195 Neurogenesis in myriapods and chelicerates and its importance for understanding arthropod relationships Angelika Stollewerk1,2 and Ariel D. Chipman Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK

Synopsis Several alternative hypotheses on the relationships between the major arthropod groups are still being discussed. We reexamine here the chelicerate/myriapod relationship by comparing previously published morphological data on neuro- genesis in the euarthropod groups and presenting data on an additional myriapod (Strigamia maritima). Although there are differences in the formation of neural precursors, most euarthropod species analyzed generate about 30 single neural precursors (insects/crustaceans) or precursor groups (chelicerates/myriapods) per hemisegment that are arranged in a regular pattern. The genetic network involved in recruitment and specification of neural precursors seems to be conserved among euarthropods. Furthermore, we show here that neural precursor identity seems to be achieved in a similar way. Besides these conserved features we found 2 characters that distinguish insects/crustaceans from myriapods/chelicerates. First, in insects and crustaceans the neuroectoderm gives rise to epidermal and neural cells, whereas in chelicerates and myriapods the central area of the neuroectoderm exclusively generates neural cells. Second, neural cells arise by stem-cell-like divisions of neuroblasts in insects and crustaceans, whereas groups of mainly postmitotic neural precursors are recruited for the neural fate in chelicerates and myriapods. We discuss whether these characteristics represent a sympleisiomorphy of myriapods and chelicerates that has been lost in the more derived Pancrustacea or whether these characteristics are a synapomorphy of myriapods and chelicerates, providing the first morphological support for the Myriochelata group.

Introduction myriapods (Field and others 1988; Turberville and The relationships between and within the major others 1991; Ballard and others 1992; Friedrich and arthropod groups have not been consistently resolved. Tautz 1995; Giribet and Ribera 1998). Recent data Several alternative hypotheses are being discussed. on comparative developmental biology support this The so-called Mandibulata hypothesis suggests a molecular sister group relationship, although the clade composed of insects, crustaceans, and myriapods synapomorphies seem to be shared mainly by insects with various ideas as to the relationships within this and malacostracans (Dohle and Scholtz 1988; Patel and clade. The Pancrustacea hypothesis assumes a crusta- others 1989; Whitington and others 1993; Osorio and cean origin of insects or a sister group relationship others 1995; Whitington 1996; Dohle 1998; Nilsson between both groups (Zrzavy´ and Sˇtys 1997; Shultz and Osorio 1998; Duman-Scheel and Patel 1999; and Regier 2000; Dohle 2001; Mallatt and others Dohle 2001). Several independent phylogenetic ana- 2004; Regier and others 2005), and the Atelocerata lyses based on molecular data support a chelicerate/ hypothesis unites insects and myriapods as a clade myriapod sister group relationship, the so-called (Snodgrass 1938, 1950, 1951; Briggs and Fortey 1989; Myriochelata hypothesis (Friedrich and Tautz 1995; Schram and Emerson 1991; Bergstro¨m 1992; Wheeler Hwang and others 2001; Kusche and Burmester WC and others 1993; Kraus O and Kraus M 1994, 1996; 2001; Nardi and others 2003; Mallatt and others Emerson and Schram 1998; Wheeler WC 1998; Wills 2004; Pisani and others 2004), a link that had never and others 1998; Bitsch C and Bitsch J 2004). Whereas been considered by comparison of morphological the Atelocerata hypothesis is mainly supported by structures. However, recent comparative studies on morphological evidence, the idea of Pancrustacea neurogenesis in the diplopod Glomeris marginata was initially based on the phylogenetic analysis of and the chilopod Lithobius forficatus have shown that ribosomal-RNA sequence data in which crustaceans the myriapods and the chelicerates share several fea- and insects grouped together to the exclusion of tures that cannot be found in homologous form in

From the symposium “The New Microscopy: Toward a Phylogenetic Synthesis” presented at the annual meeting of the Society for Integrative and Comparative Biology, January 4–8, 2005, at San Diego, California. 1 E-mail: [email protected] 2 Present address: Johannes-Gutenberg University Mainz, Department of Genetics, Johann-Joachim-Becherweg 32, 55099 Mainz, Germany. Integrative and Comparative Biology, volume 46, number 2, pp. 195–206 doi:10.1093/icb/icj020 Advance Access publication February 16, 2006 Ó The Society for Integrative and Comparative Biology 2006. All rights reserved. For permissions, please email: journals.permissions@ oxfordjournals.org. 196 A. Stollewerk and A. D. Chipman

hemisegment at the beginning of neurogenesis. In a second step, proneural gene expression is restricted to a single cell of the cluster, the future neuroblast (Cabrera and others 1987; Romani and others 1987; insects Skeath and others 1992). This process is called lateral inhibition and is mediated by the neurogenic genes Notch and Delta (Simpson 1990; Martin-Bermudo and others 1995; Heitzler and others 1996; Seugnet and others 1997). It has been predicted that proneural crustaceans gene expression is higher in a particular cell of the proneural cluster as a result of predetermination or an extrinsic signal. Since the proneural genes activate the expression of Delta, Delta is also up-regulated in this cell. Delta binds to Notch and activates Notch in chelicerates/myriapods the neighboring cells, which eventually leads to the activation of the E(spl) genes. The gene products of Fig. 1 Differences in the formation of neural precursors in this complex repress proneural gene expression, the arthropod groups. In insects and crustaceans, single which in turn leads to a down-regulation of Delta in neural precursors (neuroblasts) are specified. Whereas insect neuroblasts delaminate into the embryo shortly neighboring cells (Nakao and Campos-Ortega 1996; after formation, crustacean neuroblasts remain in the Ligoxygakis and others 1998). As a result of this feed- outer cell layer (neuroectoderm) and divide to give rise back loop, proneural gene expression is maintained in to ganglion mother cells that are pushed into the the neuroblast but down-regulated in the remaining interior of the embryo by directed mitosis. In both cells of the proneural cluster. Although this model chelicerates and myriapods, groups of neural precursors are selected and form invagination sites that eventually predicts a higher expression of Delta in single cells detach from the apical surface and differentiate into (presumptive neuroblasts), it has not been demon- neural cells. strated that Delta transcripts accumulate at higher levels in individual cells within the proneural clusters. Once a neuroblast is determined, it delaminates into insects and crustaceans. The most distinctive difference the embryo and divides asymmetrically to produce is that groups of neural precursors are singled out from ganglion mother cells (Goodman and Doe 1993). The the neuroectoderm of the spider and the myriapods, ganglion mother cells divide only once to give rise to rather than individual cells (that is, neuroblasts) as in neural cells that differentiate into neurons and glia. insects or crustaceans (Fig. 1) (Cupiennius salei: The neuroblasts do not delaminate all at once but Stollewerk and others 2001; Limulus polyphemus: in 5 discrete waves. Each neuroblast has a distinct iden- Mittmann 2002; C. salei: Stollewerk 2002; Stollewerk tity and gives rise to an invariant lineage of neural and others 2003; G. marginata: Dove and Stollewerk progenies. The identity of the neuroblasts is specified 2003; L. forficatus: Kadner and Stollewerk 2004). in the ventral neuroectoderm by segment polarity and Here we give an overview of the modes of neuro- dorsoventral patterning genes (see review in Skeath genesis in the major arthropod groups with special 1999). focus on myriapods and chelicerates. Furthermore, Neurogenesis has also been studied in insects we present new data on the geophilomorph centipede other than Drosophila. The pattern of neuroblasts is Strigamia maritima (Myriapoda) and discuss the data similar in all insects analyzed: they are arranged in in a phylogenetic context. 7 anteroposterior rows with 3–6 neuroblasts each (Bate 1976; Broadus and Doe 1995; Wheeler SR Neural precursor formation and others 2003). It has been shown in Tribolium in insects castaneum and in Schistocerca americana that single Neurogenesis has been studied in detail in the insect neuroblasts are selected in sequential waves, similar Drosophila melanogaster. The ventral neuroectoderm to D. melanogaster (Broadus and Doe 1995; Wheeler of the Drosophila embryo gives rise to both neural SR and others 2003). Within the insect group, and ectodermal cells (Jime´nez and Campos-Ortega proneural genes have been identified in several 1979, 1990; Cabrera and others 1987). The competence Diptera, a butterfly, and the flour beetle T. castaneum to adopt the neural fate depends on the presence of (Precis coenia: Galant and others 1998; Ceratitis the proneural genes achaete, scute, and lethal of scute. capitata:Wu¨lbeck and Simpson 2000; Calliphora These genes are expressed in clusters of cells in each vicina: Pistillo and others 2002; Phormia terranovae: Neurogenesis in myriapods and chelicerates 197

Skaer and others 2002; Anopheles gambiae:Wu¨lbeck 1 and Simpson 2002; T. castaneum: Wheeler SR and 1 2 others 2003). 2 3 3 4 Neural precursor formation in 4 5 malacostracan crustaceans 5 6 Neuroblasts have also been described in malacostracan 6 7 7 crustaceans and exist perhaps also in branchiopods (Leptochelia spp.: Dohle 1972; Diastylis rathkei: Dohle Glomeris marginata Cupiennius salei 1976; Neomysis integer: Scholtz 1984; Peracarida: Dohle and Scholtz 1988; Gammarus pulex: Scholtz 1990; Fig. 2 The invaginating neural precursor groups show Cherax destructor: Scholtz 1992; Leptodora kindti: a similar arrangement in chelicerates and in myriapods. Gerberding 1997; Decapoda: Harzsch and others Neural precursor groups are arranged in 7 rows with 3–6 invagination sites each, in both chelicerates and 1998; Harzsch 2001; Scholtz and Gerberding 2002; myriapods. The order of formation is different in the Harzsch 2003). However, there are several differences spider C. salei and in the centipede L. forficatus compared from insect neuroblasts. Neuroblasts in malacostracan with the diplopod G. marginata. In both Lithobius crustaceans are generated by so-called ectoteloblasts, (not shown) and Cupiennius, the first invagination sites specialized stem cells that are located in the posterior (black) arise in a coherent anterior-lateral region of region of the germ band anterior to the proctodeum each hemisegment, whereas in Glomeris the first neural precursor groups (black) are distributed over the (with the exception of Amphipoda; Scholtz 1990). hemisegment. Subsequent invagination sites (white, gray, Furthermore, crustacean neuroblasts do not delamin- striped) also arise at different positions in Cupiennius ate into the embryo but remain in the outer surface. and Glomeris. Similar to insects, crustacean neuroblasts divide asym- metrically to give rise to smaller ganglion mother cells 2002; Stollewerk 2002; Dove and Stollewerk 2003; that are pushed into the embryo by directed mitosis Kadner and Stollewerk 2004). Although the neural pre- (Scholtz 1992). The ganglion mother cells also divide cursors arise in 4 sequential waves in regions that are once to produce 2 neurons. In contrast to insects, prefigured by proneural genes, similar to Drosophila, crustacean neuroblasts can generate epidermal cells the precursor groups form invagination sites that after budding off ganglion mother cells (Scholtz persist in the ventral neuroectoderm during the entire and Gerberding 2002). Two achaete-scute homologues course of neurogenesis. Approximately 30 invagination have been identified in the branchiopod crustacean groups per hemisegment detach from the apical surface Triops longicaudatus (Wheeler SR and Skeath 2005). at about the same time, 3 days after the beginning of The expression pattern of these genes is similar to neurogenesis. Interestingly, the invagination groups the distribution of transcripts of the Achaete-Scute show a similar pattern in the myriapods and the Complex genes in Drosophila. spider (Fig. 2): they are arranged in 7 transverse rows with 3–6 invagination sites each (Stollewerk and others Neural precursor formation in 2001; Dove and Stollewerk 2003; Kadner and chelicerates and myriapods Stollewerk 2004). Although the final pattern of inva- In a few classical accounts, neuroblasts have been gination sites is similar in both groups (Fig. 2), the described in 3 chelicerate species, but it is possible order of formation of the individual invagination that the data were partly misinterpreted owing to groups is different. In both the spider C. salei and technical limitations at the time (Yoshikura 1955; the chilopod L. forficatus, the first invagination sites Mathew 1956; Winter 1980). Apart from these studies, arise in the anteriolateral region of each hemisegment, the literature suggests that neurogenesis occurs by whereas in the diplopod G. marginata, the first inva- a generalized inward proliferation of neuroecto- ginating groups are visible in the middle of each dermal cells to produce paired segmental thicken- hemisegment (Stollewerk and others 2001; Dove and ings in chelicerates and myriapods (Anderson 1973). Stollewerk 2003; Kadner and Stollewerk 2004). One can However, recent analyses of neurogenesis in 2 chelicer- be speculate that this difference in timing has an impact ates (both spiders) and 2 myriapods (a diplopod and on the identity of the neural precursors in Glomeris a chilopod) have revealed that in contrast to insects compared with the spider and the chilopod, as genes and crustaceans, groups of neural precursors are spe- that are involved in neural diversity might be expres- cified for neural fate (Fig. 1) in both myriapods and sed during different time windows in the ventral chelicerates (Stollewerk and others 2001; Mittmann neuroectoderm. 198 A. Stollewerk and A. D. Chipman

To obtain more data on diverse myriapod groups, we analyzed neurogenesis in the geophilomorph centipede S. maritima. In contrast to Lithobius and Glomeris, Strigamia undergoes so-called epimorphic development. Myriapods showing this kind of devel- opment generate all segments during embryogenesis, whereas in Lithobius and Glomeris further segments are added during posthatching larval stages. Since Strigamia does not have a considerably longer period of embryogenesis (approximately 30 days [Arthur and Chipman 2005], compared with approximately 15 days in Cupiennius, Lithobius, and Glomeris), the 50 or so segments must arise and differentiate in quick succes- sion. This raises the question of whether neural pre- cursor formation is altered in adaptation to an accelerated development of individual segments. In Strigamia segments arise from a posterior undifferentiated disc (Chipman and others 2004b). As segments are added sequentially, the older segments begin to differentiate, with the first signs of neuro- genesis becoming apparent approximately 5–6 Fig. 3 Pattern of invagination sites in the geophilomorph segments anterior to the undifferentiated area that is centipede S. maritima. Flat preparations of stage 5a the fifth or sixth youngest segment (Chipman and embryos (for staging, see Chipman and Stollewerk 2006) Stollewerk 2006). As the segments arise from the stained with phalloidin-FITC; anterior is toward the top, posterior disc, they are broad in their mediolateral medial to the left in (B–D). (A) In the posterior region of the germ band, the segments are broad in their extent and anteroposteriorly compressed (Fig. 3A mediolateral extent and anteroposteriorly compressed. and B). Shortly after their first appearance, they (B) Strigamia has about 30 invagination sites per separate into clear left and right hemisegments. hemisegment, similar to the spider and the other Morphogenetic movements cause individual segments myriapods. All invagination sites are already present in to broaden along the anteroposterior axis while the the narrow posterior segments. They are arranged in 3 rows and arise at stereotypical positions. mediolateral extent is reduced (Fig. 3C). Later in (C) Morphogenetic movements that reduce the development, after all of the segments have been mediolateral extent of the segments lead to an generated, the left and right halves of the germ band arrangement of the invagination sites that is similar to drift apart in a process known as lateral migration that in the other myriapods and the spider. In each (Kettle and others 2003; Chipman and others 2004b). anterior hemisegment, the neural precursor groups are arranged in 7 rows with 3–6 invagination sites each. Throughout development, there is an anterior to (D) Similar pattern of invagination sites in a hemisegment posterior gradient in the degree of differentiation of the diplopod G. marginata. ant, antennal segment; of individual segments, spanning a wide range of ic, intercalary segment; md, mandibular segment; stages in the neurogenic process. This allows the mx1, maxillary segment 1; mx2, maxillary segment 2; whole course of neurogenesis to be observed in a mxp, maxillipede segment; l1 to l4, trunk segments corresponding to leg pairs 1–4; l15 to l17, trunk small number of specimens. segments corresponding to leg pairs 15–17; 1–7, Similar to the spider and the other myriapods, row of invagination sites 1–7. Strigamia has approximately 30 invagination sites per hemisegment (based on counts of invagination sites at different axial positions in multiple embryos). Proneural genes in the spider and In the narrow posterior segments, they arise at stereotypical positions (see below) and are eventu- the myriapods ally arranged in 3 rows (Fig. 3B). Interestingly, the In Drosophila the proneural genes are essential for morphogenetic movements that reduce the medio- neural fate. The genes of the so-called Achaete-Scute lateral extent of the segments lead to an arrangement Complex achaete, scute, and lethal of scute are expressed of the invagination sites that is similar to the other prior to formation of the neuroblasts in the ventral myriapods and the spider: they are arranged in 7 neuroectoderm (Jime´nez and Campos-Ortega 1979, rows with 3–6 invagination sites each (Fig. 3C; 1990; Cabrera and others 1987; Martin-Bermudo compare with Glomeris Fig. 3D). and others 1991). Mutations in these genes lead to Neurogenesis in myriapods and chelicerates 199 the absence of neuroblasts. Homologues of achaete- similar to the case of Drosophila, although groups of scute have been identified in the spider C. salei neural precursors, rather than single cells, are selected. and the myriapods G. marginata and L. forficatus We have identified 1 Delta and 1 Notch homologue (Stollewerk and others 2001; Dove and Stollewerk in S. maritima (Chipman and Stollewerk 2006). 2003; Kadner and Stollewerk 2004). As in Drosophila, StmNotch shows a heterogeneous expression pattern these homologues can be detected in regions of the similar to the spider and the other myriapods (data neuroectoderm where neural precursors will be not shown). However, the expression pattern of the generated hours later. Expression is up-regulated in Strigamia Delta gene is different from that in the the neural precursor groups, and transcription is other euarthropod groups (Fig. 4). First, Delta expres- down-regulated in the surrounding cells. This expres- sion reveals that invagination sites are added continu- sion pattern is different from that in Drosophila, ously during neurogenesis. In the most posterior where proneural transcripts become restricted to single segment of the Strigamia embryo that exhibits neuro- cells (neuroblasts). Functional studies in the spider genesis (representing the earliest stages of neuro- have shown that the achaete-scute homologues genesis), 2 invagination sites are visible (Fig. 4A). In are essential for neural fate, similar to the case of the next anterior (developmentally older) segment, Drosophila (Stollewerk and others 2001). an additional invagination site has been generated, and in the next anterior segment a further 2 invagina- tion sites have been added. This pattern suggests that Neurogenic genes in the spider and the myriapods It has been shown in Drosophila that the neurogenic genes Notch and Delta are responsible for the restric- tion of proneural gene expression to a single cell of a cluster (Simpson 1990; Martin-Bermudo and others 1995; Heitzler and others 1996; Seugnet and others 1997). However, a dynamic expression of Delta (mRNA and protein) that correlates with the speci- fication of neuroblasts has not been observed in the central nervous system of fly embryos, although it is assumed that within a proneural cluster the cell expressing the highest level of Delta is selected for the neural fate. Similarly, Notch seems to be expressed at homogeneous levels in all ventral neuroecto- dermal cells, indicating that Notch transcripts are not excluded from neural precursors (Heitzler and Simpson 1993). Two Delta homologues, CsDelta1 and CsDelta2, have been identified in the spider C. salei, and 1 Delta homologue each in the myriapods G. marginata Fig. 4 Expression pattern of Strigamia Delta. Flat and L. forficatus (Stollewerk 2002; Dove and Stollewerk preparations (A and B) and sagittal section (C) of 2003; Kadner and Stollewerk 2004). In contrast to embryos (stage 5b) stained for a DIG-labeled StmDelta probe. Anterior is toward the top in (A) and (B), and the case of flies, expression of the spider CsDelta2 to the left in (C). (A) Invagination sites are added and the myriapod Delta genes can be correlated with continuously in Strigamia. In the most posterior segment the formation of neural precursors. Delta transcripts of the Strigamia embryo that exhibits neurogenesis can be detected in all neuroectodermal cells but (representing the earliest stages of neurogenesis), accumulate in the invaginating neural precursors. 2 invagination sites are visible. In the next anterior (developmentally older) segment, an additional Furthermore, the spider and myriapod Notch homo- invagination site has been generated (3) and in the next logues (1 in each species) show a heterogeneous anterior segment 2 further invagination sites have been expression pattern throughout neurogenesis. The added (4 and 5). (B and C) Flat preparation (B) and up-regulation of Notch in distinct regions in the spider sagittal section (C) through the anterior region of the and the myriapods might correlate with the forma- germ band (maxillary segment 2 to trunk segment 4). StmDelta transcripts seem to accumulate at higher levels tion of invagination sites, but this has to be analyzed in single cells within the invagination groups (arrows). in more detail. Functional studies in the spider revealed The single cells are surrounded by cells expressing that Notch and Delta mediate lateral inhibition, lower levels of Delta. 200 A. Stollewerk and A. D. Chipman invagination sites are added continuously during Specification of neuroblast identity neurogenesis, rather than in several distinct waves as in arthropods in the spider and the other myriapods. In Drosophila, segment polarity genes and dorsoventral Furthermore, StmDelta transcripts seem to patterning genes are expressed during neurogenesis accumulate at higher levels in single cells within the in the ventral neuroectoderm (see review in Skeath invagination groups (arrows in Fig. 4B and C). The 1999). These genes subdivide the neuroectoderm single cells are surrounded by cells expressing lower into a gridlike pattern, so that each proneural cluster levels of Delta (Fig. 4). To understand this expression shows a different gene expression profile. The neuro- pattern, we further analyzed the morphology of blasts maintain the specific expression pattern of the the invaginating cell groups by staining Strigamia proneural cluster from which they delaminate and give embryos with phalloidin-FITC, a dye that stains the rise to an invariant lineage of distinct neural progenies. actin cytoskeleton. An accumulation of actin around Thirty neuroblasts delaminate from the neuroectoderm single cells in the ventral neuroectoderm was observed of each hemisegment. They are arranged in 7 rows by confocal microscopy (Fig. 5A). A detailed analysis with each row expressing a different subset of segment of the morphology of the invagination groups revealed polarity genes. It has been shown in Drosophila that the that this staining is due to the cell processes of function of the segment polarity genes is specifically the cells of individual invagination groups that are required in neuroblasts. Mutations in these genes lead attached to a single cell of the group (Fig. 5B either to the absence of specific neuroblasts or to and C). These data suggest that StmDelta transcripts changes in the identity of neuroblasts (Skeath 1999). are not present at higher levels in single cells but The specification of neuroblast identity has not accumulate around single cells within invagination been analyzed in any detail in arthropods other than groups as a result of this distinct morphological Drosophila. Although segment polarity genes have arrangement. been identified in other insects, in crustaceans, in myriapods, and in a spider, their function during neurogenesis has not been studied except for engrailed (Patel and others 1992; Brown and others 1994, 1997; Dawes and others 1994; Patel 1994; Damen and others 2000; Telford 2000; Davis and others 2001; Damen 2002; Dearden and others 2002; Hughes and Kaufman 2002; Mouchel-Vielh and others 2002; Copf and others 2003; Kettle and others 2003; Chipman and others 2004a, 2004b; Eckert and others 2004; Janssen and others 2004; Peel 2004). However, Patel and coworkers (1989) investigated the expression pattern of the segment polarity gene engrailed in several insects and crustaceans and showed that engrailed expression in neuroblasts is conserved. In all species analyzed, engrailed is expressed in neuroblast rows 6 and 7, and 1 neuroblast of row 1. Fig. 5 F-actin accumulates around single cells in the We have analyzed engrailed expression in the ventral neuroectoderm of Strigamia. Flat preparations of embryos (stage 5a) stained with phalloidin-FITC spider Cupiennius and the geophilomorph centipede (A and B). Anterior is toward the top. (A) An Strigamia. In both the spider and the centipede, accumulation of F-actin can be seen at different engrailed is expressed in segmental stripes in the pos- apicobasal levels in the ventral neuroectoderm of terior region of the germ band (Fig. 6A and D) (Damen Strigamia (arrow). (B) A higher magnification of 2002; Kettle and others 2003; Chipman and others invagination groups reveals that this staining is due to the cell processes of the cells of individual invagination 2004b). In more anterior, developmentally advanced groups that are attached to a single cell of the group. segments that are undergoing neurogenesis, engrailed The white square borders an invagination group. (C) The expression covers a broader region at the posterior schematic drawing shows the distinct morphological border of the segments. In addition, in the central area arrangement of an invagination group in Strigamia. ant, of the ventral neuroectoderm the engrailed expression antennal segment; ic, intercalary segment; md, mandibular segment; mx1, maxillary segment 1; mx2, domain extends into the anterior region of the next maxillary segment 2; mxp, maxillipede segment; l1, trunk posterior segments, whereas the engrailed stripe lateral segment corresponding to leg pair 1; ml, ventral midline. to the limb buds is still restricted to a few cell rows at Neurogenesis in myriapods and chelicerates 201

We have compared the pattern of engrailed expres- sion with the pattern of invagination sites in the spider. Single-color double-staining with engrailed and anti- horseradish peroxidase, which is exclusively expressed in the cell processes of the invaginating neural pre- cursor groups at this time, revealed that engrailed is expressed in the invagination groups of rows 6 and 7 and row 1 in the spider (Fig. 7A–D). Interestingly, there are 7 invagination sites in rows 6 and 7 and this number is identical to the number of neuroblasts that are engrailed positive in rows 6 and 7 in insects and crustaceans (Patel and others 1989; Fig. 6 Comparison of engrailed expression in the spider Duman-Scheel and Patel 1999). Cupiennius and the geophilomorph centipede Strigamia. We have also compared engrailed expression with Flat preparations (A and B) and sagittal sections the position of invagination sites in the geophilomorph (C and D) of embryos stained for DIG-labeled engrailed centipede. In Strigamia the segments have a peak and probes. Anterior is toward the top in (A) and (B), and trough structure that is most obvious in sagittal sec- to the left in (C) and (D). (A) In Cupiennius, engrailed is expressed in small dorsoventral stripes in newly formed tions (Figs. 6D and 8D). In the posterior region of segments in the posterior region of the germ band the germ band the engrailed stripe divides the trough (arrowhead). In the anterior region of the germ band into 2 halves (Fig. 6D). Based on this expression (190 h after egg laying), engrailed expression covers a pattern and the fact that the posterior border of the broader region at the posterior border of the segments engrailed domain coincides with the posterior border and extends into the anterior region of the next posterior segments (arrows). The anterior expression of segments in all arthropods analyzed, we conclude domain is restricted to the mediocentral region of each that the posterior half of the trough belongs to the next hemisegment. The black lines indicate the segmental posterior segment. In more anterior segments that borders. (B) In Strigamia (stage 5b) engrailed expression exhibit neurogenesis, engrailed is expressed throughout extends into the groove (arrow). In contrast to the the trough, indicating that it is not only expressed at spider, engrailed is also expressed in the ventral midline. the posterior border of the segments but also in neural The black lines indicate the segmental borders. (C) The sagittal section through a ventral neuromere of precursors in the anterior region of each segment. the spider shows that engrailed is expressed not only in Analysis of phalloidin-FITC-stained embryos revealed the outer neuroectodermal cell layer but also in the that the neural precursor groups that belong to basally located invaginating neural precursors. the anterior row of invagination sites extend into the The arrowhead indicates the segmental border. (D) In groove (Fig. 8B and C). Similarly, the groups that Strigamia the segments have a peak and trough structure. As in Cupiennius, engrailed is expressed in a small stripe belong to the posterior rows extend into the groove in the posterior region of the germ band that coincides from the other site (Fig. 8B). Double-stainings with with the posterior border of the segments Delta and engrailed suggest that engrailed is expressed (arrowheads). In more anterior segments that are in the first anterior row of invagination groups and undergoing neurogenesis, engrailed expression also both in rows 6 and 7 (Fig. 8E). covers the anterior region of the segments (arrows).

Conclusions the posterior border of the segments (Fig. 6A and B, We have presented comparative morphological and arrows; Fig. 7A; Damen 2002). In spiders and molecular data on neurogenesis in the euarthropod myriapods, all cells of the ventral neuroectoderm groups. Although there are differences in the formation give rise to neural cells. The epidermis arises lateral of neural precursors, most arthropod species analyzed to the neuromeres only after invagination of the neural generate approximately 30 single neural precursors precursors (Stollewerk 2002; Dove and Stollewerk (insects/crustaceans) or precursor groups (chelicerates/ 2003; Stollewerk 2004). Therefore, it can be concluded myriapods) per hemisegment, which are arranged in that during neurogenesis, engrailed is specifically regular rows. Homologues of achaete-scute are neces- expressed in neural precursors. sary for the formation of neural precursors, and the In Strigamia, engrailed is expressed in the ventral neurogenic genes Notch and Delta restrict the pro- midline (Fig. 6B). Further analysis will show whether portion of cells that adopt a neural fate at a certain this expression corresponds to an accumulation of time. In insects, chelicerates, and 2 of the 3 myriapods transcripts in neural cells. analyzed, neural precursors are produced in several 202 A. Stollewerk and A. D. Chipman

Fig. 7 Comparison of the pattern of engrailed expression with the pattern of invagination sites in the spider C. salei. Flat preparations of embryos stained for DIG-labeled engrailed probes (A–C), anti-horseradish peroxidase (B and C), and with phalloidin-FITC (D). The black and white lines indicate the segmental borders. (A) Expression domain of engrailed in segments that exhibit neurogenesis (190 h after egg laying). (B and C) Single-color double-staining with engrailed and anti-horseradish peroxidase. At this stage (190 h of development) the horseradish peroxidase antigen is exclusively expressed in the cell processes of the invaginating neural precursors, as seen on the apical view of the ventral neuroectoderm (B). On the basal view of the same area, engrailed expression is visible in the neural precursors of rows 6, 7 and 1 (C). (D) Pattern of invagination sites in an embryo of the same stage stained with phalloidin-FITC. Note that there are 7 invagination sites in rows 6 and 7. The segmental border is clearly visible because of the distinct shape of the border cells: they are mediolaterally elongated. 1, 6, 7: row of invagination sites 1, 6, 7.

Fig. 8 Comparison of engrailed expression with the position of invagination sites in the geophilomorph centipede Strigamia (stage 5b). Flat preparations of embryos stained with phalloidin-FITC (A–C), flat preparation of an embryo double-stained for a DIG-labeled Stmengrailed probe (blue) and a Fluorescein-labeled StmDelta probe (red) (Chipman and others 2004a) (E), and sagittal section of an embryo stained for a DIG-labeled Stmengrailed probe (D). (A–B) Apical and basal optical section, respectively, of the same anterior region of the ventral neuroectoderm (2 hemisegments). The brackets indicate the extension of the groove. The neural precursors extend into the groove from anterior and posterior (arrows in [B]). (C) Three hemisegments of the posterior region of the germ band. The arrow indicates neural precursor groups that belong to the first row of invagination sites and extend into the groove. (D) engrailed expression in neural precursors (arrows). (E) engrailed is expressed in rows 6 and 7 and throughout the groove, indicating that transcripts also accumulate in neural precursors of row 1 (arrow). The white lines indicate the segmental borders. sequential waves. Neural precursor formation has centipede Strigamia might be an adaptation to the been analyzed in only a limited number of crustacean distinct embryonic development of this species. species (Dohle 1972, 1976; Scholtz 1984, 1990, 1992; Approximately 50 segments are generated during Harzsch and Dawirs 1994, 1996; Harzsch and others embryogenesis and differentiate in quick succession. 1998; Gerberding and Scholtz 1999; Harzsch 2001, Gene expression studies and morphological analyses 2003). Based on the small amount of data available, revealed that each segment exhibits a different differ- it can be assumed that neuroblasts in crustaceans entiation state along the anterior-posterior axis during are continuously added during neurogenesis, rather neurogenesis. Therefore, it can be concluded that each than being generated in several waves. The continuous segment initiates neurogenesis on its own, rather than addition of neural precursor groups in the being synchronized with several segments, as seen in Neurogenesis in myriapods and chelicerates 203 the spider and in the other myriapods. As has been unites myriapods and insects, to the exclusion of shown previously, neurogenesis occurs simultaneously crustaceans. These data are intriguing and warrant in the prosoma of the spider and in the head and first further research into neurogenesis in putative trunk segments of G. marginata and L. forficatus, and at arthropod sister groups. Additional data will allow a least 2 to 3 segments show the same stage of neuro- polarization of the characteristic state changes and genesis in the posterior region of the germ band help resolve the question of the relationships between (Stollewerk and others 2001; Dove and Stollewerk the major arthropod groups. 2003; Kadner and Stollewerk 2004). Despite the differ- ences in neural precursor formation in the euarthropod Acknowledgments group, neural precursor identity seems to be achieved in a similar way. In all species analyzed, engrailed is We thank the organizers of the symposium for the expressed in neural precursor rows 6, 7, and 1. opportunity to present our work. We are grateful to In addition to these conserved features, we found Michael Akam and Pat Simpson for providing lab 2 characteristics that distinguish insects/crustaceans space and for helpful discussions. Thanks to Pat from myriapods/chelicerates. First, in insects and Simpson for critical reading of the manuscript. The crustaceans the neuroectoderm gives rise to epidermal Deutsche Forschungsgemeinschaft (A.S.) and and neural cells. In contrast, there is no decision Federation of European Biochemical Societies between epidermal and neural fate in the central region (A.D.C.) supported this research. of the ventral neuroectoderm of chelicerates and myriapods. The epidermis arises lateral to the neuro- References meres and overgrows the ventral nerve cord after Anderson D. 1973. Embryology and phylogeny in annelids and formation of neural precursors. However, this charac- arthropods. Oxford: Pergamon Press. teristic seems to be ancestral (plesiomorphic), as it has been shown in onychophorans (a group that is Arthur W, Chipman AD. 2005. The centipede Strigamia maritima: what it can tell us about the development and assumed to be basal to the arthropods) that the evolution of segmentation. BioEssays 27:653–60. whole central regions of the hemisegments sink into Ballard JWO, Olsen GJ, Faith DP, Odgers WA, Rowell DM, the embryo and thus give only rise to neural cells Atkinson PW. 1992. Evidence from 12S ribosomal RNA (Eriksson and others 2003). The second distinguishing sequences that onychophorans are modified arthropods. characteristic is the presence of neuroblasts in insects/ Science 258:1345–8. crustaceans as opposed to neural precursor groups in Bate M. 1976. Embryogenesis of an insect nervous system: myriapods/chelicerates. Invaginating cell groups have I. A map of thoracic and abdominal neuroblasts in Locusta not been found in onychophorans or in tardigrades migratoria. J Embryol Exp Morphol 35:107–23. (another potential outgroup to the euarthropods Bergstro¨m J. 1992. The oldest Arthropoda and the origin of the (Eriksson and others 2003; Hejnol and Schnabel Crustacea. Acta Zool 73:287–91. 2005). However, only 2 species have been analyzed, Bitsch C, Bitsch J. 2004. Phylogenetic relationships of basal which might be derived and thus do not represent hexapods among the mandibulate arthropods: a cladistic the ancestral state. The fact that approximately 30 analysis based on comparative morphological characters. neural precursors/precursor groups per hemisegment Zool Scr 33:511–50. are arranged in a strikingly similar pattern in most Briggs DEG, Fortey RA. 1989. The early radiation and relation- euarthropod species analyzed suggests that a similar ships of the major arthropod groups. Science 246:241–3. pattern was present in the last common ancestor of Broadus J, Doe CQ. 1995. Evolution of neuroblast identity: the arthropods. The formation of neural precursor seven-up and prospero expression reveal homologous and groups could be a sympleisiomorphy of myriapods divergent neuroblast fates in Drosophila and Schistocerca. and chelicerates that has been lost in the more derived Development 121:3989–96. Pancrustacea. However, it is also possible that this Brown SJ, Parrish JK, Beeman RW, Denell RE. 1997. Molecular characteristic represents a synapomorphy of myriapods characterization and embryonic expression of the even- and chelicerates, providing the first morphological skipped ortholog of Tribolium castaneum. Mech Dev support for the Myriochelata group (Friedrich and 61:165–73. Tautz 1995; Hwang and others 2001; Kusche and Brown SJ, Parrish JK, Denell RE, Beeman RW. 1994. Genetic Burmester 2001; Nardi and others 2003; Mallatt control of early embryogenesis in the red flour beetle, and others 2004; Pisani and others 2004). At present, Tribolium castaneum. Am Zool 34:343–52. we cannot distinguish between these 2 scenarios, Cabrera CV, Martinez-Arias A, Bate M. 1987. The expression of but the data presented in this review are clearly three members of the achaete-scute gene complex correlates inconsistent with the Atelocerata hypothesis, which with neuroblast segregation in Drosophila. Cell 50:425–33. 204 A. Stollewerk and A. D. Chipman

Chipman AD, Arthur W, Akam M. 2004a. A double segment Eckert C, Aranda M, Wolff C, Tautz D. 2004. Separable stripe periodicity underlies segment generation in centipede enhancer elements for the pair-rule gene hairy in the beetle development. Curr Biol 14:1250–5. Tribolium. EMBO Rep 5:638–42. Chipman AD, Arthur W, Akam M. 2004b. Early development Emerson MJ, Schram FR. 1998. Theories, patterns, and reality: and segment formation in the centipede, Strigamia maritima Game plan for arthropod phylogeny. In: Fortey RA, (Geophilomorpha). Evol Dev 6:78–89. Thomas RH, editors. Arthropod relationships London: Chipman AD, Stollewerk A. 2006. Specification of neural pre- Chapman & Hall. p 67–86. cursor identity in the geophilomorph centipede Strigamia Eriksson BJ, Tait NN, Budd GE. 2003. Head development in the maritima. Dev Biol 290:337–350. Onychophoran Euperipatoides kanangrensis with particular Copf T, Rabet N, Celniker SE, Averof M. 2003. Posterior pat- reference to the central nervous system. J Morphol 255:1–23. terning genes and the identification of a unique body region Field KG, Olsen GJ, Lane DJ, Giovannoni SJ, Ghiselin MT, in the brine shrimp Artemia franciscana. Development Raff EC, Pace NR, Raff RA. 1988. Molecular phylogeny of 130:5915–27. the animal kingdom. Science 239:748–53. Damen W. 2002. Parasegmental organization of the spider Friedrich M, Tautz D. 1995. Ribosomal DNA phylogeny of embryo implies that the parasegment is an evolutionary the major extant arthropod classes and the evolution of conserved entity in arthropod embryogenesis. Development myriapods. Nature 376:165–7. 129:1239–50. Galant R, Skeath JB, Paddock S, Lewis DL, Carroll SB. 1998. Damen W, Weller M, Tautz D. 2000. Expression patterns of Expression pattern of a butterfly achaete-scute homolog hairy, even-skipped, and runt in the spider Cupiennius salei reveals the homology of butterfly wing scales and insect imply that these genes were segmentation genes in a basal sensory bristles. Curr Biol 8:807–13. arthropod. Proc Natl Acad Sci USA 97:4515–19. Gerberding M. 1997. Germ band formation and early neuro- Davis GK, Jaramillo CA, Patel NH. 2001. Pax group III genes genesis of Leptodora kindti (Cladocera): first evidence for and the evolution of insect pair-rule patterning. Development neuroblasts in the entomostracan crustaceans. Invertebr 128:3445–58. Reprod Dev 32:63–73. Dawes R, Dawson I, Falciani F, Tear G, Akam M. 1994. Dax, Gerberding M, Scholtz G. 1999. Cell lineage of the midline cells a locust Hox gene related to fushi-tarazu but showing no in the amphipod crustacean Orchestia cavimana (Crustacea, pair-rule expression. Development 120:1561–72. Malacostraca) during formation and separation of the germ Dearden PK, Donly C, Grbic M. 2002. Expression of pair-rule band. Dev Genes Evol 209:91–102. gene homologues in a chelicerate: early expression patterning Giribet G, Ribera C. 1998. The position of arthropods in the of the two-spotted spider mite Tetranychus urticae. animal kingdom: a search of a reliable outgroup for internal Development 129:5461–72. arthropod phylogeny. Mol Phylogenet Evol 9:481–8. Dohle W. 1972. U¨ ber die Bildung und Differenzierung des Goodman CS, Doe CQ. 1993. Embryonic development of the postnauplialen Keimstreifs von Leptochelia spec. Crustacea, Drosophila central nervous system. In: Bate M, Martinez- Tanaidacea). Zool Jb Anat 89:503–66. Arias A, editors. The development of Drosophila melanogaster. Dohle W. 1976. Die Bildung und Differenzierung des New York: Cold Spring Harbor Laboratory Press. p 1131–206. postnauplialen Keimstreifs von Diastylis rathkei (Crustacea, Harzsch S. 2001. Neurogenesis in the crustacean ventral nerve Cumacea) II. Die Differenzierung und Musterbildung des cord: homology of neuron stem cells in Malacostraca and Ektoderms. Zoomorphologie 84:235–77. Branchiopoda? Evol Dev 3:154–69. Dohle W. 1998. Myriapod–insect relationships as opposed to Harzsch S. 2003. Ontogeny of the ventral nerve cord in an insect–crustacean sister group relationship. In: Fortey RA, malacostracan crustaceans: a common plan for neuronal Thomas RH, editors. Arthropod relationships London: development in Crustacea and Hexapoda? Arthropod Chapman & Hall. p 305–15. Struct Dev 32:17–38. Dohle W. 2001. Are the insects terrestial crustaceans? A dis- Harzsch S, Dawirs RR. 1994. Neurogenesis in larval stages of cussion of some new facts and arguments and the proposal the spider crab Hyas araneus (Decapoda, Brachyura): prolif- of the proper name “Tetraconata” for the monophyletic unit eration of neuroblasts in the ventral nerve cord. Roux’s Arch Crustacea and Hexapoda. Ann Soc Entomol Fr 37:85–103. Dev Biol 204:93–100. Dohle W, Scholtz G. 1988. Clonal analysis of the crustacean Harzsch S, Dawirs RR. 1996. Neurogenesis in the developing segment: the discordance between genealogical and segmental crab brain: postembryonic generation of neurons persists borders. Development 104 (Suppl):147–60. beyond metamorphosis. J Neurobiol 29:384–98. Dove H, Stollewerk A. 2003. Comparative analysis of neuro- Harzsch S, Miller J, Benton J, Dawirs RR, Beltz B. 1998. genesis in the myriapod Glomeris marginata (Diplopoda) Neurogenesis in the thoracic neuromeres of two crustaceans suggests more similarities to chelicerates than to insects. with different types of metamorphic development. J Exp Biol Development 130:2161–71. 201:2465–79. Duman-Scheel M, Patel NH. 1999. Analysis of molecular marker Heitzler P, Bourouis M, Ruel L, Carteret C, Simpson P. 1996. expression reveals neuronal homology in distantly related Genes of the Enhancer of split and achaete-scute complexes arthropods. Development 126:2327–34. are required for a regulatory loop between Notch and Delta Neurogenesis in myriapods and chelicerates 205

during lateral signalling in Drosophila. Development Mathew AP. 1956. Embryology of Heterometrus scaber (Thorell), 122:161–71. Arachnida, Scorpionidae. Zool Mem Univ Travancore 1:1–96. Heitzler P, Simpson P. 1993. Altered epidermal growth factor- Mittmann B. 2002. Early neurogenesis in the horseshoe crab like sequences provide evidence for a role of Notch as a Limulus polyphemus and its implication for arthropod receptor in cell fate decisions. Development 117:1113–23. relationships. Biol Bull 203:221–2. Hejnol A, Schnabel R. 2005. The eutardigrade Thulinia Mouchel-Vielh E, Blin M, Rigolot C, Deutsch JS. 2002. stephaniae has an indeterminate development and the poten- Expression of a homologue of the fushi tarazu (ftz) gene in tial to regulate early blastomere ablations. Development a cirripede crustacean. Evol Dev 4:76–85. 1332:1349–61. Nakao K, Campos-Ortega JA. 1996. Persistent expression of Hughes CL, Kaufman TC. 2002. Exploring the myriapod body genes of the Enhancer of split complex suppresses neural plan: expression patterns of the ten Hox genes in a centipede. development in Drosophila. Neuron 16:275–86. Development 129:1225–38. Nardi F, Spinsanti G, Boore JL, Carapelli A, Dallai R, Frati F. Hwang UW, Friedrich M, Tautz D, Park CJ, Kim W. 2001. 2003. Hexapod origins: monophyletic or paraphyletic? Mitochondrial protein phylogeny joins myriapods with Science 299:1887–9. chelicerates. Nature 413:154–7. Nilsson DE, Osorio D. 1998. Homology and parallelism in Janssen R, Prpic N-M, Damen W. 2004. Gene expression arthropod sensory processing. In: Fortey RA, Thomas RH, suggests decoupled dorsal and ventral segmentation in the editors. Arthropod relationships London: Chapman & Hall. millipede Glomeris marginata (Myriapoda: Diplopoda). p 333–47.. Dev Biol 268:89–104. Osorio D, Averof M, Bacon JP. 1995. Arthropod evolution: great Jime´nez F, Campos-Ortega JA. 1979. A region of the Drosophila brains, beautiful bodies. Trends Ecol Evol 10:449–54. genome necessary for CNS development. Nature 282:310–12. Patel NH. 1994. Developmental evolution: insights from studies Jime´nez F, Campos-Ortega JA. 1990. Defective neuroblast of insect segmentation. Science 266:581–90. commitment in mutants of the achaete-scute complex and Patel NH, Ball EE, Goodman CS. 1992. Changing role of even- adjacent genes of Drosophila melanogaster. Neuron 5:81–9. skipped during the evolution of insect pattern formation. Kadner D, Stollewerk A. 2004. Neurogenesis in the chilopod Nature 357:339–42. Lithobius forficatus suggests more similarities to chelicerates Patel NH, Kornberg TB, Goodman CS. 1989. Expression of than to insects. Dev Genes Evol 214(8):367–79. engrailed during segmentation in grasshopper and crayfish. Kettle C, Johnstone J, Jowett T, Arthur H, Arthur W. 2003. Development 107:201–12. The pattern of segment formation, as revealed by engrailed Peel A. 2004. The evolution of arthropod segmentation expression, in a centipede with a variable number segments. mechanisms. BioEssays 26:1108–16. Evol Dev 5:198–207. Pisani D, Poling LL, Lyons-Weiler M, Hedges SB. 2004. The Kraus O, Kraus M. 1994. Phylogenetic system of the Tracheata colonization of land by animals: molecular phylogeny and (Mandibulata): on “Myriapoda”–Insecta interrelationships, divergence times among arthropods. BMC Biology 2:1. phylogenetic age and primary ecological niches. Verh Pistillo D, Skaer N, Simpson P. 2002. scute expression in Naturwiss Ver Hamburg 34:5–31. Calliphora vicina reveals an ancestral pattern of longitudinal Kraus O, Kraus M. 1996. On myriapod/insect interrelationships. stripes on the thorax of higher Diptera. Development Me´m Mus Natl Hin Nat 169:283–90. 129:563–72. Kusche K, Burmester T. 2001. Diplopod hemocyanin sequence Regier JC, Shultz JW, Kambic RF. 2005. Pancrustacean phylo- and the phylogenetic position of the Myriapoda. Mol Biol geny: hexapods are terrestrial crustaceans and maxilopods Evol 18:1566–73. are not monophyletic. Proc Biol Sci 272:395–401. Ligoxygakis P, Yu SY, Delidakis C, Baker NE. 1998. A subset of Romani S, Campuzano S, Modolell J. 1987. The achaete-scute Notch functions during Drosophila eye development require complex is expressed in neurogenic regions of Drosophila Su(H) and E(spl) gene complex. Development 125:2893–900. embryos. EMBO J. 6:2085–92. Mallatt JM, Garey JR, Shultz JW. 2004. Ecdysozoan phylogeny Scholtz G. 1984. Untersuchungen zur Bildung und and Bayesian inference: first use of nearly complete 28S and Differenzierung des postnaupliaren Keimstreifs von 18S rRNA gene sequences to classify the arthropopds and Neomysis integer Leach (Crustacea, Malacostraca, their kin. Mol Phylogenet Evol 31:178–91. Peracarida). Zool Jb Anat 112:295–349. Martin-Bermudo MD, Carmena A, Jimenez F. 1995. Neurogenic Scholtz G. 1990. The formation, differentiation and segmenta- genes control gene expression at the transcriptional level tion of the post-naupliar germ band of the amphipod in early neurogenesis and in mesectoderm specification. Gammarus pulex L. (Crustacea, Malacostraca, Peracarida). Development 121:219–24. Proc R Soc Lond B Biol Sci 239:163–211. Martin-Bermudo MD, Martinez C, Rodriguez A, Jime´nez F. Scholtz G. 1992. Cell lineage studies in the crayfish Cherax 1991. Distribution and function of the lethal of scute destructor (Crustacea, Decapoda): germ band formation, gene product during early neurogenesis in Drosophila. segmentation and early neurogenesis. Roux’s Arch Dev Biol Development 113:445–54. 202:36–48. 206 A. Stollewerk and A. D. Chipman

Scholtz G, Gerberding M. 2002. Cell lineage of crustacean Turberville JM, Pfeiffer DM, Field KG, Raff RA. 1991. The neuroblasts. In: Wiese K, editor. The Crustacean nervous phylogenetic status of arthropods as infered from 18S system. Berlin, Heidelberg, New York: Springer. p 406–416. rRNA sequences. Mol Biol Evol 8:669–86. Schram FR, Emerson MJ. 1991. Arthropod pattern theory: a new Wheeler SR, Carrico ML, Wilson BA, Brown SJ, Skeath JB. 2003. approach to arthropod phylogeny. Mem Qld Mus 31:1–18. The expression and function of the achaete-scute genes in Seugnet L, Simpson P, Haenlin M. 1997. Transcriptional Tribolium castaneum reveals conservation and variation regulation of Notch and Delta: requirement for neuroblast in neural pattern formation and cell fate specification. segregation in Drosophila. Development 124:2015–25. Development 130:4373–81. Shultz JW, Regier JC. 2000. Phylogenetic analysis of arthropods Wheeler SR, Skeath JB. 2005. The identification and expression using two nuclear protein-encoding genes supports a of achaete-scute genes in the branchiopod crustacean Triops crustacean þ hexapod clade. Proc Biol Sci 267:1011–19. longicaudatus. Gene Expr Patterns 5:695–700. Simpson P. 1990. Lateral inhibition and the development of the adult sensory bristles of the peripheral nervous system of Wheeler WC. 1998. Arthropod fossils and phylogeny. In: Drosophila. Development 109:509–19. Edgecombe GD, editor. Arthropod fossils and phylogeny. Skaer N, Pistillo D, Simpson P. 2002. Transcriptional hetero- New York: Columbia University Press. p 9–32. chrony of scute and changes in bristle pattern between two Wheeler WC, Cartwright P, Hayashi CY. 1993. Arthropod closely related species of blowfly. Dev Biol 252:31–45. phylogeny: a combined approach. Cladistics 9:1–39. Skeath JB. 1999. At the nexus between pattern formation and Whitington PM. 1996. Evolution and neural development in the cell-type specification: the generation of individual neuroblast arthropods. Semin Cell Dev Biol 7:605–14. fates in the Drosophila embryonic central nervous system. Whitington PM, Leach D, Sandeman R. 1993. Evolutionary BioEssays 21:922–31. change in neural development within the arthropods: axono- Skeath JB, Panganiban G, Selegue J, Carroll SB. 1992. Gene genesis in the embryo of two crustaceans. Development regulation in two dimensions: the proneural achaete and 118:449–61. scute genes are controlled by combinations of axis patterning genes through a common intergenic control region. Genes Wills MA, Briggs DEG, Fortey RA, Wilkinson M, Sneath PHA. Dev 6:2606–19. 1998. Arthropod fossils and phylogeny. In: Edgecombe GD, Snodgrass RE. 1938. Evolution of the Annelida, Onychophora editor. An arthropod phylogeny based on fossil and recent and Arthropoda. Smithson Misc Collect 97:1–159. taxa. New York: Columbia University Press. p 33–105. Snodgrass RE. 1950. Comparative studies on the jaws of Winter G. 1980. Beitra¨ge zur Morphologie and Embryologie mandibulate arthropods. Smithson Misc Collect 116:1–85. des vorderen Ko¨rperabschnitts (Cephalosoma) der Snodgrass RE. 1951. Comparative studies on the head of Pantopoda Gerstaecker, 1863. I. Zur Entstehung des mandibulate arthropods. Ithaca, NY: Comstock. Zentralnervensystems. Z Zool Syst Evol Forsch 18:27–61. Stollewerk A. 2002. Recruitment of cell groups through Delta/ Wu¨lbeck C, Simpson P. 2000. Expression of achaete-scute Notch signalling during spider neurogenesis. Development homologues in discrete proneural clusters on the developing 129:5339–48. notum of the medfly Ceratitis capitata, suggests a common Stollewerk A. 2004. Secondary neurons are arrested in an origin for the stereotyped bristle patterns of higher Diptera. immature state by formation of epithelial vesicles during Development 127:1411–20. neurogenesis of the spider Cupiennius salei. Front Zool 1:3. Wu¨lbeck C, Simpson P. 2002. The expression of pannier and Stollewerk A, Tautz D, Weller M. 2003. Neurogenesis in the achaete-scute homologues in a mosquito suggests an ancient spider: new insights from comparative analysis of morpho- role of pannier as a selector gene in the regulation of the dorsal logical processes and gene expression patterns. Arthrop Struct body pattern. Development 129:3861–71. Dev 32:5–16. Stollewerk A, Weller M, Tautz D. 2001. Neurogenesis in the Yoshikura M. 1955. Embryological studies on the liphistiid spi- spider Cupiennius salei. Development 128:2673–88. der, Heptathela kimurai, part II. Kumamoto J Sci B 2:1–86. Telford MJ. 2000. Evidence for the derivation of the Drosophila Zrzavy´ J, Sˇtys P. 1997. The basic body plan of arthropods: fushi tarazu gene from a Hox gene orthologous to lopho- insights from evolutionary morphology and developmental trochozoan Lox5. Curr Biol 10:349–52. biology. J Evol Biol 10:353–67.