentiated cell Cell movements Anterior segments Forming segments Ectoteloblast Anterior segments Syncytial nucleus MALACOSTRACANS A subclass of crustaceans that includes shrimps, lobsters and sandhoppers. AMPHIPODS An order of malacostracan crustaceans that includes beachhoppers. Top of the hierarchy: the maternal effect genes In D. melanogaster maternal cues trigger the pat- terning of the early embryo. Maternal transcripts are loaded into the oocyte and are specifically tar- geted to the anterior (bicoid; bcd) 43?46 or posterior (nanos; nos) 47?49 poles by the cytoskeletal machin- ery 4,50 . At fertilization these maternally provided transcripts are translated to form the source of pro- tein gradients that initiate anterior?posterior (A?P) patterning 4,50 uniF6AEBOX 1uniF6AF. Conservation and divergence of anterior patterning. At the anterior the BCD protein activates transcription of downstream segmentation genes 44 . Despite its central role in patterning D. melanogaster, an anterior gradient of the maternally derived BCD protein was probably a new invention of the higher Diptera 51?56 . bcd genes have been isolated from several Drosophila species, from houseflies 51 , and from other ?higher flies? of the suborder Cyclorrhapha 52 . However, no bcd gene has been isolated from species other than those from the Diptera 53,55 , and none has been found in the genome of the mosquito Anopheles gambiae, a distant relative within the Diptera. Instead, at the location where the bcd gene resides in the Hox cluster of D. melanogaster, there are only genes that are related to zerknullt (zen), the neighbour of bcd in the D. melanogaster genome, which is a derived Hox3 gene 53,54 . Comparison of the sequences of the Dipteran zen and bcd genes indi- cates that bcd arose within the higher Diptera (basal Cyclorrhapha) by rapid sequence divergence of a zen gene, and by the acquisition of new DNA targets through a mutation in the homeodomain that gives BCD a recog- nition sequence that is similar to that of Orthodenticle (OTX), another homeodomain-containing transcription factor 52?55 . Box 2 | Diverse cellular mechanisms of segmentation in arthropods In Drosophila melanogaster, all segments are patterned more or less simultaneously while the blastoderm is still syncytial (panel a; also see BOX 1). By contrast, most other arthropods pattern a small number of segments at the blastoderm stage, and then add posterior segments consecutively from a growth zone. The first patterned segments include at least the three anterior segments: the antennal segment, the intercalary segment (which is the second antennal segment in crustaceans) and the mandibular segment. These three segments are sometimes referred to as the naupliar segments, and are the only ones that are present in the larval stages of many crustacean groups. In many insects, in addition to the naupliar segments, two to five other segments are formed in the blastoderm stage, including up to three thoracic segments. Little is known about the mechanism behind the formation of anterior segments. It is possible that they are generated by a mechanism that is distinct from those that function during the formation of more posterior segments. In most arthropods posterior segments are generated in a cellular environment. In some cases these segments arise from a small population of posterior cells in the blastoderm, which proliferate later to generate the tissue from which segments are patterned (this occurs, for example, in Artemia franciscana and other branchiopod crustaceans, and in most ?short germ? insects). In other cases (such as in the centipede Strigamia maritima, panel b) a blastodisc that contains many thousands of apparently undifferentiated cells persists after the completion of anterior segmentation, and posterior segments are generated sequentially from this population by a combination of proliferation and cell rearrangement. In MALACOSTRACAN crustaceans, cells of the germ band organize into a square array, each row of which will give rise to a single segment through a stereotyped series of polarized cell divisions (panel c). Sometimes these rows are generated by the aggregation of cells from a preformed blastodisc (for example, in Parhyale hawaiiensis and other AMPHIPODS). In many cases they are generated by the sequential divisions of ectoteloblasts ? stem cells that lie at the posterior of the germ band. 908 | DECEMBER 2005 | VOLUME 6 www.nature.com/reviews/genetics REVIEWS � 2005 Nature Publishing Group Species Diptera Insecta Pancrustacea Coleoptera Hemiptera Orthoptera Thysanura Malacostraca Branchiopoda Myriapoda Chelicerata Available techniques Fruitfly Drosophila melanogaster Flour beetle Tribolium castaneum Milkweed bug Oncopeltus fasciatus Cricket Gryllus bimaculatus Grasshopper Schistocerca gregaria Schistocerca americana Firebrat Thermobia domestica Amphipod Parhyale hawaiiensis Brine shrimp Artemia franciscana Garden centipede Lithobius atkinsoni Coastal centipede Strigamia maritima Spider mite Tetranychus urticae Spider Cupiennius salei G, R, T G, R, T G, R G, R, T G G G, T G, R G G G G, R PREuniF6BAGNATHAL SEGMENTS The gnathal segments comprise the mandibular, maxillary and labial head segments of insects. The pre-gnathal segments lie anterior to these segments. The number of pre-gnathal segments has been debated, but probably includes at least three: the ocular, antennal and intercalary segments. BLASTODISC An undifferentiated single-cell layer in an arthropod embryo that ultimately gives rise to all embryonic structures. Details are emerging about how the embryo is patterned in some insects that lack bcd. In the beetle T. castaneum, RNAi knockdown phenotypes indicate that maternally derived OTX1 and Hunchback (HB) proteins cooperate to carry out a role that is analo- gous to BCD during embryogenesis 26,56 . Tribolium castaneum differs from D. melanogaster in that it forms its abdominal segments sequentially in a cellular environment, but is similar to D. melanogaster in that its anterior segments are patterned in a syncytium 57,58 . Therefore, the proposal that a maternally derived anterior gradient is operating in this arthropod seems feasible. It is not known how gradients of these two maternal proteins are generated in the beetle ? in both cases, their maternal RNAs are initially ubiquitously distributed in the egg 26,56,59 . However, the transcripts for two other transcription factors ? eagle (eg) and pangolin (pan) ? are maternally localized to the anterior pole in T. castaneum, which indicates that the molecular machinery required to localize mRNAs to the anterior pole predated the recruitment of bcd to anterior patterning 60 . Whether eg and pan are involved in A?P patterning in T. castaneum is unclear 60 ; neither is known to have such a role in D. melanogaster. The mechanism of anterior patterning in T. castaneum need not necessarily represent the ancestral state for insects, let alone all arthropods. Many other arthropods do not pattern their anterior segments in a syncytium. For example, cellularization occurs early in the embryonic development of the grasshopper Schistocerca gregaria 61 , before any seg- ment patterning. An anterior gradient might therefore not be required in arthropods that show sequential segmentation from a posterior growth zone, given that only a few anterior segments are patterned in the blastoderm 62,63 uniF6AEBOX 2uniF6AF. Indeed, earlier experimental studies indicate that anterior patterning gradients are widely used only among the higher (holometobolous) insects 63 . In most hemimetabolous insects, the ante- rior pole of the egg seems to have no specific role in patterning the embryo 63 . Caudal and Nanos in embryonic patterning In D. melanogaster, the Caudal (CAD) and NOS proteins both show graded distributions in the blas- toderm, with levels that are high at the posterior and decrease anteriorly 49,64?66 . Both are required for nor- mal segmentation at the posterior of the germ band uniF6AEBOX 1uniF6AF. Because cad RNA is provided both maternally and zygotically, but is not initially localized 65 , genetic screens were slow to reveal the full role of cad during segment patterning. There is now evidence that homologues of both nos and cad might be involved in patterning more anterior segments in sequentially segmenting arthropods. caudal. RNAi experiments in two sequentially segmenting insects ? T. castaneum 35 and the cricket G. bimaculatus 28 ? and in one crustacean (A. franciscana 35 ) have revealed an essential role for cad, not only in patterning posterior segments, but also in the formation of the entire segmented trunk. In G. bimaculatus 28 and T. castaneum 35 , the most extreme knockdown phenotypes eliminate all but the PREuniF6BAGNATHAL SEGMENTS. In A. franciscana, knockdown of cad in the newly hatched nauplius blocks formation of all new segments 35 uniF6AEBOX 2uniF6AF. These experiments indi- cate that cad might ancestrally have been involved in the formation of all trunk segments, but that this func- tion has been lost in D. melanogaster, perhaps by the acquisition of the long-range morphogen bcd. nanos. Evidence for the role of nos in other insects is less direct. In the locust Schistocerca americana (a species closely related to S. gregaria), the maternally derived mRNA of the nos orthologue 62 is restricted to the posterior of the early embryonic primordium, and localization of the NOS protein is consistent with a role in the translational repression of hb posterior to a gap-like domain of expression in the gnathal segments. Although it has not been directly demonstrated, the likelihood of this interaction is supported by the observation that S. americana hb mRNA retains well-conserved NOS response elements in its 3? UTR. In D. melanogaster NOS acts on maternal hb mRNA 47?49 . However, in S. americana maternal HB is provided as protein, and might be involved in defining the extent of the BLASTODISC 62 , rather than in A?P pat- terning. The role of NOS is more likely to be in regulat- ing zygotically transcribed hb RNA 62 . So the maternal provision of hb mRNA and its translational regulation by NOS could be a feature that evolved subsequently to the divergence of flies and grasshoppers. Figure 2 | Phylogenetic relationships between the arthropod species discussed in this article. The experimental techniques that are available for each organism are also shown. G, gene-expression data through RNA in situ hybridization experiments; R, RNA interference; T, germline transgenesis. NATURE REVIEWS | GENETICS VOLUME 6 | DECEMBER 2005 | 909 FOCUS ON THE BODY PLAN � 2005 Nature Publishing Group T3Lb A2 Lb A8 ab c d A1Md Md Mx T1 T1 T1 T2 T3 T2 T2 T3 A8 A1Lb HOX GENES A family of homeodomain transcription factors that are conserved across bilaterian animals; they are expressed in sequence along the A?P axis and are involved in conferring axial identity. During later embryogenesis, S. americana nos is no longer expressed in the segments that appear sequen- tially as the embryonic primordium elongates, which indicates that its role is limited to the earliest stages of embryonic patterning, and that it has no specific function in patterning posterior segments 62 . Therefore the emerging picture is that both nos and cad had an ancestral role in A?P patterning in insects and at least some crustaceans, but the extent of their influence, the identity of their regulatory targets and the nature of their roles during early development might have diverged significantly in different derived lineages. Whether an anterior pat- terning gradient was used in ancestral insect lineages seems less clear. The gap genes In D. melanogaster, the gap genes are the direct targets of maternal patterning information, and establish a series of molecularly distinct regions along the A?P axis of the blastoderm 4,67 uniF6AEBOX 1uniF6AF. The transcription fac- tors that they encode have two distinct roles at this stage of development. In combination with maternal factors, they regulate pair-rule genes through segment- specific enhancers 68 . In addition, they function with downstream segmentation genes to regulate the initial activation of HOX GENES in region-specific patterns 69,70 . In later development most of the gap genes are re-used in many other patterning processes 71 . Functional analysis of gap-gene orthologues. Ortho- logues of the D. melanogaster gap genes are relatively easy to identify in other arthropods (except knirps (kni), for which orthologues are hard to distinguish from genes that encode other kni-related transcription factors). For three genes ? hb, Kr�ppel (Kr) and giant (gt) ? both the expression patterns and functions of the orthologues have been examined in sequentially segmenting insects 25,26,30,32,33 . On the basis of expression patterns alone one might conclude that the role of these gap genes has gener- ally been conserved during insect evolution, with the caveat that there have been shifts in their precise domains of expression, and, in particular, posterior shifts within the lineage that leads to the Drosophila genus 25 . However, RNAi experiments indicate a more complex picture 25,26,30,32,33 . These gap genes seem to have a broadly conserved role in the regulation of Hox genes, but their knockdown does not always result in a true ?segment gap? phenotype, as seen in D. melanogaster ? that is, in the failure of segments to form in the region where the gap gene is normally expressed. Rather, segments might form, but have abnormal iden- tity. In these cases, corresponding and interpretable shifts in Hox gene-expression domains are detected by in situ hybridization on embryos in which these genes have been knocked down by RNAi 25,30,32 (FIG. 3). This is particularly clear in the case of the T. castaneum Kr gene, for which both RNAi and mutant data are Figure 3 | A comparison of hunchback expression and function in two sequentially segmenting insects. a,b | Panel a shows the wild-type Tribolium castaneum embryo; panel b shows the consequences of parental RNAi (pRNAi) that is targeted against hunchback in the T. castaneum embryo. pRNAi results in a canonical gap phenotype: the loss of maxillary (Mx), labial (Lb) and thoracic (T) segments. The mandibular (Md) segment is still present and the development of posterior abdominal (A) segments remains largely unaffected. Embryos are stained for engrailed expression. Hox gene expression in hunchback pRNAi embryos was not examined in this study, but homeotic transformations might be obscured by the gap phenotype. c,d | Panel c shows abdominal A (abdA) expression in the wild-type Oncopeltus fasciatus embryo; panel d shows abdA expression in the hunchback pRNAi O. fasciatus embryo. In O. fasciatus, gnathal and thoracic segments do form following hunchback pRNAi, but are transformed towards abdominal identity. Note the highly reduced labium and T1/T2 legs in the hunchback pRNAi embryo in panel d. Abdominal segments form normally, although they seem to be compacted in embryos that show strong pRNAi phenotypes (not shown). In O. fasciatus, the transformation of gnathal and thoracic segments to abdominal identity is correlated to the ectopic expression of abdA, which indicates that this Hox gene is usually repressed in the anterior by hunchback, although possibly indirectly. The anterior is to the left. Panels a and b are reproduced, with permission, from Nature REF. 26 � (2003) Macmillan Magazines Ltd. Images kindly provided by Reinhard Schr�der, Universit�t T�bingen, Germany. Panels c and d are reproduced, with permission, from REF. 32 � (2004) Company of Biologists. Images kindly provided by Thom Kaufman, Indiana University, USA. 910 | DECEMBER 2005 | VOLUME 6 www.nature.com/reviews/genetics REVIEWS � 2005 Nature Publishing Group available ? the T. castaneum jaws mutant has now been characterized as a null allele of Kr 95 . It is not always clear whether the variation across species reflects real differences in the function of gap- gene homologues or differences in the interpretation of phenotypes. Because some segments are taking on abnormal identity, whereas others are deleted, it is not easy to identify the specific segments that are lost. The penetrance of RNAi phenotypes might also vary from species to species ? for example, incomplete knock- down could be an issue with the interpretation of RNAi against hb in G. bimaculatus 30 . Several of these studies have reported an effect of knockdown on the formation of segments that appear by sequential addition from a posterior growth zone 25,30,32 . Embryos treated with RNAi are often trun- cated owing to the loss or compaction of abdominal segments, with posterior abdominal segments being affected even when they lie outside the ectodermal expression domain of the gene 25,30,32 . How gap-gene homologues mediate these apparent ?long-range? effects remains unclear. Evolution of gap-gene function. Taken at face value, the available data indicate that the function of individual gap-gene homologues has diverged significantly in dif- ferent insect lineages. If this is correct, then the regula- tion of pair-rule genes must also have changed radically during insect evolution. Indeed, a recent study in A. gambiae revealed the existence of different combina- tions of gap repressors for homologous pair-rule stripes, indicating that there has been divergence even within the Diptera 72 . Perhaps in those cases in which gap-gene homologues do not seem to function as true gap genes, other unidentified genes have analogous roles. One possibility is that in the ancestor of the insects the homologues of D. melanogaster gap genes had a role in the regulation of Hox genes and, through this, in the specification of segment identity, but not in the regulation of segmentation genes. It will be interesting to see whether gap-gene homologues are involved in the process of segment generation in arthropods other than insects. At present we have almost no data to address this. Among the many other developmental roles of the D. melanogaster gap genes, one is particularly intriguing in the context of this review. This is the role of hb and Kr during neurogenesis 71 . These two genes are among a set of transcription factors that are expressed in a stereo- typed temporal sequence in neuroblasts, where they define the temporal identity of the neuroblast progeny. The expression of hb, the first gene of the sequence, is followed slightly later by Kr ? a sequence that cor- responds to the A?P order of the expression of these genes in the D. melanogaster blastoderm 71 . Expression of hb and Kr in neuroblasts is widely conserved among arthropods 30,32,33,73 , and a temporal sequence related to that seen in D. melanogaster has been observed in centipedes, which comprise a distantly related arthro- pod lineage (A.D.C. and A. Stollewerk, unpublished observations). Perhaps these genes were recruited from neural patterning to function in segment specification in arthropods, where segments form in a temporal A?P sequence. Later, in the lineage leading to Drosophila, some of these genes (notably hb and Kr) might then have been expressed in the right place and at the right time to be recruited into regulating pair-rule gene homologues. Pair-rule genes The repetitive segment pattern of D. melanogaster is generated in the blastoderm, shortly before cellulariza- tion, by transcriptional regulation of the pair-rule genes uniF6AEBOX 1uniF6AF. The 14 parasegments of the D. melanogaster embryo are defined by 7 stripes of even skipped (eve) expression that alternate with 7 stripes of fushi tarazu (ftz) expression 4 . Every arthropod that has been examined expresses at least one homologue of a D. melanogaster pair- rule gene in a pattern that is consistent with a role in segmentation 17,31,38?40,42,74?79 . However, this does not mean that ?pair-rule patterning? is conserved in all arthropods. It is not clear whether a double-segment repeat is involved in the patterning of segments in all arthropods. We distinguish two distinct aspects of pair-rule patterning. The transcriptional network that generates the classic pair-rule stripes is only half the story (BOX 1, step 3). The final segment pattern is generated in a second step of transcriptional computation, in which the pair-rule ?codes? of the even- and odd-numbered parasegments establish a pattern that recurs identically in every segment: the initial expression of the segment polarity genes (BOX 1, step 4). In D. melanogaster many of the pair-rule genes do not provide only the inputs for this computation ? their promoters also contain regulatory elements that drive segmentally repeated expression as part of the output 4,80 . In this sense, genes such as eve are both pair-rule and segment polarity genes. Conserved expression of a pair-rule gene dur- ing segmentation does not mean that all aspects of D. melanogaster pair-rule patterning are conserved. We have noted already that the expression of some segment polarity genes is widely conserved among the arthropods 14?21 . If the segment polarity network is more ancient than the pair-rule network, then the most widely conserved roles of some pair-rule genes might lie at the level of defining the single-segment repeat, not the double-segment pre-pattern. Insects. Individual genes of the pair-rule class have diverged significantly in their expression patterns and function during arthropod evolution. For example, homologues of the pair-rule gene eve have differ- ent expression patterns in different insect species. In T. castaneum an eve homologue shows pair-rule expression, with broad pair-rule stripes splitting to form segmental stripes 78 . Inhibition of eve function generates pair-rule segmental defects, confirming that eve in this insect functions in the generation of a double-segment repeat 81 . However, in other insects NATURE REVIEWS | GENETICS VOLUME 6 | DECEMBER 2005 | 911 FOCUS ON THE BODY PLAN � 2005 Nature Publishing Group I II III I Time Time Expr ession Expr ession Anterior Neural tube Posterior PSM Tailbud Phase I Phase II Phase III Phase I SI S0 SI SII S0 SI S0 SII SI S0 b a Notch LFNG/Delta Presenilin ICN HES genes Delta HES Nucleus Cytoplasm IIII I Box 3 | Notch signalling in vertebrate somitogenesis and arthropod segmentation Vertebrate somitogenesis A segmentation clock, generated by a Notch-signalling-based oscillator, is central to somitogenesis in vertebrates 90,94 . The Notch pathway is activated by a signal from the ligand Delta in an adjacent cell. Notch then activates several downstream target genes, including those that encode transcription factors from the hairy/enhancer of split (HES) family, Lunatic fringe (LFNG) and Delta. HES family proteins repress their own expression and that of other Notch pathway genes (Lfng or Delta) (panel a). These regulatory interactions result in oscillations in the levels of the products of these genes within individual cells, which appear as anteriorly progressing waves of expression in the presomitic mesoderm (PSM; panel b). Each wave of expression precedes the formation of one somite. The extent to which intracellular versus extracellular (Notch) negative regulatory feedback loops set the period of these oscillations is still debated. Intercellular Notch signalling might couple oscillations in neighbouring cells. Opposing and antagonistic retinoic acid and fibroblast growth factor (FGF) gradients form a ?determination front?. Oscillations cease and somites are patterned in this region of the PSM ? cells that fall below a particular threshold of FGF signalling during the period of one oscillation form a somite. Recent data 91 support a role for Wnt signalling upstream of both Notch signalling (the expression of Axin2, a suppressor of Wnt signalling, oscillates in the PSM of mouse embryos) and the posterior FGF gradient (in mouse embryos there is a posterior gradient of WNT3A). Data from zebrafish 92 indicate that the Cdx genes (homologues of D. melanogaster caudal) are also downstream targets of Wnt signalling during morphogenesis of the posterior body. Segmentation of basal arthropods Recent work on segmentation in the spider Cupiennius salei indicates that a Notch-based segment-generating mechanism functions in this species as well 2 . Notch and Delta, two of the central players in the vertebrate segmentation clock, are expressed in a segmental pattern before overt segment formation, and their disruption causes segmentation defects. In addition, the disruption of two downstream targets of Notch signalling, Presenilin (Psn) and Suppressor of Hairless (Su(H)), causes severe segmentation defects. Preliminary data indicate that in the centipede Strigamia maritima, Notch target genes are involved in early segmentation. Their dynamic expression patterns indicate the existence of an underlying cycling mechanism (REF. 43; A.D.C., unpublished observations). There is also now evidence from other arthropods that Wnt signalling and caudal are crucial for sequential segmentation. However, there is currently no direct evidence for travelling waves or that gene expression oscillates in arthropods, nor is there evidence that the regulatory interactions between Notch, Wnt and caudal family genes resemble those seen in vertebrates. ICN, intracellular domain of Notch; S0, newly forming somite; SI/II, formed somites. 912 | DECEMBER 2005 | VOLUME 6 www.nature.com/reviews/genetics REVIEWS � 2005 Nature Publishing Group PROSOMA The anterior part of the body in chelicerates, including the head, the mouthparts and the walking legs. OPISTHOSOMA The posterior part of the body in chelicerates. It does not include any walking legs. GEOPHILOMORPHS A group of centipedes, normally soil dwelling, that are characterized by a long, thin body made up of many segments (27?191). species, including a hemipteran (O. fasciatus) 31 , and a parasitic hymenopteran (C. floridanium) 76 , eve homo- logues are expressed in segmental stripes. By contrast, in S. americana an eve homologue is not expressed in stripes at all, but only in a broad posterior domain 77 . A homologue of the pair-rule gene ftz is similarly not expressed in stripes in S. gregaria 82 , but does seem to be in the primitive insect Thermobia domestica 83 . There is molecular evidence that the protein domains required for efficient pair-rule function in D. melanogaster FTZ are missing in the Schistocerca spp. homologue 84 . Interestingly, ftz is expressed in a pair-rule pattern in T. castaneum 85 , but its expression is not necessary for segmentation. Notwithstanding changes in the role of individual pair-rule genes, it seems likely that the generation of segments by subdivision of a transient double segmen- tal unit is ancestral to the insects, or at least most of them. Pair-rule expression has been recorded for hairy (h) 79 and eve 78 in T. castaneum, and for a paired box gene 3/7 (pax3/7) in S. americana 75 . Non-insect arthropods. The situation in arthropods other than insects is not yet clear. Expression of pair-rule gene homologues is consistent with a role in segmenta- tion in a wide range of arthropods, including chelicer- ates 38?40 and myriapods 17,42 . However, in most cases, the expression data have been interpreted as showing segmentally repeated stripes, not pair-rule patterns 38 . One exception is the expression of a pax3/7 homologue in the PROSOMA of the spider mite Tetranychus urticae, which is pair-rule, even though expression of the same gene in the OPISTHOSOMA seems to be segmental 40 . However, in another chelicerate, the spider Cupiennius salei, expression of a pax3/7 homologue seems to be in segmental, not pair-rule, stripes 38 . Another exception is the myriapod S. maritima uniF6AEBOX 2BuniF6AF, which is a GEOPHILOMORPH centipede that generates a large number of segments as an embryo 86 . Several genes, including a homologue of the pair-rule gene odd skipped (odd), reveal that initial patterning of the entire trunk involves a double-segment repeat that is subsequently subdivided to generate individual segments 42 . The existence of pair-rule patterning in both cen- tipedes and insects, two distantly related classes of arthropods, could be taken as evidence that a double- segment repeat pattern is ancestral to arthropods. However, geophilomorph centipedes are a derived group even among the myriapods, and might be a special case. It is possible that the geophilomorphs as a group have evolved a segment-doubling step to increase segment numbers 42,87 . However, one observation argues against this. Geophilomorphs share with all centipedes the constraint that no matter how much segment num- bers vary they always possess an even number of trunk segments (including the segment carrying the poison claw plus an odd number of leg-bearing segments) 88 . If, as we have suggested, this constraint reflects the initial generation of double-segment units, then this trait too would be ancestral to the centipedes. Whether or not geophilomorphs are exceptional among myriapods, it seems extremely unlikely that pair-rule expression of genes in S. maritima is regu- lated by a series of gap genes that are analogous to those in D. melanogaster ? all the available evidence indicates that something more akin to an oscilla- tor is active during segmentation (see below). It is also far from clear whether the resolution of the pair-rule stripes to yield a single-segment repeat is homologous in any way to what happens in D. melanogaster 42,87 . Beyond the Drosophila paradigm Parallels with vertebrate somitogenesis. Until recently, comparative studies of segmentation in arthropods have focused on the homologues of D. melanogaster segmentation genes. The inherent problem with this candidate-gene approach is that genes will be over- looked if their role in segmentation has been lost or replaced by a novel mechanism in D. melanogaster. Recent discoveries in chelicerates and myriapods indi- cate that genes of the Notch signalling pathway fall into exactly this category 2,3,89 . Notch signalling is not thought to be involved in the primary segmentation process in D. melanogaster. However, in vertebrates, somite patterning uses a segmentation clock, or oscillator, which is depend- ent on the function of hairy/enhancer of split (HES) family transcription factors, and genes of the Notch pathway uniF6AEBOX 3auniF6AF. In spiders 2,3 and in the centipede Strigamia maritima (A.D.C. et al., unpublished observations) Notch pathway genes show patterned expression very early in the segmentation process, before expression of segment polarity genes. In spiders, Notch signalling is required for segment formation, and for the resolution of patterned expression of h, which is itself the homologue of a D. melanogaster pair-rule gene 2,3 . In the centipede, the expression patterns indicate that the expression is dynamic, with many cycles of gene expression generating more than 40 trunk segments. These data would be consistent with the existence of a Notch-dependent oscillator that generates the pri- mary segment pattern in myriapods and chelicerates uniF6AEBOX 3buniF6AF. This model is attractive because it explains how posterior segments can arise in the cellular environment of sequentially segmenting arthro- pods. By analogy with vertebrates, homologues of the primary pair-rule genes ? and in particular h ? might function within the clock mechanism, or downstream of it 2,3,39 . At present there is no evidence that the Notch signalling pathway is involved in the formation of posterior segments in the germ bands of insects, and for T. castaneum there are unpublished (but cited 89 ) claims that it is not. However, there is equally no evidence that the homologues of D. melanogaster gap genes function during segment patterning in chelicerates and myriapods. The discovery of similarities between the mecha- nisms that control posterior sequential segmentation in NATURE REVIEWS | GENETICS VOLUME 6 | DECEMBER 2005 | 913 FOCUS ON THE BODY PLAN � 2005 Nature Publishing Group some arthropods and that control somitogenesis in vertebrates indicates further candidate genes for which the expression pattern should be examined in arthropods. Wnts and fibroblast growth factors (FGFs) are such candidates. In vertebrates, FGFs are involved in establishing a wavefront 90 that is defined by a threshold level of signalling, below which oscilla- tions of the clock cease. The intensity of signalling is graded from a posterior source, so the level of signal- ling defines the position at which somites are stably patterned. Wnt signalling functions upstream of the Notch-signalling-dependent segmentation clock, the FGF-dependent wavefront, and posteriorly expressed cad-related genes during somitogenesis 90?92 uniF6AEBOX 3buniF6AF. Wnt signalling is also known to be involved in the A?P patterning of other deuterostomes 93 . There are already data showing that wingless function is needed for sequential segmentation in both G. bimaculatus 29 and O. fasciatus 34 , but its specific role in this process is unclear. Conclusions Our quest to understand the evolution of arthropod segmentation mechanisms is still in its infancy; some of the important questions that remain unanswered are outlined in BOX 4. However, several general conclusions can already be made. The definitive segmentation of arthropods reflects the conserved expression of en and other segment polarity genes ? a role that these genes presumably acquired before the radiation of the main arthropod groups. Homologues of some of the D. melanogaster pair-rule genes were also involved in the segmenta- tion of the arthropod common ancestor, but exactly which genes were involved, and whether that animal used a pair-rule segmental pre-pattern, remains unclear. The role of maternal factors in D. melanogaster is not representative, even of all insects. A maternally derived anterior gradient might have evolved more than once in insects, but the use of BCD for this purpose is an invention of the higher Diptera. The involvement of cad in segment patterning, or at least growth of the segmenting primordium, is probably a characteristic that is ancestral to arthropods. The ancestral role of cad probably extended more anteriorly than it does in D. melanogaster, to the development of most or all trunk segments. In a range of insects, homologues of several of the gap genes are involved in regionalization of the early embryo. A role in regulating Hox gene expres- sion seems to be broadly conserved, but it remains unclear when they acquired the role of instructing the downstream expression of segmentation genes. An alternative mechanism of segment generation might be operating in the trunk regions of chelicer- ates and myriapods. This involves the Notch signalling pathway, which might indicate an ancestral role for an arthropod segmentation clock that is at least analogous to that operating in vertebrates. Opinions differ as to how conserved the basic mechanisms of segmentation will prove to be, but it is already evident that there has been significant diver- gence during arthropod evolution in the function of some of the best-known genes of the D. melanogaster segmentation cascade. One must be careful to avoid the assumption that the D. melanogaster pattern in some way represents an evolutionary endpoint, and that other species represent intermediate stages in a progression towards this endpoint. More data will be required, and from a wider range of arthropods, before we can say with any certainty what the ancestral mechanism of arthropod segmentation might have been, and how it has been modified in different groups. Box 4 | Unresolved questions and future avenues of research The study of Drosophila melanogaster is limited in its ability to answer general questions about arthropod segmentation and its evolutionary history. However, the study of an increasing number of arthropods, using a wider range of approaches, will allow many new questions to be addressed. The advent of RNAi and transgenesis in non- model organisms will also facilitate this process. Some of the key areas that need to be addressed are as follows: ? We understand very little about the cellular dynamics of the ?growth zone? in sequentially segmenting arthropods uniF6AEBOX 2uniF6AF. A much clearer understanding of the basic embryology of some insects is required before gene-expression patterns can be properly interpreted. In many cases, reliable fate maps are desperately needed. Cell-labelling experiments will be important to generate such maps. ? Which signalling pathways are involved in posterior elongation? Are Notch, Wnt or fibroblast growth factor pathways involved in basally branching insects and other arthropods? ? Is an oscillator involved in the generation of new segments in the growth zone? Are gene or protein levels oscillating, and if so what is the primary oscillator? Cell-labelling experiments and reporter constructs can provide more data. The development of techniques for reporting gene expression in live embryos will also be invaluable for tracking dynamic gene expression. ? Is head segmentation achieved through a separate mechanism to that generating trunk segmentation? Are there two separate mechanisms in ?intermediate germ band? arthropods? ? What is the role of the mesoderm in generating posterior segments? Are the mesoderm and the ectoderm patterned independently? Is the mesoderm required for the segmentation of the ectoderm, or vice versa? These questions are of particular interest, given the similarities that are observed between posterior segmentation in some arthropods and the segmentation of the presomitic mesoderm in vertebrates. The control of segmentation in the mesoderm has been largely ignored. 914 | DECEMBER 2005 | VOLUME 6 www.nature.com/reviews/genetics REVIEWS � 2005 Nature Publishing Group 1. Aguinaldo, A. M. et al. Evidence for a clade of nematodes, arthropods and other moulting animals. Nature 387, 489?493 (1997). 2. Stollewerk, A., Schoppmeier, M. & Damen, W. G. M. Involvement of Notch and Delta genes in spider segmentation. Nature 423, 863?865 (2003). The first report of the involvement of the Notch signalling pathway in segmentation in any arthropod species. 3. Schoppmeier, M. & Damen, W. G. M. Suppressor of Hairless and Presenilin phenotypes imply involvement of canonical Notch-signalling in segmentation of the spider Cupiennius salei. Dev. Biol. 280, 211?224 (2005). 4. Lawrence, P. A. The Making of a Fly: the Genetics of Animal Design (Blackwell Scientific, Oxford, UK, 1992). 5. Maderspacher, F., Bucher, G. & Klingler, M. Pair-rule and gap gene mutants in the flour beetle Tribolium castaneum. Dev. Genes Evol. 208, 558?568 (1998). 6. Pultz et al. A genetic screen for zygotic embryonic lethal mutations affecting cuticular morphology in the wasp Nasonia vitripennis. Genetics 154, 1213?1229 (2000). 7. Pultz, M. A. & Leaf, D. S. The jewel wasp Nasonia: querying the genome with haplo-diploid genetics. Genesis 35, 185?191 (2003). 8. Bucher, G., Scholten, J. & Klingler, M. Parental RNAi in Tribolium (Coleoptera). Curr. Biol. 12, R85?R86 (2002). A report on the application of parental RNAi in a non-model arthropod species. This technique has since been used successfully in several other species. 9. Eckert, C., Aranda, M., Wolff, C. & Tautz, D. Separable stripe enhancer elements for the pair-rule gene hairy in the beetle Tribolium. EMBO Rep. 5, 638?642 (2004). 10. Pavlopoulos, A. & Averof, M. Establishing genetic transformation for comparative developmental studies in the crustacean Parhyale hawaiensis. Proc. Natl Acad. Sci. USA 102, 7888?7893 (2005). This article reports the first example of genetic transformation in the crustacean Parhyale hawaiiensis. This is one of very few species, and the only non-insect arthropod, for which this technique has been developed. 11. Shinmyo, Y. et al. piggyBac-mediated somatic transformation of the two-spotted cricket, Gryllus bimaculatus. Dev. Growth Differ. 46, 343?349 (2004). 12. Raff, R. A. The Shape of Life: Genes, Development, and the Evolution of Animal Form (Univ. Chicago Press, Chicago, 1996). 13. Martinez-Arias, A. & Lawrence, P. A. Parasegments and compartments in the Drosophila embryo. Nature 313, 639?642 (1985). 14. Damen, W. G. Parasegmental organization of the spider embryo implies that the parasegment is an evolutionary conserved entity in arthropod embryogenesis. Development 129, 1239?1250 (2002). 15. von Dassow, G., Meir, E., Munro, E. M. & Odell, G. M. The segment polarity network is a robust developmental module. Nature 406, 188?192 (2000). 16. Simonnet, F., Deutsch, J. & Queinnec, E. hedgehog is a segment polarity gene in a crustacean and a chelicerate. Dev. Genes Evol. 214, 537?545 (2004). 17. Hughes, C. L. & Kaufman, T. C. Exploring myriapod segmentation: the expression patterns of even-skipped, engrailed, and wingless in a centipede. Dev. Biol. 247, 47?61 (2002). 18. Nagy, L. M. & Carroll, S. Conservation of wingless patterning functions in the short-germ embryos of Tribolium castaneum. Nature 367, 460?463 (1994). 19. Dearden, P. K. & Akam, M. Early embryo patterning in the grasshopper, Schistocerca gregaria: wingless, decapentaplegic and caudal expression. Development 128, 3435?3444 (2001). 20. Janssen, R., Prpic, N.-M. & Damen, W. G. M. Gene expression suggests decoupled dorsal and ventral segmentation in the millipede Glomeris marginata (Myriapoda: Diplopoda). Dev. Biol. 268, 89?104 (2004). 21. Galis, F., van Dooren, T. J. & Metz, J. A. Conservation of the segmented germband stage: robustness or pleiotropy? Trends Genet. 18, 504?509 (2002). 22. Oppenheimer, D. I., MacNicol, A. M. & Patel, N. H. Functional conservation of the wingless?engrailed interaction as shown by a widely applicable baculovirus misexpression system. Curr. Biol. 9, 1288?1296 (1999). 23. Beye, M., Hartel, S., Hagen, A., Hasselmann, M. & Omholt, S. W. Specific developmental gene silencing in the honey bee using a homeobox motif. Insect Mol. Biol. 11, 527?532 (2002). 24. Davis, G. K. & Patel, N. H. Short, long, and beyond: molecular and embryological approaches to insect segmentation. Ann. Rev. Entomol. 47, 669?699 (2002). 25. Bucher, G. & Klingler, M. Divergent segmentation mechanism in the short germ insect Tribolium revealed by giant expression and function. Development 131, 1729?1740 (2004). 26. Schr�der, R. The genes orthodenticle and hunchback substitute for bicoid in the beetle Tribolium. Nature 422, 621?625 (2003). A functional analysis of the gap gene hunchback in the beetle T. castaneum. This analysis can be compared to that of the milkweed bug in reference 32. 27. Grbic, M. Polyembryony in parasitic wasps: evolution of a novel mode of development. Int. J. Dev. Biol. 47, 633?642 (2003). 28. Shinmyo, Y. et al. caudal is required for gnathal and thoracic patterning and for posterior elongation in the intermediate-germband cricket Gryllus bimaculatus. Mech. Dev. 122, 231?239 (2005). 29. Miyawaki, K. et al. Involvement of Wingless/Armadillo signaling in the posterior sequential segmentation in the cricket, Gryllus bimaculatus (Orthoptera), as revealed by RNAi analysis. Mech. Dev. 121, 119?130 (2004). 30. Mito, T. et al. Non-canonical functions of hunchback in segment patterning of the intermediate germ cricket Gryllus bimaculatus. Development 132, 2069?2079 (2005). 31. Liu, P. Z. & Kaufman, T. C. even-skipped is not a pair-rule gene but has segmental and gap-like functions in Oncopeltus fasciatus, an intermediate germband insect. Development 132, 2081?2092 (2005). 32. Liu, P. Z. & Kaufman, T. C. hunchback is required for suppression of abdominal identity, and for proper germband growth and segmentation in the intermediate germband insect Oncopeltus fasciatus. Development 131, 1515?1527 (2004). A functional analysis of the gap gene hunchback in the milkweed bug O. fasciatus. This analysis can be compared to that of the beetle T. castaneum in reference 26. 33. Liu, P. Z. & Kaufman, T. C. Kr�ppel is a gap gene in the intermediate germband insect Oncopeltus fasciatus and is required for development of both blastoderm and germband-derived segments. Development 131, 4567?4579 (2004). A functional analysis of the gap gene Kr�ppel in O. fasciatus. 34. Angelini, D. R. & Kaufman, T. C. Functional analyses in the milkweed bug Oncopeltus fasciatus (Hemiptera) support a role for Wnt signaling in body segmentation but not appendage development. Dev. Biol. 283, 409?423 (2005). 35. Copf, T., Schroder, R. & Averof, M. Ancestral role of caudal genes in axis elongation and segmentation. Proc. Natl Acad. Sci. USA 101, 17711?17715 (2004). 36. Regier, J. C., Shultz, J. W. & Kambic, R. E. Pancrustacean phylogeny: hexapods are terrestrial crustaceans and maxillopods are not monophyletic. Proc. Biol. Sci. 272, 395?401 (2005). 37. Cook, C. E., Smith, M. L., Telford, M. J., Bastianello, A. & Akam, M. Hox genes and the phylogeny of the arthropods. Curr. Biol. 11, 759?763 (2001). 38. Schoppmeier, M. & Damen, W. G. M. Expression of Pax group III genes suggests a single-segmental periodicity for opisthosomal segment patterning in the spider Cupiennius salei. Evol. Dev. 7, 160?169 (2005). This study shows that the Pax group III genes ? one member of which is a pair-rule gene in D. melanogaster ? are expressed with a single-segment periodicity in a chelicerate. 39. Damen, W. G., Weller, M. & Tautz, D. Expression patterns of hairy, even-skipped, and runt in the spider Cupiennius salei imply that these genes were segmentation genes in a basal arthropod. Proc. Natl Acad. Sci. USA 97, 4515?4519 (2000). 40. Dearden, P. K., Donly, C. & Grbic, M. Expression of pair-rule gene homologues in a chelicerate: early patterning of the two-spotted spider mite Tetranychus urticae. Development 129, 5461?5472 (2002). The authors show that at least one homologue of a D. melanogaster pair-rule gene is expressed with a double-segment periodicity in the prosoma of a chelicerate. 41. Chipman, A. D., Arthur, W. & Akam, M. Early development and segment formation in the centipede Strigamia maritima (Geophilomorpha). Evol. Dev. 6, 78?89 (2004). 42. Chipman, A. D., Arthur, W. & Akam, M. A double segment periodicity underlies segment generation in centipede development. Curr. Biol. 14, 1250?1255 (2004). The authors show that caudal and an odd-skipped- related gene are expressed with a double-segment periodicity, and suggest an involvement of the Notch signalling pathway in centipede segmentation. 43. Driever, W. & Nusslein-Volhard, C. A gradient of bicoid protein in Drosophila embryos. Cell 54, 83?93 (1988). 44. Struhl, G., Struhl, K. & Macdonald, P. M. The gradient morphogen bicoid is a concentration-dependent transcriptional activator. Cell 57, 1259?1273 (1989). 45. Wharton, R. P. & Struhl, G. Structure of the Drosophila BicaudalD protein and its role in localizing the the posterior determinant nanos. Cell 59, 881?892 (1989). 46. Pokrywka, N. J. & Stephenson, E. C. Microtubules mediate the localization of bicoid RNA during Drosophila oogenesis. Development 113, 55?66 (1991). 47. Irish, V., Lehmann, R. & Akam, M. The Drosophila posterior-group gene nanos functions by repressing hunchback activity. Nature 338, 646?648 (1989). 48. Murata, Y. & Wharton, R. P. Binding of pumilio to maternal hunchback mRNA is required for posterior patterning in Drosophila embryos. Cell 80, 747?756 (1995). 49. Gavis, E. R. & Lehmann, R. Translational regulation of nanos by RNA localization. Nature 369, 315?318 (1994). 50. St Johnston, D. & Nusslein-Volhard, C. The origin of pattern and polarity in the Drosophila embryo. Cell 68, 201?219 (1992). 51. Sommer, R. & Tautz, D. Segmentation gene expression in the housefly Musca domestica. Development 113, 419?430 (1991). 52. Stauber, M., Taubert, H. & Schmidt-Ott, U. Function of bicoid and hunchback homologs in the basal cyclorrhaphan fly Megaselia (Phoridae). Proc. Natl Acad. Sci. USA 97, 10844?10849 (2000). 53. Stauber, M., Prell, A. & Schmidt-Ott, U. A single Hox3 gene with composite bicoid and zerknullt expression characteristics in non-Cyclorrhaphan flies. Proc. Natl Acad. Sci. USA 99, 274?279 (2002). 54. Stauber, M., Jackle, H. & Schmidt-Ott, U. The anterior determinant bicoid of Drosophila is a derived Hox class 3 gene. Proc. Natl Acad. Sci. USA 96, 3786?3789 (1999). The first report that bicoid is a derived Hox gene, and that it might be specific to Diptera. 55. Brown, S. et al. A strategy for mapping bicoid on the phylogenetic tree. Curr. Biol. 11, R43?R44 (2001). 56. Lynch, J. & Desplan, C. Evolution of development: beyond bicoid. Curr. Biol. 13, R557?R559 (2003). 57. Handel, K., Grunfelder, C. G., Roth, S. & Sander, K. Tribolium embryogenesis: a SEM study of cell shapes and movements from blastoderm to serosal closure. Dev. Genes Evol. 210, 167?179 (2000). 58. Handel, K., Basal, A., Fan, X. & Roth, S. Tribolium castaneum twist: gastrulation and mesoderm formation in a short-germ beetle. Dev. Genes Evol. 215, 13?31 (2005). This paper describes the origin and behaviour of mesodermal cells during development in the flour beetle. More data of this kind are needed. 59. Wolff, C., Sommer, R., Schroder, R., Glaser, G. & Tautz, D. Conserved and divergent expression aspects of the Drosophila segmentation gene hunchback in the short germ band embryo of the flour beetle Tribolium. Development 121, 4227?4236 (1995). 60. Bucher, G., Farzana, L., Brown, S. J. & Klingler, M. Anterior localization of maternal mRNAs in a short germ insect lacking bicoid. Evol. Dev. 7, 142?149 (2005). 61. Ho, K., Dunin-Borkowski, O. M. & Akam, M. Cellularization in locust embryos occurs before blastoderm formation. Development 124, 2761?2768 (1997). 62. Lall, S., Ludwig, M. Z. & Patel, N. H. Nanos plays a conserved role in axial patterning outside of the diptera. Curr. Biol. 13, 224?229 (2003). 63. Sander, K. Specification of the basic body pattern in insect embryogenesis. Adv. Insect Physiol. 12, 125?238 (1976). 64. Dubnau, J. & Struhl, G. RNA recognition and translational regulation by a homeodomain protein. Nature 379, 694?699 (1996). 65. Rivera-Pomar, R., Niessing, D., Schmidt-Ott, U., Gehring, W. J. & Jackle, H. RNA binding and translational suppression by bicoid. Nature 379, 746?749 (1996). 66. Macdonald, P. M. & Struhl, G. A molecular gradient in early Drosophila embryos and its role in specifying the body pattern. Nature 324, 537?545 (1986). 67. Hulskamp, M. & Tautz, D. Gap genes and gradients ? the logic behind the gaps. Bioessays 13, 261?268 (1991). 68. Fujioka, M., Emi-Sarker, Y., Yusibova, G. L., Goto, T. & Jaynes, J. B. Analysis of an even-skipped rescue transgene reveals both composite and discrete neuronal and early blastoderm enhancers, and multi-stripe positioning by gap gene repressor gradients. Development 126, 2527?2538 (1999). 69. Irish, V. F., Martinez-Arias, A. & Akam, M. Spatial regulation of the Antennapedia and Ultrabithorax homeotic genes during Drosophila early development. EMBO J. 8, 1527?1537 (1989). NATURE REVIEWS | GENETICS VOLUME 6 | DECEMBER 2005 | 915 FOCUS ON THE BODY PLAN � 2005 Nature Publishing Group 70. White, R. A. & Lehmann, R. A gap gene, hunchback, regulates the spatial expression of Ultrabithorax. Cell 47, 311?321 (1986). 71. Isshiki, T., Pearson, B., Holbrook, S. & Doe, C. Q. Drosophila neuroblasts sequentially express transcription factors which specify the temporal identity of their neuronal progeny. Cell 106, 511?521 (2001). 72. Goltsev, Y., Hsiong, W., Lanzaro, G. & Levine, M. Different combinations of gap repressors for common stripes in Anopheles and Drosophila embryos. Dev. Biol. 275, 435?446 (2004). The authors show that homologous posterior pair-rule stripes of expression are probably regulated by a different spatial combination of gap-gene repressors in Anopheles gambiae, when compared with Drosophila melanogaster. 73. Patel, N. H. et al. Grasshopper hunchback expression reveals conserved and novel aspects of axis formation and segmentation. Development 128, 3459?3472 (2001). 74. Davis, G. K., D?alessio, J. A. & Patel, N. H. Pax3/7 genes reveal conservation and divergence in the arthropod segmentation hierarchy. Dev. Biol. 285, 169?184 (2005). A description of the Pax3/7 family of pair-rule gene homologues in various arthropod classes, with a discussion of their ancestral function. 75. Davis, G. K., Jaramillo, C. A. & Patel, N. H. Pax group III genes and the evolution of insect pair-rule patterning. Development 128, 3445?3458 (2001). 76. Grbic, M., Nagy, L. M., Carroll, S. B. & Strand, M. Polyembryonic development: insect pattern formation in a cellularized environment. Development 122, 795?804 (1996). 77. Patel, N. H., Ball, E. E. & Goodman, C. S. Changing role of even-skipped during the evolution of insect pattern formation. Nature 357, 339?342 (1992). 78. Patel, N. H., Condron, B. G. & Zinn, K. Pair-rule expression patterns of even-skipped are found in both short- and long-germ beetles. Nature 367, 429?434 (1994). 79. Sommer, R. J. & Tautz, D. Involvement of an orthologue of the Drosophila pair-rule gene hairy in segment formation of the short germ-band embryo of Tribolium (Coleoptera). Nature 361, 448?450 (1993). 80. Macdonald, P. M., Ingham, P. & Struhl, G. Isolation, structure, and expression of even-skipped: a second pair-rule gene of Drosophila containing a homeo box. Cell 47, 721?734 (1986). 81. Schr�der, R., Jay, D. G. & Tautz, D. Elimination of EVE protein by CALI in the short germ band insect Tribolium suggests a conserved pair-rule function for even skipped. Mech. Dev. 80, 191?195 (1999). 82. Dawes, R., Dawson, I., Falciani, F., Tear, G. & Akam, M. Dax, a locust Hox gene related to fushi-tarazu but showing no pair-rule expression. Development 120, 1561?1572 (1994). 83. Hughes, C. L., Liu, P. Z. & Kaufman, T. C. Expression patterns of the rogue Hox genes Hox3/zen and fushi tarazu in the apterygote insect Thermobia domestica. Evol. Dev. 6, 393?401 (2004). 84. Alonso, C. R., Maxton-Kuechenmeister, J. & Akam, M. Evolution of Ftz protein function in insects. Curr. Biol. 11, 1473?1478 (2001). 85. Brown, S. J., Hilgenfeld, R. B. & Denell, R. E. The beetle Tribolium castaneum has a fushi tarazu homolog expressed in stripes during segmentation. Proc. Natl Acad. Sci. USA 91, 12922?12926 (1994). 86. Arthur, W. & Chipman, A. D. The centipede Strigamia maritima: what it can tell us about the development and evolution of segmentation. Bioessays 27, 653?660 (2005). 87. Damen, W. G. Arthropod segmentation: why centipedes are odd. Curr. Biol. 14, R557?R559 (2004). 88. Arthur, W. & Farrow, M. The pattern of variation in centipede segment number as an example of developmental constraint in evolution. J. Theor. Biol. 200, 183?191 (1999). 89. Tautz, D. Segmentation. Dev. Cell 7, 301?312 (2004). 90. Aulehla, A. & Herrmann, B. G. Segmentation in vertebrates: clock and gradient finally joined. Genes Dev. 18, 2060?2067 (2004). 91. Aulehla, A. et al. Wnt3a plays a major role in the segmentation clock controlling somitogenesis. Dev. Cell 4, 395?406 (2003). 92. Shimizu, T., Bae, Y. K., Muraoka, O. & Hibi, M. Interaction of Wnt and caudal-related genes in zebrafish posterior body formation. Dev. Biol. 279, 125?141 (2005). 93. Holland, L. Z. Heads or tails? Amphioxus and the evolution of anterior?posterior patterning in deuterostomes. Dev. Biol. 241, 209?228 (2002). 94. Pourquie, O. The segmentation clock: converting embryonic time into spatial pattern. Science 328?330 (2003). 95. Cerny, A., Bucher, G., Schroder, R. & Klingler, M. Breakdown of abdominal patterning in the Tribolium Kr�ppel mutant jaws. Development (in the press). Competing interests statement The authors declare no competing financial interests. Online links DATABASES The following terms in this article are linked online to: Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query. fcgi?db=gene bcd | CAD | en | eve | ftz | gt | h | hh | kni | Kr | nos | prd | run | tll | wg | zen Access to this interactive links box is free online. 916 | DECEMBER 2005 | VOLUME 6 www.nature.com/reviews/genetics REVIEWS "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment