Autophagy induced by SAHA affects mutant P53 degradation and cancer cell survival

Missense mutations in the TP53 gene produce mutant p53 (mutp53) proteins which may acquire oncogenic properties favoring chemoresistance, cell migration, and metastasis. The exploitation of cellular pathways that promote mutp53 degradation may reduce cell proliferation and invasion as well as increase the sensitivity to anticancer drugs, with a strong impact on current cancer therapies. In the last years, several molecules have been characterized for their ability to induce the degradation of mutp53 through the activation of autophagy. Here, we investigated the correlation between autophagy and mutp53 degradation induced by suberoylanilide hydroxamic acid (SAHA), an FDA-approved histone deacetylase inhibitor. In the human cancer lines MDA-MB-231 (mutp53-R280K) and DLD1 (mutp53-S241F), SAHA induced a significant mutp53 degradation. However, such degradation correlated with autophagy induction only in MDA-MB-231 cells, being counteracted by autophagy inhibition, which also increased SAHA-induced cell death. Conversely, in DLD1 cells SAHA triggered a low level of autophagy despite promoting a strong decrease in mutp53 level, and autophagy inhibition did not change either mutp53 levels or sensitivity to this drug. We conclude that autophagy can be a relevant pathway for mutp53 degradation induced by SAHA, but its contribution to mutp53 destabilization and the consequences on cell death are likely context-dependent.


Western blot analysis and chemiluminescence detection with UVITEC
For the Western blot analysis with the UVITEC system, we set up a procedure to perform the incubation with up to 4 different antibodies on the same membrane. After protein transfer and staining with the Ponceau Red of the membrane, an image with molecular weight marker is acquired (A). The membrane is then cut in three parts, according to the molecular weight marker positions, in order to perform incubations with different antibodies (Ab) (PARP, p53 and b-actin, LC3). Following the incubation with each Ab (p53 and b-act in this figure), the chemiluminescence is acquired for protein quantification (B).
To get a precise report of all protein bands, after the acquisition of the chemiluminescence signal of each antibody, the U, M and B membrane portions are joined and an image of the whole membrane, merged with the molecular weight marker, is taken (C).

Fig. S1
Supplementary Figure S1 Full-length blots of proteins illustrated in Fig 1. For each cell lines (as indicated), the membrane loaded with protein extracts of cells treated with different SAHA concentrations were processed as described in Suppl Informations. The membranes M were first hybridized with the anti-p53 Ab (DO1) and then with the anti-b-actin Ab. After primary antibodies hybridizations, the membranes were hybridized with the common secondary Ab and the chemiluminescence from p53 and b-actin was analyzed in the same UVITEC acquisition. The membranes B were hybridized with the anti-LC3 Ab, followed by hybridization with the secondary Ab and chemiluminescence was analyzed by UVITEC. After the signal acquisition for each antibody, the M and B membranes were joined and an image of the whole membrane, merged with the molecular weight marker, was taken. The orange squares outline the cropped areas reported in  Fig 2B. A) The membrane loaded with protein extracts of MDA-MB-231 cells treated with different SAHA concentrations were processed as described in Suppl Informations. The membranes M was first hybridized with the anti-p53 Ab (DO1) and then with the anti-b-actin Ab. After primary antibodies hybridizations, the membrane was hybridized with the common secondary Ab and the chemiluminescence from p53 and b-actin was analyzed in the same UVITEC acquisition. The membranes B was hybridized with the anti-LC3 Ab, followed by hybridization with the secondary Ab and chemiluminescence was analyzed by UVITEC. After the signal acquisition for each antibody, the M and B membranes were joined and an image of the whole membrane, merged with the molecular weight marker, was taken. The orange squares outline the cropped areas reported in Fig   The effect of SAHA on the in vitro assembly of tubulin microtubules was evaluated using a fluorescence-based microtubule polymerization assay kit (Cytoskeleton Inc., Denver, USA), according to the manufacturer's protocol. Briefly, 2mg/ml porcine brain tubulin (>99% pure with minimal contamination of microtubule-associated proteins) was incubated with tubulin buffer, cushion buffer, GTP without or with 2.5-5 µM SAHA in a prewarmed half area 96-well plate (Corning Costar) and the reaction was initiated by the addition of tubulin. Three µM Taxol and 1.5 µM Vincristine were used for enhancing or suppressing tubulin polymerisation, respectively (positive controls). The plate was incubated at 37 °C in a fluorescence microplate reader (Mithras LB 940, Berthold Technologies), and microtubule assembly was monitored by measuring the increase in fluorescence due to the incorporation of a fluorescence reporter into microtubules as polymerization proceeds. Fluorescence reading (λ ex , 355 nm; λ em , 460nm) was done every 5 minutes for 60 min, with 5 sec shaking every cycle.  were mock or SAHA-treated for 24 h and fixed for immunostaining with antibodies against endogenous LC3. BafA1 was added in mock-and SAHA-treated cells 3 h before the withdrawal. Immunofluorescence and DAPI images were taken with a 40x magnification and merged. The addition of BafA1 3 h before the withdrawal of SAHA-treated cells caused a further increase of LC3positive green vacuoles (compare right upper and right lower panels); B) Full-length blots of proteins illustrated in Fig 3A. The membrane loaded with protein extracts of MDA-MB-231 cells treated with different SAHA concentrations, with or without BafA1, were processed as described in Suppl Informations. The membranes M was first hybridized with the anti-p53 Ab (DO1) and then with the anti-b-actin Ab. After primary antibodies hybridizations, the membranes was hybridized with the common secondary Ab and the chemiluminescence from p53 and b-actin was analyzed in the same UVITEC acquisition. The membrane B was hybridized with the anti-LC3 Ab, followed by hybridization with the secondary Ab and chemiluminescence was analyzed by UVITEC. The orange squares outline the cropped areas reported in Fig Fig. 3C. The membrane were processed as described in Suppl Informations. The membrane M was hybridized with the anti-p53 Ab (DO1), the secondary Ab and the chemiluminescence from p53 was acquired by UVITEC. The same membrane was then hybridized with the anti-b-actin Ab and after incubation with the secondary Ab, the chemiluminescence was again acquired with UVITEC. The membranes B was hybridized with the anti-LC3 Ab, followed by hybridization with the secondary Ab and chemiluminescence was acquired by UVITEC (B). For each Ab, the first chemiluminescence acquisition is used for protein quantification and is shown on the manuscript figure. The orange squares outline the cropped areas reported in Fig 3C. After the signal acquisition for each antibody, the M and B membranes were joined and an image of the whole membrane, merged with the molecular weight marker, was taken.

SAHA-induced cytotoxicity measured in MDA-MB-231, T1, DLD1 MCF7, HCT116 wtp53 +/+ and HCT116 wtp53 -/cell lines by colony forming and MTT assays.
For colony forming assay cells were seeded at a density of 400 cells and allowed to growth for 24 h. Cells were treated by adding the drug directly in the medium at the indicated concentrations. After 24 h of treatment, the medium was removed and replaced with complete fresh medium. After 8-10 days, the colonies were counted. For MTT, cells were seeded in 96-well and treated with different SAHA concentrations in triplicates. The percentage of viable cells was determined after 24h and 48h of treatment. The average and the standard deviations of at least three independent experiments are reported.

t 24 t 48 A) Trypan blue assay in DLD1
Cells were treated with different SAHA (S) concentrations with or without the addition of BafA1 (100 nM) or CQ (50 mM) for 24 h. The percentage of survival was determined by trypan blue exclusion method as blue/total cells.

B) xCELLigence RTCA assay in DLD1
Cell proliferation/survival assays of SAHA-treated DLD1 cells measured by xCELLigence system; the t 0 arrows indicate the time of SAHA addition.   The membrane loaded with cell extracts of HCT116 wtp53 +/+ and HCT116 wtp53 -/treated with different SAHA concentrations were processed as described in Suppl Informations. The membrane U was hybridized with the anti-PARP Ab, followed by hybridization with the secondary Ab; chemiluminescence was analyzed by UVITEC. The membrane M was first hybridized with the anti-p53 Ab (DO1) and then with the anti-b-actin Ab. After primary antibodies, the membrane was hybridized with the common secondary Ab and the chemiluminescence from P53 and b-actin was analyzed in the same UVITEC acquisition. The membrane B was hybridized with the anti-LC3 Ab and chemiluminescence was analyzed by UVITEC. The orange squares outlines the cropped areas reported in Fig S11,C; C) After the signal acquisition for each antibody, the U, M and B membranes were joined and an image of the whole membrane, merged with the molecular weight marker, was taken. (Full-length blots of proteins are presented in Fig S14). B) To better highlight the modulation of wtp53, p53-S241F and p53-R280K proteins after treatment with different SAHA concentrations, a histogram was constructed by normalizing each p53/bactin value reported above for the p53/b-actin value calculated at T 0 .

Fig. S13
Supplementary Figure S14 Full-length blots of proteins illustrated in Fig S13. Membranes (M1 and M2) loaded with cell extracts of HCT116 wtp53 -/cells transiently transfected with plasmid expressing wtp53, p53-S241F and p53-R280K proteins were first hybridized with the anti-p53 Ab (DO1) and the chemiluminescence was acquired with UVITEC. Then the same membranes were incubated with the anti-b-actin Ab and after incubation with the secondary Ab, the chemiluminescence for b-actin was acquired. The orange squares outlines the cropped areas reported in Fig S13.