What’s hiding in Waddington’s epigenetic landscape? A case study in baby cichlids
by Karina Vanadzina
30 April 2018
In his 1957 book entitled The Strategy of the Genes, British scientist Conrad Hal Waddington noted that the genetic sequence does not map directly onto the phenotype we can observe in nature. Contrary to the gene-centric views held by many of his contemporaries, Waddington emphasised that phenotypes ultimately depend on the interaction between genes and an array of often environment-sensitive developmental factors he labelled as ‘epigenetic’.1,[*] Using Waddington’s famous ‘epigenetic landscape’ metaphor, the process that connects genotype to phenotype can be described as travelling downhill through a series of valleys from a starting point at the top of the landscape. Some valleys are deep and narrow, which means that the ultimate phenotype will be robustly produced across a wide range of rearing environments. In other cases, the directing landscape is flat and can lead to a variety of end points (i.e. different phenotypes) because the development of a trait is sensitive to environmental influences, often mediated by epigenetic factors.2 Waddington went on to propose that epigenetic factors could play a role in evolution, by allowing organisms to adjust to environmental conditions. Such adaptive phenotypes are subsequently stabilised through selection on genetic variation,3 a hypothesis now known as ‘plasticity-first’ evolution.4
Today epigenetics is an active field of research. An increasing number of studies show that epigenetic mechanisms can have an impact on macro-evolutionary phenomena such as species diversification.5 In accordance with Waddington’s views, which are increasingly becoming mainstream, phenotypic traits develop as part of a complex, but flexible system that incorporates both genetic and epigenetic factors. Such a system could potentially contribute to the generation of differences between species. Darwin’s finches are a well-known example of an adaptive radiation where a number of closely related species have adapted to different ecological niches over a short period of time. A recent genome-wide comparison between genetic mutations and epimutations (changes in DNA methylation patterns) in five species of finches showed that epigenetic change was more common than genetic mutations and was associated with cellular pathways important in bird evolution.6
The species-rich clade of cichlid fishes from the rift lakes of East Africa presents another fascinating case that is being analysed through epigenetic studies. Over the last 10 million years, a small starter set has diversified into more than 2,000 species sporting a variety of feeding adaptations that help them exploit their habitat.7 The structure of cichlid skulls reflects the way these fish gather and process food. Their feeding types can range from scrapers with short and powerful jaws suitable for removing algae from stones to suction feeders with long, rapidly-rotating jaws used for capturing plankton in the water column.8
Elucidating the molecular mechanism that underlies such remarkable flexibility in the jaw structure among cichlids has proven to be difficult. A 2011 study by Roberts et al9 pinpointed the Patched1 (Ptch1) gene as an important contributor to the variation in jaw shape between species with different feeding modes. The Ptch1 gene encodes a receptor in the Hedgehog signalling pathway which ensures that the bones in the jaw develop correctly. In addition, Ptch1 is involved in a cellular process that allows bone cells to sense and react to mechanical pressure in their surroundings. Despite the crucial role of Ptch1 in bone development, the study showed that genetic variation at the Ptch1 locus accounts for only 11.3% of the total variation in cichlid jaw shape. This mismatch prompted the search for epigenetic factors that have the capacity to nudge jaw development in a certain direction with no effect on the genetic sequence.
In a recent paper10, Yinan Hu and Craig Albertson identify one such epigenetic factor in baby cichlids. They note that larvae engage in a peculiar gaping behaviour (see video below) shortly before the onset of bone deposition in the jaw. It does not seem to serve any respiratory function because, at this early stage of development, all gas exchange still occurs through the cichlid’s skin. In a series of experiments, the authors quantified the impact of gaping behaviour on the length of a bone called the retro-articular process (RA), which generates rotation as part of the mechanical linkage system that connects different bones in the lower jaw. The length of the bones that act as levers in this system changes the action of the jaw and reflects specific feeding strategies. A long RA makes jaw opening more forceful while shortening it reduces the mechanical load and results in a weaker bite.
The authors hypothesised that more vigorous gaping puts more mechanical pressure on developing bones, which leads to an increase in bone deposition and, ultimately, a longer RA in adult fish. To test their hypothesis, they assessed the natural variation of gaping frequencies in species with different feeding types. As predicted, the larvae of algae scrapers with long RAs and strong jaws gaped more vigorously than the offspring of suction feeders, which are characterised by short RAs and a weak bite.
A set of manipulation experiments gave further support to the causal link between gaping behaviour and the final shape of the skull. In fast-gaping larvae of algal scrapers (who normally have long RAs), the authors cut the ligament that connects the RA to the rest of the bones involved in jaw rotation and found this reduced RA length. In another experiment, the larvae of suction feeders (who normally have short RAs) were placed in small containers which forced them to gape at higher frequency, and they ended up with longer RAs. The larvae from the latter experiment were also assayed for the expression of the Ptch1 gene, which was previously identified as a contributor to the variation in jaw structure between species with different feeding modes. Species with more robust jaws exhibited higher levels of Ptch1 expression throughout development when compared to cichlids with a weak bite.9 Interestingly, Hu & Albertson discovered that the increase in Ptch1 expression due to more vigorous gaping in suction feeders (who naturally exhibit lower Ptch1 expression) exceeded the levels naturally expressed in algal scrapers. This indicates that the Ptch1 gene is sensitive to immediate change in the developing environment of the jaw.
Overall, the authors succeeded in showing that gaping behaviour – initially perceived as just a quirk of development – is an epigenetic factor to be reckoned with. They conclude their paper with a model where both genetic and epigenetic factors contribute to the adaptive variation in cichlid jaw structure. Sensitivity to mechanical pressure exhibited by developing bone cells offers a flexible mechanism for dealing with novel environments and foodstuffs. If the environmental conditions persist over time, the initial response will be followed by genetic change that fixes the advantageous jaw structure, in line with ‘plasticity-first’ arguments.4 This represents the first stages of divergence between the ancestral species and its descendant that has successfully conquered a new niche. Exercising jaws in advance of bone deposition strengthens the adaptive fit between the jaw structure and the new environment at early stages of development.
For more detail, read the paper here:
Hu Y, Albertson RC. 2017. Baby fish working out: an epigenetic source of adaptive variation in the cichlid jaw. Proc R Soc B 284: 20171018.
1. Waddington CH. 1957. Allen & Unwin. 2. Noble D. 2015. J Exp Biol 218:816-818. 3. Waddington CH. 1959. Nature 183:1634-1638. 4. West-Eberhard MJ. 2003. Oxford University Press. 5. Smith G & Ritchie MG. 2013. Curr Zool 59:686-696. 6. Skinner MK, et al. 2014. Genome Biol Evol 6:1972-1989. 7. Kocher TD. 2004. Nat Rev Genet 5:288-298. 8. Cooper WJ, et al. 2010. PLoS ONE 5:e9551. 9. Roberts RB, Hu Y, Albertson RC & Kocher TD. 2011. PNAS 108:13194-13199. 10. Hu Y & Albertson RC. 2017. Proc R Soc B 284: 20171018.