Determination of the characteristics of mitochondrial p53. (A) Alkali extraction showed that the mitochondrial p53 was detected in the supernatant fraction, suggesting that mitochondrial p53 retained its soluble characteristic. The membrane-integrated Cox I protein was detected in the pellet. (B) The 1% dodecymaltoside detergent extraction showed that mitochondrial p53 was more resistance than integrated Cox I protein, suggesting that both membrane-bound p53 (pellet) and free soluble form p53 (supernatant) coexisted in the mitochondria. (C) The association between p53 and phospholipids was evaluated by an in vitro lipid-binding assay. p53 displayed strong lipid-binding properties with anionic phospholipids including phosphatidylserine, cardiolipin, phosphatidylglycerol, and phosphatidic acid. (D) The p53 C-terminal domain and R175H mutant showed strong cardiolipin-binding activities. The p53 fragment containing 100 to 318 polypeptides showed mild cardiolipin-binding activity. The N-terminal domain was irrelevant to p53-cardiolipin interactions.

Down-regulation of CLS (CDS-2) expression decreased cellular p53 translocation to the mitochondria. (A) The CDS-2 shRNA constructs (1–5 µg of plasmids/6-cm transfection) were transfected into HepG2 cell for 24 hours as described in Materials and Methods. Western blots showed that the CDS-2 protein expression was down-regulated. (B) Nontransfected control, vector control, and CDS-2 shRNA-transfected cells were treated 5 µg/ml MMC for 12 hours, then total lysates were collected and analyzed by Western blots. The blots showed p53 was induced by MMC treatments. After 5 µg/ml MMC treatment for 12 hours, the mitochondria fraction was isolated as described in Materials and Methods. The contents of p53 translocation to the mitochondrial fraction significantly decreased in CDS-2 shRNA-transfected cells. The contents of mitochondrial p53 translocation were analyzed by densitometry. *P < .05, **P < .01, and ***P < .001, indicate a statistical difference with the control. ^#P < .05, ^##P < .01, and ^###P < .001, indicate a significant difference with the MMC-treated control.

Activated p53 protein was found in both nuclear and mitochondrial fractions. Western blots showed that MMC treatments induced p53 accumulation in HepG2 whole-cell lysate and mitochondrial lysate in both dose-dependent (A) and time-dependent manners (B). In dose-dependent studies, 5 and 10 µg/ml MMC was treated for 6 hours. In the time-dependent studies, 5 µg/ml MMC was treated for 3, 6, and 12 hours. The core II protein was used as a mitochondrial internal control. The nuclear PCNA protein was used to identify the fractionation quality. The contents of mitochondrial p53 translocation were analyzed by densitometry. *P < .05, **P < .01, and ***P < .001, indicate statistical significance. (C) Immunofluorescent images showed the endogenous p53 was colocalized with nuclear histone H1 (left panel) and mitochondrial Cox I (right panel) in intact A549 cells.

The expression and mitochondrial distribution of Bcl family members in HEK 293T cell stable clones expressing CDS-2 shRNA and PGS shRNA. (A) The CDS-2 shRNA- and PGS shRNA-expressing stable clone was selected in HEK 293T cells. (B) HEK 293T models have a higher p53 protein basal level. With the addition of 20 µg/ml MMC, the p53 protein content was not upregulated in 293T-wild-type, 293T-CDS-2 shRNA, and 293T-PGS shRNA cells, suggesting that the p53 status was saturated in the HEK 293T models. (C) The mitochondrial distributions of p53, Bcl-2, and Bcl-xL were analyzed in 293T cell models. The mitochondrial distribution of Bcl-2 and Bcl-xL was reduced in 293T-CDS-2 shRNA and 293T-PGS shRNA cells in relation to wild-type 293T cell. (D) The Bcl-2 and Bcl-xL mitochondrial translocations were analyzed by densitometry. The Bcl-2_{mitochondria/whole} ratio (E) and Bcl-xL_{mitochondria/whole} ratio (F) were calculated from three HEK 293T cell models and p53-null H1299 cell models and indicate a statistical difference in relation to the wild-type control. *P < .05 and ***P < .001, indicate a statistical difference between HEK 293T models. ^#P < .05 and ^###P < .001, indicate a statistical difference between H1299 models. ^aP < .05, indicates a statistical difference between H1299-wild-type control with pCEP4-p53wt expression.

Decrease mitochondrial p53 content and p53-dependent Bcl-xL or Bcl-2 mitochondrial partition enhanced MMC-induced cell death. (A) Cells were prepared and treated with MMC for 24 hours. Pretreatment with 20 µM pifithrin-α was carried out for 30 minutes before MMC incubation. The apoptotic index was analyzed by propidiumiodide staining and flow cytometry as described in Materials and Methods. ***P < .001, indicates a statistical difference with the control. ^#P < .05, ^##P < .01, and ^###P < .001, indicate a statistical difference with MMC-treated groups. (B) The expression of p53, Bax, Bcl-xL, and Bcl-2 and their mitochondrial distributions were analyzed in MMC-treated HepG2-wild-type and HepG2-CDS-2 shRNA cells. Both cell types were incubated with 5 and 20 µg/ml MMC for 6 hours. Low-dose (5 µg/ml) MMC treatment did not change Bcl-xL and Bcl-2 protein expressions in either HepG2-wild-type or HepG2-CDS-2 shRNA cells. However, as a limitation of mitochondrial p53 translocation, 5 µg/ml MMC treatment decreased Bcl-xL and Bcl-2 mitochondrial distributions in HepG2-CDS-2 shRNA cells. High-dose (20 µg/ml) MMC treatment strongly decreased Bcl-xL and Bcl-2 protein expressions in both HepG2-wild-type and HepG2-CDS-2 shRNA cells, resulting in the decrease in their mitochondrial distribution. (C and D) The Bcl-xL and Bcl-2 expression level in whole lysate (C) and their mitochondrial distributions (D) were analyzed by densitometry. *P < .05, **P < .01, ***P < .001, indicate a statistical difference with the control. ^#P < .05, indicates a statistical difference with MMC-treated groups.

Down-regulation of PGS expression decreased cellular p53 translocation to mitochondria. (A) The PGS shRNA constructs (1–5 µg of plasmids/6-cm transfection) were transfected into the HepG2 cell for 24 hours as described in Materials and Methods. Western blots showed that PGS protein expression was downregulated. (B) Nontransfected control, vector control, and PGS shRNA-transfected cells were treated 5 µg/ml MMC for 12 hours, then total lysates were collected and analyzed by Western blots. The blots showed that both p53 and PGS were induced by MMC treatments. The mitochondria fraction was isolated as described in Materials and Methods. The contents of p53 translocated to the mitochondrial fraction showed no significant difference among nontransfected control, vector control, and PGS shRNA-transfected cells. The contents of mitochondrial p53 translocation were analyzed by densitometry. *P < .05, **P < .01, and ***P < .001, indicate a significant difference with the control. ^#P < .05, ^##P < .01, and ^###P < .001, indicate a significant difference with the MMC-treated control. (C) The p53 transactivation inhibitor, pifithrin-α (20 µM), was pretreated and coincubated with 5 to 10 µg/ml MMC for 24 hours. Pifithrin-α inhibited p53-responsive Mdm2 and PGS induction but was irrelevant to p53 accumulation. (D) To determine the effect of pifithrin-α treatment on p53 mitochondrial translocation, HepG2 was pretreated with pifithrin-α for 30 minutes, followed by MMC treatment for 12 hours. The blots showed that both p53 levels in whole and in mitochondrial lysates remained unaffected by pifithrin-α. (E) The non-transfected control and PGS shRNA-transfected cells with or without pifithrin-α (20 µM) pretreatment were treated 5 µg/ml MMC for 12 hours, then total lysates were collected and analyzed by Western blots. The blots showed that PGS expression was blocked in PGS shRNA transfected cells with pifithrin-α pretreatments. The contents of p53 translocated to the mitochondrial fraction significantly decreased in PGS shRNA-transfected cells with pifithrin-α pretreatments. The contents of mitochondrial p53 translocation were analyzed by densitometry.