HeLa and NIH 3T3 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (Invitrogen, Carlsbad, CA, USA) at 37℃ with 9% CO₂. For transient transfection, cells were seeded at 10⁵ cells/well in 12-well plates. After overnight incubation, expression constructs were transfected using Lipofectamin Plus reagent (Invitrogen) according to the manufacturer's protocol. For protein stability and ubiquitination studies, cells were treated with cycloheximide (CHX) at 30 µg/mL and MG132 at 25 µM for indicated time periods.

Hes6 expression constructs were kindly provided by Dr. Sun (Korea University, Seoul, Korea). The point mutants were generated using the QuickChange site-directed mutagenesis as directed by the manufacturer's protocol (Stratagene, La Jolla, CA, USA). The polymerase chain reaction (PCR) primers used for the site-directed mutagenesis were as follows: for the K27R mutant, 5'-GCACGGGGGGACCGCA GGGCCCGGAAG CCCC-3'; for the K30R mutant, 5'-GGGACCGCAAGGCCCGGAGG CCCCTGGTGGAG-3'; for the K35R mutant, 5'-GCCCCTGGTGGAGAGGAAGCGACG CGCACGG-3'; for the K36R mutant, 5'-GCCCCTGGTGGAGAAGAGGCGACGCGCAC GG-3'; for the K60R mutant, 5'-GGTACCGAGGTGCAGGCCAGGCTAGAGAACGCCG-3'; for the 2KR (K27/30R) double mutant, 5'-CGGGGGGACCGCAGGG CCCGGAGGC CCCTGGTGG-3'. The mutagenesis was confirmed by DNA sequencing.

Cells were harvested in radioimmunoprecipitation assay buffer (50 mmol/L 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid [pH 7.4], 150 mmol/L NaCl, 1% NP-40, 1 mmol/L ethylenediaminetetraacetic acid, 1 mmol/L ethylene glycol tetraacetic acid, 1 mmol/L phenylmethanesulfonylfluoride, 0.5% sodium deoxycholate, 1 mmol/L NaF, 1 mmol/L Na3VO4 and 1×protease inhibitor cocktail [Sigma, St. Louis, MO, USA]) and centrifuged for 20 minutes at 4℃. Equal amounts of total protein were incubated with 2 µg anti-Hes6 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) or anti-Hes1 (AbFrontier, Seoul, Korea) antibodies for 1.5 hours at 4℃ and then the samples were incubated with the protein G-Sepharose bead slurry for 1 hour at 4℃. The final immune complexes were analyzed by Western blot.

For Western blot analyses, samples were resolved on 10% or 12% SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes (Immobilon P, Millipore, Bedford, MA, USA). Target proteins were detected with anti-Hes6 (anti-Hes6 antiserum was kindly provided by Dr. Sun), anti-SUMO1, anti-SUMO3 (Invitrogen), anti-Hes1 (Abfrontier), anti-HDAC3, anti-HA, or anti-actin (Santa Cruz Biotechnology) antibodies. Horseradish peroxidase-conjugated secondary antibodies were purchased from Jackson Immunoresearch Laboratories (West Grove, PA, USA) and the immune complexes were visualized with an enhanced chemiluminescent detection kit (Pierce, Rockford, IL, USA).

Chromatin immunoprecipitation (ChIP) assays were performed according to the manufacturer's instructions (Millipore). Each cross-linked sample was immunoprecipitated using anti-Hes1 (Abfrontier), anti-Hes6 (antiserum), anti-HDAC3 (Santa Cruz Biotechnology), or anti-histone H3 (acetyl K9, Abcam, Cambridge, MA, USA) antibodies and the DNA was purified by phenol/chloroform extraction followed by ethanol precipitation. For PCR, the following primers were used for the Hes1 N-box, 5'-TCCTTTTGATTGACGTTGTAGC-3' and 5'-GCACTATTCCAGGACCAAGG-3'.

To examine whether Hes6 is a substrate of sumoylation, Hes6 was co-transfected with RFP-fused SUMO paralogues (SUMO1 or SUMO3) into HeLa cells. In the presence of SUMO1 or SUMO3, sumoylation of Hes6 was observed and was enhanced by the E2-conjugating enzyme, Ubc9 (Fig. 1A). However, the Hes6 sumoylation band was not detected when co-expressed with ΔSUMO, a sumoylation defective mutant lacking the C-terminal motif (data not shown). To confirm whether this band was indeed a sumoylated form, Hes6 was co-expressed with RFP-SUMO paralogues and Ubc9 and then immunoprecipitated using anti-Hes6 antibodies. A slow migrating band of Hes6 was clearly detected by anti-Hes6 antibodies and SUMO1- or SUMO3-modified bands were detected using specific antibodies (Fig. 1B). Thus, these data suggest that Hes6 is covalently modified by either SUMO1 or SUMO3.

Next, to determine which lysine residues of Hes6 are targets of sumoylation, we next examined which lysine residues of Hes6 are targets for sumoylation. Within the bHLH domain of Hes6, there were five lysine residues positioned at K27, K30, K35, K36, and K60. We generated point mutants of Hes6 with arginine substituted for lysine residues. As shown in Fig. 2A, sumoylation of K27R or K30R single mutant was reduced in comparison with wild type Hes6, while the other mutants (K35R, K36R, or K60R mutants) did not show any decrease in sumoylation. Moreover, Hes6 mutations at both K27R and K30R (2KR mutant) completely abolished sumoylation (Fig. 2B). These data indicate that both lysine 27 and 30 of Hes6 are the target sites for sumoylation.

Different kinds of posttranslational modifications including sumoylation and ubiquitination are closely related to each other [22]. For instance, sumoylation blocks ubiquitin-dependent protein degradation of IκBα [24]. On the other hand, sumoylation of BMAL1 promotes its ubiquitination-mediated proteasomal degradation [25]. Because Hes6 also is degraded in an ubiquitin-dependent manner [19], we tested the relationship between Hes6 sumoylation and ubiquitination. First, we examined the protein stability of wild type or 2KR mutant Hes6. HeLa cells were transfected with either wild type Hes6 or the 2KR mutant and treated with CHX, an inhibitor of protein synthesis, for the indicated time periods. As shown in Fig. 3A, protein stability of wild type Hes6 was rapidly diminished, while the 2KR mutant showed much higher stability. We then examined the effect of sumoylation on ubiquitin-dependent Hes6 degradation. Wild type or the 2KR mutant Hes6 was co-transfected with HA-tagged ubiquitin in the presence or absence of RFP-SUMO. By Western blot, moderate ubiquitination of wild type Hes6 was detected by co-expression of HA-ubiquitin and further enhanced by co-expression of SUMO. Ubiquitination of wild type Hes6 was significantly decreased by SUMO protease SUSP1 (Fig. 3B). However, co-expression of either HA-ubiquitin alone or HA-ubiquitin together with SUMO failed to induce strong ubiquitination of the 2KR Hes6 mutant (Fig. 3B). Thus, these results demonstrate that sumoylation of Hes6 appears to accelerate its own ubiquitination leading to degradation via 26S proteasome pathway.

Previous studies demonstrate that gene expression of Hes family members such as Hes1 and Hes7 oscillate with a period of about 2 hours. This rhythmic expression plays an important role in the regulation of timely events in development processes [151617]. In circadian regulation, it has been reported previously that sumoylation of BMAL1 is critical for the oscillation of downstream clock genes as well as BMAL1 itself [2526]. Therefore, we examined whether sumoylation of Hes6 is involved in its rhythmic expression. We generated expression constructs of GFP-fused wild type or 2KR mutant Hes6 under the control of Hes6 promoter (Fig. 4A). These constructs were transfected into NIH 3T3 cells and the cells were incubated with serum to induce rhythmic expression of the Hes6 promoter. As shown in Fig. 4, 2-hour rhythmic expression of wild type Hes6 was induced by 50% horse serum shock (Fig. 4B, upper panel of Western blot). However, serum treatment failed to induce oscillation in the 2KR mutant Hes6 (Fig. 4B, lower panel of Western blot). Quantitation of band intensity of either wild type or the 2KR mutant Hes6 are shown in the bottom of Fig. 4B.

Hes6 is a positive regulator that derepresses Hes1 activity and thereby activates transcription of downstream genes [8]. We hypothesized that sumoylation might be important for the effect of Hes6 on Hes1-dependent transcriptional repression. As expected, Hes1-mediated transcriptional repression was relieved by Hes6 in a dose-dependent manner (Fig. 5A). Moreover, as shown in Fig. 5B, wild type Hes6 almost completely derepressed Hes1-induced transcriptional attenuation, while the 2KR mutant Hes6 exhibited partial derepression (Fig. 5B). To elucidate the molecular mechanism underlying the effect of Hes6 sumoylation on Hes1-mediated transcriptional repression, we examined interactions between Hes1 and Hes6 or Hes1 and DNA by immunoprecipitation or ChIP assay, respectively. Unexpectedly, Hes1 exhibited increased interaction with the 2KR mutant compared to wild type Hes6 (Fig. 5C). Moreover, in ChIP assays using either Hes1 or Hes6 antibodies, the Hes1/2KR Hes6 complex exhibited enhanced N-box binding than the Hes1/wild type Hes6 (Fig. 5D). Furthermore, histone H3 acetylation was decreased by co-expression of the 2KR mutant Hes6. HDAC recruitment to the N-box region was significantly induced by the 2KR mutant Hes6 (Fig. 5C, D). Taken together, these results demonstrate that sumoylation of Hes6 may play an important role in controlling Hes1 transcriptional activity by regulating interaction between Hes6 and Hes1, N-box binding of the Hes1/Hes6 complex, and recruitment of co-regulators to the promoter.