4943 Introduction In amphibians, the formation of the anterior-posterior (AP) axis is dependent on Spemann’s organizer activity (Spemann and Mangold, 1924). Classic transplantation experiments demonstrated that the inductive properties of the organizer change in the course of development. The early organizer can induce a complete secondary body axis including head, whereas the late organizer can only induce trunk-tail structures (Spemann, 1931). This led to the concept of two organizing centers: the head and the trunk-tail organizers. Recently, molecules that are expressed in the Spemann’s organizer have been identified in Xenopus (reviewed by De Robertis et al., 1997). When ectopically expressed in the ventral side of Xenopus embryos, some of these factors, like goosecoid, noggin or chordin, can induce secondary body axis (Cho et al., 1991; Smith and Harland, 1992; Sasai et al., 1994). In contrast to these axis-inducing factors, secreted proteins such as Cerberus and Dickkopf-1 are only able to induce head-like structures (Bouwmeester et al., 1996; Glinka et al., 1997). In Xenopus, cerberus is expressed in the non-involuting anterior dorsal endoderm (ADE), but not in the involuting mesoderm. The presence of the strong head-inducing factor Cerberus in the ADE raised the possibility that this region could be the head organizing center in Xenopus (Bouwmeester et al., 1996; Bouwmeester and Leyns, 1997). Biochemical analysis in Xenopus showed that Cerberus can bind to Xnr1, BMP4 and Xwnt8 and thereby blocks their function (Piccolo et al., 1999). These inhibitory properties of Cerberus are considered essential for the head inducing activity of this secreted factor. A gene homologous to cerberus has been isolated in the mouse (Belo et al., 1997; Biben et al., 1998; Shawlot et al., 1998). The expression of mouse cerberus-like (cer-l) and other markers such as Hesx1, Lim1 and Otx2 in the anterior visceral endoderm (AVE), led to the hypothesis that this region is the topological mouse equivalent of the ADE in Xenopus (Acampora et al., 1995; Thomas and Beddington, 1996; Belo et al., 1997; Bouwmeester and Leyns, 1997). Therefore, the AVE was proposed to be the head organizer in the mouse. This view is supported by the finding that in chimeric mutant mouse embryos composed of AVE that lacks either Otx2, Lim1 or Hnf3β, and wild-type epiblast, the head is not properly induced (Rhinn et al., 1998; Shawlot et al., 1999; Dufort et al., 1998). Surprisingly, in generated cer-l knockout (KO) mouse lines no phenotypic head and axis defects were observed, arguing against a role of cer-l in early embryogenesis (Belo et al., 2000; Shawlot et al., 2000; Stanley et al., 2000). In Xenopus, the endogenous function of Cerberus in the ADE remains unclear because of the lack of loss-of-function data. In order to characterize the function of Cerberus in head formation, a novel combination of strategies was employed. Endogenous Cerberus was ‘knocked down’ using an antisense morpholino oligonucleotide that specifically blocked the translation of the cerberus mRNA (CerMO). In addition, the relative levels of the signaling molecules BMP4, Xnr1 and We analyzed the endogenous requirement for Cerberus in Xenopus head development. ‘Knockdown’ of Cerberus function by antisense morpholino oligonucleotides did not impair head formation in the embryo. In contrast, targeted increase of BMP, Nodal and Wnt signaling in the anterior dorsal-endoderm (ADE) resulted in synergistic loss of anterior head structures, without affecting more posterior axial ones. Remarkably, those head phenotypes were aggravated by simultaneous depletion of Cerberus. These experiments demonstrated for the first time that endogenous Cerberus protein can inhibit BMP, Nodal and Wnt factors in vivo. Conjugates of dorsal ectoderm (DE) and ADE explants in which Cerberus function was ‘knocked down’ revealed the requirement of Cerberus in the ADE for the proper induction of anterior neural markers and repression of more posterior ones. This data supports the view that Cerberus function is required in the leading edge of the ADE for correct induction and patterning of the neuroectoderm. Key words: Cerberus, Head induction, Morpholino, Targeted activation, Xenopus laevis Summary Endogenous Cerberus activity is required for anterior head specification in Xenopus Ana Cristina Silva 1,† , Mario Filipe 1,† , Klaus-Michael Kuerner 2, *, Herbert Steinbeisser 2, * ,‡ and José António Belo 1,3,‡ 1Instituto Gulbenkian de Ciência, Rua da Quinta Grande 6, Apartado 14, 2781-901 Oeiras, Portugal 2 Max-Planck-Institut für Entwicklungsbiologie, Abt. Zellbiologie, Spemannstrasse 35, 72076 Tuebingen, Germany 3 Faculdade de Engenharia de Recursos Naturais, Universidade do Algarve, Campus de Gambelas, 8000-010 Faro, Portugal *Present address: Institute of Human Genetics, University of Heidelberg, Im Neuenheimer Feld 328, 69120 Heidelberg, Germany † These authors contributed equally to this work ‡ Authors for correspondence (e-mail: jbelo@igc.gulbenkian.pt and herbert.steinbeisser@med.uni-heidelberg.de) Accepted 4 July 2003 Development 130, 4943-4953 © 2003 The Company of Biologists Ltd doi:10.1242/dev.00705 Research article