Identification of small molecule activators of cryptochrome

Abstract

Impairment of the circadian clock has been associated with numerous disorders, including metabolic disease. Although small molecules that modulate clock function might offer therapeutic approaches to such diseases, only a few compounds have been identified that selectively target core clock proteins. From an unbiased cell-based circadian phenotypic screen, we identified KL001, a small molecule that specifically interacts with cryptochrome (CRY). KL001 prevented ubiquitin-dependent degradation of CRY, resulting in lengthening of the circadian period. In combination with mathematical modeling, our studies using KL001 revealed that CRY1 and CRY2 share a similar functional role in the period regulation. Furthermore, KL001-mediated CRY stabilization inhibited glucagon-induced gluconeogenesis in primary hepatocytes. KL001 thus provides a tool to study the regulation of CRY-dependent physiology and aid development of clock-based therapeutics of diabetes.

Fig. 1

Carbazole derivatives lengthen the circadian period. (A) The chemical structure of carbazole derivatives. (B and C) Luminescence rhythms of Bmal1-dLuc and Per2-dLuc reporter U2OS cells were monitored in the presence of various concentrations of compound. Representative traces (n = 2) are shown in (B). The changes of period and luminescence intensity (an average of 24–120 h; IC₅₀ values are indicated in insets of bottom panels) relative to DMSO control are shown in (C) as mean ± SEM (n = 4). When arrhythmic, the period was not plotted. (D and E) Luminescence rhythms of mPer2^Luc knock-in reporter in mouse SCN (D) and lung (E) explants were monitored in the presence of increasing concentration of KL001 (each for 120 h). Data are mean ± SEM (n = 6 for SCN and 5 for lung).

Fig. 2

KL001 interacts with CRY1 and CRY2. (A) Agarose-conjugated KL001-linker compound was incubated with lysate of unsynchronized U2OS cells in the presence of various concentrations of free KL001 as a competitor. Bound proteins were identified by protein immunoblotting. Asterisk indicates nonspecific band. ppt, pellet. (B) Flag-tagged PER or CRY was transiently overexpressed in HEK293T cells, and lysates containing PER (PER) or a mixture of lysates containing PER and CRY (PER-CRY) were subjected to the pull-down assay. Bound proteins were detected with anti-Flag antibody. (C) Effects of KL001, KL002, KL004 and FAD on interaction of CRY1-Flag with KL001-agarose conjugate. (D) Wild type (WT) or FAD binding site mutant (D387N or N393D) of CRY1- Flag was subjected to the pull-down assay.

Fig. 3

KL001 stabilizes CRY proteins. (A) Effect of KL001 on mPer2^Luc knock-in reporter in Cry1/2 double knockout mouse fibroblasts. Data are mean ± SEM (n = 4). (B and C) Confluent unsynchronized U2OS cells were treated with KL001 for 48 h, and then subjected to RT-qPCR (B; mean ± SEM, n = 3) or protein immunoblot (C) analysis. (D) Luciferase-fused CRY1 (CRY1-LUC), its D387N mutant (CRY1D387N-LUC) or luciferase (LUC) was transiently overexpressed in HEK293T cells. The cells were treated with KL001 for 24 h, and then cycloheximide was added for luminescence recording. Profiles are shown by setting peak luminescence as 1 (left panels). Half-life of CRY1-LUC or CRY1D387N-LUC relative to LUC is shown by setting CRY1-LUC 0 µM condition as 1 (right panel). Data are mean ± SEM (n = 8). (E) Effects of KL001, KL002 and KL004 on CRY1 and CRY2 stability in HEK293 stable cell lines expressing CRY1-LUC, CRY2- LUC or LUC. Data are mean ± SEM (n = 4). (F) Effect of FBXL3 knockdown on the action of KL001 in Bmal1-dLuc or Per2-dLuc reporter U2OS cells. Data are mean ± SEM (n = 4).

Fig. 4

Application of KL001 to define roles of CRY isoforms (A to C) and to control hepatic gluconeogenesis (D and E). (A) Scheme of the mathematical model consisting of the two parallel CRY feedback loops. (B) Effect of nuclear CRY stabilization on the period in wild type, Cry1 knockout and Cry2 knockout cells in silico. (C) Cry1 or Cry2 knockout mPer2^Luc knock-in mouse fibroblasts were stimulated with dexamethasone (Dex) for 2 h, and luminescence rhythms were monitored in the presence of KL001. Data are mean ± SEM (n = 4). (D and E) Mouse primary hepatocytes were treated with KL001 for 18 h and then stimulated with 10 nM glucagon for 2 h (D, for RT-qPCR analysis) or 3 h (E, for glucose assay). To measure glucose production, the cells were further incubated with glucose-free buffer containing 20 mM sodium lactate and 2 mM sodium pyruvate for 4 h (E). Data are mean ± SEM (n = 3).