JOURNAL OF MORPHOLOGY 275:269–294 (2014)
The Ventral Nerve Cord in Cephalocarida (Crustacea): New Insights into the Ground Pattern of Tetraconata Martin E.J. Stegner,* Georg Brenneis,# and Stefan Richter Universit€ at Rostock, Institut f€ ur Biowissenschaften, Allgemeine und Spezielle Zoologie, Universit€ atsplatz 2, 18055 Rostock, Mecklenburg-Vorpommern, Germany ABSTRACT Cephalocarida are Crustacea with many anatomical features that have been interpreted as plesiomorphic with respect to crustaceans or Tetraconata. While the ventral nerve cord (VNC) has been investigated in many other arthropods to address phylogenetic and evolutionary questions, the few studies that exist on the cephalocarid VNC date back 20 years, and data pertaining to neuroactive substances in particular are too sparse for comparison. We reinvestigated the VNC of adult Hutchinsoniella macracantha in detail, combining immunolabeling (tubulin, serotonin, RFamide, histamine) and nuclear stains with confocal laser microscopy, complemented by 3D-reconstructions based on serial semithin sections. The subesophageal ganglion in Cephalocarida comprises three segmental neuromeres (Md, Mx1, Mx2), while a separate ganglion occurs in all thoracic segments and abdominal segments 1–8. Abdominal segments 9 and 10 and the telson are free of ganglia. The maxillar neuromere and the thoracic ganglia correspond closely in their limb innervation pattern, their pattern of mostly four segmental commissures and in displaying up to six individually identified serotonin-like immunoreactive neurons per body side, which exceeds the number found in most other tetraconates. Only two commissures and two serotonin-like immunoreactive neurons per side are present in abdominal ganglia. The stomatogastric nervous system in H. macracantha corresponds to that in other crustaceans and includes, among other structures, a pair of lateral neurite bundles. These innervate the gut as well as various trunk muscles and are, uniquely, linked to the unpaired median neurite bundle. We propose that most features of the cephalocarid ventral nerve cord (VNC) are plesiomorphic with respect to the tetraconate ground pattern. Further, we suggest that this ground pattern includes more serotonin-like neurons than hitherto assumed, and argue that a sister-group relationship between Cephalocarida and Remipedia, as favored by recent molecular analyses, finds no neuroanatomical support. J. Morphol. 275:269– 294, 2014. VC 2013 Wiley Periodicals, Inc. KEY WORDS: entomostracan abdomen; mandible; maxillula; maxilla; Mandibulata; serotonergic neurons; stomatogastric nervous system; Xenocarida
INTRODUCTION General Aspects Most external anatomical features of Cephalocarida have been interpreted as being plesiomorphic with respect to the ground pattern of Crustacea, MandibuC 2013 WILEY PERIODICALS, INC. V
lata, or even Arthropoda (summarized by Olesen et al., 2011; for details, see, for example, Sanders, 1957, 1963; Hessler, 1964, 1992; Lauterbach, 1974, 1983, 1986; Walossek, 1993; Scholtz and Edgecombe, 2005). Our recent investigation into the cephalocarid brain showed that certain features in the cephalocarid olfactory system can also be traced back to the ur-mandibulate (Stegner and Richter, 2011). In contrast, other features suggested to belong to the tetraconate ground pattern (e.g., Harzsch, 2006) have been reduced in Cephalocarida. These include the optic lobes (Elofsson and Hessler, 1990; Stegner and Richter, 2011) and the central complex (Stegner and Richter, 2011; Stegner et al., in press). Since potential morphological synapomorphies between Cephalocarida and other tetraconates are scarce (see, e.g., Hessler, 1992; Jenner, 2010; Olesen et al., 2011), and molecular analyses reveal no definitive picture, the phylogenetic relationships between Cephalocarida and the rest of the Tetraconata remain a matter for debate (Koenemann et al., 2010; Regier et al., 2010; von Reumont et al., 2012; Oakley et al., 2013). In the last decade, phylogenetic and evolutionary questions have started to be addressed by looking
This article was published online on 4 November 2013. An error was subsequently identified. This notice is included in the online and print versions to indicate that both have been corrected 18 November 2013. Additional Supporting Information Figure S1 and Text S1 are found in the online version of this article. Contract grant sponsor: German Science Foundation (DFG) (RI 837/10-1,2). *Correspondence to: Martin E.J. Stegner; Institut f€ ur Biowissenschaften, Allgemeine und Spezielle Zoologie, Universit€ atsplatz 2, 18055 Rostock, Germany. E-mail: martin. [email protected]; Georg Brenneis, E-mail: georg.brenneis@ gmx.de # Present address: E-mail: [email protected] Received 2 May 2013; Revised 28 August 2013; Accepted 6 September 2013. Published online 4 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jmor.20213
270
M.E.J. STEGNER ET AL
at the arthropod VNC, and in particular, its serotonin-like immunoreactive (SL-ir) pattern (e.g., Harzsch and Waloszek, 2000; Harzsch, 2004, 2007). Our current knowledge of the cephalocarid VNC dates back to three previous studies. Elofsson and Hessler (1990) studied the species Hutchinsoniella macracantha using light and transmission electron microscopy. Although they focussed mainly on the brain, they also provided the first description of the VNC, including the pattern of segmental ganglia, commissures, and nerve roots therein. Hessler and Elofsson (1992) presented the first lateral drawing of the complete VNC, but did not provide further anatomical details. The only study so far to apply immunohistochemical techniques documented a few SL-ir and RFamide-like immunoreactive (RFL-ir) neurons in the cephalocarid VNC (Elofsson, 1992), but lacked the necessary detail to be incorporated into later comparative analyses (e.g., Harzsch and Waloszek, 2000, see below). Cephalocarid Abdomen The trunk of Cephalocarida and of other entomostracan crustaceans (i.e., non-malacostracans except Remipedia, see Walossek, 1993) is divided into a limb-bearing thorax and a limbless abdomen. The evolutionary origin of the abdomen has been tackled from different angles in the past. Authors have included in their definition of an “abdomen” not only the absence of segmental limbs but also Hox gene expression patterns (e.g., Schram and Koenemann, 2004), the nervous system (e.g., Deutsch, 2001), and muscles (Huys, 1991, but see Brenneis and Richter, 2010). In the present study, we aim to investigate to what extent segmental structures and patterns of the cephalocarid VNC are tagma-specific, to compare them between different tetraconates, and to detect potential evolutionary interdependences (i.e., coherences). Maxillar Appendage and Neuroanatomy Another topic addressed here is the morphology and evolution of the mandibulate head (traditionally termed “cephalon” in Crustacea) and its segments (see e.g., Lauterbach, 1980; Budd, 2001; Waloszek et al., 2005; Scholtz and Edgecombe, 2005, 2006; Giribet and Edgecombe, 2012; Richter et al., 2013). Although the maxillar segment (Mx2) forms part of the cephalon in all recent mandibulates, the anatomy of maxillar appendages is morphologically disparate. The hypothesis that the maxilla specialized into a feeding appendage several times independently (e.g., Walossek and M€ uller, 1990) is supported by the thoracopod-like anatomy of the “maxilla” in Cephalocarida (Sanders, 1954, 1957). According to Scholtz and Edgecombe (2006, p. 409f.), the fusion of cephalic segments during mandibulate evolution, especially of cuticular tergites and pleurites, predated the specialization of limbs. We aim to investigate how the nervJournal of Morphology
ous system relates to this scenario, that is, to find out whether the cephalocarid maxillar neuromere shows any kind of “subesophageal” specialization or whether it is more thoracic-like. Serotonin-Like Immunoreactivity and Phylogenetics Harzsch and Waloszek (2000) first noted the phylogenetic relevance of segmental individually identified SL-ir neurons in the arthropod VNC. Thanks to a growing corpus of comparable data (e.g., see Harzsch, 2002, 2003, 2004; Harzsch et al., 2005), phylogenetic inferences could soon be drawn on a larger scale. Harzsch et al. (2005) suggested that the tetraconate ground pattern featured four individually identified SL-ir bipolar neurons per hemiganglion, from which they deduced the diverging (often more simple) patterns in recent hexapods and crustaceans. The evolutionary scenarios suggested by Harzsch et al. (2005) are reconsidered here in the light of new data on Remipedia (Stemme et al., 2010, 2013) and Cephalocarida (this study). The current hypothesis of a sister-group relationship between Remipedia and Cephalocarida (Koenemann et al., 2010; Regier et al., 2010) also calls for a more detailed understanding of the cephalocarid SL-ir pattern and a detailed comparison to other tetraconates and myriapods. In order to address these phylogenetic and evolutionary problems, we reinvestigated the VNC of H. macracantha using present-day methods, namely a combination of fluorescent immunolabeling, confocal laser scanning microscopy and computer-aided 3D-reconstruction. As a first step in this study, we refine and add to earlier descriptions of the cephalocarid VNC (Elofsson and Hessler, 1990; Elofsson, 1992; Hessler and Elofsson, 1992) and better correlate nervous structures with the serotonin-like, RFamide-like, and histamine-like immunoreactive patterns. As a second step, our data are compared to that on other arthropods. Homology hypotheses are formulated and interpreted in a phylogenetic and evolutionary context. MATERIAL AND METHODS Collection, Fixation, and Storage Adults of the cephalocarid species Hutchinsoniella macracantha were collected in the daytime during the summer months from 2007 to 2011 at different spots in Buzzards Bay, MA. Samples of benthic mud were collected by boat from depths of 10–25 m using a Van Veen grab or via scuba diving. The animals were separated from the sieved samples using fine forceps or pipettes. Wildt et al. (2004) showed that the expression level of serotonin in lobsters underlies circadian rhythm, showing a peak before dusk and a trough before dawn. This phenomenon does not explain the intraspecific variability observed in the SL-ir pattern of H. macracantha (this study), however, as all the animals were fixed at the same time before dusk at the end of our collecting days. Three different fixation methods were required for the procedures used in this study. 1) For semithin sectioning, one animal was fixed and stored in Bouin’s fixative (Mulisch and Welsch,
VENTRAL NERVE CORD IN CEPHALOCARIDA 2010), the salinity of which had been increased to 3% by adding NaCl. 2) For the immunolabeling of acetylated a-tubulin and serotonin-like and RFamide-like proteins, 18 animals were fixed with a 4% paraformaldehyde solution [16% PFA stem solution (Electron Microscopy Sciences, Hatfield, PA: @ CAS #30525-89-4) diluted in phosphate buffered saline (PBS) solution which was produced by diluting 10 x PBS stock solution in filtered water from Buzzards Bay] for 30 min to 1 h. Fixed specimens were transferred to 100% methanol for storage at 4 C. Serotonin-like immunoreactivity was investigated in eleven animals, FMRFamide-like immunoreactivity was investigated in seven animals. 3) For the immunolabeling of histamine-like proteins, four animals were fixed in a 4% carbodiimid solution [400 mg of carbodiimid (Sigma-Aldrich, St. Louis, MO: @ E1769) in 10 ml of filtered water from Buzzards Bay] for 24 h at 4 C. Afterward, specimens were postfixed with a 4% paraformaldehyde solution for 1 h and stored in PBS solution at 4 C (both solutions using filtered water from Buzzards Bay).
Pretreatment of Whole Mounts for Immunolabeling After a series of dilutions from their storage medium into PBS, specimens were exposed to short pulses (
The Ventral Nerve Cord in Cephalocarida (Crustacea): New Insights into the Ground Pattern of Tetraconata Martin E.J. Stegner,* Georg Brenneis,# and Stefan Richter Universit€ at Rostock, Institut f€ ur Biowissenschaften, Allgemeine und Spezielle Zoologie, Universit€ atsplatz 2, 18055 Rostock, Mecklenburg-Vorpommern, Germany ABSTRACT Cephalocarida are Crustacea with many anatomical features that have been interpreted as plesiomorphic with respect to crustaceans or Tetraconata. While the ventral nerve cord (VNC) has been investigated in many other arthropods to address phylogenetic and evolutionary questions, the few studies that exist on the cephalocarid VNC date back 20 years, and data pertaining to neuroactive substances in particular are too sparse for comparison. We reinvestigated the VNC of adult Hutchinsoniella macracantha in detail, combining immunolabeling (tubulin, serotonin, RFamide, histamine) and nuclear stains with confocal laser microscopy, complemented by 3D-reconstructions based on serial semithin sections. The subesophageal ganglion in Cephalocarida comprises three segmental neuromeres (Md, Mx1, Mx2), while a separate ganglion occurs in all thoracic segments and abdominal segments 1–8. Abdominal segments 9 and 10 and the telson are free of ganglia. The maxillar neuromere and the thoracic ganglia correspond closely in their limb innervation pattern, their pattern of mostly four segmental commissures and in displaying up to six individually identified serotonin-like immunoreactive neurons per body side, which exceeds the number found in most other tetraconates. Only two commissures and two serotonin-like immunoreactive neurons per side are present in abdominal ganglia. The stomatogastric nervous system in H. macracantha corresponds to that in other crustaceans and includes, among other structures, a pair of lateral neurite bundles. These innervate the gut as well as various trunk muscles and are, uniquely, linked to the unpaired median neurite bundle. We propose that most features of the cephalocarid ventral nerve cord (VNC) are plesiomorphic with respect to the tetraconate ground pattern. Further, we suggest that this ground pattern includes more serotonin-like neurons than hitherto assumed, and argue that a sister-group relationship between Cephalocarida and Remipedia, as favored by recent molecular analyses, finds no neuroanatomical support. J. Morphol. 275:269– 294, 2014. VC 2013 Wiley Periodicals, Inc. KEY WORDS: entomostracan abdomen; mandible; maxillula; maxilla; Mandibulata; serotonergic neurons; stomatogastric nervous system; Xenocarida
INTRODUCTION General Aspects Most external anatomical features of Cephalocarida have been interpreted as being plesiomorphic with respect to the ground pattern of Crustacea, MandibuC 2013 WILEY PERIODICALS, INC. V
lata, or even Arthropoda (summarized by Olesen et al., 2011; for details, see, for example, Sanders, 1957, 1963; Hessler, 1964, 1992; Lauterbach, 1974, 1983, 1986; Walossek, 1993; Scholtz and Edgecombe, 2005). Our recent investigation into the cephalocarid brain showed that certain features in the cephalocarid olfactory system can also be traced back to the ur-mandibulate (Stegner and Richter, 2011). In contrast, other features suggested to belong to the tetraconate ground pattern (e.g., Harzsch, 2006) have been reduced in Cephalocarida. These include the optic lobes (Elofsson and Hessler, 1990; Stegner and Richter, 2011) and the central complex (Stegner and Richter, 2011; Stegner et al., in press). Since potential morphological synapomorphies between Cephalocarida and other tetraconates are scarce (see, e.g., Hessler, 1992; Jenner, 2010; Olesen et al., 2011), and molecular analyses reveal no definitive picture, the phylogenetic relationships between Cephalocarida and the rest of the Tetraconata remain a matter for debate (Koenemann et al., 2010; Regier et al., 2010; von Reumont et al., 2012; Oakley et al., 2013). In the last decade, phylogenetic and evolutionary questions have started to be addressed by looking
This article was published online on 4 November 2013. An error was subsequently identified. This notice is included in the online and print versions to indicate that both have been corrected 18 November 2013. Additional Supporting Information Figure S1 and Text S1 are found in the online version of this article. Contract grant sponsor: German Science Foundation (DFG) (RI 837/10-1,2). *Correspondence to: Martin E.J. Stegner; Institut f€ ur Biowissenschaften, Allgemeine und Spezielle Zoologie, Universit€ atsplatz 2, 18055 Rostock, Germany. E-mail: martin. [email protected]; Georg Brenneis, E-mail: georg.brenneis@ gmx.de # Present address: E-mail: [email protected] Received 2 May 2013; Revised 28 August 2013; Accepted 6 September 2013. Published online 4 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jmor.20213
270
M.E.J. STEGNER ET AL
at the arthropod VNC, and in particular, its serotonin-like immunoreactive (SL-ir) pattern (e.g., Harzsch and Waloszek, 2000; Harzsch, 2004, 2007). Our current knowledge of the cephalocarid VNC dates back to three previous studies. Elofsson and Hessler (1990) studied the species Hutchinsoniella macracantha using light and transmission electron microscopy. Although they focussed mainly on the brain, they also provided the first description of the VNC, including the pattern of segmental ganglia, commissures, and nerve roots therein. Hessler and Elofsson (1992) presented the first lateral drawing of the complete VNC, but did not provide further anatomical details. The only study so far to apply immunohistochemical techniques documented a few SL-ir and RFamide-like immunoreactive (RFL-ir) neurons in the cephalocarid VNC (Elofsson, 1992), but lacked the necessary detail to be incorporated into later comparative analyses (e.g., Harzsch and Waloszek, 2000, see below). Cephalocarid Abdomen The trunk of Cephalocarida and of other entomostracan crustaceans (i.e., non-malacostracans except Remipedia, see Walossek, 1993) is divided into a limb-bearing thorax and a limbless abdomen. The evolutionary origin of the abdomen has been tackled from different angles in the past. Authors have included in their definition of an “abdomen” not only the absence of segmental limbs but also Hox gene expression patterns (e.g., Schram and Koenemann, 2004), the nervous system (e.g., Deutsch, 2001), and muscles (Huys, 1991, but see Brenneis and Richter, 2010). In the present study, we aim to investigate to what extent segmental structures and patterns of the cephalocarid VNC are tagma-specific, to compare them between different tetraconates, and to detect potential evolutionary interdependences (i.e., coherences). Maxillar Appendage and Neuroanatomy Another topic addressed here is the morphology and evolution of the mandibulate head (traditionally termed “cephalon” in Crustacea) and its segments (see e.g., Lauterbach, 1980; Budd, 2001; Waloszek et al., 2005; Scholtz and Edgecombe, 2005, 2006; Giribet and Edgecombe, 2012; Richter et al., 2013). Although the maxillar segment (Mx2) forms part of the cephalon in all recent mandibulates, the anatomy of maxillar appendages is morphologically disparate. The hypothesis that the maxilla specialized into a feeding appendage several times independently (e.g., Walossek and M€ uller, 1990) is supported by the thoracopod-like anatomy of the “maxilla” in Cephalocarida (Sanders, 1954, 1957). According to Scholtz and Edgecombe (2006, p. 409f.), the fusion of cephalic segments during mandibulate evolution, especially of cuticular tergites and pleurites, predated the specialization of limbs. We aim to investigate how the nervJournal of Morphology
ous system relates to this scenario, that is, to find out whether the cephalocarid maxillar neuromere shows any kind of “subesophageal” specialization or whether it is more thoracic-like. Serotonin-Like Immunoreactivity and Phylogenetics Harzsch and Waloszek (2000) first noted the phylogenetic relevance of segmental individually identified SL-ir neurons in the arthropod VNC. Thanks to a growing corpus of comparable data (e.g., see Harzsch, 2002, 2003, 2004; Harzsch et al., 2005), phylogenetic inferences could soon be drawn on a larger scale. Harzsch et al. (2005) suggested that the tetraconate ground pattern featured four individually identified SL-ir bipolar neurons per hemiganglion, from which they deduced the diverging (often more simple) patterns in recent hexapods and crustaceans. The evolutionary scenarios suggested by Harzsch et al. (2005) are reconsidered here in the light of new data on Remipedia (Stemme et al., 2010, 2013) and Cephalocarida (this study). The current hypothesis of a sister-group relationship between Remipedia and Cephalocarida (Koenemann et al., 2010; Regier et al., 2010) also calls for a more detailed understanding of the cephalocarid SL-ir pattern and a detailed comparison to other tetraconates and myriapods. In order to address these phylogenetic and evolutionary problems, we reinvestigated the VNC of H. macracantha using present-day methods, namely a combination of fluorescent immunolabeling, confocal laser scanning microscopy and computer-aided 3D-reconstruction. As a first step in this study, we refine and add to earlier descriptions of the cephalocarid VNC (Elofsson and Hessler, 1990; Elofsson, 1992; Hessler and Elofsson, 1992) and better correlate nervous structures with the serotonin-like, RFamide-like, and histamine-like immunoreactive patterns. As a second step, our data are compared to that on other arthropods. Homology hypotheses are formulated and interpreted in a phylogenetic and evolutionary context. MATERIAL AND METHODS Collection, Fixation, and Storage Adults of the cephalocarid species Hutchinsoniella macracantha were collected in the daytime during the summer months from 2007 to 2011 at different spots in Buzzards Bay, MA. Samples of benthic mud were collected by boat from depths of 10–25 m using a Van Veen grab or via scuba diving. The animals were separated from the sieved samples using fine forceps or pipettes. Wildt et al. (2004) showed that the expression level of serotonin in lobsters underlies circadian rhythm, showing a peak before dusk and a trough before dawn. This phenomenon does not explain the intraspecific variability observed in the SL-ir pattern of H. macracantha (this study), however, as all the animals were fixed at the same time before dusk at the end of our collecting days. Three different fixation methods were required for the procedures used in this study. 1) For semithin sectioning, one animal was fixed and stored in Bouin’s fixative (Mulisch and Welsch,
VENTRAL NERVE CORD IN CEPHALOCARIDA 2010), the salinity of which had been increased to 3% by adding NaCl. 2) For the immunolabeling of acetylated a-tubulin and serotonin-like and RFamide-like proteins, 18 animals were fixed with a 4% paraformaldehyde solution [16% PFA stem solution (Electron Microscopy Sciences, Hatfield, PA: @ CAS #30525-89-4) diluted in phosphate buffered saline (PBS) solution which was produced by diluting 10 x PBS stock solution in filtered water from Buzzards Bay] for 30 min to 1 h. Fixed specimens were transferred to 100% methanol for storage at 4 C. Serotonin-like immunoreactivity was investigated in eleven animals, FMRFamide-like immunoreactivity was investigated in seven animals. 3) For the immunolabeling of histamine-like proteins, four animals were fixed in a 4% carbodiimid solution [400 mg of carbodiimid (Sigma-Aldrich, St. Louis, MO: @ E1769) in 10 ml of filtered water from Buzzards Bay] for 24 h at 4 C. Afterward, specimens were postfixed with a 4% paraformaldehyde solution for 1 h and stored in PBS solution at 4 C (both solutions using filtered water from Buzzards Bay).
Pretreatment of Whole Mounts for Immunolabeling After a series of dilutions from their storage medium into PBS, specimens were exposed to short pulses (
