Depsipeptide (FR901228) Inhibits Proliferation and Induces Apoptosis in Primary and Metastatic Human Uveal Melanoma Cell Lines

Abstract

purpose. Uveal melanoma (UM) is the most common primary malignant ocular tumor in adults. No effective chemotherapy regimens are available for either intraocular or metastatic uveal melanoma. Therefore, the ability of the histone deacetylase inhibitors (HDACIs), depsipeptide, sodium butyrate (NaB) and trichostatin A (TSA), to induce apoptosis and inhibit cell growth of UM cell lines in vitro was examined.

methods. Three primary and two metastatic UM cell lines were treated in vitro with different concentrations of histone deacetylase inhibitors (HDACIs). Cell proliferation was studied in 24-well plates. Induction of apoptosis was studied by flow cytometry. Changes in gene expression of Fas/FasL, p21^Waf/Cip1, and p27^Kip1 were studied by RT-PCR. Western blot analysis was used to study histone acetylation, Fas/FasL, p21^Waf/Cip1, p27^Kip1 and caspase-3 protein levels. Real-time PCR was used to study changes in bcl-2/bax gene expression.

results. A dose-dependent increase in histone acetylation was observed in all cell lines. This corresponded to significant inhibition of cell growth and induction of apoptosis in all melanoma cell lines in a concentration-dependent manner. Western blot analysis revealed dose-dependent increases in the amount of caspase-3, Fas/FasL, p21^Waf/Cip1, and p27^Kip1 proteins. However, no changes in bcl-2/bax gene expression were detected by real-time PCR.

conclusions. HDACIs are potent inhibitors of primary and metastatic UM cell growth in vitro. The apoptosis is probably mediated through the Fas/FasL signaling pathway, whereas bcl-2 appears not to be involved. These data support further clinical evaluation of depsipeptide and other HDACIs in patients with primary and metastatic UM.

Uveal melanoma (UM) is the most common primary intraocular malignant tumor in adults. It has a high mortality rate due to hematogenous dissemination preferentially to the liver. ^{1 2} Metastatic disease is universally fatal, usually within 6 months from the time diagnosis is made. UM has been found to be highly resistant to various forms of chemotherapy used to date. ³ Consequently, no effective chemotherapy regimens are currently available for either intraocular or metastatic UM. New therapeutic strategies that would result in a significant disease response and an increase in survival rate are therefore much needed. 

In eukaryotic cells posttranslational acetylation of lysine residues of core nucleosomal histones provides a molecular mechanism by which DNA can be rendered accessible to transcription factors, while maintaining a nucleosomal architecture. Histone acetylation is mediated by the histone acetyltransferase (HAT) activity and is often associated with transcriptional activation of genes that regulate cell cycle progression, DNA replication and apoptotic response to DNA damage. ⁴ In contrast, histone deacetylation, catalyzed by histone deacetylase (HDAC), is often associated with transcriptional silencing. Aberrant levels of HDAC activity have been found in a variety of human malignancies and result in repression of tumor-suppressor genes and promotion of tumorigenesis. Little is known, however, about the role of chromatin remodeling and the possible role(s) of HDACs and HAT in the pathogenesis of cutaneous and uveal melanoma. ⁵ In recent years, HDAC inhibitors (HDACIs) like sodium butyrate (NaB), trichostatin A (TSA), apicidin, depsipeptide (FR901228), trapoxin, oxamflatin, and m-carboxycinnamic acid bishydroxamide (CBHA) have been shown to inhibit cell proliferation, induce apoptosis and differentiation in a variety of human malignant cell lines. ^{5 6} These compounds differ, however, in their potency at inhibiting various HDACs and to induce the above-mentioned biological phenomena in eukaryotic cells. 

Depsipeptide, a very potent HDACI, is a natural tetrapeptide that was isolated from Chromobacterium violaceum No. 968. It is characterized by a bicyclic structure composed of four amino acids (d-valine, d-cysteine, dehydrobutyrine, and l-valine) and a novel acid (3-hydroxy-7-mercapto-4 heptanoic acid). ^{7 8} In previously published studies, HDACIs were shown to induce inhibition of HDAC activity and increase expression of a limited number of genes, including p21^Waf/Cip1, ^{5 6 9 10 11 12} p27^Kip1, ¹² c-myc, ¹³ plasminogen activator, ¹⁴ and gelsolin ¹⁵ in numerous cancer cell lines. 

Apoptosis is a highly conserved mechanism of programmed cell death, in which caspase enzymes play a crucial role. Those enzymes are involved in initial signaling events and in downstream proteolytic cleavages that characterize the apoptotic phenotype. ¹⁶ Extrinsic or intrinsic signaling pathway(s) can activate the apoptotic process. In extrinsic pathways, activation of one of the death receptor signaling pathways (Fas/FasL, TRAIL etc.) triggers one of the upstream signaling caspases (e.g., caspase-8 or -10) that in turn activates one of the effector caspases (e.g., caspase 3). ¹⁷ On the other hand, the intrinsic pathway, involves the disruption of the mitochondrial membrane, leakage of cytochrome c from the mitochondria and activation of caspase-9 that in turn activates one of the effector caspases. ¹⁶  

In this article we provide the first evidence that depsipeptide, NaB, and TSA strongly inhibit cell growth and induce apoptosis in primary and metastatic UM cell lines in vitro. In addition, we characterized some of the signaling mechanisms that mediate those biological effects. The ultimate goal of our study was to provide the rationale for the use of HDACIs as a new therapeutic modality in the treatment of primary and metastatic UM in humans. 

Materials and Methods

Cell Culture

Five different UM cell lines, (kind gift of Mary Hendrix, University of Iowa, Ames, IA; and June Kan-Mitchel, Wayne State University, Detroit, MI) that were isolated from primary (M619, C918, and OCM-1) or metastatic UM lesions—that is, liver metastasis (MUM-2b and -2c)—were used. These cell lines were categorized based on their proliferative and invasive potentials as low-level (OCM-1 and MUM-2c) or high-level invaders (M619, C918, and MUM-2b). ^{18 19}  

Depsipeptide, NaB, and TSA were used in cell proliferation and cell cycle analysis on all five UM cell lines. Because depsipeptide is one of the most potent HDACI available today, all the subsequent experiments were performed by treating all five cell lines with depsipeptide only. 

Cells were grown in RPMI medium (Gibco BRL-Life Technologies, Rockville, MD) with the addition of 10% heat-inactivated fetal bovine serum (FBS) and standard concentrations of streptomycin and penicillin. Cells were grown at 37°C with 5% CO₂ until 75% to 90% confluent. 

Histone Isolation and Analysis

All five melanoma cell lines were incubated in RPMI at various doses of depsipeptide for 24 hours. Cells were harvested and lysed by 20 strokes in a homogenizer (Dounce; Bellco Glass Co., Vineland, NJ) in 5 mL of buffer (10 mM Tris-HCl [pH 7.5], 1.5 mM MgCl₂, 1.0 mM CaCl₂, 0.25 M sucrose, 0.5% Triton X-100, 2.0 mM ZnSO₄, 0.2 mM phenylmethylsulfonyl fluoride [PMSF] and 1 mM benzamidine). Nuclei were collected by centrifugation at 2000g for 5 minutes and then washed twice with 5 mL of buffer without Triton X-100. Histones were then extracted by resuspending the nuclear pellet in 1 mL of 0.4 N H₂SO₄. After a 1-hour incubation at 4°C, insoluble proteins were removed by centrifugation (10,000g for 10 minutes). Histones were precipitated by the addition of trichloroacetic acid to the supernatant, to a final concentration of 20%, and incubated for 30 minutes at 4°C. The histones were collected by centrifugation at 10,000g for 10 minutes, washed twice with cold acetone (−20°C) and resuspended in 0.5 mL water. Acetylated histones H3 and H4 were analyzed by Western blot SDS-PAGE on 18% polyacrylamide gels. Gels were either stained with Coomassie blue to compare total protein levels or blotted to nitrocellulose membrane for Western blot analysis. Western blots were incubated with either 50 μg/mL of rabbit anti-human acetylated H3 (Upstate Biotechnology, Lake Placid, NY) or 2 μg/mL of rabbit anti-human acetylated H4 (Upstate Biotechnology). All cell lines were also probed with histone H4 acetylation site specific antibodies (Serotec, Raleigh, NC, and Upstate Biotechnology). 

Proliferation Assay

Attached UM cells were washed with warm PBS (Ca²⁺ and Mg²⁺ free) (Gibco BRL-Life Technologies) and detached by incubating with 2 mM sodium-EDTA in PBS (Ca²⁺ and Mg²⁺ free) for 10 minutes at 37°C. Cells were collected by centrifugation at 1500 rpm for 5 minutes. Proliferation assay was performed in triplicate in 24-well plates using colorimetric MTT-based kit (CellTiter 96 Aqueous One; Promega, Madison, WI). Cells (7500/well) were plated in complete medium, allowed to adhere, and serum starved for 8 hours before NaB (ICN Biomedicals, Irvine, CA; 0, 0.1, 1, 2, 4, and 8 mM), TSA (0, 0.001, 0.01, 0.1, 1, and 2 μM; Sigma-Aldrich, St. Louis, MO), or depsipeptide (0, 0.1, 1, 5, 10 and 100 nM) was added. Plates were incubated for 48 hours at 37°C before chromogen was added and then read on a colorimetric plate reader (HTS-7000; Perkin Elmer, Wellesley, MA) at 492 nm. 

Cell Cycle Analysis

To analyze an effect of HDACIs on cell cycle progression and apoptosis, we incubated all five cell lines with each concentration of NaB, TSA, and depsipeptide for 48 hours. DNA content profile of a given population was determined by flow cytometry. ²⁰ Treated cells were fixed with 100% ethanol overnight and treated with 0.25 μg/mL RNase (Sigma-Aldrich). Nuclei were stained with 50 μg/mL propidium iodide (Sigma-Aldrich). Cells were analyzed with a standard flow cytometer (model XL; Beckman Coulter, Hialeah, FL) with system-2 software. 

Preparation of Cell Extracts and Western Blot Analysis

Cells were incubated with various concentrations of depsipeptide for 24 hours, collected as described earlier, and homogenized in a lysis buffer (20 mM Tris, 160 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 0.1% SDS, 0.25% sodium deoxycholate, 1 mM PMSF, 1 mM NaF, 1 mM dithiothreitol (DTT), 1 mM sodium orthovanadate, pepstatin, leupeptin, and aprotinin) on ice. Cell lysate was centrifuged at 14,000g for 5 minutes at 4°C and divided into aliquots that were stored at −80°C. Protein concentration was determined by bicinchoninic acid protein assay reagent (Micro BCA; Pierce Chemical Co., Rockford, IL). For the determination of apoptotic proteins, cell lysates containing 50 μg of total protein were subjected to SDS-PAGE on 10% to 12% slab gels. The samples were then electrophoretically transferred to PVDF transfer membrane (Hybond-P; Amersham Biosciences Inc., Arlington Heights, IL) at 100 mV for 1 hour. Membranes were blocked 1 hour at room temperature (RT) with 5% dry milk in PBS with 0.1% Tween (PBS-T) and incubated overnight at 4°C with anti- p21^Waf/Cip1, (1:1000), anti- p27^Kip1 (1:1000), anti-caspase-3 (1:1000), anti-caspase 9 (1:1000), anti-PARP (1:1000), anti-Fas (1:500), anti-FasL (1:200), anti-Bcl-2 (1:1000), anti-Bax (1:1000), and anti-actin (1:1000). All primary antibodies were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Membranes were washed and incubated with 1:5000 anti-mouse or anti-rabbit IgG conjugated to horse radish peroxidase (Amersham Biosciences Inc.) for 60 minutes at RT and washed again. Bound antibodies were detected with a Western blot analysis detection system (ECL Plus; Amersham Biosciences Inc.). Actin served as a loading control. 

Caspase-3 Assay

Cells were grown as previously described and were subjected to depsipeptide treatment for 24 hours. Activity of caspase-3 in cell lysates (200 μg of protein) was measured by using caspase-3/CPP32 colorimetric proteases assay kit using DEVD-pNA as a labeled substrate, according to the manufacturer’s instructions (MBL, Nagoya, Japan). 

RNA Isolation and cDNA Synthesis

After the cells were treated with depsipeptide for 24 hours, medium was removed, and floating cells were collected by spinning at 3000 rpm for 5 minutes. Attached cells were washed with warm PBS (Ca²⁺ and Mg²⁺ free), detached by incubation with 2 mM sodium-EDTA in PBS (Ca²⁺ and Mg²⁺ free) for 10 minutes at 37°C, and collected by centrifugation at 1500 rpm for 5 minutes. Cells were resuspended in extraction reagent (Gibco BRL-Life Technologies) and snap frozen in liquid nitrogen to inhibit activity of endogenous RNases. All cells were disrupted in a tissue homogenizer (10 seconds on ice), and total RNA was isolated by phenol-chloroform extraction (Gibco BRL-Life Technologies) and precipitated with absolute ethanol. RNA samples were subsequently treated for 45 minutes with RNase-free DNase at 37°C (Message Clean; GeneHunter Corp., Nashville, TN), phenol-chloroform extracted, ethanol precipitated, and recovered in DEPC-treated water. RNA concentration was determined by spectrophotometric readings at 260 and 280 nm, respectively. 

Reverse Transcription-Polymerase Chain Reaction

For RT-PCR, a preamplification system for first-strand cDNA synthesis (SuperScript) was used as suggested by the manufacturer (Gibco BRL-Life Technologies). Random hexamer primers and 1 μg of total RNA were used in the RT step. The following primers were used for the PCR step: p21^Waf/Cip1 (279 bp) sense, 5′-GAA CTT CGA CTT TGT CAC CGA G-3′, and antisense, 5′-CGT TTT CGA CCC TGA GAG TCT C-3′; p27^Kip1 (164 bp) sense, 5′ AAC GTG CGA GTG TCT AAC GG-3′, and antisense, 5′-CTT CCA TGT CTC TGC AGT GC-3′; Fas (117 bp) sense, 5′-AAT TCT GCC ATA AGC CCT GTC CT-3′, and antisense, 5′-GGG CTT TGT CTG TGT ACT CCT TCC-3′; FasL (559 bp) sense, 5′-CGC CGC CGC CAC CAC TG-3′, and antisense 5′-CTG CTG CGG GCC CAC ATC TGC-3′. From each RNA sample, three separate cDNA samples were synthesized, and separate PCR reactions were performed for each gene. Short sequence of c-abl gene (201 bp) was coamplified in all PCR tubes as a positive internal control. The following primers were used: sense, 5′-TTC AGC GGC CAG TAG CAT CTG ACT T-3′, and antisense, 5′-TGT GAT TAT AGC CTA AGA CCC GGA G-3′. 

Quantification of Bcl-2 and Bax Gene Expression by Real-Time RT-PCR

Total cellular RNA and cDNAs were prepared as previously described. ²¹ For this analysis, samples were thawed and run in batches to eliminate variation as a consequence of personnel, cDNA preparation, or change in the composition of reagents. Each cDNA sample was used as a template in a PCR amplification reaction run in duplicate on a sequence detection system (Prism 7700; Applied Biosystems, Foster City, CA). Quantification of Bcl-2 and Bax copy number in our samples was obtained by comparing the copy number in each sample against a Bcl-2 and Bax standard curves. Copy numbers of Bcl-2 and Bax were normalized by 18S rRNA that was used as an internal control. 

The results of the real-time RT-PCR assay for each sample were reported as specific copy numbers of transcripts per nanogram of RNA. The real-time RT-PCR assay was validated by amplifying limiting dilutions of Bcl-2 and Bax plasmid standards. Linearity was observed over the concentration range of 10 to 10⁶ bcl-2 and bax copies/mL of cDNA, with the linear regression coefficients (r ²) of ≥0.98 achieved routinely. 

Results

Histone Acetylation

The ability of depsipeptide to induce histone acetylation was assessed in all five melanoma cell lines. The baseline low level of H3 and H4 acetylation detected in all cell lines significantly increased after 24-hour exposure to 5 nM depsipeptide. No additional increase in histone acetylation was observed with 10 nM treatment (Fig. 1) . When compared with no treatment, 5 nM of depsipeptide resulted in a significant increase in acetylation of lysine residues 5, 8, and 12. Depsipeptide treatment, however, did not result in an increase in acetylation of lysine residue 16 in any of the five UM cell lines. These findings strongly suggest that depsipeptide-mediated inhibition of HDAC induced a specific pattern of H4 lysine residue acetylation, which does not appear to be cell type dependent. 

Proliferation Assay

A dose-dependent decrease in cell proliferation was observed in all five UM cell lines after 48 hours of incubation with each HDACI (Fig. 2) . Each cell line showed somewhat different responses to each HDACI. Treatment with 0.1 mM NaB inhibited cell proliferation by approximately 40% to 50% in two highly invasive primary cell lines (C918 and M619) and both metastatic cell lines (MUM-2b and -2c). Fifty percent to 70% inhibition of cell proliferation was observed in all cell lines with higher concentrations of NaB (2–8 mM). In the 0.001- to 0.1-μM concentration range, TSA inhibited proliferation of primary UM cell lines by from 5% to 15%. Up to 30% inhibition was observed in metastatic cell lines in the same concentration range. The highest concentration of TSA tested in the study (2 μM) resulted in an average 70% (M619) to 90% (OCM-1 and MUM-2c) reduction of cell proliferation. In two primary cell lines (OCM-1 and M619) and one metastatic (MUM-2c) cell line, 1 nM depsipeptide inhibited proliferation by approximately 10%. In the MUM-2b cell line, 1 nM inhibited proliferation by 62%. In all cell types (except MUM-2c) maximum inhibition of cell proliferation (≈50%–70%) seemed to be achieved with 5 nM depsipeptide treatment. In those cell lines 10 and 100 nM depsipeptide treatment resulted in minimal or no increase in inhibition of cell proliferation. 

Cell Cycle Analysis

We used fluorescence-activated cell sorter analysis of propidium iodide-stained nuclei to assess the cells for the presence of the hypodiploid or sub-G₁ fraction resulting from DNA fragmentation that is indicative of apoptosis. All five cell lines were analyzed. Data for the MUM-2b cell line are shown in Figure 3 . Increasing concentrations of depsipeptide led to an incremental accumulation of the sub-G₁ fraction (apoptotic cells). The sub-G₁ fraction increased from 12.9% in untreated cells (Fig. 3A) to 49.2% in cells treated with 100 nM of depsipeptide (Fig. 3F) . Similar results were obtained with other cell lines. In all cell lines, 5 nM treatment resulted in a doubling of the proportion of apoptotic cells when compared with nontreated cells. 

Caspase-3 Activation

Western blot and ELISA-based colorimetric assays were used to confirm the effect of depsipeptide on the activation of apoptotic process in MUM-2b cells by measuring the activity of caspase-3 which is known to be the main effector caspase in most mammalian cells. As shown in Figure 4 , depsipeptide treatment resulted in the activation of caspase-3, as evidenced by conversion of the proenzyme form of caspase-3 (p32) to the catalytically active effector protease (p17). Activation of caspase-3 was also confirmed by examining poly(ADP-ribose) polymerase (PARP), a known endogenous substrate for caspase-3. Caspase-3 protease activation was accompanied by the cleavage of PARP (116 kDa) into an 85-kDa C-terminal fragment and a 28-kDa N-terminal fragment (Fig. 4) . 

Increases in caspase-3 activity associated with depsipeptide treatment were quantified by ELISA and expressed as a relative ratio compared with caspase-3 activity in untreated cells (Fig. 5) . Treatment with 10 or 100 nM resulted in an approximate 5.5-fold induction of caspase-3 activity when compared with the activity in nontreated cells. To assess the possibility that a synthesis of new proteins was involved in depsipeptide-induced apoptosis, melanoma cells were preincubated for 1 hour with 1 μg/mL cycloheximide (protein synthesis inhibitor), followed by addition of depsipeptide for 24 hours. Preincubation with cycloheximide significantly reduced depsipeptide-induced activation of caspase-3 (Fig. 5) , suggesting that de novo protein synthesis was involved in caspase-3 activation and subsequent induction of apoptosis. 

Caspase-3 can be activated by intrinsic or extrinsic pathway(s). One of the major activators of caspase-3 in the intrinsic signaling pathway is caspase-9. Use of Western blot analysis did not detect any change in the activity of caspase-9 in depsipeptide-treated MUM-2b cells (data not shown) suggesting that activation of caspase-3 in depsipeptide-treated UM cells is not mediated through the intrinsic pathway. 

RT-PCR and Western Blot Analysis of Apoptotic Proteins

To gain an insight into the depsipeptide-mediated alteration of gene expression that may be involved in the apoptotic process, we investigated the effects of depsipeptide treatment on p21^Waf/Cip1, p27^Kip1, Fas, and FasL on gene expression in all melanoma cell lines. By using RT-PCR, we observed p21^Waf/Cip1, p27^Kip1 (Fig. 6) , Fas, and FasL (Fig. 7) gene upregulation in all UM cell lines. Western blot analysis confirmed that this gene upregulation correlates with the increase in the protein content in depsipeptide-treated cells. 

p21^Waf/Cip1 has been shown to be one of the substrates for caspase-3. ²² We investigated whether p21^Waf/Cip1 could be cleaved in UM cell lines after depsipeptide treatment and found that depsipeptide treatment not only induced dose-dependent expression of p21^Waf/Cip1 but also dose-dependent cleavage of p21^Waf/Cip1 to a 14-kDa fragment (Fig. 6) . 

Depsipeptide treatment did not result in any changes in Bcl-2 or Bax gene expression (or their ratio), protein levels, or their phosphorylation (data not shown) suggesting that Bcl-2/Bax signaling pathway (part of the intrinsic signaling pathway) is not involved in depsipeptide-induced apoptosis. 

Discussion

HDACIs are emerging as a promising new therapeutic tool for treatment of a variety of human tumors. Some of them, like a depsipeptide, are being incorporated into clinical trials for solid tumors and hematologic malignancies. ^{23 24 25} Our results are the first to show that HDACIs strongly increase histone acetylation and inhibit cell proliferation in primary and metastatic UM cell lines in vitro. In addition, they are potent inducers of apoptosis in those cell lines. Our data also suggest that those two phenomena are mediated, at least in part, through induction of p21^Waf/Cip1, p27^Kip1, and Fas/FasL gene expression. 

Histone acetylation is a highly specific posttranslational modification occurring on ε-amino groups of lysine residues in the N-terminal tails of core histones (H2A, H2B, H3, and H4) located primarily in gene promoters and enhancers. This process is involved in transcriptional regulation, chromatin condensation, cell proliferation, and heterochromatin assembly in eukaryotic cells. ²⁶ Histones H2B, H3, and H4 each have four acetylatable lysines, whereas histone H2A has only two. This results in 6.7 × 10⁷ possible combinations of acetylation per one nucleosome, which translates into an equal number of different functional states of the same. ²⁷ Biological effect(s) of acetylation of each lysine residue on cell proliferation, transcription, or binding of specific transcription factor(s) was extensively studied in yeasts and Drosophila. ²⁸ In those species, acetylation of H4 K5, K8, and K12 seems to be associated with the deposition of newly synthesized histones onto newly replicated DNA, ²⁸ whereas H4 K16 acetylation has been closely linked to transcriptional activation. ^{29 30} Unfortunately, a very limited amount of data exists for the role of acetylation of specific lysine residues in human cells. ³¹  

Cell lines used in this study were isolated from primary choroidal or ciliary body melanoma (OCM-1, M619, and C918) and from metastatic liver lesions (MUM-2b and MUM-2c). In addition, those cell lines were classified as low-level (OCM-1 and MUM-2b) or high-level invaders (M619, C918B, and MUM-2b), based on their invasion properties in vitro. ^{18 19 32 33} All three HDACIs were highly efficient in inhibiting proliferation and in inducing the apoptosis in primary and metastatic cell lines, regardless of their invasion properties. Those results suggest that HDACIs could be used in the clinical setting to treat both primary and metastatic UM. 

The Fas/FasL system has been recognized as the major extrinsic apoptotic pathway in a variety of normal and malignant human cells and tissues. ³⁴ The Fas receptor and its ligand (FasL) are transmembrane proteins of the tumor necrosis factor (TNF) family of receptors and ligands. Engagement of Fas by FasL results in the activation of caspase-8 and/or -10 that activates caspase-3. FasL can act as a signaling molecule in its own right, delivering either proliferative or cytostatic signals to the cell on which it is expressed. ³⁴ Little is know about the function of the Fas/FasL signaling system in uveal melanoma. Presence of Fas and FasL in primary UM was documented by immunohistochemistry. ³⁵ In addition, Repp et al. ³⁶ showed that FasL present on UM cells induces apoptosis in Fas-positive human hepatocytes suggesting the importance of the Fas/FasL system in melanoma-induced liver damage. 

In all five UM cells used in this study, depsipeptide treatment induced gene upregulation of both Fas and FasL, thereby, potentiating their proapoptotic action(s). Similar upregulation of Fas and FasL was reported in human neuroblastoma and acute promyelocytic leukemia cell lines treated in vitro with other HDAC inhibitors such as CBHA and apicidin. ^{6 37} Depsipeptide-induced strong activation of caspase-3 detected in this study was almost completely abolished by pretreatment with cycloheximide, suggesting that de novo protein synthesis is essential for the activation of caspase-3 and initiation of apoptosis. 

Bcl-2 is an antiapoptotic protein that is located in the mitochondrial membrane. By binding and inactivating the apoptotic protein Bax, Bcl-2 blocks release of cytochrome c from the mitochondria and prevents the activation of caspase-9, which would lead to the activation of the intrinsic apoptotic pathway. ¹⁶ Several immunohistochemical studies have shown high expression of Bcl-2 in UM—that is, approximately 60% to 100% ^{38 39 40 41 42} —whereas normal choroidal melanocytes are Bcl-2 negative. The role of the Bcl-2 overexpression in the pathogenesis of UM is unknown at present. Of note, Bcl-2 expression does not correlate with clinical or morphologic features of UM. ^{38 39 40 41 42} In this study, we were unable to detect any changes in gene expression, protein content or phosphorylation status for either Bcl-2 or Bax in depsipeptide-treated UM cells. Furthermore, we were unable to detect any changes in the caspase-9 activity that leads us to believe that the intrinsic apoptotic pathway does not seem to be involved in depsipeptide-mediated apoptosis in UM cells. 

Cell cycle progression is regulated by the sequential activation of cyclin-dependent kinases (CDKs) that are subject to negative regulation by CDK inhibitors (CDKIs), such as p16^INK4, p18, p19, p21^Waf/Cip1, and p27^Kip1. The role of deregulation of cyclins, CDKs and CDKIs in the pathogenesis of uveal melanoma is poorly understood. Alterations of p16 function in UM cell lines includes abolition of p16-CDK4 interaction ⁴³ and promoter hypermethylation. ⁴⁴ Mouriaux et al. ⁴³ documented upregulation of cyclin E and downregulation of p21^Waf/Cip1 and p27^Kip1 in three UM cell lines when compared with normal choroidal melanocytes in vitro. Those results suggest that downregulation of CDKI in UM cell lines could be responsible in part for their malignant phenotype. Our results show that depsipeptide treatment results in dose-dependent upregulation of gene expression and protein content for both p21^Waf/Cip1 and p27^Kip1 in UM cells. p21^Waf/Cip1 and p27^Kip1 molecules are very potent CDKIs that inhibit multiple cyclin-CDK complexes, but seem to preferentially target CDK2. ⁴⁵ Both of those molecules can inhibit the cell cycle in G₁ phase and are able to induce cell death. In vitro upregulation of p21^Waf/Cip1 gene expression and induction of apoptosis was detected in several human malignant cell lines treated with HDACIs, such as cutaneous melanoma, ⁵ bladder carcinoma, ¹⁰ prostate carcinoma, ¹¹ lung carcinoma, ⁹ and acute leukemia. ⁶ Similar increases in expression of p27^Kip1 levels were observed in hepatocellular carcinoma, ⁴⁶ breast carcinoma, ¹² bladder carcinoma, ¹⁰ and cervical carcinoma ⁴⁷ cell lines treated with TSA and or NaB. HDAC inhibitors are known to induce upregulation of the p21^Waf/Cip1 gene by selectively inducing accumulation of acetylated histones in chromatin associated with p21^Waf/Cip1 gene promoter in human cancer cell lines. ^{9 10 11} Mechanism(s) that mediate HDACI-induced upregulation of p27^Kip1 gene transcription have not been studied thus far. 

This is the first report of biological action of depsipeptide in UM cell lines. We have demonstrated that primary and metastatic UM cell lines are exquisitely sensitive to the HDACIs including depsipeptide, NaB, and TSA. In addition, we identified two distinct signaling pathways involved in depsipeptide-mediated cell growth arrest and induction of apoptosis. The ultimate goal of our study is to provide the rationale for the future use of HDAC inhibitors as a new therapeutic modality in the treatment of primary and metastatic UM in humans. 

 Supported in part by the Ohio Lions Eye Research Foundation and The Carl M. and Grace C. Baldwin Eye Care Fund.

 Submitted for publication October 13, 2002; revised December 15, 2002; accepted January 6, 2003.

 Disclosure: D.D. Klisovic, None; S.E. Katz, None; D. Effron, None; M.I. Klisovic, None; J. Wickham, None; M.R. Parthun, None; M. Guimond, None; G. Marcucci, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Effect of depsipeptide on histone acetylation in MUM-2b cell line (Western blot).

Figure 2.

 

Effect of HDACI on proliferation of UM cell lines (A) OCM-1, (B) C918, (C) M619, (D) MUM-2b, and (E) MUM-2c. Cells were exposed to each agent for 48 hours at the indicated concentrations. Concentrations of NaB (Sodium butyrate) are expressed in millimolar, of TSA (Trichostatin) in micromolar, and of depsipeptide in nanomolar. The decrease in proliferation of treated cells is expressed as a relative ratio compared with nontreated cells.

Figure 3.

 

Effect of depsipeptide treatment at 48 hours on apoptotic DNA contents (Ap) in MUM-2b cells (the sub-G₁ fraction, by flow cytometry). Concentration of depsipeptide is given in nanomolar and apoptotic content is expressed as a percentage. (A) No treatment: Ap, 12.9%; (B) 0.1 nM: Ap, 13.1%; (C) 1 nM: Ap, 14.9%; (D) 5 nM: Ap, 25.0%; (E) 10 nM: Ap, 35.2%; and (F) 100 nM: Ap, 49.2%.

Figure 4.

 

Depsipeptide treatment induced dose-dependent activation of caspase-3 (p32) in MUM-2b cell line, evidenced by conversion of the proenzyme form p32 to the catalytically active effector protease (p17). This increase in caspase-3 activity was followed by cleavage of PARP (116 kDa) into an 85-kDa C-terminal fragment and a 28-kDa N-terminal fragment.

Figure 5.

 

Dose-dependent increase in the activity of caspase-3 in the MUM-2b cell line treated with depsipeptide. Increase in the activity observed in treated cells is expressed as a relative ratio compared with nontreated cells (▪). This increase in activity was abolished when cells were pretreated with cycloheximide (1 μg/mL) for 1 hour (▨).

Figure 6.

 

Depsipeptide induced a dose-dependent increase in gene expression (RT-PCR) and protein content (Western blot) of p21^Waf/Cip1 (A) and p27^Kip1 (B) in the MUM-2b cell line. p21^Waf/Cip1 also serves as a substrate for active caspase-3 evidenced by cleavage into 14-kDa fragment. The c-abl sequence was used as an internal RT-PCR control (C).

Figure 7.

 

Depsipeptide induces a dose-dependent increase in gene expression (RT-PCR) and protein content (Western blot) of Fas (A) and FasL (B) in the MUM-2b cell line. The c-abl sequence was used as an internal RT-PCR control (C).