Research article | Open | Published:
Expression of collier in the premandibular segment of myriapods: support for the traditional Atelocerata concept or a case of convergence?
BMC Evolutionary Biologyvolume 11, Article number: 50 (2011)
A recent study on expression and function of the ortholog of the Drosophila collier (col) gene in various arthropods including insects, crustaceans and chelicerates suggested a de novo function of col in the development of the appendage-less intercalary segment of insects. However, this assumption was made on the background of the now widely-accepted Pancrustacea hypothesis that hexapods represent an in-group of the crustaceans. It was therefore assumed that the expression of col in myriapods would reflect the ancestral state like in crustaceans and chelicerates, i.e. absence from the premandibular/intercalary segment and hence no function in its formation.
We find that col in myriapods is expressed at early developmental stages in the same anterior domain in the head, the parasegment 0, as in insects. Comparable early expression of col is not present in the anterior head of an onychophoran that serves as an out-group species closely related to the arthropods.
Our findings suggest either that i) the function of col in head development has been conserved between insects and myriapods, and that these two classes of arthropods may be closely related supporting the traditional Atelocerata (or Tracheata) hypothesis; or ii) alternatively col function could have been lost in early head development in crustaceans, or may indeed have evolved convergently in insects and myriapods.
The recent arthropods comprise four classes: the insects, the crustaceans, the myriapods and the chelicerates. In some phylogenies pycnogonids are suggested to comprise a fifth class of arthropods, in some other phylogenies they are closely grouped with the chelicerates . The sister-group of the arthropods is represented by the onychophorans that lack the most characteristic feature of the arthropods - segmentation of the appendages (arthropodization) (e.g. ). Body segmentation, tagmosis, and arthropodization are thought to be among the main causes why the arthropods became the dominating metazoan group in species number, number of individuals and morphological diversity, on our planet. A segmented body, often in combination with tagmosis, probably allowed the arthropods to adapt to new environmental situations quickly by modification of single segments and their often-specialized appendages without disturbing their general bodyplan .
Despite the biological importance of the arthropods and the enormous number of published phylogenies, the relationships of the arthropod classes remain controversial. In particular, the position of the myriapods has changed often and dramatically during the last century (reviewed in e.g. [4, 5]). The myriapods were traditionally thought to represent the sister-group of the hexapods (Atelocerata or Tracheata theory) (e.g. [6, 7]). This hypothesis is exclusively based on morphological data such as the presence of tracheae and Malpighian tubules or the appendage-less tritocerebral segment (reviewed in e.g. [8, 9]). Myriapods were even placed with onychophorans and insects (Uniramia theory), suggesting that arthropods are polyphyletic [10, 11]. This latter theory appears however to have lost its credibility (e.g. ). Another current theory places myriapods and chelicerates as closely related sister-groups (Myriochelata or Parodoxopoda theory). This theory finds support in morphological as well as in molecular studies (e.g. [2, 13–16]).
Nevertheless, most molecular and a number of morphological phylogenetic analyses argue strongly in favor of a close relationship of crustaceans and insects (either Tetraconata or Pancrustacea theory) (e.g. [17–23]). Note that it is important to distinguish a true sister-group relationship of insects and crustaceans (= Tetraconata) and an in-group relationship of insects and crustaceans (= Pancrustacea). Morphological features supporting the Atelocerata are now often considered to have convergently evolved. Tracheae, Malpighian tubules and the loss of the tritocerebral appendage for example are thought to represent independent adaptations in insects and myriapods necessary for a life on land [4, 5, 24–26].
A number of genes involved in the formation of the head segments have been identified in Drosophila (e.g. [27–30]) and subsequent studies suggested that these factors may play widely conserved roles in insects (e.g. [31–33]). One of the key players in anterior head development is the COE-family HLH transcription factor collier (aka knot) . Flies deficient for collier (col) function lack ectodermal structures of the intercalary segment, and the expression of segment defining genes like engrailed and wingless is disturbed .
Very recently a study on function and expression of the orthologs of col in insects, a crustacean and a chelicerate suggested that early col function in the development of the intercalary segment is only present in insects . In their paper Schaeper and colleagues conclude that the early function of col in head segmentation is most probably an insect novelty. In accord with the Pancrustacea hypothesis, the development of the limbless tritocerebral segment in myriapods is most likely convergent and thus likely based on a different genetic mechanism .
Our data on col expression in two myriapod species, the millipede Glomeris marginata and the distantly related centipede Lithobius forficatus now show that the early expression of col is present in both the insects and the myriapods. This finding may be seen as support for the traditional Atelocerata hypothesis, and thus arguing against a true in-group relationship of insects and crustaceans in the sense of the widely accepted Pancrustacea concept, or alternatively that the early expression of col in the tritocerebral segment of insects and myriapods may represent a case of convergence in gene deployment.
Species husbandry and embryo treatment
The handling of Glomeris marginata, Lithobius forficatus and Euperipatoides kanangrensis specimens is described in ,  and  respectively. After oviposition embryos of both myriapod species were allowed to develop at room temperature. Staging was done after  for Glomeris, after  for Lithobius and after  for Euperipatoides. The developmental stage of all embryos was determined by using the dye DAPI (4'-6-Diamidino-2-phenylindole).
A fragment of the collier gene was isolated from Glomeris, Euperipatoides and Lithobius each with degenerate primers from cDNA (SuperScript First Strand kit, Invitrogen). The primers col_fw1 (GCN CAY TTY GAR AAR CAR CC) and col_bw1 (TTR TTR TGN ACR AAC ATR TTR TC) for the initial PCR and col_fw1 and col_bw2 (GAT RTC NCK NGG RTT NCC NGC) for a semi-nested PCR were used to isolate the Glomeris fragment. The Euperipatoides fragment was isolated using primers col_fw1 and col_bw1 in a single PCR reaction. The Lithobius fragment was isolated using primers col_fw1 and col_bw1 in a first and col_fw2 (CAR GGC CAR CCN GTN GAR ATH GAR) and col_bw1 in a semi-nested PCR.
Sequences of the fragments were determined from both strands by means of Big Dye chemistry on an ABI3730XL analyser by a commercial sequencing service (Macrogen, Korea). Sequences are available in GenBank under the accession numbers AM279685 (Gm-col), FN827160 (Lf-col), FN827161 (Ek-col).
In situ hybridization and nuclei staining
Whole mount in situ hybridization for all species was performed as described for Glomeris in . The inner membrane of Lithobius embryos is (or becomes) very fragile after fixation. As a consequence it is often hard to remove the membrane completely. Unlike the case for Glomeris, however, this membrane does not disturb the in-situ hybridization procedure; it does not stain unspecifically or inhibit detection of specific staining. Embryos were analyzed under a Leica dissection microscope equipped with either an Axiocam (Zeiss) or a Leica DC100 digital camera. Brightness, contrast, and colour values were corrected in all images using the image processing software Adobe Photoshop CS2 (Version 9.0.1 for Apple Macintosh).
cDNA fragments of the ortholog of the Drosophila gene collier (col) have been amplified by RT-PCR from the myriapods Glomeris marginata (millipede) and Lithobius forficatus (centipede) and the onychophoran Euperipatoides kanangrensis. Orthology of the gene fragments has been assessed by comparison with published collier sequences from various metazoan species. There appears to be no risk of mistaking the isolated fragments with genes other than collier; no other similar sequences or indeed paralog of col is present in the published genomes of any protostome species . We therefore designate the corresponding genes as Gm-collier, Lf-collier and Ek-collier respectively.
collier expression in Glomeris
Expression of Glomeris collier is already detectable at the blastoderm stage (stage 0) as a broad closed ring surrounding the egg perpendicularly to the anterior-posterior axis of the embryo. This ring is situated in the anterior of the future head. Expression of the ring is weaker but broader in the dorsal of the embryo (this expression later disappears completely) (Additional file 1: Figure S1A, B, D-F). Expression of col is possibly already present at earlier stages, but in situ hybridization experiments for such stages are not workable. The dorsal extra-embryonic tissue does no longer express col (Figure 1A, a). At stage 1 the stripe starts fading from ventral tissue (Figure 1B), and at later stages up to stage 2 col is only visible as two separated shorter stripes (Figure 1C, C'). Finally, at stage 3, this expression disappears completely and at this point the embryos do not express col at all. At late stage 4, expression appears de novo in form of dots in the trunk lateral to the appendages. Soon after, at stage 5, stripes of expression extend from there to a position anterior to the limb buds (Figure 1D, D'). At this point expression in the central nervous system of the trunk also appears (Figure 1D, D'). In stage 6 embryos this expression intensifies; in addition the nervous system of the head segments also expresses col in a complex pattern (Figure 1E), and two thin stripes of col appear at the anterior rim of the head (Figure 1E).
The early stripe of col expression is situated in the anterior part of the mandibular (md) segment and the posterior part of the premandibular (pmd) segment (intercalary segment in insects) (Figure 1A). This is clear from the position of the col-stripe at later stages when the intersegmental indentations form (Figure 1B). We also provided a one-colour double staining using the segmental marker engrailed (en) in a series of early stage embryos (Figure 1F-H). Note that at this stage the en-stripe of the pmd segment has not yet formed (cf. ). Therefore it is clear that in the shown embryos the stripe between the antennal and md en-stripes represents expression of col. The area expressing early col (Figure 1A) is homologous to parasegment 0 of Drosophila (e.g. ). At subsequent stages the anterior-most and posterior-most expression of col disappears, so that as a consequence col expression does not abut en expression in the md segment any longer (Figure 1F-H). Instead, a clear gap is seen between the expression of en in these two segments and the expression of col covering the pmd/md boundary (Figure 1F-H).
collier expression in Lithobius
As in the millipede Glomeris, an early stripe of collier is also detectable in the centipede Lithobius at the blastoderm stage (Figures 2A, A' and Additional file 1: Figure S1C, G-I). In the embryo shown (Figure 2A), expression is somewhat weaker in the ventral part of the future germband, similar to the situation in Glomeris (Figure 2A, A'). We find that as in Glomeris this stripe forms a closed ring at the early blastoderm stage and, similar as in Glomeris, is weaker but broader in the future dorsal tissue (Additional file 1: Figure S1J-L). Because of the limited number of available young Lithobius embryos, we were not able to trace this early expression to unambiguously determine in which segment it lies. However, some aspects suggest that the expression in Glomeris and Lithobius at early developmental stages is indeed in a homologous position. First, the stripe/ring of expression lies in the anterior of the blastoderm stage embryo (Additional file 1: Figure S1). Second, the expression is weaker in dorsal tissue (Additional file 1: Figure S1). Third, the expression is broader in dorsal tissue compared to ventral tissue (Additional file 1: Figure S1). Fourth, the most ventral expression starts disappearing at an early stage in both myriapod species. At later stages the stripe of col expression disappears from the anterior of the future head. In stage 1 embryos col is not expressed and in stage 2 embryos expression is only seen faintly anterior to the labrum. The latter expression differs from the situation in Glomeris, where col is not expressed anterior to the developing labrum (Figure 2B, B'). Later, with beginning of blastokinesis and at least up to stage 6, expression appears and persists in the developing brain, the central nervous system and in segmental patches between the developing limbs (Figure 2C, C', D). Altogether the expression profile of Lithobius col is very similar to that of Glomeris col with the one exception of dominant expression anterior to the labrum in Lithobius.
Collier expression in Euperipatoides
In contrast to col expression in myriapods, onychophoran collier is not expressed at early blastoderm stages in form of an anterior transversal stripe. However col is expressed in a fuzzy domain around the blastopore in an early gastrulation embryo (Figure 3A, A') (here referred to as stage 0 embryo; cf. e.g.  for early stages). The earliest expression in the head appears much later as two lateral domains in the future brain anlage of the early stage II embryo (Additional file 2: Figure S2). Soon after in late stage II embryos this pattern transforms into a complex pattern in the developing brain (Figure 3B). This expression profile is similar to the expression in Glomeris and Lithobius at later developmental stages (cf. Figures 1E and 2C). In early stage III embryos col is also expressed in the anterior of the developing limbs (Additional file 2: Figure S2). In late stage III embryos col is expressed in spots in the limb buds and along the trunk ventral to the limbs (Additional file 2: Figure S2). The latter expression may be associated with the developing neuropil . In early stage IV embryos this expression profile persists, with exception of the dots in the limbs that disappear again (Figure 3C). At late stage IV the initially continuous expression along the trunk disappears from the position of the limbs (Figure 3C, D). At this point expression in the head transforms into broad domains in the developing brain (Figure 3D).
Faint staining also appears at this stage in the tips of the legs, the slime papillae, the jaws, and in the ventral nervous system (or ventral organs; for a discussion on the contribution of this tissue to the nervous system see e.g. [42, 43]) (Figure 1D). Note that this staining as well as the staining in the antennae may be unspecific due to the beginning of cuticle development.
Conserved and derived expression patterns of collier in arthropods
Data on collier expression and function are now available from a wide range of metazoan animals. These data suggest that the unifying theme, the ancestral function of col, is associated with the development of the nervous system [29, 32, 33, 44–50].
Including this study, collier orthologs have been examined in representatives of all extant arthropod classes [29, 32, 33]. Function of col in muscle differentiation and wing patterning appears to be arthropod or even only Drosophila specific [47, 51]. In addition the function of col in the patterning of the head segments was argued to be an insect-specific feature . In order to gain information on the ancestral expression patterns of col in arthropods we examined its expression in the onychophoran Euperipatoides kanangrensis. The onychophorans represent the sister-group to the arthropods and can therefore serve as outgroup to distinguish ancestral from derived features in arthropods (e.g. [2, 16, 21, 52]). Most of the observed expression patterns of col in the onychophoran Euperipatoides may be associated with the development of the nervous system. We can however not totally exclude the possibility that some of the col-expressing cells are involved in the development of other tissues than the nervous system, for example the mesoderm. Overall we find that most aspects of col expression seem to be conserved among arthropods and onychophorans, for example, expression in: 1) the anterior rim of the head lobes; 2) the developing brain; 3) the central nervous system of the trunk and 4) dorsolateral patches of the trunk. An obvious exception is the prominent expression of col anterior to the labrum in the centipede Lithobius. However the observed expression patterns in arthropods + onychophorans suggest at least partially conserved functions of col in this group.
The involvement of col in head segmentation in insects and myriapods represents a novelty, since the expression of col is absent from the crustacean, the chelicerate and the onychophoran. The question now is how likely it is that such novelty would have evolved independently in these two assumed rather distantly related arthropod groups, i.e. is due to convergent evolution (discussed below).
Early expression of collier in insects and myriapods: Support for the traditional Atelocerata concept?
It has long been known from manipulation studies in the fly Drosophila melanogaster that collier plays a crucial role in anterior head patterning and that a loss of col-function causes the loss of the head regions expressing col [29, 34, 53]. The recruitment of col expression in patterning the anterior head and the coincident formation of the limb-less intercalary segment was recently argued to represent a developmental novelty in insects . This idea was supported by the finding that col has no early expression and consequently also no early function in head development in a chelicerate and a crustacean that both have retained their tritocerebral appendage, the pedipalp and the second antenna respectively .
Our findings in two distantly related myriapods, the millipede Glomeris and the centipede Lithobius (note that the Lithobius data are less well worked-out than the Glomeris data), contradict this assumption and instead argue in favour of a conserved expression of col in head patterning in both, hexapods and myriapods. Together with the data provided by  on a chelicerate and a crustacean, our onychophoran data further support the idea that such early expression of col is not a plesiomorphic character for arthropods but a derived character.
Though the unique absence of the tritocerebral appendages in insects and myriapods is a long discussed common feature of these two arthropod classes, it was often considered as a mere convergence and not as a synapomorphy (e.g. [13, 24–26]). One of the strongest arguments for this assumption was that the "simple" loss of an appendage could easily be caused by any disturbance or mutation of the underlying genetic network needed for limb development [2, 39]. That arthropods lose or modify appendages is indeed frequent; in millipedes for example - but not in centipedes - the second maxilla is also missing. Consequently the lack of the tritocerebral appendage as possible synapomorphy for insects and myriapods was often, and obviously with some justification, understated in phylogenetic discussions (e.g. [54, 55]).
Now however it appears that the col gene may be involved in the development of the limb-less tritocerebral segment in both insects and myriapods. The finding that the same genetic factor(s)/mechanism(s) are possibly involved in the formation of the tritocerebral segment elevates this feature from a likely convergence to a possible synapomorphic character. This may therefore add a molecular perspective to the body of hitherto exclusively morphological data supporting the Atelocerata. In fact the most parsimonious scenario in terms of requiring the fewest number of evolutionary events affecting col would be the single recruitment of col in the formation of the tritocerebral segment (Figure 4). Since this would argue against an insect-crustacean relationship in the sense of the now widely accepted Pancrustacea concept, a case of gene convergence must still be considered likely for the expression of col in the tritocerebral segment of insects and myriapods (Figure 4D).
The conserved expression of col in the tritocerebral segments in insects may thus indeed represent an evolutionary novelty, but then the presence of col in the homologous region in myriapods has to be considered as another independently evolved evolutionary novelty as well.
Contradictory data in arthropod phylogeny: A case of homology versus convergence
The data presented here support the traditional Atelocerata theory, as do a number of morphological studies. Other data support the Myriochelata hypothesis joining chelicerates with myriapods. However the majority of data available today, including some morphological studies and most nucleotide sequence analysis, clearly support the close relationship of insects and crustaceans (Tetraconata) or even consider the insects as an in-group of the crustaceans (Pancrustacea). Consequently some of the data supporting contradicting evolutionary relationships must be considered to be either artificial, incorrectly interpreted or the result of convergent evolution. Convergent evolution, or convergence, is a much-discussed possibility to explain the presence of morphological data contradicting the Tetraconata/Pancrustacea hypothesis. It describes a scenario where similar morphological structures evolved independently in not (closely) related organisms as a response to similar environmental conditions. But convergent evolution is of course not restricted to morphological features only but must also be reflected by the underlying genetic levels controlling morphology. It is often argued that single genes or even genetic networks, or part of it, may be involved in the development of non-homologous structures (e.g. [56, 57]). In other words during evolution a single gene may be recruited independently because of its given function. Likewise gene networks may be recruited because of the conserved interaction of genes (e.g. [3, 58, 59]).
For the given case described in this paper this would mean that the collier gene could have been recruited independently in the formation of the appendage-less tritocerebral segment in insects and myriapods (Figure 4D). In that case the genetic network or at least part of it (the action of the collier gene) would be conserved (homologous), but the resulting modification of the tritocerebral segment, the lack of an appendage on this segment, would not.
Further investigation of the function of collier, and the genetic network within which it operates, may answer this question in the future. If, as seems most likely, the formation of the appendage-less tritocerebal segment is convergent in myriapods and insects, the hint of the same genetic mechanism behind this convergence offers a rare and important insight into the genetic basis of convergence. The degree to which the genetic patterning mechanism matches the two cases may offer important insights into how genes and their regulatory apparatus are recruited during the origin of novelties.
One of the key players in the development of the limb-less tritocerebral segment in insects (intercalary segment), the COE-family HLH transcription factor collier, is also specifically expressed in the homologous limb-less segment in myriapods. This finding contradicts the suggestion that the role of col in the development of the anterior head is an insect novelty .
Historically insects and myriapods have been united in the Atelocerata (or Tracheata), and the morphology of the tritocerebral segments was used as the main synapomorphy to support this group. Modern sequence-based phylogenetic analysis, however, now rather suggests a sister- or even in-group relationship of insects to crustaceans (Tetraconata or Pancrustacea). The apparently synapomorphic limb-less tritocerebral segment has been explained as an example of convergent evolution, since it appeared likely that a structure (like one of many appendages) could easily be lost independently. Our data question this argumentation, because we show that it is not only the mere loss of an appendage, but also the involvement of a specific gene that may argue in favour of the Atelocerata.
This study shows that comprehensive data (and taxon) sampling is often crucial to allow secure evolutionary statements. Although in line with the current opinion, i.e. the Pancrustacea/Tetraconata hypothesis, the data by  somewhat prematurely concluded that the involvement of col in the formation of the tritocerebral segment in insects would represent an evolutionary novelty.
Our data strengthen a possible synapomorphy (limb-less tritocerebral segment) for the unlikely Atelocerata concept, either challenging modern phylogenies, or presenting a complex case of parallel evolution. To find out which of either is the case must be subject of future investigation including an in-depth analysis of the genetic network involved in the formation of the tritocerebral segment in arthropods.
Dunlop JA, Arango CP: Pycnogonid affinities: a review. J Zool Syst Evol Res. 2005, 43: 8-21. 10.1111/j.1439-0469.2004.00284.x.
Janssen R, Eriksson BJ, Budd GE, Akam M, Prpic NM: Gene expression patterns in an onychophoran reveal that regionalization predates limb segmentation in pan-arthropods. Evol Dev. 2010, 12: 363-372. 10.1111/j.1525-142X.2010.00423.x.
Budd GE: Why are arthropods segmented?. Evol Dev. 2001, 5: 332-342. 10.1046/j.1525-142X.2001.01041.x.
Shear WA, Edgecombe GD: The geological record and phylogeny of the Myriapoda. Arthropod Struct Dev. 2010, 39: 174-190. 10.1016/j.asd.2009.11.002.
Edgecombe DG: Arthropod phylogeny: an overview from the perspectives of morphology, molecular data and the fossil record. Arthropod Struct Dev. 2010, 39: 74-87. 10.1016/j.asd.2009.10.002.
Snodgrass RE: Evolution of the Annelida, Onychophora and Arthropoda. Smithson Misc Collect. 1938, 97: 1-159.
Hennig W: Insect Phylogeny. 1981, John Wiley, New York
Bitsch C, Bitsch J: Phylogenetic relationships of basal hexapods among the mandibulate arthropods: a cladistic analysis based on comparative morphological characters. Zoologica Scripta. 2004, 33: 511-550. 10.1111/j.0300-3256.2004.00162.x.
Grimaldi DA: 400 million years on six legs: On the origin and early evolution of Hexapoda. Arthropod Struct Dev. 2009, 39: 191-203. 10.1016/j.asd.2009.10.008.
Tiegs OW, Manton SM: The evolution of Arthropoda. Biol Rev. 1958, 33: 255-337. 10.1111/j.1469-185X.1958.tb01258.x.
Manton SM: Arthropod Phylogeny - Modern synthesis. J Zoology. 1973, 171: 111-130. 10.1111/j.1469-7998.1973.tb07519.x.
Shear WA: End of the 'Uniramia' taxon. Nature. 1992, 359: 477-478. 10.1038/359477a0.
Friedrich M, Tautz D: Ribosomal DNA phylogeny of the major extant arthropod classes and the evolution of myriapods. Nature. 1995, 376: 165-167. 10.1038/376165a0.
Mallatt JM, Garey JR, Shultz JW: Ecdysozoan phylogeny and Bayesian inference: first use of nearly complete 28S and 18S rRNA gene sequences to classify the arthropods and their kin. Mol Phylogenet Evol. 2004, 31: 178-191. 10.1016/j.ympev.2003.07.013.
Pisani D, Poling LL, Lyons-Weiler M, Hedges SB: The colonization of land by animals: molecular phylogeny and divergence times among arthropods. BMC Biol. 2004, 2: 1-10.1186/1741-7007-2-1.
Mayer G, Whitington PM: Velvet worm development links myriapods with chelicerates. Proc R Soc B. 2009, 276: 3571-3579. 10.1098/rspb.2009.0950.
Regier JC, Shultz JW, Kambic RE: Pancrustacean phylogeny: hexapods are terrestrial crustaceans and maxillopods are not monophyletic. Proc R Soc B. 2005, 272: 395-401. 10.1098/rspb.2004.2917.
Harzsch S: Neurophylogeny: architecture of the nervous system and a fresh view on arthropod phylogeny. Integr Comp Biol. 2006, 46: 162-194. 10.1093/icb/icj011.
Strausfeld NJ, Strausfeld MC, Stowe S, Rowell D, Loesel R: The organization and evolutionary implications of neuropoils and their neurons in the brain of the onychophorans Euperipatoides rowelli. Arthropod Struct Dev. 2006, 135: 169-196. 10.1016/j.asd.2006.06.002.
Ungerer P, Scholtz G: Filling the gap between identified neuroblasts and neurons in crustaceans adds new support for Tetraconata. Proc R Soc B. 2008, 275: 369-376. 10.1098/rspb.2007.1391.
Dunn CW, Hejnol A, Matus DO, Pang K, Browne WE, Smith SA, Seaver E, Rouse GW, Obst M, Edgecombe GD, Sörensen MV, Haddock SHD, Schmidt-Rhaesa A, Okusu A, Kristensen RM, Wheeler WC, Martindale MQ, Giribet G: Broad phylogenomic sampling improves resolution of the animal tree of life. Nature. 2008, 452: 745-749. 10.1038/nature06614.
Bourlat SJ, Nielsen C, Economou AD, Telford MJ: Testing the new animal phylogeny: a phylum level molecular analysis of the animal kingdom. Mol Phylogenet Evol. 2008, 49: 23-31. 10.1016/j.ympev.2008.07.008.
Rota-Stabelli O, Kayal E, Gleeson D, Daub J, Boore JL, Telford MJ, Pisani D, Blaxter M, Lavrov DV: Ecdysozoan mitogenomics: evidene for a common origin of the legged invertebrates, the Panarthropoda. Genome Biol Evol. 2010, 2: 425-440. 10.1093/gbe/evq030.
Averof M, Akam M: Insect-crustacean relationships: insights from comparative developmental and molecular studies. Phil Trans R Soc B. 1995, 347: 293-303. 10.1098/rstb.1995.0028.
Boore JL, Lavrov DV, Brown WM: Gene translocation links insects and crustaceans. Nature. 1998, 392: 667-668. 10.1038/33577.
Dohle W: Are the insects terrestrial crustaceans? A discussion of some new facts and arguments and the proposal of the proper name "Tetraconata" for the monophyletic unit Crustacea and Hexapoda. Ann Soc Entomol Fr. 2001, 37: 85-103.
Mohler J, Mahaffey JW, Deutsch E, Vani K: Control of Drosophila head segment identity by the bZIP homeotic gene cnc. Development. 1995, 121: 237-247.
Gallitano-Mendel A, Finkelstein R: Novel segment polarity gene interactions during embryonic head development in Drosophila. Dev Biol. 1997, 192: 599-613. 10.1006/dbio.1997.8753.
Crozatier M, Valle D, Dubois L, Ibnsouda S, Vincent A: collier, a novel regulator of Drosophila head development, is expressed in a single mitotic domain. Curr Biol. 1996, 6: 707-718. 10.1016/S0960-9822(09)00452-7.
Finkelstein R, Perrimon N: The orthodenticle gene is regulated by bicoid and torso and specifies Drosophila head development. Nature. 1990, 346: 485-488. 10.1038/346485a0.
Lynch JA, Brent AE, Leaf DS, Pultz MA, Desplan C: Localized maternal orthodenticle patterns anterior and posterior in the long germ wasp Nasonia. Nature. 2006, 439: 728-732. 10.1038/nature04445.
Economou AD, Telford MJ: Comparative gene expression in the heads of Drosophila melanogaster and Tribolium castaneum and the segmental affinity of the Drosophila hypopharyngeal lobes. Evol Dev. 2009, 11: 88-96. 10.1111/j.1525-142X.2008.00305.x.
Schaeper ND, Pechmann M, Damen WGM, Prpic NM, Wimmer EA: Evolutionary plasticity of collier function in head development of diverse arthropods. Dev Biol. 2010, 344: 363-376. 10.1016/j.ydbio.2010.05.001.
Crozatier M, Valle D, Dubois L, Ibnsouda S, Vincent A: Head versus trunk patterning in the Drosophila embryo: collier requirement for formation of the intercalary segment. Development. 1999, 126: 4385-4394.
Janssen R, Prpic NM, Damen WGM: Gene expression suggests decoupled dorsal and ventral segmentation in the millipede Glomeris marginata (Myriapoda: Diplopoda). Dev Biol. 2004, 268: 89-104. 10.1016/j.ydbio.2003.12.021.
Janssen R, Budd GE: Gene expression suggests conserved aspects of Hox gene regulation in arthropods and provides additional support for monophyletic Myriapoda. EvoDevo. 2010, 1: 4-10.1186/2041-9139-1-4.
Kadner D, Stollewerk A: Neurogenesis in the chilopod Lithobius forficatus suggests more similarities to chelicerates than to insects. Dev Genes Evol. 2004, 214: 367-379. 10.1007/s00427-004-0419-z.
Walker MH, Tait NN: Studies of embryonic development and the reproductive cycle in ovoviviparous Australian Onychophora (Peripatopsidae). J Zool. 2004, 264: 333-354. 10.1017/S0952836904005837.
Prpic NM, Tautz D: The expression of the proximodistal axis patterning genes Distal-less and dachshund in the appendages of Glomeris marginata (Myriapoda: Diplopoda) suggests a special role of these genes in patterning the head appendages. Dev Biol. 2003, 260: 97-112. 10.1016/S0012-1606(03)00217-3.
Jackson DJ, Meyer NP, Seaver E, Pang K, McDougall C, Moy VN, Gordon K, Degnan BM, Martindale MQ, Burke RD, Peterson KJ: Developmental expression of COE across the Metazoa supports a conserved role in neuronal cell-type specification and mesodermal development. Dev Genes Evol. 2010, 220: 221-234. 10.1007/s00427-010-0343-3.
Eriksson BJ, Tait NN, Budd GE, Akam M: The involvement of engrailed and wingless during segmentation in the onychophoran Euperiaptoides kanangrensis (Peripatopsidae: Onychophora) (Reid 1996). Dev Genes Evol. 2009, 219: 249-264. 10.1007/s00427-009-0287-7.
Mayer G, Whitington PM: Neural development in Onychophora (velvet worms) suggests a step-wise evolution of segmentation in the nervous system of Panarthropoda. Dev Biol. 2009, 335: 263-275. 10.1016/j.ydbio.2009.08.011.
Eriksson BJ, Tait NN, Budd GE: Head development in the onychophoran Euperipatoides kanangrensis with particular reference to the central nervous system. J Morphol. 2003, 255: 1-23. 10.1002/jmor.10034.
Prasad BC, Ye B, Zackhary R, Schrader K, Seydoux G: unc-3, a gene required for axonal guidance in Caenorhabditis elegans, encodes a member of the O/E family of transcription factors. Development. 1998, 125: 1561-1568.
Dubois L, Enriquez J, Daburon V, Crozet F, Lebreton G, Crozatier M, Vincent A: collier transcription in a single Drosophila muscle lineage: the combinatorial control of muscle identity. Development. 2007, 134: 4347-4355. 10.1242/dev.008409.
Baumgardt M, Miguel-Aliaga I, Karlsson D, Ekman H, Thor S: Specification of neuronal identities by feedforward combinatorial coding. PLoS Biol. 2007, 5: e37-10.1371/journal.pbio.0050037.
Crozatier M, Vincent A: Requirement for the Drosophila COE transcription factor Collier in formation of an embryonic muscle: transcriptional response to Notch signalling. Development. 1999, 126: 1495-1504.
Crozatier M, Vincent A: Control of multidendritic neuron differentiation in Drosophila: The role of Collier. Dev Biol. 2008, 315: 232-242. 10.1016/j.ydbio.2007.12.030.
Dubois L, Vincent A: The COE- Collier/Olf1/EBF- transcription factors: structural conservation and diversity of developmental functions. Mech Dev. 2001, 108: 3-12. 10.1016/S0925-4773(01)00486-5.
Pang K, Matus DQ, Martindale MQ: The ancestral role of COE genes may have been in chemoreception: evidence from the development of the sea anemone, Nematostella vectensis (Phylum Cnidaria; Class Anthozoa). Dev Genes Evol. 2004, 214: 134-138. 10.1007/s00427-004-0383-7.
Vervoort M, Crozatier M, Valle D, Vincent A: The COE transcription factor Collier is a mediator of short-range hedgehog-induced patterning of the Drosophila wing. Curr Biol. 1999, 9: 632-639. 10.1016/S0960-9822(99)80285-1.
Regier JC, Shultz JW, Zwick A, Hussey A, Ball B, Wetzer R, Martin JW, Cunningham CW: Arthropod relationships revealed by phylogenomic analysis of nuclear protein-coding sequences. Nature. 2010, 463: 1079-1083. 10.1038/nature08742.
Seecoomar M, Agarwal S, Vani K, Yang G, Mohler J: knot is required for the hypopharyngeal lobe and its derivatives in the Drosophila embryo. Mech Dev. 2000, 91: 209-215. 10.1016/S0925-4773(99)00305-6.
Klass KD, Kristensen NP: The ground plan and affinities of hexapods: recent progress and open problems. Ann Soc Entomol Fr. Edited by: Deuve T. 2001, Origin of the Hexapoda, 37: 265-581.
Carapelli A, Lio P, Nardi F, van der Wath E, Frati F: Phylogenetic analysis of mitochondrial protein coding genes confirms the reciprocal paraphyly of Hexapoda and Crustacea. BMC Evol Biol. 2007, 7: S8-10.1186/1471-2148-7-S2-S8.
Couso JP: Segmentation, metamerism and the Cambrian explosion. Int J Dev Biol. 2009, 53: 1305-1316. 10.1387/ijdb.072425jc.
Chipman AD: Parallel evolution of segmentation by co-option of ancestral gene regulatory networks. BioEssays. 2010, 32: 60-70. 10.1002/bies.200900130.
True JR, Carroll SB: Gene co-option in physiological and morphological evolution. Annu Rev Cell Dev Biol. 2002, 18: 53-80. 10.1146/annurev.cellbio.18.020402.140619.
Saeko SV, Vrench V, Brakefield PM, Beldade P: Conserved developmental processes and the formation of evolutionary novelties: examples from butterfly wings. Phil Trans R Soc B. 2008, 363: 1549-1555. 10.1098/rstb.2007.2245.
Janssen R, Le Gouar M, Pechmann M, Poulin F, Bolognesi R, Schwager EE, Hopfen C, Colbourne JK, Budd GE, Brown SJ, Prpic NM, Kosiol C, Vervoort M, Damen WG, Balavoine G, McGregor AP: Conservation, loss, and redeployment of Wnt ligands in protostomes: implications for understanding the evolution of segment formation. BMC Evol Biol. 2010, 10: 374-10.1186/1471-2148-10-374.
Mayer G, Kato C, Quast B, Chrisholm RH, Landman KA, Quinn LM: Growth patterns in Onychophora (velvet worms): lack of a localized posterior proliferation zone. BMC Evol Biol. 2010, 10: 339-10.1186/1471-2148-10-339.
We wish to thank all reviewers involved in the publication of this paper. Their comments were helpful and most appreciated. This work has been supported by the Swedish Research Council (VR: grant to GEB), the European Union via the Marie Curie Training network "ZOONET" (MRTN-CT-2004-005624 (to GEB, WGMD and RJ)) and the DFG via SFB 572 of the University of Cologne (to WGMD and RJ). The authors wish to thank Jean Joss, Rolf Ericsson and especially Noel Tait for their help during onychophoran collection.
RJ designed the study, conducted the experiments and wrote the first draft manuscript. WGMD and GEB were involved in data discussion and writing the final version of the manuscript. All authors approved the final version of the manuscript.