Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 21 (2), 226-237

AKT Methylation by SETDB1 Promotes AKT Kinase Activity and Oncogenic Functions

Affiliations

AKT Methylation by SETDB1 Promotes AKT Kinase Activity and Oncogenic Functions

Jianping Guo et al. Nat Cell Biol.

Abstract

Aberrant activation of AKT disturbs the proliferation, survival and metabolic homeostasis of various human cancers. Thus, it is critical to understand the upstream signalling pathways governing AKT activation. Here, we report that AKT undergoes SETDB1-mediated lysine methylation to promote its activation, which is antagonized by the Jumonji-family demethylase KDM4B. Notably, compared with wild-type mice, mice harbouring non-methylated mutant Akt1 not only exhibited reduced body size but were also less prone to carcinogen-induced skin tumours, in part due to reduced AKT activation. Mechanistically, the interaction of phosphatidylinositol (3,4,5)-trisphosphate with AKT facilitates its interaction with SETDB1 for subsequent AKT methylation, which in turn sustains AKT phosphorylation. Pathologically, genetic alterations, including SETDB1 amplification, aberrantly promote AKT methylation to facilitate its activation and oncogenic functions. Thus, AKT methylation is an important step, synergizing with PI3K signalling to control AKT activation. This suggests that targeting SETDB1 signalling could be a potential therapeutic strategy for combatting hyperactive AKT-driven cancers.

Figures

Fig. 1
Fig. 1. Akt methylation promotes its activity and oncogenic functions
a, A schematic workflow of IAP-LC-mass spectrometry (MS)/MS experiments. OVCAR5 cell lysates were proteolytically digested to perform IAP-LC-MS/MS assays. b, Alignment of MS-characterized Akt1 putative methylation residues among different species, Akt2 and Akt3. c, Immunoblot (IB) analysis of whole cell lysates (WCL) derived from DLD1-AKT1/2−/− cells infected with indicated Akt1 encoding virus and selected with hygromycin (200 μg/ml) for 72 hrs before harvesting. Data shown represent two independent experiments. d-e, cells generated in c were subjected to colony formation and soft agar assays. The experiment was performed twice independently with three repeats, and exhibited similar results (d). Representative images were shown in d and relative colony numbers derived from two independent experiments were plotted in e. f-h, Cells generated in (c) were subjected to mouse xenograft assays. Tumor sizes were monitored (f), and dissected tumors were weighed (g, h). Error bars in f and h are mean ± s.e.m, n = 7 mice. i-k, K140R and K142R mutations of Akt1 were genetically engineered in HEK293 cells by the CRISPR/CAS9-based technique. Resulting cells were serum-starved for 12 hrs, and harvested for IB analysis at different time points after stimulation with insulin (100 nM) (i). The experiment was performed twice independently with similar results (i). Cells generated in (i) were assessed for proliferation (j) assays. The experiment in j was performed three times independently and exhibited similar results. Error bars in j are mean ± s.e.m, n = 3 independent experiments. k,l,m, Cells generated in i were subjected to colony formation (k), glucose uptake (l) and lactate production (m) assays. The experiment was performed twice independently with three repeats, and exhibited similar results (k,l,m). Relative colony numbers, glucose and lactate levels derived from two independent experiments were plotted in (k,l,m). Two-way ANOVA analysis was performed in (f,j) to calculate the P value. Detailed statistical tests are described in the Methods. Source data for e, f, h and j-m are shown in Supplementary Table 2. Scanned images of unprocessed blots are shown in Supplementary Fig. 8.
Fig. 2
Fig. 2. Methylation deficient Akt1 knock-in mice display reduced body size/weight and resist to chemical carcinogen-induced skin tumorigenesis in vivo
a, Mice derived from the same litter were imaged at age of 4 weeks old. b, The mice derived from the age of 4 weeks were weighed (including 15 male mice with 5 WT, 5 heterogeneous, and 5 homogeneous-Akt1K140/142R genetic background status). Error bars are mean ± s.e.m, n = 5 mice. P values were calculated using two-tailed unpaired Student’s t test. c, The mice in (b) were euthanized and their organs were dissected and weighed. Error bars are mean ± s.e.m, n = 5 mice. P values were calculated using two-tailed unpaired Student’s t test. d, IB analysis of WCL derived from livers or hearts of WT (+/+), Akt1K140/142R-knock-in heterogeneous (+/KI) or homogeneous (KI/KI) mice from the same litter at age of 4 weeks. The experiment was performed twice, independently, with similar results. e, Graphic representation of H&E and IHC staining of heart and liver tissues derived from WT or Akt1K140/142R-KI mice. Scale bar, 50 μm. This experiment was performed twice, independently, with similar results. f, The side view of 12-weeks old mice derived from WT or Akt1K140/142R knock-in mice were treated with chemical carcinogen (DMBA following with TPA) (n = 16 for WT mice; n = 11 for Akt1-K140/142-KI mice). The neoplasm lesions were arrowed. The tumor incidence (g) and lesion numbers (h) of the mice described in (f) were calculated and plotted. Error bars are mean ± s.e.m (for WT n = 16 mice; for Akt1-K140/142-KI n=11 mice). After treatment 12 weeks with DMBA/TPA, the mice were euthanized and the H&E and IHC staining were performed (i). Scale bar, 1 mm. The experiment in (i) was performed twice, independently, with similar results. Statistical source data for b,c,g and h are shown in Supplementary Table 2. Scanned images of unprocessed blots are shown in Supplementary Fig. 8.
Fig. 3
Fig. 3. SETDB1 methylates Akt on K140 and K142 to promote its kinase activity
a-d,f,h, IB analysis of immunoprecipitates (IP), GST pulldown products and WCL derived from A375 and OVCAR5 (a) or HEK293 cells transfected with indicated constructs (b-d, f) or AKT1K140/142R-KI cells stably expressed WT or catalytically inactive SETDB1-H1224K (d). IgG was used as a negative control. e, In vitro methylation assays were performed with recombinant His-Akt1 proteins purified from insect cells as substrates and purified Flag-SETDB1 from HEK293T cells as the source of methyltransferase in the presence/absence of 3H-SAM. g, In vitro methylation assays were performed with IP Flag-SETDB1 derived from HEK293T cells as the source of methyltransferase, and the synthetic Akt1 peptides containing K140 and K142 as the substrate. Akt1-K140-me3 peptides were used as a positive control. h, IB analysis of IP products and WCL derived from A375 cells lentivirally transfected with shRNAs against SETDB1. Resulting cells were selected with puromycin for 72 hrs before harvesting. j, Setdb1 conditional knockout MEFs were treated with or without 4-OHT (500 nM) for 48 hrs to deplete endogenous Setdb1, then resulting cells were serum-starved for another 20 hrs and stimulated with insulin (100 nM) for 15 min before being harvested and subjected to IP and IB analysis. k, IB analysis of IP products and WCL derived from HEK293 cells transfected with indicated constructs. l, Setdb1 conditional knockout MEFs were infected with WT or H1224K-SETDB1 encoding virus and selected with puromycin for 72 hrs, then treated with or without 4-OHT (500 nM) for another 48 hrs before harvesting for IB analysis. m-q, Image illustration of the immunohistochemistry (IHC) staining for SETDB1, pS473-Akt1 and K140-me3-Akt1 in melanoma TMA (m,o). Scale bar, 50 μm. The distribution of SETDB1 staining was plotted in (n) (n = 97 tissue specimens). The correlation of pS473-Akt or K140-me3-Akt with SETDB1 were plotted in (p,q) (n = 95 tissue specimens). Western-blots in (a-l) were performed twice, independently, with similar results. P values were calculated using Chi-Square test in (n-q). Statistical source data for n-q are shown in Supplementary Table 2. Scanned images of unprocessed blots are shown in Supplementary Fig. 8.
Fig. 4
Fig. 4. Oncogenic function of SETDB1 depends on the activation of Akt
a-c, Human immortalized melanocytes (HIM) were lentivirally infected with indicated constructs, and selected with puromycin and hygromycin for 72 hrs before harvesting for IB analysis (a). Resulting cells were subjected to soft agar assays (b). d, IB analysis of CRISPR/CAS9-mediated AKT1 knockout and parental HEK293 cells which were lentivirally infected with the constructs encoding SETDB1. e,f, Cells described in d were subjected to colony formation assays (e). g-j, DLD1-AKT1/2−/− cells were infected with virus encoding WT or mutated Akt1, and selected with hygromycin for 72 hrs. Resulting cells were lentivirally infected with shRNA against SETDB1 (with shCtr as a negative control) and selected with puromycin for 72 hrs, and were harvested for IB analysis (g), cell proliferation (h), colony formation and soft agar (i) assays. k-n, AKT1 K140/142R-edited and parental HEK293 cells were lentivirally infected with SETDB1-WT or SETDB1-H1224K encoding constructs and selected with puromycin for 72 hrs before harvesting for IB analysis (k). Resulting cells were subjected to proliferation (l) and colony formation (m) assays. The experiments in (b,e,i,m) were performed twice independently with three repeats, and exhibited similar results. Relative colony numbers derived from two independent experiments were plotted in (c,f,j,n). The cell proliferation assays in (h,l) were performed three times independently, and cell numbers were quantified in (h,l). Error bars are mean ± s.e.m, n = 3 independent experiments. Two-way ANOVA analysis was performed in (h,l) to calculate the P value. Detailed statistical tests are described in the Methods. Source data for c,f,h,j,l and n are shown in Supplementary Table 2. Western-blots in (a,g,k) were performed twice, independently, with similar results. Scanned images of unprocessed blots are shown in Supplementary Fig. 8.
Fig. 5
Fig. 5. SETDB1-medidated methylation of Akt synergizes with PI3K to activate Akt
a,b, A375 cells were serum-starved for 20 hrs, then stimulated with insulin (50 nM) at different time points (a) or post-treatment of various PI3K inhibitors (b) before harvesting for IP and IB analysis. c,d, IB analysis of IP products and WCL derived from HEK293 cells transfected with indicated constructs stimulated without (c) or with IGF (100 ng/ml) (d) before harvesting. e, In vitro binding assays were performed with recombinant GST-Akt1 protein purified from mammalian cells, and flag beads bound SETDB1. The binding was performed in 4°C for 4 hrs incubated with or without PIP3 (20 μM) and subjected to IB analysis. f-h, IB analysis of Akt1-IP and WCL derived from HEK293 cells infected with indicated SETDB1 encoding constructs (f), AKT1K140/142R and its parental HEK293 cells (g) and SETDB1 depleted A375 cells (h). i,j, IB analysis of PIP3 pull-down products and WCL derived from Setdb1 conditional knockout MEFs treated with or without 4-OHT (500 nM) for 48 hrs (i) or from HEK293 cells transfected with indicated constructs (j). Where indicated, empty beads (Ctr) serve as a negative control. k,l, IB analysis of cell fractionations separated from Setdb1 conditional knockout MEFs treated with or without 4-OHT (500 nM) for 48 hrs (k) or from AKT1K140/142R-edited and parental HEK293 cells (l). All Western-blots above were performed twice, independently, with similar results. Scanned images of unprocessed blots are shown in Supplementary Fig. 8.
Fig. 6
Fig. 6. KMD4B demethylates Akt to inhibit Akt kinase activity
a,c-e, IB analysis of Akt1-IP products and WCL derived from A375, OVACR5 cells (a), SETDB1 expressing HEK293 cells (c), AKT1K140/142R-edited and parental HEK293 cells (d) and SETDB1-depleted A375 cells (e). IgG was used as a negative control. b, IB analysis of in vitro de-methylation assays performed with synthetic Akt1-K140-me3 peptides as substrate, and bacterially purified catalytic domain of KDM4B as the source of demethylase. f, IB analysis of WCL derived from primary Kdm4b conditional knockout MEFs infected with or without phage-Cre for 48 hrs before harvesting. g, A375 cells were lentivirally infected with shRNA against KDM4B. Resulting cells were serum starved for 20 hrs, then stimulated with IGF (100 ng/ml) before harvesting for IP and IB analysis. h, IB analysis of IP products and WCL derived from A375 cells infected with lentivirus against KDM4B. i, IB analysis of cell fractionations separated from A375 cells lentivirally infected with shRNA against KDM4B or KDM4A. j,k, HEK293 cells were transfected with indicated constructs and treated with different PI3K inhibitors for 1 hr (j) before subjected to GST pull-down assay and IB analysis. l-o, IHC staining of KDM4B, pS473-Akt1 and K140-me3-Akt1 in melanoma TMA (l). Scale bar, 50 μm. The distribution of KDM4B staining was plotted in (m, n = 97 tissue specimens). The correlations of pS473-Akt or K140-me3-Akt1 with KDM4B were plotted in (n, n = 96 tissue specimens; o, n = 95 tissue specimens). All P values were calculated using Chi-Square test. Detailed statistical tests were described in Methods. Western-blots were performed twice, independently, with similar results. Statistical source data for m-n are shown in Supplementary Table 2. Scanned images of unprocessed blots are shown in Supplementary Fig. 8.
Fig. 7
Fig. 7. Deficiency of SETDB1 inhibits Akt kinase activity and oncogenic function
a-d, SETDB1-depleted A375 and control cells were subjected to mouse xenograft assays. Tumor sizes were monitored (a,b). Tumors were dissected (c) and tumor mass were weighed (d). Error bars are mean ± s.e.m, n = 7 mice. P values were calculated by using two-way ANOVA analysis (b) and two-tailed unpaired Student’s t test (d). e, The phosphorylation status of Akt1 (pT308-Akt) and methylation of H3K9 (H3K9me3) were detected by IB analysis with WCL derived from recovered xenografted tumors. f, IB analysis of IP products and WCL derived from A375 cells treated with different doses of Mithramycin A for 72 hrs before harvesting. g, HEK293 cells were transfected with indicated constructs and treated with different doses of Mithramycin A for 72 hrs before harvesting for GST pull-down assays and IB analysis. h-j, AKT1K140/142R-edited and parental HEK293 cells were treated with different doses of Mithramycin A for 72 hrs and subjected to IB analysis (h). Meantime, resulting cells were subjected to colony formation (i) assay. The experiment was performed twice independently with three repeats, and exhibited similar results (i). Representative images were shown in (i) and relative colony numbers derived from two independent experiments were plotted in (j). k-n, Mithramycin A treatment reduced in vivo tumorigenesis of xenografted A375 cells. When the tumors of xenografted A375 cells reached 100 mm3, the mice were treated with Mithramycin A (0.2 mg/kg) or PBS (as a negative control). Tumor sizes were monitored in (k,l) and tumor mass were weighed and presented in (m,n). Error bars are mean ± s.e.m, n = 8 mice. P values were calculated by using two-way ANOVA analysis (l) and two-tailed unpaired Student’s t test (n). Detailed statistical tests are described in the Methods. Source data for b, d, j, l and n are shown in Supplementary Table 2. Scanned images of unprocessed blots are shown in Supplementary Fig. 8.

Comment in

Similar articles

See all similar articles

Cited by 12 PubMed Central articles

See all "Cited by" articles

References

    1. Jones PA, Issa JP & Baylin S Targeting the cancer epigenome for therapy. Nat Rev Genet 17, 630–641 (2016). - PubMed
    1. You JS & Jones PA Cancer genetics and epigenetics: two sides of the same coin? Cancer Cell 22, 9–20 (2012). - PMC - PubMed
    1. Glaser KB HDAC inhibitors: clinical update and mechanism-based potential. Biochem Pharmacol 74, 659–671 (2007). - PubMed
    1. Fahy J, Jeltsch A & Arimondo PB DNA methyltransferase inhibitors in cancer: a chemical and therapeutic patent overview and selected clinical studies. Expert Opin Ther Pat 22, 1427–1442 (2012). - PubMed
    1. Delmore JE et al. BET bromodomain inhibition as a therapeutic strategy to target c-Myc. Cell 146, 904–917 (2011). - PMC - PubMed

Publication types

MeSH terms

Feedback