Adult marbled and crested newts have two versions – a long and a short one – of their largest chromosome: chromosome 1. They randomly transmit either the long or the short version to each of their sex cells, resulting in an equal ratio of sex cells with the long or short version. When egg and sperm cells fuse upon fertilization, by chance half of the resulting embryos will have either two long or two short versions. Such individuals die before they even hatch. A situation in which only individuals with two distinct versions of a chromosome survive – and the ones that have the same version twice perish – is called a balanced lethal system.

In a new paper out in Philosophical Transactions of the Royal Society B: Biological Sciences we provide a new hypothesis on how balanced lethal systems could evolve. Balanced lethal systems pose an evolutionary paradox, because they are extremely wasteful, but must have evolved in the face of natural selection. Emma Berdan did a postdoc in my lab on my ERC Starting Grant project. Together with Alexandre Blanckaert, Emma took the lead in this study (actually part of a special issue that Emma and I co-edited). We tested if a balanced lethal system could be a potential end point of a ‘supergene system’.

Supergenes are large chunks of DNA that contain many genes and come in different versions. Crucially, the different versions of the same supergene cannot exchange DNA anymore because, for example, one of them has been flipped around in the genome. Due to the different orientation of the two supergene versions, they are not recognized as equivalent during the production of sex cells, when typically the DNA of both parents gets reshuffled to produce a new, unique genetic combination, different from each parent (a process known as recombination). As a consequence of suppressed recombination in supergenes, alleles (gene variants) of many genes can co-evolve independently on the different versions and, together, eventually encode widely different phenotypes. This can be a good thing: distinct phenotypes may provide unique benefits. If this were the case, then balancing selection would tend to preserve both supergene versions in the same population.

However, the lack of recombination between the supergene versions also comes at a cost: when genes on one supergene version get broken, they cannot be replaced by working copies on the alternative supergene version anymore – and the other way around. The official term for this irreversible accumulation of broken genes is Muller’s Ratchet. Now what if both supergene versions acquire unique broken genes? Then you are only viable if you possess both supergene versions, and a balanced lethal system is born! We wanted to see what it would take for this to happen in nature.

We simulate a situation in which two different versions of a supergene first evolve in separate populations and randomly acquire bad mutations here. The two are subsequently united, because one population donates its supergene version to the population with the alternative supergene version via hybridization. Now each supergene version can correct for the bad mutations on the other one. We explore how the two supergene versions could be maintained together by balancing selection long enough for them to degrade to the point at which each of them cannot function on its own anymore because genes have become broken. The tricky thing is that, as soon as one supergene version starts to perform relatively worse, natural selection tends to remove it from the population. Degradation of the supergene versions needs to be symmetrical! Only when the population is tiny, and the broken genes are fully compensated by their complements on the alternative supergene version, do we sometimes see balanced lethal systems appear in our simulations.

We need to look at Triturus newts and see if we can find evidence of hybridization associated with chromosome 1: could it be that one of the versions of chromosome 1 was donated to the ancestral Triturus population by another newt species? Actually, introgression (genetic exchange between species) abounds in the salamander family that the crested and marbled newts belong to. When we get our hands on Triturus genomes we can also test for the signature left by the tiny population size that we predict is required for the origin of the balanced lethal system. Did the ancestral Triturus population go through such a bottleneck? And when we home in on the broken genes on each version of chromosome 1, we expect that the task of these genes is fully taken over by their counterparts on the alternative version of chromosome 1. Lots of exciting ideas to test!

Reference: Berdan, E.L., Blanckaert, A., Butlin, R.K., Flatt, T., Slotte, T., Wielstra, B. (2022). Mutation accumulation opposes polymorphism: Supergenes and the curious case of balanced lethals. Philosophical Transactions of the Royal Society B 377(1856): 20210199.

This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant Agreement No. 802759).