Integrative and Comparative Biology Advance Access published online on April 4, 2008
Integrative and Comparative Biology, doi:10.1093/icb/icn012
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Head organization and the head/trunk relationship in protochordates: problems and prospects
Biology Department, University of Victoria, Victoria, BC, Canada, V8W-3N5
Correspondence: 1E-mail: lacalli{at}uvic.ca
 |
SYNOPSIS |
|---|
TOP
SYNOPSIS
Reconstructing the evolutionary...The fossil record has been an invaluable aid for reconstructing the major events of vertebrate evolution. There is no comparable record for protochordates, however, which severely limits our knowledge of their ancestral morphology, habits, and mode of life. The alternative is inference based on an interpretation of living protochordates but this is fraught with problems, not least being our own biases of what we think an ancestral chordate ought to look like. Relevant to the present symposium is the problem of head/trunk relationships and whether or not the myotomes of the trunk originally extended into the head in vertebrates. I will review what is currently known of patterns of innervation in tunicates and amphioxus in relation to Romer's somaticovisceral concept of the vertebrate body to show how little progress has been made in resolving this problem. There are, in contrast, surprisingly good prospects for solving some other puzzles concerning chordate origins. Dorsoventral inversion provides a good example. A consensus is now emerging, based largely on molecular data from hemichordates that casts new light on the asymmetry of the head in amphioxus. Specifically, the morphogenetic growth process that reestablishes symmetry in late-stage larvae can now be seen, at least in part, as a recapitulation of past evolutionary events, and this has important implications for the origin and basic organization of the brain.
|
Reconstructing the evolutionary past |
|---|
Protochordates are crucially important to comparative zoologists for the clues they provide concerning the course of early chordate evolution and the nature of ancestral vertebrates. Only two taxa have survived to modern times, cephalochordates (amphioxus) and urochordates (tunicates). Past uncertainty about which of these is closer to vertebrates seems to be resolved in favor of tunicates, which are now considered to be the sister group of vertebrates (Holland 2007
). Establishing the correct branching order among these three lineages is a significant step forward, but there is a still larger problem. This is to determine how, and how much, tunicates and amphioxus have changed in >500 million years since they first appeared. We otherwise have no way of knowing how much reliance can be placed on either group as models for ancestral chordates. Fossils would help enormously, but the fossil record for protochordates is fragmentary and difficult to interpret (see below), which severely complicates matters.
Consider tunicates: at one time they, or more specifically their larvae, were accepted as good stand-ins for the first motile chordates, which were supposed to have originated by paedomorphosis from the larvae of a sedentary ancestor. This scenario was popularized by Romer (1967), in conjunction with his concept of vertebrates as "dual" animals, in which visceral and somatic functions are segregated to the head and trunk, respectively (Romer 1972). The basic idea predates Romer's writings; however, the earliest complete exposition being by McMurrich (1912), for which information I am indebted to Nick Holland. Modern molecular studies have more recently tended to push tunicates from center stage, and it is now difficult to see either the genome (Holland 2007) or the body plan (Lacalli 2005) as other than secondarily quite modified, and thus of limited value as a guide to the ancestral form.
Amphioxus is more conservative in terms of its genome, and a case can be made that it must certainly preserve the ancestral condition more fully than do tunicates, but can one be sure? Consider the origin and evolution of serial repeats in the vertebrate head, the subject of this symposium. There have been various proposals in the past, some based on the arrangement of skeletal structures, others on nerves, muscles, or branchial structures (Gilland and Baker 1993; Kuratani et al. 1997, 1999; Northcutt 2008), but without a final resolution. Tunicates and amphioxus provide two very different models for the ancestors of vertebrates. In tunicates, the head and trunk are quite separate, and there is no overlap between serially repeating structures from one to the other. In amphioxus, the head and trunk are more fully integrated, and the somite series extends forward into the head. Does one of these reflect the ancestral condition, or are both secondary specializations? Here, I will look first at the fossil evidence, limited though it is, and then patterns of innervation, for what they may reveal.
|
The importance of fossils |
|---|
Fossils are essential for reconstructing evolution, but I suspect we tend to forget how essential. A simple thought experiment will illustrate this. By way of analogy with the situation in protochordates, assume that the living vertebrates comprised assorted species from a small number of diverse groups, as in Fig. 1, and that everything else was extinct. Assume also that the morphology, development, and molecular biology of the surviving species could be examined as much as you like, but you would have no knowledge of more basal forms, lampreys, and such-like (or, at least, forms you knew to be basal), and no reliable fossils. How easily could you then make a convincing case for even the most basic facts that we know to be true about vertebrate origins, for example, that the first vertebrates were fish-like and aquatic rather than terrestrial or aerial. This would be especially problematic if the only outgroups available among the invertebrates were terrestrial or, like Drosophila, able to fly. For protochordates, the available outgroups are the mostly burrowing hemichordates and the echinoderms. The former are as yet poorly studied and the latter, although well studied, are endlessly puzzling and difficult to interpret in a phylogenetic context. Molecular analysis of the examples in Fig. 1 would probably show fish to be the most derived group, so they would sort out at the base of any molecular tree, but this is no more useful than knowing that larvaceans fall to the base of the tunicates in molecular phylogenies if we do not know why. If it is because they are more divergent in molecular terms than other tunicates, then they may be rather unreliable models for anything ancestral. What about developmental anatomy? Here, one can cite the example of gill slits in amniote embryos as perhaps the most incontrovertible evidence that vertebrates were originally aquatic. But there are always caveats, in this case that gill slits, along with other relicts of our aquatic past, could instead be embryonic adaptations that preadapted non-aquatic ancestors for a secondary invasion of watery habitats. This is a reminder, and an important one in my view, that evolutionary sequences we take for granted as being correct can prove, on closer inspection, to be convincing only because the alternatives are simply untenable in the face of fossil evidence. In the absence of such evidence, there seems no other option but to continue the task of identifying, by whatever means possible, structural or functional features for which it can plausibly be argued that they, like gill slits, are relicts of the evolutionary past. I will return to this point at the end of this essay.
|
|
Putative early chordate fossils: problematic but suggestive |
|---|
There are a number of fossils from the Lower and Middle Cambrian deposits that may be either stem-group chordates or their close relatives, but it is seldom obvious exactly what they are. A representative selection is shown in Fig. 2. Vetulicola is a vetulicolian, a group characterized by an enlarged anterior end bearing rows of structures thought to be respiratory openings, i.e., gill slits, behind which is a segmented trunk (reviewed by Aldridge et al. 2007
). Vetulicola is a rather robust animal, probably with a stiffened external covering, while other species in the group are more soft-bodied. Suggested affinities for these peculiar animals range from arthropods, despite the absence of appendages, to basal deuterostomes of a roughly hemichordate level of complexity to early tunicates. Proper tunicates are already present in these same deposits, however (Chen et al. 2003), and indeed, all the main chordate groups probably coexisted in the Lower Cambrian, including vertebrates (Shu et al. 1999). Pikaia, along with the apparently related Cathaymyrus (Shu et al. 1996), is more overtly chordate-like, with a series of what appear to be muscle bands, suggestive of myotomes, and an anterior head region with lateral projections that could be gills. A definitive analysis of this animal remains to be done, and nonchordate affinities, e.g., with annelids, have been suggested. Nevertheless, Pikaia remains a strong candidate for a basal chordate representing something close to an amphioxus level of organization. Yunnaozoans, represented in Fig. 2 by Haikouella (Chen et al. 1999), have a well-defined head region bearing structures that are almost certainly gills. The trunk has a prominent sail-like structure made of repeating elements. Traces of other repeating elements occur along the sides of the trunk, but these do not indicate very clearly how the trunk is organized, e.g., whether it is segmented or not. Interpretations vary as to whether Yunnanozoans are quite advanced animals, i.e., vertebrates or close to them, which is likely if Mallatt and Chen (2003) are correct in their interpretation of eyes, brain and notochord. A position closer to the base of the chordates, perhaps as basal deuterostomes, has also been suggested (Shu 2003).
|
The feature of interest in relation to the present symposium is that, all the above have a separate anterior head region bearing 4–6 gill-like structures, while the locomotory elements form a separate series restricted to the trunk. If at least some of these animals are indeed basal chordates or related to them, we would have direct evidence for a separate origin for the repeated elements of the branchial region and trunk, as Romer's somaticovisceral concept requires. The superposition of these two series in amphioxus would then be a secondary and later specialization. It is possible, however, that lumping these animals together as putative ancestral chordates is simply an indication of our preconceptions concerning the latter. Nevertheless, on balance, the fossil evidence is at least consistent with the supposition that vertebrates evolved from animals in which the locomotory apparatus, including somites, was restricted to the trunk. Serial structures in the head would, in such a situation, be entirely associated with visceral functions, including gills and support structures associated with them.
|
Stretching and compressing the branchio-visceralmotor series: amphioxus versus tunicates |
|---|
Innervation patterns have frequently been used to clarify the nature and registry of serially repeating structures in the vertebrate head, and for evidence of homology between these and serial repeats in protochordates (Fritzsch and Northcutt 1993
; Kuratani et al. 1997). The ability to identify motoneuron subtypes unambiguously by their gene expression profiles during development has put this type of analysis on a much firmer footing. Of special interest is the recent work on branchio-visceralmotor (bvm) neurons, which characteristically express Phox2b in both vertebrates and tunicates (Dufour et al. 2006; Hirsch et al. 2007). Amphioxus also has a distinct population of bvm neurons, located in the ventral midline of the nerve cord at the junctions between somites (Fig. 3). The series begins at the junction between myotomes 2 and 3, and extends from there through many somites (Wicht and Lacalli 2005). The branchial and visceral organs are thus innervated from far more of the cord than the 7–8 anterior-most segments that, on the basis of gene expression data, are considered to be the amphioxus homolog of the vertebrate hindbrain. Bvm axons in amphioxus pass out the dorsal nerves, which also carry incoming sensory fibers from epithelial sensory cells. There are no ventral roots in the vertebrate sense. Instead, the somatomotor (sm) neurons, which are ventrolateral in position, innervate the myotome directly across the basal lamina without leaving the confines of the nerve cord. There are both fast and slow fibers in the myotome, and these are innervated by separate subpopulations of sm neurons (Lacalli and Kelly 1999). In comparison with vertebrates, where the differentiated bvm neurons lie dorsal to sm neurons, the pattern is reversed in amphioxus. However, it is now clear from tracing studies in vertebrate hindbrain that the bvm neuronal precursors lie ventral to the precursors of sm neurons, and only secondarily move to a more dorsal location (Ericson et al. 1997). There is thus convincing support for homology between bvm neurons in vertebrates and tunicates and, pending molecular confirmation, potentially for amphioxus as well.
|
In amphioxus, both the pharyngeal slits and the nerve supplying them form an extended series over many somites, and the atrium has a comparable axial extent. This contrasts with the situation in tunicates, in which both the visceral nerves and the branchial structures they innervate are reduced in number and shifted forward. The source of bvm neurons in tunicate larvae, based on Phox2b data, is the narrow "neck" of tissue lying between the sensory vesicle and the tail ganglion (Dufour et al. 2006
), and it is from here that the whole of the adult brain develops (Fig. 4). Patterns of gene expression indicate that the "neck" includes a region corresponding roughly to the midbrain–hindbrain boundary (MHB) of vertebrates (Castro et al. 2006; Lacalli 2006) and at least some of the hindbrain, although precisely how much of the latter is currently unresolved (Dufour et al. 2006; Ikuta and Saiga 2007). There would seem, therefore, to be two alternatives: either the source of visceral fibers in tunicates has shifted forward compared with vertebrates, and now resides in a region that corresponds to the MHB and anterior hindbrain of the latter, or the whole of the hindbrain has been compressed forward in conjunction with a reduction in the number of visceral nerves arising from it. Regardless of which is correct, the development of the pharynx suggests a reason for such changes: initially there are only one or at most two pairs of slits in tunicates, called protostigma, which become definitive slits in taxa with few such slits (larvaceans and salps) or are repeatedly subdivided in tunicates with a more elaborate branchial filter and many slits (ascidians). In other words, where multiple slits are present, they are not part of an axially extended series as they are in amphioxus. In addition, from patterns of Hox gene expression, it is clear that opening of the atrium in tunicates also lies quite far forward, near the level of the anterior hindbrain (Lacalli 2005), which accords with the idea that slit number is increased in tunicates by replicating the anterior-most members of the series, rather than by extending it caudally. It is therefore hard to see the forward shift of the visceral innervation in tunicates as anything other than a specialized condition, one in fact implying that ancestral tunicates may have been quite small organisms, as they are specialized in ways consistent with this supposition. If this is indeed the case, two options remain: ancestral vertebrates either had many gill slits as in amphioxus, which were then reduced, or they had a smaller number, closer to what we see in the various fossils pictured above.
|
|
Somatomotor pathways: why ventral roots? |
|---|
A second neuroanatomical issue concerns the nature of dorsal and ventral pathways from the protochordate CNS and the relation these have to vertebrate dorsal and ventral roots. Given that the ventral pathway is exclusively associated with somatomotor (sm) innervation in amphioxus, the main evolutionary question is whether the presence of sm nerves in the vertebrate cranial series necessarily means that head somites were present in ancestral vertebrates. In other words, have somatic motor nerves always and only been associated with somite derivatives?
Dorsal spinal nerves in amphioxus, like the mixed cranial nerves in vertebrates, combine sensory and motor functions. Sm innervation is ventral, but the sm axons are confined within the cord, synapsing across the basal lamina to cellular extension from the myotomal muscle cells. This is a peculiar arrangement, but amphioxus is sufficiently small, and the myotome is so closely apposed to the nerve cord and notochord, that the distances involved are not great. Indeed, amphioxus locomotion appears to depend on an arrangement whereby the myotomes, notochord, and nerve cord are tightly bound together by a continuous sheath of connective tissue matrix to form a single functional unit. There are no gaps in the sheath surrounding the nerve cord through which nerves can leave except dorsally (Fig. 3), via pathways established early in development by the first sensory nerves. The pioneering nerve fibers reach the cord by passing above the nascent somites. In the adult, the nerves pass between the somites in the intermyotomal septa, but this is a secondary condition resulting from the subsequent upward growth of the myotomes. All the peripheral nerves in amphioxus, whatever their targets, thus originate as dorsal roots. The question then is whether the ventrally directed axons of the sm neurons within the cord are restricted to innervating somites because this is the ancestral condition, as most proposals regarding somite evolution have assumed (Bardet et al. 2005, Holland et al. 2008), or in contrast, whether the complete enclosure of the cord by myotomes has blocked the innervation of other targets, not derived from somites, that sm nerves or their antecedents once innervated.
Tunicates are not particularly informative on this issue because their bvm and sm nerve homologs arise from separate parts of the nerve cord, and it is not easy even to determine if one is more dorsal than the other. Indeed, it is not yet clear whether this is a meaningful distinction in the tunicate CNS. The salp ganglion (Fig. 4) illustrates this: in this animal the neurons providing branchial and visceral innervation are lateral in position, and their axons pass out roughly midway between the top and bottom of the ganglion (Lacalli and Holland 1998). Above the exit point are the cells of the eye rudiment and below are cells of unknown fate, not all of which are necessarily neurons. The peripheral nerves thus appear to be neither dorsal nor ventral in a conventional sense. Instead, they are roughly intermediate in position, but, because the ganglion projects above the level of the body, this is simply the most direct route into the CNS for any nerve fibers approaching it. Nothing is yet known in any detail about how patterns of innervation are established in these animals. Internal landmarks in the ganglion are probably involved, but the morphology suggests that the initial point of contact by pioneering fibers growing into the ganglion may be quite important. On a broader, phylogenetic scale, this implies a degree of flexibility and opportunism in the establishing points of nerve entry into, and exit from, the evolving CNS, and one can argue that the situation in amphioxus tends to confirm this. Pharyngeal slits, digestive tracts, and the nerves that supply both of these undoubtedly predate the origin of chordate somites, which implies that the restrictions the latter impose on patterns of peripheral nerve outgrowth are more recent.
What about the sm neurons themselves? We associate these in chordates with the innervation of somites or their derivatives, but sm neurons could predate somites and have been co-opted secondarily for their innervation. If so, they would once have had other targets. It is thus quite important to determine if sm-type neurons or their antecedents occur in hemichordates or echinoderms, and if so, what their targets are. Our understanding of the functional neuroanatomy of either group is unfortunately rather limited, particularly for hemichordates, despite their being the most promising outgroup for comparison with chordates. If, as above, pathways from the nerve cord are positioned more as a matter of convenience rather than being a fixed feature of cord architecture, it is easier to envisage an ancestral condition in which there are multiple routes out of the nerve cord, any one of which could have evolved into what we see today as the ventral somatomotor pathway. The problem with amphioxus is that all routes except the dorsal-most one are blocked by the myotome, so no evidence remains of any other routes to the periphery that may once have existed. The absence of true ventral roots in amphioxus may not therefore reflect the ancestral condition, which means we cannot rule out the possibility that there were once nonsomitic muscle blocks in the head that received sm-type innervation. If this were the case, the presence of the latter in the vertebrate head could be explained without recourse to ancestral, amphioxus-type head segments.
To explore this further, it is worth considering two ways that vertebrates and amphioxus differ. On one hand, vertebrates have a more elaborate skeletal support system to which the myotomes attach, rather than attaching directly to the notochord and nerve cord. Second, vertebrate embryos are much larger than their amphioxus counterparts (Fig. 5). Given the size difference between the postembryonic stages, the amphioxus arrangement would not suit vertebrates, as it provides much less leverage about the central axis than does attachment to a skeleton. It is the larger size of vertebrate embryos and their prolonged period of development that makes such alternatives possible, e.g., in allowing for the evolution of the sclerotome and its derivatives. Consequently, the main components of the vertebrate locomotory system can be larger than in amphioxus, and have more time to form complex structures and associations before they have to function as a unitary whole. Sizeable zones of loose connective tissue are present, which permits the growth of nerves into the zone between the nerve cord and myotome. The neural crest is important, as it is a key occupant of this zone, where it participates in the formation of peripheral ganglia. Various suggestions have been made as to how the neural crest first evolved. What is perhaps less appreciated is that unless, and until, an amphioxus-type association between cord and somites was replaced by something less restrictive, the space required for the migration and condensation of neural crest cells would not have existed. There is, in other words, a linkage between increased egg size, embryo size, and the consequent delay in larval function, which provides both time and space for novel structures and associations between cells to evolve in vertebrates.
|
|
Inversion and asymmetry: tracking mouth migration |
|---|
Although tracking evolutionary change among basal chordates is fraught with problems, the above analysis points up two potentially useful approaches. First, it is crucial to pay attention to the most relevant outgroup, amphioxus in the case of vertebrates, and (most likely) hemichordates in the case of amphioxus (Gerhart et al. 2005
). Second, as mentioned in the first section, it is useful to seek out evolutionary relicts, either of structures or physiological processes, where there is a prospect of confirming these as being ancestral rather than derived. Basically, as in the case of embryonic gill slits, one is looking for examples of recapitulation of one sort or another. To illustrate these two points, and to show that they are more than just arm-waving, I would like to cite an example where both come usefully into play. It is related to the evolution of the vertebrate head, although not specifically to head segmentation. I refer to the issue of dorsoventral inversion of chordates relative to other bilaterians, and the subsequent migration required of the mouth to move it to the new ventral surface of the body. The idea that chordates are dorsoventrally inverted relative to protostome invertebrates now has strong support from molecular data. Recent work on hemichordates indicates that the inversion happened, not at the base of the deuterostomes, but after the split between the ambulacraria [hemichordates + echinoderms] and chordates (Lowe et al. 2006). Not only are dorsal and ventral inverted between these groups, but left and right are also exchanged, just as one would predict (Hibino et al. 2006).
This puts an entirely new twist on an old puzzle concerning amphioxus, the asymmetry of the mouth. The larval mouth develops on the left side, while the gill slits are on the right (Fig. 6A), but the latter are left gill slits not right ones. They subsequently shift position due to differential growth during the late larval phase, moving in a clockwise direction as seen from behind. They end up on the left, while a new series of right slits develops on the right. The ventral midline thus starts out also on the right (Lacalli 2008a), as does the anus (Stokes and Holland 1995). The movement of the mouth to the ventral midline is, in contrast, anticlockwise. All of this should, in principle, require a very complex process of positional specification and growth control, so as to move some ventral structures in different direction than others. Dorsoventral inversion makes everything much simpler. If the mouth was originally dorsal, as the inversion hypothesis requires, then its site of formation in amphioxus, on the left side, is part of an overall anticlockwise rotation of the whole of the dorsoventral patterning system. Thus, the initial sites of formation of the mouth and gill slits are shifted by essentially similar amounts, and in the same direction. The mouth continues to move in an anticlockwise direction, apparently independently of the symmetry-restoring clockwise shift in all the other pharyngeal landmarks. If this interpretation is correct, then the movement of the mouth, from the left side of the head to the ventral midline, is in effect a recapitulation of a past evolutionary change. Note that the main support for this proposal comes ultimately from comparative molecular analysis, which has established inversion as a convincing hypothesis, and especially from work on hemichordates, which are outgroup of choice in this instance. As an aside, note also that the above bears on the question of brain origins, since the brain in this scenario derives from an epithelial domain located either behind the mouth or around it (Fig. 6B). One consequence of the latter is that the brain could only have coalesced and been internalized as a unitary structure after the mouth was moved out of the way, which suggests an epithelial brain may be an ancestral deuterostome feature. While this is all still hypothetical, it shows that there has been significant progress, if not towards an immediate solution, at least towards a redefinition of the problem and, hence, a deeper understanding of its subtleties.
|
|
Acknowledgments |
|---|
I thank Nick Holland and Chris Lowe for stimulating conversations on these topics. This study was supported by NSERC Canada.
Conflict of interest: None declared.
|
FOOTNOTES |
|---|
From the symposium "Vertebrate Head Segmentation in a Modern Evo-Devo Context" presented at the annual meeting of the Society for Integrative and Comparative Biology, January 2–6, 2008 at San Antonio, Texas.
|
References |
|---|
Aldridge RJ, Xian-Guang H, Siveter DJ, Siveter DJ, Gabbott SE. The systematics and phylogenetic relationships of vetulicolians. Palaeontol (2007) 50::131–168.[CrossRef]
Bardet PL, Schbert M, Horard B, Holland LZ, Laudet V, Holland ND, Vanacker JM. Expression of estrogen-receptor related receptors in amphioxus and zebrafish: implications for the evolution of posterior brain segmentation at the invertebrate-to-vertebrate transition. Evol Dev (2005) 7::223–233.[CrossRef][Web of Science][Medline]
Castro LC, Rasmussen S, Holland PWH, Holland ND, Holland LZ. A Gbx homeobox gene in amphioxus: insights into ancestry of the ANTP class and evolution of the midbrain/hindbrain boundary. Dev Biol (2006) 295::40–51.[CrossRef][Web of Science][Medline]
Chen JY, Huang DY, Li CW. An early Cambrain craniate-like chordate. Nature (1999) 402::518–522.[CrossRef]
Chen JY, Huang DY, Peng QQ, Chi HM, Qwang X, Feng M. The first tunicate from the early Cambrian of South China. Proc Natl Acad Sci USA (2003) 100::8314–8318.
Dufour HD, Chettouh Z, Deyts C, de Rosa R, Gordis C, Joly JS, Brunet JF. Precraniate origin of cranial motoneurons. Proc Natl Acad Sci USA (2006) 103::8727–8732.
Ericson J, Rashbass P, Shedl A, Brenner-Morton S, Kawakami A, van Heyningen V, Jessell TM, Briscoe J. Pax6 controls progenitor cell identity and neuronal fate in response to graded Shh signaling. Cell (1997) 90::169–180.[CrossRef][Web of Science][Medline]
Gerhart J, Lowe C, Kirschner M. Hemichordates and the origin of chordates. Curr Opin Gen Dev (2005) 15::461–467.[CrossRef][Web of Science][Medline]
Gilland E, Baker R. Conservation of neuroepithelial and mesodermal segments in the embryonic head. Acta Anat (1993) 148::110–123.[Web of Science][Medline]
Fritzsch B, Northcutt RG. Cranial and spinal nerve organization in amphioxus and lampreys: evidence for an ancestral craniate pattern. Acta Anat (1993) 148::96–109.[Web of Science][Medline]
Hibino T, Nishino A, Amemiya S. Phylogenetic correspondence of the body axis in bilaterians is revealed by the right-sided expression of Pitx genes in echinoderm larvae. Dev Gr Diff (2006) 48::587–595.[CrossRef]
Hirsch MR, Glover JC, Dufour HD, Brunet JF, Gordis C. Forced expression of Phox2 homeodomain transcription factors induces a branchio-visceromotor axonal phenotype. Dev Biol (2007) 303::687–702.[CrossRef][Web of Science][Medline]
Holland LZ. A chordate with a difference. Nature (2007) 447::1153–1155.
Holland LZ, Rasmussen SLK, Beaster-Jones L, Koop D. Amphioxus and the evolution of head segmentation. (2008) Proceedings of the Society for Integrative and Comparative Biology, January 2–6: in San Antonio, Texas. (http://www.sicb.org/meetings/2008/schedule/).
Ikuta T, Saiga H. Dynamic change in the expression of developmental genes in the ascidian central nervous system: revisit to the tripartite model and the origin of the midbrain-hindbrain boundary region. Dev Biol (2007) 312::631–643.[CrossRef][Web of Science][Medline]
Kuratani S, Ueki T, Aizawa S, Hirano S. Peripheral development of cranial nerves in a cyclostome, Lampetra japonica: morphological distribution of nerve branches and the vertebrate body plan. J Comp Neurol (1997) 384::483–500.[CrossRef][Web of Science][Medline]
Kuratani S, Ueki T, Hirano S, Aizawa S. Rostral truncation of a cyclostome, Lampetra japonica, induced by all-trans retinoic acid defines the head/trunk interfacee of the vertebrate body. Dev Dynam (1998) 211::35–51.[CrossRef][Web of Science][Medline]
Kuratani S, Horigome N, Hirano S. Developmental morphology of the head mesoderm and reevaluation of segment theories of the vertebrate head: evidence from embryos of an agnathan vertebrate, Lampetra japonica. Dev Biol (1999) 210::381–400.[CrossRef][Web of Science][Medline]
Lacalli TC. Protochordate body plan and the evolutionary role of larvae: old controversies resolved? Can J Zool (2005) 83::216–224.[CrossRef]
Lacalli TC. Prospective protochordate homologs of the vertebrate midbrain and MHB, with some thoughts on MHB origins. Int J Biol Sci (2006) 2::1004–1009.
Lacalli TC. Mucus secretion and transport in amphioxus larvae: organization and ultrastructure of the food trapping system, and implications for head evolution. Acta Zool (2008a) (in press) [doi:10.1111/j.1463-6395.2007.00310.x] [Epub ahead of print].
Lacalli TC. Basic features of the ancestral chordate brain: a protochordate perspective. Brain Res Bull (2008b) 75::319–323.[CrossRef][Medline]
Lacalli TC, Holland LZ. The developing dorsal ganglion of the salp Thalia democratica, and the nature of the ancestral chordate brain. Phil Trans R Soc Lond B (1998) 353::1943–1967.
Lacalli TC, Kelly SJ. Somatic motoneurons in amphioxus larvae: cell types, cell position and innervation patterns. Acta Zool (1999) 80::113–124.[CrossRef]
Lowe CJ, et al. Dorsoventral patterning in hemichordates: insights into early chordate evolution. PLoS Biol (2006) 4::1603–1619.[Web of Science]
Mallatt J, Chen JY. Fossil sister group of craniates: predicted and found. J Morph (2003) 258::1–31.[CrossRef][Medline]
McMurrich JP. The problem of the vertebrate head in the light of comparative anatomy. In: Proceedings of the Seventh International Zoological Congress (1912) Boston. Cambridge: Cambridge University Press. 167–176.
Northcutt RG. Historical hypotheses regarding vertebrate head organization. (2008) Proceedings of the Society for Integrative and Comparative Biology, January 2–6: in San Antonio, Texas. (http://www.sicb.org/meetings/2008/schedule/).
Romer AS. Major steps in vertebrate evolution. Science (1967) 158::1629–1637.
Romer AS. The vertebrate as dual animal – somatic and visceral. Evol Biol (1972) 6::121–156.
Shu DG. A paleontological perspective of vertebrate origin. Chin Sci Bull (2003) 48::725–735.[CrossRef]
Shu DG, Conway Morris S, Zhang L. A Pikaia-like chordate from the lower Cambrian of China. Nature (1996) 384::157–158.[CrossRef]
Shu DG, Luo HL, Conway Morris S, Zhang XL, Hu SX, Chen L, Han J, Zhu M, Li Y, Chen LZ. Lower Cambrian vertebrates from south China. Nature (1999) 402::42–46.[CrossRef]
Stokes MD, Holland ND. Embryos and larva of a lancelet, Branchiostoma floridae, from hatching through metamorphosis: growth in the laboratory and external morphology. Acta Zool (1995) 76::105–120.
Wicht H, Lacalli TC. The nervous system of amphioxus: structure, development and evolutionary significance. Can J Zool (2005) 83::122–150.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||