Cyclin I-like (CCNI2) is a cyclin-dependent kinase 5 (CDK5) activator and is involved in cell cycle regulation
In contrast to conventional cyclin-dependent kinases that are important for mitotic cell division, cyclin-dependent kinase 5 (CDK5) is predominantly activated in post-mitotic cells and is involved in various cellular events. The kinase activity of CDK5 is tightly regulated by specific activators including p35, p39, and cyclin I (CCNI). Here we show that cyclin I-like (CCNI2), a homolog of CCNI, interacts with CDK5 and activates the kinase activity of CDK5. Different from CCNI, which colocalizes with CDK5 in the nuclei in transfected cells, CCNI2 mainly retains CDK5 in the cytoplasm as well as on the cell membrane. Furthermore, although the expression level of CCNI2 mRNA and CCNI2 protein do not change significantly during cell cycle, depletion of CCNI2 with siRNA affects cell cycle progression as well as cell proliferation. In conclusion, our data strongly suggest that CCNI2 is a novel CDK5 activator and is involved in cell cycle regulation.In an effort to identify new CDK5 binding-proteins, we performed yeast two-hybrid screening of a chicken cochlear cDNA library using CDK5 as bait. The identified positive clones encode two proteins, cyclin I (CCNI, GenBank accession number XP_420590) and cyclin I-like (CCNI2, GenBank accession number XP_001234830) (Table 1). CCNI was originally identified based on its similarity to other known cyclins18, and has been shown to bind and activate CDK517. CCNI2 is considered as a homolog of CCNI, although the overall similarity between these two proteins is relatively low except for the cyclin box (Fig. 1A,B). At present, the physiological function of CCNI2 is largely unknown, and the interaction between CCNI2 and CDKs has not been reported.Sequence analysis revealed that CCNI2 is not conserved during evolution. CCNI2 gene is present in human, chicken and zebrafish genomes, but not in mouse and rat genomes. The expression pattern of CCNI2 protein in different chicken tissues was examined by western blot with a polyclonal CCNI2 antibody. The results showed that chicken CCNI2 is highly expressed in neural tissues including cerebrum and cerebellum, and weakly expressed in kidney and cochlea, whereas undetected in liver and heart (Fig. 1C).We then performed co-immunoprecipitation (co-IP) experiment to confirm the interaction between CCNI2 and CDK5. CCNI was also included in this experiment for comparison. The results revealed that Flag-tagged human CDK5 is co-immunoprecipitated with Myc-tagged human CCNI or CCNI2, but not cyclin A1 (CCNA1) (Fig. 2A). Noticeably, our results indicated that the interaction between CDK5 and CCNI2 is stronger than that between CDK5 and CCNI. The interaction was also confirmed by Co-IP when chicken CDK5, CCNI, and CCNI2 were used (Fig. 2B).As an important kinase, CDK5 requires other proteins to activate its kinase activity. Several CDK5 activators have been identified, including p35, p39, and CCNI6,7,11,17. Here we demonstrate that CCNI2 functions as a novel CDK5 activator. CCNI2 physically interacts with CDK5 and activates its kinase activity. Moreover, compared to the previously identified CDK5 activator CCNI, CCNI2 exhibits higher binding affinity with CDK5 and is more potent to activate CDK5.Different from other known CDK5 activators, CCNI2 is not conserved during evolution. CCNI2 gene and CCNI2 protein are missing in many species, including mouse and rat. Although CCNI2 is present in human and chicken, the N-terminus and C-terminus differ significantly among these species. Human CCNI2 has a unique long N-terminal fragment, which is not present in chicken CCNI2 and in CCNI. In addition, unlike chicken CCNI2 and CCNI, human CCNI2 lacks the C-terminal PEST region. Nevertheless, both human and chicken CCNI2 share a conserved cyclin box, which is responsible for binding and activating CDKs. Indeed, our data clearly show that both human and chicken CCNI2 bind and activate CDK5.All animal experiments were approved by the Ethics Committee of Shandong University and were performed in accordance with the relevant guidelines and regulations. The coding sequences of human CDK5 and CCNI, chicken CDK5, CCNI, and CCNI2 were amplified by PCR and cloned into expression vectors pEGFP-C2 or modified pEGFP-C2 (EGFP replaced with Myc or Flag). Human CCNI2 cDNA was obtained from Cyagen Biosciences Inc. Human CCNA1 cDNA was obtained from ViGene Biosciences Inc. Rabbit polyclonal anti-CDK5 antibody (Cat. No. sc-173, 1:1000 diluted) was from Santa Cruz, and rabbit polyclonal anti-CCNI2 antibody (Cat. No. 18822-1-AP, 1:1500 diluted) was from Proteintech. Rabbit monoclonal anti-phospho-CDK5 substrate antibody mix (Cat. No. 9477, 1:1000 diluted) was from Cell Signaling Technology. Rabbit polyclonal anti-ACTIN antibody (Cat. No. P30002, 1:5000 diluted) and mouse monoclonal anti-GFP antibody (Cat. No. M20004, 1:5000 diluted) were from Abmart. Mouse polyclonal anti-GAPDH antibody (Cat. No. MAB374, 1:5000 diluted) was from Millipore. Mouse monoclonal anti-Myc antibody (Cat. No. M4439, 1:5000 diluted) and mouse monoclonal anti-Flag antibody (Cat. No. F1804, 1:5000 diluted) were from Sigma-Aldrich.We thank Dr. Deli Shi for critical reading of the manuscript. This work was supported by grants from the National Basic Research Program of China (2013CB967700 to Z.X.), the National Natural Science Foundation of China (31371355 to Z.X. and 31401007 to Y.W.), the China Postdoctoral Science Foundation (2014M560550 to Y.W.), and the Fundamental Research Funds of Shandong University (2014HW011 to Y.W.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.Author Contributions Z.X. conceived and designed the experiments; C.L., X.Z., and B.Z. performed the experiments; C.L., Y.W., and Z.X. analyzed the data; C.L. and Z.X. wrote the paper.