CDK16 Phosphorylates and Degrades p53 to Promote Radioresistance and Predicts Prognosis in Lung Cancer.

Rationale: Radioresistance is considered the main cause of local relapse in lung cancer. However, the molecular mechanisms of radioresistance remain poorly understood. This study investigates the role of CDK16 in radioresistance of human lung cancer cells.
Methods: The expression levels of CDK16 were determined by immunohistochemistry in lung cancer tissues and adjacent normal lung tissues. Immunoprecipitation assay and GST pulldown were utilized to detect the protein-protein interaction. The phosphorylation of p53 was evaluated by in vitro kinase assay. Poly-ubiquitination of p53 was examined by in vivo ubiquitination assay. Cell growth and apoptosis, ROS levels and DNA damage response were measured for functional analyses.
Results: We showed that CDK16 is frequently overexpressed in lung cancer cells and tissues, and high levels of CDK16 are correlated with lymph node stage and poor prognosis in lung cancer patients. Furthermore, we provided evidence that CDK16 binds to and phosphorylates p53 at Ser315 site to inhibit transcriptional activity of p53. Moreover, we uncovered that this phosphorylation modification accelerates p53 degradation via the ubiquitin/proteasome pathway. Importantly, we demonstrated that CDK16 promotes radioresistance by suppressing apoptosis and ROS production as well as inhibiting DNA damage response in lung cancer cells in a p53-dependent manner.
Conclusion: Our findings suggest that CDK16 negatively modulates p53 signaling pathway to promote radioresistance, and therefore represents a promising therapeutic target for lung cancer radiotherapy.

Introduction

Lung cancer continues to be the deadliest human malignancy worldwide with 5-year overall survival rate of less than 15% 1, 2. About 85% of lung cancer cases are classified as non-small cell lung cancer (NSCLC), of which adenocarcinoma accounts for 50% of histologic subtypes of NSCLC 3. Radiotherapy has been considered a promising and major local treatment strategy for lung cancer. However, local relapse occurs frequently due to radioresistance for most patients. Therefore, it is critical to elucidate the underlying mechanisms and overcome radioresistance in lung cancer.

p53, the most extensively studied tumor suppressor gene, is mutated or functionally inactive in approximately 50% of human cancers 4-6. p53 is labile and is usually ubiquitinated and targeted for degradation by MDM2 E3 ubiquitin ligase under normal physiological conditions 7, 8. Following various stress signals, such as DNA damage, hypoxia, and oncogene activation, p53 is stabilized, activated, and translocated into the nucleus, where it induces the transactivation of many downstream target genes involved in cell-cycle arrest, apoptosis, DNA repair, and senescence. Accumulating evidence suggest that p53 plays essential roles in a wide spectrum of biological processes, including apoptosis, autophagy, metabolism, immune response, DNA damage repair, and tumorigenesis 9-11. Additionally, p53 also functions as a key player that determines radiosensitivity by modulating cell cycle redistribution and DNA damage 12-14. Therefore, p53 is regarded as a potential target for radiation therapy.

CDK16, also termed PCTAIRE1 or PCTK1, is a newly identified member of the cyclin-dependent kinases family (CDK) 15. CDK16 is activated by binding to cyclin Y (CCNY) or its homologue cyclin Y-like 1 (CCNYL1) proteins 16-19. CDK16 has been found to be involved in vesicle trafficking 20, 21, neurite outgrowth 22, spermatogenesis 16, glucose transportation 23, 24, skeletal myogenesis 25, and spindle orientation 26. In addition, recent studies demonstrated indispensable roles of CDK16 in tumorigenesis. CDK16 silencing has been reported to suppress cell growth and proliferation in multiple human cancers, such as prostate cancer, breast cancer, and colorectal cancer 27-29. These studies suggest that CDK16 may act as an oncoprotein in several types of cancers. However, an in-depth exploration of its role in lung cancer has not yet been performed.

In this study, we found that CDK16 is overexpressed in lung cancer and predicts poor prognosis in patients. Furthermore, we showed that CDK16 binds to and phosphorylates p53 at Ser315 site to trigger p53 degradation via the ubiquitin/proteasome pathway. Moreover, we demonstrated that CDK16 depletion enhances radiosensitivity in a p53-dependent manner in lung cancer cells, suggesting that CDK16 is an attractive target for cancer radiotherapy.

Materials and Methods

Cell culture and transfection

Cells (BEAS, A549, H292, H1299, H358, H1975, HCC827) were obtained from ATCC and cultured in RPMI-1640 supplemented with 10% fetal bovine serum (FBS) and 100 U/mL penicillin and streptomycin in 5% CO2 at 37 °C. HEK293T, U2OS and MCF-7 cells were purchased from ATCC and cultured in DMEM supplemented with 10% FBS and 100 U/mL penicillin and streptomycin in 5% CO2 at 37 °C. Plasmid transfections were performed using Lipofectamine 2000 (Invitrogen).

Constructs

CDK16 plasmid was purchased from DF/HCC DNA Resource Core. The construct was subcloned into pDONR201 entry vector and transferred to destination vectors with the indicated SFB or Myc tag using Gateway Technology (Invitrogen). Plasmids encoding p53 and HA-ubiquitin have been described previously 30. Mutations were introduced using the Quik-Change Site-Directed Mutagenesis Kit (Stratagene) and verified by DNA sequencing.

Antibodies and reagents

Anti-human CDK16 antibodies were purchased from Sigma and Proteintech. Anti-p53, anti-Myc, anti-GAPDH, anti-CDK1, anti-CDK2, anti-CyclinB1, anti-CyclinD1, and anti-MDM2 antibodies were obtained from Santa Cruz Biotechnology. Anti-human p27 was from Abgent, anti-Flag antibody was from Sigma, and anti-human p53-S315 P, anti-p53-S33P, anti-p53-S46P phospho-specific antibodies and anti-HA antibody were from Cell Signaling Technology. MG132 and Cycloheximide were purchased from Calbiochem and Sigma respectively.

RNAi treatment

The following target siRNA sequences were as follows: SiCDK16#1, CCACUGAGGACAUCAACAA and SiCDK16#2, GGAGAUCAGACUGGAACAU, which was previously described 27. Sip53: AAGACTCCAGTGGTAATCTAC, was described previously 31. SiRNA transfection (50 nM siRNA) was performed using Lipofectamine RNAiMAX reagent (Invitrogen). 48 h post-transfection, cells were harvested and analyzed.

CRISPR/Cas9 gene-editing technology to generate CDK16 knockout cells

This assay was performed as described previously 32. In Brief, lenti-CRISPR sgRNAs targeting CDK16 were co-transfected into HEK293T cells with the packaging plasmids pSPAX2 and pMD2G. After 48 h, supernatants were collected and used to infect A549 cells in the presence of 10 μg/mL Polybrene (Sigma). Stable CDK16 knockout A549 cells were selected in the addition of 2 μg/mL puromycin. The sgRNA sequences were as follows: CDK16-sgRNA#1: 5'-GGAGATTCAGCTACAAAAGG-3'; CDK16-sgRNA#2: 5'-ATAGGCCTGGATGAGAGTGG-3'.

Western blotting and immunoprecipitation

Cells were lysed in NETN buffer (20 mM Tris-HCl [pH 8.0], 1 mM EDTA, 100 mM NaCl, and 0.5 % Nonidet P-40), and the clarified lysates were resolved by SDS-PAGE and transferred to PVDF membranes for Western blotting using ECL detection reagents. For immunoprecipitation, the supernatants were first incubated with S-protein agarose (Novagen) overnight at 4 °C, and the precipitates were washed five times with NETN buffer. To detect endogenous interaction, the clarified supernatants were first incubated with anti-p53 antibody and then protein A/G-agaroses overnight. After being washed five times with NETN buffer, the samples were collected and analyzed by Western blotting.

GST pull down assay

GST-vector or GST-p53 fusion proteins purified from bacteria were immobilized on GST beads (GE Healthcare) and incubated with lysates prepared from HEK293T cells transiently transfected with SFB-tagged CDK16 for 2 h at 4 °C. The samples were washed five times and then analyzed by Western blotting.

Cytoplasmic and nuclear protein fractionation

A549 cells were transfected with scramble or CDK16 siRNAs. After 48 h, the cells were collected. Cytoplasmic and nuclear fractions were separated using NE-PER Nuclear Cytoplasmic Extraction Reagent kit (Thermo Fisher Scientific) according to the manufacturer's procedures.

In vivo ubiquitination assay

A549 cells were transfected with indicated siRNAs or constructs. MG132 (10 μM) was added 4 h before harvesting. Cells were then lysed with NETN buffer. The supernatants were incubated with S-protein agarose and then analyzed by Western blotting.

In vitro kinase assay

HA-p53 and HA-p53-S315A were synthesized in vitro using the High-yield protein expression system (Promega). SFB-CDK16 was obtained by immunoprecipitation using anti-Flag agarose from lysates of cells overexpressing SFB-CDK16. The samples were then incubated with HA-p53-WT or HA-p53-S315A in 50 μL of kinase buffer (Cell Signaling Technology) containing 10 μM ATP for 30 min at 30 °C. The samples were subsequently analyzed by Western blotting.

Real-time quantitative PCR

This assay was performed as described previously 30. Briefly, total RNA from A549 cells was isolated using Trizol reagent (Invitrogen), and reverse transcription assay was carried out using the ReverTra Ace qPCR RT Kit (Toyobo, Japan). The relative mRNA levels of each gene were measured as two power values of ΔCt (Ct of GAPDH minus the Ct of the target genes). Primer sequences are listed in Table .

Cell clonogenic survival assays

A549 cells transfected with indicated siRNAs were seeded into 6-well plates and irradiated with doses as indicated. Cells were then cultured for 14 days. After fixation and staining, the colonies containing more than 50 cells were calculated.

Cell cycle analysis

A549 cells were transfected with scrambled or CDK16 siRNAs. 48 h later, cells were first fixed with 70% ethanol and then treated with propidium iodide (PI) and RNase A for 30 min. The samples were analyzed by flow cytometry.

Apoptosis assay

A549 cells were transfected with scrambled or CDK16 siRNAs. 48 h post-transfection, Cells were harvested by trypsin without EDTA, washed with PBS, and then stained with annexin V-EGFP and propidium iodide. The samples were analyzed by flow cytometry.

ROS measurement

A549 cells transfected with indicated siRNAs were harvested and washed three times with PBS buffer. Cells were then incubated with fresh phenol red-free DMEM containing 10% FBS and 25 mM DCF (2',7'-dichlorofluorescein diacetate) (Sigma) for 20 min and were collected. ROS levels were measured by flow cytometry.

Immunofluorescence staining

A549 Cells transfected with indicated siRNAs were cultured on coverslips and then exposed to 2 Gy IR. Cells were then fixed with 4% paraformaldehyde for 15 min at various time points after radiation (0 min, 30 min, 4 h and 24 h) and permeabilized in 0.2% Triton X-100 for 5 min. After blocking with 5% bovine serum albumin, the samples were incubated with γ-H2AX antibody overnight and secondary antibody for 1h. Cells were counterstained with DAPI for 10 min and visualized by fluorescence microscopy.

Lung Cancer Tissue Microarray and Immunohistochemistry Staining

A lung cancer tissue microarray was purchased from Shanghai Outdo Biotech (Shanghai, China), which contained 90 carcinoma tissues and paired para-carcinoma tissues. All patients had been pathologically diagnosed with adenocarcinoma after operation. The immunohistochemistry (IHC) analysis was performed as described previously 33, 34. Briefly, the tissue sections were blocked with goat serum and then incubated with anti-CDK16 antibody (Sigma) overnight at 4 ºC. The sections were stained with 3, 3-diaminobenzidine and counterstained with hematoxylin after being incubated with secondary antibody. PBS was used as a negative control. Both staining intensity and positive percentage were used to examine the expression of CDK16 in lung cancer tissue: the IHC staining score (values 0-12) was calculated by multiplying the scores for intensity of positive staining (negative = 0, weak = 1, moderate = 2, or strong = 3) and the percentage of positive-stained cells (0-25% = 1, 26-50% = 2, 51-75% = 3, >75% = 4). IHC scores ≥ 8 defined high expression, while scores < 8 defined low expression.

Data analysis

In our study, three or more independent experiments were carried out for each assay. Results are presented as the mean ± SD. An unpaired Student's t-test was used for statistical comparison in each group. χ2 test was used to determine the correlation between CDK16 expression and clinicopathologic variables. The Kaplan-Meier method was used to evaluate the overall survival. P value <0.05 was considered to be statistically significant.

Results

CDK16 is frequently overexpressed and predicts poor prognosis in lung cancer

To investigate the expression of CDK16 in lung cancer, we first examined the protein level of CDK16 in different lung cancer cell lines. As shown in Figure , CDK16 was significantly upregulated in most lung cancer cell lines compared to normal human lung epithelial cells BEAS-2B. Next, we performed immunohistochemical staining for CDK16 in a human lung cancer tissue microarray, which contained 90 carcinoma tissues and paired adjacent lung tissues. As shown in Figure , CDK16 was mainly located in the cytoplasm and overexpressed in lung cancer compared to adjacent lung tissues. The positive rate of CDK16 staining in lung cancers (71.7%) was significantly higher than that in adjacent lung tissues (0%) (P < 0.01). Furthermore, we found that high CDK16 expression significantly correlated with lymph node stage in clinicopathologic characteristics (P < 0.05) (Table . The Kaplan-Meier analysis demonstrated that the patients with high CDK16 expression had a significantly shorter overall survival than those with low CDK16 expression (Figure . Taken together, these data indicate that CDK16 overexpression may contribute to tumorigenesis and elevated levels of CDK16 predict adverse prognosis in lung cancer patients.

CDK16 negatively regulates p53 stability in multiple p53 wild-type cell lines

Our clinical data suggested CDK16 upregulation may promote tumorigenesis in lung cancer. To further explore the underlying mechanisms, we examined a few tumour suppressive and oncogenic proteins that are known to regulate cell growth and cell cycle progression in CDK16-depleted A549 lung cancer cells. Consistent with a previous study 27, we observed increased level of p27 (Figure , which has been considered a CDK16 downstream effector. Interestingly, p53 level was also substantially augmented when CDK16 was depleted (Figure . Moreover, restoring siRNA-resistant CDK16 expression remarkably reduced p53 protein level in CDK16-depleted cells (Figure ). Additionally, we employed a CRISPR/Cas9-based gene editing strategy to deplete CDK16 and found that loss of CDK16 dramatically resulted in the accumulation of endogenous p53 in A549 cells (Figure . These findings strongly support the notion that the stability of p53 is regulated by CDK16 in lung cancer cells. Furthermore, we showed that ectopically expressed CDK16 led to decreased level of p53 (Figure , further supporting that p53 abundance is negatively controlled by CDK16. More importantly, this phenomenon was also observed in several other cancer cell lines expressing wild-type p53 (Figure , implying that p53 regulation by CDK16 may be a common feature. Notably, loss of CDK16 did not affect the mRNA level of p53 (Figure , suggesting that CDK16 may control p53 protein level. Furthermore, we found that p53 protein level was not altered by CDK16 depletion in cells harboring p53 mutations (Figure . Together, these results suggest that CDK16 negatively regulates p53 stability at the post-translational level.

CDK16 directly interacts with p53 in vivo and in vitro

To elucidate how the protein kinase CDK16 controls p53 stability, we first examined whether CDK16 would interact with p53 in cells. As expected, the exogenously expressed CDK16 co-immunoprecipitated with p53 and vice versa (Figure . To further confirm this interaction, we performed an endogenous co-immunoprecipitation experiment and detected a complex containing CDK16 and p53 (Figure . Importantly, we found that recombinant GST-tag-expressed p53 immunoprecipitated with ectopically expressed CDK16 (Figure , indicating CDK16 directly associates with p53 in vitro. Next, we generated a series of p53 deletion mutations (Figure and showed that the core DNA binding domain of p53 (residues 100-300) was capable of binding to CDK16 (Figure . Notably, the mobility of HA-p53 (100-393) fragment (294aa) was much faster than HA-p53 1-300) fragment (300aa). We speculated that HA-p53 (100-393) fragment presents a lower frequency of proline since the N-terminus of p53 contains a proline-rich region (residues 61-94). In fact, this mobility difference was reported in our previous work (30. Together, these data strongly suggest that CDK16 directly binds to p53 in vivo and in vitro.

CDK16 phosphorylates p53 at Ser315 site and controls its transcriptional activity

Our data above suggest that CDK16 interacts with p53 and modulates its stability. Given that CDK16 is a protein kinase, we asked whether p53 is a substrate of CDK16. It is well established that most CDK substrates contain S/T-P-X-K/R motif 35. We first examined three known p53 phosphorylation sites that contain the S/T-P motif. As shown in Figure , phospho-p53-S315 level was reduced in CDK16 knockdown cells, while phospho-p53-S33 and phospho-p53-S46 levels remained constant, suggesting that Ser315 site may be a candidate phosphorylation site by CDK16. Notably, the Ser315 site is evolutionally conserved (Figure . To further investigate whether CDK16 directly phosphorylates p53, we performed an in vitro kinase assay and showed that CDK16 could phosphorylate wild-type p53 at Ser315 site (Figure . Anti-phospho-p53-S315 antibody could not detect p53 when the Ser315 site was mutated to alanine (S315A) (Figure , indicating that this antibody is highly specific for the S315 phosphorylated form of p53. Collectively, these findings suggest that CDK16 directly phosphorylates p53 at Ser315 site in vivo and in vitro.

Because p53 phosphorylation has been reported to associate with its nuclear localization 36, we therefore speculated that CDK16 might also affect this process. Indeed, CDK16 knockdown led to p53 accumulation in the nucleus (Figure . Nuclear p53 functions as a transcriptional factor and induces the expression of p53 downstream genes. As shown in Figure , CDK16 knockdown clearly increased the expression levels of MDM2 and PUMA by over 2-fold, while those of NOXA and P21 were not remarkably altered. Notably, CDK16 depletion did not affect the protein level of MDM2, which functions as ubiquitin ligase to target p53 for degradation (Figure ), suggesting that MDM2 may not feedback to degrade p53 under this condition. Taken together, our results indicate that the phosphorylation of p53 by CDK16 inhibits its nuclear accumulation and activation of its transcriptional activity.

CDK16 regulates p53 stability via the ubiquitin/proteasome pathway

Considering that CDK16 negatively regulates p53 accumulation and p53 is a labile protein degraded by the proteasome, we speculated that CDK16 may affect p53 stability via the ubiquitin/proteasome pathway. Indeed, there was no further increase of p53 protein level when proteasome inhibitor MG132 was added in CDK16-depleted cells (Figure , suggesting that CDK16 knockdown inhibits p53 degradation by the ubiquitin/proteasome pathway. Moreover, we showed that the half-life of endogenous p53 protein was longer in CDK16-knockdown cells when compared to that in control cells (Figure . We also performed in vivo ubiquitination assays and showed that p53 poly-ubiquitination was decreased when CDK16 was inhibited (Figure , indicating that p53 degradation promoted by CDK16 is ubiquitination-dependent.

Since CDK16 phosphorylates p53 and then leads to its degradation, we speculated that the CDK16-dependent degradation of p53 also requires its ability to phosphorylate p53 at Ser315 site. As shown in Figure , the S315A mutant displayed significantly reduced p53 ubiquitination when compared to wild-type p53 in cells with ectopic expression of CDK16. Moreover, wild-type CDK16, but not its kinase dead mutant (K194M) 26, 27, promoted poly-ubiquitination of p53 (Figure . Taken together, our results suggest that CDK16 negatively controls the stability of p53 by promoting its polyubiquitination and degradation in a phosphorylation-dependent manner.

Knockdown of CDK16 enhanced radiosensitivity of lung cancer cells

To explore the biological significance of CDK16 in lung cancer cells, we examined cell proliferation and apoptosis since the p53 target genes MDM2 and PUMA were upregulated when CDK16 was inhibited. As shown in Figure , cell colony formation was substantially decreased in CDK16-depleted cells, suggesting that CDK16 contributes to cell proliferation in lung cancer. Moreover, apoptosis was significantly increased after CDK16 depletion (Figure , which was further confirmed by elevated sub-G1 population (Figure , indicating that knockdown of CDK16 promotes apoptosis.

p53 has been shown to modulate reactive oxygen species (ROS) generation 37; therefore, we measured ROS levels in CDK16 knockdown lung cancer cells. As shown in Figure , CDK16 silencing led to increased production of ROS. p53 also participates in DNA damage response 38, 39. Therefore, we carried out a γH2AX foci assay and showed that IR-induced γ-H2AX foci were significantly increased 4 h and 24 h after irradiation in CDK16-depleted lung cancer cells (Figure . Additionally, loss of CDK16 significantly enhanced cellular sensitivity to radiation (Figure . Together, these findings indicate that CDK16 promotes radioresistance by suppressing apoptosis and ROS production as well as inhibiting DNA damage response in lung cancer cells.

p53 mediates the biological effects of CDK16 in lung cancer cells

To confirm the functional roles of p53 in radioresistance caused by CDK16 knockdown, A549 cells were transfected with siRNAs targeting CDK16, p53 or both (Figure . As shown in Figure , p53 knockdown partially rescued the defects in cell colony formation in CDK16-depleted cells. Furthermore, we found that apoptosis caused by CDK16 silencing was inhibited by p53 depletion (Figure . Moreover, the increased ROS levels and γ-H2AX foci induced by CDK16 depletion were partially reversed when p53 was co-depleted (Figure . We also showed that p53 knockdown partially restored cell survival after irradiation in CDK16-depleted cells (Figure . Together, these data suggest that CDK16 exerts its functions mainly via its ability to inhibit p53.

Discussion

In this study, we demonstrated that CDK16 phosphorylates p53 at Ser315 and promotes ubiquitination and subsequent degradation of p53. We also showed that CDK16 promotes radioresistance by suppressing apoptosis and ROS production as well as inhibiting DNA damage response in a p53-dependent manner. Importantly, CDK16 is overexpressed in lung cancer and predicts unfavorable prognosis, implying that CDK16 may be a promising therapeutic target for lung cancer radiotherapy.

As a key tumor suppressor protein, p53 and its associated activities are tightly controlled by its interactions with other proteins, its subcellular localization and its post-translational modifications 34, 40, 41. For example, our previous study has shown that hSSB1 interacts with p53 and regulates the transcriptional activity and stability of p53 30. Moreover, phosphorylation of p53 at Ser15 site by ATM kinase leads to p53 stabilization by blocking p53 ubiquitination in response to DNA damage 41. In this study, we identified CDK16 as a key upstream kinase responsible for p53 phosphorylation at Ser315 site. We further demonstrated that p53 phosphorylation by CDK16 leads to destabilization of p53 and inhibits p53 transcriptional activity. In addition, MDM2 ubiquitin ligase has been shown to be mainly responsible for p53 degradation 41. Thus, we speculate that MDM2 E3 ligase may also be involved in the ubiquitination and degradation of p53 promoted by CDK16. Collectively, our data suggest that CDK16 is a key regulator of p53 and promotes the degradation of p53 in a phosphorylation-dependent manner.

Elevated expression of CDK16 has been observed in some types of human cancer 27-29. We demonstrated that CDK16 is overexpressed in lung cancer and is associated with poor clinical outcomes. Our follow-up functional studies have revealed that CDK16 accelerates cell proliferation and impairs apoptosis, further supporting that CDK16 may act as an oncoprotein and overexpression of CDK16 could contribute to tumorigenesis. Furthermore, we uncovered a new role for CDK16 in the regulation of radioresistance although CDK16 may not be a DNA damage-responsive protein (Figure ). We uncovered that CDK16 depletion enhanced radiosensitivity of lung cancer cells by promoting apoptosis and ROS production as well as inhibiting DNA damage repair. Importantly, p53 knockdown could partially rescue these phenotypes caused by CDK16 depletion, suggesting that p53 is the major downstream effector of CDK16, which functions in radioresistance in lung cancer.

In conclusion, our study identified p53 as a novel substrate of CDK16 and elucidated the mechanism by which p53 is regulated by CDK16 in lung cancer. Furthermore, we demonstrated that CDK16 promotes radioresistance in lung cancer by facilitating p53 degradation, as proposed in Figure . Several CDK16 inhibitors have been developed and exhibited anti-tumor activities 42, 43. Our data further support the development of CDK16 inhibitors and suggest that these inhibitors may be effective for cancer radiotherapy, especially for lung cancer displaying elevated levels of CDK16.