This page has been archived and is no longer updated
Segmental patterning of the vertebrate embryonic axis
Author: Mary-Lee Dequéant
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"Vertebrate somites are epithelial blocks of mesoderm containing the precursors of the vertebrae and the skeletal muscles (FIG. 1). They form rhythmically from the presomitic mesoderm (PSM) at a time period that is characteristic of the species, ranging from 30 minutes in zebrafish embryos, 90 minutes in chicken and 120 min- utes in mouse, to approximately 4?5 hours in humans 1 . Pairs of somites regularly pinch off synchronously from the anterior tip of the PSM in an anterior-to-posterior sequence until a defined number, also characteristic of the species, is reached. Although this number is usually highly constrained within a given species, it varies widely between species, ranging from approximately 30 pairs of somites in some fish to several hundred in snakes 2 . The somitic lineage is part of the paraxial mesoderm in amni- ote species such as chicken, mouse and human, and is part of the dorsal mesoderm in lower vertebrates such as fish and frog. In all vertebrates, the somitic series begins anteriorly immediately caudal to the otic vesicle (FIG. 1) and runs posteriorly on both sides of the neural tube and notochord to the caudal tip of the embryo. The first five somites are incorporated into the basi-occipital bone at the base of the skull, whereas the more posterior somites form the vertebral column 3 . Somitogenesis in amniotes can be subdivided into three main phases (FIG. 1). First, during the specifica- tion phase, the descendants of the epiblast and later on of the tail bud acquire a paraxial mesoderm identity. These cells are progressively added to the posterior tip of the embryo to form the PSM. During the second phase, a segmental pre-pattern manifests as a stripe of gene expression that is established in the anterior PSM. This pre-pattern provides the blueprint from which the morphological segment ? the somite ? will be formed during the final phase. Then, the rostrocaudal polarity of the future somite is established in the newly speci- fied segment. This rostrocaudal subdivision of somites controls the segmentation of the nervous system by restricting migration of neural crest cells and axons to the anterior part of the somites 4 . This subdivision is also responsible for the definitive patterning of verte- brae that form when the posterior part of one somite fuses to the anterior part of the consecutive somite during a process called resegmentation. Finally, the formation of the morphological boundaries results in the separation of the epithelial somite from the PSM 5 . Soon after their formation, somites subdivide into the ventral sclerotome that contains the precursors of the axial skeleton and the dorsal dermomyotome that con- tributes to the myotome and dermatome, which form the skeletal muscles and dermis of the back, respec- tively 6 . Finally, depending on their position along the anteroposterior (AP) axis, somite derivatives acquire a defined anatomical identity that is imposed mainly by Hox genes that control their subsequent regional differentiation 7,8 . The mechanisms underlying the ros- trocaudal subdivision and maturation of the somites have been reviewed 6,9?11 and will not be discussed here further. Instead, we focus on the mechanisms involved in the generation of the metamery in the PSM, an aspect of somitogenesis that has received the most attention over the past 10 years since the identification of an oscillator associated with this process called the segmentation clock. *Stowers Institute for Medical Research and ? Howard Hughes Medical Institute, 1000 East 50 th Street, Kansas City, Missouri 64110, USA. Correspondence to O.P. e-mail: olp@stowers-institute.org doi:10.1038/nrg2320 Published online 15 April 2008 Somites Embryonic segments (epithelial blocks of tissue surrounding a cavity called somitocoele) giving rise to the sclerotome (precursors of the axial skeleton) and dermomyotome (precursors of the dermis of the back and skeletal muscles). Presomitic mesoderm A mesoderm-derived mesenchymal tissue lying on both sides of the neural tube that gives rise to the somites. Paraxial mesoderm A mesodermal tissue comprising the head mesoderm and the somitic mesoderm. Otic vesicle One of the paired sacs of invaginated ectoderm that develops into the inner ear. Segmental patterning of the vertebrate embryonic axis Mary-Lee Dequ�ant* and Olivier Pourqui�* ? Abstract | The body axis of vertebrates is composed of a serial repetition of similar anatomical modules that are called segments or metameres. This particular mode of organization is especially conspicuous at the level of the periodic arrangement of vertebrae in the spine. The segmental pattern is established during embryogenesis when the somites ? the embryonic segments of vertebrates ? are rhythmically produced from the paraxial mesoderm. This process involves the segmentation clock, which is a travelling oscillator that interacts with a maturation wave called the wavefront to produce the periodic series of somites. Here, we review our current understanding of the segmentation process in vertebrates. R E V I E W S 370 | MAY 2008 | VOLUME 9 www.nature.com/reviews/genetics Amniotes Group of tetrapod vertebrates including mammals, reptiles and birds, the embryo of which is protected by a membrane called the amnion, in particular from dehydration. Epiblast Tissue precursor of the three germ layers during gastrulation. Metamery A segmented organization of the body plan along the anterior?posterior axis. The segmentation clock oscillator During the past three decades, the fly embryo has been the major paradigm for the study of segmentation. The pioneering screens for segmentation defects in Drosophila melanogaster led to a thorough understanding of the molecular cascade controlling the establishment of the segmental pattern in this organism 12,13 . The proc- ess is initiated by gradients of maternal gene products, such as bicoid and nanos, which then activate a series of zygotic gap genes such as hunchback and Kruppel, the expression domains of which are sequentially organized in broad stripes along the AP axis of the embryo. The combinatorial expression of the gap genes then results in the periodic expression of the pair-rule genes, which include hairy, even-skipped and runt, in seven alternating domains that pre-figure the 14 embryonic segments. The combinatorial expression of the pair-rule genes, in turn, activates the segment polarity genes (such as engrailed, wingless and hedgehog) that establish the definitive segmental pattern of the embryo. However, in contrast to the fly, most other segmented species add segments sequentially as the embryonic axis progressively elongates posteriorly (FIG. 1). Such a rhythmic mode of segment formation is observed in vertebrates. It inspired theoretical models such as the ?clock and wavefront? model 14 , which proposed that PSM cells forming a somite undergo an abrupt change in cellular properties that could be formalized by a par- ticular type of mathematical catastrophe (FIG. 2a). Such a catastrophe can be explained by a bistable transition between two steady states, allowing an abrupt switch from one particular state to another 15 . To account for the periodic occurrence of the catastrophe, the existence of an oscillator that controls the response of PSM cells to the mechanism triggering the catastrophe was pos- tulated (FIG. 2a). A number of subsequent models were proposed, many of which also relied upon the conver- sion of a temporal oscillation into a spatial periodic pattern 16,17 . The first evidence of the existence of an oscillator coupled to somitogenesis was provided by the periodic expression of the mRNA encoding the transcription fac- tor HES1 (hairy and enhancer of split 1) in the chicken embryo PSM 18 . During the formation of each somite, the PSM is swiped by a dynamic wave of HES1 mRNA expression (FIG. 2b). These transcriptional oscillations of HES1 that occur with the same periodicity as the somi- togenesis process were proposed to identify a molecular oscillator ? termed the segmentation clock ? acting in PSM cells. Subsequently, several other genes exhibit- ing such a cyclic behaviour were identified in fish, frog and mouse, indicating that the oscillator is conserved in vertebrates 19?22 . These genes are now referred to as cyclic genes and, as we shall see below, the vast major- ity of them belong to the Notch, Wnt and fibroblast growth factor (FGF) signalling pathways. Much of the recent research in the vertebrate segmentation field has focused on the identification of the pacemaker that triggers the rhythmic expression of the cyclic genes in the PSM. This has led to several hypotheses that are discussed in this Review. The zebrafish oscillator In zebrafish, all the cyclic genes identified so far belong to the Notch pathway and comprise the Notch downstream targets: her1 (hairy and enhancer of split-related 1), her7, her11, her12 and her15 (which are homologous to the chicken HES1) 19,23?28 , as well as the Notch ligand DeltaC 21 . Large genetic screens carried out in zebrafish have identified a handful of mutants in which somito- genesis is disrupted 29 . These mutants show alterations Nature Reviews | Genetics SII SI S0 S?I S?II Rostrocaudal patterning Segmental determination Paraxial mesoderm production Axis elongation Somites Otic vesicle Neural tube Head mesoderm Head PSM Tail bud Tail Somite formation Neural folds Figure 1 | Establishment of segmentation during embryogenesis. Dorsal view of a 4-week-old human embryo showing somites and the presomitic mesoderm (PSM) forming the paraxial mesoderm that flanks the axial neural tube. The different phases of paraxial mesoderm patterning leading to somite formation are indicated as: paraxial mesoderm production from the progenitor pool localized in the tail bud, segmental determination, rostro-caudal patterning and somite formation. The anterior-most somites give rise to the basi-occipital bone of the skull and to the anterior-most cervical vertebrae. The subsequent somites generate the vertebral column. Prospective somites in the PSM are numbered in a rostrocaudal series beginning with somite S0, which is the forming somite, in negative roman numerals (for example, ?I, ?II) 132 . The segmented somites (for example, SI, SII) are numbered according to Ordahl 133 . Arrows indicate the movement of paraxial mesoderm cells from the tail bud into the PSM. R E V I E W S NATURE REVIEWS | GENETICS VOLUME 9 | MAY 2008 | 371 Morpholinos Synthetic molecules of antisense oligonucleotides used for gene expression knock-down. in genes that encode components of the Notch path- way, such as the Notch1A receptor, the ligands DeltaD and DeltaC, and the Mindbomb ubiquitin ligase that is required for Delta endocytosis and Notch activa- tion 19,30?32 . In these mutants, the dynamic wave of cyclic gene expression in the PSM is disrupted and replaced by a salt-and-pepper expression pattern 21 . This char- acteristic expression pattern of the cyclic genes in the Notch mutants was proposed to reflect desynchronized oscillations in PSM cells 21 , suggesting a role for Notch signalling in the synchronization of the oscillations among PSM cells (BOX 1). Experiments in zebrafish embryos in which the function of the Her genes was blocked by mutation or by morpholino knock-down, or constitutively activated by overexpression 19,24,25,30,33?37 led to a simple oscillator model that essentially relied on the Her1 and Her7 transcriptional repressors. In this model, oscillations are generated by a negative feedback loop in which the Her genes are directly repressed by their own protein prod- ucts 36 (BOX 1). To generate oscillations, the model takes into account a defined time delay in the auto-inhibitory circuit that occurs from the beginning of transcription of the Her RNA until the Her protein binds to the Her gene promoter. Mathematical modelling showed that oscillations can be sustained, but only if the half-lives of the gene transcripts and proteins are short compared with the sum of the transcriptional and translational delays 36 . Using plausible numerical values for the model parameters, oscillations exhibiting a period consistent with that observed in zebrafish could be obtained 36 . Several of the kinetic parameters (such as transcrip- tional delays and the stability of the RNAs and proteins) that were initially roughly estimated, were subsequently validated in the embryo and shown to be consistent with the estimated values 35 . Surprisingly, whereas a com- plete disruption of segmentation would be expected in zebrafish mutant embryos lacking both her1 and her7, or in embryos injected with her1 and her7 morpholinos, these embryos still form abnormal somites 25 . However, redundancy with the other Her genes could account for this surprisingly mild phenotype. This Her1?Her7 intracellular oscillator was proposed to be linked to an intercellular oscillator involving the Notch signalling pathway. Her1 and Her7 negatively regulate deltaC, thus potentially triggering oscillations of this Notch ligand (BOX 1) that should, in turn, result in periodic Notch activation in neighbouring cells 24,31 . This coupling provides a basis for maintaining the synchrony between oscillations of neighbouring cells 21 (BOX 1). It remains unclear whether the Notch pathway is required for Her oscillations. The fact that the first oscillatory cycles require Her genes but not the Notch pathway 38 argues that Notch is not part of the clock pacemaker. Accordingly, Her gene expression is not abolished in the Notch pathway mutants or when treating embryos with the ?-secretase inhibitor DAPT (N-[N-(3,5-difluorophenacetyl)-l.sc-alanyl]-S-phenyl glycine t-butyl ester) 19,24,38 . However, a constitutively active form of Notch1A results in overexpression of her1 in the PSM 33 , suggesting that these genes are targets of the Notch pathway. This is also consistent with the observation that in many biological systems, genes of the Her family of transcription factors function as down- stream targets of Notch 39 . Furthermore, it has been pro- posed that the Notch ligand DeltaD is required for the initiation of oscillations in the tail bud 40 . The Her1?Her7 oscillations require the Hes6-related gene, Her13.2 (REF. 41). Expression of this transcription factor is regulated by FGF signalling. It can form a heterodimer with Her1, enhancing the ability of Her1 to negatively regulate its own promoter 41 (BOX 1). Thus, although in fish the Her1?Her7 negative feedback loop might constitute the core of the segmentation clock pacemaker, it requires additional input from several signalling pathways 42 . Nature Reviews | Genetics Expression waveb a Oscillator De v e lopment s T SI S0 SII S0 SI Figure 2 | The clock and wavefront model and the segmentation clock oscillator. a | Topological representation of the somitogenesis model showing a section of the embryonic axis (corresponding to the posterior part of the embryo, including the presomitic mesoderm (PSM) and the few somites most recently formed) plotted in real space (s) (in head?tail axis), real developmental time (T) (that is, indicating the onset of each somitic cycle) and a dimension representing intracellular development (the vertical arrows falling from the fold-edge symbolize the catastrophe and correspond to abrupt cellular changes associated with somite formation). The oscillator (circle) was postulated to ensure the periodic occurrence of the catastrophe (in the fold- edge, along the dashed line) that corresponds to an abrupt transition between two cellular states. b | Evidence for an oscillator underlying vertebrate segmentation. Periodic waves of transcriptional expression of the hairy1 gene (blue) in PSM cells are associated with the formation of each pair of somites added sequentially 18 . Part a modified with permission from REF. 14 � (1976) Elsevier Ltd. R E V I E W S 372 | MAY 2008 | VOLUME 9 www.nature.com/reviews/genetics The amniote oscillator The first cyclic genes identified in amniotes also belong to the Notch pathway 18 . As in zebrafish, oscillations of the hairy and enhancer of split (called Hes in amniotes) homologues are detected in chicken (HES1, HAIRY2 and HEY2) and in mouse (Hes1, Hes7, Hes5 and Hey1) 18,43?46 . Oscillations of the Notch ligand-encoding Dll1 (delta-like 1) have also been reported in mouse 47 . Cyclic expression of other Notch pathway genes such as Lfng (lunatic fringe), which is a glycosyl-transferase that modifies the Notch receptor, is detected only in amniotes 20,48,49 and not in lower vertebrates 50 . As the first cyclic genes identified in amniotes were targets of the Notch pathway, the first molecular models placed Notch signalling as a central component of the seg- mentation clock 51,52 . This idea was supported by the observation that Lfng and Hes1 expression was lost in the PSM of mouse Notch pathway mutants such as Rbpjk (recombination signal binding protein for immuno- globulin kappa J region) and Dll1 homozygotes 43,53 . Furthermore, periodic expression in the mouse PSM of the cleaved activated intracytoplasmic form of the Notch1 receptor is detected by immunohistochemistry, providing direct evidence for rhythmic activation of the pathway 54,55 . In the chicken embryo, LFNG is periodi- cally activated by Notch signalling in the PSM and, in turn, inhibits Notch signalling, thereby establishing a negative feedback loop 56 (FIG. 3). The expression of the LFNG protein cycles with the same period as somi- togenesis in the PSM, indicating a short half-life 56 . In mouse Lfng-null mutants, constitutive expression of the activated form of Notch is detected in the PSM, Nature Reviews | Genetics Her13.2 Her1?Her7 Her11?Her12? Her15 DeltaC DeltaC DeltaC Notch1A DeltaC Notch1A Notch1A Notch1A The zebr a f i s h o s c i l l a t or FGF SI PSM S0 DeltaCFGFR1 Box 1 | Synchronization of the presomitic mesoderm (PSM) cellular oscillators A remarkable property among neighbouring PSM cells is their synchronized oscillations, resulting in the smooth transcriptional wave that sweeps through the PSM (cyclic gene expression is shown in blue in PSM cells that are represented as octagons). In zebrafish, the role of the Notch pathway in synchronizing oscillators was first proposed based on the typical salt-and-pepper expression pattern of cyclic genes in PSM cells of Notch segmentation mutants 19,21 , interpreted as desynchronized oscillations 21 . In these mutants in which the first somites segment normally, oscillations would be set initially, but owing to the lack of Notch-dependent coupling, they would progressively drift out of synchrony, resulting in segmentation failure 21,38 . The role of the Notch pathway in coupling oscillators has been mathematically modelled by connecting the zebrafish Her1 (hairy and enhancer of split-related 1)?Her7 intracellular oscillator to the Notch?DeltaC (DeltaC is a Notch ligand) intercellular loop 35,36,130 . The transcription factors Her1 and Her7 establish a negative feedback loop that is proposed to function as the zebrafish clock pacemaker and to control the periodic repression of DeltaC, allowing the synchronous activation of Notch signalling in neighbouring cells. In addition to receiving inputs from Notch signalling, the Her1?Her7 oscillator requires the Her13.2 partner, which is downstream of FGF (fibroblast growth factor) signalling. Experimental evidence supports the role of cell?cell communication through the Notch?DeltaC loop in maintaining synchronized oscillations among nearby cells. Implantation of cells from a zebrafish embryo overexpressing DeltaC can desynchronize the waves of cyclic gene expression, resulting in the shifting of somitic boundaries on the injected side 128,129 . Consistently, dissociation of the PSM cells rapidly results in a loss of synchronized oscillations in chicken 130 and mouse 131 . The coupling between cellular oscillators provided by the Notch?Delta intercellular loop is thought to confer robustness to the synchronized clock oscillations 38 (E. Ozbudak and J. Lewis, personal communication) against developmental noise such as cell proliferation, cell movement or stochastic gene expression. FGFR1, FGF receptor 1. R E V I E W S NATURE REVIEWS | GENETICS VOLUME 9 | MAY 2008 | 373 supporting a conserved role for this negative feed- back loop in amniote segmentation 55 (FIG. 3). However, constitutive overexpression of Lfng in the PSM of transgenic mice does not prevent oscillations of the endogenous Lfng or Hes7 (REF. 57), suggesting that this negative feedback loop is, by itself, insufficient to drive the oscillatory network. In mouse, the HES protein family has a key role in the control of oscillations by implementing a negative feedback loop similar to the zebrafish Her-based loop 58 . The genes Hes7, Hes1, Hes5 and Hey2 encode transcrip- tional repressors that are expressed periodically in the mouse PSM 43?46 . The Hes1, Hes5 and Hey2-null mutants do not show any somitic phenotype, but in Hes7-null mutants somitogenesis is defective and oscillations of cyclic genes such as Lfng are disrupted 59?62 . Furthermore, transcription of both Hes7 and Lfng is upregulated in the Hes7 mutants, consistent with the idea that Hes7 represses its own transcription in the PSM as well as that of Lfng 58 (FIG. 3). Analysis of the regulatory region of the Lfng promoter confirmed that it includes CSL (CBF1, Su(H), LAG1) binding sites that are required for regulation by Notch signalling 63 , as well as binding sites for bHLH (basic helix-loop-helix) proteins, such as members of the Hes family 64 . Deletion of these regula- tory sites (clock elements) blocks Lfng oscillations in the posterior PSM 65 . This leads to a ubiquitous activation of Notch in the PSM and to severe vertebral anomalies. Strikingly, however, this phenotype is not observed in the sacral (posterior-most region of the trunk) and caudal (tail) region, suggesting that the clock regulation might vary along the AP body axis. Lfng Hes1?7?5?Hey Nkd1 Nrarp Nature Reviews | Genetics Dusp6/4 Sprouty2 Snail1/2 Shp2 Axin2 Dact1 Dkk1 Myc Sp5 Tnfrsf19 Perlecan Axin2 ?-catenin ERK SHP2 DUSP6 FGF DLL1 DLL1 Notch1 LFNG LRP6 DKK1 Frizzled Wnt FGFR1 Perlecan Sprouty2MEK GSK3 APC DACT1DSH NKD1FGF Notch Wnt Ras Raf FRS2SOS GRB2 NICD Figure 3 | The mouse oscillator. Cyclic genes belonging to the Notch and FGF (fibroblast growth factor) pathways (the products of which are indicated in red) oscillate in opposite phase to cyclic genes of the Wnt pathway (blue). A large number of the cyclic genes are involved in negative feedback loops. The basic circuitry of the three signalling pathways is represented. Dashed lines correspond to modes of regulation inferred from work in other systems or based on microarray data 70 . APC, adenomatous polyposis coli; DACT1, dapper homologue 1; DKK1, dickkopf homologue 1; DLL1, delta-like 1; DSH, dishevelled; DUSP6, dual specificity phosphatase 6; ERK, mitogen-activated protein kinase 1; FGFR1, FGF receptor 1; GRB2, growth factor receptor-bound protein 2; GSK3, glycogen synthase kinase 3; Hes1, hairy and enhancer of split- related 1; LFNG, lunatic fringe; LRP6, low density lipoprotein receptor-related protein 6; MEK, mitogen-activated protein kinase kinase 1; NICD, Notch intracellular domain; NKD1, naked cuticle 1 homologue; Nrarp, Notch-regulated ankyrin repeat protein; SHP2, Src homology region 2-containing protein tyrosine phosphatase 2; SOS, son of sevenless; Sp5, trans-acting transcription factor 5; Tnfrsf19, tumour necrosis factor receptor superfamily, member 19. R E V I E W S 374 | MAY 2008 | VOLUME 9 www.nature.com/reviews/genetics Mathematical models, based on a delayed negative feedback loop that is controlled by Hes-family members similar to those proposed to drive fish cyclic gene oscil- lations, have been proposed for the mouse embryo 66,67 . In these models, the production of oscillations directly depends on the half-life of the Her/Hes RNA and protein. The rapid clearing of the cyclic genes from the PSM is consistent with a short half-life of the mRNA of the cyclic genes. The mouse HES7 and HES1 proteins also have a short half-life that is actively controlled by the protea- some and estimated to be approximately 22 minutes in vitro 58,66,68 . To test the importance of protein stability in the generation of oscillations, the half-life of HES7 was genetically modified in the mouse embryo. A point muta- tion conferring a small half-life increase from 22 to 30 minutes leads to a failure of Hes7 oscillations and somite segmentation, as predicted by the time-delay model 66 . Unlike in zebrafish, several other signalling path- ways show periodic activity in the mouse PSM during the segmentation clock cycle. Identification in mouse of oscillations of Axin2, a key negative feedback inhibitor of the Wnt pathway, together with the observation that in mice with the hypomorphic Wnt3a mutation vestigial tail (vt) both Axin2 and Lfng oscillations are disrupted, suggested the implication of Wnt signalling in the oscil- lator mechanism 69 . A microarray-based approach led to the identification of a much larger number of cyclic genes that are associated with the segmentation clock in the mouse PSM transcriptome 70 (FIG. 3). The cyclic genes that were identified fell into two main clusters of gene expression profiles that oscillate in antiphase (FIG. 3). One cluster contained a large number of cyclic genes, most of which are linked to the Wnt signalling pathway (shown in blue in FIG. 3). These cyclic genes correspond to downstream targets of the Wnt pathway such as the transcription factors SP5 (trans-acting transcription factor 5) 71 and MYC (myelocytomatosis oncogene) 70,72 . These also include negative feedback inhibitors such as AXIN2 (REF. 69) and DKK1 (dickkopf homologue 1) 73 . Although Axin2 ?/? mutants do not exhibit any somitic phenotype 74 , inactivation of many of the newly identi- fied cyclic genes in the Wnt cluster (Dkk1 (REF. 75), Sp5 (REF. 76), Myc (REF. 77) and has2 (hyaluronan synthase 2) 78 ) produces segmentation defects, supporting a role for Wnt signalling during segmentation in mouse. Wnt activation in cells results in stabilization of ?-catenin, which in turn enters the nucleus to activate the expression of target genes. Therefore, oscillations of Wnt inhibitors such as DKK1 or DACT1 (dapper homologue 1) should, in principle, result in the rhyth- mic fluctuation of ?-catenin expression levels. As Wnt signalling has been shown to function upstream of both Wnt and Notch oscillations 69 , such ?-catenin oscilla- tions seem to be a good candidate for a pacemaker that entrains the Notch signalling loop. Accordingly, Notch and Wnt cyclic gene oscillations are lost in a conditional deletion of ?-catenin in the PSM 79 . However, in a mouse mutant in which ?-catenin is made constitutively stable in the PSM, expression of Wnt and Notch pathway genes still oscillates 79,80 , indicating that ?-catenin signalling is necessary but insufficient for driving the expression of cyclic genes. Therefore, these experiments argue against a role for the periodic destabilization of ?-catenin in the control of clock oscillations. The second cyclic gene cluster that was identi- fied in the microarray study 70 contains known cyclic genes of the Notch pathway as well as other genes of this pathway that had not been previously associated with the oscillator (FIG. 3). These include Nrarp (Notch- regulated ankyrin repeat protein), which is a direct target of Notch signalling that functions as a negative regulator of the Notch pathway 81?83 . In parallel, NRARP can also positively regulate the Wnt pathway by stabiliz- ing the transcription factor LEF1 (lymphoid enhancer binding factor 1) 84 . The previously identified cyclic gene naked cuticle 1 homologue (Nkd1), an inhibitor of Wnt signalling regulated by Notch signalling 85 , was also iden- tified as part of the Notch cluster (FIG. 3). These genes might provide a functional link between the Notch and Wnt arms of the oscillator. A novel class of cyclic genes involved in FGF sig- nalling and oscillating in phase with the Notch cyclic genes was identified in the same microarray study 70 . Two negative feedback inhibitors of the FGF pathway, Spry2 (sprouty homologue 2) and Dusp6 (dual specifi- city phosphatase 6), show a clearly periodic profile in the array series (FIG. 3). The FGF targets Snai1 (snail homologue 1), in mouse, and SNAI2, in chicken 86 , as well as the Dusp4 negative feedback inhibitors of the FGF pathway 87 , also exhibit periodic expression in mouse and chicken PSM. Furthermore, periodic phosphorylation of ERK (extracellular signal-regulated kinase) in the mouse PSM supports periodic FGF signalling activity 87 . Despite the synchronized activation of Notch and FGF activity in the clock cycle, expression of FGF- regulated genes seems to be largely independent of Notch signalling as Spry2 expression is still dynamic in Rbpjk ?/? mutant mice 70 . Furthermore, Dusp4 expression is still cyclic in Lfng ?/? , Dll1 ?/? and Rbpjk ?/? mutants as well as in mouse PSM treated with DAPT 87 . However, conditional deletion of Fgfr1 (FGF receptor 1) in the PSM blocks the oscillations of the FGF, Notch and Wnt pathways in the PSM 87,88 . Moreover, whereas treatment of mouse tail explants with the FGFR1 inhibitor SU5402 quickly abol- ishes cyclic expression of Axin2 and Spry2, it only blocks Lfng oscillations after a one-cycle delay, indicating that FGF indirectly regulates the Notch oscillations 88 . The requirement of Wnt signalling for Lfng oscillations 89,90 suggests that FGF functions upstream of Wnt signalling, which, in turn, controls Notch oscillations. In mouse, the Notch target Hes7 also requires FGF signalling 87 . FGF was proposed to be required to initiate Hes7 pulses of expression in the tail bud, whereas propagation of Hes7 oscillations in the more anterior PSM requires Notch signalling 87 . Such a two-step model is consistent with the observed delay of Lfng oscillation inhibition when FGFR1 is inhibited. However, Lfng stripes are observed in the PSM of a mouse conditional Fgfr1 ?/? mutant with a constitutively stable form of ?-catenin 80 . Therefore, constitutive ?-catenin restores Notch oscillations in the absence of FGF signalling, arguing against a role for FGF as the periodic input controlling Wnt R E V I E W S NATURE REVIEWS | GENETICS VOLUME 9 | MAY 2008 | 375 Mesenchyme Tissue consisting of loosely packed cells. Basal lamina A layer of extracellular matrix that underlies the epithelium and is secreted by the epithelial cells. oscillations. Therefore, these experiments argue that nei- ther periodic Wnt nor FGF signalling triggers the rhyth- mic expression of cyclic genes such as Lfng. Together with observations indicating that Axin2 oscillations are maintained in Notch pathway mutants 69 , this suggests that none of the three oscillating pathways of the seg- mentation clock functions as a pacemaker. Therefore, it cannot be ruled out that these oscillatory networks of signalling genes correspond only to outputs of an as yet unidentified pacemaker. On the other hand, the evolu- tionary conservation of the role of the Her/Hes proteins is consistent with their potential role in the pacemaker of the segmentation clock. A striking feature of the cyclic gene network in amni- otes is its apparent redundancy. Among mouse cyclic genes are several negative feedback inhibitors for each of the Notch, FGF and Wnt pathways (FIG. 3) that potentially account for the robustness of the network. Other cyclic genes are likely to be just downstream targets of the clock and could merely represent outputs of the oscillator. The role of the segmentation clock might be to deliver coordinated pulses of Notch, FGF and Wnt signalling that are, in turn, used for the appropriate patterning of the seg- ments. Below we discuss how this periodic signalling is translated into a coordinated striped gene activation that defines the segmental domain and its boundaries. Translating the clock pulse into segments In the original clock and wavefront model, Cooke and Zeeman postulated the existence of a front of maturation ? the wavefront ? that slowly moves posteriorly along the embryo 14 . When PSM cells in the permissive phase of the clock oscillation cycle are passed by the wave- front, they undergo an abrupt transition (a catastrophe) that leads to somite formation (FIG. 2a). Therefore, the wavefront serves to translate the rhythmic pulse of the clock into the spatial periodic series of segments. In the Cooke and Zeeman model, the wavefront was positioned at the anterior-most level of the PSM where somites form. This positioning was subsequently challenged by heat-shock experiments in Xenopus 91 that identified and positioned the hidden wavefront of cellular change more posteriorly than initially proposed. In chicken embryos, microsurgical inversions of small fragments of the PSM along the AP axis demonstrated the existence of a virtual boundary ? called the determination front ? in this tissue 92 . This boundary separates the posterior PSM domain (where inverted blocks of cells form seg- ments according to their new position) from an anterior domain (where cells are committed to their original segmental fate). Therefore, the determination front was defined as the level at which PSM cells first acquire their segmental identity, and is therefore conceptually similar to the wavefront. At the molecular level, the position of the determination front corresponds to the posterior boundary of the Mesp2 (mesoderm posterior 2) stripe that marks the first evidence of a segmental prepattern in the PSM. Interestingly, this transition in the PSM also cor- responds to a morphological transition at the cellular level. Whereas the posterior PSM is a loose mesenchyme, the cells located anterior to the determination front become progressively epithelialized. This transition is also accompanied by a slowing down of PSM cell move- ments 93 . The mesenchymal?epithelial transition that occurs concomitantly with the segmental patterning of the anterior PSM correlates with a downregulation of the Snai genes that are regulated by FGF signalling and are associated with a mesenchymal state in many systems 86 . Downregulation of Snai genes at the deter- mination front correlates with the expression of several adhesion molecules such as integrins and cadherins, which progressively increases in the anterior PSM as cells become polarized 94?96 . This transition is also accom- panied by the deposition of a basal lamina, containing laminin and fibronectin, that surrounds the anterior PSM 95 . This epithelialization process requires the bHLH transcription factor TCF15 (transcription factor 15, also known as Paraxis) that is expressed anteriorly to the determination front in the anterior PSM and somites 97 . TCF15 controls the activity of Rho GTPases, such as RAC1 (Ras-related C3 botulinum substrate 1) and CDC42 (cell division cycle 42 homologue), which have been shown to mediate the mesenchymal?epithelial transition during somite formation 98 . The position of the determination front is defined by specific thresholds of FGF and Wnt signalling activi- ties 69,92,99 (FIG. 4). The FGF gradient was first described as a posterior-to-anterior gradient of Fgf8 mRNA in the PSM of chicken, fish and mouse embryos 92,99,100 . This mRNA gradient is subsequently translated into a protein gradient and then into a MAPK (mitogen- activated protein kinase)?AKT activity gradient along the PSM 93,99,100 . The role of the FGF signalling gradient in positioning the determination front was first dem- onstrated by experiments that perturbed the slope of the gradient in chicken embryos. This was achieved by grafting FGF8-soaked beads next to the PSM or by over- expressing an FGF8-expressing construct in the PSM by electroporation 92 . This resulted in an anterior extension of posterior PSM markers, such as Brachyury, and in downregulation of segmentation and differentiation markers such as paraxis, Mesp2 and Myod (myogenic differentiation 1) 92,93 . Fgf8 loss-of-function mutations have proved to be more problematic to interpret owing to redundancy in the FGF pathway. Neither the zebrafish fgf8 mutant ace 101 nor the conditional deletion of Fgf8 in the mouse PSM 102 shows a segmentation phenotype, suggesting that FGF8 is not the only ligand involved in setting the FGF-signalling gradient along the PSM gradient. Fgf3, Fgf4 and Fgf18 are expressed in the PSM and tail bud region of the mouse embryo 88 . In zebrafish, fgf8 func- tions redundantly with fgf24 to promote the formation of the posterior PSM 103 . Inhibition of FGF signalling was achieved by treating chicken embryos with phar- macological inhibitors 92 . This resulted in a posterior shift of the anterior boundary of the expression domain of genes such as FGF8 that are associated with a poste- rior identity. Such a posterior shift was also observed for the expression domain of posterior markers Fgf8 and Msgn1 (mesogenin 1) in mouse mutant embryos R E V I E W S 376 | MAY 2008 | VOLUME 9 www.nature.com/reviews/genetics with a conditional deletion in the paraxial mesoderm of FGFR1, the only FGF receptor that is expressed in the PSM 88 . Together, these data suggest that high levels of FGF signalling are required to maintain the poste- rior identity of PSM cells 92 . This further led to the idea that the progressive decrease in FGF signalling activity along the PSM defines a specific threshold below which the cells become competent to respond to the signal- ling pulse that is delivered by the segmentation clock (FIG. 4). The position of this threshold was proposed to correspond to the determination front 92 . Wnt genes, such as Wnt3a, are expressed in the tail bud and posterior PSM 69 . Furthermore, a gradient of nuclear ?-catenin extends from the tail bud to the determination front 80 . Together with the graded expres- sion of Wnt targets, such as Axin2, along the PSM, they identify a posterior-to-anterior Wnt signalling gradient in the PSM parallel to the FGF gradient 69 . Fgf8 expres- sion is absent in Wnt3a mutants, indicating that Wnt signalling is required for the expression of FGF ligands in the PSM 69 . However, only a partial FGF gain of func- tion is observed in the PSM of mouse embryos over- expressing a constitutively stable ?-catenin, suggesting that Wnt signalling is insufficient for Fgf8 expression in the PSM 80 . The posterior ?-catenin gradient was recently shown to define the size of the oscillatory field in the PSM, thereby controlling the position of the wavefront where the oscillations stop 80 . Furthermore, ?-catenin gain of function in the PSM prevents activa- tion of MESP2 targets, indicating that downregulation of Wnt signalling at the determination front is required for normal segmentation to proceed 79,80 . Whereas both FGF and Wnt signalling are character- ized by posterior-to-anterior gradients of activity in the posterior PSM, some of their targets exhibit an oscilla- tory expression, which seems paradoxical. The ?-catenin gain-of-function experiments demonstrate a role for the nuclear ?-catenin gradient in the control of the matura- tion of cells along the PSM, but they indicate that oscil- lation of Wnt targets, such as Axin2, results from an oscillating input that is independent of ?-catenin and FGF signalling 80 . Such an input could be provided by a pacemaker that is external to the cyclic gene network. Although similar gain-of-function experiments remain to be carried out for FGF signalling, such a pacemaker could also function on the FGF cyclic genes and explain the coexistence of a graded signal and an oscillatory response. Retinoic acid (RA) 104?107 was also proposed to have a role in positioning the determination front as an anterior- to-posterior gradient of RA opposing the Wnt?FGF gra- dient 108,109 (FIG. 4). RALDH2, the RA biosynthetic enzyme, is expressed in the anterior-most PSM and segmented region, and is excluded from the tail bud and posterior PSM 110 . Using a RARE (RA response element)?LacZ reporter mouse, RA signalling was found to be restricted to the anterior PSM and segmented region, and absent from the posterior PSM and tail bud 104 where CYP26, an enzyme of the cytochrome P450 family that is involved in RA degradation, is expressed downstream of FGF 111 . In chicken, treatment of posterior PSM explants with RA agonists can downregulate FGF8 expression, and a graft of an FGF8-soaked bead in the PSM represses RALDH2 expression in the embryo 108 . Furthermore, in Raldh2 Nature Reviews | Genetics Somitogenesis Determination front Axis extension Retinoic acid FGF/Wnt T=0 T=1 cycle T=2 cycles Figure 4 | A somitogenesis model integrating the segmentation clock and determination front. The system of opposing gradients of FGF (fibroblast growth factor)?Wnt signalling (purple) and retinoic acid (green) signalling was proposed to position the determination front (black line) along the presomitic mesoderm (PSM). This particular level is characterized by a signalling threshold at which the cells become competent to respond to the segmentation clock signal and is conceptually similar to the wavefront of the original Cooke and Zeeman model 14 . The clock signal is still poorly characterized but probably involves three signalling pathways experiencing periodic activity: FGF, Wnt and Notch. The wave of cyclic gene expression controlled by the segmentation clock oscillator is shown in orange on the left side of the embryos. When competent cells that pass through the determination front receive the clock signal, they simultaneously activate Mesp2 (mesoderm posterior 2; shown in black), thereby defining the future segmental domain as shown on the right side of the embryos. In this model, the size of the segment (the future somite) is defined by the distance travelled by the wavefront during one oscillation of the segmentation clock. However, the role of retinoic acid in this model remains debated (see text). During the next cycle Mesp2 expression becomes restricted to the anterior compartment of S?I (grey). T, time in segmentation clock cycle unit. R E V I E W S NATURE REVIEWS | GENETICS VOLUME 9 | MAY 2008 | 377 mouse mutants and in chicken or quail embryos that are deprived of RA, the Fgf8 expression domain is extended along the PSM 105,108 . A similar antagonistic action of FGF and RA gradients was also observed in Xenopus, suggest- ing that this gradient system is conserved among verte- brates 109 . These experiments led to a proposal that the mutual inhibition of the FGF and RA gradients has a role in positioning the determination front. However, this is difficult to reconcile with the observation that somites do form in the Raldh2 mouse mutant in which no RA signalling is detected 112 , suggesting that RA signalling is dispensable for somite formation. Furthermore, in the Fgfr1 conditional knock-out, no significant posterior shift of the RARE?lacZ domain is observed, suggesting that FGF is not the only antagonist of the RA gradi- ent 88 . Whether the posterior Wnt gradient, by itself, can antagonize RA signalling remains to be investigated. RA has also been implicated in the control of the symmetry of the somitogenesis process 104?107 . Somite formation is asymmetric in embryos that are deprived of RA. This lateralized desynchronization of somitogen- esis occurs in response to the activation of the left?right machinery that is involved in the asymmetric positioning of the internal organs. RA prevents the paraxial meso- derm from responding to the asymmetric signal down- stream of Nodal, thereby maintaining the symmetry of the somitogenesis process at early stages. In the embryo, the segmentation process is tightly coordinated with axis elongation through the formation of the FGF signalling gradient in the posterior PSM 113 . As the embryo grows posteriorly owing to axis elonga- tion, new cells enter the posterior PSM, compensating for the loss of the anterior PSM cells that form somites. Transcription of the Fgf8 mRNA is restricted to the PSM precursors in the tail bud, and it ceases when their descendents enter the posterior PSM. Therefore, as the axis elongates, cells become located progressively more anteriorly in the PSM and their Fgf8 mRNA content progressively decays. This results in the establishment of an Fgf8 mRNA gradient that is converted into a graded distribution of ligand and FGF activity 93,99,113 . A similar mechanism is assumed to be responsible for establishing the Wnt gradient 89 . As a result of the progressive decay of FGF?Wnt family mRNA and proteins in PSM cells, the determination front is constantly displaced posteriorly, and the speed of this displacement defines the speed that somitogenesis progresses along the AP axis (FIG. 4). This mechanism ensures a tight coordination between axis elongation and segmentation during embryogenesis. A clock and wavefront-based model Experimental perturbations of the shape of the FGF, Wnt and RA gradients led to specific somite defects that revealed how the clock and the wavefront could interact. In zebrafish and chicken embryos, FGF- and Wnt-bead graft experiments shift the determination front anteri- orly and lead to the formation of smaller somites 69,92,99 . Similarly, RA loss of function in mouse Raldh2 mutants or in RA-deprived chicken or quail embryos results in an FGF gain of function in the PSM, leading to the same phenotype 104,105,108 . Conversely, inhibition of FGF signalling shifts the determination front posteriorly, resulting in the formation of larger somites 92,99 . Conditional deletion of Fgfr1 in the PSM also results in transient formation of larger somites followed by disruption of segmentation 88 . These results gave rise to a new segmentation model integrating the original clock and wavefront concepts 14 . In this model, the wavefront corresponds to the trav- elling determination front, defined as a threshold of Wnt?FGF?RA signalling, the position of which moves posteriorly and accompanies the posterior regression of these gradients. During one segmentation clock oscil- lation, the determination front moves posteriorly along the AP axis by a distance that corresponds to approxi- mately one somite (FIG. 4). We proposed that when PSM cells are passed by the determination front, they become competent to respond to a periodic signal delivered by the segmentation clock 92 . In response to this signal, the cohort of cells located between the determination front and the posterior boundary of the segment that was determined in the previous segmentation cycle simultaneously activate Mesp2, resulting in the forma- tion of a stripe of Mesp2 expression that prefigures the future segment. Once expressed, Mesp2 stabilizes Lfng expression in the newly formed striped domain, leading to an inhibition of Notch signalling in this territory 55 . Because Notch is activated in the posterior part of the segmental domain that is located immediately anterior to the Mesp2?Lfng domain 114 , this mechanism gener- ates an interface between cells activating and repress- ing Notch 55 . This interface marks the level of the future somite boundary. During the next oscillation cycle, the newly specified segmental domain becomes located more anteriorly in the PSM by one somite. Cells in this territory begin to activate a complex genetic pro- gramme downstream of MESP2. These cells activate the transcriptional repressor Ripply1 that establishes a negative feedback loop shutting down Mesp2 expression in the future posterior compartment 115,116 . MESP2 also activates the expression of genes, such as Epha4 (Eph receptor A4), that are involved in boundary forma- tion 117 . This complex genetic cascade ultimately results in the specification of the anterior and posterior somite compartments and of the somite boundaries. A striking feature of the segmental patterning proc- ess is the highly synchronized periodic gene activation that occurs in the stripes of cells that define the future segments. The signalling pulse that is delivered by the segmentation clock is a good candidate to trigger this periodic gene activation. The synchronization of the response to this signal in the future segment was pro- posed to reflect a molecular switch that simultaneously triggers Mesp2 expression in the cohort of competent cells that passed the determination front 15 . Mathematical modelling shows that the mutual inhibition of FGF and RA signalling can define a bistability domain along the PSM in which such a switch behaviour can be observed 15 (FIG. 5). In response to an appropriate signal, cells located in the bistability domain (in fact, in the area between the determination front and the last specified segment) can abruptly switch from the FGF-dominated steady R E V I E W S 378 | MAY 2008 | VOLUME 9 www.nature.com/reviews/genetics Suprathreshold stimulation Stimulation of sufficient strength to produce a perceptible effect; in the current context a catastrophe leading to somite determination. Urbilateria Hypothetical last common ancestor of all bilaterians. Bilateria Members of the animal kingdom that have bilateral symmetry ? the property of having two similar sides, with definite upper and lower surfaces, and anterior and posterior ends. Ultradian oscillator Oscillator with a period of less than 24 hours. state to the other, RA-dominated steady state 15 (FIG. 5). The signalling pulse that is generated by the segmenta- tion clock is a good candidate to trigger the switch-like transition. This transition would result in a synchronous exposure of cells of the future segmental domain to RA signalling, thereby explaining the collective gene activa- tion in the stripe of cells. This hypothesis is consistent with the observation that Mesp2 and thylacine expres- sion is repressed by FGF and controlled by Notch and RA signalling 93,109,114 . Remarkably, a similar bistable behaviour working together with an autonomous clock is observed in in silico simulations of segmentation controlled by a moving gradient, as in the clock and wavefront model 118 . Conclusion Studies of the segmentation clock oscillator in verte- brates have begun to shed light on the complex mecha- nism that is involved in generating the characteristic periodic pattern of the vertebrate body axis. A common strategy, based on an oscillator (the segmentation clock) that generates a temporal periodicity and a travelling maturation front that converts the signalling pulse into a spatial periodic pattern, was identified in fish, frog, chicken and mouse embryos, supporting some conserva- tion of the segmentation mechanism among vertebrates. A metameric pattern similar to that of vertebrates is also found in many invertebrate phyla such as arthropods and annelids 119 . Some arthropods in which axis forma- tion and segmentation proceeds sequentially as in ver- tebrates, such as spiders, do show dynamic expression of Notch pathway genes during segment formation, suggesting that a molecular oscillator could operate in these species 120 . However, a very different segmentation machinery lacking cyclic genes has been identified in D. melanogaster. Furthermore, other invertebrates (including molluscs, nematodes and urochordates) are not segmented, raising the question of the conserva- tion of segmentation in evolution 119 . So, it still remains unclear whether segmentation appeared independently in different phyla during evolution or whether it repre- sents an ancestral feature of urbilateria, the ancestor of bilaterian animals. Other examples of periodic structures generated by an oscillator include the well-characterized circadian pattern of sporulation in Neurospora crassa, which results in the formation of stripes of spores that are deposited daily in the race tubes 121 . In plants, many structures, such as shoots and roots, are produced by progressive growth from a terminal growth zone (the meristem). This mechanism resembles that involved in producing the vertebrate AP body axis. Shoots and roots are often subdivided into repeated units or segments, as, for example, in bamboo. Whether the mechanisms gov- erning the establishment of this segmentation pattern in plants are related to that observed in animals is currently unknown. Given the similarity between the patterning of plant and animal body axes, it is possible that the clock and wavefront strategy for generating periodic patterns identified in vertebrates reflects a very general patterning principle for metazoans. Whether the ultradian oscillator identified in verte- brate segmentation is strictly dedicated to this process or whether it reflects a more fundamental cellular proc- ess at work in the embryo remains an open question. Strikingly, Hes1 oscillations with a period similar to that of somite segmentation in mouse can be recapitulated in cultured cells in response to serum stimulation 68 . Microarray studies of these serum-induced oscilla- tions in mouse C3H10T1/2 fibroblasts and human mesenchymal stem cells identified ultradian oscil- lations of Smad (MAD homologue) and Stat (signal transducer and activator of transcription) signalling pathways 122,123 . So, the molecular circuitry underlying these oscillations shows some differences from that of the segmentation clock. The key remaining task is the identification of the clock pacemaker that drives the oscillations of cyclic genes in somitogenesis and in these cultured cells. Whereas much has been learned since the identification of the segmentation clock, it currently remains unclear whether the cyclic genes are part of the segmentation clock pacemaker or if they merely reflect an output of a yet-to-be-identified pacemaker. S0 S?ISI S?IISII S0 S?ISI S?IISII SI S0SII S?I S?IISIII Nature Reviews | Genetics Anterior Posterior FGF/Wnt dominates FGF/Wnt dominates Bistability window RA dominates RA dominates FGF/Wnt dominates RA dominates Figure 5 | Model for segment determination. The system of opposing FGF (fibroblast growth factor)?Wnt (purple) and retinoic acid (RA) (green) gradients was proposed to define a bistability window (dashed rectangle) in which cells can adopt either of two distinct steady states (FGF-dominated or RA-dominated 15 ). In the bistability window, upon suprathreshold stimulation, cells that are in the FGF- dominated state can abruptly switch to the RA-dominated steady state, resulting in the simultaneous exposure of a cohort of cells (the future segment, in hatched orange) to RA signalling. This stimulation was proposed to be provided by the periodic signalling pulse delivered by the segmentation clock. Owing to the posterior extension of the axis and the decay of the FGF?Wnt mRNA and ligands in the PSM, the bistability window constantly moves posteriorly. The next cohort of cells to be simultaneously determined to form the future segment is hatched in blue. In this model, the posterior edge of the bistability window (bifurcation point) corresponds to the determination front. R E V I E W S NATURE REVIEWS | GENETICS VOLUME 9 | MAY 2008 | 379 In humans, severe disruptions of the segmentation pattern of the vertebrae lead to congenital scoliosis, which is a rare deformity of the spine that occurs in 1?2 per 10,000 births 124 . Most forms of congenital scoliosis are thought to be sporadic, but, in fact, little informa- tion on familial incidence is available. So far, traditional linkage analysis in families with individuals affected with congenital scoliosis has led to the identification of three genes, all associated with the segmentation clock. Mutations in Dll3, Mesp2 and Lfng were shown to lead to familial forms of spondylocostal dysostosis ? a form of congenital scoliosis 125?127 . The fact that all of the genes that have been associated with familial congenital scoliosis so far are linked to the segmentation clock suggests that these anomalies result from defects in the somitogenesis process 124 . In addition, it suggests that the oscillator also operates in human embryos to control segmentation. Studies of the somitogenesis process in mouse embryos point to a number of genes in which mutation results in phenotypes that resemble human congenital scoliosis, providing interesting can- didate genes that might carry mutations in the patients. Deciphering the segmentation clock mechanism in model organisms will help improve our knowledge of these diseases, which, in turn, could lead to improved clinical management of these patients. 1. Sadler, T. W. in Langman?s Medical Embryology (ed. Wilkins, W.A.) (Lippincott Williams & Wilkins, Baltimore, 2000). 2. Richardson, M. K., Allen, S. P., Wright, G. M., Raynaud, A. & Hanken, J. Somite number and vertebrate evolution. Development 125, 151?160 (1998). 3. Couly, G., Coltey, P. & Le Douarin, N. M. The triple origin of skull in higher vertebrates: a study in quail?chick chimeras. Development 117, 409?429 (1993). 4. Kuan, C. Y., Tannahill, D., Cook, G. M. & Keynes, R. J. Somite polarity and segmental patterning of the peripheral nervous system. Mech. Dev. 121, 1055?1068 (2004). 5. Kulesa, P. M. & Fraser, S. E. Cell dynamics during somite boundary formation revealed by time-lapse analysis. Science 298, 991?995 (2002). 6. Hirsinger, E., Jouve, C., Dubrulle, J. & Pourqui�, O. Somite formation and patterning. Int. Rev. Cytol. 198, 1?65 (2000). 7. Wellik, D. M. Hox patterning of the vertebrate axial skeleton. Dev. Dyn. 236, 2454?2463 (2007). 8. Kmita, M. & Duboule, D. Organizing axes in time and space; 25 years of colinear tinkering. Science 301, 331?333 (2003). 9. Saga, Y. & Takeda, H. The making of the somite: molecular events in vertebrate segmentation. Nature Rev. Genet. 2, 835?845 (2001). 10. Brent, A. E. & Tabin, C. J. Developmental regulation of somite derivatives: muscle, cartilage and tendon. Curr. Opin. Genet. Dev. 12, 548?557 (2002). 11. Bothe, I., Ahmed, M. U., Winterbottom, F. L., von Scheven, G. & Dietrich, S. Extrinsic versus intrinsic cues in avian paraxial mesoderm patterning and differentiation. Dev. Dyn. 236, 2397?2409 (2007). 12. Nusslein-Volhard, C. & Wieschaus, E. Mutations affecting segment number and polarity in Drosophila. Nature 287, 795?801 (1980). 13. Wilkins, A. S. The Evolution Of Developmental Pathways (Sinauer Associates, Sunderland, 2001). 14. Cooke, J. & Zeeman, E. C. A clock and wavefront model for control of the number of repeated structures during animal morphogenesis. J.Theor. Biol. 58, 455?476 (1976). The authors propose a theoretical model for somitogenesis postulating the existence of a clock and a wavefront. 15. Goldbeter, A., Gonze, D. & Pourqui�, O. Sharp developmental thresholds defined through bistability by antagonistic gradients of retinoic acid and FGF signaling. Dev. Dyn. 236, 1495?1508 (2007). 16. Dale, J. K. & Pourqui�, O. A clock-work somite. Bioessays 22, 72?83 (2000). 17. Kulesa, P. M., Schnell, S., Rudloff, S., Baker, R. E. & Maini, P. K. From segment to somite: Segmentation to epithelialization analyzed within quantitative frameworks. Dev. Dyn. 236, 1392?1402 (2007). 18. Palmeirim, I., Henrique, D., Ish-Horowicz, D. & Pourqui�, O. Avian hairy gene expression identifies a molecular clock linked to vertebrate segmentation and somitogenesiss. Cell 91, 639?648 (1997). This work provides the first molecular evidence of the existence of a molecular oscillator ? called the segmentation clock ? linked to somitogenesis, on the basis of the periodic transcriptional expression in the presomitic mesoderm of HES1. 19. Holley, S. A., Geisler, R. & Nusslein-Volhard, C. Control of her1 expression during zebrafish somitogenesis by a delta-dependent oscillator and an independent wave- front activity. Genes Dev. 14, 1678?1690 (2000). 20. Forsberg, H., Crozet, F. & Brown, N. A. Waves of mouse Lunatic fringe expression, in four-hour cycles at two-hour intervals, precede somite boundary formation. Curr.Biol. 8, 1027?1030 (1998). 21. Jiang, Y. J. et al. Notch signalling and the synchronization of the somite segmentation clock. Nature 408, 475?479 (2000). The authors propose a role for Notch signalling in the synchronization between neighbouring cellular oscillators in the PSM. 22. Li, Y., Fenger, U., Niehrs, C. & Pollet, N. Cyclic expression of esr9 gene in Xenopus presomitic mesoderm. Differentiation 71, 83?89 (2003). 23. Muller, M., v Weizsacker, E. & Campos-Ortega, J. A. Expression domains of a zebrafish homologue of the Drosophila pair-rule gene hairy correspond to primordia of alternating somites. Development 122, 2071?2078 (1996). 24. Oates, A. C. & Ho, R. K. Hairy/E(spl)-related (Her) genes are central components of the segmentation oscillator and display redundancy with the Delta/Notch signaling pathway in the formation of anterior segmental boundaries in the zebrafish. Development 129, 2929?2946 (2002). 25. Henry, C. A. et al. Two linked hairy/Enhancer of split- related zebrafish genes, her1 and her7, function together to refine alternating somite boundaries. Development 129, 3693?3704 (2002). 26. Gajewski, M., Elmasri, H., Girschick, M., Sieger, D. & Winkler, C. Comparative analysis of her genes during fish somitogenesis suggests a mouse/chick-like mode of oscillation in medaka. Dev. Genes Evol. 216, 315?332 (2006). 27. Shankaran, S. S. et al. Completing the set of h/E(spl) cyclic genes in zebrafish: her12 and her15 reveal novel modes of expression and contribute to the segmentation clock. Dev. Biol. 304, 615?632 (2007). 28. Sieger, D., Tautz, D. & Gajewski, M. her11 is involved in the somitogenesis clock in zebrafish. Dev. Genes Evol. 214, 393?406 (2004). 29. van Eeden, F. J. et al. Mutations affecting somite formation and patterning in the zebrafish, Danio rerio. Development 123, 153?164 (1996). 30. Holley, S. A., Julich, D., Rauch, G. J., Geisler, R. & Nusslein-Volhard, C. her1 and the Notch pathway function within the oscillator mechanism that regulates zebrafish somitogenesis. Development 129, 1175?1183 (2002). 31. Julich, D. et al. beamter/deltaC and the role of Notch ligands in the zebrafish somite segmentation, hindbrain neurogenesis and hypochord differentiation. Dev. Biol. 286, 391?404 (2005). 32. Itoh, M. et al. Mind bomb is a ubiquitin ligase that is essential for efficient activation of Notch signaling by Delta. Dev. Cell 4, 67?82 (2003). 33. Takke, C. & Campos-Ortega, J. A. her1, a zebrafish pair-rule like gene, acts downstream of notch signalling to control somite development. Development 126, 3005?3014 (1999). 34. Gajewski, M. et al. Anterior and posterior waves of cyclic her1 gene expression are differentially regulated in the presomitic mesoderm of zebrafish. Development 130, 4269?4278 (2003). 35. Giudicelli, F., Ozbudak, E. M., Wright, G. J. & Lewis, J. Setting the tempo in development: an investigation of the zebrafish somite clock mechanism. PLoS Biol. 5, e150 (2007). 36. Lewis, J. Autoinhibition with transcriptional delay: a simple mechanism for the zebrafish somitogenesis oscillator. Curr. Biol. 13, 1398?1408 (2003). This paper proposes a mathematical model for the segmentation clock oscillator in zebrafish, based on a negative feedback loop with delay. 37. Oates, A. C., Mueller, C. & Ho, R. K. Cooperative function of deltaC and her7 in anterior segment formation. Dev. Biol. 280, 133?149 (2005). 38. Riedel-Kruse, I. H., Muller, C. & Oates, A. C. Synchrony dynamics during initiation, failure, and rescue of the segmentation clock. Science 317, 1911?1915 (2007). 39. Davis, R. L. & Turner, D. L. Vertebrate hairy and Enhancer of split related proteins: transcriptional repressors regulating cellular differentiation and embryonic patterning. Oncogene 20, 8342?8357 (2001). 40. Mara, A., Schroeder, J., Chalouni, C. & Holley, S. A. Priming, initiation and synchronization of the segmentation clock by deltaD and deltaC. Nature Cell Biol. 9, 523?530 (2007). 41. Kawamura, A. et al. Zebrafish hairy/enhancer of split protein links FGF signaling to cyclic gene expression in the periodic segmentation of somites. Genes Dev. 19, 1156?1161 (2005). 42. Lewis, J. & Ozbudak, E. M. Deciphering the somite segmentation clock: beyond mutants and morphants. Dev. Dyn. 236, 1410?1415 (2007). 43. Jouve, C. et al. Notch signalling is required for cyclic expression of the hairy-like gene HES1 in the presomitic mesoderm. Development 127, 1421?1429 (2000). 44. Leimeister, C. et al. Oscillating expression of c-hey2 in the presomitic mesoderm suggests that the segmentation clock may use combinatorial signaling through multiple interacting bHLH factors. Dev. Biol. 227, 91?103 (2000). 45. Dunwoodie, S. L. et al. Axial skeletal defects caused by mutation in the spondylocostal dysplasia/pudgy gene Dll3 are associated with disruption of the segmentation clock within the presomitic mesoderm. Development 129, 1795?1806 (2002). 46. Bessho, Y., Miyoshi, G., Sakata, R. & Kageyama, R. Hes7: a bHLH-type repressor gene regulated by Notch and expressed in the presomitic mesoderm. Genes Cells 6, 175?185 (2001). 47. Maruhashi, M., Van De Putte, T., Huylebroeck, D., Kondoh, H. & Higashi, Y. Involvement of SIP1 in positioning of somite boundaries in the mouse embryo. Dev. Dyn. 234, 332?338 (2005). 48. McGrew, M. J., Dale, J. K., Fraboulet, S. & Pourqui�, O. The lunatic fringe gene is a target of the molecular clock linked to somite segmentation in avian embryos. Curr. Biol. 8, 979?982 (1998). 49. Aulehla, A. & Johnson, R. L. Dynamic expression of lunatic fringe suggests a link between notch signaling and an autonomous cellular oscillator driving somite segmentation. Dev. Biol. 207, 49?61 (1999). 50. Prince, V. E. et al. Zebrafish lunatic fringe demarcates segmental boundaries. Mech. Dev. 105, 175?180 (2001). 51. Pourqui�, O. Notch around the clock. Curr. Opin. Genet. Dev. 9, 559?565 (1999). 52. Jiang, Y. J., Smithers, L. & Lewis, J. The clock is linked to notch signalling. Curr. Biol. 8, R868?R871 (1998). R E V I E W S 380 | MAY 2008 | VOLUME 9 www.nature.com/reviews/genetics 53. del Barco Barrantes, I. et al. Interaction between Notch signalling and Lunatic Fringe during somite boundary formation in the mouse. Curr. Biol. 9, 470?480 (1999). 54. Huppert, S. S., Ilagan, M. X., De Strooper, B. & Kopan, R. Analysis of Notch function in presomitic mesoderm suggests a gamma-secretase-independent role for presenilins in somite differentiation. Dev. Cell 8, 677?688 (2005). 55. Morimoto, M., Takahashi, Y., Endo, M. & Saga, Y. The Mesp2 transcription factor establishes segmental borders by suppressing Notch activity. Nature 435, 354?359 (2005). The authors characterize the role of Mesp2 in defining the segment boundary. 56. Dale, J. K. et al. Periodic Notch inhibition by Lunatic Fringe underlies the chick segmentation clock. Nature 421, 275?278 (2003). 57. Serth, K., Schuster-Gossler, K., Cordes, R. & Gossler, A. Transcriptional oscillation of Lunatic fringe is essential for somitogenesis. Genes Dev. 17, 912?925 (2003). 58. Bessho, Y., Hirata, H., Masamizu, Y. & Kageyama, R. Periodic repression by the bHLH factor Hes7 is an essential mechanism for the somite segmentation clock. Genes Dev. 17, 1451?1456 (2003). 59. Ishibashi, M. et al. Targeted disruption of mammalian hairy and Enhancer of split homolog-1 (HES-1) leads to up-regulation of neural helix-loop-helix factors, premature neurogenesis, and severe neural tube defects. Genes Dev. 9, 3136?3148 (1995). 60. Ohtsuka, T. et al. Hes1 and Hes5 as notch effectors in mammalian neuronal differentiation. EMBO J. 18, 2196?2207 (1999). 61. Kokubo, H., Miyagawa-Tomita, S., Nakazawa, M., Saga, Y. & Johnson, R. L. Mouse hesr1 and hesr2 genes are redundantly required to mediate Notch signaling in the developing cardiovascular system. Dev. Biol. 278, 301?309 (2005). 62. Bessho, Y. et al. Dynamic expression and essential functions of Hes7 in somite segmentation. Genes Dev. 15, 2642?2647 (2001). 63. Morales, A. V., Yasuda, Y. & Ish-Horowicz, D. Periodic lunatic fringe expression is controlled during segmentation by a cyclic transcriptional enhancer responsive to Notch signaling. Dev. Cell 3, 63?74 (2002). 64. Cole, S. E., Levorse, J. M., Tilghman, S. M. & Vogt, T. F. Clock regulatory elements control cyclic expression of lunatic fringe during somitogenesis. Dev. Cell 3, 75?84 (2002). 65. Shifley, E. T. et al. Oscillatory lunatic fringe activity is crucial for segmentation of the anterior but not posterior skeleton. Development 135, 899?908 (2008). 66. Hirata, H. et al. Instability of Hes7 protein is crucial for the somite segmentation clock. Nature Genet. 36, 750?754 (2004). 67. Monk, N. A. M. Oscillatory expression of Hes1, p53 and NF-kB driven by transcriptional time delays. Curr. Biol. 13, 1409?1413 (2003). 68. Hirata, H. et al. Oscillatory expression of the bHLH factor Hes1 regulated by a negative feedback loop. Science 298, 840?843 (2002). This work shows that Hes1 oscillations are induced by serum stimulation in mouse fibroblasts, with a period similar to the mouse segmentation clock. 69. Aulehla, A. et al. Wnt3a plays a major role in the segmentation clock controlling somitogenesis. Dev. Cell 4, 395?406 (2003). The authors first report the implication of Wnt signalling in the mouse segmentation clock. 70. Dequeant, M. L. et al. A complex oscillating network of signaling genes underlies the mouse segmentation clock. Science 314, 1595?1598 (2006). This work provides the first systematic approach of the segmentation clock regulatory network uncovering the existence of a large number of cyclic genes participating in negative feedback loops linked to the Notch, Wnt and FGF pathways. 71. Weidinger, G., Thorpe, C. J., Wuennenberg-Stapleton, K., Ngai, J. & Moon, R. T. The Sp1-related transcription factors sp5 and sp5-like act downstream of Wnt/beta- catenin signaling in mesoderm and neuroectoderm patterning. Curr. Biol. 15, 489?500 (2005). 72. He, T. C. et al. Identification of c-MYC as a target of the APC pathway. Science 281, 1509?1512 (1998). 73. Glinka, A. et al. Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature 391, 357?362 (1998). 74. Yu, H. M. et al. The role of Axin2 in calvarial morphogenesis and craniosynostosis. Development 132, 1995?2005 (2005). 75. MacDonald, B. T., Adamska, M. & Meisler, M. H. Hypomorphic expression of Dkk1 in the doubleridge mouse: dose dependence and compensatory interactions with Lrp6. Development 131, 2543?2552 (2004). 76. Harrison, S. M., Houzelstein, D., Dunwoodie, S. L. & Beddington, R. S. Sp5, a new member of the Sp1 family, is dynamically expressed during development and genetically interacts with Brachyury. Dev. Biol. 227, 358?372. (2000). 77. Trumpp, A. et al. c-Myc regulates mammalian body size by controlling cell number but not cell size. Nature 414, 768?773 (2001). 78. Camenisch, T. D. et al. Disruption of hyaluronan synthase-2 abrogates normal cardiac morphogenesis and hyaluronan-mediated transformation of epithelium to mesenchyme. J. Clin. Invest. 106, 349?360 (2000). 79. Dunty, W. C., Jr. et al. Wnt3a/beta-catenin signaling controls posterior body development by coordinating mesoderm formation and segmentation. Development 135, 85?94 (2008). 80. Aulehla, A. et al. A beta-catenin gradient links the clock and wavefront systems in mouse embryo segmentation. Nature Cell Biol. 10, 186?193 (2008). 81. Lamar, E. et al. Nrarp is a novel intracellular component of the Notch signaling pathway. Genes Dev. 15, 1885?1899 (2001). 82. Krebs, L. T., Deftos, M. L., Bevan, M. J. & Gridley, T. The Nrarp gene encodes an ankyrin-repeat protein that is transcriptionally regulated by the notch signaling pathway. Dev. Biol. 238, 110?119 (2001). 83. Pirot, P., van Grunsven, L. A., Marine, J. C., Huylebroeck, D. & Bellefroid, E. J. Direct regulation of the Nrarp gene promoter by the Notch signaling pathway. Biochem. Biophys. Res. Commun. 322, 526?534 (2004). 84. Ishitani, T., Matsumoto, K., Chitnis, A. B. & Itoh, M. Nrarp functions to modulate neural-crest-cell differentiation by regulating LEF1 protein stability. Nature Cell Biol. 7, 1106?1112 (2005). 85. Ishikawa, A. et al. Mouse Nkd1, a Wnt antagonist, exhibits oscillatory gene expression in the PSM under the control of Notch signaling. Mech. Dev. 121, 1443?1453 (2004). 86. Dale, J. K. et al. Oscillations of the snail genes in the presomitic mesoderm coordinate segmental patterning and morphogenesis in vertebrate somitogenesis. Dev. Cell 10, 355?366 (2006). 87. Niwa, Y. et al. The initiation and propagation of Hes7 oscillation are cooperatively regulated by Fgf and notch signaling in the somite segmentation clock. Dev. Cell 13, 298?304 (2007). 88. Wahl, M. B., Deng, C., Lewandoski, M. & Pourqui�, O. FGF signaling acts upstream of the NOTCH and WNT signaling pathways to control segmentation clock oscillations in mouse somitogenesis. Development 134, 4033?4041 (2007). 89. Aulehla, A. & Herrmann, B. G. Segmentation in vertebrates: clock and gradient finally joined. Genes Dev. 18, 2060?2067 (2004). 90. Nakaya, M. A. et al. Wnt3a links left-right determination with segmentation and anteroposterior axis elongation. Development 132, 5425?5436 (2005). 91. Elsdale, T., Pearson, M. & Whitehead, M. Abnormalities in somite segmentation following heat shock to Xenopus embryos. J. Embryol. Exp. Morphol. 35, 625?635 (1976). 92. Dubrulle, J., McGrew, M. J. & Pourqui�, O. FGF signaling controls somite boundary position and regulates segmentation clock control of spatiotemporal Hox gene activation. Cell 106, 219?232 (2001). This work reports the first evidence of the role of FGF signalling as a key component of the wavefront in the chicken embryo. 93. Delfini, M. C., Dubrulle, J., Malapert, P., Chal, J. & Pourqui�, O. Control of the segmentation process by graded MAPK/ERK activation in the chick embryo. Proc. Natl Acad. Sci. USA 102, 11343?11348 (2005). 94. Linask, K. K. et al. N.-cadherin/catenin-mediated morphoregulation of somite formation. Dev. Biol. 202, 85?102 (1998). 95. Duband, J. L. et al. Adhesion molecules during somitogenesis in the avian embryo. J. Cell Biol. 104, 1361?1374 (1987). 96. Horikawa K, R. G., Takeichi M, Chisaka O. Adhesive subdivisions intrinsic to the epithelial somites. Dev. Biol. 215, 182?189 (1999). 97. Burgess, R., Cserjesi, P., Ligon, K. L. & Olson, E. N. Paraxis: a basic helix-loop-helix protein expressed in paraxial mesoderm and developing somites. Dev. Biol. 168, 296?306 (1995). 98. Nakaya, Y., Kuroda, S., Katagiri, Y. T., Kaibuchi, K. & Takahashi, Y. Mesenchymal?epithelial transition during somitic segmentation is regulated by differential roles of Cdc42 and Rac1. Dev. Cell 7, 425?438 (2004). 99. Sawada, A. et al. Fgf/MAPK signalling is a crucial positional cue in somite boundary formation. Development 128, 4873?4880 (2001). 100. Dubrulle, J. & Pourqui�, O. fgf8 mRNA decay establishes a gradient that couples axial elongation to patterning in the vertebrate embryo. Nature 427, 419?422 (2004). 101. Reifers, F. et al. Fgf8 is mutated in zebrafish acerebellar (ace) mutants and is required for maintenance of midbrain-hindbrain boundary development and somitogenesis. Development 125, 2381?2395 (1998). 102. Perantoni, A. O. et al. Inactivation of FGF8 in early mesoderm reveals an essential role in kidney development. Development 132, 3859?3871 (2005). 103. Draper, B. W., Stock, D. W. & Kimmel, C. B. Zebrafish fgf24 functions with fgf8 to promote posterior mesodermal development. Development 130, 4639?4654 (2003). 104. Vermot, J. et al. Retinoic acid controls the bilateral symmetry of somite formation in the mouse embryo. Science 308, 563?566 (2005). 105. Vermot, J. & Pourquie, O. Retinoic acid coordinates somitogenesis and left-right patterning in vertebrate embryos. Nature 435, 215?220 (2005). 106. Kawakami, Y., Raya, A., Raya, R. M., Rodriguez-Esteban, C. & Belmonte, J. C. Retinoic acid signalling links left-right asymmetric patterning and bilaterally symmetric somitogenesis in the zebrafish embryo. Nature 435, 165?171 (2005). 107. Sirbu, I. O. & Duester, G. Retinoic-acid signalling in node ectoderm and posterior neural plate directs left-right patterning of somitic mesoderm. Nature Cell Biol. 8, 271?277 (2006). 108. Diez del Corral, R. et al. Opposing FGF and retinoid pathways control ventral neural pattern, neuronal differentiation, and segmentation during body axis extension. Neuron 40, 65?79 (2003). 109. Moreno, T. A. & Kintner, C. Regulation of segmental patterning by retinoic acid signaling during Xenopus somitogenesis. Dev. Cell 6, 205?218 (2004). 110. Niederreither, K., Fraulob, V., Garnier, J. M., Chambon, P. & Dolle, P. Differential expression of retinoic acid-synthesizing (RALDH) enzymes during fetal development and organ differentiation in the mouse. Mech. Dev. 110, 165?171 (2002). 111. Abu-Abed, S. et al. The retinoic acid-metabolizing enzyme, CYP26A1, is essential for normal hindbrain patterning, vertebral identity, and development of posterior structures. Genes Dev. 15, 226?240 (2001). 112. Niederreither, K. et al. Genetic evidence that oxidative derivatives of retinoic acid are not involved in retinoid signaling during mouse development. Nature Genet. 31, 84?88 (2002). 113. Dubrulle, J. & Pourqui�, O. Coupling segmentation to axis formation. Development 131, 5783?5793 (2004). 114. Takahashi, Y. et al. Mesp2 initiates somite segmentation through the Notch signalling pathway. Nature Genet. 25, 390?396. (2000). 115. Morimoto, M. et al. The negative regulation of Mesp2 by mouse Ripply2 is required to establish the rostro- caudal patterning within a somite. Development 134, 1561?1569 (2007). 116. Kawamura, A. et al. Groucho-associated transcriptional repressor ripply1 is required for proper transition from the presomitic mesoderm to somites. Dev. Cell 9, 735?744 (2005). 117. Nakajima, Y., Morimoto, M., Takahashi, Y., Koseki, H. & Saga, Y. Identification of Epha4 enhancer required for segmental expression and the regulation by Mesp2. Development 133, 2517?2525 (2006). 118. Francois, P., Hakim, V. & Siggia, E. D. Deriving structure from evolution: metazoan segmentation. Mol. Syst. Biol. 3, 154 (2007). 119. Davis, G. K. & Patel, N. H. The origin and evolution of segmentation. Trends Cell Biol. 9, 68?72 (1999). 120. Stollewerk, A., Schoppmeier, M. & Damen, W. G. Involvement of Notch and Delta genes in spider segmentation. Nature 423, 863?865 (2003). 121. Dunlap, J. C. & Loros, J. J. The neurospora circadian system. J. Biol. Rhythms 19, 414?424 (2004). 122. Yoshiura, S. et al. Ultradian oscillations of Stat., Smad, and Hes1 expression in response to serum. Proc. Natl Acad. Sci. USA 104, 11292?11297 (2007). R E V I E W S NATURE REVIEWS | GENETICS VOLUME 9 | MAY 2008 | 381 123. William, D. A. et al. Identification of oscillatory genes in somitogenesis from functional genomic analysis of a human mesenchymal stem cell model. Dev. Biol. 305, 172?186 (2007). 124. Turnpenny, P. D. et al. Abnormal vertebral segmentation and the notch signaling pathway in man. Dev. Dyn. 236, 1456?1474 (2007). 125. Bulman, M. P. et al. Mutations in the human delta homologue, DLL3, cause axial skeletal defects in spondylocostal dysostosis. Nature Genet. 24, 438?441 (2000). 126. Whittock, N. V. et al. Mutated MESP2 causes spondylocostal dysostosis in humans. Am. J. Hum. Genet. 74, 1249?1254 (2004). 127. Sparrow, D. B. et al. Mutation of the LUNATIC FRINGE gene in humans causes spondylocostal dysostosis with a severe vertebral phenotype. Am. J. Hum. Genet. 78, 28?37 (2006). 128. Horikawa, K., Ishimatsu, K., Yoshimoto, E., Kondo, S. & Takeda, H. Noise-resistant and synchronized oscillation of the segmentation clock. Nature 441, 719?723 (2006). This work provides experimental evidence supporting the role of Notch in the synchronization between neighbouring PSM cells oscillations. 129. Ishimatsu, K., Horikawa, K. & Takeda, H. Coupling cellular oscillators: a mechanism that maintains synchrony against developmental noise in the segmentation clock. Dev. Dyn. 236, 1416?1421 (2007). 130. Maroto, M., Dale, J. K., Dequeant, M. L., Petit, A. C. & Pourqui�, O. Synchronised cycling gene oscillations in presomitic mesoderm cells require cell-cell contact. Int. J. Dev. Biol. 49, 309?315 (2005). 131. Masamizu, Y. et al. Real-time imaging of the somite segmentation clock: revelation of unstable oscillators in the individual presomitic mesoderm cells. Proc. Natl Acad. Sci. USA 103, 1313?1318 (2006). This work provides the first real-time imaging of the segmentation clock oscillations in mouse presomitic mesoderm using a luciferase-based reporter fused to the Hes1 promoter. 132. Pourqui�, O. & Tam, P. P. A nomenclature for prospective somites and phases of cyclic gene expression in the presomitic mesoderm. Dev. Cell 1, 619?620 (2001). 133. Ordahl, C. P. in Molecular Basis of Morphogenesis (ed. Bernfield, M.) pp165?176 (John Wiley and Sons, New York, 1993). Acknowledgements The authors thank A. Aulehla and to E. Ozbudak for critical reading of the manuscript. Work in the Pourqui� laboratory is supported by NIH grant R02HD043,158 to O.P. and by Stowers Institute for Medical Research. O.P. is a Howard Hughes Medical Institute Investigator. DATABASES Entrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query. fcgi?db=gene Axin2 | bicoid | Brachyury | Dll1 | Dusp4 | Dusp6 | engrailed | Epha4 | even-skipped | Fgf3 | Fgf4 | Fgf8 | Fgf18 | fgf24 | Fgfr1 | hairy | has2 | hedgehog | her1 | her7 | her11 | her12 | her15 | Hes5 | Hey1 | HEY2 | hunchback | Kruppel | Lfng | Mesp2 | Msgn1 | Myod | nanos | Nkd1 | Nrarp | Rbpjk | runt | Snai1 | SNAI2 | Spry2 | wingless | Wnt3a UniProtKB: http://ca.expasy.org/sprot ?-catenin | CDC42 | CYP26 | DACT1 | DeltaC | DeltaD | DKK1 | ERK | LEF1 | Mindbomb | MYC | Nodal | Notch1A | RAC1 | RALDH2 | Ripply1 | SP5 | TCF15 FURTHER INFORMATION Olivier Pourqui�?s homepage: http://research.stowers-institute.org/pourquielab/ ALL LINKS ARE ACTIVE IN THE ONLINE PDF R E V I E W S 382 | MAY 2008 | VOLUME 9 www.nature.com/reviews/genetics "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment