DNA-dependent protein kinase catalytic subunit (DNA-PKcs)-SIN1 association mediates ultraviolet B (UVB)-induced Akt Ser-473 phosphorylation and skin cell survival

Activation of Akt/mTOR pathway is the major regulator of apoptosis resistance and cancer transformation in UVB irradiated skin cells [8, 20‐23]. It is known that mTOR kinase forms at least two distinct multi-protein complexes, namely mTOR complex 1 (mTORC1), or the rapamycin sensitive complex composed of mTOR, Raptor and PRAS40 [24], as well as mTOR mTORC2 [25‐27]. mTORC1 phosphorylates S6K and 4E-BP1 [24]. MTORC2 promotes Akt Ser-473 phosphorylation, and increases its activity [19]. Thus, we tested Akt and mTOR activation in primary cultured skin keratinocytes. Akt activation was reflected by phospho(p)-Akt (Ser-473 and Thr-308) and its downstream p-GSK3β (Ser 9), mTORC1 activation was shown by p-S6K (Thr 389), p-S6 (Ser 235/236), p-4EBP1 (Ser 65) and p-Raptor (Ser 792). The activation of mTORC2 was confirmed by p-Akt (Ser-473) and p-Rictor (Tyr 1135). Immunoblotting results in Figure 1 showed that UVB radiation induced significant activation of Akt (Figure 1A and B), mTORC1 (Figure 1C-F) and mTORC2 (Figure 1A, B, E and F) in primary cultured skin keratinocytes. The effect of UVB was time- (Figure 1A, C and E) and dose-dependent (Figure 1B, D and E). Thus, our results confirmed that UVB activates Akt, mTORC1 and mTORC2 signalings in primary skin keratinocytes.

The key issue of this study is to understand the potential role of DNA-PKcs in UVB-induced Akt activation. As shown in Figure 2A, UVB induced DNA-PKcs activation in primary keratinocytes, which was demonstrated by its phosphorylation at Thr 2647 and Thr 2609 (Figure 2A). Importantly, DNA-PKcs siRNA knockdown significantly inhibited UVB-induced Akt Ser-473 phosphorylation, while leaving Akt Thr-308 phosphorylation unaffected (Figure 2B). We used two distinct, non-overlapping siRNAs against DNA-PKcs, and both siRNAs inhibited DNA-PKcs expression and Akt Ser-473 phosphorylation (Figure 2B). Similar results were also seen in transformed keratinocytes (HaCaT cells), as DNA-PKcs siRNA knockdown dramatically inhibited UVB-induced Akt Ser-473 (but not Thr-308) phosphorylation in HaCaT cells (Figure 2C). NU7026, the DNA-PKcs kinase inhibitor [28, 29], almost abolished UVB-induced Akt Ser-473 phosphorylation (Figure 2D), indicating that DNA-PKcs kinase activity was required for Ser-473 phosphorylation by UVB. To further support this, we found that UVB-induced Ser-473 phosphorylation was suppressed by DNA-PKcs kinase-dead mutation (T2609A), while over-expression of WT-DNA-PKcs enhanced Ser-473 phosphorylation (Figure 2E). In both conditions, Akt Thr-308 phosphorylation was not affected (Figure 2E). Note that UVB-induced S6 phosphorylation, an indicator of mTORC1 activation, and Erk1/2 phosphorylation were not affected by NU7026 or DNA-PKcs mutation (Figure 2D and E). Thus, our results suggest that DNA-PKcs activation is required for UVB-induced Akt Ser-473 phosphorylation, which is further supported by the fact that DNA-PKcs knockout (KO) mouse embryonic fibroblasts (MEFs) showed a defective Akt Ser-473 phosphorylation after UVB radiation (Figure 2F). Note that Ku70 siRNA knockdown had almost no effect on UVB-induced Akt Ser-473 phosphorylation in cultured skin keratinocytes, suggesting that DNA-PKcs, but not Ku70 is required for UVB-induced Akt Ser 473 phosphorylation (Additional file 1: Figure S1). Further, insulin-induced Akt Ser-473 and Erk1/2 phosphorylation was not affected by DNA-PKcs deficiency in MEFs (Figure 2G).

The above results suggested that DNA-PKcs is required for UVB-induced Akt Ser-473 phosphorylation. As discussed, mTORC2 is the known Ser-473 kinase. Thus, we tested whether a physical interaction between DNA-PKcs and mTORC2 components (i.e. SIN1 [30]) existed in UVB-treated cells. CoIP experiment results in Figure 3A and B confirmed DNA-PKcs-SIN1 association in UVB-irradiated keratinocytes, which was abolished by DNA-PKcs inhibitor NU7026 (Figure 3A and B). NU7026 had no affect on DNA-PKcs/mTOR expression or S6 phosphorylation (Figure 3C). Results from SIN1 KO MEFs confirmed that SIN1 was required for Akt Ser-473 phosphorylation by UV (Figure 3D). Notably, UVB-induced DNA-PKcs-SIN1 association was inhibited by DNA-PKcs mutation or silencing (Figure 3E), confirming the requirement of DNA-PKcs activity for the complexation. Importantly, DNA-PKcs was found to translocate from cell nuclei to cytosol after UVB radiation, while SIN1 stayed in the cytosol (Figure 3F). Thus, the binding between DNA-PKcs and SIN1, and the following Akt phosphorylation (Ser-473) were likely happening in the cytosol, although a small fraction of Akt was also found to shuttle into nuclear after UVB radiation (Figure 3F).

Trans-activation of EGFR by UVB mediates activation of Akt and other signaling pathways [7, 8]. We then wanted to know if EGFR was also important for UVB-induced DNA-PKcs/SIN1 association. As expected, AG1478, the EGFR inhibitor blocked UVB-induced mTOR activation (Ser 2448 phosphorylation) and Akt phosphorylation (Ser-473) (Figure 4A input). Importantly, UVB-induced DNA-PKcs/SIN1/mTOR association was also blocked (Figure 4A). Intriguingly, EGFR was not in the complex of DNA-PKcs/SIN1/mTOR (Figure 4A), suggesting that EGFR mediated DNA-PKcs/SIN1 complexation was an indirect effect. Further, EGFR depletion abolished UVB-induced DNA-PKcs/SIN1/mTOR association (Figure 4C), mTOR phosphorylation (Figure 4B) and Akt/Erk activation in MEFs (Figure 4E). However, UVB-induced DNA-PKcs activation (phosphorylation) was not affected by either AG1478 or EGFR depletion (Figure 4A and D), thus UVB-induced DNA-PKcs activation was not dependent on EGFR. SIN1 siRNA knockdown had no effect on UVB-induced DNA-PKcs and mTOR phosphorylation, suggesting that SIN1 likely lies downstream of DNA-PKcs (Figure 4B).

As shown in our previous study, activation of Akt by UVB radiation contributes to apoptosis-resistance and promotes survival of damaged cells [9]. The Akt inhibitor perifosine significantly enhances UVB-induced skin cell apoptosis and reduces DNA damages through Akt in-activation along with other actions [9]. Since DNA-PKcs mediates UVB-induced Akt Ser-473 phosphorylation, we then tested whether DNA-PKcs activation was also important for UVB-induced anti-apoptosis effect. As shown in Figure 5A, the DNA-PKcs kinase inhibitor NU7026 enhanced UVB-induced death of keratinocytes, which was shown by the decreased cell viability. Meanwhile, UVB-induced apoptosis was significantly enhanced by NU7026 (Figure 5B and C). These results suggested that DNA-PKcs activation is important for apoptosis resistance against UV. The fact that DNA-PKcs depletion also enhanced UVB-induced MEFs death (Figure 5D) further supported our proposal. Notably, SIN1 KO cells showed similar phenotype as DNA-PKcs KO cells, and were more sensitive to UVB-induced death (Figure 5D). Thus, DNA-PKcs/SIN1 inhibition exacerbates UVB-induced cell death and apoptosis.

NU7026 and AG1478 were purchased from Calbiochem (Shanghai, China). Annexin V apoptosis kit was obtained from Promega (Shanghai, China).Insulin, 3-(4,5-dimethyl-thiazol-2-yl)2,5-diphenyl tetrazolium bromide (MTT) dye was purchased from Sigma (Shanghai, China).

Antibodies again Erk1/2, S6, Ku70, p-DNA-PKcs (Thr 2609) and Akt (1/2) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse anti-β-actin antibody was obtained from Sigma (Shanghai, China). Antibodies against phospho(p)-Akt (Ser 473), p-Akt (Thr 308), p-mTOR (Ser 2448), mTOR, p-S6K (ribosomal p70 S6 kinase 1,Thr 389), S6K, p-S6 (Ser 235/236), p-GSK3β (Ser 9), GSK3β, p-4E-BP1 (Ser 65), DNA-PKcs, EGFR, p-EGFR (Tyr 1068), Rictor, p-Rictor (Tyr 1135), Raptor, p-Raptor (Ser 792) were purchased from Cell Signaling Technology (Denver, MA). Anti-p-DNA-PKcs (Thr 2647) antibody was purchased from Abcam (Shanghai, China).

Primary human epidermal keratinocytes, HaCaT cells as well as various mouse embryonic fibroblasts (MEFs) were maintained in DMEM medium supplemented with a 10% FBS, with anti-biotic in a CO₂ incubator at 37°C. UVB radiation equipments and procedure were described in [41, 42]. EGFR wild-type (WT) and knockout (KO) MEFs were reported [7]. DNA-PKcs WT and DNA-PKcs MEFs were from Dr. Yang’s Lab [43]. SIN1 WT and SIN1 KO MEFs were gifts from Dr. Li at Nanjing Medical University.

Western blot and cell viability (MTT) assay were described in our previous studies [44, 45]. Nuclear and cytosol fraction of cells were isolated through the nuclear and cytosol separation kit purchased from Nanjing Kai-ji Biotech (Nanjing, China) according to the protocols. Band intensity was quantified with ImageQuant V5.1 software, and was normalized to control group (Molecular Dynamics, Piscataway, NJ).

The Cell Apoptosis Histone-DNA ELISA PLUS Kit (Roche, Palo Alto, CA) was used to detect apoptosis in skin cells after different treatments as previously described [44, 45].

After indicated treatment/s, keratinocytes were washed with cold PBS and incubated with the binding buffer attached, containing 2 μg/ml of annexin V-FITC for 10 min. Cells were then washed with PBS and resuspended. A total of 15,000 cells of each event were analyzed by flow cytometry (Beckton Dickinson FACScan). Annexin V (apoptotic) percentage was recorded.

DNA-PKcs-siRNA-1 was purchased from Santa Cruz (sc-35200 h). DNA-PKcs-siRNA-2 (leading strand: GAUCGCACCUUACUCUGUUdTdT), targeting 352 bases downstream from the start codon, was purchased from Dharmacon Research Inc [46]. Scramble siRNA, Ku 70 siRNA (sc-29383) and SIN1 (MAPKAP1) siRNA (sc-60984) were also purchased from Santa Cruz. SiRNA transfection was performed through Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to recommendation procedure, see [44].

A 3-kb Hind III fragment of DNA-PKcs cDNA covering Thr 2609 was used as the template for generating the T2609A mutation of DNA-PKcs cDNA. Site-directed mutagenesis was performed using the Quik Change site-directed mutagenesis kit (Stratagene) and the forward (TCCGATTTTGTGGAGsGACCAGGCCTCCCAGGGC) and reverse (GCCCTG GGAGGCCTGGTCCTCCACAAACATCGGA) primers [47]. The mutated DNA-PKcs cDNA fragment was assembled back into the full-length DNA-PKcs cDNA as described previously [47]. T2609A DNA-PKcs or the WT-DNA-PKcs cDNA were introduced into pSV2 neo plasmid, which was transfected into the keratinocytes through Lipofectamine 2000 protocol [44]. Forty-eight hours after transfection, cells were re-plated on selection medium containing 500 μg/mL of G418 for 8–10 days. Individual colonies were isolated and further characterized for expression.

As previously reported [45], keratinocytes or MEFs with indicated treatments were lysed with lysis buffer containg 0.3% CHAPS and protease inhibitor cocktails (Roche Diagnostics, Indianapolis, IN). Aliquots of 1000 μg of proteins from each sample were pre-cleared by incubation with 15 μl of protein A/G Sepharose (beads) (Amersham, IL) for 1 hour at 4°C. Precleared samples were incubated with specific antibody (1–2 μg/sample) in lysis buffer (1000 μl) overnight at 4°C. To this was added 30 μl of protein A/G beads and the samples were incubated for 2 hours at 4°C. The beads were washed five times with phosphate-buffered saline (PBS) and once with lysis buffer, boiled, separated by 10% SDS-PAGE, and transferred onto a PVDF membrane followed by immunoblotting analysis.