View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Dispatch R593

action-specific reward values in the 13. Rabin, M. (1993). Incorporating Department of Psychological and Brain striatum. Science 310, 1337–1340. Fairness into Game Theory and Sciences and Krieger Mind/Brain 11. Matsumoto, K., Suzuki, W., and Economics. Am. Econ. Rev. 83, Institute, Johns Hopkins University, Tanaka, K. (2003). Neuronal correlates of 1281–1302. goal-based motor selection in the 14. de Quervain, D.J., Fischbacher, U., 3400 North Charles Street, Baltimore, prefrontal cortex. Science 301, Treyer, V., Schellhammer, M., Maryland 21218, USA. 229–232. Schnyder, U., Buck, A., and Fehr, E. E-mail: [email protected] 12. Camerer, C.F. (2003). Behavioral Game (2004). The neural basis of altruistic Theory (Princeton: Princeton University punishment. Science 305, Press). 1254–1258. DOI: 10.1016/j.cub.2006.07.005

Phylogeny: The Continuing chaetognaths, although it bears some resemblance to the nervous Classificatory Conundrum of systems of nematodes, kinorhynchs and priapulids [1]. The Chaetognaths age of the phylum is borne out by fossils from the Cambrian Burgess Shale which show a remarkable The phylogenetic conundrum posed by the Chaetognatha, a cryptic resemblance to modern phylum consisting largely of planktonic predators, is the subject of two chaetognaths [3] (Figure 2), short papers in this issue of Current Biology. These analyses go some suggesting chaetognaths have way towards defining the phylogenetic position of the chaetognaths, existed without major which possess features apparently spanning the protostome/ morphological change for at least deuterostome divide. 500 million years. Chaetognaths are unusual among marine Eldon E. Ball1,* and David J. Miller2 of the chaetognaths (Figure 1A,B) invertebrates in that they are direct is relatively constant across the developers, hatching from the egg It would be difficult to invent a more phylum, although details may vary essentially as small adults rather bizarre group of creatures than the considerably. The structure of the than as a distinct larval stage. chaetognaths or arrow worms, nervous system reveals little about Chaetognaths appear to rely (literally ‘bristle jaws’). The the evolutionary affinities of mainly on vibrations for information chaetognaths are an ancient lineage of invertebrates that shares Figure 1. Chaetognath mor- A BC phology. some characteristics with just Grasping (A) The deep water chaetog- spines about every other major Ciliate corona invertebrate phylum and has nath Eukrohnia sp. from the Sargasso Sea (Photo Collarette consequently puzzled taxonomists courtesy Russ Hopcroft/ Intestine ever since its original description in UAF/NOAA/CoML). In con- 1769. Darwin described trast to most shallow water forms, this species is chaetognaths as ‘‘remarkable for Hair fans pigmented, making the in- the obscurity of their affinities’’ and Muscles they have puzzled a succession of ternal organs easily visible. (B) This labelled drawing of Lateral eminent zoologists ever since [1]. the pelagic chaetognath, fin Though unfamiliar to most Heterokrohnia, provides a good introduction to the biologists, chaetognaths are Ovary typically the most abundant basic features of chaetog- Anus planktonic predators, sometimes nath anatomy. Some au- Posterior septum accounting for more than 10% of thors divide the body into head and trunk, while others Testis zooplankton biomass and being add a tail, dividing the trunk outnumbered only by their major at the posterior septum. Muscles prey, the copepods. Approximately There are three body cavi- Median 20 genera have been described, ties, which, together with mesentery containing more than 100 the cuticle and the longitudi- Caudal species ([2], see also: http:// nal muscles, form a hydro- static skeleton. One or two fin academic.evergreen.edu/t/ pairs of longitudinal fins pro- Current Biology thuesene/chaetognaths/ vide stabilisation, while the chaetognaths.htm). flattened caudal fin provides locomotory power. There is no circulatory system and no Although most chaetognaths specialised excretory system. The anterior grasping spines used in prey capture are are only a few mm in size and shown spread, although they are normally folded under a hood, which presumably planktonic, there are some inshore serves the dual purpose of protecting them and streamlining the head during swimming. Eggs are released into the thoracic coelomic space and sperm into the coelomic space of benthic and deep-water species the tail. Modified from [20], with permission. (C) Drawing of Sagitta hispida showing the that may be much larger, up to 12 closely spaced hair fans which presumably provide the sensory input required for such cm in length. The basic body plan complex behaviours as mating and prey capture. Modified with permission from [4]. Current Biology Vol 16 No 15 R594

Grasping Figure 2. Fossil chaetog- months in the upper 100 m, spines naths. descending to the 100–250 m The assumption that mod- zone during the winter. Transverse ern chaetognaths epitomize muscles All chaetognath species that an ancient and successful Intestine anatomical design, which have been studied are has remained unchanged hermaphrodites with testes in the Ventral ganglion since the Cambrian, is con- tail and ovaries in the posterior firmed by beautiful fossils trunk, the latter maturing only after Ovaries from the Burgess Shale, the tail coeloms are filled with Anus such as this Oesia disjuncta, sperm. Self-fertilization has been in which many modern Lateral described in Sagitta, but it is fins anatomical features have been tentatively identified unclear whether this occurs in B Seminal (A–C). Modified with permis- nature. Mating specimens of vesicle sion from [3]. Spadella cephaloptera become Caudal A fin aligned head to tail in parallel and C simultaneously exchange sperm Current Biology packets [9], although one-way transfer has also been reported. about their environment and they or glass probe vibrating at an The most amazing mating can hunt in darkness. Their appropriate frequency and behaviour, starting with a dance vibration sense is mediated amplitude with the same ferocity as and ending with a leap to through fans of mechanoreceptive it will attack a copepod [6]. deposit the sperm bundle or hairs [4] (Figure 1C) which Chaetognaths swallow their prey spermatophore, has been apparently play critical roles in all whole and will apparently eat described for Paraspadella gotoi behaviours. In addition to these anything that can be [10] (pictured at http:// hairs, chaetognaths have two eyes. accommodated in their mouth. As www.uibk.ac.at/zoology/ultra/ In some genera the eyes are the chaetognath strikes its prey, evol_ecol/former_research/ inverted, with the photoreceptor the hood over the grasping spines paraspadella/mating/ cells pointing inward and is withdrawn, allowing them to fly mating_behaviour.html). This embedded in a single large outward and form a net that complex behaviour is pigment cell while in others the surrounds the prey and brings it probably mediated by the eyes are everted with the into contact with the formidable mechanoreceptive hair fans [4]. photoreceptors pointing outward. teeth around the mouth. As if these Probably the strangest The eyes have relatively few impressive weapons weren’t chaetognath of all is a deep-sea receptor units and lack lenses, so enough, at least some form, Bathybelos typhlops, they apparently are not chaetognaths appear to use the described originally from a single image-forming. However, two potent neurotoxin tetrodotoxin to specimen from a depth of 2,500 m features that appear to be unique stun potential prey [7]. The exact in the Gulf of Mexico. This to chaetognath eyes are the source of the venom was not specimen showed many unusual conical body, for which some identified but it is most likely features including rayless lateral have suggested an optical role, produced by symbiotic bacteria fins, an elongated head and and the unusual lamellar and injected from one of the a dorsal nervous system, in organization of the distal segment structures surrounding the contrast to the usual ventral one. of at least some species [5].In mouth. Bathybelos is apparently addition, their photoreceptor cell is Chaetognaths seem to spend a bottom-living form, living with clearly ciliary, in contrast to the most of their time gliding on their the ventral side pressed into rhabdomeric photoreceptors fins, until eventually they begin to bottom sediments containing very found in many invertebrates. sink. Sinking is counteracted by low oxygen [11]. Chaetognaths Based on this eye structure, swimming with a series of short, lack a circulatory system, so their chaetognaths should have an rapid darts, followed by another tissues are dependent on opsin characteristic of ciliary type glide. Some species show diurnal diffusion for oxygen. Under such eyes, thus adding to the other vertical migrations, tending to conditions, there would molecular evidence against maintain themselves in a constant presumably be selective pressure ecdysozoan affinities. twilight zone. There is also a for the normally ventral nervous tendency for larger chaetognaths system to move dorsally. Behaviour – Feeding, Sexual to occur at greater depths. More specimens of benthic and Other Weirdness According to Marshall [8], Sagitta deep-water chaetognaths are Chaetognaths are remarkable gazellae in the Southern Ocean eagerly awaited. killers that appear to hunt primarily sink deeper in the water column by sensing vibration using the as they mature, releasing their Where Do Chaetognaths Fit in hair fans (http://www.cc.mie-u.ac. eggs at about 1000 m. The eggs the Scheme of Living Things? jp/wsagitta/lab.html). The are buoyant and float upward, Ever since their discovery, the chaetognath Spadella hatching at about 250 m. The phylogenetic position of cephaloptera will attack a metal hatchlings spend the summer chaetognaths has posed Dispatch R595 a conundrum. In particular, it has gene set. The fact that ‘blind’ particularly of benthic deep-water never been clear to which of the analyses based on many other species, are eagerly awaited. It is two major subdivisions of the genes are not consistent with the humbling to note that in spite of all animal kingdom they belong: Are new phylogeny has led to its of the recent molecular progress chaetognaths protostomes or rejection by some authors. we are still not far beyond the deuterostomes? Two short papers As there appear to be some conclusion of Libbie Hyman [1] [12,13] appearing in this issue of characteristic differences in the who concluded after a life spent Current Biology summarise some Hox clusters of lophotrochozoans studying invertebrate diversity, of the reasons for this obscurity and ecdysozoans this may be ‘‘It seems probable that the and shed some light on a fruitful and informative area for chaetognaths should be regarded chaetognath evolutionary affinities. future work. Le Parco’s group has as having diverged at an early stage By reporting moderate sized EST done some preliminary work [15] from the primitive ancestor of the datasets for Spadella cephaloptera on the Spadella cephaloptera Bilateria’’. [12] and Flaccisagittaenflata [13] Hox genes, but much more and applying some innovative detailed analyses are clearly References methods of analysis, these papers warranted. 1. Hyman, L.H. (1959). Phylum significantly advance the field. The Gene expression patterns could Chaetognatha. In The Invertebrata: Smaller Coelomate Groups, Vol. V analyses in one of the papers [12] be another useful type of (New York: McGraw-Hill), pp. 1–71. were based on 78 ribosomal information for clarifying the 2. Bieri, R. (1991). Systematics of the protein sequences, using the novel evolutionary relationships of Chaetognatha. In The Biology of Chaetognaths, Q. Bone, H. Kapp, and approach of making comparisons chaetognaths. Unfortunately, very A.C. Pierrot-Bults, eds. (Oxford: Oxford with composite sequence few chaetognath gene expression University Press), pp. 122–136. 3. Szaniawski, H. (2005). Cambrian collections consisting of the patterns have been reported to chaetognaths recognized in Burgess closest matching sequences date, and the take-home message Shale fossils. Acta Palaeontologica across each of 14 animal phyla. The from those that have, e.g. Polonica 50, 1–8. 4. Feigenbaum, D.L. (1978). Hair fan patterns second paper [13] includes Brachyury [16], is that the in the Chaetognatha. Can. J. Zool. 56, analyses based on 72 genes chaetognaths are mainstream 536–546. 5. Goto, T., Takasu, N., and Yoshida, M. known from the previous work of bilaterians. Hopefully the (1984). A unique photoreceptive structure Herve Philippe [14] to be availability of the new EST datasets in the arrowworms Sagitta crassa and appropriate for deep level will lead to many more expression Spadella schizoptera (Chaetognatha). Cell Tissue Res. 235, 471–478. phylogenetics. patterns being determined, which 6. Horridge, G.A., and Boulton, P.S. (1967). Both analyses clearly also could advance our Prey detection by Chaetognatha via a vibration sense. Proc. Roy. Soc. B 168, demonstrate that chaetognaths are understanding of several 413–419. protostomes, despite showing important aspects of chaetognath 7. Thuesen, E.V. (1991). The tetrodotoxin deuterostome-like features of early biology. venom of chaetognaths. In The Biology of Chaetognaths, Q. Bone, H. Kapp, and embryonic development. However, We have, as yet, a quite limited A.C. Pierrot-Bults, eds. (Oxford: Oxford their position within the view of the biology of the phylum University Press), pp. 55–60. 8. Marshall, N.B. (1979). Developments in protostomes remains unclear and Chaetognatha as a whole and Deep-Sea Biology (Poole, Dorset.: controversial. In part, this reflects much of what we know, particularly Blandford Press). the debate still raging over concerning behaviour, is derived 9. Ghirardelli, E. (1968). Some aspects of the biology of Chaetognaths. Adv. Marine acceptance of the ‘new’ from a handful of genera. The only Biol. 6, 271–376. phylogeny — the splitting of the extensive molecular phylogenetic 10. Goto, T., and Yoshida, M. (1985). The mating sequence of the benthic protostomes into Ecdysozoa analyses conducted to date have arrowworm Spadella schizoptera. Biol. and Lophotrochozoa. It is hardly been based on rDNA sequences Bull. 169, 328–333. surprising that some of the ‘minor’ [17] and are complicated by the 11. Bieri, R., and Thuesen, E.V. (1990). The strange worm Bathybelos. Am. Sci. 78, phyla, such as the chaetognaths, presence of paralogous rDNA 542–549. cannot yet be accurately placed clusters. However, the 12. Marle´ taz, F., Martin, E., Perez, Y., Papillon, D., Caubit, X., Fasano, L., when there is still disagreement mitochondrial genomes of Lowe, C.J., Freeman, B., Dossat, C., over the broad framework, with Spadella cephaloptera [18] and Wincker, P., et al. (2006). Chaetognath many authors remaining Paraspadella gotoi [19] are highly phylogenomics: A protostome with deuterostome-like development. Curr. unconvinced that the ‘old’ diverged in spite of having Biol. 16, R577–R578. phylogeny needs revision. The new common membership of a single 13. Matus, D.Q., Copley, R.R., Dunn, C.W., Hejnol, A., Eccleston, H., Halanych, K.M., phylogeny recognizes a clade of family in one of the two recognised Martindale, M.Q., and Telford, M.J. (2006). moulting animals (the Ecdysozoa) orders (and one of four clades Broad taxon and gene sampling indicate within the protostomes, uniting based on rDNA sequence data; that chaetognaths are protostomes. Curr. Biol. 16, R575–R576. some phyla previously considered [17]). Given their major differences 14. Philippe, H., Lartillot, N., and to be only very distantly related in mitochondrial gene Brinkmann, H. (2005). Multigene analyses of bilaterian animals corroborate the (e.g. nematodes and arthropods), arrangements, despite similar monophyly of Ecdysozoa, and also splits some traditional overall morphology, it seems likely Lophotrochozoa and Protostomia. Mol. groupings (e.g. arthropods and that the chaetognaths are a Biol. Evol. 22, 1246–1253. 15. Papillon, D., Perez, Y., Fasano, L., annelids). At the core of the genetically diverse phylum. More Le Parco, Y., and Caubit, X. (2003). disagreement is the fact that the gene expression data, more Hox gene survey in the chaetognath Spadella cephaloptera: evolutionary new phylogeny is revealed only sequences and anatomical implications. Dev. Genes Evol. 213, with a specific and rather limited analyses of more specimens, 142–148. Current Biology Vol 16 No 15 R596

16. Takada, N., Goto, T., and Satoh, N. (2002). mitochondrial genome. Mol. Biol. Evol. 21, 1Centre for Molecular Genetics of Expression pattern of the Brachyury gene 2122–2129. Development & Research School of in the arrow worm paraspadella 19. Helfenbein, K.G., Fourcade, H.M., Biological Sciences, Australian National gotoi (chaetognatha). Genesis 32, Vanjani, R.G., and Boore, J.L. (2004). 240–245. The mitochondrial genome of University, P.O. Box 475, Canberra, 2 17. Papillon, D., Perez, Y., Caubit, X., and Paraspadella gotoi is highly reduced A.C.T. 2601, Australia. Comparative Le Parco, Y. (2006). Systematics of and reveals that chaetognaths are Genomics Centre, and Centre of Chaetognatha under the light of molecular a sister group to protostomes. Proc. Excellence for Coral Reef Studies, data, using duplicated ribosomal 18S DNA Natl. Acad. Sci. USA 101, James Cook University, Townsville, Qld. sequences. Mol. Phylogenet. Evol. 38, 10639–10643. 621–634. 20. Shinn, G.L. (1997). Chaetognatha. In 4811, Australia. 18. Papillon, D., Perez, Y., Caubit, X., and Microscopic Anatomy of Invertebrates, *E-mail: [email protected] Le Parco, Y. (2004). Identification of Vol. 15, F.W. Harrison and E.E. Ruppert, chaetognaths as protostomes is eds. (New York: Wiley-Liss), supported by the analysis of their pp. 103–220. DOI: 10.1016/j.cub.2006.07.006

Declarative Memory: Sleep coherent enough to be accessed during sleep. Protects New Memories from A new study by Ellenbogen et al. [9], published recently in Current Interference Biology, directly explores the role of sleep in protecting new memories from interference. To accomplish Interference is one of the most fundamental phenomena in memory this goal, the authors used the research: acquiring new memories causes forgetting of other, related AB–AC word pair learning memories. A new study shows that sleep, interposed between learning paradigm [10]. In this paradigm, episodes, can mitigate the extent to which new (post-sleep) learning subjects learn a set of word pairs, interferes with recall of previously acquired knowledge. such as shoe–banana; call this the A–B set. Next, subjects learn a new Kenneth A. Norman existing memories [4–6], it is also set of word pairs, the A–C set, clear that the interference effects where the ‘A’ words from before are How does the brain protect observed in behavioral studies are paired with new words, for example existing memories from being much smaller than those predicted shoe–stapler. Later, subjects are destroyed or distorted by new by the simple neural network given the ‘A’ words, such as shoe, learning? This puzzle, labeled the models mentioned above. and asked to recall both words that stability-plasticity dilemma by [1], This discrepancy has led were paired with each ‘A’ word. has been at the forefront of researchers to puzzle over what Previous studies using this memory research for several kinds of mechanisms (not present paradigm have found that learning decades. In the 1970s and 1980s, in these simple models) the brain the A–C items impairs subsequent when researchers started to build might use to protect stored recall of the A–B items [4,10]. explicit computer models of how memories. Computational To address the role of sleep in the brain stores memories, they neuroscientists have come up with mitigating interference, Ellenbogen found that these neural network several different ideas about how et al. [9] set up their study such that models of memory tend to show the brain avoids catastrophic one group of subjects slept catastrophic levels interference (for example, see between learning the A–B pairs and of memory interference [2,3]: [1,7,8]). One of the most intriguing learning the A–C pairs, and another implanting new memories in the of these ideas is that sleep might group of subjects did not sleep network causes distressingly fast play a role in reinforcing and during that interval; subjects were forgetting of previously stored protecting existing knowledge. The tested on A–B and A–C pairs after memories. This problem occurs gist of the idea is that sleep learning the A–C pairs. The because memories are stored in provides an opportunity for the experiment also included ‘no an overlapping fashion in these brain to ‘think about what it already interference’ control conditions models, as they are in the brain. knows’, strengthening and that were identical to the A given synapse might participate potentially also refining existing aforementioned conditions, except in storing multiple related memories so they are less likely subjects did not learn any A–C memories (for example, memory to be disrupted by new learning. items. The authors found that the for your breakfast today versus Some researchers [8] also have negative effect of A–C learning yesterday), so adjusting the argued that learning during sleep on A–B recall was much smaller strength of that synapse to can serve to repair damage to in the sleep condition than in the strengthen one memory might existing memories; just as a no-sleep condition. Thus, it distort another memory that relies building with a crumbling appears that something happens on that synapse. While numerous infrastructure can be repaired (if it during sleep — when interposed psychological studies have is accessed before it collapses), a between A–B and A–C learning — demonstrated that new learning damaged memory can be repaired that makes the A–B trace less does, in fact, interfere with recall of so long as the memory is still vulnerable to interference.