Circadian clocks are fundamental to the biology of most eukaryotes, coordinating behavior and physiology to resonate with the environmental cycle of day and night through complex networks of clock-controlled genes1-3. A fundamental knowledge gap exists however, between circadian gene expression cycles and the biochemical mechanisms that ultimately facilitate circadian regulation of cell biology4,5. Here we report circadian rhythms in the intracellular concentration of magnesium ions, [Mg 2+ ] i , which act as a cell-autonomous timekeeping component to determine key clock properties in both a human cell line and a unicellular alga that diverged from metazoans more than 1 billion years ago6. Given the essential role of Mg 2+ as a cofactor for ATP, a functional consequence of [Mg 2+ ] i oscillations is dynamic regulation of cellular energy expenditure over the daily cycle. Mechanistically, we find that these rhythms provide bilateral feedback linking rhythmic metabolism to clock-controlled gene expression. The global regulation of nucleotide triphosphate turnover by intracellular Mg 2+ availability has potential to impact upon many of the cell's >600 MgATP-dependent enzymes7 and every cellular system where MgNTP hydrolysis becomes rate limiting. Indeed, we find that circadian control of translation by mTOR8 is regulated through [Mg 2+ ] i oscillations. It will now be important to identify which additional biological # To whom correspondence should be sent: Gerben.vanOoijen@ed.ac.uk, oneillj@mrc-lmb.cam.ac.uk. Author contributions GvO and JSO conceived the approach and designed the study. LFL and COY generated the Neurospora result. JD and LE performed ICP analyses. GvO and LLH performed Ostreococcus experiments. Human U2OS cell experiments were performed by KAF. MP performed mouse fibroblast experiments. NPH provided analytical and intellectual contributions. GvO and JSO wrote the manuscript.

[1] A method has been developed for rapid and precise simultaneous determination of nine element/Ca ratios in foraminiferal tests directly from intensity ratios using external, matrix-matched standards on a quadrupole inductively coupled plasma-mass spectrometer (ICP-MS). All quantification isotopes are determined in pulse mode to avoid cross-calibration. Small argide ( Zn are corrected by using two additional Mg and Zn standards. A stable signal, conducive for high-precision measurements, is obtained by cone conditioning. Variable calcium concentration has negligible effect on Li, Al, Mn, and Sr, but Ca concentrations for standards and samples need to be constrained at a similar level for precise measurements of Zn, Cd, and U. Aliquots of samples are first analyzed for Ca concentrations on an inductively coupled plasma-atomic emission spectrometer (ICP-AES), and the remaining solutions are diluted to Ca concentration of 100 ppm for ratio measurements to assure data quality. The long-term reproducibility of the method yielded precisions of Li/Ca = 2.4%, B/Ca = 4.2%, Mg/Ca = 1.4%, Al/Ca = 14%, Mn/Ca = 0.9%, Zn/Ca = 2.8% (1.2$7.8 mmol/mol) and 5.1% (0.5$1.2 mmol/mol), Sr/Ca = 0.9%, Cd/Ca = 2.4% (0.07$0.24 mmol/mol) and 4.8% (0.01$0.07 mmol/mol), and U/Ca = 2.5% for foraminiferal samples as small as 60 mg.

[1] Analyses of basalts from the Galápagos archipelago and adjacent spreading center have shown that the underlying mantle plume is isotopically heterogeneous over length scales of 10s to 100s of km. We show that the convecting mantle is also compositionally heterogeneous on smaller length scales (kms). Our evidence comes from recent small-volume mildly alkaline and tholeiitic basalts on Isla Santiago, central Galá-pagos. Flows from volcanic vents <5 km apart are homogeneous in terms of incompatible-trace element and isotopic ratios but inter-vent variations in these ratios are large, such that Santiago basalts display some of the most extensive ranges known for any Galápagos island. Geochemical indexes of depth of melting correlate with an eastward decrease in geophysical estimates of lithospheric thickness-from 55 to 43 km over a $10 km horizontal distance beneath Isla Santiago-suggesting that melts have not undergone significant lateral transport in the underlying crust. This lithospheric 'step' below the center of the island generally results in a greater proportion of enriched mantle melts contributing to basalts from west Santiago, than to those in the east, due to less melting of more depleted mantle as the plume upwells beneath thicker lithosphere. Nevertheless, the sporadic occurrence of isotopically enriched flows with low [Sm/Yb] n, and more isotopically depleted compositions with elevated [Sm/Yb] n, across Santiago suggests that portions of the underlying Galápagos plume are compositionally heterogeneous in terms of enriched and depleted reservoirs over short length scales (kilometers). In this respect the Galápagos mantle plume is similar to plumes beneath Hawaii and Canary Islands.

No abstract

[1] We have investigated the effect of cleaning procedures on eight element/Ca ratios in three widely used benthic foraminifera species from core top and down core sediments. Two cleaning techniques were employed: (1) comparison between ''Mg-cleaning'' and ''Cd-cleaning'' methods and (2) comparison between the various constituent reagents. Li/Ca, B/Ca, and Sr/Ca ratios remained unchanged for samples subjected to different treatments, but Mg/Ca, Mn/Ca, Zn/Ca, Cd/Ca, and U/Ca were substantially decreased when foraminifera shells were cleaned with reagents containing citrate. In contrast, no significant decreases in element/Ca ratios were observed for samples cleaned in N 2 H 4 without citrate, indicating that the role of N 2 H 4 was insignificant during reductive cleaning and that citrate is responsible for decreases of Mg/Ca, Mn/Ca, Zn/Ca, Cd/Ca, and U/Ca. Decreases in these ratios are most likely due to (1) removal of contaminant particles enriched in Mg, Mn, Zn, Cd and U and/or (2) preferential leaching of CaCO 3 by the formation of stable metal complexes through chelation between citrate and metals heterogeneously distributed in foraminiferal shells. Due to the negligible effect of N 2 H 4 and preferential dissolution of CaCO 3 during reductive cleaning, it is suggested that the reductive cleaning step should be omitted and the ''Mg-cleaning'' method [Barker et al., 2003] could be employed to clean foraminiferal shells for trace element measurements.