LIN-9 phosphorylation on threonine-96 is required for transcriptional activation of LIN-9 target genes and promotes cell cycle progression

Abstract

Cell cycle transitions are governed by the timely expression of cyclins, the activating subunits of Cyclin-dependent kinases (Cdks), which are responsible for the inactivation of the pocket proteins. Overexpression of cyclins promotes cell proliferation and cancer. Therefore, it is important to understand the mechanisms by which cyclins regulate the expression of cell cycle promoting genes including subsequent cyclins. LIN-9 and the pocket proteins p107 and p130 are members of the DREAM complex that in G0 represses cell cycle genes. Interestingly, little is know about the regulation and function of LIN-9 after phosphorylation of p107,p130 by Cyclin D/Cdk4 disassembles the DREAM complex in early G1. In this report, we demonstrate that cyclin E1/Cdk3 phosphorylates LIN-9 on Thr-96. Mutating Thr-96 to alanine inhibits activation of cyclins A2 and B1 promoters, whereas a phosphomimetic Asp mutant strongly activates their promoters and triggers accelerated entry into G2/M phase in 293T cells. Taken together, our data suggest a novel role for cyclin E1 beyond G1/S and into S/G2 phase, most likely by inducing the expression of subsequent cyclins A2 and B1 through LIN-9.

Figure 1. LIN-9 is phosphorylated by and is found in a trimeric complex with Cdk3/cyclin E1.

(A) GST-LIN-9 expressed in bacteria was subjected to in vitro kinase assays using indicated kinase complexes purchased from ProQinase. Asterisk indicates GST-cyclin autophosphorylation. The lower panel shows Coomassie blue staining of the same gel. (B) GST-LIN-9 was subjected to in vitro kinase assays using Flag-Cdk3 kinase complexes (Flag-IP) from human 293Tcells. Autoradiography visualizes incorporation of phospholabel and asterisks (white, placed to the left of the corresponding band) indicate autophosphorylation of cyclin E1 and cyclin A2, respectively (upper panel). Coomassie Brillant Blue (CBB) staining depicts loading of GST-LIN-9 and Flag-Cdk3 (second panel). Ten percent of IP material was subjected to Western blot to visualize amounts of co-precipitated cyclin A2 (third panel), cyclin E1 (fourth panel) and Flag-Cdk3 (lower panel). (C) 293T cells were cotransfected with Flag-Cdk3 and GFP-LIN-9, followed by Flag-IP. IP material was subjected to Western blots using LIN-9 antibody (upper panel), cyclin E1 antibody (middle panel) and Flag-antibody (lower panel).

Figure 2. Thr-96 is the major phosphorylation site in LIN-9 for Cdk3/cyclin E1.

Indicated GST-LIN-9 deletion constructs were expressed in bacteria, purified, and subjected to in vitro kinase assays using Flag-IPs from 293T cells cotransfected with Flag-Cdk3 and untagged cyclin E1. Autoradiogram shows phosphorylation of recombinant LIN-9 proteins (upper panel). CBB staining of the same gel was used to visualize the amounts of GST-LIN-9 constructs in the reaction (lower panel) Asterisks indicate the respective GST-LIN-9 constructs. (B) Indicated GST-LIN-9 full-length mutants expressed in bacteria were subjected to in vitro kinase assays using Flag-Cdk3/cyclin E precipitated from 293T cells, transfected with FLAG-Cdk3 and cyclin E1, via Flag-antibody as a kinase. LIN-9 phosphorylation (upper panel), Flag-Cdk3 and cyclin E1 autophosphorylation (middle panel) were monitored by autoradiography. Loading of GST-LIN-9 mutants was determined by CBB staining of the same gel (lower panel). (C) GST-LIN-9-1-109 and its mutants (S76A and T96A) were subjected to kinase assays as in B. GST-LIN-9 phosphorylation was again detected by autoradiography (upper panel) and equal loading was confirmed by CBB staining of the same gel (lower panel). (D) Alignment of LIN-9 amino acid residues surrounding Thr-96 among species. Identical residues are depicted in red, strongly similar residues are depicted in green, and non-similar residues are depicted in blue. Arrow indicates position of Thr-96. Accession numbers of aligned amino acid sequences are: Homo sapiens: AAV41873.1; Mus musculus: NP_001096652.1; Xenopus laevis: NP_001121233.1 and Danio rerio: NP_001038411.1.

Figure 3. cyclin E is essential for phosphorylation of Thr-96 in LIN-9 in vivo in primary cells.

(A) p-LIN-9^Thr-96 antibody is specific. 293T cells were cotransfected with plasmids encoding GFP-LIN-9, FLAG-Cdk3 and cyclin E, and cell extracts were subjected to immunoprecipitation using a monoclonal LIN-9 antibody. IP material was split into two, and incubated for 1 hour at 30°C with or without lambda phosphatase. Samples were subjected to Western Blot using our p-LIN-9^Thr-96 antibody (upper panel). Equal loading of LIN-9 was confirmed by incubating the same membrane with a rabbit antibody against LIN-9 (lower panel). (B) HUVECs were transfected with siRNA control or siRNA directed against Cyclin E1 (Santa Cruz) using Dharmafect 4 (Dharmacon) and harvested after 48 hours. Western Blots was performed with the indicated antibodies. (C) T98G cells were starved in serum-free medium for 36 hours followed by the addition of growth medium (supplemented with 10% serum). Cells were harvested by tripsynization at the indicated time after the addition of serum and use for Western blot or FACS analysis. Western blots were performed as described in Methods.

Figure 4. Phosphorylation of Thr-96 in LIN-9 is essential for transcriptional activation of cell cycle genes.

(A) Indicated GFP-LIN-9 constructs were cotransfected into 293T cells, together with luciferase under the control of the promoters of cdc6, cyclin B1 or cyclin A2. Cell lysates were subjected to reporter gene assays using Promega Glo (Promega). Experiment was performed in triplicates. For the cdc6 promoter: n = 4, for the cyclin A and cyclin B promoters: n = 2; *p<0.05, **p<0.005, ***p<0.0005 (B) One of the three samples used in A was subjected to Western blot, using GFP antibody to detect expression of GFP-LIN-9 constructs. GFP-LIN-9^T96D migrates faster and is always expressed at lower levels than GST-LIN-9 wild type and T96A.

Figure 5. Phosphorylation of Thr-96 in LIN-9 accelerates entry into G2/M.

293T cells were transfected with the indicated LIN-9 constructs or GFP control and 24 hours later synchronized by serum starvation. Cells were released into the cell cycle by the addition of complete growth medium and harvested at 0, 6 and 12 hours for FACS analysis.