<<

Fuxianhuiid ventral nerve cord and early evolution in

Jie Yanga, Javier Ortega-Hernándezb,1, Nicholas J. Butterfieldb, Yu Liua,c,d, George S. Boyanc, Jin-bo Houa, Tian Lane, and Xi-guang Zhanga,2

aYunnan Key Laboratory for Paleobiology, Yunnan University, Kunming 650091, China; bDepartment of Earth Sciences, University of Cambridge, Cambridge CB2 3EQ, United Kingdom; cDevelopmental Neurobiology Group, Biocenter, Ludwig-Maximilians-Universität, 82152 Martinsried, Germany; dGeoBio-Center Ludwig-Maximilians-Universität, Munich 80333, Germany; and eCollege of Resources and Environmental Engineering, Guizhou University, Guiyang 550003, China

Edited by Gregory D. Edgecombe, The Natural History Museum, London, United Kingdom, and accepted by the Editorial Board January 29, 2016 (received for review November 14, 2015) Panarthropods are typified by disparate grades of neurological feature as the VNC. This interpretation is supported by com- organization reflecting a complex evolutionary history. The fossil parisons with other preserved components of the internal anat- record offers a unique opportunity to reconstruct early character omy. For instance, the VNC can be readily distinguished from evolution of the nervous system via exceptional preservation in the digestive tract of C. kunmingensis, which is expressed as a extinct representatives. Here we describe the neurological archi- comparatively larger (maximum width, ∼860 μm) but fully tecture of the ventral nerve cord (VNC) in the upper-stem group compressed, linear structure running almost the entire length of euarthropod Chengjiangocaris kunmingensis from the early Cam- the (23, figure 1 d and e). The VNC extends from at least brian Xiaoshiba Lagerstätte (South China). The VNC of C. kunmin- the five anteriormost reduced trunk tergites (i.e., dorsal exo- gensis comprises a homonymous series of condensed ganglia that skeletal plates) to tergite T23 at the posterior end of the trunk extend throughout the body, each associated with a pair of bi- (Figs. 1 and 2 and Figs. S1, S2, and S3). Although the VNC ramous limbs. Submillimetric preservation reveals numerous seg- would have continued into the head region in life, our material mental and intersegmental nerve roots emerging from both sides does not preserve any anterior neurological structures that have of the VNC, which correspond topologically to the peripheral been previously identified as the dorsal brain or nerves leading to nerves of extant Priapulida and Onychophora. The fuxianhuiid the antennae and specialized postantennal appendages (8). The VNC indicates that ancestral neurological features of Ecdysozoa absence of fossilized brains in these otherwise exceptionally persisted into derived members of stem-group Euarthropoda but preserved specimens can be attributed to the small sample size were later lost in crown-group representatives. These findings il- and moderate postmortem disarticulation; likewise, other studies luminate the VNC ground pattern in Panarthropoda and suggest addressing neurological structures in Cambrian fossils rarely the independent secondary loss of cycloneuralian-like neurological report the CNS preserved in its entirety (8–11). The VNC also characters in Tardigrada and Euarthropoda. expresses a distinctive dark/light color banding throughout its length, with the dark bands varying from black to reddish-brown stem-group Euarthropoda | Onychophora | phylogeny | Cambrian Explosion | Xiaoshiba Lagerstätte Significance

he nervous system represents a critical source of phylogenetic Understanding the evolution of the CNS is fundamental for Tinformation and has been used extensively for exploring the resolving the phylogenetic relationships within Panarthropoda evolutionary relationships of extant Panarthropoda (i.e., Ony- – (Euarthropoda, Tardigrada, Onychophora). The ground pattern chophora, Tardigrada, Euarthropoda) (1 7). Identification of of the panarthropod CNS remains elusive, however, as there is fossilized nervous tissues has provided a unique perspective on uncertainty on which neurological characters can be regarded early euarthropod brain neuroanatomy and suggests that broad as ancestral among extant phyla. Here we describe the ventral patterns of extant neurological diversity were already in place by – nerve cord (VNC) in Chengjiangocaris kunmingensis, an early the Cambrian (8 11). The ventral nerve cord (VNC) reflects Cambrian euarthropod from South China. The VNC reveals fundamental aspects of panarthropod body organization that extraordinary detail, including condensed ganglia and regu- complement the organization of the brain and together illuminate – – larly spaced nerve roots that correspond topologically to the the evolution of the CNS (1 3, 5, 7, 12 16). The early evolutionary peripheral nerves of Priapulida and Onychophora. Our findings history of the panarthropod postcephalic CNS, however, remains demonstrate the persistence of ancestral neurological features obscure due to the exclusive preservation of brains in most available of Ecdysozoa in early euarthropods and help to reconstruct the fossils (8, 10, 11). Moreover, the unresolved phylogenetic relation- VNC ground pattern in Panarthropoda. ships within Panarthropoda complicate accurate reconstruction of – the CNS ground pattern (16 22). In this study, we demonstrate the Author contributions: J.Y., N.J.B., and X.-g.Z. designed research; J.Y., J.O.-H., N.J.B., Y.L., exceptional preservation of postcephalic neurological features in the J.-b.H., T.L., and X.-g.Z. performed research; Y.L. and G.S.B. contributed new reagents/ early Cambrian fuxianhuiid Chengjiangocaris kunmingensis,anup- analytic tools; J.O.-H. analyzed data; J.O.-H. and N.J.B. wrote the paper; J.Y. collected and per stem-group euarthropod (17) from the Xiaoshiba Lagerstätte, prepared all the fossil material; J.O.-H. performed light photography; N.J.B. and X.-g.Z. discussed and approved the manuscript; Y.L. and G.S.B. performed immunohistochemistry South China (23). These fossils clarify the neurological organization and living animal microscopy; and J.-b.H. and T.L. collected fossil material and of the VNC in early euarthropod ancestors, thereby polarizing the performed photography. evolution of the panarthropod CNS. The authors declare no conflict of interest. This article is a PNAS Direct Submission. G.D.E. is a guest editor invited by the Editorial Results Board. Five individuals of C. kunmingensis display a narrow (maximum 1Present address: Department of Zoology, University of Cambridge, Cambridge CB2 3EJ, width, ∼170 μm) and slightly convex rope-like structure with a United Kingdom. metameric pattern that extends medially throughout the body 2To whom correspondence should be addressed. Email: [email protected]. (Figs. 1 and 2, Figs. S1 and S2, and Table S1); its segmental This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. organization and position at the ventral midline identifies this 1073/pnas.1522434113/-/DCSupplemental.

2988–2993 | PNAS | March 15, 2016 | vol. 113 | no. 11 www.pnas.org/cgi/doi/10.1073/pnas.1522434113 Downloaded by guest on September 23, 2021 Fig. 1. VNC in C. kunmingensis. Anterior at top. (A) YKLP 12023 (Holotype), complete specimen preserved in dorsal view with taphonomically dissected head shield (hs) showing internal organization of anterior region. (B) YKLP 12324, complete specimen preserved in dorsolateral view showing preserved VNC extending throughout the almost the entire length of the body. (C) YKLP 12324, magnification of VNC on the anterior trunk region showing the differential preservation of the condensed ganglia (ga) and longitudinal connectives (cn) as dark and light colored bands, respectively, and one-to-one correspondence between the ganglia and the walking legs (wl) (upper box in B). (D) YKLP 12024, magnification of the VNC on the posterior trunk region showing the progressive reduction of size of the ganglia and connectives toward the rear end of the body (lower box in B). Dotted lines highlight the anterior and posterior tergal borders of T15 for comparative purposes with the number of preserved ganglia. , antennae; SPA, specialized postantennal appendage; Tn, trunk tergites.

between specimens (Figs. 1 and 2 and Figs. S1 and S2C); this remain separate throughout the entire VNC, with no indication variability in color reflects differences in the extent of weathering of fusion or specialization. There is a progressive reduction in between individual specimens. Raman spectroscopy indicates the the size of individual ganglia along the body, with the anterior- presence of residual organic carbon associated with the dark- most being ∼3 times longer and 1.5 times wider than the pos- colored bands (Fig. S4) but not in the light ones, likely mirroring terior ones (Fig. 1 C and D). The preserved proximal portions of differences in their original histology and early diagenesis. In the trunk endopods indicate that each ganglion was associated addition to the carbonaceous films, the VNC is typified by with a single pair of biramous appendages (Fig. 1C and Fig. S2 A modest relief relative to adjacent exoskeletal features, attesting and B), which, like the ganglia, become progressively smaller to a degree of early diagenetic permineralization before com- posteriorly (Fig. 1D). As such, each of the five anteriormost plete degradational collapse (24). reduced tergites (Fig. 1A) correlates with an individual ganglion Neurologically, the C. kunmingensis VNC consists of an and leg pair, whereas the comparatively larger tergites T6–T26 interconnected series of separate metameric ganglia (dark col- overlie up to four ganglia and their corresponding appendages ored bands) (Figs. 1 and 2 and Figs. S1 and S3). Longitudinal (Fig. 1D and Figs. S2 A and B and S3). This organization reflects connectives are not morphologically discrete, but their presence the notable dorsoventral segmental mismatch that typifies the EVOLUTION is strongly suggested by the alternating light colored bands be- fuxianhuiid trunk (23, 25). tween the condensed ganglia. The distinct preservation style of The unique disarticulation pattern of the head shield in the these neurological structures likely stems from differences in Xiaoshiba fuxianhuiids (23)—aided by mechanical preparation— their original histology; whereas the condensed ganglia are reveals submillimetric neurological detail in C. kunmingensis. enriched in lipids due to the presence of abundant somata, the In specimen YKLP 12026, the exposed VNC extends from the connectives consist of neurites and lack cell bodies (1, 3–5). The oral region into tergite T6, comprising a total of seven ganglia ganglia have an elongate subelliptic outline (maximum length, (Fig. 2 A–C and Fig. S1). Close inspection reveals the presence ∼600 μm; 3.5:1 length/width) and are roughly three times longer of delicate nerve roots (maximum length, ∼209 μm; width, 9 μm) than the connectives (maximum length, ∼206 μm). The ganglia emerging at both sides of the VNC. These nerve roots are

Yang et al. PNAS | March 15, 2016 | vol. 113 | no. 11 | 2989 Downloaded by guest on September 23, 2021 Fig. 2. Fine neurological organization of the VNC in C. kunmingensis, YKLP 12026. Anterior is to the left. (A) Completely articulated specimen preserved in laterodorsal orientation with displaced head shield exposing VNC on anterior trunk region. (B) VNC showing the presence of seven sets of condensed ganglia (ga) linked by longitudinal connectives (cn) (box in A). (C) Composite fluorescence photograph of VNC (box in A). (D) Magnification of the VNC (box in B) showing regularly spaced peripheral nerve roots (arrowheads) emerging from the condensed ganglia and connective. (E) Composite fluorescence photograph magnification of the VNC (box in C).

2990 | www.pnas.org/cgi/doi/10.1073/pnas.1522434113 Yang et al. Downloaded by guest on September 23, 2021 regularly spaced (∼17–20 μm) and originate from both the context of extant representatives, the postcephalic CNS of ganglia (i.e., segmental) and interganglia connectives (i.e., in- C. kunmingensis exhibits notable similarities with the thoracic tersegmental) at acute angles (∼40–60°) relative to the VNC VNC of the notostracan Triops cancriformis, which consists of a (Fig. 2 C–E and Figs. S1 B and C and S3). The emergent chev- homonymous series of separate ganglia linked by short connec- ron-like pattern most likely resulted from moderate postmortem tives that become progressively smaller toward the posterior end displacement of the distal portions of the peripheral nerves rel- (1, 3). The overall morphology of T. cancriformis further re- ative to the VNC. There are at least a dozen paired sets of roots sembles C. kunmingensis—and generally other fuxianhuiids—in associated with each ganglion and four to six nerve roots asso- the presence of dorsoventral trunk segmental mismatch ciated with the longitudinal connectives. Preferential preserva- expressed as appendage polypody (23, 25, 26). A similarly un- tion of the proximal bases of the nerve roots precludes the specialized VNC is also found in typified by a trunk identification of precise relationships between these structures region with undifferentiated limb-bearing segments, such as the and the exoskeletal morphology, such as the walking legs. The remipedes (5, 7) (Fig. S5C). Unlike C. kunmingensis and lack of distal neurological preservation suggests that, similarly to T. cancriformis, however, the body organization of remipedes 3D guts in Burgess Shale fossils (24), the VNC and the proximal does not evince any type of trunk segmental mismatch. Thus, the nerve roots were permineralized during early diagenesis due to paleoneurological data available indicate that the fuxianhuiid the chemically reactive composition of adjacent lipid-rich ganglia CNS combines a relatively simple VNC without signs of func- (8). Fluorescence microscopy corroborates the anatomical and tional specialization (Figs. 1–3) with a tripartite brain bearing compositional continuity of the regularly spaced nerve roots and both olfactory and optic lobes (8). Considering the position of VNC throughout the body (Fig. 2 C and E and Dataset S1). fuxianhuiids within upper-stem Euarthropoda (17), this char- acter combination could suggest the evolutionary convergence of Discussion either the VNC of Chengjiangocaris with Branchiopoda or the The Xiaoshiba fossils provide previously unidentified insights on complex brain of Fuxianhuia with Malacostraca. Alternatively, the neurological diversity of Cambrian euarthropods. Within the the fuxianhuiid CNS may approximate the complex brain (sans EVOLUTION

Fig. 3. Simplified cladogram showing the evolution of the postcephalic CNS in Panarthropoda. Detailed results of the phylogenetic analysis are provided in Fig. S6 and SI Text. The topology supports a single origin for the condensed ganglia (ga) in the VNC in a clade including Tardigrada and Euarthropoda; note that the presence of multiple intersegmental peripheral nerves (ipn) in C. kunmingensis represents an ancestral condition. Given the morphological similarity between peripheral and leg nerve roots, the presence of a single pair of leg nerves (lgn) in C. kunmingensis is hypothetical (dashed lines) and based on the condition observed in crown-group Euarthropoda. †, fossil taxa; ?, uncertain character polarity within total-group Euarthropoda. asn, anterior segmental nerve; cn, longitudinal connectives; co, commissure; dln, dorsolateral longitudinal nerve; ico, interpedal median commissure; irc, incomplete ring commissure; pn, peripheral nerve; psn, posterior segmental nerve; rc, ring commissure. Reconstruction of VNC in Onychophora adapted from ref. 13.

Yang et al. PNAS | March 15, 2016 | vol. 113 | no. 11 | 2991 Downloaded by guest on September 23, 2021 optic lobes) and unspecialized VNC of remipedes (5, 7). If the (31) and Onychophora (12, 13, 15, 29) but are otherwise greatly neurological organization of fuxianhuiids reflects an ancestral reduced in number or completely lost in crown-group Euar- condition as suggested by their phylogenetic position, it would thropoda and Tardigrada, respectively (1, 5, 16, 28) (Fig. S5); the support the symplesiomorphic nature of a malacostracan-like secondary reduction/loss is recognized as a result of parallel brain for Pancrustacea/Tetraconata (6, 8) and also imply an in- evolution between these lineages. The phylogenetic analysis also dependent increase in VNC complexity among extant indicates that several neurological characters shared between lineages. However, the lack of consensus pertaining to the phylo- Onychophora and Tardigrada are actually symplesiomorphic and genetic placement of Branchiopoda, , and Malacostraca likely reflect the broader ground pattern of Panarthropoda (Fig. 3); relative to Hexapoda based on molecular and neuromophological these include the median interpedal commissures, an orthogonal- data (5–7, 27) complicates deciding between these hypotheses based like organization, and paired segmental leg nerves (SI Text). A on the available paleoneurological information. lateralized VNC is resolved as synapomorphic for Onychophora The neurological organization of the fuxianhuiid VNC leads to (SI Text). Taken together, these findings argue against the sec- more informative comparisons with other panarthropod groups. ondary loss of morphologically discrete segmental ganglia in the Apart from Euarthropoda (1–5, 9), the presence of segmental VNC of Onychophora (30). By contrast, the evolution of the ganglia is also characteristic of the CNS in Tardigrada (18, 28) VNC in crown-group Euarthropoda is the result of secondary (Fig. S5B) and has been used to advocate the sister-group re- simplification relative to the CNS of cycloneuralians and other lationship between these clades (16, 17, 20). Thus, the presence panarthropods, as informed by the transitional neurological or- of segmental ganglia in C. kunmingensis provides a minimum ganization of C. kunmingensis (Fig. 3). phylogenetic threshold for the evolution of this feature in total- The integration of paleoneurological data with the present un- group Euarthropoda (SI Text). derstanding of the developmental biology of extant groups illumi- At a broader phylogenetic level, the neurological organization nates major issues pertaining the evolution of the nervous system of C. kunmingensis also exhibits marked similarities with the in Panarthropoda. Our results indicate that the ground pattern of VNC of Onychophora (Fig. 3 and Fig. S5A), namely the pres- the panarthropod postcephalic CNS incorporates a paired, but not ence of numerous regularly spaced peripheral nerve roots lateralized, VNC with median interpedal commissures, an or- (12–15, 29, 30). A major difference, however, is that the ony- thogonal organization with complete ring-commissures, paired leg chophoran postcephalic CNS is lateralized and lacks morpho- nerves, and intersegmental peripheral nerves (SI Text); in this logically discrete segmental ganglia despite clearly having a context, an orthogonal VNC with complete ring commissures is a segmented organization (SI Text). Although the proximal preser- symplesiomorphy inherited from a cycloneuralian-like ancestor. vation of the peripheral nerves makes it uncertain whether the The presence of condensed segmental ganglia, a lateralized or- C. kunmingensis VNC possessed complete ring commissures like ganization, anteriorly displaced leg nerves following the para- – those in onychophorans (12 14), this feature could potentially add segments, and a stomatogastric ganglion (SI Text) (Fig. 3), represent morphological support for the sister-group relationship between derived features that were acquired later in the evolutionary Euarthropoda and Onychophora suggested by molecular phylog- history of the different panarthropod phyla. enies (21, 22). Among crown-group euarthropods, there are only a restricted number of intersegmental peripheral nerves emerging Methods from the longitudinal connectives in the VNC of some pan- All of the described material is deposited at the Key Laboratory for Paleo- crustaceans (1, 4, 5) (SI Text,Fig.3,andFig. S5 C and D). Alter- biology, Yunnan University, Kunming, China (YKLP). natively, it is possible that the presence of numerous peripheral nerves in the VNC of C. kunmingensis and Onychophora is sym- Imaging. Fossils were photographed with a Nikon D3X fitted with a Nikon plesiomorphic, as comparable structures are also found in the or- AF-S Micro Nikkor 105-mm lens. For close-up images, a LEICA M205-C stereo- thogonal CNS of Priapulida (31) and other protostomes (12, 13, 15). microscope fitted with a Leica DFC 500 digital camera was used under di- In any event, the postcephalic CNS of C. kunmingensis reveals a rectional illumination provided by a LEICA LED5000 MCITM. Geochemical unique combination of neurological characters otherwise unknown analyses were performed with an inVia Raman microscope (Renishaw). in any extant group within Panarthropoda (Fig. S3). Fluorescence photography was performed as described in ref. 32. To test these different hypotheses and to clarify the polarity of neurological characters in extant and extinct groups, we per- Phylogenetic Analysis. The data matrix includes 50 taxa and 95 characters (Dataset S2 and SI Text). The analysis was run in TNT (33) under New Tech- formed a comprehensive phylogenetic analysis incorporating the nology Search, using Driven Search with Sectorial Search, Ratchet, Drift, and available fossil data on the postcephalic CNS of total-group Tree fusing options activated in standard settings (34, 35). The analysis was Panarthropoda (SI Text and Dataset S2). The results support the set to find the minimum tree length 100 times and to collapse trees after sister-group relationship between Tardigrada and Euarthropoda each search. All characters were treated as unordered. For an initial analysis, (Fig. 3 and Figs. S6 and S7), corroborated by several unambig- all characters were treated as equally weighted (Fig. S6A); subsequent rep- uous synapomorphies of the VNC. These synapomorphies in- etitions with variable concavity values (k) were used to explore the effect of clude the presence of condensed segmental ganglia connected by different degrees of homoplasy penalization to test the robustness of the commissures, the anterior shift of the leg nerves following a dataset (Fig. S6 B–D) (36). parasegmental organization, and the presence of a stomatogas- ACKNOWLEDGMENTS. We thank K.-S. Du and J.-F. He for assistance with fossil tric ganglion associated with the tritocerebral segment (SI Text) collection. B. J. Eriksson (University of Vienna), G. Mayer (University of Leipzig), (14, 16, 18, 19, 28). In this context, the presence of numerous and G. Bicker (University of Veterinary Medicine Hannover) generously segmental and intersegmental peripheral nerves in C. kunmin- contributed photographic material for Fig. S5. This work was supported by gensis (Fig. 2 B–E) is recognized as an ancestral condition that National Natural Science Foundation of China (NSFC) Grants 41472022 and U1402232 (to J.Y. and X.-g.Z.), a research fellowship at Emmanuel College persisted in derived members of upper stem-group Euar- and a Herchel Smith fellowship (both University of Cambridge; to J.O.-H.), thropoda (SI Text). Among extant groups, homologs of the in- and a Ludwig Maximilians Universität München excellent Junior Researcher tersegmental peripheral nerves are expressed in both Priapulida fund and NSFC Grant 41528202 (to Y.L.).

1. Harzsch S, Waloszek D (2000) Serotonin-immunoreactive neurons in the ventral nerve Xiphosura): Evidence for a conserved protocerebral architecture in Euarthropoda. cord of Crustacea: A character to study aspects of phylogeny. Arthropod Arthropod Struct Dev 34(3):327–342. Struct Dev 29(4):307–322. 3. Fritsch M, Richter S (2010) The formation of the nervous system during larval devel- 2. Harzsch S, Wildt M, Battelle B, Waloszek D (2005) Immunohistochemical localization opment in Triops cancriformis (Bosc) (crustacea, Branchiopoda): An immunohisto- of in the nervous system of larval Limulus polyphemus (, chemical survey. J Morphol 271(12):1457–1481.

2992 | www.pnas.org/cgi/doi/10.1073/pnas.1522434113 Yang et al. Downloaded by guest on September 23, 2021 4. Richter S, et al. (2010) neurophylogeny: Suggested terms and definitions 27. von Reumont BM, et al. (2012) Pancrustacean phylogeny in the light of new phylo- for a neuroanatomical glossary. Front Zool 7:29. genomic data: Support for Remipedia as the possible sister group of Hexapoda. Mol 5. Stemme T, et al. (2013) Serotonin-immunoreactive neurons in the ventral nerve cord Biol Evol 29(3):1031–1045. of Remipedia (Crustacea): Support for a sister group relationship of Remipedia and 28. Schulze C, Neves RC,Schmidt-RhaesaA (2014) Comparative immunohistochemical in- Hexapoda? BMC Evol Biol 13:119. vestigation on the nervous system of two species of Arthrotardigrada (Hetero- 6. Strausfeld NJ, Andrew DR (2011) A new view of -crustacean relationships I. In- tardigrada, Tardigrada). Zool Anz 253(3):225–235. ferences from neural cladistics and comparative neuroanatomy. Arthropod Struct Dev 29. Eriksson BJ, Tait NN, Budd GE (2003) Head development in the onychophoran Eu- 40(3):276–288. peripatoides kanangrensis with particular reference to the . – 7. Fanenbruck M, Harzsch S (2005) A brain atlas of Godzilliognomus frondosus Yager, J Morphol 255(1):1 23. 1989 (Remipedia, Godzilliidae) and comparison with the brain of Speleonectes tulu- 30. Eriksson BJ, Larson ET, Thörnqvist PO, Tait NN, Budd GE (2005) Expression of engrailed in the developing brain and appendages of the onychophoran euperipatoides ka- mensis Yager, 1987 (Remipedia, Speleonectidae): Implications for arthropod rela- nangrensis (Reid). J Exp Zoolog B Mol Dev Evol 304(3):220–228. tionships. Arthropod Struct Dev 34(3):343–378. 31. Rothe BH, Schmidt-Rhaesa A (2010) Structure of the nervous system in Tubiluchus 8. Ma X, Edgecombe GD, Hou X, Goral T, Strausfeld NJ (2015) Preservational pathways troglodytes (Priapulida). Invertebr Biol 129(1):39–58. of corresponding brains of a Cambrian euarthropod. Curr Biol 25(22):2969–2975. 32. Liu Y, Scholtz G, Hou X (2015) When a 520 million-year-old Chengjiang fossil meets a 9. Tanaka G, Hou X, Ma X, Edgecombe GD, Strausfeld NJ (2013) Chelicerate neural modern micro-CT–a case study. Sci Rep 5:12802. ground pattern in a Cambrian great appendage arthropod. Nature 502(7471): 33. Goloboff PA, Farris JS, Nixon KC (2008) TNT, a free program for phylogenetic analysis. 364–367. Cladistics 24(5):415–428. 10. Cong P, Ma X, Hou X, Edgecombe GD, Strausfeld NJ (2014) Brain structure resolves the 34. Goloboff PA (1999) Analyzing large datasets in reasonable times: Solutions for – segmental affinity of anomalocaridid appendages. Nature 513(7519):538 542. composite optima. Cladistics 15(4):415–428. 11. Ortega-Hernández J (2015) Homology of head sclerites in Burgess Shale euarthro- 35. Nixon KC (1999) The parsimony ratchet, a new method for rapid parsimony analysis. – pods. Curr Biol 25(12):1625 1631. Cladistics 15(4):407–414. 12. Mayer G, Harzsch S (2007) Immunolocalization of serotonin in Onychophora argues 36. Yang J, et al. (2015) A superarmored lobopodian from the Cambrian of China and against segmental ganglia being an ancestral feature of . BMC Evol Biol 7: early disparity in the evolution of Onychophora. Proc Natl Acad Sci USA 112(28): 118. 8678–8683. 13. Mayer G, Harzsch S (2008) Distribution of serotonin in the trunk of Metaperipatus 37. Mittmann B, Scholtz G (2003) Development of the nervous system in the “head” of blainvillei (Onychophora, Peripatopsidae): Implications for the evolution of the ner- Limulus polyphemus (Chelicerata: Xiphosura): Morphological evidence for a corre- vous system in Arthropoda. J Comp Neurol 507(2):1196–1208. spondence between the segments of the chelicerae and of the (first) antennae of 14. Mayer G, Whitington PM, Sunnucks P, Pflüger H-J (2010) A revision of brain com- Mandibulata. Dev Genes Evol 213(1):9–17. position in Onychophora (velvet worms) suggests that the tritocerebrum evolved in 38. Eriksson BJ, Budd GE (2000) Onychophoran cephalic nerves and their bearing on our arthropods. BMC Evol Biol 10:255. understanding of head segmentation and stem-group evolution of Arthropoda. 15. Whitington PM, Mayer G (2011) The origins of the arthropod nervous system: Insights Arthropod Struct Dev 29(3):197–209. from the Onychophora. Arthropod Struct Dev 40(3):193–209. 39. Møller OS, Olesen J, Høeg JT (2003) SEM studies on the early larval development of 16. Mayer G, et al. (2013) Selective neuronal staining in and onychophorans Triops cancriformis (Bosc) (Crustacea: Branchiopoda, Notostraca). Acta Zool 84(4): provides insights into the evolution of segmental ganglia in panarthropods. BMC Evol 267–284. Biol 13:230. 40. Suzuki Y, Kondo A, Bergström J (2008) Morphological requirements in limulid and 17. Ortega-Hernández J (2016) Making sense of ‘lower’ and ‘upper’ stem-group Euar- decapod gills: A case study in deducing the function of lamellipedian exopod la- – thropoda, with comments on the strict use of the name Arthropoda von Siebold, mellae. Acta Palaeontol Pol 53(2):275 283. 1848. Biol Rev Camb Philos Soc 91(1):255–273. 41. Ortega-Hernández J, Legg DA, Braddy SJ (2013) The phylogeny of aglaspidid ar- – 18. Gross V, Mayer G (2015) Neural development in the Hypsibius dujardini thropods and the internal relationships within Artiopoda. Cladistics 29(1):15 45. 42. Schulze C, Schmidt-Rhaesa A (2013) The architecture of the nervous system of Echi- based on anti-acetylated α-tubulin immunolabeling. Evodevo 6:12. niscus testudo (Echiniscoidea, Heterotardigrada). J Limnol 72(1):44–53. 19. Mayer G, Kauschke S, Rüdiger J, Stevenson PA (2013) Neural markers reveal a one- 43. Persson DK, Halberg KA, Jørgensen A, Møbjerg N, Kristensen RM (2012) Neuroanat- segmented head in tardigrades (water bears). PLoS One 8(3):e59090. omy of Halobiotus crispae (Eutardigrada: Hypsibiidae): Tardigrade brain structure 20. Smith MR, Ortega-Hernández J (2014) Hallucigenia’s onychophoran-like claws and supports the clade Panarthropoda. J Morphol 273(11):1227–1245. the case for Tactopoda. Nature 514(7522):363–366. 44. Hou X-G, Ma X-Y, Zhao J, Bergström J (2004) The lobopodian Paucipodia inermis from 21. Campbell LI, et al. (2011) MicroRNAs and phylogenomics resolve the relationships of the lower Cambrian Chengjiang fauna, Yunnan, China. Lethaia 37(3):235–244. Tardigrada and suggest that velvet worms are the sister group of Arthropoda. Proc 45. Zantke J, Wolff C, Scholtz G (2008) Three-dimensional reconstruction of the central Natl Acad Sci USA 108(38):15920–15924. nervous system of Macrobiotus hufelandi (Eutardigrada, Parachela): Implications for 22. Borner J, Rehm P, Schill RO, Ebersberger I, Burmester T (2014) A transcriptome ap- the phylogenetic position of Tardigrada. Zoomorphology 127(1):21–36. – proach to ecdysozoan phylogeny. Mol Phylogenet Evol 80:79 87. 46. Boyan GS, Williams JLD, Herbert Z (2008) Fascicle switching generates a chiasmal 23. Yang J, Ortega-Hernández J, Butterfield NJ, Zhang X-G (2013) Specialized append- neuroarchitecture in the embryonic central body of the grasshopper Schistocerca ages in fuxianhuiids and the head organization of early euarthropods. Nature gregaria. Arthropod Struct Dev 37(6):539–544. – 494(7438):468 471. 47. Stegner ME, Brenneis G, Richter S (2014) The ventral nerve cord in Cephalocarida 24. Butterfield NJ (2002) Leanchoilia guts and the interpretation of three-dimensional (Crustacea): New insights into the ground pattern of Tetraconata. J Morphol 275(3): structures in Burgess Shale-type fossils. Paleobiology 28(1):155–177. 269–294. 25. Ortega-Hernández J, Brena C (2012) Ancestral patterning of tergite formation in a 48. Maas A, Waloszek D (2001) Cambrian derivatives of the early arthropod stem lineage, centipede suggests derived mode of trunk segmentation in trilobites. PLoS One 7(12): pentastomids, tardigrades and lobopodians: An ‘Orsten’ perspective. Zool Anz 240(3): e52623. 451–459. 26. Olesen J, Møller OS (2013) Notostraca. Atlas of Crustacean Larvae, eds Martin JW, 49. Legg DA, Sutton MD, Edgecombe GD (2013) Arthropod fossil data increase congru- Olesen J, Høeg T (John Hopkins Univ Press, Baltimore), pp 40–46. ence of morphological and molecular phylogenies. Nat Commun 4:2485. EVOLUTION

Yang et al. PNAS | March 15, 2016 | vol. 113 | no. 11 | 2993 Downloaded by guest on September 23, 2021