Biophysics and population size constrains speciation in an evolutionary model of developmental system drift

Abstract

Understanding the origin of species is as Darwin called it "that mystery of mysteries". Yet, how the processes of evolution give rise to non-interbreeding species is still not well understood. In an empirical search for a genetic basis, transcription factor DNA binding has been identified as an important factor in the development of reproductive isolation. Computational and theoretical models based on the biophysics of transcription factor DNA binding have provided a mechanistic basis of such incompatibilities between allopatrically evolving populations. However, gene transcription by such binding events occurs embedded within gene regulatory networks, so the importance of pair-wise interactions compared to higher-order interactions in speciation remains an open question. Theoretical arguments suggest that higher-order incompatibilities should arise more easily. Here, we show using simulations based on a simple biophysical genotype phenotype map of spatial patterning in development, that biophysics provides a stronger constraint, leading to pair-wise incompatibilities arising more quickly and being more numerous than higher-order incompatibilities. Further, we find for small, drift dominated, populations that the growth of incompatibilities is largely determined by sequence entropy constraints alone; small populations give rise to incompatibilities more rapidly as the common ancestor is more likely to be slightly maladapted. This is also seen in models based solely on transcription factor DNA binding, showing that such simple models have considerable explanative power. We suggest the balance between sequence entropy and fitness may play a universal role in the growth of incompatibilities in complex gene regulatory systems.

Synthetic genotype networks

“…The epistatic relation between GRNs 2c.6 and 3.2 also exemplifies how hybrid incompatibilities could arise: certain crossings between individuals carrying these GRNs could lead to GRNs without a stripe phenotype. This is consistent with system drift being an important source of hybrid incompatibilities that cause reproductive isolation (Johnson and Porter, 2000, 2001; Khatri and Goldstein, 2019).…”

Reproductive barriers as a byproduct of gene network evolution

“…However, the pathway framework can be readily modified to include divergent selection and will almost certainly result in higher degrees of reproductive isolation. Fifth, recent work suggests that demographic complexity such as a varying population size can adds constraints on the process of speciation (Harvey et al, 2019;Khatri and Goldstein, 2019;Yamasaki et al, 2020;Momigliano et al, 2020). Future research may combine the pathway framework with additional demographic information to further investigate how ancestral genetic variation is sorted throughout the evolutionary trajectories.…”

Genetic drift promotes and recombination hinders speciation on holey fitness landscapes

“…One question raised by founder effect speciation models is whether fluctuations of population size, such as those caused by founder events, may have a different effect on the accumulation of genetic incompatibilities when compared to constant (low) population size, as in our model. Khatri and Goldstein showed that genetic incompatibilities accumulate more quickly in smaller populations in several models of gene regulation 27,28,50 . Like our model, their models included strong incompatibilities that create holes in the fitness landscape.…”

“…Khatri and Goldstein showed that genetic incompatibilities accumulate more quickly in smaller populations in several models of gene regulation 27, 28, 50 . Like our model, their models included strong incompatibilities that create holes in the fitness landscape.…”

“…Empirical tests of this prediction confirmed that genetic incompatibilities accumulated faster than linearly for inviability in species of Drosophila 12, 13 , male sterility in subspecies of Mus musculus 15 , and both seed and pollen sterility in species of Solanum 11, 14 . Computational studies using models of RNA folding 26 , transcription factor-DNA binding 27 , and spatial gene regulation during development 28 also found that genetic incompatibilities snowballed. These empirical and computational studies were all conducted over relatively short evolutionary distances.…”