Cycling of CRYPTOCHROME proteins is not necessary for circadian-clock function in mammalian fibroblasts

Abstract

Background: An interlocked transcriptional-translational feedback loop (TTFL) is thought to generate the mammalian circadian clockwork in both the central pacemaker residing in the hypothalamic suprachiasmatic nuclei and in peripheral tissues. The core circadian genes, including Period1 and Period2 (Per1 and Per2), Cryptochrome1 and Cryptochrome2 (Cry1 and Cry2), Bmal1, and Clock are indispensable components of this biological clockwork. The cycling of the PER and CRY clock proteins has been thought to be necessary to keep the mammalian clock ticking.

Results: We provide a novel cell-permeant protein approach for manipulating cryptochrome protein levels to evaluate the current transcription and translation feedback model of the circadian clockwork. Cell-permeant cryptochrome proteins appear to be functional on the basis of several criteria, including the abilities to (1) rescue circadian properties in Cry1(-/-)Cry2(-/-) mouse fibroblasts, (2) act as transcriptional repressors, and (3) phase shift the circadian oscillator in Rat-1 fibroblasts. By using cell-permeant cryptochrome proteins, we demonstrate that cycling of CRY1, CRY2, and BMAL1 is not necessary for circadian-clock function in fibroblasts.

Conclusions: These results are not supportive of the current version of the transcription and translation feedback-loop model of the mammalian clock mechanism, in which cycling of the essential clock proteins CRY1 and CRY2 is thought to be necessary.

Figure 1. Uptake of Cell-Permeant Cryptochrome Proteins into Rat-1 Fibroblasts

(A) Schematic structure of the recombinant CP proteins. (B) Affinity-purified CP-CRY1 and CP-CRY2 proteins expressed in E. coli. (C) Rat-1 cells were incubated with 50 nM of fluorescein-labeled CP-CRY1 (C₁), CP-CRY2 (C₂), CP-mutCRY1 (C₃), CP-CRE (C₄), or FITC (no protein) (C₅). After a 30 min incubation, cells were treated with proteinase K to remove any fluorescein-labeled protein that nonspecifically adhered to the surface of the cells.

Figure 2. Repression of BMAL1/CLOCK-Mediated Transcription by CP-CRY1 and CP-CRY2

HEK293 cells were transfected with haBmal1, mouse Clock, and an E box-containing promoter P_PK2-driven luciferase reporter (P_PK2::luc). (A) Intracellularly expressed CRY (P_CMV::Cry1) and extracellularly applied CP-CRYs on P_PK2 activity were evaluated with a luciferase assay of PK2 promoter activity. The concentrations of CP-CRY1, CP-CRY2, and CP-CRE proteins were 250 nM. (B) Both transfected P_CMV::Cry1 and extracellularly applied CP-CRY can inhibit P_PK2 activity. Extracellularly applied CP-CRY1 and CP-mutCRY1 both inhibit P_PK2 activity in a dose-dependent fashion, but CP-mutCRY1 is less effective than CP-CRY1. Protein concentrations are as follows: +, 5 nM; ++, 50 nM; and +++, 250 nM. The plotted values are mean±SD of three replicates. Asterisks indicate significant differences from haBmal1/mClock controls at the p < 0.01 level (Student’s t test).

Figure 3. Circadian Properties Can Be Rescued by CP-CRYs in Cry1^−/−Cry2^−/− Knockout Mice Fibroblasts

(A) mPER1 expression level is high and arrhythmic in Cry1^−/−Cry2^−/− mouse fibroblasts. Treatment with CP-CRY1 and CPCRY2 reduces mPER1 expression levels to wild-type levels. (B) Quantification of results shown in (A). Protein levels of mPER1 were normalized to the actin loading control. (C) “Rescue” of circadian rhythmicity in Cry1^−/−Cry2^−/− knockout mice fibroblasts. The P_Bmal1::Fluc luminescence reporter was introduced to the fibroblasts by nucleofection. After CP-CRY1 and CP-CRY2 treatment, the P_Bmal1::Fluc expression level is elevated and rhythmicity is recovered. In the wild-type mouse fibroblasts, rhythms are sustained in both the untreated and the CPCRY1- + CP-CRY2-treated cell lines. Data were detrended with LumiCycle software. Colors represent the following: wild-type (blue), wild-type + CP-CRYs (green), knockout (red), and knockout + CP-CRYs (black).

Figure 4. Impact of CP-CRY1 and CP-CRY2 Treatment upon BMAL1-Protein Abundance in Rat-1 Fibroblasts as Measured by Immunoblotting

(A) BMAL1-protein expression in response to continuous treatment of CP-CRY1, CP-CRY2, or CP-CRE. PBS is the phosphate-buffered saline control. (B) BMAL1 expression in response to 3 hr or 6 hr pulse treatments of CP-CRY1. (C) Comparison of the BMAL1 expression level of Rat-1 cells treated with different concentrations of CP-mCRY1 or CP-mutCRY1 as compared with the actin control.

Figure 5. Phase Shifts Induced by Step Treatments of CP-CRY1 and CP-mCRY2

Rhythms in Rat-1 cells with stably transfected P_Per2::Fluc reporters were initiated by a 2 hr forskolin pulse (10 μM). (A) CP-CRY1 (50 nM) was added to the assay medium at the phases indicated by arrows. Addition of CP-CRE (50 nM) or PBS were controls. (B) CP-CRY1 is more effective than CP-mutCRY1 as a phase-shifting agent at a 5 nM concentration. (C) Regression analyses of CP-CRY1-induced phase shifts. The left panel shows phase advance, and the right panel shows phase delay; open symbols are CP-CRE-treated controls, and filled symbols are CP-CRY1-treated samples. (D) PRCs to step treatments of CP-CRY1 and CP-CRY2 on Rat-1 fibroblasts. Black and blue symbols are CP-CRY1 and CP-CRY2 treatments, respectively. Red symbols are CP-CRE treatments. Error bars represent ± SD.

Figure 6. Administration of CP-CRY Proteins Does Not Suppress Circadian Rhythmicity of the P_Per2::Fluc Reporter in Rat-1 Cells

CP-proteins were added at the time indicated by the arrow (concentration = 50 nM for each protein).