Refining the Ciona intestinalis Model of Central Nervous System Regeneration Carl Dahlberg1.*, Helene �� ` Auger2.*, Sam Dupont1, Yasunori Sasakura3, Mike Thorndyke1, Jean-Stephane�� Joly2 1 Department of Marine Ecology, Goteborg �� University, Fiskebackskil, �� Sweden, 2 U1126/INRA ����Morphogenese ` du systeme ` nerveux des chordes���� �� group, DEPSN, UPR2197, Institut Fessard, CNRS, Gif sur Yvette, France, 3 Shimoda Marine Research Center, University of Tsukuba, Shimoda, Shizuoka, Japan Abstract Background: New, practical models of central nervous system regeneration are required and should provide molecular tools and resources. We focus here on the tunicate Ciona intestinalis, which has the capacity to regenerate nerves and a complete adult central nervous system, a capacity unusual in the chordate phylum. We investigated the timing and sequence of events during nervous system regeneration in this organism. Methodology/Principal Findings: We developed techniques for reproducible ablations and for imaging live cellular events in tissue explants. Based on live observations of more than 100 regenerating animals, we subdivided the regeneration process into four stages. Regeneration was functional, as shown by the sequential recovery of reflexes that established new criteria for defining regeneration rates. We used transgenic animals and labeled nucleotide analogs to describe in detail the early cellular events at the tip of the regenerating nerves and the first appearance of the new adult ganglion anlage. Conclusions/Significance: The rate of regeneration was found to be negatively correlated with adult size. New neural structures were derived from the anterior and posterior nerve endings. A blastemal structure was implicated in the formation of new neural cells. This work demonstrates that Ciona intestinalis is as a useful system for studies on regeneration of the brain, brain-associated organs and nerves. Citation: Dahlberg C, Auger H, Dupont S, Sasakura Y, Thorndyke M, et al. (2009) Refining the Ciona intestinalis Model of Central Nervous System Regeneration. PLoS ONE 4(2): e4458. doi:10.1371/journal.pone.0004458 Editor: Hernan Lopez-Schier, Centre de Regulacio Genomica, Spain Received August 5, 2008 Accepted November 28, 2008 Published February 12, 2009 Copyright: �� 2009 Dahlberg et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: The work carried out at Gif was supported by INRA, CNRS, the French GIS Institut de la Genomique �� Marine, the Plurigenes STREP project LSHG-CT-2005- 018673, the Marine Genomics Network of Excellence (EU-FP6 contract no. GOCE-CT-2004-505403). H. Auger held a PhD fellowship from MRT and ARC and a fellowship from the Marine Genomics Network for Women in Science Gender Action Plan. Y. Sasakura was supported by Grants-in-Aid for Scientific Research for a Priority Area from MEXT (17018018) and the NIG Cooperative Research Program (2008-B02). The E15 transgenic line was provided by the National Bioresource Project (07128019 and 07128021). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: carl.dahlberg@marecol.gu.se (CD) auger@inaf.cnrs-gif.fr (HA) . These authors contributed equally to this work. Introduction Few model systems are currently available for studies of central nervous system (CNS) regeneration. Retina regeneration and partial brain ablations in teleostean fish are two situations in which defined parts of the central nervous system reform after removal [1,2]. The Teleostean CNS also maintains numerous neurogenic zones in adulthood that generate new neurons throughout life [3��� 5]. Urodeles and frogs can regenerate the lens and spinal cord at tadpole stages. However, this capacity is lost in adults, and this loss is correlated with a decrease in neurogenesis [4]. The well studied organisms Caenorhabditis elegans and Drosophila melanogaster do not regenerate nervous system structures or display adult neurogenesis. However, some insects with longer lifespans, such as crickets, have centers of adult neurogenesis [6]. Planarians can regenerate almost complete nervous systems from multipotent neoblasts [7,8]. Finally, it has been shown that some colonial tunicates can regenerate their whole body from multipotent cells [9,10]. Ciona intestinalis (hereafter referred to as Ciona) is a tunicate deuterostome, and is thus a member of the clade corresponding to the closest extant relatives of vertebrates [11,12]. Its simplicity and chordate characters have made Ciona one of the most widely studied invertebrate deuterostomes, particularly from a develop- mental point of view (http://genome.jgi-psf.org/Cioin2/Cioin2. home.html http://hoya.zool.kyoto-u.ac.jp/cgi-bin/gbrowse/ci http://crfb.univ-mrs.fr/aniseed/index.php). Following the sequencing of two Ciona genomes, facilitating the characterization of regulatory elements, and improvements in methods for raising tunicates in the laboratory, many lines of transgenic animals expressing fluorescent markers in a wide array of patterns under the control of endogenous or exogenous promoters have been generated and are available [13,14]. The tunicate nervous system comprises the cerebral ganglion, the nerves of the body wall, the visceral nerve, the dorsal strand plexus and sensory structures connected to the siphons [15,16]. The neural gland, which connects to the pharyngeal lumen via the ciliated duct and the ciliated funnel (partially filled by the dorsal tubercle) is located close to the ganglion. Together, these organs form the neural complex (Fig. 1). It has recently been suggested that the neural gland and ciliated duct may be homologous to the circumventricular PLoS ONE | www.plosone.org 1 February 2009 | Volume 4 | Issue 2 | e4458

organs (e.g. the choroid plexus) of vertebrates, with a putative role in controlling the homeostasis of the fluid surrounding the neural complex [17,18] The cerebral ganglion includes most of the neural cell bodies of the central nervous system (CNS). Nerves leaving from the ganglion innervate the siphons, body wall and caudal viscera. Following ablation of the entire neural complex, animals appear normal and continue to filter water and to feed. However, this ablation has some detectable effects, such as altered behavioral responses to stimulations. Squirting, consisting of retraction of the whole body when one of the siphon tips is touched or the crossed siphon reflexes when the tentacles are touched are among the responses that are impaired (reviewed in [19,20,21]). CNS regeneration in Ciona has been studied in detail since its discovery by Schultze (1899). Following complete ablation of the neural complex, including part of the pharynx and the body wall, the neural complex regenerates completely within about a month [22]. Bollner et al. have shown that the expression patterns of certain transmitters (i.e. GNRH, 5-HT, SP) are recovered in the new cerebral ganglion, which, however, remains smaller than the original [23,24]. The rate of regeneration in an animal depends on many factors, including temperature, food, stress, and the environment [25���27]. Some studies in echinoderms have indicated that the proportion of tissue lost is also an important regulator of the rate of regeneration [28]. It has been shown in vertebrates that young individuals have higher regenerative capacities [29,30]. However, no systematic investigation has addressed the correlation between size and regeneration in ascidians. We analyzed the regenerative process with a high level of temporal, morphological and functional resolution, by live imaging and functional analysis of the same animals over the course of one month of regeneration. We also compared the rates of regeneration in regenerating animals of different sizes. We used transgenic animals to achieve a higher cellular resolution of the regenerating nerves. This made it possible to show that proliferation occurred at the tip of the nerves during regeneration, consistent with each nerve having a regenerative blastema. Results Wildtype regenerative stages We divided regeneration into stages, to obtain a clearer view of the complex temporal and spatial changes occurring during this process. Regeneration has previously been divided into stages on a purely temporal basis (i.e. first week, second week, etc. [22,31]). However, regeneration speed varies considerably and is particu- larly sensitive to rearing conditions (e.g. temperature), resulting in imprecise staging. We identified and rigorously defined four different stages of morphological regeneration: healing (I), nerve merging (II), structural regeneration (III) and functional regeneration (IV). Figure 2 illustrates this generalized regenerative process, based on the analysis of more than 100 regenerating neural complexes. All images were obtained after carefully opening the tunic and photographing the regenerating neural complex from the outside through the body wall. A description of the events taking place during each stage is provided in Table 1 Characters and Stages. Stage I ������healing������. This stage was characterized by the closing of the ablation wound by the tunic. A thicker transient tissue was initially formed. Complete tunic healing then occurred, with the new tunic becoming indistinguishable from the normal tunic (Fig. 2). Stage II ������nerve merging������. Soon after ablation, the sectioned nerves began to extend towards the wounded area. Adjacent nerve ends merged into networks with a higher density of regeneration foci in posterior than in anterior areas. The ends of the nerve fibers then thickened. We defined stage II ������nerve merging������ as the time at which the main branches of each of the four paired nerves could be seen to have merged in the regenerating area (Fig. 2). The healing of the pharynx took place during this stage and may be defined as a process in which the edges of the pharyngeal wall around the hole and epithelium resulting from the ablation procedure drawn together to close the wound. We defined ������pharynx healing������ as the stage at which the pharynx wound was closed, rather than the stage at which all muscles were repaired and in place. The timing of this process differed between individual animals. Stage III ������structural regeneration������. This stage was defined by the completion of four characters, all pertaining to neural structures: (1) The anterior nerves grew and sent out fine fibers in posterior and lateral directions. These fibers frequently formed a loose net-like structure of connecting lateral nerve endings (referred to here as the lateral net). More nerves extended in a posterior direction from the anterior network. Figure 1. Ciona neural complex anatomy. A Neural complex (NC) of a wild-type animal, located between the oral siphon (OrS) and the atrial siphon (AtS). B Higher magnification of the neural complex (NC) composed of the neural ganglion (GG), the neural gland (NG) and the dorsal tubercle (DT). C Nomenclature used to name the nerves emerging from the ganglion. Peripharyngeal band (PPB), left anterior nerve (LA), right anterior nerve (RA), left posterior nerve (LP), right posterior nerve (RP), dorsal strand plexus (DSP). doi:10.1371/journal.pone.0004458.g001 Neural Regeneration in Ciona PLoS ONE | www.plosone.org 2 February 2009 | Volume 4 | Issue 2 | e4458