Serine/arginine protein-specific kinase 2 promotes leukemia cell proliferation by phosphorylating acinus and regulating cyclin A1

Abstract

Serine/arginine (SR) protein-specific kinase (SRPK), a family of cell cycle-regulated protein kinases, phosphorylate SR domain-containing proteins in nuclear speckles and mediate the pre-mRNA splicing. However, the physiologic roles of this event in cell cycle are incompletely understood. Here, we show that SRPK2 binds and phosphorylates acinus, an SR protein essential for RNA splicing, and redistributes it from the nuclear speckles to the nucleoplasm, resulting in cyclin A1 but not A2 up-regulation. Acinus S422D, an SRPK2 phosphorylation mimetic, enhances cyclin A1 transcription, whereas acinus S422A, an unphosphorylatable mutant, blocks the stimulatory effect of SRPK2. Ablation of acinus or SRPK2 abrogates cyclin A1 expression in leukemia cells and arrest cells at G(1) phase. Overexpression of acinus or SRPK2 increases leukemia cell proliferation. Furthermore, both SRPK2 and acinus are overexpressed in some human acute myelogenous leukemia patients and correlate with elevated cyclin A1 expression levels, fitting with the oncogenic activity of cyclin A1 in leukemia. Thus, our findings establish a molecular mechanism by which SR splicing machinery regulates cell cycle and contributes to leukemia tumorigenesis.

Figure 1. Acinus binds SRPK2

(A). Yeast-two-hybrid screen searching for the binding targets of the CTD of acinus. (B). The CTD of acinus associates with the middle spacer in SRPK2. Various GST-tagged acinus constructs were cotransfected with SRPK2 into HEK293 cells. Transfected acinus proteins were pulled down with glutathione beads. The C-terminal end of but not the N-terminal domain of acinus associates with SRPK2 (top left panel). GST-tagged SRPK2 fragments were cotransfected into HEK293 cells with Flag-acinus. SRPK2 proteins were pulled down with glutathione beads. The middle region of SRPK2 from 308 to 383 interacted with acinus (top middle panel). The expression of the transfected constructs was confirmed (middle and bottom middle panels). SRPK1 does not bind to acinus (right panels). (C). Endogenous acinus binds to SRPK2 in mouse brain. Acinus coimmunoprecipitated with SRPK2 regardless of acinus or SRPK2 antibody employed. (D). PI 3-kinase signaling mediates the association between SRPK2 and acinus. K562 cells were pretreated with various pharmacological inhibitors (20 nM Wortmannin, 10 μM LY294002, 10 μM PD98059) for 30 min, followed by 50 ng/ml EGF for 10 min. Endogenous acinus was immunoprecipitated with anti-acinus antibody. PI 3-kinase inhibitors but not MEK1 pretreatment abolished SRPK2 binding to acinus (top panel). Acinus S422 and Akt S473 phosphorylation were verified (2^nd and 3^rd panels). Equal amount of acinus was immunoprecipitated (bottom panel).

Figure 2. SRPK2 phosphorylates acinus on Serine 422

(A). Diagram of Acinus-S. Acinus-S possesses three RS motifs as indicated. The three fragments with each containing the RS dipeptide repeat motif are indicated with residue numbers. (B). In vitro SRPK2 kinase assay. Purified recombinant GST-fusion proteins were incubated with purified His-SRPK2 at 30 °C for 30 min. Both fragments B and C were robustly phosphorylated, while fragment A was not (left panel). S422 residue in acinus-S was phosphorylated by SRPK2. Purified GST-acinus proteins were incubated with purified SRPK2 in the presence of [γ-³²P]-ATP. S422A mutant substantially decreased acinus phosphorylation (middle and right panels). (C). Wild-type but not kinase-dead SRPK2 phosphorylates acinus. GST-acinus wt and S422A were transfected into HEK293 cells in the presence or absence of SRPK2. Transfected acinus was pulled down with glutathione beads. While S422 site was markedly phosphorylated in wild-type acinus, no phosphorylation was detected in S422A mutant (top left panel). The expression of transfected constructs was verified (2^nd to bottom left panels). Flag-acinus and Myc-SRPK2 wt or KD were transfected into HEK293 cells. Acinus was immunoprecipitated with anti-Flag antibody and probed with anti-phospho-S422 antibody. Wild-type SRPK2 potently phosphorylated acinus, whereas SRPK2-KD failed (top right panel). The expression of transfected constructs was verified (2^nd to bottom right panels). (D). Acinus-S can be phosphorylated in intact cells. HEK293 cells were transfected with si-RNA for SRPK2 or Akt, and followed by serum starvation overnight. In control samples, serum triggered potent S422 phosphorylation. Knocking down of SRPK2 or Akt blocked acinus S422 phosphorylation (top panel). The depletion of SRPK2 and Akt was confirmed (2^nd and 3^rd panels).

Figure 3. SRPK2 but not Akt redistributes acinus in the nucleus

(A). Acinus phosphorylation mimetic mutant S422D redistributes in the nucleus. Wild-type acinus and S422A mutants resided in the nuclear speckles, whereas S422D occurred in the whole nucleoplasm. Wild-type SRPK2 mainly localized in the cytoplasm, and a portion of it also distributed in the nucleus. SRPK2-KD exclusively localized in the cytoplasm. SRPK2 phosphorylation triggers acinus relocation from the nuclear speckle to the nucleoplasm. Wild-type acinus redistributed in the nucleoplasm when cotransfected with wild-type SRPK2, and it localized in the nuclear speckles when cotransfected with SRPK2-KD. S422D resided in the nucleoplasm regardless of SRPK2 wild-type or KD. (B). Akt can not relocate acinus from the nuclear speckles. All Akt proteins (wild-type, CA and KD) and acinus-S colocalized in the nuclear speckles of transfected cells. (C). S422A exhibited lower affinity to SRPK2. Myc-SRPK2 was cotransfected into HEK293 cells with various GST-tagged acinus constructs. Transfected acinus proteins were pulled down with glutathione beads and probed with anti-myc antibody. S422A mutants exhibited decreased binding activity to SRPK2 (top panel). The expression of transfected constructs was confirmed (middle and bottom panels). (D). Akt enhances the interaction between SRPK2 and acinus, when SRPK2 kinase activity is low. Acinus and SRPK2 were cotransfected into HEK293 cells, followed by knocking down of Akt with si-RNA. Wild-type and SRPK2-KD displayed the similar affinity to wild-type acinus and lower binding activity to S422A. Depletion of Akt slightly decreased the affinity of wild-type SRPK2 to acinus, while SRPK2 KD binding to acinus wild-type was evidently reduced and completely abolished to acinus S422A (top panel). The expression of transfected constructs and Akt protein level were confirmed (2^nd to bottom panels).

Figure 4. SRPK2 is required for cyclin A1 expression

(A). SRPK2 overexpression upregulates cyclin A1 expression. HeLa cells and K562 cells were transfected with SRPK2 wild-type and KD. The expression of various cell-cycle and apoptosis-related proteins was monitored by immunoblotting. Cyclin A1 but not A2 or B1 was selectively increased in SRPK2 wild-type cells, and the stimulatory effect was lost in KD sample (2^nd, 3^rd, and 4^th panels). Interestingly, cyclin D1 was also weakly enhanced in SRPK2 overexpressed cells (5^th panel). CDK4 and DFF/ICAD expression levels were not affected by SRPK2 (6^th and 7^th panels). (B). SRPK2 regulates cyclin A1 promoter activity. Different amounts of SRPK2 wild-type and KD were coexpressed with a cyclin A1 promoter construct (335-bp fragment). Empty vector was used to match the same total amount of DNA in all experiments. The mean and standard error for 3 independent experiments are shown. SRPK2 mediated cyclin A1 promoter activation in a dose-dependent manner, and KD lost its activity. SRPK2 is required for cyclin A1 promoter activation. Endogenous SRPK2 was depleted from HEK293 cells, transfected with cyclin A1 promoter construct. Ablation of SRPK2 decreased cyclin A1promoter luciferase activity. Values are means (± SD) of three independent experiments. (C). Acinus mediates SRPK2's activity on cyclin A1 expression. Various acinus constructs and siRNA of acinus were cotransfected with SRPK2 wild-type into HEK293 cells. Depletion of acinus blocked SRPK2's activity. Unphosphorylatable mutant S422A decreased SRPK2's effect, whereas S422D substantially increased SRPK2's activity. Values are means (± SD) of three independent experiments. (D). SRPK2 controls cyclin A1 expression in human leukemia cells. SRPK2 siRNA and control RNAi were transfected into HeLa, HL-60 and K562 cells. The total RNA was extracted and RT-PCR was conducted. Ablation of SRPK2 abolished cyclin A1 but not cyclin A2 expression (top and 2^nd panels). The cell lysates were analyzed with immunoblotting with anti-SRPK2, cyclin A1 and cyclinA2 antibodies, respectively. Depletion of SRPK2 substantially attenuated cyclin A1 but not cyclin A2 expression.

Figure 5. Acinus phosphorylation by SRPK2 regulates its effect on cyclin A1 expression

(A). Acinus mediates cyclin A1 promoter activity, which is regulated by S422 phosphorylation. Different amounts of acinus wild-type and phosphorylation mutants were coexpressed with a cyclin A1 promoter construct (335-bp fragment). Empty vector was used to match the same total amount of DNA in all experiments. Values are means (± SD) of three independent experiments. Acinus mediated cyclin A1 promoter activation in a dose-dependent manner. Acinus S422A lost its activity, whereas S422D strongly elevated cyclin A1 promoter activity. (B). Acinus is required for cyclin A1 expression. Endogenous acinus was knocked down from HEK293 cells, transfected with cyclin A1 promoter construct. Depletion of acinus reduced cyclin A1 promoter activity. Values are means (± SD) of three independent experiments. (C). SRPK2 plays a more important role in activating acinus's stimulatory activity. Constitutively active Akt-CA overexpression evidently increased cyclin A1 promoter activity (lane 4), but the effect was not as much as acinus overexpression (lane 5). Coexpression of Akt with acinus slightly increased acinus's activity (lanes 6 and 7), which was almost completely abrogated in SRPK2-depleted samples (lanes 9 and 10). Akt-KD almost had no effect on acinus activity (lane 8). Values are means (± SD) of three independent experiments. (D). Acinus controls cyclin A1 expression in human leukemia cells. Acinus siRNA and control RNAi were transfected into HeLa, HL-60 and K562 cells. The RT-PCR was conducted. Knocking down of acinus abolished cyclin A1 but not cyclin A2 expression (top and 2^nd panels). The cell lysates were analyzed with immunoblotting with anti-acinus, cyclin A1 and cyclinA2 antibodies, respectively. Depletion of SRPK2 prominently decreased cyclin A1 but not cyclin A2 expression.

Figure 6. SRPK2 and acinus regulates cell cycle profile and leukemia cell proliferation

(A). Ablation of acinus or SRPK2 attenuates cyclin A1 expression. Immunoblotting analysis of K562 cells transfected with siRNA of acinus or SRPK2. (B). Depletion of SRPK2 or acinus strongly decreases K562 cell proliferation. K562 cells were treated with a control RNAi, acinus RNAi, SRPK2 RNAi or Akt1 RNAi. The cells were stained with crystal violet 3 days after siRNA treatment. (C). Depletion of SRPK2 or acinus decreased Brdu incorporation. K562 cells treated with a control RNAi, acinus RNAi, SRPK2 RNAi or Akt1 RNAi. The cells were labeled with Brdu and stained 1 day after RNAi treatment. (D). Inactivation of acinus or SRPK2 induces G1 arrest in K562 cells. K562 cells were treated with various siRNA and the cell cycle profiles were analyzed with FACS.

Figure 7