Dev Genes Evol (2000) 210:518–521
© Springer-Verlag 2000
EXPRESSION NOTE
Robert D. Knight · Georgia D. Panopoulou Peter W.H. Holland · Sebastian M. Shimeld
An amphioxus Krox gene: insights into vertebrate hindbrain evolution
Received: 10 April 2000 / Accepted: 25 May 2000
Abstract The transcription factor Krox-20 has roles in the maintenance of segmentation and specification of segment identity in the vertebrate hindbrain. Overt hindbrain segmentation is a vertebrate novelty, and is not seen in invertebrate chordates such as amphioxus and tunicates. To test if the roles of Krox-20 are also derived, we cloned a Krox-20 related gene, AmphiKrox, from amphioxus. AmphiKrox is related to a small family of vertebrate Krox genes and is expressed in the most anterior region of the amphioxus brain and in the club shaped gland, a secretory organ that develops in the anterior pharynx. Neither expression domain overlaps with the expression of AmphiHox-1, -2, -3 or -4, suggesting that the roles of Krox-20 in hindbrain segmentation and in Hox gene regulation were acquired concomitant with the duplication of Krox genes in vertebrate evolution. Key words Krox-20 · Hox · amphioxus · Hindbrain · Vertebrate evolution
Introduction During the development of vertebrate embryos, the hindbrain forms up to eight segments called rhombomeres (Lumsden 1990). Each rhombomere gives rise to specific derivatives, including the cranial nerves and cranial neural crest cells, of which the latter migrate to contribute to craniofacial structures (Le Douarin 1981). A number of genes are implicated in the direct control of segmentation, including the kreisler/valentino gene (Frohman et Edited by R. Balling R.D. Knight · P.W.H. Holland · S.M. Shimeld (✉) School of Animal and Microbial Sciences, University of Reading, Whiteknights, PO Box 228, Reading RG6 6AJ, UK e-mail:
[email protected] Tel.: +44-118-9875123 ext. 7084, Fax: +44-118-9310180 G.D. Panopoulou Max-Planck-Institut für Molekulare Genetik, 14195 Berlin-Dahlem, Germany
al. 1993; Moens et al. 1998), members of the ephrin/eph gene family (Xu et al. 1995) and Krox-20 (Wilkinson et al. 1989a). Krox-20 encodes a C2H2 zinc finger transcription factor and is a member of the Egr gene family, which in mice and humans contains four closely related genes known as Egr-1 (Krox-24), Egr-2 (Krox-20), Egr-3 and Egr-4 (Chavrier et al. 1988; Crosby et al. 1991; Patwardhan et al. 1991; Sukhatme et al. 1988). Krox-20 is expressed in rhombomeres 3 and 5 in all vertebrates examined (Oxtoby and Jowett 1993; Wilkinson et al. 1989a) and, where tested, is required for their maintenance, demonstrating a requirement for Krox-20 for correct segmentation of the hindbrain (Schneider-Maunoury et al. 1993, 1997). Each rhombomere of the vertebrate hindbrain has a specific identity, and this appears to be determined by the nested expression of Hox genes (McGinnis and Krumlauf 1992). Control of identity is tightly linked to segmentation, since Hox gene expression limits come to be defined by rhombomere boundaries (Wilkinson et al. 1989b). Krox-20 is also directly involved in this process, since it has a conserved function in vertebrates in directly activating vertebrate Hox-2 group genes in rhombomeres 3 and 5 (Nonchev et al. 1996; Vesque et al. 1996). Krox-20 thus plays essential roles in both the segmentation of the hindbrain and the specification of segment identity via regulation of Hox gene expression. In the phylum chordata, overt neural segmentation, in which discrete boundaries form between segments, is a character seen only in vertebrates, although repetition of structures along the anterior-posterior neural axis is also seen in both amphioxus and ascidians. In amphioxus this takes the form of regular repetition of dorsal nerve exit points and of particular cell types (Bone 1960; Fritsch and Northcutt 1993; Yasui et al. 1998). In ascidians, repeated expression of a Pax-3/7 gene has been interpreted as evidence for segmentation (Wada et al. 1996). Both taxa also have nested neural expression of Hox genes (Gionti et al. 1998; Katsuyama et al. 1995; Locascio et al. 1999; Wada et al. 1999). Neither taxon, however, shows definitive neural compartments separated by
519
boundaries, as is seen for rhombomeres in the vertebrate hindbrain. Consequently, this is likely to be a vertebrate innovation. We wished to test whether the roles of Krox-20 in the vertebrate hindbrain are also derived, and potentially linked to the evolution of hindbrain segmentation. Several hypotheses are possible. One is that the regulation of Hox-2 genes by Krox-20 is ancestral, but that its role in neural segmentation is new. A second is that a role in neural segmentation is ancestral (either throughout the length of the neural cord, or confined to anterior regions), but that regulation of Hox-2 genes is new. A third is that both characters are derived. These alternative hypotheses make specific predictions regarding the ancestral relationship between Hox and Krox expression in a non-vertebrate chordate.
Materials and methods A Branchiostoma floridae cDNA library (MPIMG531) constructed from 26 h neurula stage embryos was screened at high stringency (65°C) in Church and Gilbert’s buffer [50% (v/v) 0.5 M Na2HPO4 (pH 7.2), 7% (w/v) SDS, 1% (w/v) BSA, 1 mM EDTA (pH 8)], using a 200 bp zebrafish Egr-1 clone comprising the highly conserved zinc finger region (Lanfear et al. 1991). Alignments of the zinc finger regions of amino acid sequences were performed using CLUSTALW (Thompson et al. 1994) and edited by eye. Maximum likelihood trees were constructed using Puzzle, allowing rate heterogeneity between sites (eight variable gamma-distributed rates and one fixed rate) from the aligned zinc finger regions (Strimmer and von Haesler 1996). In situ hybridisations of amphioxus embryos and larvae were performed according to Holland (1999) with the following modifications. Proteinase K incubation treatment of embryos was performed at 37°C with incubation times of 7–15 min depending on the embryonic stage. Hybridisation and washing of embryos was performed at 55°C, with three successive washes of 30 min duration in 50% formamide, 5× SSC, 0.1% Tween 20; 50% formamide, 2× SSC, 0.1% Tween 20; 50% formamide, 1× SSC, 0.1% Tween 20.
Results and discussion Two amphioxus cDNA clones of approximate sizes 2.2 kb and 1.2 kb (clone identification numbers MPMGp531K2070 and MPMGp531H08121 respectively), hybridised strongly in secondary screens to zebrafish Egr-1, and when sequenced were found to be derived from the same gene. Similarity comparisons of these two clones when compared against the TREMBL and EMBL databases using BLAST showed closest matches to vertebrate Egr genes. The amphioxus gene encodes a putative protein of 299 amino acids, including three C2H2 zinc finger motifs. The zinc finger region shows 90–95% similarity to the vertebrate Egr genes and Drosophila stripe over the zinc finger region; we therefore denote the gene AmphiKrox. Figure 1A shows an alignment of the amino acid sequence of AmphiKrox, vertebrate Egr-1, Egr-2, Egr-3 and Egr-4, Drosophila stripe and a Caenorhabditis elegans putative orthologue. All residues putatively in-
Fig. 1 A Alignment of the zinc finger region of AmphiKrox with selected Egr family members, including Drosophila Stripe and a putative Caenorhabditis elegans family member (accession number Q18250). The AmphiKrox sequence is available from the EMBL database, accession number AJ278149. B Phylogenetic tree of the Egr gene family and WT-1 constructed from the amino acid sequences of the zinc finger region by maximum likelihood using Puzzle. Numbers at the nodes are percentage quartet puzzling support values and the scale indicates maximum likelihood branch lengths. AmphiKrox groups robustly with the vertebrate Egr sequences; however, resolution within this cluster is poor
volved in DNA sequence recognition are conserved between invertebrates and vertebrates, as are several residues adjacent to the zinc fingers, suggesting functional roles for these regions. No other conserved domains were found between AmphiKrox and the vertebrate Egr proteins. Phylogenetic trees were constructed from an alignment of vertebrate Egr-1, -2, -3 and -4, Drosophila
520
Fig. 2A–C Analysis of AmphiKrox gene expression by wholemount in situ hybridisation. Anterior is to the right. A Right-sided view of an embryo at the end of neurulation. AmphiKrox expression is visible in the cerebral vesicle (cv) and in confined region of pharynx (ph) endoderm where the club shaped gland (csg) will form. n Notochord, nt neural tube. B Right-sided view of an early larva. Weak expression of AmphiKrox remains in the cv, while strong expression is observed in the developing csg. hp Hatschek’s pit. C Ventral view of a larva with two gill slits. AmphiKrox expression is confined to cell of the dorsal club shaped gland (dcsg) on the right hand side. Ventral club shaped gland (vcsg) cells, which form the duct, are also visible
stripe, a C. elegans orthologue, AmphiKrox and the Wilms Tumour genes (WT-1) from human and alligator. WT-1 genes have only been isolated from vertebrates and possess four zinc fingers, of which the last three fingers show high similarity to the zinc fingers of the Egr genes (Call et al. 1990). Therefore WT-1 genes were used as outgroups for phylogenetic tree construction. A maximum likelihood tree was constructed with Puzzle allowing for rate heterogeneity between sites (Fig. 1B). This showed clustering of the vertebrate Egr sequences and AmphiKrox, with strong support. The relationship between AmphiKrox and the vertebrate Egr genes is not completely resolved in this tree, probably as a consequence of the high degree of conservation over the zinc finger region. In this tree and a tree constructed in a similar manner using only Egr family sequences, AmphiKrox falls basally to the vertebrate Egr genes. These trees show that AmphiKrox is equally related to vertebrate genes Egr-1, -2 and -3, suggesting that these genes are descended from the same ancestral gene in the most recent common ancestor of cephalochordates and vertebrates. This is supported by the observation that Egr-1, -2 and -3 lie on separate chromosomes in well defined
paralogy regions, suggesting that these genes arose through gene duplication in the vertebrate lineage (Pebusque et al. 1998). The phylogeny also suggests that the Egr-4 genes are as closely related to AmphiKrox as are Egr-1, -2 and -3. However the branch length of the Egr-4 group is long compared to the other vertebrate Egr sequences. This probably reflects a relatively rapid rate of evolution for these genes in the vertebrate lineage. In summary, these analyses show that AmphiKrox is a genuine member of the Egr gene family and is equally related to the vertebrate Egr-1, -2, -3 and -4 genes, consistent with AmphiKrox being descended from the same ancestral gene as these vertebrate genes. The expression of AmphiKrox was examined by whole mount in situ hybridisation (Fig. 2). AmphiKrox expression was not detected until mid neurula stages (prior to neural tube closure). Initial expression is strong in the anterior cerebral vesicle (CV), including the prospective frontal eye region, and in the club shaped gland (CSG). Expression in the CV gradually fades in larval stages (after 30 h) and is undetectable by 48 h post-fertilisation. The expression in the CSG becomes restricted dorsally as the gland extends ventrally and forms an external opening on the left side. This expression remains strong in all larval stages examined (up to 48 h post-fertilisation). Other than the anterior cerebral vesicle, we did not observe AmphiKrox expression elsewhere in the developing nervous system. The CSG of amphioxus is an enigmatic structure. It is a larval specific structure which is lost on metamorphosis and is secretory, releasing mucous secretions into the lumen of the gut (Olsson 1983). Based on its position relative to the gill slits and its mode of development, it has been suggested to have evolved by modification of the first gill slit (Goodrich 1930). No morphological differences between the dorsal and ventral cells of the CSG have previously been described. The restriction of AmphiKrox expression to the dorsal part of the CSG described here, however, suggests there is dorsalventral regionalisation of the CSG. This expression domain of AmphiKrox is likely to be a derived feature of amphioxus development. Vertebrate Egr genes do show expression in the brain (Beckmann and Wilce 1997), so expression of AmphiKrox in the CV may represent an ancestral character. Wada et al. (1999) analysed the expression of amphioxus Hox genes of groups 1–4 at comparable developmental stages, and identified expression of Hox-1, Hox-2 and Hox-4 in the neural tube posterior to the cerebral vesicle and of Hox-2 in Hatschek’s pit. None of these expression domains overlaps with the expression of AmphiKrox, as determined in this study. This contrasts with the situation in the vertebrate hindbrain where Hox genes of groups 1–3 do show overlap with Krox-20 expression, refuting their functional interaction in amphioxus. Thus, we see neither repetitive neural expression of AmphiKrox, nor overlap of expression with amphioxus Hox gene expression. We conclude that the roles of Krox-20 in rhombomere segmentation and Hox gene
521
regulation in the vertebrate hindbrain are both likely to be vertebrate novelties. Since Krox-20 is the only one of four vertebrate Egr gene family members to be involved in these processes, the evolution of these new functions may have depended on the duplication of an ancestral Egr family gene early in vertebrate evolution. Acknowledgements We would like to thank William Jackman and Ragnar Olsson for discussions and for communicating unpublished data, and John Lawrence for generous loan of laboratory space in Tampa. This work was funded by the MRC and BBSRC and was facilitated by a BBSRC ISIS award and by the British Council/German Academic Collaboration programme.
References Beckmann AM, Wilce PA (1997) Egr transcription factors in the nervous system. Neurochem Int 31:477–510 Bone Q (1960) The central nervous system in amphioxus. J Comp Neurol 115:27–64 Call K, Glaser T, Ito C, Buckler A, Pelletier J, Haber D, Rose E, Kral A, Yeger H, Lewis W, Jones C, Housman DE (1990) Isolation and characterization of a zinc finger polypeptide gene at the human chromosome 11 Wilms’ tumor locus. Cell 60: 509–520 Chavrier P, Lemaire P, Revelant O, Bravo R, Charnay P (1988) Characterization of a mouse multigene family that encodes zinc finger structures. Mol Cell Biol 8:1319–1326 Crosby S, Puetz J, Simburger K, Fahrner T, Milbrandt J (1991) The early response gene NGFI-C encodes a zinc finger transcriptional activator and is a member of the GCGGGGGCG (GSG) element-binding protein family. Mol Cell Biol 11: 3835–3841 Fritsch B, Northcutt RG (1993) Cranial and spinal nerve organization in amphioxus and lampreys: evidence for an ancestral craniate pattern. Acta Anat 148:96–109 Frohman MA, Martin GR, Cordes SP, Halamed LP, Barsh GS (1993) Altered rhombomere-specific gene expression and hyoid bone differentiation in the mouse segmentation mutant, kreisler. Development 117:925–936 Gionti M, Ristoratore F, Di Gregorio A, Aniello F, Branno M, Di Lauro R (1998) Cihox5, a new Ciona intestinalis Hox-related gene, is involved in regionalization of the spinal cord. Dev Genes Evol 207:515–523 Goodrich ES (1930) The development of the Club-shaped Gland in amphioxus. Q J Microsc Sci 74:155–164 Holland PWH (1999) In situ hybridisation to amphioxus embryos. In: Sharpe PT, Mason IJ(eds) Molecular embryology – methods and protocols. Humana Press, Totowa, pp 641–644 Katsuyama Y, Wada S, Yasugi S, Saiga H (1995) Expression of the labial group Hox gene HrHox-1 and its alteration induced by retinoic acid in development of the ascidian Halocynthia roretzi. Development 121:3197–3205 Lanfear J, Jowett T, Holland PWH (1991) Cloning of fish zinc-finger genes related to Krox-20 and Krox-24. Biochem Biophys Res Commun 179:1220–1224 Le Douarin N (1981) The neural crest. Cambridge University Press, Cambridge Locascio A, Aniell F, Alessandro A, Manzanares M, Krumlauf R, Branno M (1999) Patterning of the ascidian nervous system: structure, expression and transgenic analysis of the CiHox3 gene. Development 126:4737–4748 Lumsden AGS (1990) The development and significance of hindbrain segmentation. Semin Dev Biol 1:117–125
McGinnis W, Krumlauf R (1992) Homeobox genes and axial patterning. Cell 68:283–302 Moens CB, Cordes SP, Giorgiani MW, Barsh GS, Kimmel CB (1998) Equivalence in the genetic control of hindbrain segmentation in fish and mouse. Development 125:381–391 Nonchev S, Maconochie M, Vesque C, Aparicio S, ArizaMcNaughton L, Manzanares M, Maruthainar K, Kuroiwa A, Brenner S, Charnay P, Krumlauf R (1996) The conserved role of Krox-20 in directing Hox gene expression during vertebrate hindbrain segmentation. Proc Natl Acad Sci USA 93:9339–9345 Olsson R (1983) Club-shaped Gland and endostyle in larval Branchiostoma lanceolatum (Cephalochordata). Zoomorphology 103:1–13 Oxtoby E, Jowett T (1993) Cloning of the zebrafish Krox-20 gene and its expression during hindbrain development. Nucleic Acids Res 21:1087–1095 Patwardhan S, Gashler A, Siegel M, Chang L, Joseph L, Shows T, Le Beau M, Sukhatme V (1991) EGR3, a novel member of the Egr family of genes encoding immediate-early transcription factors. Oncogene 6:917–928 Pebusque MJ, Coulier F, Birnbaum D, Pontarotti P (1998) Ancient large-scale genome duplications: phylogenetic and linkage analyses shed light on chordate genome evolution. Mol Biol Evol 15:1145–1159 Schneider-Maunoury S, Topilko P, Seitandou T, Levi G, CohenTannoudji M, Pournin S, Babinet C, Charnay P (1993) Disruption of Krox-20 results in alteration of rhombomeres 3 and 5 in the developing hindbrain. Cell 75:1199–1214 Schneider-Maunoury S, Seitanidou T, Charnay P, Lumsden A (1997) Segmental and neuronal architecture of the hindbrain of Krox-20 mouse mutants. Development 124:1215–1226 Strimmer K, von Haesler A (1996) Quartet Puzzling: a quartet maximum-likelihood method for reconstructing tree topologies. Mol Biol Evol 13:964–969 Sukhatme V, Cao X, Chan L, Tsai-Morris C, Stamenkovich D, Ferreira P, Cohen D, Edwards S, Shows T, Curran T (1988) A zinc finger-encoding gene coregulated with c-fos during growth and differentiation, and after cellular depolarisation. Cell 53:37–43 Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680 Vesque C, Maconochie M, Nonchev S, Ariza-McNaughton L, Kuroiwa A, Charnay P, Krumlauf R (1996) Hoxb-2 transcriptional activation in rhombomeres 3 and 5 requires an evolutionarily conserved cis-acting element in addition to the Krox-20 binding site. EMBO J 15:5383–5396 Wada H, Holland PWH, Satoh N (1996) Origin of patterning in neural tubes. Nature 384:123 Wada H, Garcia-Fernandez J, Holland PWH (1999) Colinear and segemntal expression of amphioxus Hox genes. Dev Biol 213: 131–141 Wilkinson D, Bhatt S, Chavrier P, Bravo R, Charnay P (1989a) Segment-specific expression of a zinc-finger gene in the developing nervous system of the mouse. Nature 337:461–464 Wilkinson DW, Bhatt S, Cook M, Boncinelli E, Krumlauf R (1989b) Segmental expression of Hox 2 homeobox-containing genes in the developing mouse hindbrain. Nature 341:405–409 Xu Q, Alldus G, Holder N, Wilkinson DG (1995) Expression of truncated Sek-1 receptor tyrosine kinase disrupts the segmental restriction of gene expression in the Xenopus and zebrafish hindbrain. Development 121:4005–4016 Yasui K, Tabata S, Ueki T, Uemura M, Zhang SC (1998) Early development of the peripheral nervous system in a lancelet species. J Comp Neurol 393:415–425
