A fundamental question in developmental biology is how the pace of developmental processes is controlled from a cell biological, molecular, and genetic perspective. Significant progress has been made in identifying the timekeeping mechanisms that regulate individual developmental events. For instance, the segmentation clock is a molecular oscillator that controls the rhythmic production of somites in vertebrate embryos 1, 2. A similar transcriptional oscillator times the progression between larval stages in the nematode Caenorhabditis elegans 3, 4. Other developmental events are gated by timers, such as the progressive degradation of maternal replication initiation factors that regulates the mid-blastula transition in Xenopus [5]. More recently, however, researchers have turned their attention toward the species-specific control of developmental pace 6, 7, 8, 9, 10, 11. It is striking that the absolute speed at which developmental clocks and timers tick differs significantly between species despite the coarse conservation of their molecular wiring. For instance, oscillations of the segmentation clock occur every 2 hours in mouse embryos but every 5 hours in human 12, 13, 14, 15, 16. Species-specific tempos often scale with the overall duration of embryogenesis within a given species (allochrony) but can also be specifically accelerated or slowed down relative to organismal development to achieve particular outcomes (heterochrony) 7, 17.

The rate at which clock genes flow through parts of the central dogma of molecular biology seems to correlate with developmental tempo. In the case of the segmentation clock, which scales allochronically with embryogenesis in most species [18], clock genes are spliced 18••, 19, 20 exported [19], translated [21], and degraded 18••, 20, 21•• faster in species with shorter clock periods. Importantly, however, these patterns do not arise from sequence differences in clock gene orthologs but rather from differences in the intracellular environment between species 20, 22. In fact, even exogenous genes such as fluorescent proteins are produced and degraded with species-specific kinetics depending on the species of the cell where they are expressed 20, 22. Species-specific biochemical reaction rates along the central dogma are broadly accepted to be the most proximate regulators of developmental speed. Nevertheless, given that species-specific changes in segmentation clock period could presumably be achieved by targeting any one of these biochemical steps, it is puzzling that all steps are always observed to scale concomitantly. For instance, accelerating splicing by removing introns from the Hes7 gene is sufficient to accelerate the clock without simultaneously changing the rate of nuclear export, protein synthesis, or protein degradation [23]. These observations have therefore prompted the search for a more ultimate, upstream regulator that could potentially control the pace of such varied molecular processes simultaneously. Here, we review recent attempts to identify ultimate regulators of developmental speed with a focus on mitochondrial metabolism.

This review disproportionately focuses on the role of mitochondrial metabolism in regulating developmental speed simply because this was the assigned article topic in the context of a special edition on developmental timing. However, this focus should not be misinterpreted as an undue bias in favor of this particular hypothesis.