PKD phosphorylation and COP9/Signalosome modulate intracellular Spry2 protein stability

Abstract

Spry2 is a molecular modulator of tyrosine kinase receptor signaling pathways that has cancer-type-specific effects. Mammalian Spry2 protein undergoes tyrosine and serine phosphorylation in response to growth factor stimulation. Spry2 expression is distinctly altered in various cancer types. Inhibition of the proteasome functionality results in reduced intracellular Spry2 degradation. Using in vitro and in vivo assays, we show that protein kinase D (PKD) phosphorylates Spry2 at serine 112 and interacts in vivo with the C-terminal half of this protein. Importantly, missense mutation of Ser112 decreases the rate of Spry2 intracellular protein degradation. Either knocking down the expression of all three mammalian PKD isoforms or blocking their kinase activity with a specific inhibitor contributes to the stabilization of Spry2 wild-type protein. Downregulation of CSN3, a component of the COP9/Signalosome that binds PKD, significantly increases the half-life of Spry2 wild-type protein but does not affect the stability of a Spry2 after mutating Ser112 to the non-phosphorylatable residue alanine. Our data demonstrate that both PKD and the COP9/Signalosome play a significant role in control of Spry2 intracellular stability and support the consideration of the PKD/COP9 complex as a potential therapeutic target in tumors where Spry2 expression is reduced.

Fig. 1. PKD phosphorylates Spry2 at serine 112 in vitro and in vivo.

A PKD phosphorylation consensus site around serine 112 in Spry2 and Spry1, and serine 80 in Spry3. B PKD phosphorylates Spry2 in vitro. HEK293T cells were transfected with 5 µg of pECFL-AU5-Spry2 or AU5-vector, and with pEFBOS-GFP-PKD1-KD (GFP-PKD-KD) or pEFBOS-GFP-PKD1-CA (GFP-PKD-CA) for 48 h. Then, cells were serum-starved for 18 h, lysated and 200 µg of whole-cell lysates (WCL) were immunoprecipitated with anti-AU5 and anti-GFP antibodies. The expression levels of transfected proteins were detected by immunoblotting with specific antibodies after WCL were resolved by SDS-PAGE. Autoradiogram of in vitro kinase assays performed in the immunoprecipates using [γ³²P]-ATP showing phosphorylated AU5-Spry2 and autophosphorylated GFP-PKD-CA. C PKD phosphorylates Spry2 in vivo at serine 112. HEK293T cells were transfected with 5 µg of AU5-Spry2 WT or AU5-Spry2 S112A together with pEFBOS-GFP-PKD1-CA (GFP-PKD-CA). Cells were in vivo labeled using ³²P_i as described in Materials and Methods. AU5-immunoprecipates were resolved by SDS-PAGE. Radioactive bands in the autoradiogram correspond to AU5-Spry2 and autophosphorylated GFP-PKD-CA. Expression levels of transfected proteins were detected by immunoblotting with specific antibodies after WCL were resolved by SDS-PAGE.

Fig. 2. Spry2 associates with PKD through its C-terminal region.

A PKD and Spry2 co-localize in vivo. HeLa cells transiently co-transfected with pECFL-AU5-Spry2 WT and pcDNA3-PKD1 WT were used for immunofluorescence-confocal microscopy analysis. Spry2 is visualized in green (by anti-AU5 FITC), PKD1 in red (by anti-PKD1 with a far-red-emitting fluorophore conjugate) and nuclei are stained with DAPI in blue. Panels show different localization and intensity profiles of PKD1 (red) and Spry2 (green), along different straight lines (white) of a representative cell. B The C-terminal region of Spry2 mediates its association with PKD in vivo. Cell extracts from transiently transfected HEK293T cells overexpressing Spry2 WT, Spry2 Cter (truncated mutant containing residues from 177 to 315) or Spry2 Nter (residues from 1 to 177), with HA epitope tag together with GFP-PKD WT or GFP vector alone, were incubated with anti-GFP polyclonal rabbit antibody or with unspecific rabbit IgG. Anti-GFP immunoprecipitates or WCL were then analyzed by immunoblotting using anti-GFP and anti-HA antibodies as described in Materials and Methods.

Fig. 3. Spry2 S112A and S112E mutant proteins are more stable than Spry2 WT.

A SW480-ADH cells constitutively overexpressing Spry2 WT, Spry2 S112A, or Spry2 S112E with AU5 epitope were treated with cycloheximide (CHX) 10 µg/ml for different times up to 8 h. AU5-Spry2 and tubulin levels were assessed by immunoblotting with specific antibodies after resolving WCL by SDS-PAGE. The images shown are representative of three independent experiments. B Quantitation of AU5-Spry2 protein levels normalized to tubulin levels. The histograms represent the average and SD of three separate assays (^**p < 0.01, ^***p < 0.001).

Fig. 4. PKD negatively affects the stability of Spry2 WT but not of mutant Spry2 S112A.

A SW480-ADH cells constitutively overexpressing AU5-Spry2 WT or AU5-Spry2 S112A were transfected with 37.5 nM control siRNA scramble (siRNA control), or 5 nM siRNAs PKD1 and 20 nM siRNAs PKD2 and 12.5 nM siRNAs PKD3 for 48 h. Cells were then treated with CHX (10 µg/ml) for 0, 4, and 6 h. Levels of AU5-Spry2 WT, AU5-Spry2 S112A, PKD1, PKD2, PKD3, and tubulin were detected by immunoblotting with specific antibodies after WCL were resolved by SDS-PAGE. The data shown are representative of three separate experiments. B Quantitation of AU5-Spry2 protein levels normalized to tubulin levels. Histograms represent the average and SD of three separate analysis (ns not significance, ^**p < 0.01). C SW480-ADH cells constitutively overexpressing AU5-Spry2 WT were treated with CHX (10 µg/ml) for 0, 2, 4, and 6 h ±PKD-inhibitor 3.5 µM kbNB 142-70. D Quantitation of AU5-Spry2 levels normalized to tubulin levels. Histograms represent the average and SD of three separate assays (^*p < 0.05, ^**p < 0.01).

Fig. 5. CSN3 negatively regulates the stability of Spry2 WT.

A SW480-ADH cells, constitutively overexpressing AU5-Spry2 WT or AU5-Spry2 S112A, were transfected with either 60 nM control siRNA scramble (siRNA control), or a pool of four specific CSN3 siRNA (siRNA CSN3) for 48 h. Cells were then treated with CHX (10 µg/ml) up to 8 h. Levels of AU5-Spry2 proteins, CSN3 and tubulin were detected by immunoblotting with specific antibodies after whole-cell lysates were resolved by SDS-PAGE. Images shown are representative of three independent experiments. B Quantitation of AU5-Spry2 levels normalized to tubulin levels. Histograms represent the average and SD of three separate assays (ns not significance, ^***p < 0.001). C SW480-ADH cells constitutively overexpressing AU5-Spry2 WT were transfected with either 60 nM control siRNA scramble (siRNA control), or specific CSN3 siRNAs (siRNA CSN3). 48 h after the transfection the cells were treated as indicated with CHX (10 µg/ml) and/or MG132 (10 µM) up to 6 h. Levels of Spry2, CSN3, and tubulin were detected by immunoblotting with specific antibodies after WCL were resolved by SDS-PAGE. The result shown is representative of three separate assays. D Quantitation of AU5-Spry2 levels normalized to tubulin levels. Histograms represent the average and SD of three separate assays (^*p < 0.05).