Metabolic regulation of species-specific developmental rates

Background

Rate of embryonic development varies widely across species. In mammals, large-bodied species tend to have slower development rates and increased lifespan compared to small-bodied species. For example, the rate of human embryo development is about 2-3 times slower than mouse embryos even though their overall embryonic size is similar, and they undergo the same series of developmental steps. Recent work has identified the underlying cause of these differences as coming from differential rates of protein production and degradation between the two species^1,2. However, what leads to these different biochemical speeds in the two species remains unknown. In this preprint, Diaz-Cuadros et al. provide the first clues towards understanding this phenomenon by investigating the role of metabolism in regulating development rates within a specific developmental process, termed the segmentation clock.

The segmentation clock is a tissue-scale rhythmic patterning system in vertebrates. Locally synchronized waves of genetic transcription sweep from the posterior of the embryo to the anterior providing spatiotemporal information to morphological somite formation. Somites are segmented tissues in the embryonic mesoderm, which give rise to the adult musculoskeletal system. Somites form in a periodic manner – about every 5 hours in human embryos and about every 2.5 hours in mouse embryos – and this periodicity emerges from periodic transcriptional waves. Given the difference in periodicity between humans and mice, the segmentation clock can be used as a powerful system to probe underlying processes driving differential rates during development. To follow periodic transcriptional waves, the authors had previously established a methodology to recapitulate the segmentation clock in vitro from pluripotent stem cells (PSC)³. Here, the same in vitro system was used to investigate the role of metabolism in the differential periodicity of the segmentation clock.

Key Findings

The authors first developed a protocol to differentiate mouse and human PSCs towards presomitic mesoderm (cell type which exhibits segmentation clock oscillations) fate under identical media conditions. An accelerated differentiation efficiency, a faster cell cycle rate and more frequent oscillations in mouse cells indicated an approximately two-fold difference in developmental rate between mouse and human cells (Fig. 1A). The observed developmental rate was cell autonomous as the segmentation clock period did not change in isolated cells nor when human cells were co-cultured at low density with mouse cells.

To investigate the source of this differential developmental rate, oxygen consumption rate (OCR) and glycolytic proton efflux rate (glycoPER), which both serve as proxies for metabolic rate, were determined. Surprisingly, both rates were similar between mouse and humans. However, as the total mass and volume of human cells were twice that of mouse cells, it was concluded that a good comparison is only possible by considering mass-specific metabolic rates, which represents the rate at which cells consume energy per gram of cell weight. Accordingly, mass-specific OCR and glycoPER were found to be twice as fast in mouse cells as in human cells (Fig. 1B). A faster mass-specific metabolic rate in mouse cells was further confirmed through multiple independent measurements of glucose consumption, lactate secretion and glutamine consumption, as well as through comparison of mouse and human neural progenitors differentiated in vitro from PSCs.

Figure 1. Key findings from the preprint. Adapted from Figs 1-4, Diaz-Cuadros et al. 2021

The authors then proceeded to test which aspect of metabolism controls developmental rate. Given that segmentation clock oscillations are known to be affected upon perturbation of respiration but not glycolysis, human presomitic mesoderm (PSM) cells were treated with small molecule inhibitors of the electron transport chain (ETC). Inhibiting ETC complexes I, III, and IV in human PSM cells resulted in a significant lengthening of the segmentation clock and damped oscillatory dynamics, while inhibition of the ATP synthase had no effects. These results suggest that ETC, rather than ATP synthase, activity is involved in the regulation of the segmentation clock. To further test the hypothesis that cellular ATP levels do not regulate the segmentation clock, the authors cultured human PSM cells in multiple conditions that increased the cellular concentration of ATP. These conditions did not shorten the segmentation clock period and increased cell cycle length, indicating slower proliferation. Therefore, the authors concluded that increased levels of ATP do not mediate the accelerated developmental rate of mouse cells.

Considering the link between a high NAD+/NADH ratio and increased proliferation rates⁴, the authors then tested the importance of NAD+ availability on the segmentation clock. Decreasing NAD+/NADH within human PSM cells pharmacologically or through manipulation of the cell culture medium led to a lengthening of the segmentation clock period. This effect could be rescued by regenerating NAD+ levels, suggesting the segmentation clock period depends on NAD+ levels (Fig. 1C).

Previous studies have suggested that differences in developmental rate between species was partly due to differences in protein production^1,2. The authors validated this and showed that mouse PSM cells have a mass-specific translation rate that is twice as fast as human PSM cells (Fig. 1D). In addition, slowing down translation led to a significant extension of the segmentation clock period of human PSM cells. Importantly, ETC inhibition, which also lengthened the segmentation clock period, decreased translation rates, with the magnitude of the effect on clock period scaling with magnitude of translation rate inhibition. Further experiments suggested that mitochondrial activity acts upstream of translation rate to regulate the clock period. In general, a change in the clock period could also emerge from a change in protein degradation rate. However, clock oscillations were found to be much less sensitive to perturbations of proteasome activity compared to perturbations of protein production. Taken together, the authors concluded that mitochondrial activity drives the translation rate which regulates the segmentation clock period and therefore, developmental rate.

Why we like this preprint

1. This is the first study that systematically compares mass-specific metabolic rates across species during embryonic development. By doing so, this work has shown Kleiber’s law, a long-standing allometric relationship between body mass and metabolic rate, to be valid during embryonic development as well.
2. The discovery that developmental rate, at least in the context of segmentation clock period, is sensitive to NAD+ levels is interesting from a perspective of understanding cancer cell proliferation as well as aging-related processes, both of which are known to depend on NAD+ levels.
3. This work potentially sets the stage for developing stem cell-based therapies in the future where manipulation of differentiation rate could aid in accelerating disease modeling.

Open questions for the authors

1. Why are control segmentation clock periods (compare Fig. 3b, 3e and 3i) variable across experiments? If you run a statistical test across controls, would they be significantly different? On the same note, clock period under DCA-treated conditions matches the control clock period from Fig. 3b. Given this, how do you interpret changes in clock period under these conditions?
2. In vivo, the electron transport chain and oxidative metabolism was previously shown to have a gradient with a higher activity in the anterior^5,6. Considering results from this work, does it mean that a differential clock period between mouse and human cells can be explained as a function of metabolic rate only in the anterior?
3. Do the authors plan to study whether mutations in mitochondrial/metabolic genes are linked to segmentation clock or somitogenesis defects? Have any links already been established?
4. From an evolutionary perspective, it still remains unclear why different species have widely varying metabolic rates. Could the authors comment on this?
5. Any speculation on why embryos from a particular species have a changed protein production rate rather than degradation rate to modulate overall developmental speed?

References:

1. Matsuda, M., et al., Species-specific segmentation clock periods are due to differential biochemical reaction speeds. Science, 2020. 369(6510): p. 1450.
2. Rayon, T., et al., Species-specific pace of development is associated with differences in protein stability. Science, 2020. 369(6510): p. eaba7667.
3. Diaz-Cuadros, M., et al., In vitro characterization of the human segmentation clock. Nature, 2020. 580(7801): p. 113-118.
4. Luengo, A., et al., Increased demand for NAD(+) relative to ATP drives aerobic glycolysis. Mol Cell, 2021. 81(4): p. 691-707.e6.
5. Özbudak et al., Spatiotemporal compartmentalization of key physiological processes during muscle precursor differentiation. PNAS, 2010. 107(9): p. 4224-4229
6. Oginuma et al., A gradient of glycolytic activity coordinates FGF and Wnt signaling during elongation of the body axis in amniote embryos. Dev. Cell, 2017. 40: p. 342-353

Tags: differentiation rate, electron transport chain, genetic oscillation, metabolism, pluripotent stem cells, somitogenesis

doi: https://doi.org/10.1242/prelights.30666

Read preprint (No Ratings Yet)