Vol 454|17 July 2008
NEWS & VIEWS DEVELOPMENTAL BIOLOGY
Serpent clocks tick faster Freek J. Vonk and Michael K. Richardson
Snakes have graceful, elongated bodies containing hundreds of vertebrae. This extreme of morphology stems from evolutionary changes in a developmental clock that regulates body patterning. Evolution has produced an astonishing numbers of vertebrae in the snakes. diversity of body shapes adapted to The authors found that the snake different lifestyles. An important PSM develops more slowly and has a b contributor to the shape of animals longer lifespan than in short-bodied a with backbones is the number of animals. Despite this longer lifespan, c bones (vertebrae) that make up this the snake PSM does not undergo many structure. Some animals have gone to more cell divisions to make the backbone. There are roughly 21 rounds of extremes (Fig. 1), none more so than snakes, which have more vertebrae cell division in snakes, 16 in chicks and than any other living animal — often 13 in mice. This is too small a variamore than 300, with some species1 tion to fully explain the high vertebral having more than 500. On page 335 of count in snakes. But in addition, the clock component of the clock-andthis issue, Gomez et al.2 describe their d wavefront mechanism is accelerated investigations into how snakes develop — ticking around four times faster, this spectacular number of vertebrae. Their data suggest that a developmenrelative to growth rate2. Snake embryos tal ‘clock’, which regulates key steps in thus segment their available backbone body patterning, ticks faster (relative to tissue more rapidly than other animals, growth rate) in snakes than in shorterforming smaller and more numerous bodied animals. This is a dramatic somites (Fig. 2b). example of heterochrony3, in which The clock-and-wavefront mechanism may be a ‘hotspot’ for natural adjustments in developmental timing Figure 1 | Serpentine silhouettes. Body elongation, including selection, because changes in vertelead to morphological change. At first glance, a snake looks like increased numbers of vertebrae and a longer backbone, has bral number can greatly affect fitness. evolved independently many times among vertebrates. a, Eastern a long tail with a head at one end; glass lizard (Ophisaurus ventralis). b, European eel (Anguilla The number of vertebrae an animal but snakes are not ‘all tail’. Their tail anguilla). c, Sao Tome caecilian (Schistometopum thomense). has can influence such factors as loco— defined as the part of the body that d, Malayan pit viper (Calloselasma rhodostoma). Photos not to motor speed and even fecundity (giving follows the cloaca (the opening at the scale. (a,b,d, F.J.V. & M.K.R.; c, S. Blair Hedges.) more room in a body for eggs7), and may end of the genital and intestinal tracts) underlie the evolution of some species — is in fact relatively short. In front of the tail, the PSM from tail to head. These oscillations groups. Serpentine body shapes have evolved the snake has a neck, thorax and abdomen like are halted when they meet a wave of maturation independently in groups such as eels, caecilany other reptile, although the boundaries that sweeps through the PSM in the opposite ian amphibians and some lizards, as well as in between these regions are not obvious. Loss direction, from head to tail. Each cycle results snakes (Fig. 1). In these long-bodied animals, of these boundaries in the snake body, along in a new pair of somites, the process ceasing there would have been an adaptive advantage with its loss of limbs, is thought to be due to when the wave of maturation catches up with in evolving an extreme body shape, perhaps evolutionary changes in the workings of the the ‘tail’ end of the PSM. allowing different ecological niches to be Hox developmental genes4. How snakes make Gomez and colleagues2 compared corn conquered. such large numbers of vertebrae is a separate snake embryos with those of shorter-bodied Snakes put their long bodies to good use question. animals (mice, chicken and zebrafish), which when moving through dense vegetation, Vertebrae develop from segments of tissue have significantly fewer vertebrae, to deter- along tree branches or in water, which they called somites, which form, one after another, mine whether the large number of vertebrae do with incredible grace. Constricting snakes in a head-to-tail sequence in the embryo in snakes is generated by a variation on the use their coils to suffocate prey, whereas ven(Fig. 2a). They bud off from the ‘head’ end of clock-and-wavefront model or by some fun- omous snakes are like a coiled spring that can the presomitic mesoderm (PSM), an immature damentally different mechanism. They found strike prey at a distance and with terrifying tissue fated to generate the somites. This bud- the same molecular toolkit for body segmenta- speed. Snakes are among the most successful ding is regulated by a ‘clock-and-wavefront’ tion in corn snake embryos as in other verte- of vertebrate groups, both in terms of number model, which was first proposed by Cooke brates: a cyclic pattern of clock-gene expression of species and geographic distribution8. Elonand Zeeman5 in 1976. Cyclical waves of gene in the developing PSM. However, subtle vari- gated bodies have certainly contributed to this expression, controlled by a molecular oscillator ation in this common theme for body seg- success, and the fact that heterochrony may called the segmentation clock6, spread through mentation in vertebrates produced the greater be involved is particularly topical given the 282
NEWS & VIEWS
NATURE|Vol 454|17 July 2008
PSM ‘head end’
Somites
PSM ‘tail end’
a
Tail growth New somite formation
Snakes (~4 times faster) Clock-gene expression
Clock-gene expression
b Short-bodied vertebrates
Waves of clockgene expression
Pair of somites
Pair of somites
Figure 2 | Speeding-up somitogenesis. a, Somites, the regular embryonic body segments from which some bones (including vertebrae), muscles and other tissues develop in vertebrates, form by a general ‘clock-and-wavefront’ model5. Waves of cyclical gene expression, driven by ‘clock’ genes, mature at the ‘head’ end of the presomitic mesoderm (PSM) to form paired blocks of spatially distinct somitic tissues. b, In snakes (right), the clock components seem to tick around four times faster (relative to growth rate) than in shorter-bodied animals (left), leading to many more, though smaller, somites2.
current interest in chronomics, the study of time regulation in biological systems. Model species (such as mice) will always be the mainstay of research because scientists can engineer changes in their genomes and look for effects on the body plan under controlled conditions9. Biologists are nevertheless increasingly considering non-model organisms. In snakes, natural selection has done
the genetic engineering for us, and so we can study them as experiments performed by evolution. Investigating snakes hand-in-hand with model species will provide a holistic view of evolution and development. Snake embryos are not easy to work with in the lab because it is difficult to open their eggs without damaging the embryo, and obtaining eggs from snakes is not always easy. But if these technical
MICROSCOPY
Spot the atom John Silcox Heavy atoms can be detected by electron microscopy, but lighter atoms, such as carbon or hydrogen, are more elusive. These bashful atoms can now be pinpointed if they are adsorbed to a single layer of graphite. Visualizing individual atoms is hard, not least because free atoms move rapidly at room temperature. The problem is finding a way to keep them motionless for long enough to observe them, without the resulting image being swamped by images of surrounding atoms. Ideally, they would be made to sit still on a transparent substrate in a near-perfect vacuum, so that the resulting image would be an array of sharp spots, rather like bright stars in a cloudless night sky. Reporting in this issue (page 319), Meyer et al.1 bring that ideal closer to reality. They have used electron microscopy to observe carbon atoms sitting on a graphene sheet (a single layer of carbon atoms arranged in
a ‘honeycomb’ array). Even more impressively, they have used image recording and computer simulations to identify hydrogen atoms — the smallest and most difficult atoms to spot — adsorbed to the sheet. There are several reasons why graphene is superb as a backdrop for Meyer and colleagues’ microscope images. First, under the conditions used for electron microscopy it traps molecules or atoms that land on its surface, so providing a way of immobilizing the specimens for long enough to get a decent image. Second, a single atomic layer is the thinnest, most transparent support possible, so any clouding of the image by the substrate is kept to a minimum. But despite being extremely thin, graphene
challenges are overcome, and snake genome sequences become widely available, a new era of ‘evo-devo’ research may open up. The question remains whether the somite clock also ticks faster in other long-bodied animals, or whether natural selection has exploited alternative ways of creating a serpentine body. For example, we know that abnormal incubation temperatures can influence vertebral count in embryos (reviewed in ref. 10), and it would be interesting to know how this relates to the ticking of the somite clock. Even in snakes, alternative strategies for body elongation are possible: real giants, such as anacondas (Eunectes murinus), which can reach lengths of 8 metres or more, do not have more vertebrae than other snakes, they just have larger ones11. Further comparative studies in non-model species may show us how natural selection has tinkered with developmental mechanisms and pathways to produce different body plans. ■ Freek J. Vonk and Michael K. Richardson are at the Institute of Biology, Leiden University, PO Box 9516, 2300 RA Leiden, the Netherlands. e-mails:
[email protected];
[email protected] 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
Marx, H. & Rabb, G. B. Fieldiana Zool. 63, 1–321 (1972). Gomez, C. et al. Nature 454, 335–339 (2008). Richardson, M. K. BioEssays 21, 604–613 (1999). Cohn, M. J. & Tickle, C. Nature 399, 474–479 (1999). Cooke, J. & Zeeman, E. C. J. Theor. Biol. 58, 455–476 (1976). Palmeirim, I. et al. Cell 91, 639–648 (1997). Shine, R. J. Evol. Biol. 13, 455–465 (2000). Greene, H. W. Snakes: Evolution of Mystery in Nature (Univ. California Press, 1992). Hérault, Y. et al. Nature Genet. 20, 381–384 (1998). Fowler, J. A. Q. Rev. Biol. 45, 148–164 (1970). Head, J. J. & Polly, D. Biol. Lett. 3, 296–298 (2007).
is also strong — certainly robust enough to survive three hours or more of irradiation by high-energy electrons, which are sometimes required to spot small atoms. Meyer et al.1 used an electron microscope that cannot identify carbon atoms in a graphene lattice directly, but detects the change in intensity that results when an atom binds to the graphene. The difference in intensity is caused by the extra scattering of incident electrons by the adsorbed atom, which adds to the number of electrons scattered by the graphene layer itself. In the authors’ images, the carbon atoms show up as dark spots against a bright background — a negative image of a night sky (see Fig. 2a on page 320). Meyer and colleagues’ technique allows real-time observation of atomic and molecular dynamics, and of chemical reactions that occur under electron irradiation. The authors demonstrate this by recording a series of images showing the formation of vacancies in the graphene sheet — holes in the lattice caused by missing atoms — as the carbon monolayer is bombarded with electrons. They even record one of these vacancies disappearing as it is filled by a carbon atom adsorbed on the graphene sheet. No doubt 283
