Species-specific developmental timing is associated with global differences in protein stability in mouse and human

Introduction

“Diese Verschiebung kann sowohl den Teile Ort, als die Zeit der Erscheinung betreffen.

Jene erstere nennen wir Heterotopie, diese letztere Heterochronie”

(“This shift can affect both the location and the time of appearance.

We call the former heterotopy, the latter heterochrony.”)

Haeckel, 1875

Ernst Haeckel coined the term “heterochrony” to describe temporal deviations from his infamous recapitulation theory or biogenetic law. Its modern use, however, refers to any evolutionary change in developmental timing [1]. Changes in developmental timing between species are readily evident and have fascinated embryologists for decades (if not centuries). Before the genomic era, heterochrony was perceived as the temporal shift in the appearance of a morphological trait, when comparing embryos of different species (e.g. the appearance of the heart between human and fish). Now it is known that the underlying gene regulatory machinery to form homologous structures is greatly conserved between species, and heterochronic shifts can be recognised by comparing the temporal dynamics of gene expression through the development of such structures. The mechanisms responsible for these shifts have remained elusive, however.

About the preprint

Rayon and collaborators investigated the temporal variation in the differentiation of motor neurons (MNs) between mouse and human. MNs develop from progenitors in the embryonic spinal cord after being exposed to Sonic Hedgehog emanating from the notochord. In both species, MN progenitors can be identified by the expression of the genes Olig2 and Nkx6.1 due to the conservation of its Gene Regulatory Network (GRN). Although the temporal sequence of gene expression is conserved in both species, its rate is different. In mouse, MN differentiation takes 3-4 days while in human it takes ~2 weeks. The authors decided to use the term “developmental allochrony” instead of heterochrony, to highlight that in this case there is a scaling of developmental timing between these species without any apparent morphological trait alteration (see Author’s response below).

To investigate what could be causing this timing difference they decided to use MN differentiation in vitro for both species. First, they confirmed that their in vitro system could recapitulate the species-specific differentiation timing, assayed by the expression of marker genes such as Olig2 and Nkx6.1. Then they showed, by comparing bulk transcriptomes, that this difference in developmental tempo was present on a global scale as a 2.5 fold decrease in the gene expression rate in human compared to mouse (see Figure 1). So what could be causing this global change in the differentiation of MNs? The authors elegantly performed a series of tests to try to identify the culprit.

 

First, they tested if the slower pace in human could be explained by a reduced response to Shh signalling. To do this they increased the amount of Shh signaling in human MN progenitors (via addition of smoothened agonists). The results showed that species-specific sensitivity to Shh is not responsible for the difference in developmental tempo. Secondly, they tested if species-specific differences in the genetic sequence of Olig2 (the major regulator of MN progenitors) could explain the global tempo shift. For this, they used a mouse ESC line that contains the arm of chromosome 21 where the gene Olig2 is found. They found that these cells containing the human version of Olig2 had the same expression dynamics of cells with the mouse version of it, indicating that the temporal control of gene expression depends on the cellular environment and not on inter-specific differences in the genetic sequence of Oligo2 or its cis-regulatory regions.

Finally, Rayon and collaborators set out to determine if the degradation rate of transcripts and proteins could be involved in the temporal shift of MN differentiation. An assay of global mRNA stability showed no differences between mouse and human neural progenitors. A protein stability assay showed, however, that the half-life of the proteome in mouse neural progenitors was shorter than in human progenitors, by approximately 2.5 fold (Figure 2).

 

They used a mathematical model of the MN GRN, previously developed by the same group, to test if increasing the stability of transcription factors could account for the observed temporal differences. Indeed, the simulations showed that increasing the stability of the TF genes present in the GRN resulted in a slower pace of gene expression sequence, similar to the experimentally observed one (Figure 3 ). Given that the cell cycle also shows a similar temporal difference between species, the authors suggest that differences in cell cycle rate between mouse and human cells could also be a consequence of a global change in protein stability.

 

A previous analysis [2] demonstrated that the speed of the biochemical reactions (including protein degradation) involved in the intracellular network that drives gene expression oscillation in the presomitic mesoderm could explain the 2.5 fold timing difference of the segmentation clocks between the mouse and human (see prelights [3] and [4]). The results of Rayon and collaborators show that the timing differences between mouse and human motor neuron differentiation is linked to differences in protein stability between these species, suggesting that protein stability could regulate developmental timings at a global scale. How the protein stability can be regulated at a global level remains to be determined.

Questions to the authors:

1. Did you come up with the term “developmental allochrony”? I could not find any previous usage of it.

2. The GRN of MN differentiation consists of TF mutual inhibitions. Do you think this GRN structure is especially sensitive to protein stability? Could other type of GRNs (e.g. a more hierarchical GRN) be affected less by protein stability?

3. The idea of protein stability affecting both developmental timing and cell cycle rate is attractive as it could act as a mechanism to couple patterning with cell cycle dynamics during development. Do you think this could have evolved as a scaling mechanism of developmental timing, or is it an intrinsic property of developmental systems?

 

References

[1] Gould SJ. The Uses of Heterochrony. In: McKinney ML, ed. Heterochrony in Evolution: A Multidisciplinary Approach. Boston, MA: Springer US; 1988:1-13. doi:10.1007/978-1-4899-0795-0_1

[2] Matsuda M, Hayashi H, Garcia-Ojalvo J, et al. Species-specific oscillation periods of human and mouse segmentation clocks are due to cell autonomous differences in biochemical reaction parameters. bioRxiv. 2019. https://www.biorxiv.org/content/10.1101/650648v1.

[3] https://prelights.biologists.com/highlights/species-specific-oscillation-periods-of-human-and-mouse-segmentation-clocks-are-due-to-cell-autonomous-differences-in-biochemical-reaction-parameters/

[4] https://prelights.biologists.com/highlights/species-specific-oscillation-periods-of-human-and-mouse-segmentation-clocks-are-due-to-cell-autonomous-differences-in-biochemical-reaction-parameters-2/

Tags: allochrony, developmental time

doi: https://doi.org/10.1242/prelights.16298

Read preprint (No Ratings Yet)