p8 silencing causes an increase of autophagy. (A) p8 RNAi in U2OS cells. Cells were transfected with two different p8 siRNAs (OL1 and OL2) or with a mix of two nontargeting siRNA as control (ctrl). After 24 h total RNA was analyzed by RT-PCR, by using p8-specific and actin-specific oligonucleotides. (B) p8 RNAi causes LC3-II processing in U2OS cells. Western blot analysis of total cellular lysates from U2OS cells after RNAi as described in A, and either left untreated or treated with 100 nM rapamycin (Rapa) or serum starved (SF) for additional 24 h. Membranes were developed using anti LC3B, p-S6, p-Akt, and actin antibodies. In this experiment, p-S6, p-Akt antibodies were used as treatment controls for serum starvation and Rapa addition. The LC3B antibody has a stronger reactivity for LC3B-II form. Film was quantitated by densitometry and fold induction of LC3-II, normalized to total actin, is shown. (C) p8 silencing induces autophagosome formation in U2OS cells. Red-LC3–expressing U2OS cells after either OL1 or OL2 or crtl RNAi. The cells were fixed for the analysis after 48 h. When indicated, 10 mM 3MA or 100 nM rapamycin were added for 24 h. (C and D), Representative immunofluorescent images (63×; bar, 20 μm), and the percentage of cells with LC3 puncta is shown. (E) Percentage of cells with LC3 puncta in experiments as described in C and D having pepstatin A and E64D added for 4 h before analysis. The values represent the mean ± SEM for three experiments. Samples were subjected to unpaired Student's t test.

p8 interacts with and represses Foxo3 transactivation. (A and B) p8 represses FoxO3 transactivation in a dose dependent manner. U2OS cells were transfected with a FoxO transcription factors reporter and a combination of pcDNAp8, pEGFPp8, and pCMVFoxO3 wild type (wt). The next day, cells were serum starved (0.1% FBS) for 16 h. Luciferase activity was measured in triplicate and normalized to β-galactosidase activity. The FoxO3 concentration having maximal transactivation was cotransfected with increasing amounts of pcDNAp8, in B. Mean ± SEM for four experiments is shown. (C) U2OS cells were infected either with a control silencing lentivirus (LV sicrtl) or a lentivirus silencing human p8 (LV sip8). Confluent cells were subjected to nuclear/cytosolic separation (N and C). The Western blots were carried out with the indicated antibodies, quantitated by densitometry, and the normalized values are indicated as nuclear/cytosol ratio in the histograms (mean ± SEM for three experiments; p < 0.05). Separate cultures were analyzed for total protein expression (right). (D) p8 binds to endogenous FoxO3. U2OS cells expressing enhanced green fluorescent protein (EGFP) or EGFPp8 were either left untreated (10%) or serum starved (SF) for 24 h before immunoprecipitation with anti-p8 antibodies. The western blots were carried out with anti-FoxO3 and p8, whereas total lysates were stained with GFP.

Cellular energy stress acts on p8 expression and stability. (A) Endogenous p8 is down-regulated by AMPK-signaling in U2OS cells. Analysis of U2OS cells stimulated for the indicated times with 1 mM AICAR. The Western blots were carried out with anti-p8, Bnip3, LC3B, p-AMPKα, p-AMPKβ, AMPK-cleaved PARP, and actin antibodies. For p8 and Bnip3, values were quantified as fold of induction relative to the untreated cells levels. (B) Ectopic expressed p8 is stabilized by 2DG and AICAR. 293 cells were cotransfected with pcDNAp8 and a vector expressing GST protein (pEGB). After 2 d, the cells were treated for 8 h with 20 mM 2DG, 1 mM AICAR, 20 ng/ml rapamycin, and 20 ng/ml TNF (positive control) either in the presence (10% FBS) or absence (SF) of serum in the culture medium. The Western blot analysis of total lysates was performed using anti-p8, anti-p-AMPK, p-S6, and GST as treatments and efficiency of transfection controls, respectively. (C) p8 overexpression blocks endogenous Bnip3 up-regulation by AICAR. Analysis of EGFP- or EGFPp8-expressing cells stimulated for the indicated times with AICAR as described in A. The Western blots were made with anti-Bnip3 and tubulin antibodies.

Cardiac tissue from p8 −/− mice exhibits higher basal autophagy and present increased levels of both Bnip3 RNA and protein. (A) p8 is expressed in the heart. RT-PCR analysis of total RNA from LVs of wilt type (p8 +/+) and knockout mice (p8 −/−). p8 and actin were coamplified, and PCR products were separated on 1.8% agarose gel. (B) p8 −/− mice have higher LC3-II and ATG12–5 levels. Western blot analysis of LV extracts from p8 +/+ and p8 −/− mice by using anti-LC3B, anti-ATG12-5, and β-actin antibodies. The quantifications are shown in the histograms. The values represent the mean ± SEM. Samples were subjected to unpaired Student's t test (p < 0.03 and p < 0.02, respectively; n = 5). (C) p8 −/− mice have higher bnip3 expression. bnip3, lc3, and actin were amplified from LV RNA of p8 +/+ and p8 −/− mice. PCR products were quantified and subjected to unpaired Student's t test (p < 0.01; n = 6). (D) p8 −/−mice have higher Bnip3 protein levels. Western blot analysis of LV extracts from p8 +/+ and p8 −/− mice by using anti Bnip3 and β-actin antibodies, as indicated. The quantifications are shown in the histograms (p < 0.05; n = 4). The values represent the means ± SEM.

p8 presence protects from autophagy-induced apoptosis. (A–C) p8 RNAi decreases cellular viability in response to autophagy stimuli. U2OS were cultured for 48 h after p8 or ctrl RNAi, and then the cells were cultured for further 48 h with 10% FBS or without FBS (SF) before cell viability assays. When indicated, 1 mM AICAR (AICAR) or 100 nM rapamycin (RAPA) or 10 mM 3MA was added. (B) Values were normalized to ctrl-silenced cells in serum-free conditions. Values represent the mean ± SEM for five experiments in triplicate. Samples were subjected to unpaired Student's t test. #p < 0.05. (C) Western blot analysis of the levels of LC3-II in total cellular lysates of U2OS treated as described in B. (D) Decrease of total cell number following p8 RNAi. U2OS were cultured for 96 h after p8 (sip8) or ctrl (sictrl) RNAi, and cell number was evaluated using a hemocytometer. The values represent the mean ± SEM for three experiments. (F) p8 RNAi-induced LC3 processing is blocked by autophagy inhibitors. Western blot analysis after p8 or ctrl RNAi as described in Figure 1A and serum starvation (SF) or treatment with 10 mM 3MA or 10 mM methylpyruvate (Py) for 24 h. Western blots were performed with anti-LC3B and actin antibodies. (E) p8 RNAi induces autophagy and concomitant apoptosis in U2OS cells. U2OS were cultured in 10% FBS for 48 and 72 h after p8 (sip8) or ctrl (sictrl) RNAi. Total cellular lysates were analyzed with antibodies to autophagy, survival, and apoptosis markers, as indicated. Actin was used to normalize the loading. Representative Western blots are shown. (G) p8 RNAi-induced LC3 processing is not blocked by apoptosis inhibitors. Analysis after p8 or ctrl RNAi as in Figure 1A and treatment with 20 μM zVADfmk (ZVAD) caspase inhibitor alone or with ZVAD and 10 mM 3MA combined for 16 h in 10% FBS. The Western blots were carried out with LC3B, caspase-3, cleaved PARP, and tubulin antibodies.

p8 RNAi increases FoxO3 association to bnip3 promoter resulting in higher bnip3 RNA and protein levels. (A) Increased FoxO3 association to endogenous bnip3 promoter after p8 RNAi. U2OS 24 h after p8 or crtl RNAi were either left untreated or serum starved (SF) for additional 20 h before harvesting. Chromatin immunoprecipitations of endogenous FoxO3: either anti FoxO3 (αFoxO3) or a nonimmune serum (αNS-Rabbit) were used. Total cellular chromatin (INPUT). Two FoxO-responsive elements in the bnip3 promoter (A and B in the bnip3 promoter scheme) were analyzed. p8 RNAi increases histone H3 acetylation on bnip3 promoter. ChIPs were performed from U2OS chromatin by using anti-lysine 9-acetylated histone H3 antibodies. PCR reactions were made using B oligonucleotides. (B) p8 RNAi enhances bnip3 RNA expression. U2OS were treated as described in Figure 2A, and RT-PCRs were run using bnip3, p8, and actin oligonucleotides. The histograms show the mean values ± SEM for three experiments. p < 0.05. (C) p8 RNAi increases Bnip3 protein levels. Western blot analysis of total cellular lysates from U2OS cells treated as described in Figure 5A, using either anti Bnip3 or actin antibodies, as loading control. (D) Knockdown of bnip3 decreases cell death associated with p8 silencing. U2OS were cultured for 48 h after ctrl, p8 alone, or p8 and bnip3 RNAi. Cellular viability assays (MTT) were performed after additional 48 h. The values represent the mean ± SEM for three experiments in triplicate. Samples were subjected to unpaired Student's t test. (E) p8 RNAi induces atg5-dependent autophagy and bnip3-dependent apoptosis. U2OS were cultured for 48 h in serum-free after ctrl, atg5, p8, and bnip3 alone or after combined p8+atg5, p8+bnip3, and atg5+bnip3 RNAi. Western blot analysis of total cellular lysates using LC3B, anti-cleaved PARP, caspase-3, and tubulin antibodies. The efficiency of atg5 and bnip3 knockdown is shown in Supplemental Figure 2C. The quantification of two separate Western blots is shown on the right.

p8 silencing in cardiovascular cells. (A–C) p8 silencing increases autophagy and apoptosis in cardiac cells. p8 RNAi in primary neonatal rat cardiomyocytes (A and B) and in rat H9C2 cardiac cells (C). Cells were cultured for 72 h in 10% FBS after (rat) p8 or ctrl RNAi. When indicated, both pepstatin A and E64D were added for 4 h (+inh) or 10 mM 3-methyladenine (+3MA) was added for 16 h before analysis. The Western blots were performed with anti-LC3B, PARP, rat-specific cleaved (clvd) caspase-9, and total and cleaved (clvd) caspase-3 antibodies for cardiomyocytes and LC3B and cleaved (clvd) PARP and caspase-3 antibodies for H9C2 cells. Tubulin was used as loading control. (D) p8 represses FoxO3 transactivation in a dose-dependent manner in H9C2 cells. H9C2 cells were transfected in triplicate with a FoxO luciferase reporter and pcDNAp8 and pCMVFoxO3 wt as described for Figure 3A). In separate experiments, FoxO3 was cotransfected with increasing amounts of pcDNAp8. Luciferase activity was measured in triplicate and normalized to coexpressed β-galactosidase activity. Means ± SEM for two experiments are shown. (E) p8 silencing increases autophagy in primary human cells. p8 RNAi in primary hCFs and in primary HAECs. Cells were cultured for 72 h in 10% FBS after p8 or ctrl RNAi, and the Western blot analysis was performed as described in A.

p8 genetic deletion is associated with a cardiac phenotype. (A) p8 −/− hearts display chamber dilation. Echocardiographic data from LV of p8 +/+ and p8 −/− mice (*p < 0.05); n = 15 for each group). LVEDD and LVESD are measured in millimeters. (B) p8 −/− hearts display LV wall thinning. AWT and PW, anterior and posterior wall thickness (in millimeters) (*p < 0.05 for PWT +/+ vs. −/−; n = 15 for each group). (C) p8 −/− mice have lower fractional shortening when compared with p8 +/+ mice. FS, fractional shortening, calculated from A as (LVEDD − LVESD/LVEDD) × 100% (*p < 0.01 +/+ vs. −/−; n = 15). Numeric values for echocardiographic data are shown in the right panel. (D) Scheme of the mechanism regulating autophagy by p8.