The Deubiquitinase USP2 Modulates Photic Entrainment of the Circadian Clock at the Level of the Suprachiasmatic Nucleus

Abstract

Ubiquitin-specific peptidase 2 (USP2) is a deubiquitinase (DUB) with a diversity of functions in physiology. One of these functions is the regulation of circadian rhythms, which are physiological rhythms with a period of ~24 h. Previous studies have indicated a role for USP2 in photic entrainment, the process by which circadian clocks synchronize to environmental light cues. Here, we investigated the implication of USP2 in this process, using Usp2 knockout (KO) mice. Using different light treatments and running wheel recordings, we established that USP2 controls entrainment of the clock to light cues at dusk. Further, we showed that Usp2 is expressed throughout the suprachiasmatic nucleus (SCN), the site of the central clock, and in the retina. This raised the question of where USP2 acts on circadian photoreception. We found that it is not within the retina, as retinas of Usp2 KO mice have an intact structure and unaltered photoreception through intrinsically photosensitive retinal ganglion cells. Moreover, KO of Usp2 within the retina does not alter clock entrainment to light. In contract, KO of Usp2 in the SCN causes a light entrainment phenotype similar to full-body KO mice, showing that the action of USP2 in modulating photic entrainment predominantly takes place in the SCN. Finally, within the SCN, we found that induction of clock gene Per1 and activation of MAPK/ERK pathway in response to light were blunted in Usp2 KO mice. Altogether, we established a key role for USP2 in regulating photic entrainment by modulating light-responsive pathways within the SCN.

Keywords: USP2; circadian clock light; knock‐out mouse; retina; suprachiasmatic nucleus.

FIGURE 1

USP2 differentially affects entrainment at different times at night. (A) Representative actograms for wheel‐running activity of WT and Usp2 KO mice light pulsed at CT 12 or CT 14. Schematic created in Biorender. (B) Phase response curve for the response of WT and Usp2 KO mice to a light pulse. Data are represented as mean ± SEM (n = 3–8 mice per CT). Two‐way ANOVA, with Holm‐Šídák post hoc test. *p < 0.05 (C) Representative actograms for WT and Usp2 KO mice going through the modified skeleton photoperiod protocol. Schematic created in Biorender. (D, E) Quantification of two parameters of entrainment, interdaily stability (D) and phase angle of entrainment (E). ASPP = phase advance with mSPP; DSPP, phase delay with mSPP; SPP, skeleton photoperiod. Individual data points represent individual mice (n = 15 mice per group), and data are represented as mean ± SEM. Mixed model two‐way ANOVA, with Holm‐Šídák post hoc test. *p < 0.05, **p < 0.01.

FIGURE 2

USP2 affects rhythmic gene expression in the retina, but not retinal structure. (A, B) Quantification of gene expression of clock genes (A) and retinal opsin genes (B) in the retina. Data are represented as mean ± SEM (n = 3–5 retinas per group). Data with significant cosine curve fitting are represented by a continuous curve; data with no significant cosine curve fitting are represented by dots connected by lines. P values in the graphs are for the cosinor analysis. The results of the 2‐way ANOVA analysis (for effects of time and genotype) are in Table S3. (C) Representative hematoxylin/eosin‐stained images of the retinas of WT and Usp2 KO mice. (D) Representative fluorescence microscopy images of OTX2‐stained retinas of WT and Usp2 KO mice. Scale bar, 50 μm. (E) Quantification of outer nuclear layer (ONL) thickness across the retina of WT and Usp2 KO mice. (F) Quantification of OTX2 positive cells in the INL of WT and Usp2 KO mice (n = 4 per group). Data are represented as mean ± SEM. Two‐tailed t‐test, ns.

FIGURE 3

USP2 does not affect circadian photoreception at the level of the retina. (A) Representative images of pupils of WT and Usp2 KO mice undergoing the PLR test. Pre‐stim picture at time = 5 s and stim picture at time = 10 s. (B) Average pupillary constriction throughout the PLR experiment. The lighter umbra around the line represents SEM. “Stim” represents the time interval during which the light stimulation was provided (C) quantification of pupil constriction in response to the light stimulation. Individual datapoints represent individual mice (n = 7–9 mice per group), and data are represented as mean ± SEM. Two‐tailed t‐test, ns. (D) Representative wheel‐running actograms of AAV‐Sham‐ and AAV‐Cre‐injected floxed Usp2 mice subjected to the modified skeleton photoperiod protocol. Schematic created in Biorender. (E) Quantification of phase angle of entrainment. ASPP = phase advance with mSPP; DSPP, phase delay with mSPP, SPP, skeleton photoperiod. Individual data points represent individual mice (n = 5 mice per group), and data are represented as mean ± SEM. Mixed model two‐way ANOVA, ns. (F) Quantification of Usp2 gene expression levels in the eyes of AAV‐Sham‐ and AAV‐Cre‐injected mice using qRT‐PCR. Individual data points represent individual mice (n = 5 mice per group), and data are represented as mean ± SEM. Two‐tailed t‐test with Welch correction, **p < 0.01.

FIGURE 4

USP2 varies differently in distinct neuronal subpopulations within the SCN and affects photic entrainment at the level of the SCN. (A) Representative RNAscope images of Usp2, Avp, and Vip, and merged with DAPI at CT 1, 7, 13, and 19. Scale bar, 100 μm. (B) Quantification of the percentage of SCN cells that express Usp2 across time points. (C) Quantification of the percentage of cells in the SCN that express Usp2 along with the specific neuropeptide(s). Individual points represent individual SCN lobes. All data are represented as mean ± SEM (n = 5–8 SCN lobes from 3 to 4 mice per group). Significant rhythms are represented by a smooth Cosinor fit line. (D) Representative images showing SCN‐specific targeting of AAV‐Cre and AAV‐Sham (green channel). Scale bar, 100 μm. (E) Representative wheel‐running actograms of AAV‐Sham‐ and AAV‐Cre‐injected mice subjected to the modified skeleton photoperiod protocol. Schematic created in Biorender. (F) Quantification of phase angle of entrainment. SPP = skeleton photoperiod, DSPP = phase delay with mSPP, and ASPP = phase advance with mSPP. Individual data points represent individual mice (n = 8–10 mice per group), and data are represented as mean ± SEM. Mixed model two‐way ANOVA, with Holm‐Šídák post hoc test. *p < 0.05.

FIGURE 5

USP2 affects light‐responsive signaling pathways within the SCN in a region‐specific manner. (a) Representative images for Per1 in situ hybridization at different time points around the modified skeleton photoperiod dusk light pulse. The red line represents the SCN region of interest. Scale bar, 100 μm. (B) Quantification of Per1 staining in WT and Usp2 KO mice. (C) Quantification of Per1 staining in the dorsal and ventral 50% of cells of the SCN. In (B, C), data points are represented as mean ± SEM (n = 3–6 mice per group). Two‐way ANOVA, with Holm‐Šídák post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001. (D) Representative fluorescence microscopy images of phosphorylated ERK1/2 (pERK) staining 30 min after a light pulse and merged images with DAPI staining. Scale bar, 100 μm. (E) Quantification of pERK staining averaged across the SCNs of WT and Usp2 KO mice. Individual data points represent individual mice (n = 3 mice per group). All data are represented as mean ± SEM. Two‐tailed t‐test, ns.