JMJD6 promotes colon carcinogenesis through negative regulation of p53 by hydroxylation

Abstract

Jumonji domain-containing 6 (JMJD6) is a member of the Jumonji C domain-containing family of proteins. Compared to other members of the family, the cellular activity of JMJD6 is still not clearly defined and its biological function is still largely unexplored. Here we report that JMJD6 is physically associated with the tumor suppressor p53. We demonstrated that JMJD6 acts as an α-ketoglutarate- and Fe(II)-dependent lysyl hydroxylase to catalyze p53 hydroxylation. We found that p53 indeed exists as a hydroxylated protein in vivo and that the hydroxylation occurs mainly on lysine 382 of p53. We showed that JMJD6 antagonizes p53 acetylation, promotes the association of p53 with its negative regulator MDMX, and represses transcriptional activity of p53. Depletion of JMJD6 enhances p53 transcriptional activity, arrests cells in the G1 phase, promotes cell apoptosis, and sensitizes cells to DNA damaging agent-induced cell death. Importantly, knockdown of JMJD6 represses p53-dependent colon cell proliferation and tumorigenesis in vivo, and significantly, the expression of JMJD6 is markedly up-regulated in various types of human cancer especially in colon cancer, and high nuclear JMJD6 protein is strongly correlated with aggressive clinical behaviors of colon adenocarcinomas. Our results reveal a novel posttranslational modification for p53 and support the pursuit of JMJD6 as a potential biomarker for colon cancer aggressiveness and a potential target for colon cancer intervention.

Figure 1. JMJD6 is physically associated with p53 in vivo and in vitro.

(A) Cellular extracts from HCT116 cells stably expressing vector or FLAG-JMJD6 were immunopurified with anti-FLAG affinity columns and eluted with FLAG peptide. The eluates were resolved by SDS-PAGE and silver-stained. The proteins bands were retrieved and analyzed by mass spectrometry. (B) HCT116 cell lysates were immunoprecipitated with antibodies against JMJD6 followed by immunoblotting with antibodies against p53 (FL-393), or they were immunoprecipitated with antibodies against p53 (FL-393) followed by immunoblotting with antibodies against JMJD6. (C) Immunofluorescence-stained endogenous JMJD6 (red) and p53 (green) were visualized by confocal microscopy. DAPI staining was included to visualize the cell nucleus (blue). Scale bar, 25 µm. (D) Mapping the domain of p53 that is required for its interaction with JMJD6. GST pull-down experiments were performed with GST-fused JMJD6 and in vitro transcribed/translated Myc-tagged full-length p53 or deletions of p53.

Figure 2. JMJD6 hydroxylates p53 in vivo and in vitro.

(A) Recombinant p53 was incubated with or without recombinant JMJD6 in the presence or absence of α-ketoglutarate (2-OG) and Fe(II). The mixture was then separated on SDS-PAGE, and the band corresponding to the molecular weight of p53 was excised and digested with trypsin and analyzed by LC-MS/MS. Inserts show the doubly charged peptide precursor ions that were fragmented. The relevant ion fragments are labeled, and the corresponding peptide positions are illustrated. K, lysine; K(OH), hydroxylated lysine; the expected increase in mass by hydroxylation modification is 16 Dalton. M, methionine; m, randomly oxidized methionine, which results in a +16 Dalton shift in mass. (I) Experimental group with 2-OG, Fe(II), and JMJD6; (II) negative control group without JMJD6; (III) negative control group without Fe(II); (IV) negative control group without 2-OG. (B) Wild-type JMJD6 hydroxylates p53^381–393 at K382 of p53 in the presence of 2-OG and Fe(II) in vitro. The peptides corresponding to amino acids 381–393 of p53 (wild-type p53, p53K382R, or p53K382A) were incubated with or without recombinant JMJD6 or JMJD6(H187A/D189A) in the presence or absence of 2-OG and Fe(II) for 2 h at 37°C. The mixture was then analyzed by MALDI/TOF. (C) Hydroxylation of p53 at K382 in vivo. Lysates from HCT116 cells were immunoprecipitated with anti-p53 monoclonal antibody-conjugated agarose. Bound proteins were eluted with p53 peptide, separated on SDS-PAGE, and analyzed by LC-MS/MS. Inserts show the doubly charged peptide precursor ions that were fragmented. The relevant ion fragments are labeled and the corresponding peptide positions are illustrated. Analysis by LC-MS/MS revealed the presence of modified p53^382–393 peptide (M+2H)²⁺ containing hydroxylation of K382. (D) Extracted ion chromatogram (XIC) of nonmodified (upper panel) and hydroxylated p53K382 (lower panel) extracted from vector (black) or JMJD6 (red) transfected HCT116 cells.

Figure 3. Negative regulation of p53 transcriptional activity by JMJD6.

(A) Measurement of mRNA (left panel) and protein (right panel) levels of p21 and PUMA by real-time RT PCR and Western blotting in HCT116 cells that were transfected with JMJD6 siRNAs and/or JMJD6 siRNA-1–resistant JMJD6 form (rJMJD6) followed by treatment with or without VP-16. Each bar represents the mean ± S.D. for triplicate measurements. *p<0.05. (B) HCT116 cells were treated with control siRNA or JMJD6 siRNAs and challenged with or without VP-16. Real-time RT PCR was performed using exon–exon junction-specific or intron–exon junction-specific primers to measure spliced and unspliced mRNA levels of p21 and PUMA by RT-qPCR analysis to determine splicing efficiency of p21 (left panel) and PUMA mRNA (right panel). Each bar represents the mean ± S.D. for triplicate measurements. (C) Reporter assays in HCT116 cells that were transfected with JMJD6 siRNAs and/or rJMJD6 together with p21 promoter-driven luciferase reporter construct and challenged with or without VP-16. *p<0.05. (D) qChIP was performed in HCT116 cells treated with control siRNA or JMJD6 siRNA with indicated antibodies. (E) HCT116 cells transfected with JMJD6 siRNAs and/or rJMJD6 were synchronized by double thymidine block and released into the cell cycle. Cells were collected for cell cycle analysis by flow cytometry. Experiments were repeated three times and the data from a representative experiment are shown. *p<0.05. (F) HCT116 cells were transfected with control siRNA or JMJD6 siRNAs and challenged with or without VP-16 for 24 h. Annexin V/PI staining and flow cytometry were performed to assess the effect of JMJD6 on the apoptosis of HCT116 cells. Experiments were repeated three times, and the data from a representative experiment are shown. Each bar represents the mean ± S.D. for triplicate experiments. *p<0.05.

Figure 4. The negative impact of JMJD6 on p53 pathway is through its effect on p53 protein.

(A) HCT116 p53^+/+ or HCT116 p53^−/− cells were transfected with control siRNA or JMJD6 siRNAs. The mRNA levels of p21 and PUMA were detected by real-time RT PCR and the levels of the indicated proteins were detected by Western blotting. (B) HCT116 p53^+/+ or HCT116 p53^−/− cells were transfected with control siRNA or JMJD6 siRNAs together with p21 promoter-driven luciferase construct. Cells were then harvested and luciferase activity was measured and normalized to that of renilla. Each bar represents the mean ± S.D. for triplicate experiments. (C) HCT116 p53^−/− cells were transfected with control siRNA or JMJD6 siRNAs and were synchronized by double thymidine block and released into the cell cycle before cell cycle analysis by flow cytometry. Experiments were repeated three times and the data from a representative experiment are shown. (D) HCT116 p53^−/− cells were transfected with control siRNA or JMJD6 siRNAs, and annexin V/PI staining and flow cytometry were performed to assess cell apoptosis. (E) HCT116 p53^+/+ or HCT116 p53^−/− cells were infected with lentivirus carrying a control siRNA or JMJD6 siRNAs, and were subcutaneously injected into the right anterior armpit of BALB/c nude mice. Tumors were measured weekly with Vernier calipers, and volume was calculated using the formula π/6×length×width². Each point represents the mean ± S.D. for different animal measurements (n = 6). The levels of the indicated proteins extracted from xenograft tumor were detected by Western blotting. p values were determined by Student's t test. *p<0.01. (F) Immunohistochemical staining was performed in xenograft tumor sections using antibody against Ki-67. Scale bar, 36 µm. Proliferation was assessed by counting the number of Ki-67 positively stained nuclei and total number of cancer cells at 400× magnification in five representative regions of the tumor.

Figure 5. The negative effect of JMJD6 on p53 pathway depends on its hydroxylase activity.

(A) HCT116 cells were transfected with vector, FLAG-JMJD6, or FLAG-JMJD6(H187A/D189A) (mut-JMJD6). The levels of the indicated proteins were detected by Western blotting. The mRNA levels of p21 and PUMA were detected by real-time RT PCR. (B) HCT116 cells were transfected with p21 promoter-driven luciferase construct together with vector, FLAG-JMJD6, or FLAG-JMJD6(H187A/D189A) (mut-JMJD6) plasmids. Cells were then harvested and luciferase activity was measured and normalized to that of renilla. Each bar represents the mean ± S.D. for triplicate experiments. (C) HCT116 cells transfected with vector or FLAG-JMJD6, or FLAG-JMJD6(H187A/D189A) (mut-JMJD6) were synchronized by double thymidine block and released into the cell cycle. Cells were collected for cell cycle analysis by flow cytometry. Experiments were repeated three times and the data from a representative experiment are shown. (D) HCT116 cells were transfected with vector, FLAG-JMJD6, or FLAG-JMJD6(H187A/D189A) (mut-JMJD6), and challenged with VP-16. Annexin V/PI staining and flow cytometry were performed to assess the effect of JMJD6 on the apoptosis of HCT116 cells.

Figure 6. The interplay between hydroxylation and acetylation of p53.

(A) Knockdown of JMJD6 promotes acetylation of p53^K382. Lysates from HCT116 cells treated with control siRNA or JMJD6 siRNA-1 were immunoprecipitated with antibodies against p53 followed by immunoblotting with antibodies against acetyl-K382 of p53 (p53^K382ac). (B) HCT116 cells were transfected with CBP and different amounts of FLAG-JMJD6 or FLAG-JMJD6(H187A/D189A) (mut-JMJD6) expression constructs together with p21 promoter-driven luciferase reporter. Cells were then harvested and luciferase activity was measured and normalized to that of renilla. Each bar represents the mean ± S.D. for triplicate experiments. (C) The peptide p53^381–393 without (upper) or with (lower) K382 acetylation was incubated with recombinant JMJD6 in the presence of 2-OG and Fe(II). The mixture was then analyzed by MALDI/TOF. (D) qChIP assay was done to measure p53^K382ac bound to its response element and to a control site in p21 promoter in HCT116 cells treated with control siRNA or JMJD6 siRNA-1 and challenged with or without VP-16. (E) Lysates from HCT116 cells treated with control siRNA or JMJD6 siRNA-1 were immunoprecipitated with antibodies against p53 followed by immunoblotting with antibodies against MDMX or MDM2. (F) JMJD6 did not affect p53 ubiquitination in vivo. HCT116 p53^−/− cells were co-transfected with FLAG-JMJD6, HA-ubiquitin, wild-type p53, or p53 mutant (p53K382R). Forty-eight hours after transfection, cells were treated with MG132 for 6 h before cellular extracts were prepared for co-immunoprecipitation assays with anti-p53 followed by immunoblotting with anti-HA. (G) JMJD6 could not hydroxylate K382-methylated p53^381–393 peptides. The peptide p53^381–393 with or without K382 monomethylation was incubated with recombinant JMJD6 in the presence of α-ketoglutarate and Fe(II). The mixture was then analyzed by MALDI/TOF. (I) JMJD6 hydroxylates K382 of p53^381–393 peptides; (II) JMJD6 could not hydroxylate K382-methylated p53^381–393 peptides.

Figure 7. JMJD6 is a potential biomarker for colon cancer aggressiveness.

(A) Immunohistochemical staining of JMJD6 in paired samples of breast ductal carcinoma (1), hepatocellular carcinoma (2), lung adenocarcinoma (3), lung squamous carcinoma (4), suprarenal epithelioma (5), pancreatic ductal carcinoma (6), colon adenocarcinoma (7), esophageal squamous carcinoma (8), rectal adenocarcinoma (9), and gastric adenocarcinoma (10) versus adjacent normal tissues. Representative tumor and adjacent normal sections stained with JMJD6 antibody are shown (magnification, ×25; scale bar, 200 µm). Each type of carcinoma included at least six paired samples and the scores were determined by evaluating the extent and intensity of immunopositivity. (B) Immunohistochemical staining of JMJD6 in 90 samples of colon adenocarcinomas paired with adjacent normal tissues. Representative sections from colon cancer (upper, tubular adenocarcinoma; lower, poorly differentiated adenocarcinoma) or adjacent normal tissue stained with JMJD6 antibody are shown (magnification, ×100; scale bar, 100 µm). The scores were determined by evaluating the extent and intensity of immunopositivity and were analyzed by paired-samples t test (***p<0.001). (C) Representative sections of histological grade I, II, and III of colon adenocarcinomas that were stained with JMJD6 antibody are presented (magnification, ×100; scale bar, 100 µm). The scores were determined by evaluating the extent and intensity of immunopositivity and were analyzed by two-tailed unpaired t test (***p<0.001). (D) Time-to-event data were plotted using Kaplan–Meier curves, and the 5-year survival rate of different groups was compared using the Mantel–Cox log-rank test (***p = 0.001). The y-axis represents the percentage of patients, and the x-axis represents the survival in months. (E) High expression of JMJD6 protein is found only in the base of intestinal glands (crypt of Lieberkuhn). Representative images of immunohistochemical staining of JMJD6 in normal intestinal glands are shown. Upper—magnification, ×25; scale bar, 200 µm. Lower—magnification, ×400; scale bar, 50 µm.