Diverse and rhythmic expression patterns of zebrafish cry genes.

(A) Rhythmic expression of zebrafish cry1aa determined by qRT-PCR. (B) Rhythmic expression of zebrafish cry1ab determined by qRT-PCR. (C) Rhythmic expression of zebrafish cry1ba determined by qRT-PCR. (D) Rhythmic expression of zebrafish cry1bb determined by qRT-PCR. (E) Rhythmic expression of zebrafish cry2 determined by qRT-PCR. (F) Rhythmic expression of zebrafish cry3 determined by qRT-PCR. Each value is the mean ± SD of the three independent experiments. (G) Lateral views (anterior to left) of expression patterns of zebrafish cry1aa, cry1ab, cry1ba, cry1bb, cry2 and cry3 shown by in situ hybridization. Whole-mount in situ hybridization was performed to examine expression patterns of six cry genes from 72hpf to 116hpf (Day 4 and Day 5 post fertilization, at 4h intervals each day). Black in the horizontal bars at the top represent darkness, and white bars indicate light. Zebrafish cry1aa, cry1ab, cry1ba, cry1bb, cry2 and cry3 display diverse patterns of rhythmic expression in larvae. cry1aa and cry1ab display distinct but similar rhythmic expression patterns, both reaching a trough at ZT15-19. The oscillating patterns of cry1ba and cry1bb are similar, both peaking at ZT13-15. The oscillating patterns of cry2 and cry3 differed from each other: cry2 peaked at ZT23-1, while cry3 peaked at ZT9. In situ hybridization results are consistent with those of qRT-PCR.

Repression of zebrafish Cry proteins on different Clock:Bmal combinations.

(A) Clock1a:Bmal1a combination, (B) Clock1a:Bmal1b combination and (C) Clock1a:Bmal2 combination. (D) All possible Clock:Bmal combinations. Each value is the mean ± SD of the three independent experiments. Results were analyzed by ANOVA. One star on the line indicates 0.01 < p < 0.05, two stars on the line p < 0.01, and three stars on the line p < 0.001. (E) Sub-cellular localizations of zebrafish Cry1ab, Cry2 and Cry3. Cry1ab-EGFP, Cry2-EGFP or Cry3-EGFP was transfected into HEK293 cells, respectively, and then the cells were stained with Hoechst and observed by an inverted fluorescent microscopy. The cytoplasmic or nuclear distribution of zebrafish Cry1ab, Cry2 or Cry3 was detected by GFP signal (green); DNA was stained by Hoechst (blue), respectively. The leader lines and letters N, C indicate the location of nucleus and cytoplasm, respectively. (F) Quantitative analysis of the subcellular localization of each Cry proteins. In each experiment, 50–100 cells were evaluated for nuclear fluorescence (N > C, blue bars), nuclear-cytoplasmic fluorescence (N = C, purple bars), and cytoplasmic fluorescence (N < C, pink bars). The figure shows a representative of the three independent experiments.

Identification of a novel NLS (Nuclear Localization Signal) sequence of zebrafish Cry1bb.

(A) Schematic diagrams of the full length of zebrafish Cry1bb (1–638aa) and Cry1bb truncation constructs. The yellow bar indicates the DNA photolyase domain, the green bar indicates the FAD binding domain of Cry proteins and the red bar indicates the predicted NLS sequence (VRREQPGPSGAKHR, Ensembl Peptide ID: ENSDARP00000090995) according to an online program (www.moseslab.csb.utoronto.ca/NLStradamus)⁴⁶ (see also Supplementary Fig. S7). Cry1bbΔ1 contained 499amino acids, truncated at amino acid position number 499; Cry1bbΔ2 contained 536amino acids truncated at amino acid position number 536 just before the predicted NLS signal; Cry1bbΔ3 contained 546amino acids truncated at amino acid position number 546 in the middle of the predicted NLS signal; and Cry1bbΔ4 contained 555amino acids truncated at amino acid position number 555 just after the predicted NLS signal. (B) Sub-cellular localization of zebrafish Cry1bb and its four truncated fragments. Cry1bb-EGFP or Cry1bb-EGFP truncated fragments were transfected into HEK293 cells, respectively, and then the cells were stained with Hoechst. The cytoplasmic or nuclear distributions of zebrafish Cry1bb and its four truncated fragments were detected by GFP signals (green); the nuclei were identified by Hoechst (blue). The leader lines and letters N, C indicate nuclear and cytoplasmic localizations, respectively. (C) Quantitative analysis of the subcellular localization of each Cry proteins. In each experiment, 50–100 cells were evaluated for nuclear fluorescence (N > C, blue bars), nuclear-cytoplasmic fluorescence (N = C, purple bars), and cytoplasmic fluorescence (N < C, pink bars). The figure shows a representative of the three independent experiments. (D) Repressive activities of zebrafish Cry1bb and its four truncated fragments determined by dual Luciferase reporter assays. The full length of Cry1bb and its four truncated fragments were co-transfected with per2-luc reporter and Bmal1b:Clock1a combination, respectively. Among the four Cry1bb truncated fragments, only Cry1bbΔ4 retained the repressive ability of the full length Cry1bb on Bmal1b:Clock1a-mediated transcription. A Renilla Luciferase was added in each transfection to normalize transfection efficiency. The figure shows the mean and SD (error bar) of two independent experiments (triplicate for each experiment). Results were analyzed by ANOVA.

Mechanisms for non-repression of zebrafish Cry2 and Cry3 on Clock:Bmal mediated transcription.

(A) Co-immunoprecipitation (Co-IP) experiments. Zebrafish cry1ab and cry2 constructs were co-transfected into HEK293 cells with per2-Luc reporter and bmal1a, and clock1a, respectively. Then the anti-Bmal1a polyclonal antibody was used to pull down proteins bound to Bmal1a, which was identified by the anti-Cry1ab and anti-Cry2 polyclonal antibodies, respectively. (B) The relative protein binding affinity of Cry1ab and Cry2 to Bmal1a-Clock1a complex. The relative protein binding affinity was quantified with Image pro-Plus6.0. Each value is the mean ± SD of three independent experiments. Results were analyzed by Student′s t-test. Horizontal lines indicate columns compared for significance. One star on the line shows 0.01 < p < 0.05. (C) Protein binding affinity comparison of Cry1ab and Cry2 to Bmal1a:Clock1a complex. Zebrafish cry1ab and cry2 constructs were co-transfected into HEK293 cells with per2-luc reporter and Bmal1a-Clock1a combination, respectively. Then anti-Cry1ab or anti-Cry2 polyclonal antibody was used to collect the proteins bound to Bmal1a, which was identified by the anti-Bmal1a polyclonal antibody. (D) The relative protein binding affinity of Cry1ab and Cry2 to the Clock1a-Bmal1a complex from Fig 8C. The relative protein binding affinity was quantified by Image pro-Plus6.0. Each value is the mean ± SD of three independent experiments. Results were analyzed by Student′s t-test. Horizontal lines indicate columns compared for significance. Two stars on the line means p < 0.01, (E) Sub-cellular localization of zebrafish Cry3 fused to either Cry1bb NLS from Fig 7A or a SV40 NLS. The Cytoplasmic-nuclear distribution of these two DNA constructs was detected by GFP signals (green) and the nucleus identified by Hoechst (blue). The leader lines and letters N, C indicate the location of nucleus and cytoplasm, respectively. (F) Inhibitory activities of zebrafish Cry3 fused to either Cry1bb NLS or a SV40 NLS determined by dual Luciferase reporter assays. The two DNA constructs were co-transfected with per2-Luc reporter and a Clock1a-Bmal1b combination, respectively. A Renilla Luciferase was added in each transfection to normalize transfection efficiency. The figure shows the mean and SD (error bar) of the two independent experiments (triplicate for each experiment). Results were analyzed by ANOVA.

Dorsal views of expression patterns of zebrafish cry1aa, cry1ab, cry1ba, cry1bb, cry2 and cry3 shown by in situ hybridization.