Phosphorylation of islet-1 serine 269 by CDK1 increases its transcriptional activity and promotes cell proliferation in gastric cancer

Abstract

Background: Despite recent advances in diagnostic and therapeutic approaches for gastric cancer (GC), the survival of patients with advanced GC remains very low. Islet-1 (ISL1) is a LIM-homeodomain transcription factor, which is upregulated and promotes cell proliferation in GC. The exact mechanism by which ISL1 influences GC development is unclear.

Methods: Co-immunoprecipitation (co-IP) and glutathione S-transferase (GST)-pulldown assays were employed to evaluate the interaction of ISL1 with CDK1. Western blot and immunohistochemistry analyses were performed to evaluate the ability of CDK1 to phosphorylate ISL1 at Ser 269 in GC cell and tissue specimens. Chromatin immunoprecipitation (ChIP), ChIP re-IP, luciferase reporter, and CCK-8 assays were combined with flow cytometry cell cycle analysis to detect the transactivation potency of ISL1-S269-p and its ability to promote cell proliferation. The self-stability and interaction with CDK1 of ISL1-S269-p were also determined.

Results: ISL1 is phosphorylated by CDK1 at serine 269 (S269) in vivo. Phosphorylation of ISL1 by CDK1 on serine 269 strengthened its binding on the cyclin B1 and cyclin B2 promoters and increased its transcriptional activity in GC. Furthermore, CDK1-dependent phosphorylation of ISL1 correlated positively with ISL1 protein self-stability in NIH3T3 cells.

Conclusions: ISL1-S269-p increased ISL1 transcriptional activity and self-stability while binding to the cyclinB1 and cyclinB2 promoters promotes cell proliferation. ISL1-S269-p is therefore crucial for tumorigenesis and potentially a direct therapeutic target for GC.

Keywords: CDK1; Gastric cancer; ISL1; Phosphorylation; Transcription activity.

Fig. 1

ISL1 interacts with CDK1. a–c Co-IP assays were performed to measure the ISL1/CDK1/cyclinB1 complex in MGC803 cells. Lysate was immunoprecipitated with antibodies against ISL1, CDK1, or cyclinB1 and analyzed by immunoblotting. IgG: immunoglobulin G. d Full-length or truncated ISL1 was used to construct GST-fusion proteins (Top) for pull-down assays with CDK1 protein (Bottom)

Fig. 2

ISL1 is serine-phosphorylated. a MGC803 cells lysates were immunoprecipitated (IP) using anti‑ISL1 and analyzed by immunoblotting using antibody against phosphorylated serine and ISL1 (HC: heavy chain, LC: light chain). b Alignments of the cluster of ISL1; S269 fit the CDK1 phosphorylation consensus, and (S/T)PX(R/K) sequences (red shade) are shown

Fig. 3

CDK1 mediates ISL1 phosphorylation at serine 269. a MGC803 cells transfected with plasmid encoding CDK1 or control vector (left), negative control or CDK1 si-RNA (right). Lysate was analyzed by immunoblotting with antibodies against the indicated proteins. b MGC803 cells were transfected with plasmids encoding dominant‑negative mutant CDK1 (DN‑CDK1) or CDK1, and lysate was analyzed by immunoblotting with antibodies against the indicated proteins. c MKN28 cells expressing wild-type ISL1 or ISL1-S269A were transfected with plasmid encoding CDK1 or control vector. Cell lysates were immunoblotted with antibodies against the indicated proteins. d MKN28 cells were transfected with plasmids encoding wild-type ISL1 or ISL1-S269A and treated with CDK1 inhibitor RO-3306 or DMSO. Lysates were analyzed by immunoblotting with antibodies against the indicated proteins. GAPDH served as the internal control. e Representative IHC staining of ISL1-S269-p expression in GC (right, n = 60) and paired adjacent tissues (left, n = 60). f Relationship between expression of ISL1-S629-p and overall survival in gastric cancer

Fig. 4

Phosphorylated ISL1 showed improved localization on cyclinB1 and cyclinB2 promoters. a ChIP assay analysis and PCR gel electrophoresis analysis of ISL-1 recruitment to the cyclinB1 and cyclinB2 promoters in MGC803 cells. b ChIP-re-IP assay was performed with anti-ISL-1 or rabbit IgG antibodies and then with anti-CDK1 or IgG antibodies for immunoprecipitation using chromatin harvested from MGC803 cells. c, d Quantitative chromatin immunoprecipitation analysis of cyclinB1 and cyclinB2 promoters in MKN28 cell line transfected with control vector or expressing wild-type ISL1 or ISL1-S269A plasmids. Data are expressed as mean ± s. d. from three individual experiments. (*p < 0.05, WT vs. control, ^#p < 0.05, S269A vs. control, ^$p < 0.05, WT vs. S269A)

Fig. 5

Phosphorylation Ser 269 of ISL1 increases transactivation potency. a, b Effect of CDK1 activity on the transcriptional activity of wild-type ISL1 or S269A ISL1. Transcriptional activity of wild-type ISL1 and S269A ISL1 on cyclinB1-luc and cyclinB2-luc determined by luciferase reporter assay in MKN28 cells. c Cell proliferation was studied by CCK-8 analysis of MKN28 cells transfected with plasmids encoding control vector and wild-type ISL1 or ISL1-S269A. d Cell cycle distributions were analyzed by flow cytometry in MKN28 transfected with plasmids encoding control vector and wild-type ISL1 or ISL1-S269A. The data represent 3 independent experiments, each performed in triplicate. Each bar represents mean ± SD. (*p < 0.05, WT vs. control, ^#p < 0.05, S269A vs. control, ^$p < 0.05, WT vs. S269A)

Fig. 6

Phosphorylation of ISL1 increases self-stability but does not increase its interaction with CDK1. a NIH3T3 cells were transfected with wild-type ISL1 or ISL1-S269A plasmids. After 48 h, cells were treated with CHX (100 µg/ml) for 12 h. Whole cell extracts were harvested for western blotting. b Co-IP assays were performed in NIH3T3 cells to detect the interaction between ISL1 (wild-type or mutants) and CDK1. GAPDH served as a loading and negative control. c Schematic representation of positive feedback regulation by the ISL1-CDK1/cyclin B1/2 complex in GC. Full arrow: confirmed; broken arrow: to be verified