SAHA induces caspase-independent, autophagic cell death of endometrial stromal sarcoma cells by...

Journal of PathologyJ Pathol 2008; 216: 495–504Published online 26 August 2008 in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/path.2434

Original Paper

SAHA induces caspase-independent, autophagic celldeath of endometrial stromal sarcoma cells by influencingthe mTOR pathway

A Hrzenjak,* † M-L Kremser, B Strohmeier, F Moinfar, K Zatloukal and H Denk†

Saldow Research Unit for Molecular Pathology of Gynecologic Tumors, Department of Pathology, Medical University of Graz, Graz, Austria

*Correspondence to:A Hrzenjak, Saldow ResearchUnit for Molecular Pathology ofGynecologic Tumors, Departmentof Pathology, Medical Universityof Graz, Auenbruggerplatz 25,8036 Graz, Austria.E-mail:andelko.hrzenjak@meduni-graz.at

†Both authors should beconsidered as senior authors.

No conflicts of interest weredeclared.

Received: 9 June 2008Revised: 28 July 2008Accepted: 7 August 2008

AbstractEndometrial stromal sarcomas are rare and molecular mechanisms involved in their patho-genesis are poorly understood. Covalent modifications of histone proteins, in particularde/acetylation of lysine residues, play an important role in the regulation of gene transcrip-tion in normal and neoplastic cells, but there are only limited data about these processesin solid mesenchymal tumours. We treated endometrial stromal sarcoma cells (ESS-1) andnon-malignant human endometrial stromal cells (HESCs) with suberoylanilide hydroxamicacid (SAHA), a histone deacetylase inhibitor. SAHA was able to mediate the cell cycle andexpression of genes related to the malignant phenotype of endometrial stromal tumours, egp21WAF1 and HDAC7. SAHA led to dose-dependent differentiation and death of ESS-1 cellsbut not of HESCs. Exposure of HESCs to SAHA resulted only in slightly decreased cell pro-liferation. SAHA also increased the p21WAF1 expression and caused significant changes in thecell cycle by inhibiting the G1/S transition in ESS-1 cells. Recovery experiments indicatedthat these changes became irreversible when the tumour cells were treated with SAHA forlonger than 24 h. In our experimental system, not apoptotic but autophagic processes wereresponsible for the cell death. Monodansyl cadaverine accumulation in treated ESS-1 cellsand decreased expression of the mTOR and phospho-S6 ribosomal protein (S6rp) addition-ally supported this observation. Taken together, our study indicates that HDACs might beconsidered as potential drug targets in the therapy of stromal sarcomas and that SAHAmight be a promising therapeutic agent for endometrial stromal sarcoma.Copyright 2008 Pathological Society of Great Britain and Ireland. Published by JohnWiley & Sons, Ltd.

Keywords: uterus; endometrial stromal sarcoma; histone deacetylase; SAHA; G1 arrest;autophagy; mTOR

Introduction

Endometrial stromal sarcomas are very rare uterinetumours of mesenchymal origin. These neoplasmscomprise benign stromal nodules, low-grade endome-trial stromal sarcomas (ESSs), and undifferentiatedendometrial sarcomas (UESs). The last two enti-ties show different histological features and clinicalbehaviour [1]. Their heterogeneity and the low inci-dence of these malignancies make investigations dif-ficult. This is also reflected in the diagnostic methodsand therapy of these tumours, which is primarily lim-ited to surgery. However, recurrences occur often aftera prolonged period of time and metastases are not lim-ited to the peritoneum but can be found in differentorgans involving especially the lung [2–4] and bones[5–7]. The overall 5-year survival rate of ESS patientsranges from 60% to 90% [8–10], whereas the 5-yearsurvival rate of UES patients is markedly lower.

Histone deacetylases (HDACs) and histone acetyltransferases (HATs) are responsible for covalent mod-ification of histone proteins and consequential changesin chromatin architecture and gene expression. HDACsplay a critical role in the pathogenesis of varioustumours; for example, certain types of haematolog-ical malignancies and solid tumours. The familyof HDACs contains 18 enzymes divided into fourclasses, based on sequence homology to their yeastcounterparts. Especially enzymes of classes I andII can be efficiently inhibited by different HDACinhibitors [11–13]. Our recent work revealed thatHDAC2 is considerably overexpressed in 80% ofESSs, when compared with non-neoplastic endome-trial stroma [14]. Meanwhile, a few HDAC inhibitorsexist with high specificity and few side effects withintheir therapeutic range. One of the most interestinginhibitors is suberoylanilide hydroxamic acid (SAHA).SAHA inhibits HDACs, resulting in the accumulationof acetylated histones H2a, H2b, H3, and H4 [15,16]

496 A Hrzenjak et al

through direct interaction with the catalytic site of theenzyme [17]. This drug has been recently approved bythe FDA for therapy of cutaneous T-cell lymphomaand it is also used in clinical trials in patients withother malignancies including solid tumours, such asmesothelioma, medulloblastoma, and prostate and thy-roid cancer [18–20]. When administered intravenouslyor orally, SAHA induces cell differentiation, cell cyclearrest, and apoptosis in different tumour cells. Thesetherapeutic effects are observed at well-tolerated doseswith only mild side effects including fatigue, diar-rhoea, malaise, vomiting, and thrombocytopenia [21].

Here we tested the effect of SAHA on endometrialstromal sarcoma cells (ESS-1) and non-neoplasticendometrial stromal cells (HESCs). Our results showthat SAHA very efficiently inhibits cell growth andcauses cell differentiation and autophagic cell deathof ESS-1, but not of HESCs. These data indicate thatSAHA might be a promising therapeutic agent for thetreatment of endometrial stromal sarcoma.

Materials and methods

Chemicals and cell lines

All chemicals and media were purchased from Sigma(Sigma-Aldrich Handels GmbH, Vienna, Austria).SAHA was purchased from Alexis Biochemicals(Lausen, Switzerland). The human endometrial stro-mal sarcoma cell line (ESS-1), established byGunawan et al [22], was purchased from the Ger-man Collection of Microorganisms and Cell Cultures(Braunschweig, Germany). Human endometrial stro-mal cells (HESCs; ATCC No CRL-4003) were estab-lished by Krikun et al [23]. This study was approvedby the Medical Ethics Committee and all experimentswere performed according to local ethical guidelines.

Measurements of apoptosis

ESS-1 cells and HESCs were treated with 3 µM SAHAfor 24, 48, and 72 h. Harvested cells were incubatedwith 10 µl of Cleaved Caspase-3 (Asp175) Antibody-Alexa Fluor 488 Conjugate (Cell Signaling Technol-ogy, No 9669) for 60 min and finally measured byflow cytometry (BD FACSCalibur, BD Biosciences),using staurosporine-treated cells (0.1 µM; 4 h) as apositive control.

Immunoblotting

The following antibodies and concentrations/dilutionswere used: rabbit anti-HDAC1 (1 µg/ml); rabbit anti-HDAC2 (1 µg/ml); rabbit anti-HDAC3 (9 µg/ml); rab-bit anti-HDAC7 (3 µg/ml); mouse anti-p21 (0.5 µg/ml); rabbit anti-mTOR (1 : 1000); and mouse anti-p-S6rp (1 : 500). As secondary antibodies, we used rabbitanti-mouse and swine anti-rabbit HRP-coupled anti-bodies at a final concentration of 1 µg/ml. Specific

protein bands were visualized by enhanced chemi-luminescence assay (ECL; Amersham Biosciences,Buckinghamshire, UK). All western blots were probedfor β-tubulin to demonstrate equal loading of proteinsamples.

Recovery experiments

ESS-1 cells (1 × 105 cells per well) and HESCs (4 ×104 cells per well) were seeded into 12-well platesand left for 24 h to attach. After 24 h, both celllines were treated with growth media containing 3 µM

SAHA. The SAHA-containing medium was removedand replaced by fresh growth medium after 24, 48, and72 h, respectively. Treated cells were further grown inculture for a total of 6 days with daily observation.

Clonogenic assay

ESS-1 cells and HESCs were seeded in 6 cm culturedishes (300 cells per plate) and treated with 3 µM

SAHA for 24, 48, and 72 h. Afterwards fresh mediumwas added and the cells were cultured for another14 days, followed by fixation with butanol–acetic acid(3 : 1) and staining with 0.5% crystal violet.

Cell observer

ESS-1 cells and HESCs treated with SAHA weregrown directly on an Axiovert 200M inverse lightmicroscope (Zeiss, Vienna, Austria) equipped witha cell chamber, under standard conditions (37 ◦C,5% CO2). Every 15 min, one photograph (1388 ×1040 pixels; 16 bit black and white) was taken usingan A-Plan 10×/0.25 Ph1 lens mounted on the Axio-CamHR. In each well, two different positions werephotographed with an exposition time for every pho-tograph of 12 ms. During 72 h, a total of 289 pho-tographs per experiment were taken for visualizationand photo/film editing. Axio Vision software providedby Zeiss was used.

Electron microscopy

ESS-1 cells were grown on glass coverslips in 12-welldishes (7 × 104 cells per well), treated with SAHA,harvested by trypsinization, and fixed with ice-coldglutaraldehyde (2.5% in 0.1 M cacodylate buffer, pH7.4) for 30 min. After fixation, the samples were post-fixed in 1% OsO4 in the same buffer for 30 min,washed twice with cacodylate buffer, and rehydratedthrough a series of increasing alcohol concentrations(70%, 80%, 90%, 95% ethanol, 5 min each). Ultra-thin sections were stained with uranyl acetate and leadcitrate, and viewed with a Philips CM100 transmissionelectron microscope. Photographs were developed onKodak SO-163 Electron Image Film (Kodak, Vienna,Austria).

SAHA selectively activates autophagy in ESS cells 497

Visualization of autophagic vacuoles withmonodansylcadaverine (MDC) and fluorometricquantitation of MDC staining

Labelling of autophagic vacuoles was achieved asdescribed previously [24]. Cells were immediatelyanalysed by fluorescence microscopy (Axiophoto,Zeiss, Germany) and photographed by an AxioCamMRC5 (Zeiss) using the AxioVision rel. 4.6 soft-ware package supplied by the manufacturer. In orderto quantify the intracellular MDC, cells were anal-ysed by fluorescence spectrophotometry (Hitachi F-2500) using 380 nm as the excitation wavelength and525 nm as the emission wavelength. Photometric datawere normalized to the total number of cells per welland incorporated MDC was expressed in arbitraryunits.

Statistical analysis

Unless otherwise stated, all values represent means ofat least three independent experiments ± SD. Valueswere compared using Student’s t-test. p ≤ 0.05 wasconsidered statistically significant.

Results

SAHA inhibits G1/S transition in ESS-1 cells

We have already shown that the HDAC inhibitorvalproic acid inhibits G1/S transition in ESS-1 cells[15]. Since the growth of ESS-1 cells was even moreefficiently inhibited by SAHA, we checked whetherSAHA treatment influenced the cell cycle. As shownin Figure 1A, in SAHA-treated cells there was a pro-nounced shift in the G1/S ratio. Our time–courseexperiments showed that 3 µM SAHA caused pro-nounced inhibition of the G1/S transition in ESS-1cells as early as 24 h after commencement of the treat-ment (Figure 1B). FACS analysis showed typical G1arrest and concomitant decrease of the cell number inthe S phase, which reached its lowest level (4% of thetotal cell number) 24 h after starting the treatment. TheG1/G2 ratio continued to change slightly during thenext 24 h, until 86% and 10% of all cells were trappedin the G1 and G2 phase, respectively. G1 arrest wasalso associated with strongly deregulated expressionof p21WAF1, a protein that decreases cyclin-dependentkinase activity and thereby leads to growth arrest. Itsexpression was increased after SAHA treatment inESS-1 cells, but it was not influenced by SAHA innon-malignant HESCs (Figure 2).

SAHA decreases HDAC7 expression

The expression of different members of class I andII HDACs (eg HDAC1, 2, 3, and 7) was analysed.Decreased HDAC7 expression in both ESS-1 cells andHESCs was related to SAHA treatment. Other HDACswere only slightly affected. After SAHA treatment for

Figure 1. SAHA inhibits the G1/S transition in ESS-1 cells.Tumour cells were treated with different SAHA concentrationsfor 24 h. Low concentrations (eg 1 µM) did not cause theshift and no further increment was shown with high SAHAconcentrations (eg 5 or 10 µM). Therefore, 3 µM SAHA wasused as the working concentration in further experiments,since this concentration efficiently reduces cell growth andinhibits the G1/S transition (A). These effects are already visibleat SAHA treatment for 24 h (B)

48 h, there was a slight decrease in HDAC3 expressionin ESS-1 cells. However, this effect was absent inHESCs. Interestingly, in HESCs, HDAC2 expressionwas slightly decreased during the SAHA treatment(Figure 2).

SAHA selectively decreases the number of viableESS-1 cells

Decreased proliferation of SAHA-treated ESS-1 cellswas always associated with pronounced cell enlarge-ment; ESS-1 cells that survived SAHA treatment werecharacterized by enlarged nuclei and were consid-erably larger than non-treated, spindle-shaped ESS-1 cells (Figures 3 and 4). The selectivity of SAHAaction for ESS-1 tumour cells in comparison to HESCswas performed by colony assays. In ESS-1 cells, 24 hSAHA treatment resulted in an approximately 20%reduction in colony formation, whereas treatment for48 and 72 h resulted in 60% and 80% reduction,respectively (Figures 4A and 4B). This clearly indi-cates the ability of SAHA to kill ESS-1 cells andto prevent colony formation. In comparison, HESCswere relatively resistant to SAHA-induced cell death.

Figure 2. SAHA decreases the HDAC7 protein level in ESS-1cells. SAHA-treated and control ESS-1 cells and HESCs wereanalysed by immunoblotting. In ESS-1 cells, expression of HDAC3 and HDAC7 is mainly affected by SAHA treatment, whereasin HESCs SAHA treatment has more influence on HDAC2expression. In ESS-1 cells, in contrast to HESCs, the expressionof p21WAF1 is strongly affected by SAHA treatment. β-tubulinimmunoblotting was used as a loading control

In HESCs, the same SAHA concentration caused onlya very slight reduction in the colony number, whichwas visible as early as 72 h after starting the treat-ment. Microscopically small HESC colonies addition-ally support our observation that prolonged incubationwith SAHA slows down the proliferation of HESCs.Furthermore, using cell observer, we were able torecord that SAHA specifically killed ESS-1 cells butnot HESCs (see Supporting information, Supplemen-tary data).

Recovery of ESS-1 cells depends on the durationof SAHA treatment

As shown in Figure 4C, ESS-1 cells growing inSAHA-containing medium for 24 h showed enlargednuclei and irregularly shaped cytoplasm. However,they recovered quite efficiently in SAHA-free medium.Five days after changing the medium, there was nomorphological difference between untreated ESS-1cells and ESS-1 cells that survived 24 h SAHA treat-ment. The ESS-1 cells treated with SAHA for 48 and

72 h showed signs of more severe damage includ-ing conspicuous enlargement of cell diameters, nuclearenlargement, and many cytoplasmic extensions. Thesecells were unable to recover during incubation inSAHA-free medium. This experiment supports ourobservations with the colony-forming assays, show-ing that there is a crucial switch between 24 and 48 hof SAHA treatment which causes irreversible changesin the metabolism of ESS-1 cells.

SAHA-induced ESS-1 cell death iscaspase-3-independent

To determine whether SAHA causes apoptotic celldeath of ESS-1 cells, we measured activated caspase3, one of the most frequently used apoptotic marker.FACS analysis of activated caspase 3 in ESS-1 cellsand HESCs did not disclose activation of caspase 3after SAHA treatment (Table 1). In parallel experi-ments, ESS-1 cells showed a pronounced apoptoticeffect after treatment with 0.1 µM staurosporin, a well-known inducer of apoptosis. HESCs were more resis-tant to staurosporin. Moreover, DNA fragmentationanalysis and detection of the sub-G1 apoptotic peak byFACS analysis were also employed and did not pro-vide evidence for apoptosis in SAHA-treated ESS-1cells (data not shown).

Monodansylcadaverine (MDC) accumulation inESS-1 cells increases during SAHA treatment

MDC accumulates specifically in autophagic vac-uoles, especially in autolysosomes [24]. As shownin Figure 5A, SAHA-treated cells showed an accu-mulation of MDC in the perinuclear area. Both thenumber of vacuoles and their size increased con-siderably in SAHA-treated cells. As a positive con-trol, we used ESS-1 cells treated with 100 nM

rapamycin, a well-known inducer of autophagy. Inorder to quantify the MDC accumulation, we mea-sured the amount of intracellular MDC by fluorom-etry. These measurements had to be done immedi-ately after incubation since MDC fluorescence fadesquickly. After 24 h treatment with 3 µM SAHA, theMDC accumulation in ESS-1 cells was only slightlyincreased. In cells treated for 48 h, MDC accumu-lation increased more than two-fold and after 72 h,

Table 1. Induction of apoptosis by SAHA in ESS-1 cellsand HESCs

Proportion of apoptotic cells (%)

Cell line Treatment 24 h 48 h 72 h

ESS-1 Untreated 0.7 1.73 1.44HESCs 1.73 0.54 0.26

ESS-1 Staurosporine (0.1 µM) 26.89 49.15 34.4HESCs 3.10 3.34 3.47

ESS-1 SAHA (3 µM) 8.93 6.14 8.19HESCs 2.30 3.13 10.55

Figure 3. Survival of ESS-1 cells after SAHA treatment is concentration- and time-dependent. (A) ESS-1 cells were treated withdifferent SAHA concentrations for 24, 48, and 72 h. (B) Morphology of ESS-1 cells and HESCs after 48 h treatment with 3 µMSAHA. Note that untreated ESS-1 cells are predominantly spindle-shaped with scant cytoplasm, whereas treated ESS-1 cells arerich in cytoplasm and show a polygonal configuration. Furthermore, most treated ESS-1 cells are dead and surviving cells areenlarged. Unlike the expectations, characteristic morphological features of potential apoptosis, eg cytoplasmic shrinkage, nuclearcondensation, and/or DNA fragmentation, are absent. HESCs are affected by SAHA to a much lower extent, showing slightlydecreased proliferation without cell death. The figures on the right show cell debris in order to demonstrate dead ESS-1 cellsswimming in the medium

more than four fold in comparison to untreated con-trol cells (Figure 5B). At the same time, the numberof cells that survived SAHA treatment decreased bymore than 70% in comparison to control cells. Trans-mission electron microscopy of ESS-1 cells treatedfor 48 h with SAHA revealed extensive vacuoliza-tion as a hallmark of autophagy (Figures 5C and 5D).Furthermore, a decreased number of mitochondria,cell swelling, and nuclear enlargement were observed.Untreated ESS-1 cells showed normal morphology(Figure 5E).

mTOR and phospho-mTOR expression in ESS-1cells is decreased upon SAHA treatment

Among other things, mTOR plays a role in the ini-tiation and maturation of autophagy by controllingsignal transduction cascades involved in this pro-cess [25]. As shown in Figures 6A and 6B, both

mTOR and phospho-mTOR (p-mTOR) expression wasstrongly decreased in SAHA-treated ESS-1 cells ina concentration-dependent manner. In order to deter-mine whether SAHA treatment also influences theproduction and/or stability of mTOR mRNA, RT-PCRwas performed with RNA probes isolated from con-trol and SAHA-treated cells. We found that mTORmRNA in ESS-1 cells was not affected by SAHAtreatment (Figure 6C). Thus, it seems that in ESS-1cells, SAHA modulates the stability and/or the degra-dation mechanisms of mTOR at the protein level.SAHA also exerted an influence on mTOR phospho-rylation. The level of phospho-S6rp, a protein of the40S ribosomal unit, which plays a regulatory rolein the mTOR pathway, was also affected by SAHA(Figure 6A). The same results were found in ESS-1 cells treated by rapamycin, a well-known mTORinhibitor.

Figure 4. SAHA decreases colony-forming ability and induces irreversible cell damage in ESS-1 cells. Cells were treated with3 µM SAHA for the indicated time periods and stained with 0.5% crystal violet after growing for an additional 2 weeks. Theeffect of SAHA treatment on the formation of cell colonies is shown as a percentage of untreated control cells (100%) (A). In B,representative images of crystal violet-stained cell colonies assayed in A are shown. Interestingly, the size of treated HESC coloniesvaried considerably in comparison to ESS-1 cells. HESC colonies were small, each colony consisting of as few as 3–10 cells;therefore a microscopic evaluation was necessary in order to count stained colonies. For recovery experiments, the cells wereseeded into 12-well plates and allowed to attach for 24 h. Afterwards they were treated with growth medium containing 3 µMSAHA. SAHA-containing medium was replaced by fresh SAHA-free RPMI medium after 24, 48, and 72 h, respectively, and cellswere cultivated for an additional 120 h. Cells treated with SAHA for 24 h show enlarged nuclei and irregularly shaped cytoplasmbut they fully recovered in normal growth medium. ESS-1 cells treated for 48 and 72 h with SAHA show severe alterations bothin nuclei and in cytoplasm. These cells were not able to recover in SAHA-free growth medium

Discussion

Endometrial stromal sarcomas represent less than1% of all uterine malignancies [26]. The molecularmechanisms involved have to be elucidated in orderto improve therapeutic options predominantly limitedto surgical excision. Therapeutic alternatives suchas aromatase inhibitors and hormonal therapy aresometimes used for selected ESS cases. However,their efficiency seems to be largely case-specific andthese studies are additionally hampered by the limited

number of cases [27,28]. No efficient adjuvant therapyexists for a high-grade variant of these malignancies.

Among other mechanisms, transcriptional regula-tion in eukaryotic cells is exerted by acetylation anddeacetylation of histone proteins [29]. The expressionof different genes involved in cell growth, apoptosis,cell migration, and tumour growth are consequentlyaffected by these modifications. Therefore, in the lastfew years, HDACs and HDAC inhibitors have becomehot topics in cancer therapy. Today, HDAC inhibitorsrepresent a potent group of anti-cancer agents [30].

Figure 5. SAHA induces autophagic cell death in ESS-1 cells. Following treatment with either 3 µM SAHA or 100 nm rapamycin,the ESS-1 cells were additionally incubated for 1 h at 37 ◦C with 0.05 mM MDC and analysed by fluorescence microscopy (A).Note the MDC accumulation and characteristic punctate appearance in autophagic vacuoles. Significant cell enlargement is alsovisible. In untreated confluent ESS-1 cells kept in cell culture for 72 h, MDC staining was also slightly increased. This is expected,since autophagy also occurs under physiological conditions. The data were evaluated by fluorometry and normalized to thecell number (B). Additionally, ESS-1 cells were analysed by transmission electron microscopy. In SAHA-treated cells (C, D),augmented autophagic structures (arrows) and a reduced number of mitochondria (M) were detected. (D) Micrograph of theboxed area of C at a higher magnification showing autophagosomes. (E) Micrograph of control, untreated ESS-1 cells. N = nucleus;CHR = condensed chromatin

They belong to different structural classes and show abroad range of anti-tumour activity involving apop-tosis, cell cycle arrest, and terminal differentiationof cultured tumour cells [31,32]. Different HDACinhibitors revealed promising results during stage I/IIclinical trials [33,34]. One of the most promising com-pounds is suberoylanilide hydroxamic acid (SAHA), ahydroxamate-based HDAC inhibitor [35], which hasrecently been approved by the FDA for cutaneousT-cell lymphoma therapy [36]. This drug is avail-able under the generic name Vorinostat (Merck & Co,Inc, NJ, USA) and is orally administered at a dailydose of 300–400 mg. In comparison to other HDACinhibitors, SAHA shows good bioavailability and well-known molecular mechanisms of action. Up to now,no studies testing the effects of SAHA in the therapyof gynaecological malignancies exist.

Here we have shown that SAHA inhibits the G1/Stransition in ESS-1 cells and specifically kills tumourcells in a concentration- and time-dependent manner.It is interesting that, in contrast to data published byother authors [37–42] using different cell models, in

our ESS-1 model SAHA does not cause apoptosis.Although it is known that tumour cells sometimeshave a defective apoptotic pathway, our experimentswith doxorubicine showed that ESS-1 cells are able toundergo apoptosis. These findings indicate that SAHA,although it inhibits cell growth and induces cell deathin a broad range of tumour cells, affects different regu-latory mechanisms in different cell models. AlthoughSAHA generally acts as an inhibitor of class I andII HDACs, in our system it especially influences theexpression of HDAC7, with little effect on the expres-sion of other HDACs. Similar findings were recentlypublished by Dokmanovic et al for other cell models[43]. It is, however, interesting that although HDAC7expression is affected both in malignant and in non-malignant cells, the ultimate effects of SAHA differ.This indicates that the final SAHA effects depend onthe interplay between different HDACs, rather thanon deregulation of a single HDAC species. The elu-cidation of the exact role of HDAC7 in different cellmodels and of its influence on the expression of dif-ferent regulatory proteins requires further studies.

Figure 6. Expression of mTOR, a main modulator of autophagy,is decreased in ESS-1 cells upon SAHA treatment. Both mTORand phospho mTOR (p-mTOR) are decreased in SAHA-treatedESS-1 cells in a concentration-dependent manner (A, B). Usingan antibody against the phosphorylated form of S6 ribosomalprotein (S6rp), we showed that the phosphorylation atSer235/236 was significantly decreased upon SAHA treatment.However, the S6rp phosphorylation was less affected bySAHA than by rapamycin. mTOR mRNA was not affectedby the SAHA treatment, suggesting that SAHA modulates thestability and/or the degradation mechanisms of mTOR at theprotein level (C). These data were confirmed by quantitativeRT-PCR on a Taq-Man basis, showing no significant differencesin mTOR mRNA synthesis in untreated and SAHA-treatedESS-1 cells at different time points (data not shown). 1, 4 and7 = untreated cells; 2, 5 and 8 = SAHA (3 µM)-treated cells; 3,6 and 9 = rapamycin (100 nM)-treated cells as a positive control

Autophagy is one of the main mechanisms respon-sible for the degradation of cellular organelles andlong-lived proteins. It also induces cell death that iscontrolled by processes different from those involvedin apoptosis and is therefore described as type IIprogrammed cell death [25,44,45]. According to ourresults, autophagic processes seem to be the main rea-son for cell death caused by SAHA in ESS-1 cells.These results are consistent with data published byShao et al, who showed that in HeLa cells, SAHA caninduce autophagic cell death independent of caspaseactivation [46]. Moreover, this is additionally sup-ported by our observation that in ESS-1 cells, SAHAtreatment decreased expression and phosphorylation ofmTOR. The serine/threonine kinase mTOR is one ofthe key molecules in the autophagic pathway. Fewexisting data show that also in other systems, mTOR

expression can be influenced by SAHA treatment [47].In ESS-1 cells, the down-regulation of mTOR bySAHA is connected with decreased cell proliferation.Furthermore, the phosphorylation of the S6 ribosomalprotein (S6rp) in ESS-1 cells is also strongly influ-enced by SAHA. Our results are supported by datashowing that S6rp phosphorylation is directly involvedin the control of autophagic processes [48]. mTORcontrols cell growth through the phosphorylation ofp70S6K kinase, which phosphorylates S6rp. S6rp thenregulates the synthesis of other proteins which controlthe cell cycle progression and G1/S transition [49,50].In this context, SAHA-induced inhibition of mTORand S6rp phosphorylation (Ser235/236) induces G1arrest and autophagy in ESS-1 cells but not in HESCs.Hence, further investigations are needed to elucidatethe exact pharmacological mechanisms of SAHA indifferent experimental systems and to answer the ques-tion of whether SAHA modulates the expression andphosphorylation of mTOR directly or acts somewhereupstream of mTOR.

In conclusion, ESS-1 tumour cells are more sensitiveagainst SAHA than non-malignant human endometrialstromal cells. SAHA significantly inhibits the prolifer-ation of ESS-1 cells by inducing cell cycle arrest andactivating autophagic mechanisms in a concentrationrange which resembles that used in pre-clinical andclinical therapeutic trials involving other malignancies.Our results indicate that up-regulation of autophagymay have therapeutic benefit in uterine sarcomas ingeneral and endometrial stromal sarcomas in particu-lar. This makes SAHA a promising candidate for thetreatment of endometrial stromal sarcomas. Ongoinginvestigations are currently being performed in ourlaboratory in order to determine the in vivo efficacy ofSAHA treatment for ESS and other uterine sarcomas.

Acknowledgements

This work was supported by Lore Saldow Research Fund.All experiments were done within the Lore Saldow ResearchUnit ‘Molecular Pathology of Gynecologic Tumors’. We thankMarkus Absenger from the Core Facility for CF Microscopy(Center for Medical Research — ZMF, Medical UniversityGraz) and Andrea Koschell from the Laboratory for ElectronMicroscopy, Institute of Pathology, for excellent technicalassistance. This paper is dedicated to the memory of Mrs LoreSaldow.

Supporting information

Supporting information may be found in the onlineversion of this article.

References

1. Lax SF. Molecular genetic changes in epithelial, stromal andmixed neoplasms of the endometrium. Pathology 2007;39:46–54.

2. Goff BA, Rice LW, Fleischhacker D, Muntz HG, Falkenberry SS,Nikrui N, et al. Uterine leiomyosarcoma and endometrial stromal

sarcoma: lymph nodetastases and sites of recurrence. GynecolOncol 1993;50:105–109.

3. Michalas S, Creatsas G, Deligeoroglou E, Markaki S. High-gradeendometrial stromal sarcoma in a 16-year-old girl. Gynecol Oncol1994;54:95–98.

4. Inayama Y, Shoji A, Odagiri S, Hirahara F, Ito T, Kawano N,et al. Detection of pulmonary metastasis of low-grade endometrialstromal sarcoma 25 years after hysterectomy. Pathol Res Pract2000;196:129–134.

5. Date I, Yagyu Y, Bukeo T. Endometrial stromal sarcomametastatic to the skull — case report. Neurol Med Chir1986;26:571–574.

6. Matsuura Y, Yasunaga K, Kuroki H, Inagaki H, Kashimura M.Low-grade endometrial stromal sarcoma recurring with multiplebone and lung metastases: report of a case. Gynecol Oncol2004;92:995–998.

7. Al-Salam S, El-Terifi H, Ghazal-Aswad S. Low-grade endometrialstromal sarcoma with sex cord-like differentiation metastatic to thethoracic spines. APMIS 2006;114:651–655.

8. Ashraf-Ganjoei T, Behtash N, Shariat M, Mosavi A. Low gradeendometrial stromal sarcoma of uterine corpus, a clinico-pathological and survey study in 14 cases. World J Surg Oncol2006;4:1–5.

9. Bodner K, Bodner-Adler B, Obermair A, Windbichler G, Petru E,Mayerhofer S, et al. Prognostic parameters in endometrial stromalsarcoma: a clinicopathologic study in 31 patients. Gynecol Oncol2001;81:160–165.

10. Haberal A, Kayikcioglu F, Boran N, Caliskan E, Ozgul N,Kose MF. Endometrial stromal sarcoma of the uterus: analysis of25 patients. Eur J Obstet Gynecol Reprod Biol 2003;109:209–213.

11. Richon VM, Webb Y, Merger R, Sheppard T, Jursic B, Ngo L,et al. Second generation hybrid polar compounds are potentinducers of transformed cell differentiation. Proc Natl Acad SciU S A 1996;93:5705–5708.

12. Marks PA, Richon VM, Rifkind RA. Histone deacetylaseinhibitors: inducers of differentiation or apoptosis of transformedcells. J Natl Cancer Inst 2000;92:1210–1216.

13. Davie JR. Inhibition of histone deacetylase activity by butyrate. JNutr 2003;133:2485–2493.

14. Hrzenjak A, Moinfar F, Kremser ML, Strohmeier B, Staber PB,Zatloukal K, et al. Valproate inhibition of histone deacetylase 2affects differentiation and decreases proliferation of endometrialstromal sarcoma cells. Mol Cancer Ther 2006;5:2203–2210.

15. Richon VM, Sandhoff TW, Rifkind RA, Marks PA. Histonedeacetylase inhibitor selectively induces p21WAF1 expression andgene-associated histone acetylation. Proc Natl Acad Sci U S A2000;97:10014–10019.

16. Benjamin D, Jost JP. Reversal of methylation-mediated repressionwith short-chain fatty acids: evidence for an additional mechanismto histone deacetylation. Nucleic Acids Res 2001;29:3603–3610.

17. Finnin MS, Donigian JR, Cohen A, Richon VM, Rifkind RA,Marks PA, et al. Structures of a histone deacetylase homologuebound to the TSA and SAHA inhibitors. Nature 1999;401:188–193.

18. Kelly WK, Richon VM, O’Connor O, Curley T, MacGregor-Curtelli B, Tong W, et al. Phase I clinical trial of histone deacety-lase inhibitor: suberoylanilide hydroxamic acid administered intra-venously. Clin Cancer Res 2003;9:3578–3588.

19. Kelly WK, O’Connor OA, Krug ML, Chiao JH, Heaney M,Curley T, et al. Phase I study of an oral histone deacetylaseinhibitor, suberoylanilide hydroxamic acid, in patients withadvanced cancer. J Clin Oncol 2005;23:3923–3931.

20. Spiller SE, Ravanpay AC, Hahn AW, Olson JM. Suberoylanilidehydroxamic acid is effective in preclinical studies of medulloblas-toma. J Neurooncol 2006;79:259–270.

21. O’Connor OA, Heaney ML, Schwartz L, Richardson S, Willim R,MacGregor-Cortelli B, et al. Clinical experience with intravenousand oral formulation of the novel histone deacetylase inhibitorsuberoylanilide hydroxamic acid in patients with advancedhematologic malignancies. J Clin Oncol 2006;24:166–172.

22. Gunawan B, Braun S, Cortes MJ, Bergmann F, Karl C, Fuzesi L.Characterization of a newly established endometrial stromalsarcoma cell line. Int J Cancer 1998;77:424–428.

23. Krikun G, Mor G, Alvero A, Guller S, Schatz F, Sapi E,et al. A novel immortalized human endometrial stromalcell line with normal progestational response. Endocrinology2004;145:2291–2296.

24. Biederbick A, Kern HF, Elsasser HP. Monodansylcadaverine(MDC) is a specific in vivo marker for autophagic vacuoles. EurJ Cell Biol 1995;66:3–14.

25. Yang YP, Liang ZQ, Gu ZL, Qin ZH. Molecular mecha-nism and regulation of autophagy. Acta Pharmacol Sinica2005;26:1421–1434.

26. Mansi JL, Ramachandra S, Wiltshaw E, Fisher C. Endometrialstromal sarcomas. Gynecol Oncol 1990;36:113–118.

27. Pink D, Lindner T, Mrozek A, Kretzschmar A, Thuss-PatiencePC, Dorken B, et al. Harm or benefit of hormonal treatment inmetastatic low-grade endometrial stromal sarcoma: single centerexperience with 10 cases and review of the literature. GynecolOncol 2006;101:464–469.

28. Amant F, De Knijf A, Van Calster B, Leunen K, Neven P,Berteloot P, et al. Clinical study investigating the role oflymphadenectomy, surgical castration and adjuvant hormonaltreatment in endometrial stromal sarcoma. Br J Cancer2007;97:1194–1199.

29. Lehrmann H, Pritchard LL, Harel-Bellan A. Histone acetyltrans-ferases and deacetylases in the control of cell proliferation anddifferentiation. Adv Cancer Res 2002;86:41–65.

30. Minucci S, Pelicci PG. Histone deacetylase inhibitors and thepromise of epigenetic (and more) treatments for cancer. NatureRev Cancer 2006;6:38–51.

31. Marks PA, Richon VM, Miller T, Kelly WK. Histone deacetylaseinhibitors. Adv Cancer Res 2004;91:137–168.

32. Xu WS, Parmigiani RB, Marks PA. Histone deacetylase inhibitors:molecular mechanisms of action. Oncogene 2007;26:5541–5552.

33. Gabrielli BG, Johnstone RW, Saunders NA. Identifying moleculartargets mediating the anticancer activity of histone deacetylaseinhibitors: a work in progress. Curr Cancer Drug Targets2002;2:337–353.

34. Marks PA, Richon VM, Breslow R, Rifkind RA. Histone deacety-lase inhibitors as new cancer drugs. Curr Opin Oncol2001;13:477–483.

35. Richon VM, Emiliani S, Verdin E, Webb Y, Breslow R,Rifking RA, et al. A class of hybrid polar inducers of transformedcell differentiation inhibits histone deacetylases. Proc Natl AcadSci U S A 1998;95:3003–3007.

36. Marks PA, Breslow R. Dimethyl sulfoxide to Vorinostat: develop-ment of this histone deacetylase inhibitor as an anticancer drug.Nature Biotechnol 2007;25:84–90.

37. Gillenwater AM, Zhong M, Lotan R. Histone deacetylase inhibitorsuberoylanilide hydroxamic acid induces apoptosis through bothmitochondrial and Fas (Cd95) signaling in head and necksquamous carcinoma cells. Mol Cancer Ther 2007;6:2967–2975.

38. Fedier A, Dedes KJ, Imesch P, Von Bueren AO, Fink D.The histone deacetylase inhibitors suberoylanilide hydroxamic(Vorinostat) and valproic acid induce irreversible and MDR1-independent resistance in human colon cancer cells. Int J Oncol2007;31:633–641.

39. Tong A, Zhang H, Li Z, Gou L, Wang Z, Wei H, et al. Proteomicanalysis of liver cancer cells treated with suberonylanilide hydrox-amic acid. Cancer Chemother Pharmacol 2008;61:791–802.

40. Kumagai T, Wakimoto N, Yin D, Gery S, Kawamata N, Takai N,et al. Histone deacetylase inhibitor, suberoylanilide hydroxamicacid (Vorinostat, SAHA), profoundly inhibits the growth of humanpancreatic cancer cells. Int J Cancer 2007;121:656–665.

41. Emanuele S, Lauricella M, Carlisi D, Vassallo B, D’Anneo A,Di Fazio P, et al. SAHA induces apoptosis in hepatoma cells andsynergistically interacts with the proteasome inhibitor Bortezomib.Apoptosis 2007;12:1327–1338.

42. Jiang X, Tsang YH, Yu Q. c-Myc overexpression sensitizes Bim-mediated Bax activation for apoptosis induced by histonedeacetylase inhibitor suberoylanilide hydroxamic acid (SAHA)

through regulating Bcl-2/Bcl-xL expression. Int J Biochem CellBiol 2007;39:1016–1025.

43. Dokmanovic M, Perez G, Xu W, Ngo L, Clarke C, ParmigianiRB, et al. Histone deacetylase inhibitors selectively suppressexpression of HDAC7. Mol Cancer Ther 2007;6:2525–2534.

44. Bursch W, Ellinger A, Gerner C, Schulte-Hermann R. Autophago-cytosis and programmed cell death. In Autophagy, Klion-sky DJ (ed). Landes Bioscience: Georgetown, TX, 2004;287–303.

45. Mizushima N, Levine B, Cuervo AM, Klionsky DJ. Autophagyfights disease through cellular self-digestion. Nature2008;451:1069–1075.

46. Shao Y, Gao Z, Marks PA, Jiang X. Apoptotic and autophagiccell death induced by histone deacetylase inhibitors. PNAS2004;101:18030–18035.

47. Kawamata N, Chen J, Koeffler HP. Suberoylanilide hydroxamicacid (SAHA, Vorinostat) suppresses translation of cyclin D1 inmantle cell lymphoma cells. Blood 2007;110:2667–2673.

48. Blommaart EF, Luiken JJ, Blommaart PJ, van Woerkom GM,Meijer AJ. Phosphorylation of ribosomal protein S6 is inhibitoryfor autophagy in isolated rat hepatocytes. J Biol Chem1995;270:2320–2326.

49. Bjornsti M, Houghton PJ. The TOR pathway: a target for cancertherapy. Nature Rev Cancer 2004;4:335–348.

50. Iwenofu OH, Lackman RD, Staddon AP, Goodwin DG, HauptHM, Brooks JSJ. Phospho-S6 ribosomal protein: a potential newpredictive sarcoma marker for targeted mTOR pathway. ModPathol 2008;21:231–237.

SAHA induces caspase-independent, autophagic cell death of endometrial stromal sarcoma cells by...

Documents

FIP-gts Potentiate Autophagic Cell Death Against Cisplatin

Glutamate transporter-dependent mTOR phosphorylation in Müller glia cells

Apoptotic and autophagic responses to Bcl-2 inhibition and photodamage

Roles of the PI3K/AKT/mTOR signalling pathways in

Saha Health And Herbal LLP - IndiaMART

NIT KUMAR SAHA - (1922-1991) Elected Fellow 1959 - Indian

mindset dan pengus un n peril saha ja

Chapter 12 Monitoring Autophagic Degradation of p62/SQSTM1

SAHA: sistema para acompanhamento holístico de atletas

Acid sphingomyelinase-dependent autophagic degradation of

mTOR regulates phagosome and entotic vacuole fission

Das Ritual saHa kA sHn.t als Tempelfest des Gottes Min

Plasticity and mTOR: towards restoration of impaired synaptic plasticity in mTOR-related neurogenetic disorders

TBK1 regulates prostate cancer dormancy through mTOR inhibition

Metabolic clock generates nutrient anticipation rhythms in mTOR signaling

autophagic vacuoles in human red cells - NCBI

PROSPEK TUS PENA W AR AN UMUM PERD ANA SAHA M

Book - MEGHNAD SAHA: A GREAT SCIENTIST & VISIONARY

2-Hydroxyglutarate Inhibits ATP Synthase and mTOR Signaling

Gopal B. Saha - Fundamentals of Nuclear Pharmacy