Fig. 2: Circadian and ultradian rhythms underlie coupling of intracranial activity and interstitial glucose dynamics. | Nature Communications

Fig. 2: Circadian and ultradian rhythms underlie coupling of intracranial activity and interstitial glucose dynamics.

From: Spectro-spatial features in distributed human intracranial activity proactively encode peripheral metabolic activity

Fig. 2

A Wavelet coherence spectrum between hypothalamic HFA and interstitial glucose levels revealing high coherence at circadian periodicity. Arrows denote phase lag. B Corresponding coherence periodogram showing peaks in ultradian and circadian periodicity. C Circadian coherence between HFA and interstitial glucose across subcortical and corticolimbic regions (Nsubj = 3, Nchannels = 288, one-way ANOVA; F(6,287) = 13, P < 0.001). D Average cortico-cortical evoked potentials (CCEP) elicited by left hypothalamic stimulation in S1 stratified by whether channels’ HFA were significantly coupled to glucose in a circadian manner. Channels with significant glucose coupling showed an overall larger CCEP with left hypothalamic stimulation. Shaded error bars indicate error of the mean (SEM). E Average CCEP magnitude (0.01–0.1 s) was significantly higher in channels with significant circadian coherence to glucose (two-sample t-test, t(118) = 2.9, P < 0.001). F Left hypothalamic CCEP magnitude correlates directly with HFA-glucose lag-corrected correlation across all channels in S1 (Nchannels = 121, Pearson’s R = 0.42, P < 0.001). G Wavelet decomposition of circadian and non-circadian rhythms in interstitial glucose dynamics and mean hypothalamic HFA (Subject 1). Combining the circadian and non-circadian rhythms reconstruct the original signal trace. H Removal of circadian contribution to HFA and interstitial glucose variations resulted in significant reduction of correlation across 3 subjects (Paired t-test with statistic shown for each subject). I Lag-corrected non-circadian correlation between HFA and interstitial glucose across subcortical and corticolimbic regions (Nsubj = 3, Nchannels = 288, one-way ANOVA; F(6,287) = 3, P = 0.007). J Subcortical and corticolimbic lag-corrected non-circadian HFA-glucose correlation stratified by sleep and wake states across three subjects (Nchannels per region is noted by the degree of freedom, all tests were Paired t-test with statistics and P-value shown per region). All error bars indicate SEM. Source data are provided as a Source data file. *P < 0.05, **P < 0.01, ***P < 0.001.

Back to article page