A cellular basis for the mammalian nocturnal-diurnal switch

Andrew D Beale; Matthew J Christmas; Nina M Rzechorzek; Andrei Mihut; Aiwei Zeng; Christopher Ellis; Nathan R James; Nicola J Smyllie; Violetta Pilorz; Rose Richardson; Mads F Bertelsen; Shaline V Fazal; Zanna Voysey; Kevin Moreau; Jerry Pelletier; Priya Crosby; Sew Y Peak-Chew; Rachel S Edgar; Madeline A Lancaster; Roelof A Hut; John S O’Neill

doi:10.1126/science.ady2822

. Author manuscript; available in PMC: 2026 Mar 10.

Published in final edited form as: Science. 2026 Feb 26;391(6788):eady2822. doi: 10.1126/science.ady2822

A cellular basis for the mammalian nocturnal-diurnal switch

Andrew D Beale ^1,^*, Matthew J Christmas ^2,^#, Nina M Rzechorzek ^1,^3,^#, Andrei Mihut ^1,^#, Aiwei Zeng ^1,^12,¹³, Christopher Ellis ¹, Nathan R James ¹, Nicola J Smyllie ¹, Violetta Pilorz ⁴, Rose Richardson ⁵, Mads F Bertelsen ⁶, Shaline V Fazal ⁷, Zanna Voysey ⁷, Kevin Moreau ⁸, Jerry Pelletier ^9,^‡, Priya Crosby ^10,¹¹, Sew Y Peak-Chew ¹, Rachel S Edgar ^12,¹³, Madeline A Lancaster ¹, Roelof A Hut ¹⁴, John S O’Neill ^1,^*

PMCID: PMC7618833 EMSID: EMS212587 PMID: 41747039

Abstract

Early mammals were nocturnal while dinosaurs dominated the daytime. Mammalian transition to daytime activity accelerated after the Cretaceous-Paleogene extinction, but the underlying mechanisms remain unclear. We identified a conserved cell-intrinsic, thermodynamic mechanism that likely facilitated this shift. In cells from diurnal mammals, protein synthesis, phosphorylation and circadian timing were less sensitive to temperature changes than cells from nocturnal mammals. Comparative genomics revealed accelerated evolution within essential signaling pathways, including mechanistic Target of Rapamycin (mTOR), that increase the robustness of diurnal cellular clocks to thermal and osmotic perturbation. In nocturnal mice, mTOR inhibition shifted cells, tissues and behavior towards diurnal activity. These findings uncover a genetic and biochemical basis for nocturnal-diurnal switching, emphasizing how cellular signaling networks can encode complex phenotypes such as temporal niche selection.

Early mammals were nocturnal (active at night) until extinction of the diurnal (active during the day) dinosaurs facilitated a rapid expansion into daytime niches (1, 2). Diurnality subsequently arose multiple times, independently, from diverse and distant nocturnal lineages (1–3). No mechanistic basis for the switch between nocturnality and diurnality is known, though evidently some change in the relation between internal circadian clocks and external daily rhythms is required (4).

Despite the 76 million years that separate nocturnal mice and diurnal humans from their common ancestor (5), the same cell-autonomous circadian clock mechanism operates in both mouse and human cells (6). Daily rhythms of gene expression, proteome renewal, and myriad cellular functions depend on cell-intrinsic ~24 hour oscillations in the production of PERIOD (PER) proteins (6); where the changing activity of PER over time effectively determines the biological time-of-day (7). Similarly, the hypothalamic suprachiasmatic nucleus (SCN) performs an equivalent function in diurnal and nocturnal mammals, receiving light input directly from the eyes to generate an internal representation of solar time (4, 8, 9). However, unlike the SCN, PER oscillations in peripheral cells and tissues are oppositely organized between diurnal and nocturnal mammals (9), and instead vary with daily systemic signals that coincide with the transition from resting, fasting, and lower body temperature to activity, feeding, and higher body temperature (10), rather than external solar time. Thus, excepting the SCN, the major behavioral and physiological daily rhythms in mammals are set to opposite times of day in nocturnal and diurnal mammals (Fig. 1A), perhaps through mechanisms downstream of the SCN (4). Understanding how diurnal mammals integrate the same environmental cues to achieve an inversion of organismal and cellular physiology compared with nocturnal mammals is critical for understanding the internal synchrony that is pivotal for long-term health (11).

Fig. 1 — **(A)** Organismal and cellular physiology differ between nocturnal mice and diurnal humans, despite both species sharing the same cell-autonomous circadian clock mechanism. In both groups, the retinorecipient hypothalamic SCN shows daytime peaks of neuronal firing and PER expression, serving as an internal representation of the day-night cycle. Outside the SCN, however, rhythms in behavior, physiology, and cellular PER oscillations are inverted between nocturnal and diurnal mammals, aligning instead with systemic cues that coincide with rest-fast and activity-feeding transitions. This suggests a switch, downstream of the SCN, that controls the appearance of diurnality. Mouse icon created by Phạm Thanh Lộc from Noun Project CC-BY-3.0.

**(B)** Mouse (grey) and human (orange) primary fibroblasts (N = 3 individuals each, mean ± SEM) expressing *Bmal1:luc* were entrained to 7 × 12 h:12 h temperature cycles (37°C, red; 32°C, blue; final 3 temperature cycles shown) and released into constant 37°C. (Left) Bioluminescence was detrended and normalized to aid visualization of circadian phase. A vertical dotted line at t = 32 h illustrates species-specific phase of entrainment. (Right, above) Circadian period varied between individuals, as previously reported (97), but did not significantly differ between the species (Student’s t-test). (Right, below) Circadian phase, defined as the time of the peak of *Bmal1:luc* relative to the last transition to 37°C and plotted on a circular axis corresponding to a 24 h day, differed significantly between the species (Watson-Williams test; mean ± SEM).

**(C)** Primary fibroblasts from striped mouse (*Rhabdomys pumilio*), marmoset (*Callithrix jacchus*), gibbon (*Hylobates lar*), sheep (*Ovis aries*), rat (*Rattus norvegicus*) and lemur (*Lemur catta*) expressing *Bmal1:luc* were entrained as in (B) (final 1.5 temperature cycles shown) and released into constant 37°C (n = 6-8 per species, mean ± SEM). Time is plotted as circadian time, normalized to each species’ intrinsic period (0–24 h), such that one full cycle corresponds to circadian time 0-24 h, to aid cross-species comparison. A vertical dotted line at t = 32 h illustrates the divergent phases (left). (Right) Circadian phase of all species (mean ± SEM).

**(D)** Circadian phase in constant conditions from (B) and (C) for all 8 species (mean ± SEM). Two opposing clusters of phases correspond to diurnal vs nocturnal temporal niche (Watson–Williams). Phase did not correlate with ranked body size (circular Pearson’s).

**(E)** Cells from mouse (fibroblasts, immortalized) and human (U2OS) expressing PER2-LUC from the endogenous locus, entrained to 12 h:12 h cycles (37°C:32°C, 3 cycles shown) with opposite phases (n = 8, mean ± SEM). Note the high amplitude rhythms during temperature cycles that damp upon entry into constant temperature condition, and the different phase of entrainment between the two different reporters, *Bmal1:luc* (B) and PER2-LUC (E).

**(F)** Mouse (fibroblasts, immortalized) and human (U2OS) PER2-LUC cells were synchronized by medium change at t = -186h and subjected 12 h:12 h cycles (37°C:35.5°C, 3 cycles shown) shifted 6 h from the original phase (mean ± SEM). Both species re-entrained within 7 days and maintained opposite phases in constant conditions.

**(G)** Circadian phase for mouse (grey) and human (orange), calculated from (D) and (E), is given as peak of PER2-LUC expression relative to the last transition to 37°C (mean ± SEM). (Above) The phase difference between mouse and human was similar at 5°C cycles (solid circles) and 1.5°C cycles (open circles) (t-test; mean ± SEM). (Below) Phase differed significantly between mouse and human at each cycle magnitude (Watson-Williams; mean ± SEM).

At the cellular level, increase in synthesis of PER or global translation shifts the timing (or phase) of subsequent PER oscillations in both mouse and human cells (12–14). Physiologically, daily PER oscillations in cells throughout the body are synchronized and amplified by behavioral cycles of feeding-fasting, rest-activity, light-dark and stress exposure acting via specific systemic signals (7, 14–17), a process known as circadian entrainment (18). Hormonal cues such as insulin (feeding-fasting) (10, 14, 19, 20) and glucocorticoids (light-dark) (15, 16) entrain cellular clocks through conserved mechanisms in human and mouse (14, 21), reinforcing the behavioral patterns that drive their daily release. Cellular clocks can also be synchronized by daily rhythms in body temperature, which reflect heat production from activity and feeding, and cooling through peripheral vasodilation during sleep (22–27). Whether such temperature-mediated timing cues act equivalently on cells from diurnal and nocturnal mammals remains unknown.

Circadian synchronization by temperature is typically weaker than hormonal stimulation, with heat shock pathways (25, 26), cold shock proteins (28), cdc-like kinases (29), and upstream open reading frames (uORFs) (30, 31) having each been proposed to communicate temperature change to the cellular clock by a range of transcriptional and post-transcriptional mechanisms (32, 33). As these pathways are evolutionarily conserved, circadian responses to temperature change are assumed to operate analogously in mouse and humans and other mammals. Mouse and human biology can differ markedly, however, beyond obvious developmental differences. For example, mouse and human cells exhibit profoundly different biochemical reaction rates (34–37).

We found that the sensitivity of mTOR signaling and bulk protein synthesis to temperature are fundamentally different between nocturnal mice and diurnal humans, with physiological consequences that include the nocturnal-diurnal switch. We identified cell-autonomous differences between nocturnal and diurnal mammals in their response to thermal and osmotic challenge by specific (PER2 protein synthesis) and general (global phosphorylation and translation rate) mechanisms. We recapitulated temporal niche selection in vitro and found its cellular and molecular bases may be explained thermodynamically, not kinetically. We tested the functional consequences of modifying protein synthesis rates on temporal niche in vivo, and propose mTOR activity as a component of a signaling nexus that may integrate bioenergetic and thermodynamic cues into the cellular clock.

Cellular circadian rhythms of nocturnal and diurnal mammals are differentially entrained by daily temperature cycles

Circadian entrainment is defined by a stable relationship between circadian clock and external stimulus that persists after the cue is removed, distinguishing it from transient responses. Most systemic timing cues tested in vitro, such as hormones or metabolites, entrain the circadian clocks of mouse and human cells similarly, reflecting the activation of conserved cell-autonomous response pathways ((14, 21); fig. S1, A and B). Daily temperature cycles can also synchronize circadian clocks in mammalian cells in culture (23, 25, 26). We were therefore surprised to find that 12 hour 37°C:12 hour 32°C temperature cycles, which model daily body temperature rhythms (22, 38), synchronize primary skin fibroblasts from mice and humans to opposite phases (Fig. 1B, fig. S1C). Bmal1:luciferase (Bmal1:luc) waveforms, a conventional reporter of circadian rhythms, became antiphasic between species during the temperature cycle and remained so in constant temperature, confirming genuine opposite entrainment (Fig. 1B, fig, S1C).

This opposite entrainment mirrors the opposing temporal niches of the two species – nocturnal mice and diurnal humans – so we tested the generality of our findings in other nocturnal and mammals. Primary fibroblasts from nocturnal species (rats, mice, lemurs (39, 42)) and diurnal species (humans, gibbons, marmosets (39), sheep (40), striped mice (41)) consistently entrained in opposite phases under identical temperature cycles (Fig. 1C) without significant differences in circadian period (fig. S1, D and E) and irrespective of body size (Fig. 1D).These oppositely phased oscillations may reveal a fundamental difference in the way cells from different species respond to temperature.

To validate our findings and facilitate deeper mechanistic investigation, we repeated these experiments using an alternative reporter, PER2-LUCIFERASE (PER2-LUC aka PER2::LUC). PER2-LUC directly reports nascent production of endogenous PER2 protein, reflecting the molecular circadian clock in mammalian cells and tissues (43, 44). We compared fibroblasts from PER2-LUC mice (43) with CRISPR-Cas9 knock-in human PER2-LUC U2OS cells (fig. S1, F-G). Under a 5°C daily temperature cycle, mouse and human cells quickly developed and maintained oppositely phased rhythms (Fig. 1E, fig. S1, H and I).

To confirm our findings were not attributable to thermal stress responses we repeated the experiments using smaller 1.5°C temperature cycles. Again, we observed that human PER2-LUC showed an inverted phase relative to mouse PER2-LUC rhythms (Fig 1, F and G), with the only difference being a shift in absolute phase relative to the temperature cycle (Fig. 1G) and consistent with the theory that phase of entrainment varies with strength of stimulus (18). From these observations, we infer the existence of a cell-intrinsic signal inverter that reverses the response of diurnal and nocturnal mammalian cells. Understanding this signal inverter might provide insight into the nature of the mammalian nocturnal-diurnal switch.

Diurnal cellular clocks are buffered against temperature change

In vivo, the phase of cellular clocks is adjusted daily by systemic signals, dependent on both the magnitude of each stimulus and the relative biological time (circadian phase) at which they are received (14, 18). Single time-restricted stimuli, such as discrete temperature shifts, are sufficient to adjust cellular clock phase and quantified as the phase response (25, 31). Having found opposite responses of circadian clocks from diurnal and nocturnal cells to daily temperature cycles in the absence of other stimuli, we sought to elucidate the mechanism of signal inversion by using thermal changes as a tool. PER2-LUC-expressing immortalized mouse fibroblasts and human knock-in U2OS cells were subjected to single temperature increases or decreases at different circadian phases (fig. S2, A and B). Phase response curves (fig. S2, C and D) revealed species-specific sensitivity to single temperature shifts: although the qualitative direction of response was similar, over most of the circadian cycle human cells showed larger phase advances to temperature increases, while mouse cells exhibited larger phase delays to temperature decreases. These differences were sufficient to produce opposite entrainment under daily temperature cycles, (fig. S2E) explaining the cellular phenomenon but not its mechanistic origin.

Acute changes in PER protein production shift the phase of cellular clocks in vitro and in vivo (12, 14). We therefore tested whether differential sensitivity of mammalian cellular clocks to temperature change was reflected at the level of PER synthesis (Fig. 2). We drew on understanding of firefly luciferase enzyme kinetics (44, 45) to deconvolve the acute response to a rapid 5°C temperature increase (Fig. 2A) into two components. First, change in the baseline due to change in catalytic turnover of luciferase, which was not different between mouse and human cells (Fig. 2B); second, the change in total and peak luminescence that reflects the induction of PER2-LUC protein synthesis, which occurred more rapidly and produced more nascent PER2 in mouse than in human cells (Fig. 2C). At lower luciferin concentrations, which reflect steady-state PER2-luciferase concentration (44), the luciferase signal did not change with temperature over these short timescales (fig. S3, A and B). Thus it may be the synthesis of PER2 protein that responds to temperature change with a magnitude that differs between mouse than human cells. Over several cycles then, in principle, species-specific differences in the thermal sensitivity of PER protein production could function cumulatively to invert cellular clock timing. We therefore explored mechanisms by which temperature-dependent translation of PER2 might differ between species. We considered this could either occur by mechanisms that selectively regulate PER or by more general mechanisms that include changes in PER expression and activity.

**(A)** Raw bioluminescence data (arbitrary luminescence units, LU) from cells shifted from 32°C to 37°C at t = 0 h, trough of PER2-LUC expression (n = 8). Change in baseline (Δ baseline) reflects the temperature dependence of luciferase kinetics, and total signal (AUC synthesis, shaded) reflects new PER2-LUC synthesis.

(B and C) Quantification of change in baseline (fold, B), and synthesis (AUC, C), after a temperature shift at 0 h in mouse and human cells (unpaired Student’s t-test; mean ± SEM).

**(D)** (Left) U2OS cells stably expressing constitutive PER2-LUC fusions – human PER2-LUC (orange), mouse PER2-LUC (grey) or codon-optimized human PER2-LUC (teal) – were kept for 3 days at 32°C or 37°C before shifting temperature at t = 0 min. Fold induction of luminescence is shown relative to t = 0 min. (Right) Rate of induction (fold/h) was calculated from the gradient of the straight line fit from non-linear regression and compared between reporters (two-way ANOVA with by Šídák’s post-hoc test; mean ± SEM).

**(E)** U2OS cells expressing HaloTag from the endogenous PER2 locus were treated with 1 µM HaloPROTAC3 or DMSO control. (Left) Cells were lysed after 24 h for immunoblotting with anti-HaloTag and anti β-actin as a loading control. (Right) Cells ± 1 µM PROTAC cultured under 7 × 12 h:12 h temperature cycles (37°C, red; 32°C, blue; final 3 temperature cycles shown) and released into constant 37°C (n = 4, mean ± SEM). Difference in entrained phase between control and PROTAC treatment is given. PROTAC treatment advanced entrained phase relative to control.

**(F)** Mouse (NIH 3t3) and human (U2OS) cells (n=3) stably expressing constitutive LUC as a reporter of protein synthesis were exposed to the same temperature conditions as in (D). (Right) Rate of induction (fold/h) of constitutive LUC was calculated as in (D) (two-way ANOVA (TWA) with Šídák’s post-hoc test; mean ± SEM).

**(G)** Mouse and human primary fibroblasts were kept for 7 days at 32°C or 37°C before shifting temperature. Puromycin (10 µg/ml) was added 1.5h after temperature shift and cells lysed 30 min later. Representative immunoblot (left) shows one of three biological replicates (mouse or human) under four conditions (constant 37°C, shift down 37°C to 32°C, constant 32°C, shift up 32°C to 37°C). Fold change puromycin incorporation (N = 3 individuals, n = 4) was calculated relative to control and analyzed by two-way mixed-effects ANOVA with Šídák’s post-hoc test (interaction species × temperature: F(1,44) = 30.47, p < 0.0001; mean ± SEM).

**(H)** Fold change in protein synthesis rate between constant 37°C vs constant 32°C in primary fibroblasts of mouse (grey, N = 3 individuals, n = 4), human (orange, N = 3 individuals, n = 4), rat (light grey, n = 3), and striped mouse (maroon, n = 3) from Fig. 2G and fig. S6C (one sample t-test H₀ = 1; mean ± SEM).

**(I)** Mouse (grey) and human (orange) primary fibroblasts (N = 3 individuals, n = 6) were kept for 7 days at 32°C or 37°C. (Left) Detrended and normalized *Bmal1:luc* for the final 2 days in constant temperature is shown (mean ± SEM). Arrowheads indicate timing of peak expression of *Bmal1:luc* at 37°C (red arrowheads) and 32°C (blue arrowheads) and illustrate the difference in circadian period between the two temperatures. (Right) Circadian period (cycles/day) at constant 37°C (red) or 32 °C (blue) of the two species (two-way ANOVA with Šídák’s post-hoc test reported, interaction species x temperature: F(1, 67) = 56.75, p < 0.0001; mean ± SEM).

Compared with the clear mouse-human difference in PER2 translation and consistent with previous reports, we found no evidence for equivalent differences in the acute transcriptional response of Per2 to temperature change (fig. S3, C and D (Miyake et al, 2023)). Thus, signal inversion appears to occur post-transcriptionally. Per2 mRNA contains a temperature-responsive upstream open reading frame (uORF) that modulates translation of PER2 protein after physiological temperature increases (30, 31). The Per2 uORF is highly conserved among nocturnal and diurnal mammals however (fig. S3E and (31), and so is not an attractive candidate for species-specific differences in PER2 translation. In contrast, mouse or human PER2-LUC ORF expressed constitutively in mouse or human cells recapitulated the acute response of endogenous PER2-LUC to temperature change (Fig. 2D, fig. S4, A and B), indicating differential thermal sensitivity of PER2 translation is largely intrinsic to the coding region without requiring 5’- or 3’-UTR regulation. Rare codon usage and RNA secondary structure are common mechanisms of translational regulation that affect the synthesis of many proteins (46), including those with circadian function (47). Consistent with this, compared with wild type, codon-optimized hPER2 (hPER2-CO, fig. S4C) with less predicted mRNA structure (fig. S4D) showed minimal sensitivity to temperature change (Fig. 2D).

On the other hand, we identified only small differences in codon usage and predicted mRNA structure between mouse and human PER2 (fig. S4, C and D), with mouse PER2 slightly more structured than human. Moreover, PER1 in both species was highly similar (fig. S4, E and F). It was therefore important to distinguish whether PER2 itself was essential for signal inversion, or else simply associated with it. To directly test the contribution of PER2 to circadian synchronization by temperature cycles, we used CRISPR-edited human U2OS cells in which endogenous HALO-tagged PER2 could be acutely depleted using HALO-PROTAC3 (Fig. 2E, and fig. S5). Critically, when PER2 was acutely depleted, we found significant but small differences in the effect of daily temperature cycles on the diurnal cellular clock (Fig. 2E), consistent with previous reports (31). Therefore, although species-specific differences in PER2 translation may contribute to differential effects of temperature, they cannot be the sole basis for cellular signal inversion. From these data, we do not discount differences in the individual contributions of many other proteins, such as PER1. However, an alternative hypothesis is that general diurnal-nocturnal differences in the temperature-dependence of the translational machinery underlie the differences observed for PER2. Marked disparities in global biochemical reactions exist between species, with humans exhibiting generally slower rates and more stable proteins than mice (34–36). We therefore tested whether broader differences in the translational response to temperature change might underpin our observations.

Using constitutively expressed luciferase as a reporter for bulk 5'-cap-dependent translation, we found mouse cells were much more sensitive to temperature change than were human cells (Fig. 2F). Mouse cellular translation increased with temperature increase, and vice versa. By contrast, human cells showed an inverted response with reduced magnitude: reduced translation for temperature increase and no significant change for temperature decrease. This inverted response of protein synthesis to temperature was particularly stark over repeated temperature cycles (fig. S6A). We validated these findings by quantifying nascent polypeptide production with puromycin-labelling in primary fibroblasts (Fig. 2G, and fig. S6B). The differential effect of temperature on translation rate was also observed over longer timescales: after 1 week at constant 32°C or 37°C, mouse cell protein synthesis was temperature-dependent, faster at the higher temperature, whereas human cells showed no significant difference between the two temperatures (Fig. 2H). Translation in nocturnal rat cells likewise showed greater long-term temperature sensitivity compared with cells from the similarly sized but diurnal striped mouse (Fig. 2H, fig. S6C).

Circadian rhythms exhibit temperature compensation, in which, unlike most biological processes, the ~24h period of oscillation is only slightly affected by a change in ambient temperature (Q₁₀ of 0.8-1.2) (48). However, consistent with their increased translational sensitivity to temperature, the cellular circadian rhythms of nocturnal mammals showed increased temperature dependence relative to that of diurnal mammals. Mouse and rat circadian rhythms ran significantly faster at 37°C than 32°C, whereas circadian rhythms in human and striped mouse cells ran significantly slower at the higher temperature (Fig. 2I, fig. S6D). Thus, in nocturnal mammals, biochemical reactions appear to be more sensitive to temperature and run faster at higher temperatures than those of diurnal mammals. This provides an additional insight into cellular signal inversion that is complementary to the acute differences in thermosensitivity described above: during daily temperature cycles nocturnal cellular clocks accelerate at the higher temperature, whereas diurnal ones tend to slow down.

Potential role of mTOR and WNK in global species differences in the response to temperature change

Protein synthesis is principally controlled by phosphorylation of proteins of the translational apparatus, including members of the cap-binding complex eIF4F, 43S preinitiation complex, and the elongation factor eEF2 (49). To gain insight into differential responses to temperature, we performed quantitative (phospho)proteomics on biological replicates of primary mouse and human fibroblasts subjected to high or low temperature over either acute or extended timeframes (fig. S7A). We reasoned that thermosensitive phosphosites could impart directionality to temperature signals, including those that collectively control translation rate. To identify potential thermosensitive phosphosites we focused our analysis on those in which phosphorylation changed in proportion (fold change increases with temperature increase and vice versa) or in inverse proportion (fold change decreases with temperature increase and vice versa) with acute temperature change or longer-term temperature adaptation (Fig. 3, A and B and table S1).

Acute temperature shifts elicited marked phosphoproteomic differences between species (Fig. 3A): in mice, phosphorylation levels generally decreased at higher temperatures and increased at lower temperatures, whereas in humans the opposite pattern was observed. This inverse relationship mirrored their respective phase responses (fig. S2). A similar pattern was evident in the temperature-adapted phosphoproteome (Fig. 3B), consistent with widespread differences in the homeostatic mechanisms underlying ambient thermo-adaptation. In contrast, protein abundances in both species were much less sensitive to either acute or longer-term temperature changes (fig. S7, B to D, and table S1).

There was, however, little overlap between mouse and human cells in the identity of temperature-dependent phosphosites and of the proteins to which they belong (fig. S7E). The pathways previously identified as regulators of circadian temperature response, HSF1 signaling (25, 26) via Heat Shock Protein (HSP) 70 and HSP90, or the RNA binding proteins Cold-inducible RNA binding protein (CIRBP) and RNA-binding motif protein 3 (RBM3) (28) had similar proteomic responses to acute or long-term temperature change between mouse and human cells (fig. S7F). This aligns with the expected strong evolutionary conservation of the cellular response to temperature (50, 51), but not with a role in cell-intrinsic signal inversion. Therefore, to examine alternative regulators, we performed motif analysis for amino acids surrounding the phosphoacceptor (S or T or Y) to identify the kinases and/or phosphatases that might drive differential phosphoproteomic responses to temperature changes. We observed differences between mouse and human cells in both the direction and magnitude of response: in mouse, phosphosites surrounded by basic residues were highly enriched for inversely proportional phosphorylations; in humans this trend was reversed, apparent only in the -2/-3 positions, and with smaller magnitude (Fig. 3C, fig. S7G).

Among the many basophilic kinases that recognize basic residue motifs (52), two systems emerged as dominant in our data. The first, the AGC family, includes central regulators and effectors of the mechanistic target of rapamycin (mTOR) pathway – a major pathway for controlling protein synthesis, macromolecular crowding and cellular metabolism, that integrates diverse metabolic and extracellular signals (53–55). The second, the with-no-lysine (WNK) kinases, are master sensor-effectors of the WNK, SPS1-related proline/alanine-rich kinase (SPAK) and SPAK homolog oxidative stress-responsive kinase 1 (OSR1) pathway that maintains intracellular water balance (56–58). Although only a small fraction of homologous phosphosites were conserved between mouse and human and showed opposite temperature-dependent changes (acute: <3%; adaptation: <1%; fig. S7E), these rare sites were notably enriched for regulatory components of WNK and mTOR signalling (Fig. 3, D and E). WNK1 and mTOR are highly conserved, essential proteins that function as major determinants of translation and cellular homeostasis whose activities are co-regulated and circadian-modulated in cells and in vivo (53, 58, 59). Thus, nocturnal-diurnal differences in their activities could plausibly participate in the phenotypic switch.

Convergent evolution of diurnal response to thermodynamic perturbation

In concentrated macromolecular solutions like the cytosol, small changes in temperature elicit large changes in the total thermodynamic potential energy of water, or water potential. As temperature decreases, more water molecules become bound within macromolecular hydration layers, reducing the fraction of ‘free’ bulk solvent and lowering the potential energy available for cellular work (60). This relationship deviates from simple linearity predicted by Van’t Hoff’s equation, both in magnitude and direction. Thermosensitivity in biological systems can therefore arise either from direct kinetic effects or from components that respond to changes in solvent thermodynamics. These mechanisms can be distinguished by testing whether an equivalent change in water potential mimics a temperature shift (60). For example, increasing external osmolarity should phenocopy cooling, as osmotic water loss lowers intracellular water potential by removing free water. WNK and mTOR signaling pathways are sensitive to such changes in water potential: phosphorylation of OXSR1-S339 and AKT1-T450 increases with extracellular osmolarity and decreases with temperature (60). We therefore tested whether temperature entrainment might operate by a thermodynamic mechanism. Indeed, mouse and human cells entrained to opposite phases under daily extracellular osmolality cycles (Fig. 3F; fig. S8, A to D). By contrast, a purely kinetic mechanism should depend on absolute temperature. When subjected to cycles of the same 5°C amplitude but a lower mid-point (28°C:33°C vs 32°C:37°C), mouse and human cells still entrained to opposing phases, inconsistent with simple kinetic explanations (Fig. 3G).

Collectively, these results supported a model in which WNK and mTOR pathways contribute to an intrinsic nocturnal-diurnal switch by virtue of species-specific differences in their response to thermodynamic changes in the intracellular environment. Diurnality evolved several times, likely acting through complementary changes at many genetic loci that are assumed to differ between diurnal lineages. However, if changes in WNK and mTOR activity are an efficient evolutionary means to select for diurnal phenotypes then convergent evolution might be detected by comparative genomics. We therefore mined the Zoonomia comparative genomics resource of placental mammals (61) to examine whether members of these pathways are amongst those genes that evolved particularly quickly in the genomes of diurnal mammals relative to nocturnal mammals. Of the 242 species analyzed, 77 and 109 were categorized as definitively diurnal and nocturnal, respectively, from prior literature (fig. S8E and table S2). We restricted the analysis to widely expressed genes – excluding tissue-specific genes such as olfactory receptors (fig. S8, E and F) – to identify candidates that could contribute cellular phenotypes. Among genes predicted to have evolved significantly faster in diurnal mammals we found members of the WNK-SPAK/OSR1 pathway (WNK1 and Solute carrier family 12 member 1 (SLC12A1)), and regulators of mTOR and protein synthesis (Ras-related GTP-binding protein B (RRAGB), a core regulator of mTOR complex 1 (mTORC1) activity (53), and translational quality control factor ZNF598 (62)) (Fig. 3H). Faster evolutionary rates in an additional key regulator of mTORC, Tuberous sclerosis complex 2 (TSC2), and a second paralogue of WNK1, WNK4, correlated with diurnality, but lay just outside the phylogeny-corrected significance threshold suggesting they might have evolved faster in a subset of related diurnal mammals (table S3).

Comparative analyses suggested the emergence of diurnality in mammals was linked with differential sensitivity of mTOR and WNK pathways to applied changes in solvent thermodynamics. In general, more disordered proteins have higher propensity to form supramolecular assemblies when the water availability becomes limiting (60). mTOR activity is governed by many subunits and interactions that could individually or cooperatively affect sensitivity to solvent thermodynamics so we focused on WNK1 to validate our findings, because its self-assembly and activation is directly and acutely sensitive to water potential (57, 60). In cells, reductions in water potential due to increased macromolecular crowding, cooling or hyperosmolarity promote WNK condensation and subsequent activation via its intrinsically disordered C-terminal tail (57, 60). We therefore tested whether the lower responsiveness of diurnal species to temperature change is reflected in lower intrinsic disorder of WNK1. Consistent with this model, WNK1 from diurnal mammals showed lower predicted disorder than WNK1 from nocturnal mammals (Fig. 3I, fig. S8G), consistent with reduced sensitivity.

Comparative genomics therefore was consistent with our hypothesis that the nocturnal-diurnal switch arose convergently and independently through multiple complementary mutations that act together to alter the cellular sensitivity (e.g. WNK pathway) and responsiveness (e.g. mTOR pathway) to perturbation of cellular thermodynamic equilibria by modulating the favorability of key macromolecular interactions. This differentially affects circadian phase through a combination of specific (PER synthesis) and more general mechanisms (basophilic kinase activity, bulk translation) that ultimately renders human circadian clocks more robust to thermal and osmotic perturbation than those in mice. Under repeated daily thermodynamic perturbations, this results in entrainment to opposing phases (fig. S2E). We propose that cellular clocks respond to crowding-related changes in macromolecular hydration and supramolecular assembly rather than changes in solute kinetic energy, as was implicitly assumed.

Perturbation of mTOR activity and translational initiation makes nocturnal cells behave like diurnal cells

Decreasing the basal protein synthesis rate by inhibiting mTOR activity should attenuate the capacity of nocturnal circadian timekeeping to respond to thermal challenge more than it does diurnal circadian timekeeping, rendering them more diurnal-like by reducing the relative magnitude of temperature-dependent differences in translation between the two. Whereas a panel of small molecule inhibitors of proteins and kinases previously implicated in circadian post-translational regulation revealed only small effects on entrained phase under daily temperature cycles (fig. S9, A to D), inhibitors targeting the mTOR signaling pathway showed large effects on entrained phase (Fig. 4A and fig. S9, E and F), with selective mTOR inhibition by sapanisertib (INK128) showing the largest effect. When mTOR activity was decreased, mouse cellular rhythms shifted to become similar to those of human cells (Fig. 4A). INK128 treatment of fibroblasts from another nocturnal mammal (rat) and diurnal mammal (striped mouse) gave comparable entrainment phenotypes, demonstrating functional conservation of this pathway in temperature signaling (Fig. 4B). Growth factor signaling acts through mTORC1 to control protein synthesis rates (53), and can be manipulated in cell culture by changing serum concentration. In lower serum concentrations, mouse cellular rhythms were delayed by up to 6 hours under daily temperature cycles compared with high serum control conditions, whereas human cells were not (Fig. 4C). In all cases, suppression of mTOR activity made cells from nocturnal mammals behave more like cells from diurnal mammals, with PER2-LUC peaks selectively shifting towards the early warm portion of the temperature cycle, whereas the contrary was not true for diurnal cells.

Fig. 4 — (A and B) Mouse (fibroblasts, immortalized) and human (U2OS) cells expressing PER2-LUC (A) and striped mouse and rat fibroblasts expressing *Bmal1:luc* (B) were entrained to 7 × 12 h:12 h temperature cycles (37°C, red; 32°C, blue; final 4 temperature cycles shown) in the presence of mTORC1/2 inhibitor (1 µM INK128) or vehicle control. (Left) Bioluminescence was recorded, detrended and normalized to visualize circadian phase during entrainment. (Right, above) Circadian phase under control (black) and INK128 (blue) conditions is shown for human and striped mice (circles) and mouse and rat (triangles) as circle plots. (Right, below) Change in phase relative to control (n = 4 to 6 each condition, two-way ANOVA with Šídák’s post-hoc test between species reported; mean ± SEM).

**(C)** Human and mouse PER2-LUC fibroblasts, cultured in decreasing concentrations of serum, were entrained to 7 days of temperature cycles before transfer to constant conditions. Dashed line at the peak of 10% serum control is shown for illustration purposes. (Right, above) Circadian phase under control (black) and 1% serum (medium grey) and 0% serum (light grey) conditions is shown for human (circles) and mouse (triangles). Colored lines indicate human (orange) and mouse phases (grey). Change in phase, relative to 10% serum control, is shown below (n = 6 human, n = 19 to 24 mouse, two-way ANOVA, interaction species x condition: F(1, 52) = 21.54, p < 0.0001; mean ± SEM).

**(D)** Human and mouse PER2-LUC cells, kept in constant conditions, were treated with inhibitors of the cap-binding complex that target eIF4A (rocaglamide, rocA; hippuristanol) at the trough of PER2-LUC at t=-12h, indicated by an arrow. A vertical dashed line at the peak of DMSO control is shown for illustration purposes. (Right, above) Circadian phase under control (black), rocA (pink) and hippuristanol (purple) treatment is shown for human (circles) and mouse (triangles). (Right, below) Change in phase, expressed as a delay relative to control, in mouse and human after treatment (n = 4, two-way ANOVA, interaction species x condition: F(1, 17) = 387.1, p < 0.0001; mean ± SEM).

mTOR inhibition was not sufficient to make nocturnal cells completely phenocopy diurnal cells, however, and we note several other essential genes showed faster evolution in diurnal than nocturnal species including translational regulators (table S3). We therefore assessed how robust human versus mouse cellular circadian rhythms are to acute perturbation of translation rate by pharmacological attenuation of 5'-cap-dependent translational initiation, independently of mTOR or temperature. Circadian clocks drive and are driven by daily cycles of protein synthesis (13, 63, 64), amplified in vivo by daily timing cues such as insulin signaling linked with feeding and fasting cycles, which act through the translational machinery (14). When treated at the same circadian phase, significant differences were observed in the magnitude and direction of circadian phase shifts between mouse and human cells in response to direct inhibitors of eIF4A, rocaglamide A (RocA) and hippuristanol (Fig. 4D). Again, the cellular clock in mouse cells was much more sensitive than in human cells, consistent with the idea that natural selection has led to increased increased robustness of the cellular clockwork in diurnal mammals.

mTOR regulation of circadian phase is maintained from cells to tissues

The function of the mTOR pathway as a cellular signaling nexus for translational regulation is conserved across eukaryotes and essential in mammals. We propose that differences in mTOR regulation and activity constitute a major element of the nocturnal-diurnal switch. If so, mTOR inhibition should render circadian clocks in mouse tissues more diurnal in their response to daily temperature cycles, both ex vivo and in vivo.

To test this, we subjected tissue explants from adult PER2-LUC mice to daily temperature cycles with or without mTOR inhibition. As expected (25), high amplitude PER2-LUC oscillations were observed in neuroendocrine (pituitary) and non-neuronal (lung, adrenal) tissues, with PER2 consistently peaking around the warm-to-cold transition (Fig. 5A), as in mouse fibroblasts (Fig. 1E). Inhibition of mTOR activity by INK128 resulted in significant phase delays of 8-12h with PER2 now peaking near the cold-to-warm transition (Fig. 5A), and so more similar to human than mouse cells in vitro (Fig. 1E).

Fig. 5 — **(A)** Pituitary (square), adrenal (triangle) or lung (circle) tissues were dissected from PER2-LUC mice at lights on (rest phase onset for nocturnal mice), cultured *ex vivo* with luciferin, treated with DMSO (black) or INK128 (blue), and exposed to 3 × 12 h:12 h temperature cycles (37°C, red; 32°C, blue) which were antiphasic to previous activity patterns. (Left) PER2-LUC bioluminescence on 2^nd and 3rd temperature cycle and first 3 days of constant conditions are shown for each tissue and condition (mean ± SEM). (Center) Circadian phase of entrainment, calculated given relative to the final transition from 32°C to 37°C, for each tissue under each condition is shown, with arrows indicating the direction of the phase shift. (Right) Change in phase, expressed as a delay relative to control (n = 4 to 6, two-way ANOVA, interaction tissue x condition: F(2, 25) = 18.04, p < 0.0001; mean ± SEM).

**(B)** SCN tissue was dissected from PER2-LUC mice, sliced, and placed into *ex vivo* culture in the presence of luciferin. Slices were entrained for 7 days in antiphasic temperature cycles (12h 37 °C:12h 32°C, black; reversed, grey; mean ± SEM) before release into constant 37 °C. (Right) Circadian phase relative to t = 0 h (n = 4 to 5, Student’s t-test, unpaired).

**(C)** SCN slices from PER2-LUC mice kept under light-dark cycles were kept at constant temperature (37 °C) or exposed to 3 × 12 h:12 h temperature cycles (37°C, red; 32°C, blue) in antiphase to the LD cycle from which the mice were taken. *Ex vivo* culture started at 0 h when slices were given either INK128 (blue) or DMSO (black). (Left) PER2-LUC bioluminescence is shown relative to the subjective light period (0 h = light on; light on, pale yellow; light off, dark blue; mean ± SEM). (Right, above) Circadian phase of entrainment, relative to subjective light on. Subjective light cycle is shown underlying data points. (Right, below) Change in phase, relative to control (n = 3 to 5, Student’s t-test, unpaired; mean ± SEM).

**(D)** (Top) Adult male mice were singly housed with wheels and food pellets delivered in proportion to wheel revolutions under 12 h light: 12 h dark cycles by Hut et al. (72). In the baseline period, 1 pellet/100 revolutions was provided; in the WFF group the ratio gradually increased to 1 pellet/300 revolutions, reducing mTOR-dependent processes (71). Running wheel activity and core body temperature were recorded throughout. (Middle) Average daily activity profile across 7-day baseline and test periods in the WFF group (N=11 mice, mean ± SEM) where time 0 h represents midday. (Bottom) Acrophase of activity (left) and diurnality index (% of activity in light, right) for control (N = 10 mice) and WFF groups (N = 11 mice) (two-way ANOVA with Šídák’s post-hoc test between baseline and test period reported; mean ± SEM).

**(E)** Adult mice (N = 23, 12 male, 11 female) were singly housed in cages in long day conditions (14 h light: 4 h twilight: 6 h dark) for 1 week, then split into control (N = 11 mice) and low methionine (low met, 25% of control; N = 12 mice) diet groups for 4 weeks. Food and water was provided *ad libitum* throughout. (Below) mTOR activity in liver and brain was assessed by immunoblot in control and low methionine mice. mTOR activity is expressed as a % of control (two-way ANOVA with Šídák’s post-hoc test between control and low met reported; mean ± SEM).

**(F)** Representative actograms for control and low methionine mice. Actograms are double plotted for visualization, where each row represents two days, the second of which is presented on the following row directly beneath the first day. White, light grey and dark grey represent light, twilight and dark respectively.

**(G)** (Left) Average daily activity profile across days 15 to 37 for all mice (N = 11 control, N = 12 low methionine, mean ± SEM). Light conditions are indicated relative to midday as t = 0h with twilight (light grey) and darkness (dark grey) periods shaded. (Right) Activity acrophase (top) and onset (bottom) for control (con) and low methionine (low met) groups (Welch’s t-test).

**(H)** Model: the mTOR pathway acts as a cellular hub to integrate thermodynamic challenges and energy balance to regulate cell-intrinsic circadian clocks. Species-specific alterations in genes including RRAGB, WNK1, and ZNF598 ultimately render diurnal cellular clocks more robust to perturbation than nocturnal, an evolutionary change that can be recapitulated in vitro and in vivo through reduction in mTOR activity. Mouse icon created by Katy Lawler from PhyloPic CC-BY-4.0.

The hypothalamic suprachiasmatic nuclei (SCN) of nocturnal mammals are remarkable for PER rhythms that are essentially opposite to almost all other tissues, in the same phase as SCN of diurnal mammals (Fig. 1A). This is consistent with the SCN's conserved function in all mammals as a dedicated photic timekeeper, responsible for encoding and communicating anticipated photoperiod. Interneuronal coupling renders SCN PER rhythms more robust than other tissues and much more sensitive to photic cues than to systemic signals such as temperature (14, 25, 65). Adult SCN are sensitive to temperature, however, and explants stably entrain to the same daily temperature cycles used in this study (Fig. 5B) (66, 67). After 7 daily temperature cycles, SCN entrained with a phase that was later than that of other tissues, with PER2 peaking late in the cold phase (Fig. 5B), at the end of the subjective day (Fig. 5C), reminiscent of the difference in circadian timing between SCN and other mouse tissues in vivo. The phase of SCN rhythms remained unaltered by mTOR inhibition (Fig. 5C). Uniquely for nocturnal tissues then, SCN phenocopies the increased robustness of cellular and tissue clocks from diurnal species, likely through reduced sensitivity to changes in mTOR activity, which diminishes the responsiveness of the protein synthesis machinery to temperature change. Indeed, it is long established that SCN timekeeping is robust to abrupt changes in translation rate through network coupling (64). Because SCN activity in nocturnal mammals aligns with daytime, as for diurnal mammals (Fig 1A), mouse brain temperature rhythms would be expected to reinforce, rather than disrupt, the SCN's established relationship with the light:dark cycle in vivo. Our findings support a model in which all mammalian SCN maintain an mTOR-insensitive representation of daytime, whereas the timing of behavior and physiology outside the SCN is governed by the interaction between cell-autonomous timekeeping and timing cues – such as temperature, osmolarity and growth factors – that regulate global and specific (PER) protein synthesis via mTOR pathway signalling.

mTOR activity contributes to temporal niche selection

We used a pharmacological active site inhibitor of mTOR (INK128) (68) to demonstrate that, in vitro, modifying the basal activity of this pathway differentially alters cell-intrinsic responses in nocturnal vs diurnal mammals, implicating mTOR pathway activity as contributing to the nocturnal-diurnal switch. However, diurnality or nocturnality are behavioral phenotypes in which the timing of locomotor activity defines temporal niche classification; and ultimately occurs in the brain in vivo. We considered that if mTOR pathway activity within the central nervous system contributes to temporal niche selection, then perturbation of mTOR activity at the organismal scale would be expected to alter the daily organization of locomotor activity.

Under dietary-restricted conditions, as may occur in the wild, mTORC1 activity and protein synthesis are greatly reduced (53, 69). Accordingly, mouse behavior becomes more diurnal than when animals are fed ad libitum (70). We sought to replicate these observations in mice under laboratory conditions. Precise control of energy balance in mice can be achieved in the Work for Food (WFF) paradigm, under which food is limited (Fig. 5D, fig. S10A) and mice lose body mass according to the negative energy balance imposed upon them (fig. S10B) (71). Under these conditions, mTOR activity is decreased in multiple brain and peripheral tissues (fig. S11). Compared with control conditions (food ad libitum) where mice are nocturnally active (fig. S10, C and D), under the negative energy balance conditions of WFF, mice apportion more of their activity to the daytime, like a diurnal mammal, (Fig. 5D, fig. S10, C to E, (72)), which is matched by advanced timing of core body temperature rhythms towards daylight hours (fig. S10, F-H).

WFF demonstrates that, in principle and without affecting the SCN ((73); fig. S11), it is possible for a nocturnal mouse to behave more diurnally under conditions when mTOR pathway activity is reduced (fig. S11). WFF has many more consequences than reduced mTOR activity however (71, 74). Therefore, to validate these results we targeted the mTOR pathway more selectively via isocaloric reduction of dietary methionine. Methionine restriction inhibits mTOR activity through amino acid sensing by a Rag-dependent signaling pathway (53, 75); fig. S10I) and is well-tolerated in mice over several weeks with minimal weight loss (fig. S10, J-M), unlike methionine starvation or sustained pharmacological inhibition. Over four weeks on an isocaloric ad libitum methionine-restricted diet, mTOR activity in the brain was significantly reduced (Fig. 5E). Under these conditions, both the onset and peak of mouse locomotor activity was significantly phase-advanced into the daylight hours relative to control (Fig. 5, F and G) with no change in locomotor period (fig. S10M). These activity shifts towards earlier behavior evoked by partial attenuation of mTOR activity are consistent with the cellular data, and support a molecular mechanism whereby the basal mTOR activity modulates the response to physiological entraining cues. We do not suggest that differential mTOR pathway activity is the sole basis for diurnality in mammals, WNK pathway is also evolutionarily implicated, for example. This is the first relevant mechanism, however, whose biophysical and cellular basis can be understood to act akin to a nocturnal-diurnal switch in vitro and in vivo (Fig. 5H).

Discussion

Mammalian colonization of the daytime accelerated when its previous occupants, the dinosaurs, became extinct (1). Subsequently, mammals came to occupy all temporal niches, frequently switching between them as life history and environment dictated (4). The specific mechanism governing the switch between nocturnality and diurnality was previously unknown, but had been assumed to emerge from specific diurnal neuronal circuits rather than being intrinsic to the biology of all diurnal cells, including neurons. We identified an apparent cell-intrinsic inversion of the response of the molecular circadian clockwork to entraining cues that correlated with the donor’s temporal niche. Complete inversion required several days, dependent on the magnitude of the entraining stimulus, and was largely attributable to differences in the activity and sensitivity of the mTOR and WNK pathways. This leads to broad consequences for cellular protein synthesis and phosphorylation, that ultimately render diurnal cellular clocks more robust to perturbation than nocturnal. Both pathways showed accelerated molecular evolution in diurnal compared with nocturnal mammals, and experimental modulation of mTOR activity in cultured cells, tissues or in vivo could recapitulate the switch from nocturnal towards more diurnal activity.

A comparable principle has been recently recognized in developmental biology, where mammalian species show marked differences in global biochemical reaction rates which correlate with developmental tempo (34–36). Our findings extend this concept to encompass osmotic homeostasis, signal transduction and circadian physiology of non-dividing, terminally differentiated cells. Nocturnal-diurnal differences were particularly apparent in sensitivity to perturbations that affected intracellular solvent thermodynamics, appearing to converge on mTOR activity as a plausible and major component of the nocturnal-diurnal switch. mTOR complexes 1 and 2 have multiple substrate effectors as part of the large and interlinked PI3K-AKT-mTOR pathway, which regulates and is regulated by multiple cellular signaling systems (53), including the WNK-SPAK/OSR pathway and the cellular circadian clockwork itself (58, 59). Of course, other temperature-sensitive kinases, phosphatases, and signaling mechanisms, acting upstream, downstream or in parallel may also contribute. Moreover, the slow re-organization of physiology under amino acid restriction or WFF occurs over several weeks, and does not completely recapitulate diurnal behavior (72, 74), suggesting the involvement of additional processes that may include hypothalamic neuroplasticity, melatonin signal inversion or direct photic modulation of locomotor activity, for example (76, 77).

Ultimately though, any switching mechanism that arose evolutionarily must have a genetic basis. We demonstrate this through a genome-wide comparison of diurnal and nocturnal mammals, which provides complimentary genetic evidence for the importance of mTOR activity with key proteins, RRAGB and WNK1, having faster evolutionary rates in diurnal versus nocturnal mammals. We consider genetic mechanisms of diurnality may be broadly dispersed and polygenic, and to this end we also found evidence for faster evolutionary rates in olfactory pathway genes (fig. S9) and phototransduction genes (78).

At the whole organism level, our findings agree with the circadian thermo-energetics (CTE) hypothesis for conditional niche-switching in several different mammals (79). CTE states that nocturnal activity patterns for homeothermic mammals are more costly than diurnal patterns, because nocturnal animals have higher energy requirements to mitigate the greater heat loss of being active during the (cold) night (73, 79). Diurnality arises as an energy saving measure when food availability is scarce, which outcompetes the extra predation pressure of being active by day (79). At the cellular level, these results support the bioenergetic hypothesis for circadian and other biological rhythms (80), where oscillations primarily function to minimize the high cost of maintaining protein homeostasis. In this context, the increased resistance to translational perturbation in cells from diurnal mammals is thus an energy saving measure, and will diminish the cellular challenge of conflicting timing cues. It would be interesting to investigate whether birds – which independently evolved diurnality, homeothermy and have a higher basal core body temperature than mammals (81), as well as marked heat stress and specialized thermoregulation during flight (82) – use the same mechanism.

Overall, our findings indicate marked nocturnal-diurnal differences in cellular biochemistry and global pathway activity, influencing cellular circadian timing and daily physiology in vivo. They add to a growing literature demonstrating species-specific differences in molecular activity which map to cellular or organismal phenotypes (34–37, 83, 84). Strikingly, many of these differences involve global regulation of protein turnover and the mTOR pathway, integrating metabolic status with functional output (84, 85). By linking cellular biochemistry, genomics, and organismal behavior, our results reveal that the timing of life itself is embedded in species-specific cellular machinery, providing a mechanistic bridge between molecular clocks, metabolism, and the evolution of daily activity patterns.

Materials and Methods

Cell culture

Primary fibroblasts from mice (Mus musculus, wild-type C57BL/6), humans (Homo sapiens), common marmoset (Callithrix jacchus), white-handed gibbon (Hylobates lar), ring tailed lemur (Lemur catta), common sheep (Ovis aries), and four-striped grass mouse (Rhabdomys pumilio) were obtained as following. Tissue biopsies were taken post-mortem under local ethical approval from adult animals: from a male gibbon and a male lemur, both euthanized for management reasons at Copenhagen Zoo; from two male and two female mice at the MRC Laboratory of Molecular Biology; from a sheep and from a marmoset (sex not recorded) who died of natural causes at University of Cambridge. An ear punch biopsy was used from a striped mouse at University of Manchester. Fibroblasts were isolated following standard culture (86) from skin (human, mouse and striped mouse), lung (mouse), heart (lemur, sheep) or muscle (marmoset, gibbon). Adult human skin (dermal) fibroblasts were obtained from healthy volunteers (two male, one female) at the John van Geest Centre for Brain Repair (Cambridge, UK), under local ethical approval (REC 09/H0311/88) using a previously described method (87). Rat (Rattus norvegicus) fibroblasts were derived from Rat-1 cells (male, RRID:CVCL 0492). Fibroblasts were kept in Dulbecco's modified Eagle medium (DMEM) high glucose, GlutaMAX(TM), pyruvate medium (Gibco, #31966) supplemented with 10% fetal bovine serum (FBS, Gibco, #10437028) or HyClone FetalClone III serum (Cytiva Life Sciences, #SH30109) and 1% penicillin/streptomycin (P/S, final concentration: 100 units/mL of penicillin and 100 µg/mL of streptomycin, Gibco, #15140122) and passaged when they reached 80–90% confluency. All lines were cultured to at least passage 2, tested for mycoplasma using MycoAlert™ mycoplasma detection kit (Lonza, #LT07), and maintained for routine culture in DMEM supplemented with 10-20% FetalClone III serum and P/S at 37°C, 5 % O₂, 5% CO₂. For creation of bioluminescent lines, cells were transduced with pLV6-Bmal-luc lentivirus (gift of Steven Brown, also found at Addgene plasmid #68833) and selected for stable cultures by blasticidin (blasticidin-HCL, Gibco) selection at 4 µg/ml. Selection was performed for between 1 and 2 weeks until non-transduced control cells were dead, and colonies were established in transduced lines. Lines were subsequently maintained at 2 µg/ml blasticidin in DMEM containing 10% FetalClone III serum and penicillin/streptomycin at 37°C, 5% CO₂. Experiments were performed at between 1 and 5 passages post transduction, omitting blasticidin in the experimental culture medium.

Immortalized lung fibroblasts were obtained from lung tissue of adult PER2-LUC mice (43) and immortalized by serial passage as described in Crosby et al. (14). Immortalized fibroblast lines were cultured as described previously (14), cultured in DMEM/10% FetalClone III/P/S as above for primary lines, but maintained at atmospheric O₂.

Human U2OS (also known as U-2 OS) cells, (ATCC, #HTB-96, RRID:CVCL 0042), were cultured in DMEM (Gibco, #31966) supplemented with 10% HyClone FetalClone II serum (Cytiva Life Sciences, #SH30066) and P/S at 5% CO₂. For creation of stably transfected lines, cells were transfected with linearized plasmid DNA with Genejuice (Merck, #70967) at 3 µl/µg DNA according to manufacturer’s protocol. Between 24 and 48 h after transfection, cells were selected for stable transfection by supplementing medium with puromycin (puromycin dihydrochloride, Gibco, #12122530) at 4 µg/ml.

U2OS cells endogenously expressing the PER2-LUCIFERASE or PER2-HaloTag fusion proteins were obtained as following. Guide RNA for Cas9 editing at the stop codon of PER2 were designed using a combination of CHOPCHOP v2 (88) and ATUM gRNA design tool (atum.bio/eCommerce/cas9/input) and default parameters: 5’ATGGATCCCCCTTGAATCAC-3'. Guide RNA sequence, lacking the PAM sequence, were inserted as annealed complementary single-stranded oligos into BbsI-digested pSpCas9-2A-GFP (PX458, Addgene). HDR templates, inserting the coding region lacking the initial methionine codon of human codon-optimized firefly luciferase (fig. S2A) or human codon-optimized HaloTag (Promega, fig. S6A) immediately 5’ of the endogenous stop codon, were created by NEB Hifi assembly or direct gene synthesis respectively. Homology arms of ~1000 bp were included either side of the insert coding sequence. U2OS cells were transfected with a 3:1 molar ratio of linearized HDR template:Cas9 sgRNA using PEI at 3ug/ug DNA. On the third day after transfection, GFP+ve cells were single-cell sorted and expanded as clonal cell lines in 96-well plates. Clonal lines were expanded until confluent, changed into medium containing 1mM firefly luciferin (Biosynth, #L-8220) and placed in an ALLIGATOR Luminescence System (“Alligator”, Cairn) at 37°C and 5% CO₂ for screening of positive colonies (89). Positive clones were identified by presence of luminescence, which oscillated in brightness with a period of approximately 24 h in constant conditions. Positive clones were validated by PCR (fig. S2B) and functional assays (fig. S2, C to E).

Plasmids

Constitutive expression constructs for mouse PER2-LUC, human PER2-LUC and human PER2_codon optimized-LUC (PER2_CO) were generated as following. PER2 ORF from mouse (GenBank NM_011066.3) or human (NM_022817) was synthesized by Twist Bioscience gene synthesis. Codon optimized human PER2 was synthesized by Twist Bioscience gene synthesis, and codons were optimized using Twist Bioscience Codon Optimization tool. This tool limits the selection of codons for an amino acid with a cutoff frequency of 8%. Codon usage frequency was assessed using Codon Plot from the Sequence Manipulation Suite (https://www.bioinformatics.org/sms2/codon_plot.html) using codon frequencies for Homo sapiens from the Codon Usage Database (https://www.kazusa.or.jp/codon/). mRNA structure was assessed for the short translational ramp of 30 amino acids (90) using RNAFold web server (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi) using minimum free energy (MFE) function settings. ORFs were synthesised flanked with SpeI and NotI sites and with no stop codon. ORF-LUC fusions were generated in a stepwise subcloning protocol, following MXS chaining assembly protocol as following. ORFs were cloned into a backbone containing SV40 promoter. Luciferase, plus SV40pA sequence, was subcloned into this backbone with NotI and SalI to generate an in-frame fusion of ORF and luciferase, under the control of SV40 promoter, in a vector containing puromycin resistance. Constitutive LUC was generated similarly, omitting the ORF subcloning step.

Experimental cell culture - static

For static culture bioluminescent experiments, cells were plated at confluence in transparent bottom black 96-well plates (Greiner, µClear #655096) at 100 µl/well in standard culture medium: DMEM, 10% serum (FetalClone II for U2OS, FetalClone III for fibroblasts), 1X P/S. The following day, medium was replaced with experimental culture medium which was identical to the standard culture medium except for the addition of D-luciferin potassium salt (Biosynth, #L-8220) at 0.037 mM to 1 mM (to deconvolve luciferase signal into steady state and nascent production (44)), and serum concentrations from 0% to 10%, both as described in figure legends. For examining the effect of amino acid restriction on cellular rhythms, DMEM (Gibco, #31966) was diluted to 10% concentration with EBSS (Gibco, #24010), and then supplemented with sodium bicarbonate (final concentration 44 mM), pyruvate (final concentration 1 mM), glucose (final concentration 25 mM), P/S (final concentration 1X) and serum (FetalClone III, final concentration 1%). For bioluminescence recording, luciferin was added to final concentration of 0.3 mM. Osmolality of experimental medium was measured with the Osmomat 3000 freezing-point-depression osmometer (Gonotec) and adjusted to 330 ± 10 mOsm/kg using NaCl. Osmolality of media was controlled in all conditions to remain isosmotic before and after treatment, except where change in extracellular osmolality was the treatment variable. Here, osmolality was adjusted with raffinose (D-(+)-Raffinose pentahydrate; Sigma, #R0514).

Entrainment was performed as described in figure legends and described below. For temperature entrainment, cells were placed in humidified incubators at 5% CO2 which cycled 12h 37°C: 12h 32°C. At the transition from 32°C to 37°C or 37°C to 32°C, plates were transferred isothermally into dark, light-tight, humified and 5% CO₂ Alligator at 37 °C or 32°C respectively. Where indicated, temperature entrainment took place in humified and 5% CO₂ Alligators (cycling 12h 37°C: 12h 32°C; 12h 37°C: 12h 35.5°C; or 12h 33°C: 12h 28°C) to enable recording of bioluminescence under cycling conditions. In the latter case, cells were allowed to adapt to the lower temperature of 33°C for 5 days before cycling commenced. In cases where entrainment took place in the Alligator, cells were not disturbed at the transition from cycling to constant conditions.

For dexamethasone or serum synchronizations, after plating, the following day medium was replaced with experimental culture medium containing 10% serum ± 100 nM dexamethasone. Serum synchronization and dexamethasone treatment sets mouse and human cells to identical phases (fig. S1A). Cells were then moved isothermally into humified and 5% CO₂ Alligator at 37 °C. The effect on phase of acute translation initiation inhibition under constant temperature conditions was determined by treating with a 10X spike of inhibitor (rocaglamide A or Hippuristanol) isothermally at the trough of PER2 expression. Plates were gently mixed before placing back in Alligators. For re-entrainment, cells were given a final medium change containing 10% serum at ZT0, ZT6 or ZT12, where ZT0 indicates the start of the hot cycle. Cells were transferred isothermally into a humified and 5% CO₂ Alligator, which was programmed to cycle 12h 37°C: 12h 32°C 12h for 7 days before maintaining constant 37°C for the remainder of the recording. For re-entrainment, drug treatments were given at the final medium change before entering temperature cycles or constant conditions as indicated. In all cases, bioluminescence was recorded at half-hourly intervals. Where indicated, in order to aid visual comparison of phase, time was rescaled to circadian time by correcting elapsed hours against the free-running period for each animal, with circadian time defined as (elapsed time / free-running period) × 24. Thus, one full cycle corresponds to circadian time 0–24.

The following drugs were used: Apigenin (Cayman, #10010275), H89 (Apex Bio, #B2190), Hippuristanol (lab of Jerry Pelletier, (91)), INK128 (Selleckchem, #S2811), LY3214996 (Selleckchem, #8534), Radicicol (Tocris, #1589), Ravoxertinib (Selleckchem, #S7554), Rocaglamide (Sigma, #SML0656), SCH772984 (Selleckchem, #S7101), SR3029 (Medchem Express, #HY100011), TG003 (Apex Bio, #B1431), Torin1 (Selleckchem, #S2827), TTP22 (Tocris, #4432), VER144008 (Selleckchem, #S7751).

Experimental cell culture - flow

For bioluminescent experiments under continuous flow, cells were plated at confluence in transparent glass bottom 6-channel slides (µ-Slide VI 0.5 Glass Bottom, ibidi) in standard culture medium (DMEM Gibco #31966, 10% FetalClone III serum, 1X P/S) supplemented with 0.3 mM luciferin (Biosynth, #L-8220) at 40 µl/channel and placed in humidified incubators at 5% CO2 and 37°C. After cell attachment, channels were topped up with 120 µl/channel of the same medium. After 24-72h, channel slides were connected in series to a pressurized microfluidics flow system (Elveflow OB1 Mk3), with thermal flow rate sensor feedback control (Elveflow MFS 0-80 µl/min), inside an Alligator bioluminescence incubator set to 37°C and 5% CO₂. “Flow medium”, a modified version of the static culture experimental medium (DMEM #31966 Gibco, 1% FetalClone III, 1X P/S, 0.5X Mycozap-plus PR Lonza, 0.3mM luciferin), was delivered to cells via a distribution valve (10:1 MUX bidirectional distributer) at 9-11 µl/min, which was programmed to select medium from pressurized microfluidic reservoirs for durations and intervals as indicated in figure legends. FGF2-containing medium contained 20 ng/ml fibroblast growth factor-basic (FGF2; PeproTech, #450-33). Hyperosmotic medium (+ 50 mOsm) contained 50 mM raffinose (D-(+)-Raffinose pentahydrate; Sigma, #R0514). Bioluminescence was monitored throughout.

Animals and organotypic culture

C57BL/6 (RRID:MGI:2159769) and reporter homozygote mPer2^Luc/Luc (also known as PER2-LUC) mice (originally supplied by Joe Takahashi, University of Texas Southwestern (43), RRID:IMSR JAX:006852) were housed in a specified pathogen free barrier facility. For husbandry and non-experimental housing, mice were group-housed with environmental enrichment under 12h light:12h dark (LD) cycles with lights on at 7am. For experiments, mice were singly housed in individual cages containing environmental enrichment and equipped with running wheels to enable monitoring throughout the experiment. All animal experiments were licensed under 1986 Home Office Animal Procedures Act (UK) and carried out in accordance with local animal welfare committee guidelines (MRC-LMB Animal Welfare and Ethical Review Body).

For organotypic culture of pituitary, adrenal glands and lungs, adult PER2-LUC mice, aged between 8-12 weeks, 8 females, 16 males, were housed in individual cages placed in LD (lights on 10am, lights off 10pm, 12 mice, 4 females, 8 males) or DL (lights on 10pm, lights off 10am, 12 mice, 4 females, 8 males). Mice were euthanized by cervical dislocation and exsanguination at ZT 0.5 (LD) or ZT 23.5 (DL), ensuring that mice did not experience a light-to-dark or dark-to-light transition before euthanizing. Tissues were dissected from the mice in 1X HBSS on ice. Whole pituitary, paired adrenal glands and lung tissue were cut into 1-3 mm pieces in HBSS with a surgical knife. Diced tissue was cultured on Millicell culture membranes (Millipore, #PICMORG50) in 35 mm dishes with 1.2 ml of medium consisting of DMEM (#31966, Gibco), 5% FBS, 100 IU/ml and 100 µg/ml penicillin/streptomycin, 300 µM D-luciferin potassium salt (Biosynth, #L-8220), and 1 µM INK128 (N=6 mice) or DMSO control (N=6 mice).

SCN slices were prepared as previously described (92). Briefly, brains were removed from male and female PER2-LUC mice aged between 9-10 weeks and transferred to ice-cold dissection medium: GBSS (Sigma, #G9779), containing 5 mg/mL glucose, and 100 nM MK801 (Sigma, #M-107), 3 mM MgCl₂ and 0.05 mM APV (Sigma, #A5282) to block excitotoxicity. After trimming, 300 μm coronal brain slices were prepared on a McIlwain tissue chopper. Non-SCN tissue was removed from slices, before they were transferred to Millicell culture membranes (Millipore, #PICMORG50) placed on top of tissue culture medium containing luciferin (50% Eagle’s basal medium, 25% EBSS, 25% heat inactivated horse serum (Invitrogen, USA), 5 mg/mL D-glucose (Sigma), 25 μg/mL Penicillin/Streptomycin, 1% GlutaMAX (Invitrogen), 300 µM luciferin) adjusted to pH7.2 and osmolality 315-320 mOsm/kg, and supplemented with 100 nM MK801, 3 mM MgCl₂ and 0.05 mM APV. After 2-4h, slices were transferred into 35 mm dishes containing fresh culture medium excluding MK801, MgCl₂ and APV, for experiments.

Once prepared, tissues were placed in constant conditions (37°C, 5% CO₂) or temperature cycling (12h 32°C:12h 37°C) humidified Alligator for 3-7 days before release into constant conditions of 37°C. Bioluminescence was recorded throughout temperature cycling and constant conditions at half-hourly intervals.

Temperature shift experiments

For temperature shift experiments (described in Fig. 3 and fig. S7A), cells were plated at confluence and maintained at ‘origin’ temperature of either 32°C or 37°C in incubators humified and at 5% CO₂ for one week. 24h before temperature switch, at t=-24h, cells were re-synchronized with 100 nM dexamethasone, to ensure all cells were in the same circadian phase. At t=0h, cells were either switched to the ‘destination’ temperature of 37°C or 32°C respectively, or maintained at the original constant temperature, for four conditions: constant ‘high’ of 37°C, ‘shift up’, ‘shift down’ or constant ‘low’ of 32°C. Parallel bioluminescent recordings were made in Alligators, humidified and at 5% CO2, at 32°C or 37°C and parallel plates remained in Alligators or switched between them to match the above conditions. Cells were recorded for at least 4 days post shift to allow assessment of phase and period of the circadian rhythm. For phase response curves, 6 x 96-well plates were consecutively shifted between temperatures in Alligators every 4h for 24h.

Puromycin labelling

Puromycin dihydrochloride (#12122530, Gibco) was added directly to cells in culture medium as 10x bolus to a final concentration of 10 μg/ml puromycin. Labelling proceeded for 30min, after which cells were lysed and puromycin detected by Western blotting using anti-puromycin (PMY-2A4-2 from Developmental Studies Hybridoma bank, RRID:AB 2619605, at 1:1000). Fold change puromycin incorporation was calculated as ^{incorporation in temperature shifted}/^{incorporation in constant temperature control}.

Protein extraction

For each condition, cells were washed twice in isosmotic ice-cold PBS plus phosphatase and protease inhibitors at 1X (PhosSTOP, Roche, #04906837001 and cOmplete, Roche #4693116001,) and then lysed at room temperature in a urea/thiourea buffer (UTS, 7 M urea, 2 M thiourea, 1 % Na deoxycholate, 20 mM Tris-HCl, 5 mM TCEP), including phosphatase and protease inhibitors. Lysis (500 µl for 15 cm dish for (phospho)proteomics, 150 µl for 6-well plate for puromycin immunoblot), proceeded for 30 min at room temperature. The lysis buffer was prepared the day before sampling began and frozen in 1 ml aliquots. After lysis, the cells were scraped, transferred into Protein LoBind tubes (#0030108116, Eppendorf) and sonicated for 3 min with 30s on/ 30s off sonication (Bioruptor Plus, Diagenode). Samples were then spun at 15,000 rpm at 4°C for 20 min and the supernatant was taken to a fresh tube. Supernatant was flash-frozen in liquid nitrogen and stored at −80°C for subsequent analysis and processing.

For tissues, on dry ice, brain and liver were chopped into ~5 mm pieces, weighed, and 100 mg transferred to 2 ml tubes containing CK14 ceramic beads (Bertin Instruments) with 1 ml UTS buffer (7 M urea, 1% sodium deoxycholate, 20 mM tris pH 8, 5 mM TCEP, 5 mM DTT). Tissues were homogenized using a Precellys 24 Tissue Homogenizer (Bertin Instruments) for 3 x 15 s at 5,000 RPM with 30 s breaks. Lysates were centrifuged at 20,000 x g for 1 min at 4°C, sonicated, centrifuged at 20,000 x g for 30 minutes at 4°C, and supernatants collected.

Protein concentration was measured on thawed samples using Pierce 660 nm Protein Assay Kit (#22662, Thermo Scientific) in U-bottom 96-well plates (Costar) using a Tecan Spark 10M microplate reader, diluting samples 1 in 4 with dH2O to reduce sodium deoxycholate to compatible levels.

Proteomics and phosphoproteomics

Protein samples for (phospho)proteomics were processed as follows.

Digestion

Protein samples (150 µg in UTS buffer) were reduced with 5 mM DTT at 37°C for 45 min and alkylated with 10 mM iodoacetamide in the dark at room temperature for 30 min. Samples were then diluted to 2.7 M urea and digested with Lys-C (Promega) for 4 h at 25°C. Next, samples were further diluted to 1.5 M urea and digested with trypsin (Promega) overnight at 30°C. After digestion, samples were acidified with formic acid (FA) to a final concentration of 0.5% to precipitate SDC, mixed by shaking for 1 min and centrifuged at 12000 x g for 10 min. The supernatants were transferred to new tubes and desalted using home-made C18 stage tips (3M Empore) packed with porous R3 resin (Thermo Scientific). The stage tips were equilibrated with 80% acetonitrile (MeCN) / 0.5% FA, followed by 0.5% FA. Bound peptides were eluted with 30-80% MeCN / 0.5% FA and lyophilized.

Tandem mass tag (TMT) labelling

The lyophilized peptides from each sample were resuspended in 40 μl of 200 mM Hepes, pH 8.5. 20 μl (500 μg) TMTpro 12plex reagent (Thermo Fisher Scientific), reconstituted in anhydrous MeCN according to manufacturer’s instructions was added. Peptides from each time point were labelled with a distinct TMT tag for 60 min at room temperature. The labelling reaction was quenched by incubation with 4 μl 5% hydroxylamine for 30 min. Each set of 12 labelled peptides (4 conditions, triplicates) for human and mouse were combined into two single samples, one per species, and partially dried to remove MeCN in a SpeedVac (Savant). Sample was then acidified and centrifuged at 16000 x g for 10 min. Supernatant was desalted using the same stage tips method as above and lyophilized.

Titanium dioxide (TiO₂) – Enrichment of phosphopeptides

Phosphopeptides were enriched using TiO₂ titansphere-chromatography (GL Science Inc. Japan). Dried peptides were resolubilized in 50% MeCN containing 2 M lactic acid (loading buffer) and incubated with TiO₂ beads (1: 5, peptides: TiO₂, w/w) that were prewashed with loading buffer. After 30 min, the TiO₂ beads were centrifuged at 9000 x g for 2 min, the supernatant was transferred to fresh TiO₂ beads for a second round of enrichment. After second incubation, TiO₂ beads were loaded onto C8 stage tips (3M Empore) and washed twice with loading buffer and once with 50% MeCN/0.5% FA. Phosphopeptides were eluted sequentially with 100 mM K₂HPO₄, pH 10; 50% MeCN/50 mM K₂HPO₄, pH 10 and 50% MeCN/0.5%FA. The eluates were acidified, partially dried in a SpeedVac, desalted with C18 stage tip (3M Empore) same as above and lyophilized.

Off-line high pH reverse-phase peptides fractionation

200 μg of the labelled peptides were separated on an off-line, high pressure liquid chromatography (HPLC). The experiment was carried out using XBridge BEH130 C18, 5 μm, 2.1 x 150 mm column (Waters), connected to an Ultimate 3000 analytical HPLC (Thermo Fisher Scientific). Peptides were separated with a gradient of 1-90% buffer A and B (A: 5% MeCN, 10 mM ammonium bicarbonate, pH 8; B: MeCN, 10 mM ammonium bicarbonate, pH8, [9:1]) for 60 min at a flow rate of 250 μl/min. A total of 54 fractions were collected, which were then combined into 18 fractions and lyophilized. Dried peptides were resuspended in 1% MeCN / 0.5% FA, and desalted using C18 stage tips, ready for mass spectrometry analysis. Dried phosphopeptides were also fractionated using the same set up as peptides. The collected fractions were combined into 14 fractions and lyophilized. Each fraction was resuspended in 30 ul 20% MeCN/0.1% FA. MeCN was subsequently removed by vacuum centrifugation.

LC-MS/MS

The fractionated peptides were analyzed by LC-MS/MS using a fully automated Ultimate 3000 RSLC nano System fitted with a 100 μm x 2 cm PepMap100 C18 nano trap column (Thermo Fisher Scientific) and a 75 μm x 25 cm, nanoEase M/Z HSS C18 T3 column (Waters). Peptides were separated using a binary gradient consisting of buffer A (0.1% FA) and buffer B (80% MeCN, 0.1% FA) at 300 nL/min, column temperature of 40 °C. Eluted peptides were introduced directly via a nanoFlex ion source into an Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific). The mass spectrometer was operated in real-time database search (RTS) with synchronous-precursor selection (SPS) -MS3 analysis for reporter ion quantification. MS1 spectra were acquired with a resolution of 120K; mass range = 400-1400 m/z; AGC target = 4e5; maximum injection time (MaxIT) = 50 ms; dynamic exclusion was set at 60 s. MS2 analysis were carried out with collision induced dissociation (CID) activation; ion trap detection; AGC = 1e4; MaxIT = 50 ms and isolation window = 0.7 m/z. RTS-SPS was set up to search Uniprot Mus musculus or Homo sapiens proteomes. In MS3 scans, the selected precursors were fragmented by higher-energy collision dissociation (HCD) with collision energy = 55% and analyzed using the orbitrap; resolution = 50K; scan range = 110-500 m/z and MaxIT = 150 ms. For phosphoproteomcs, the mass spectrometer was operated in DDA mode, performed MS1 scan events using the same parameters as above. MS2 spectra were acquired with a 50K resolution; MaxIT= 200 ms; AGC = 1e5; collision energy = 35% and isolation width = 0.7 m/z.

Raw MS data processing

The acquired raw files from LC-MS/MS were processed using MaxQuant (Cox & Mann, 2008) with the integrated Andromeda search engine (v1.6.17.0). MS/MS spectra were quantified with reporter ion MS3 (proteomics) and MS2 (phosphoproteomics) from TMTpro experiments and searched against UniProt Mus musculus or Homo sapiens Reviewed (Nov 2020) Fasta databases. Carbamidomethylation of cysteines was set as a fixed modification, while methionine oxidation, protein N-terminal acetylation and phosphorylation (STY) (for phosphoproteomic group only) were set as variable modifications. Protein quantification requirements were set at 1 unique and razor peptide. In the identification tab, second peptides and match between runs were not selected. Other parameters in MaxQuant were set to default values. The MaxQuant output file was then processed with Perseus (v1.6.15.0). Reporter ion intensities for protein group table were uploaded to Perseus. The data was filtered: identifications from the reverse database, only identified by site, potential contaminants were removed. Then all columns with an intensity “less or equal to zero” were converted to “NAN” (not a number) and exported. The MaxQuant output file with phospho (STY) sites table was also processed with Perseus software (v1.6.15.0). The data was filtered: identifications from the reverse database, potential contaminants were removed, and we only considered phosphopeptides with localization probability ≥ 0.75. Finally, all columns with intensity “less or equal to zero” were converted to “NAN” and exported. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD067660.

Proteomics data analysis

After the MaxQuant search, all subsequent proteomics data processing and analysis was performed in R (v4.3.2) with R Studio (2023.09.1+494). The custom scripts are freely available via Zenodo at doi: 10.5281/zenodo.17053463. Peptide level information from MaxQuant was used as a starting point. Contaminants and reverse hits were removed. Sample loading normalization was performed on proteomic and phosphoproteomic data, applying a scaling factor to equalize total summed intensity across individual TMT channels for the proteome, and using the same scaling factor for the phosphoproteome. In this way, real and biological differences in total phosphorylation between samples was maintained, while correcting for sample loading. Peptides and phosphopeptides were filtered to leave only those detected in all conditions. To assess the proteins and phosphopeptides that changed significantly between conditions, the data were processed in the following manner. For temperature shifts, the raw intensity data were converted to fold change (FC) by matched replicate (i.e. shift up FC = intensity at 32 °C to 37 °C “up”/ intensity at 32 °C “low”; shift down FC = intensity at 37 °C to 32 °C/ intensity at 37 °C). These fold changes were log₂ transformed (upfc#rep_log2 or downfc#rep_log2) before performing repeated measures two-way ANOVA (TWA) on matched pairs of conditions. For adaptation comparisons (i.e. 32 °C vs 37 °C), raw data were log₂ transformed before performing repeated measures two-way ANOVA (TWA) on matched pairs of conditions. Multiple comparisons were corrected for by controlling the false discovery rate by the two-stage step-up method of Benjamini, Krieger and Yekutieli, with Q = 0.01. Figures show mean log₂ fold changes, plotted for each protein/phosphoprotein. The hypothesis that mTOR- and WNK-related phosphosites were differentially responsive to acute temperature change or adapted temperature change in the two species was tested by TWA or t-test respectively for each phosphosite. For acute temperature response, species and fold change upon shift (up or down) and their interaction were variables; the interaction p value presented above each subpanel. For adapted temperature response, fold change (high/low) was tested against species.

Phosphopeptide motif analysis was performed on the subset of singly phosphorylated phosphoproteins (4399 sites mouse, 3713 sites humans). Enrichment-depletion sequence logos were created using dagLogo R package, with foreground and background datasets, and a cutoff of p < 0.05. Foreground datasets were created by subsetting total data into subsets that changed in a proportional (log fold change up > 0 AND log fold change down < 0) or inversely proportional (log fold change up < 0 AND log fold change down > 0) direction. Background datasets were created for each direction comprising the total dataset. Enrichment-depletion sequence logos were separated to obtain enrichment motifs (Fig. 3C) and depletion motifs (fig. S7G) for each direction.

To compare the magnitude of proteome and phosphoproteome change, vectors were calculated for each protein or phosphosite comprising their log₂ fold change under shift up and shift down conditions as illustrated (fig. S7B). Vector magnitude for all proteins was compared with that for phosphosites by Wilcoxon signed-rank test.

Comparison of cellular phosphoproteome temperature responses between mouse and human was performed in a stepwise manner. Mouse phosphosites were mapped to their human orthologs using a curated Uniprot ID gene mapping file. Analyses were performed primarily at the protein level, defining overlapping phosphoproteins as those detected in both species. Temperature-dependent phosphosites were identified using a q-value threshold of 0.05, and the subset mapping to shared phosphoproteins was used for cross-species comparison. To investigate site-level conservation, phosphosites on shared proteins were exported and manually curated based on sequence homology to identify truly homologous positions. Opposite regulatory responses (opposite direction of temperature-dependent phosphorylation) were annotated for these homologous sites.

Immunoblotting

Samples for denaturing polyacrylamide gel electrophoresis (SDS-PAGE) were prepared by diluting lysates with reduced NuPage™ LDS sample buffer and heating at 70°C for 10 min. Samples were run on NuPage™ Novex™ 4-12% Bis-Tris protein gels in MES buffer. For Western blotting for HaloTag or puromycin, chemiluminescence detection was used. Proteins were transferred from the gels to nitrocellulose membranes using an iBlot system (ThermoFisher). Membranes were stained by Ponceau as control for total protein loading, then washed, blocked, and incubated with primary antibody in the blocking buffer at 4°C overnight. Mouse anti-puromycin (PMY-2A4-2 from Developmental Studies Hybridoma bank, RRID:AB_2619605; at 1:1000), rabbit anti-HaloTag (Promega, #G9221, RRID:AB_2688011; 1:3000), rabbit anti-S6K (Abcam, #ab32359, 1:1000), rabbit anti-S6K pT389 (Cell Signaling, #9234S, RRID:AB_777802; 1:1000), anti-4EBP1 (Cell Signaling, #9644S, RRID:AB_2097841 ; 1:1000), rabbit anti-4EBP1 pT37/T46 (Cell Signaling, #2855S, RRID:AB_560835; 1:1000), or anti-β-actin (SCBT, #SC-47778 (C4), RRID:AB_626632; 1:1000) were used with 5% milk in TBST blocking buffer, and an anti-mouse or anti-rabbit HRP-conjugated secondary antibody. Immobilon HRP substrate reagents (Millipore, #WBKLS0500) were used to detect chemiluminescence (ChemiDoc MP; Bio-Rad). Gels were stained with coomassie blue (Severn Biotech, #30-38-50). Images were analyzed by densitometry in ImageLab v4.1 (BioRad).

Comparative genomic analysis

We used RERconverge (93) to identify genes whose evolutionary rates significantly correlate with diurnality in mammals using the Zoonomia database (61). Genes with low relative evolutionary rates in diurnal mammals compared to nocturnal mammals are likely to be under greater evolutionary constraint in diurnal species. Higher gene relative evolutionary rates in diurnal compared to nocturnal mammals may be due to relaxed evolutionary constraint (implying decreased functional importance) or positive selection (implying divergent functional importance). Filtered alignments for gene amino acid sequences were used to estimate tree branch lengths (61), with 16,209 total genes or 11,259 widely expressed genes as defined by expression in fibroblasts within 1.5 SD of the mean expression across 68 tissues and cell lines from RNA-Seq data from the Genotype-Tissue Expression (GTEx) Project. Median RNA-Seq data was accessed from the GTEx Portal on 12 Nov 2024 (GTEx_Analysis_v10_RNASeQCv2.4.2_gene_median_tpm.gct.gz).

We ran RERconverge using default parameters on a set of 186 mammal species for which phenotypes could be confidently defined and had genome information. Each species was classified as using one of four temporal niches – nocturnal (NOC), diurnal (DIU), crepuscular (CRE), or cathemeral (CAT) – using an extensive literature search of books, primary and secondary journal articles and their supplementary information. A foreground set of 77 species was defined as diurnal or crepuscular and the remaining 109 species defined as nocturnal (table S2). Full data, including references for activity patterns, are available at doi: 10.5281/zenodo.16901437. Relative evolutionary rates were calculated using ‘getAllResiduals’ with the gene trees as input. The binary trait tree was defined using the ‘foreground2Tree’ function. We set the transition argument to ‘transition = “unidirectional”‘ allow only transitions to the foreground state. We set the clade argument to “clade=all”, where maximum parsimony is used to infer where transitions to diurnality occur on the tree and set those branches, along with all daughter lineages, to foreground. Paths were then generated from the trait tree using the ‘tree2paths’ function. We performed correlation analyses between the paths and the relative evolutionary rate of each gene using the correlateWithBinaryPhenotype function, with the minimum number of species in a gene tree for inclusion in the analysis set to 20 (‘min.sp=20’), the minimum number of foreground species set to 5 (‘min.pos=5’). P values were adjusted for multiple comparisons using the Benjamini-Hochberg procedure. The hypothesis that mTOR- and WNK-related genes were targets for evolutionary change was tested by filtering the total dataset for genes in these pathways and repeating multiple comparisons adjustment using the Benjamini-Hochberg procedure. To account for any phylogenetic signal that may lead to significant associations that are independent from the diurnality trait we conducted phylogeny-aware permutations (‘permulations’) using the getPermsBinary function with default settings and 1,000 permutations. We considered any genes with both a adjusted p value < 0.05 from the correlation tests and a permuted p value < 0.05 to be significantly associated with being a diurnal mammal (table S3).

Gene Ontology and KEGG pathway analysis was performed using g:Profiler. Redundant GO:Process terms were removed using REVIGO. Top terms were retained and plotted according to whether they were conserved (Rho < 0) or evolved (Rho > 0) in diurnal mammals.

WNK1 protein disorder was calculated per residue per species (fig. S8G) using Metapredict v3 via a Google Colab notebook (94). The activity phenotype-WNK1 disorder relationship was assessed by calculating median ensemble disorder score of full-length WNK1 for each species, and comparing the median disorder score of diurnal vs nocturnal species by Wilcoxon signed-rank test (Fig. 3I).

In vivo mTOR activity reduction

To assess the effect of reduced mTOR activity on diurnal activity pattern, mice were fed a methionine-reduced diet (methionine content 14% of control; methionine reduced: #A11051301Bi, control: #A11051302Bi, Research Diets Inc). We performed a priori power analysis using ClinCalc using effect sizes from pilot data published in (95). Specifically, we used the estimated ratio of day:night activity for mice on control and methionine-reduced diet with pooled variance to perform a power analysis (two-tailed t-test, α = 0.05, power = 0.8). This yielded a required sample size of n = 12 per group. 24 PER2-LUC mice (12 males, 12 females) were singly housed in running-wheel cages (Actimetrics Inc.) with environmental enrichment in light-tight circadian cabinets operated by ClockLab Chamber Control (Actimetrics Inc.). Mice were assigned to experimental groups without reference to phenotype or other criteria; Animal allocation was known to the researcher performing the experimental procedures; however, the researcher was blinded to the underlying hypothesis to minimize bias. Long day lighting conditions with natural twilights (96) were programmed, corresponding to the following lighting schedule: 2h light increasing:14h light:2h light decreasing:6h dark. Mouse activity was monitored throughout through wheel running and IR beam breaking; wheel running (revolutions) was used in downstream analysis. Mice were fed ad libitum with standard diet (Teklad global 18% protein, Inotiv) for 1 week to acclimatize to single housing. One female mouse was excluded from the experiment at this stage due to failure to adapt to single housing and excessive weight loss (> 10% within 7 days). After 1 week, mice were switched onto control or methionine reduced diet (N = 5-6/sex/group) ad libitum for 4 weeks and monitored by behavioral analysis under LD conditions. After 4 weeks, mice were switched into constant darkness, continuing with control or methionine reduced diet, ad libitum for 1 week with behavioral monitoring. At the end of this period, at CT6 on day 42, all mice were culled by cervical dislocation and exsanguination, with whole brain and liver tissue taken and snap frozen in liquid nitrogen for subsequent protein extraction and immunoblotting. Mice were weighed weekly throughout to monitor weight loss. Mouse work was overseen by the Animal Welfare and Ethical Review Body of the MRC Laboratory of Molecular Biology.

For work for food (WFF), data was extracted from previously published work (72) for reanalysis, with experiments described as in (72). Experiments were approved by the University of Groningen ethical committee (DEC 5011, 5114, 5454) and carried out following national animal welfare regulations. Adult male mice (N = 24, CBA/CaJ) were kept singly housed in individual cages containing environmental enrichment, water ad libitum and equipped with running wheels to enable monitoring throughout the experiment. Mice were implanted under isoflurane (Abbott Laboratories Ltd., Kent, UK) inhalation anesthesia (2%) with customized temperature loggers (Thermochron iButton, DS1922L, Maxim Integrated Products Inc., Sunnyvale CA, USA) into their abdomen to enable core body temperature monitoring. Cages were placed in 12 h:12 h light:dark cycles for the duration of the experiment (48 days). Mice were split into two groups: control, fed ad libitum for the duration, and work for food, where food availability was linked to wheel running and delivered after a set number of wheel revolutions. On days 1-10, 100 revolutions were required per food pellet, equivalent to an ad libitum food condition. At day 10, workload was increased daily by an extra 10–20 revolutions per pellet up to 260–330 revolutions per pellet by day 40. 260-300 rev/pellet was maintained for one week during days 40 to 47 during the test period. Throughout the experiment body mass was closely monitored at least every 3rd day at different times of day and high workload levels were individually titrated to keep the mice above 18 grams and 75% of their initial body mass.

To create an average daily activity, activity (running wheel counts) was grouped into 24 x 1h time bins for each day and time bins averaged across 7 days of the baseline (WFF: days 2 to 9) and test (WFF: days 40 to 47; methionine restriction: days 21 to 28) periods of the experiment. Averaged counts per time bin were subsequently normalized by the total counts per day, to give hourly activity as a proportion of the total activity in each day. Acrophase was determined by curve fitting (as described below) to normalized activity (counts/sum of counts during time window) and body temperature (WFF only) during baseline (WFF only) and test periods of the experiment. Phase was calculated as described below, where the peak of the curve fit represents the timing of the highest activity or body temperature. Activity onset was determined using automated onset detection in ClockLab (Actimetrics Inc.). To assess diurnality index (WFF only, for activity), normalized activity in daylight hours (6am to 6pm) of the average day was summed and expressed as a % of total activity occurring during the day.

Statistics

For bioluminescence data, all raw data was detrended using a 24 h moving average and normalized (where 0 = average detrended luminescence and 1 = maximum detrended luminescence). For curve fitting, the following equation was used: y=(mx + c) + ae^kxcos(2πx−r/p) where m is the baseline, c is the offset from 0 in y-axis, a is the amplitude, k is the damping rate, r is the phase (in radians), and p is the period (89). Curve fitting was performed for constant condition data at least 12 h after the last transition in temperature, or treatment, until 96h in constant conditions. Circadian phase was calculated from the timing of peaks of the fitted curve (designated as > 10% above baseline) using the area under the curve test on Prism. The time in hours of the first peak after entry to constant conditions was converted into circadian phase in hours by dividing by circadian period, and is given relative to the last transition between cold and hot such that the start of the hot cycle is set as 0 h. Phase shift for the phase response curve was calculated as the difference in phase between cells that changed temperature and cells that stayed at constant temperature, and was given by phase shift = phase_shifted - phase_control.

Circadian phases and circadian periods between species ± conditions were compared by either Watson-Williams test for homogeneity of means (R package “circular”: watson.williams.test), Correlation Coefficient for Angular Variables (R package “circular”: cor.circular with permutations), or two-way ANOVA (TWA) with Šídák’s post-hoc test for individual comparisons as stated in figure legends.

Mean ± SEM are reported throughout unless otherwise stated in figure legends. Error bars are not displayed if they are smaller than the size of the point to which they relate, due to a restriction in GraphPad Prism (v10, GraphPad Software). Statistical tests were performed using GraphPad Prism (v10) and R v4 and are indicated in figure legends. P values are either reported in figures directly or annotated with asterisks: * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001; n.s. not significant, p > 0.05. Number of replicates are reported as n or N (for biological replicates and number of individuals, respectively) in the figures.

Supplementary Material

Supp Mat

EMS212587-supplement-Supp_Mat.pdf^{(6.5MB, pdf)}

Acknowledgements

We thank Ken Wright for intellectual contributions in the genesis of the project; Rajesh Narasimamurthy and David Virshup for pilot experiments; Michael Hastings, Robert Lucas and Roger Barker for resources and discussion; Tim Eppley for valuable discussions on activity patterns in Lemurs; Joana Frankel for general discussion on diurnal models; Jo Menzies for discussion on luciferase kinetics; and Aymen al-Rawi for valuable contributions to proteomic analysis. We thank all past and current members of O’Neill lab and David Bechtold, for their valuable feedback. We thank biomedical services group staff at Medical Research Council (MRC) Ares facility and LMB facilities for assistance. This project is supported through a research collaboration between AstraZeneca UK Limited and the Medical Research Council, reference BSF38. At the time of writing, Nina Rzechorzek was undertaking an AZ-MRC Industry Partnership for Academic Clinicians, partly funded by AstraZeneca and the Medical Research Council (MC_EX_MR/Y013018/1).

For the purpose of open access, the MRC Laboratory of Molecular Biology applies a CC BY public copyright license to any Author Accepted Manuscript version arising.

Funding

Medical Research Council MC_UP_1201/4 and AstraZeneca BlueSky 2.0 (J.S.O.)

Medical Research Council MR/S022023/1 and MC_EX_MR/S022023/1 (N.M.R.)

Medical Research Council MC_UP_1201/9 and European Research Council ERC STG 757710 (M.A.L.)

Wellcome Trust Investigator Award to Robert Lucas, University of Manchester 210684/Z/18/Z (R.R.)

Royal Society - Wellcome Sir Henry Dale Fellowship 208790/Z/17/Z and UKRI Future Leaders Fellowship MR/Y017552/1 (R.S.E.)

Dutch Research Council (NWO) to the BioClock Consortium Grant No. NWA.1292.19.077 (R.A.H.)

Footnotes

Author contributions

Conceptualization: A.D.B., J.S.O., N.M.R., P.C.

Formal Analysis: A.D.B., N.M.R., M.J.C., S.Y.P.-C., V.P., R.A.H.

Funding acquisition: J.S.O., N.M.R., M.A.L., R.A.H.

Investigation: A.D.B., M.J.C., N.M.R., A.Z., A.M., C.E., N.J.S., V.P., N.R.J., S.Y.P.-C., R.A.H.

Methodology: A.D.B., J.S.O.

Project administration: A.D.B., J.S.O.

Resources: N.J.S., R.R., M.A.L., M.F.B., S.V.F., Z.V., K.M., J.P., R.S.E.

Software: A.D.B.

Supervision: J.S.O.

Visualization: A.D.B.

Writing – original draft: A.D.B.

Writing – review & editing: A.D.B., J.S.O., N.M.R., A.M., R.S.E., M.A.L., M.J.C., A.Z., R.A.H., N.J.S.

Competing interests

The authors declare that they have no competing interests.

Data and materials availability

All data are available in the manuscript or supplementary materials. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD067660. Code for analysis of proteomics are freely available at doi: 10.5281/zenodo.17053463. Materials are available upon request and are subject to materials transfer agreements.

References and Notes

1.Gerkema MP, Davies WIL, Foster RG, Menaker M, Hut RA. The nocturnal bottleneck and the evolution of activity patterns in mammals. Proc Royal Soc B Biological Sci. 2013;280:20130508. doi: 10.1098/rspb.2013.0508. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Maor R, Dayan T, Ferguson-Gow H, Jones KE. Temporal niche expansion in mammals from a nocturnal ancestor after dinosaur extinction. Nat Ecol Evol. 2017;1:1889–1895. doi: 10.1038/s41559-017-0366-5. [DOI] [PubMed] [Google Scholar]
3.Bennie JJ, Duffy JP, Inger R, Gaston KJ. Biogeography of time partitioning in mammals. Proc Natl Acad Sci. 2014;111:13727–13732. doi: 10.1073/pnas.1216063110. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.van der Veen DR, et al. Flexible clock systems: adjusting the temporal programme. Philosophical Transactions Royal Soc B Biological Sci. 2017;372:20160254. doi: 10.1098/rstb.2016.0254. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.dos Reis M, et al. Phylogenomic datasets provide both precision and accuracy in estimating the timescale of placental mammal phylogeny. Proc Royal Soc B Biological Sci. 2012;279:3491–3500. doi: 10.1098/rspb.2012.0683. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Partch CL, Green CB, Takahashi JS. Molecular architecture of the mammalian circadian clock. Trends Cell Biol. 2014;24:90–99. doi: 10.1016/j.tcb.2013.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Mihut A, Partch CL, Crosby P. PERspectives on circadian cell biology. Philos Trans B. 2025;380:20230483. doi: 10.1098/rstb.2023.0483. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Challet E. Minireview: Entrainment of the suprachiasmatic clockwork in diurnal and nocturnal mammals. Endocrinology. 2007;148:5648–55. doi: 10.1210/en.2007-0804. [DOI] [PubMed] [Google Scholar]
9.Mure LS, et al. Diurnal transcriptome atlas of a primate across major neural and peripheral tissues. Science. 2018;359 doi: 10.1126/science.aao0318. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Jha PK, Challet E, Kalsbeek A. Circadian rhythms in glucose and lipid metabolism in nocturnal and diurnal mammals. Mol Cell Endocrinol. 2015;418:74–88. doi: 10.1016/j.mce.2015.01.024. [DOI] [PubMed] [Google Scholar]
11.Cederroth CR, et al. Medicine in the Fourth Dimension. Cell Metabolism. 2019;30:238–250. doi: 10.1016/j.cmet.2019.06.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.D’Alessandro M, et al. A tunable artificial circadian clock in clock-defective mice. Nat Commun. 2015;6:8587. doi: 10.1038/ncomms9587. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Feeney KA, et al. Daily magnesium fluxes regulate cellular timekeeping and energy balance. Nature. 2016;532:375–379. doi: 10.1038/nature17407. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Crosby P, et al. Insulin/IGF-1 Drives PERIOD Synthesis to Entrain Circadian Rhythms with Feeding Time. Cell. 2019;177:896–909.:e20. doi: 10.1016/j.cell.2019.02.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Buijs RM, et al. Anatomical and functional demonstration of a multisynaptic suprachiasmatic nucleus adrenal (cortex) pathway. Eur J Neurosci. 1999;11:1535–1544. doi: 10.1046/j.1460-9568.1999.00575.x. [DOI] [PubMed] [Google Scholar]
16.Balsalobre A, et al. Resetting of Circadian Time in Peripheral Tissues by Glucocorticoid Signaling. Science. 2000;289:2344–2347. doi: 10.1126/science.289.5488.2344. [DOI] [PubMed] [Google Scholar]
17.Ota SM, Kong X, Hut R, Suchecki D, Meerlo P. The impact of stress and stress hormones on endogenous clocks and circadian rhythms. Frontiers Neuroendocrinol. 2021;63:100931. doi: 10.1016/j.yfrne.2021.100931. [DOI] [PubMed] [Google Scholar]
18.Roenneberg T, Daan S, Merrow M. The Art of Entrainment. J Biol Rhythm. 2003;18:183–194. doi: 10.1177/0748730403018003001. [DOI] [PubMed] [Google Scholar]
19.Damiola F, et al. Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Gene Dev. 2000;14:2950–2961. doi: 10.1101/gad.183500. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Stokkan K-A, Yamazaki S, Tei H, Sakaki Y, Menaker M. Entrainment of the Circadian Clock in the Liver by Feeding. Science. 2001;291:490–493. doi: 10.1126/science.291.5503.490. [DOI] [PubMed] [Google Scholar]
21.Isojima Y, et al. CKIε/δ-dependent phosphorylation is a temperature-insensitive, period-determining process in the mammalian circadian clock. Proc National Acad Sci. 2009;106:15744–15749. doi: 10.1073/pnas.0908733106. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Refinetti R, Menaker M. The circadian rhythm of body temperature. Physiol Behav. 1992;51:613–637. doi: 10.1016/0031-9384(92)90188-8. [DOI] [PubMed] [Google Scholar]
23.Brown SA, Zumbrunn G, Fleury-Olela F, Preitner N, Schibler U. Rhythms of mammalian body temperature can sustain peripheral circadian clocks. Curr Biol. 2002;12:1574–1583. doi: 10.1016/s0960-9822(02)01145-4. [DOI] [PubMed] [Google Scholar]
24.Prolo LM, Takahashi JS, Herzog ED. Circadian Rhythm Generation and Entrainment in Astrocytes. J Neurosci. 2005;25:404–408. doi: 10.1523/JNEUROSCI.4133-04.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Buhr ED, Yoo S-H, Takahashi JS. Temperature as a Universal Resetting Cue for Mammalian Circadian Oscillators. Science. 2010;330:379–385. doi: 10.1126/science.1195262. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Saini C, Morf J, Stratmann M, Gos P, Schibler U. Simulated body temperature rhythms reveal the phase-shifting behavior and plasticity of mammalian circadian oscillators. Gene Dev. 2012;26:567–580. doi: 10.1101/gad.183251.111. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Beale AD, et al. Mechanisms and physiological function of daily haemoglobin oxidation rhythms in red blood cells. EMBO J. 2023:e114164. doi: 10.15252/embj.2023114164. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Morf J, et al. Cold-inducible RNA-binding protein modulates circadian gene expression posttranscriptionally. Science. 2012;338:379–383. doi: 10.1126/science.1217726. [DOI] [PubMed] [Google Scholar]
29.Haltenhof T, et al. A Conserved Kinase-Based Body-Temperature Sensor Globally Controls Alternative Splicing and Gene Expression. Mol Cell. 2020;78:57–69.:e4. doi: 10.1016/j.molcel.2020.01.028. [DOI] [PubMed] [Google Scholar]
30.Doi M, et al. Non-coding cis-element of Period2 is essential for maintaining organismal circadian behaviour and body temperature rhythmicity. Nat Commun. 2019;10:2563. doi: 10.1038/s41467-019-10532-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Miyake T, et al. Minimal upstream open reading frame of Per2 mediates phase fitness of the circadian clock to day/night physiological body temperature rhythm. Cell Reports. 2023:112157. doi: 10.1016/j.celrep.2023.112157. [DOI] [PubMed] [Google Scholar]
32.Kornmann B, Schaad O, Bujard H, Takahashi JS, Schibler U. System-driven and oscillator-dependent circadian transcription in mice with a conditionally active liver clock. Plos Biol. 2007;5:0179-0189. doi: 10.1371/journal.pbio.0050034. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Reinke H, et al. Differential display of DNA-binding proteins reveals heat-shock factor 1 as a circadian transcription factor. Gene Dev. 2008;22:331–345. doi: 10.1101/gad.453808. [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Matsuda M, et al. Species-specific segmentation clock periods are due to differential biochemical reaction speeds. Science. 2020;369:1450–1455. doi: 10.1126/science.aba7668. [DOI] [PubMed] [Google Scholar]
35.Rayon T, et al. Species-specific pace of development is associated with differences in protein stability. Science. 2020;369 doi: 10.1126/science.aba7667. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Lázaro J, et al. A stem cell zoo uncovers intracellular scaling of developmental tempo across mammals. Cell Stem Cell. 2023;30:938–949.:e7. doi: 10.1016/j.stem.2023.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Matsuda M, Lazaro J, Ebisuya M. Metabolic activities are selective modulators for individual segmentation clock processes. Nat Commun. 2025;16:845. doi: 10.1038/s41467-025-56120-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Rzechorzek NM, et al. A daily temperature rhythm in the human brain predicts survival after brain injury. Brain. 2022;145:2031–2048. doi: 10.1093/brain/awab466. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Santini L, Rojas D, Donati G. Evolving through day and night: origin and diversification of activity pattern in modern primates. Behav Ecol. 2015;26:789–796. [Google Scholar]
40.Mendel VE, Raghavan GV. A study of diurnal temperature patterns in sheep. J Physiology. 1964;174:206–216. doi: 10.1113/jphysiol.1964.sp007482. [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Schumann DM, Cooper HM, Hofmeyr MD, Bennett NC. Circadian rhythm of locomotor activity in the four-striped field mouse, Rhabdomys pumilio: A diurnal African rodent. Physiol Behav. 2005;85:231–239. doi: 10.1016/j.physbeh.2005.03.024. [DOI] [PubMed] [Google Scholar]
42.Donati G, Santini L, Razafindramanana J, Boitani L, Borgognini-Tarli S. (Un-)expected nocturnal activity in “Diurnal” Lemur catta supports cathemerality as one of the key adaptations of the lemurid radiation. Am J Phys Anthropol. 2013;150:99–106. doi: 10.1002/ajpa.22180. [DOI] [PubMed] [Google Scholar]
43.Yoo S-H, et al. PERIOD2∷LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. P Natl Acad Sci Usa. 2004;101:5339–5346. doi: 10.1073/pnas.0308709101. [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Feeney KA, Putker M, Brancaccio M, O’Neill JS. In-depth Characterization of Firefly Luciferase as a Reporter of Circadian Gene Expression in Mammalian Cells. J Biol Rhythm. 2016;31:540–550. doi: 10.1177/0748730416668898. [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Gandelman O, Allue I, Bowers K, Cobbold P. Cytoplasmic factors that affect the intensity and stability of bioluminescence from firefly luciferase in living mammalian cells. J Biolumin Chemilumin. 1994;9:363–371. doi: 10.1002/bio.1170090603. [DOI] [PubMed] [Google Scholar]
46.Liu Y, Yang Q, Zhao F. Synonymous but not Silent: The Codon Usage Code for Gene Expression and Protein Folding. Annu Rev Biochem. 2021;90:1–27. doi: 10.1146/annurev-biochem-071320-112701. [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Zhou M, et al. Non-optimal codon usage affects expression, structure and function of clock protein FRQ. Nature. 2013;495:111–115. doi: 10.1038/nature11833. [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Pittendrigh CS. Circadian Rhythms and the Circadian Organization of Living Systems. Cold Spring Harb Sym. 1960;25:159–184. doi: 10.1101/sqb.1960.025.01.015. [DOI] [PubMed] [Google Scholar]
49.Proud CG. Phosphorylation and Signal Transduction Pathways in Translational Control. Csh Perspect Biol. 2019;11:a033050. doi: 10.1101/cshperspect.a033050. [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Liu X, Liu PCC, Santoro N, Thiele DJ. Conservation of a stress response: human heat shock transcription factors functionally substitute for yeast HSF. The EMBO J. 1997;16:6466–6477. doi: 10.1093/emboj/16.21.6466. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Al-Fageeh MB, Smales CM. Control and regulation of the cellular responses to cold shock: the responses in yeast and mammalian systems. Biochem J. 2006;397:247–259. doi: 10.1042/BJ20060166. [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Johnson JL, et al. An atlas of substrate specificities for the human serine/threonine kinome. Nature. 2023:1–8. doi: 10.1038/s41586-022-05575-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Valvezan AJ, Manning BD. Molecular logic of mTORC1 signalling as a metabolic rheostat. Nat Metab. 2019;1:321–333. doi: 10.1038/s42255-019-0038-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Delarue M, et al. mTORC1 Controls Phase Separation and the Biophysical Properties of the Cytoplasm by Tuning Crowding. Cell. 2018;174:338–349.:e20. doi: 10.1016/j.cell.2018.05.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Xie Y, et al. Polysome collapse and RNA condensation fluidize the cytoplasm. Mol Cell. 2024;84:2698–2716.:e9. doi: 10.1016/j.molcel.2024.06.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Akella R, et al. Osmosensing by WNK Kinases. Mol Biol Cell. 2021;32:1614–1623. doi: 10.1091/mbc.E20-01-0089. [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Boyd-Shiwarski CR, et al. WNK kinases sense molecular crowding and rescue cell volume via phase separation. Cell. 2022;185:4488–4506.:e20. doi: 10.1016/j.cell.2022.09.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Subramanya AR, Boyd-Shiwarski CR. Molecular Crowding: Physiologic Sensing and Control. Annu Rev Physiol. 2024;86:429–452. doi: 10.1146/annurev-physiol-042222-025920. [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Stangherlin A, et al. Compensatory ion transport buffers daily protein rhythms to regulate osmotic balance and cellular physiology. Nat Commun. 2021;12:6035. doi: 10.1038/s41467-021-25942-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Watson JL, et al. Macromolecular condensation buffers intracellular water potential. Nature. 2023;623:842–852. doi: 10.1038/s41586-023-06626-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Christmas MJ, et al. Evolutionary constraint and innovation across hundreds of placental mammals. Science. 2023;380:eabn3943. doi: 10.1126/science.abn3943. [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Juszkiewicz S, et al. ZNF598 Is a Quality Control Sensor of Collided Ribosomes. Mol Cell. 2018;72:469–481.:e7. doi: 10.1016/j.molcel.2018.08.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Khalsa SB, Whitmore D, Block GD. Stopping the circadian pacemaker with inhibitors of protein synthesis. Proc National Acad Sci. 1992;89:10862–10866. doi: 10.1073/pnas.89.22.10862. [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Yamaguchi S, et al. Synchronization of Cellular Clocks in the Suprachiasmatic Nucleus. Science. 2003;302:1408–1412. doi: 10.1126/science.1089287. [DOI] [PubMed] [Google Scholar]
65.Abraham U, et al. Coupling governs entrainment range of circadian clocks. Mol Syst Biol. 2010;6:438. doi: 10.1038/msb.2010.92. [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Ruby NF, Burns DE, Heller HC. Circadian Rhythms in the Suprachiasmatic Nucleus are Temperature-Compensated and Phase-Shifted by Heat Pulses In Vitro. J Neurosci. 1999;19:8630–8636. doi: 10.1523/JNEUROSCI.19-19-08630.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Pilorz V, et al. A novel mechanism controlling resetting speed of the circadian clock to environmental stimuli. Curr Biology Cb. 2013;24:766–73. doi: 10.1016/j.cub.2014.02.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Hsieh AC, et al. The translational landscape of mTOR signalling steers cancer initiation and metastasis. Nature. 2012;485:55–61. doi: 10.1038/nature10912. [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Naito T, Kuma A, Mizushima N. Differential Contribution of Insulin and Amino Acids to the mTORC1-Autophagy Pathway in the Liver and Muscle*. J Biol Chem. 2013;288:21074–21081. doi: 10.1074/jbc.M113.456228. [DOI] [PMC free article] [PubMed] [Google Scholar]
70.Daan S, et al. Lab Mice in the Field: Unorthodox Daily Activity and Effects of a Dysfunctional Circadian Clock Allele. J Biol Rhythm. 2011;26:118–129. doi: 10.1177/0748730410397645. [DOI] [PubMed] [Google Scholar]
71.van Rosmalen L, et al. Multi-organ transcriptome atlas of a mouse model of relative energy deficiency in sport. Cell Metab. 2024;36:2015–2037.:e6. doi: 10.1016/j.cmet.2024.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Hut RA, Pilorz V, Boerema AS, Strijkstra AM, Daan S. Working for Food Shifts Nocturnal Mouse Activity into the Day. Plos One. 2011;6:e17527. doi: 10.1371/journal.pone.0017527. [DOI] [PMC free article] [PubMed] [Google Scholar]
73.van der Vinne V, et al. Cold and hunger induce diurnality in a nocturnal mammal. Proc National Acad Sci. 2014;111:15256–15260. doi: 10.1073/pnas.1413135111. [DOI] [PMC free article] [PubMed] [Google Scholar]
74.van Rosmalen L, et al. Energy balance drives diurnal and nocturnal brain transcriptome rhythms. Cell Rep. 2024;43:113951. doi: 10.1016/j.celrep.2024.113951. [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Gu X, et al. SAMTOR is an S-adenosylmethionine sensor for the mTORC1 pathway. Science. 2017;358:813–818. doi: 10.1126/science.aao3265. [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Doyle SE, Yoshikawa T, Hillson H, Menaker M. Retinal pathways influence temporal niche. Proc National Acad Sci. 2008;105:13133–13138. doi: 10.1073/pnas.0801728105. [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Schoonderwoerd RA, et al. Inhibitory responses to retinohypothalamic tract stimulation in the circadian clock of the diurnal rodent Rhabdomys pumilio. FASEB J. 2022;36:e22415. doi: 10.1096/fj.202200477R. [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Richardson R, et al. The genomic basis of temporal niche evolution in a diurnal rodent. Curr Biol. 2023;33:3289–3298.:e6. doi: 10.1016/j.cub.2023.06.068. [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Riede SJ, van der Vinne V, Hut RA. The flexible clock: predictive and reactive homeostasis, energy balance and the circadian regulation of sleep–wake timing. J Exp Biol. 2017;220:738–749. doi: 10.1242/jeb.130757. [DOI] [PubMed] [Google Scholar]
80.Stangherlin A, Seinkmane E, O’Neill JS. Understanding circadian regulation of mammalian cell function, protein homeostasis, and metabolism. Curr Opin Syst Biology. 2021;28:100391. doi: 10.1016/j.coisb.2021.100391. [DOI] [PMC free article] [PubMed] [Google Scholar]
81.Moreira MO, Qu Y, Wiens JJ. Large-scale evolution of body temperatures in land vertebrates. Evol Lett. 2021;5:484–494. doi: 10.1002/evl3.249. [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Lewden A, Bishop CM, Askew GN. How birds dissipate heat before, during and after flight. J R Soc Interface. 2023;20:20230442. doi: 10.1098/rsif.2023.0442. [DOI] [PMC free article] [PubMed] [Google Scholar]
83.Lindhout FW, Krienen FM, Pollard KS, Lancaster MA. A molecular and cellular perspective on human brain evolution and tempo. Nature. 2024;630:596–608. doi: 10.1038/s41586-024-07521-x. [DOI] [PubMed] [Google Scholar]
84.Iyer DP, et al. mTOR activity paces human blastocyst stage developmental progression. Cell. 2024;187:6566–6583.:e22. doi: 10.1016/j.cell.2024.08.048. [DOI] [PMC free article] [PubMed] [Google Scholar]
85.Iwata R, Vanderhaeghen P. Metabolic mechanisms of species-specific developmental tempo. Dev Cell. 2024;59:1628–1639. doi: 10.1016/j.devcel.2024.05.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
86.Seluanov A, Vaidya A, Gorbunova V. Establishing Primary Adult Fibroblast Cultures From Rodents. J Vis Exp. 2010 doi: 10.3791/2033. [DOI] [PMC free article] [PubMed] [Google Scholar]
87.Drouin-Ouellet J, et al. REST suppression mediates neural conversion of adult human fibroblasts via microRNA-dependent and -independent pathways. EMBO Mol Medicine. 2017;9:1117–1131. doi: 10.15252/emmm.201607471. [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Labun K, Montague TG, Gagnon JA, Thyme SB, Valen E. CHOPCHOP v2: a web tool for the next generation of CRISPR genome engineering. Nucleic Acids Res. 2016;44:W272–W276. doi: 10.1093/nar/gkw398. [DOI] [PMC free article] [PubMed] [Google Scholar]
89.Zeng A, O’Neill JS. Using ALLIGATORs to Capture Circadian Bioluminescence. Circadian Regulation of Methods in Molecular Biology. 2022;2482:125–135. doi: 10.1007/978-1-0716-2249-0_8. [DOI] [PubMed] [Google Scholar]
90.Verma M, et al. A short translational ramp determines the efficiency of protein synthesis. Nat Commun. 2019;10:5774. doi: 10.1038/s41467-019-13810-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Cencic R, Pelletier J. Hippuristanol - A potent steroid inhibitor of eukaryotic initiation factor 4A. Translation. 2016;4:e1137381. doi: 10.1080/21690731.2015.1137381. [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Hastings MH, Reddy AB, McMahon DG, Maywood ES. Analysis of Circadian Mechanisms in the Suprachiasmatic Nucleus by Transgenesis and Biolistic Transfection. Methods Enzymol. 2005;393:579–592. doi: 10.1016/S0076-6879(05)93030-9. [DOI] [PubMed] [Google Scholar]
93.Kowalczyk A, et al. RERconverge: an R package for associating evolutionary rates with convergent traits. Bioinformatics. 2019;35:4815–4817. doi: 10.1093/bioinformatics/btz468. [DOI] [PMC free article] [PubMed] [Google Scholar]
94.Lotthammer JM, et al. Metapredict enables accurate disorder prediction across the Tree of Life. bioRxiv. 2024:2024.11.05.622168 [Google Scholar]
95.Bárcena C, et al. Methionine Restriction Extends Lifespan in Progeroid Mice and Alters Lipid and Bile Acid Metabolism. Cell Rep. 2018;24:2392–2403. doi: 10.1016/j.celrep.2018.07.089. [DOI] [PMC free article] [PubMed] [Google Scholar]
96.Comas M, Hut RA. Twilight and Photoperiod Affect Behavioral Entrainment in the House Mouse (Mus musculus) J Biol Rhythm. 2009;24:403–412. doi: 10.1177/0748730409343873. [DOI] [PubMed] [Google Scholar]
97.Brown SA, et al. The Period Length of Fibroblast Circadian Gene Expression Varies Widely among Human Individuals. PLoS Biol. 2005;3:e338. doi: 10.1371/journal.pbio.0030338. [DOI] [PMC free article] [PubMed] [Google Scholar]
98.Pelham JF, Dunlap JC, Hurley JM. Intrinsic disorder is an essential characteristic of components in the conserved circadian circuit. Cell Commun Signal. 2020;18:181. doi: 10.1186/s12964-020-00658-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
99.Düvel K, et al. Activation of a Metabolic Gene Regulatory Network Downstream of mTOR Complex 1. Mol Cell. 2010;39:171–183. doi: 10.1016/j.molcel.2010.06.022. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Mat

EMS212587-supplement-Supp_Mat.pdf^{(6.5MB, pdf)}

Data Availability Statement

[R1] 1.Gerkema MP, Davies WIL, Foster RG, Menaker M, Hut RA. The nocturnal bottleneck and the evolution of activity patterns in mammals. Proc Royal Soc B Biological Sci. 2013;280:20130508. doi: 10.1098/rspb.2013.0508. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Maor R, Dayan T, Ferguson-Gow H, Jones KE. Temporal niche expansion in mammals from a nocturnal ancestor after dinosaur extinction. Nat Ecol Evol. 2017;1:1889–1895. doi: 10.1038/s41559-017-0366-5. [DOI] [PubMed] [Google Scholar]

[R3] 3.Bennie JJ, Duffy JP, Inger R, Gaston KJ. Biogeography of time partitioning in mammals. Proc Natl Acad Sci. 2014;111:13727–13732. doi: 10.1073/pnas.1216063110. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.van der Veen DR, et al. Flexible clock systems: adjusting the temporal programme. Philosophical Transactions Royal Soc B Biological Sci. 2017;372:20160254. doi: 10.1098/rstb.2016.0254. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] 5.dos Reis M, et al. Phylogenomic datasets provide both precision and accuracy in estimating the timescale of placental mammal phylogeny. Proc Royal Soc B Biological Sci. 2012;279:3491–3500. doi: 10.1098/rspb.2012.0683. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Partch CL, Green CB, Takahashi JS. Molecular architecture of the mammalian circadian clock. Trends Cell Biol. 2014;24:90–99. doi: 10.1016/j.tcb.2013.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Mihut A, Partch CL, Crosby P. PERspectives on circadian cell biology. Philos Trans B. 2025;380:20230483. doi: 10.1098/rstb.2023.0483. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Challet E. Minireview: Entrainment of the suprachiasmatic clockwork in diurnal and nocturnal mammals. Endocrinology. 2007;148:5648–55. doi: 10.1210/en.2007-0804. [DOI] [PubMed] [Google Scholar]

[R9] 9.Mure LS, et al. Diurnal transcriptome atlas of a primate across major neural and peripheral tissues. Science. 2018;359 doi: 10.1126/science.aao0318. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] 10.Jha PK, Challet E, Kalsbeek A. Circadian rhythms in glucose and lipid metabolism in nocturnal and diurnal mammals. Mol Cell Endocrinol. 2015;418:74–88. doi: 10.1016/j.mce.2015.01.024. [DOI] [PubMed] [Google Scholar]

[R11] 11.Cederroth CR, et al. Medicine in the Fourth Dimension. Cell Metabolism. 2019;30:238–250. doi: 10.1016/j.cmet.2019.06.019. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] 12.D’Alessandro M, et al. A tunable artificial circadian clock in clock-defective mice. Nat Commun. 2015;6:8587. doi: 10.1038/ncomms9587. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] 13.Feeney KA, et al. Daily magnesium fluxes regulate cellular timekeeping and energy balance. Nature. 2016;532:375–379. doi: 10.1038/nature17407. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Crosby P, et al. Insulin/IGF-1 Drives PERIOD Synthesis to Entrain Circadian Rhythms with Feeding Time. Cell. 2019;177:896–909.:e20. doi: 10.1016/j.cell.2019.02.017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Buijs RM, et al. Anatomical and functional demonstration of a multisynaptic suprachiasmatic nucleus adrenal (cortex) pathway. Eur J Neurosci. 1999;11:1535–1544. doi: 10.1046/j.1460-9568.1999.00575.x. [DOI] [PubMed] [Google Scholar]

[R16] 16.Balsalobre A, et al. Resetting of Circadian Time in Peripheral Tissues by Glucocorticoid Signaling. Science. 2000;289:2344–2347. doi: 10.1126/science.289.5488.2344. [DOI] [PubMed] [Google Scholar]

[R17] 17.Ota SM, Kong X, Hut R, Suchecki D, Meerlo P. The impact of stress and stress hormones on endogenous clocks and circadian rhythms. Frontiers Neuroendocrinol. 2021;63:100931. doi: 10.1016/j.yfrne.2021.100931. [DOI] [PubMed] [Google Scholar]

[R18] 18.Roenneberg T, Daan S, Merrow M. The Art of Entrainment. J Biol Rhythm. 2003;18:183–194. doi: 10.1177/0748730403018003001. [DOI] [PubMed] [Google Scholar]

[R19] 19.Damiola F, et al. Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Gene Dev. 2000;14:2950–2961. doi: 10.1101/gad.183500. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Stokkan K-A, Yamazaki S, Tei H, Sakaki Y, Menaker M. Entrainment of the Circadian Clock in the Liver by Feeding. Science. 2001;291:490–493. doi: 10.1126/science.291.5503.490. [DOI] [PubMed] [Google Scholar]

[R21] 21.Isojima Y, et al. CKIε/δ-dependent phosphorylation is a temperature-insensitive, period-determining process in the mammalian circadian clock. Proc National Acad Sci. 2009;106:15744–15749. doi: 10.1073/pnas.0908733106. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Refinetti R, Menaker M. The circadian rhythm of body temperature. Physiol Behav. 1992;51:613–637. doi: 10.1016/0031-9384(92)90188-8. [DOI] [PubMed] [Google Scholar]

[R23] 23.Brown SA, Zumbrunn G, Fleury-Olela F, Preitner N, Schibler U. Rhythms of mammalian body temperature can sustain peripheral circadian clocks. Curr Biol. 2002;12:1574–1583. doi: 10.1016/s0960-9822(02)01145-4. [DOI] [PubMed] [Google Scholar]

[R24] 24.Prolo LM, Takahashi JS, Herzog ED. Circadian Rhythm Generation and Entrainment in Astrocytes. J Neurosci. 2005;25:404–408. doi: 10.1523/JNEUROSCI.4133-04.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Buhr ED, Yoo S-H, Takahashi JS. Temperature as a Universal Resetting Cue for Mammalian Circadian Oscillators. Science. 2010;330:379–385. doi: 10.1126/science.1195262. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] 26.Saini C, Morf J, Stratmann M, Gos P, Schibler U. Simulated body temperature rhythms reveal the phase-shifting behavior and plasticity of mammalian circadian oscillators. Gene Dev. 2012;26:567–580. doi: 10.1101/gad.183251.111. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] 27.Beale AD, et al. Mechanisms and physiological function of daily haemoglobin oxidation rhythms in red blood cells. EMBO J. 2023:e114164. doi: 10.15252/embj.2023114164. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Morf J, et al. Cold-inducible RNA-binding protein modulates circadian gene expression posttranscriptionally. Science. 2012;338:379–383. doi: 10.1126/science.1217726. [DOI] [PubMed] [Google Scholar]

[R29] 29.Haltenhof T, et al. A Conserved Kinase-Based Body-Temperature Sensor Globally Controls Alternative Splicing and Gene Expression. Mol Cell. 2020;78:57–69.:e4. doi: 10.1016/j.molcel.2020.01.028. [DOI] [PubMed] [Google Scholar]

[R30] 30.Doi M, et al. Non-coding cis-element of Period2 is essential for maintaining organismal circadian behaviour and body temperature rhythmicity. Nat Commun. 2019;10:2563. doi: 10.1038/s41467-019-10532-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Miyake T, et al. Minimal upstream open reading frame of Per2 mediates phase fitness of the circadian clock to day/night physiological body temperature rhythm. Cell Reports. 2023:112157. doi: 10.1016/j.celrep.2023.112157. [DOI] [PubMed] [Google Scholar]

[R32] 32.Kornmann B, Schaad O, Bujard H, Takahashi JS, Schibler U. System-driven and oscillator-dependent circadian transcription in mice with a conditionally active liver clock. Plos Biol. 2007;5:0179-0189. doi: 10.1371/journal.pbio.0050034. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] 33.Reinke H, et al. Differential display of DNA-binding proteins reveals heat-shock factor 1 as a circadian transcription factor. Gene Dev. 2008;22:331–345. doi: 10.1101/gad.453808. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Matsuda M, et al. Species-specific segmentation clock periods are due to differential biochemical reaction speeds. Science. 2020;369:1450–1455. doi: 10.1126/science.aba7668. [DOI] [PubMed] [Google Scholar]

[R35] 35.Rayon T, et al. Species-specific pace of development is associated with differences in protein stability. Science. 2020;369 doi: 10.1126/science.aba7667. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] 36.Lázaro J, et al. A stem cell zoo uncovers intracellular scaling of developmental tempo across mammals. Cell Stem Cell. 2023;30:938–949.:e7. doi: 10.1016/j.stem.2023.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] 37.Matsuda M, Lazaro J, Ebisuya M. Metabolic activities are selective modulators for individual segmentation clock processes. Nat Commun. 2025;16:845. doi: 10.1038/s41467-025-56120-5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] 38.Rzechorzek NM, et al. A daily temperature rhythm in the human brain predicts survival after brain injury. Brain. 2022;145:2031–2048. doi: 10.1093/brain/awab466. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] 39.Santini L, Rojas D, Donati G. Evolving through day and night: origin and diversification of activity pattern in modern primates. Behav Ecol. 2015;26:789–796. [Google Scholar]

[R40] 40.Mendel VE, Raghavan GV. A study of diurnal temperature patterns in sheep. J Physiology. 1964;174:206–216. doi: 10.1113/jphysiol.1964.sp007482. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] 41.Schumann DM, Cooper HM, Hofmeyr MD, Bennett NC. Circadian rhythm of locomotor activity in the four-striped field mouse, Rhabdomys pumilio: A diurnal African rodent. Physiol Behav. 2005;85:231–239. doi: 10.1016/j.physbeh.2005.03.024. [DOI] [PubMed] [Google Scholar]

[R42] 42.Donati G, Santini L, Razafindramanana J, Boitani L, Borgognini-Tarli S. (Un-)expected nocturnal activity in “Diurnal” Lemur catta supports cathemerality as one of the key adaptations of the lemurid radiation. Am J Phys Anthropol. 2013;150:99–106. doi: 10.1002/ajpa.22180. [DOI] [PubMed] [Google Scholar]

[R43] 43.Yoo S-H, et al. PERIOD2∷LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. P Natl Acad Sci Usa. 2004;101:5339–5346. doi: 10.1073/pnas.0308709101. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R44] 44.Feeney KA, Putker M, Brancaccio M, O’Neill JS. In-depth Characterization of Firefly Luciferase as a Reporter of Circadian Gene Expression in Mammalian Cells. J Biol Rhythm. 2016;31:540–550. doi: 10.1177/0748730416668898. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R45] 45.Gandelman O, Allue I, Bowers K, Cobbold P. Cytoplasmic factors that affect the intensity and stability of bioluminescence from firefly luciferase in living mammalian cells. J Biolumin Chemilumin. 1994;9:363–371. doi: 10.1002/bio.1170090603. [DOI] [PubMed] [Google Scholar]

[R46] 46.Liu Y, Yang Q, Zhao F. Synonymous but not Silent: The Codon Usage Code for Gene Expression and Protein Folding. Annu Rev Biochem. 2021;90:1–27. doi: 10.1146/annurev-biochem-071320-112701. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] 47.Zhou M, et al. Non-optimal codon usage affects expression, structure and function of clock protein FRQ. Nature. 2013;495:111–115. doi: 10.1038/nature11833. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R48] 48.Pittendrigh CS. Circadian Rhythms and the Circadian Organization of Living Systems. Cold Spring Harb Sym. 1960;25:159–184. doi: 10.1101/sqb.1960.025.01.015. [DOI] [PubMed] [Google Scholar]

[R49] 49.Proud CG. Phosphorylation and Signal Transduction Pathways in Translational Control. Csh Perspect Biol. 2019;11:a033050. doi: 10.1101/cshperspect.a033050. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] 50.Liu X, Liu PCC, Santoro N, Thiele DJ. Conservation of a stress response: human heat shock transcription factors functionally substitute for yeast HSF. The EMBO J. 1997;16:6466–6477. doi: 10.1093/emboj/16.21.6466. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R51] 51.Al-Fageeh MB, Smales CM. Control and regulation of the cellular responses to cold shock: the responses in yeast and mammalian systems. Biochem J. 2006;397:247–259. doi: 10.1042/BJ20060166. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] 52.Johnson JL, et al. An atlas of substrate specificities for the human serine/threonine kinome. Nature. 2023:1–8. doi: 10.1038/s41586-022-05575-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R53] 53.Valvezan AJ, Manning BD. Molecular logic of mTORC1 signalling as a metabolic rheostat. Nat Metab. 2019;1:321–333. doi: 10.1038/s42255-019-0038-7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R54] 54.Delarue M, et al. mTORC1 Controls Phase Separation and the Biophysical Properties of the Cytoplasm by Tuning Crowding. Cell. 2018;174:338–349.:e20. doi: 10.1016/j.cell.2018.05.042. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R55] 55.Xie Y, et al. Polysome collapse and RNA condensation fluidize the cytoplasm. Mol Cell. 2024;84:2698–2716.:e9. doi: 10.1016/j.molcel.2024.06.024. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R56] 56.Akella R, et al. Osmosensing by WNK Kinases. Mol Biol Cell. 2021;32:1614–1623. doi: 10.1091/mbc.E20-01-0089. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R57] 57.Boyd-Shiwarski CR, et al. WNK kinases sense molecular crowding and rescue cell volume via phase separation. Cell. 2022;185:4488–4506.:e20. doi: 10.1016/j.cell.2022.09.042. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R58] 58.Subramanya AR, Boyd-Shiwarski CR. Molecular Crowding: Physiologic Sensing and Control. Annu Rev Physiol. 2024;86:429–452. doi: 10.1146/annurev-physiol-042222-025920. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R59] 59.Stangherlin A, et al. Compensatory ion transport buffers daily protein rhythms to regulate osmotic balance and cellular physiology. Nat Commun. 2021;12:6035. doi: 10.1038/s41467-021-25942-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R60] 60.Watson JL, et al. Macromolecular condensation buffers intracellular water potential. Nature. 2023;623:842–852. doi: 10.1038/s41586-023-06626-z. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R61] 61.Christmas MJ, et al. Evolutionary constraint and innovation across hundreds of placental mammals. Science. 2023;380:eabn3943. doi: 10.1126/science.abn3943. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R62] 62.Juszkiewicz S, et al. ZNF598 Is a Quality Control Sensor of Collided Ribosomes. Mol Cell. 2018;72:469–481.:e7. doi: 10.1016/j.molcel.2018.08.037. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R63] 63.Khalsa SB, Whitmore D, Block GD. Stopping the circadian pacemaker with inhibitors of protein synthesis. Proc National Acad Sci. 1992;89:10862–10866. doi: 10.1073/pnas.89.22.10862. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R64] 64.Yamaguchi S, et al. Synchronization of Cellular Clocks in the Suprachiasmatic Nucleus. Science. 2003;302:1408–1412. doi: 10.1126/science.1089287. [DOI] [PubMed] [Google Scholar]

[R65] 65.Abraham U, et al. Coupling governs entrainment range of circadian clocks. Mol Syst Biol. 2010;6:438. doi: 10.1038/msb.2010.92. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R66] 66.Ruby NF, Burns DE, Heller HC. Circadian Rhythms in the Suprachiasmatic Nucleus are Temperature-Compensated and Phase-Shifted by Heat Pulses In Vitro. J Neurosci. 1999;19:8630–8636. doi: 10.1523/JNEUROSCI.19-19-08630.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R67] 67.Pilorz V, et al. A novel mechanism controlling resetting speed of the circadian clock to environmental stimuli. Curr Biology Cb. 2013;24:766–73. doi: 10.1016/j.cub.2014.02.027. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R68] 68.Hsieh AC, et al. The translational landscape of mTOR signalling steers cancer initiation and metastasis. Nature. 2012;485:55–61. doi: 10.1038/nature10912. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R69] 69.Naito T, Kuma A, Mizushima N. Differential Contribution of Insulin and Amino Acids to the mTORC1-Autophagy Pathway in the Liver and Muscle*. J Biol Chem. 2013;288:21074–21081. doi: 10.1074/jbc.M113.456228. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R70] 70.Daan S, et al. Lab Mice in the Field: Unorthodox Daily Activity and Effects of a Dysfunctional Circadian Clock Allele. J Biol Rhythm. 2011;26:118–129. doi: 10.1177/0748730410397645. [DOI] [PubMed] [Google Scholar]

[R71] 71.van Rosmalen L, et al. Multi-organ transcriptome atlas of a mouse model of relative energy deficiency in sport. Cell Metab. 2024;36:2015–2037.:e6. doi: 10.1016/j.cmet.2024.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R72] 72.Hut RA, Pilorz V, Boerema AS, Strijkstra AM, Daan S. Working for Food Shifts Nocturnal Mouse Activity into the Day. Plos One. 2011;6:e17527. doi: 10.1371/journal.pone.0017527. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R73] 73.van der Vinne V, et al. Cold and hunger induce diurnality in a nocturnal mammal. Proc National Acad Sci. 2014;111:15256–15260. doi: 10.1073/pnas.1413135111. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R74] 74.van Rosmalen L, et al. Energy balance drives diurnal and nocturnal brain transcriptome rhythms. Cell Rep. 2024;43:113951. doi: 10.1016/j.celrep.2024.113951. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R75] 75.Gu X, et al. SAMTOR is an S-adenosylmethionine sensor for the mTORC1 pathway. Science. 2017;358:813–818. doi: 10.1126/science.aao3265. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R76] 76.Doyle SE, Yoshikawa T, Hillson H, Menaker M. Retinal pathways influence temporal niche. Proc National Acad Sci. 2008;105:13133–13138. doi: 10.1073/pnas.0801728105. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R77] 77.Schoonderwoerd RA, et al. Inhibitory responses to retinohypothalamic tract stimulation in the circadian clock of the diurnal rodent Rhabdomys pumilio. FASEB J. 2022;36:e22415. doi: 10.1096/fj.202200477R. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R78] 78.Richardson R, et al. The genomic basis of temporal niche evolution in a diurnal rodent. Curr Biol. 2023;33:3289–3298.:e6. doi: 10.1016/j.cub.2023.06.068. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R79] 79.Riede SJ, van der Vinne V, Hut RA. The flexible clock: predictive and reactive homeostasis, energy balance and the circadian regulation of sleep–wake timing. J Exp Biol. 2017;220:738–749. doi: 10.1242/jeb.130757. [DOI] [PubMed] [Google Scholar]

[R80] 80.Stangherlin A, Seinkmane E, O’Neill JS. Understanding circadian regulation of mammalian cell function, protein homeostasis, and metabolism. Curr Opin Syst Biology. 2021;28:100391. doi: 10.1016/j.coisb.2021.100391. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R81] 81.Moreira MO, Qu Y, Wiens JJ. Large-scale evolution of body temperatures in land vertebrates. Evol Lett. 2021;5:484–494. doi: 10.1002/evl3.249. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R82] 82.Lewden A, Bishop CM, Askew GN. How birds dissipate heat before, during and after flight. J R Soc Interface. 2023;20:20230442. doi: 10.1098/rsif.2023.0442. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R83] 83.Lindhout FW, Krienen FM, Pollard KS, Lancaster MA. A molecular and cellular perspective on human brain evolution and tempo. Nature. 2024;630:596–608. doi: 10.1038/s41586-024-07521-x. [DOI] [PubMed] [Google Scholar]

[R84] 84.Iyer DP, et al. mTOR activity paces human blastocyst stage developmental progression. Cell. 2024;187:6566–6583.:e22. doi: 10.1016/j.cell.2024.08.048. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R85] 85.Iwata R, Vanderhaeghen P. Metabolic mechanisms of species-specific developmental tempo. Dev Cell. 2024;59:1628–1639. doi: 10.1016/j.devcel.2024.05.027. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R86] 86.Seluanov A, Vaidya A, Gorbunova V. Establishing Primary Adult Fibroblast Cultures From Rodents. J Vis Exp. 2010 doi: 10.3791/2033. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R87] 87.Drouin-Ouellet J, et al. REST suppression mediates neural conversion of adult human fibroblasts via microRNA-dependent and -independent pathways. EMBO Mol Medicine. 2017;9:1117–1131. doi: 10.15252/emmm.201607471. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R88] 88.Labun K, Montague TG, Gagnon JA, Thyme SB, Valen E. CHOPCHOP v2: a web tool for the next generation of CRISPR genome engineering. Nucleic Acids Res. 2016;44:W272–W276. doi: 10.1093/nar/gkw398. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R89] 89.Zeng A, O’Neill JS. Using ALLIGATORs to Capture Circadian Bioluminescence. Circadian Regulation of Methods in Molecular Biology. 2022;2482:125–135. doi: 10.1007/978-1-0716-2249-0_8. [DOI] [PubMed] [Google Scholar]

[R90] 90.Verma M, et al. A short translational ramp determines the efficiency of protein synthesis. Nat Commun. 2019;10:5774. doi: 10.1038/s41467-019-13810-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R91] 91.Cencic R, Pelletier J. Hippuristanol - A potent steroid inhibitor of eukaryotic initiation factor 4A. Translation. 2016;4:e1137381. doi: 10.1080/21690731.2015.1137381. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R92] 92.Hastings MH, Reddy AB, McMahon DG, Maywood ES. Analysis of Circadian Mechanisms in the Suprachiasmatic Nucleus by Transgenesis and Biolistic Transfection. Methods Enzymol. 2005;393:579–592. doi: 10.1016/S0076-6879(05)93030-9. [DOI] [PubMed] [Google Scholar]

[R93] 93.Kowalczyk A, et al. RERconverge: an R package for associating evolutionary rates with convergent traits. Bioinformatics. 2019;35:4815–4817. doi: 10.1093/bioinformatics/btz468. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R94] 94.Lotthammer JM, et al. Metapredict enables accurate disorder prediction across the Tree of Life. bioRxiv. 2024:2024.11.05.622168 [Google Scholar]

[R95] 95.Bárcena C, et al. Methionine Restriction Extends Lifespan in Progeroid Mice and Alters Lipid and Bile Acid Metabolism. Cell Rep. 2018;24:2392–2403. doi: 10.1016/j.celrep.2018.07.089. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R96] 96.Comas M, Hut RA. Twilight and Photoperiod Affect Behavioral Entrainment in the House Mouse (Mus musculus) J Biol Rhythm. 2009;24:403–412. doi: 10.1177/0748730409343873. [DOI] [PubMed] [Google Scholar]

[R97] 97.Brown SA, et al. The Period Length of Fibroblast Circadian Gene Expression Varies Widely among Human Individuals. PLoS Biol. 2005;3:e338. doi: 10.1371/journal.pbio.0030338. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R98] 98.Pelham JF, Dunlap JC, Hurley JM. Intrinsic disorder is an essential characteristic of components in the conserved circadian circuit. Cell Commun Signal. 2020;18:181. doi: 10.1186/s12964-020-00658-y. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R99] 99.Düvel K, et al. Activation of a Metabolic Gene Regulatory Network Downstream of mTOR Complex 1. Mol Cell. 2010;39:171–183. doi: 10.1016/j.molcel.2010.06.022. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

A cellular basis for the mammalian nocturnal-diurnal switch

Andrew D Beale

Matthew J Christmas

Nina M Rzechorzek

Andrei Mihut

Aiwei Zeng

Christopher Ellis

Nathan R James

Nicola J Smyllie

Violetta Pilorz

Rose Richardson

Mads F Bertelsen

Shaline V Fazal

Zanna Voysey

Kevin Moreau

Jerry Pelletier

Priya Crosby

Sew Y Peak-Chew

Rachel S Edgar

Madeline A Lancaster

Roelof A Hut

John S O’Neill

Abstract

Fig. 1. Entrainment to temperature is a cellular correlate of behavioral temporal niche.

Cellular circadian rhythms of nocturnal and diurnal mammals are differentially entrained by daily temperature cycles

Diurnal cellular clocks are buffered against temperature change

Fig. 2. Differential response to temperature is both a specific property of PER2 and global translation.

Potential role of mTOR and WNK in global species differences in the response to temperature change

Fig. 3. Differential mTOR pathway activity as the basis of the nocturnal-diurnal switch.

Convergent evolution of diurnal response to thermodynamic perturbation

Perturbation of mTOR activity and translational initiation makes nocturnal cells behave like diurnal cells

Fig. 4. Manipulation of mTOR pathway activity alters phase of entrainment.

mTOR regulation of circadian phase is maintained from cells to tissues

Fig. 5. mTOR pathway activity underlies phase of entrainment in tissues and the nocturnal-diurnal switch in vivo.

mTOR activity contributes to temporal niche selection

Discussion

Materials and Methods

Cell culture

Plasmids

Experimental cell culture - static

Experimental cell culture - flow

Animals and organotypic culture

Temperature shift experiments

Puromycin labelling

Protein extraction

Proteomics and phosphoproteomics

Digestion

Tandem mass tag (TMT) labelling

Titanium dioxide (TiO2) – Enrichment of phosphopeptides

Off-line high pH reverse-phase peptides fractionation

LC-MS/MS

Raw MS data processing

Proteomics data analysis

Immunoblotting

Comparative genomic analysis

In vivo mTOR activity reduction

Statistics

Supplementary Material

Acknowledgements

Funding

Footnotes

Data and materials availability

References and Notes

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

Titanium dioxide (TiO₂) – Enrichment of phosphopeptides