Antileukemic Scalarane Sesterterpenoids and Meroditerpenoid from Carteriospongia (Phyllospongia) sp., Induce Apoptosis via Dual Inhibitory Effects on Topoisomerase II and Hsp90

Two new scalarane sesterterpenoids, 12β-(3′β-hydroxybutanoyloxy)-20,24-dimethyl-24-oxo-scalara-16-en-25-al (1) and 12β-(3′β-hydroxypentanoyloxy)-20,24-dimethyl-24-oxo-scalara-16-en-25-al (2), along with one known tetraprenyltoluquinol-related metabolite (3), were isolated from the sponge Carteriospongia sp. In leukemia Molt 4 cells, 1 at 0.0625 μg/mL (125 nM) triggered mitochondrial membrane potential (MMP) disruption and apoptosis showing more potent effect than 2 and 3. The isolates inhibited topoisomerase IIα expression. The apoptotic-inducing effect of 3 was supported by the in vivo experiment through suppressing the volume of xenograft tumor growth (47.58%) compared with the control. Compound 1 apoptotic mechanism of action in Molt 4 cells was further elucidated through inducing ROS generation, calcium release and ER stress. Using the molecular docking analysis, 1 exhibited more binding affinity to N-terminal ATP-binding pocket of Hsp90 protein than 17-AAG, a standard Hsp90 inhibitor. The expression of Hsp90 client proteins, Akt, p70S6k, NFκB, Raf-1, p-GSK3β, and XIAP, MDM 2 and Rb2, and CDK4 and Cyclin D3, HIF 1 and HSF1 were suppressed by the use of 1. However, the expression of Hsp70, acetylated tubulin, and activated caspase 3 were induced after 1 treatment. Our results suggested that the proapoptotic effect of the isolates is mediated through the inhibition of Hsp90 and topoisomerase activities.

Heat shock proteins (Hsp) comprise a certain group of highly conserved stress proteins which attracted attention due their overexpression in cancer tissues 1,2 . Overexpression of these proteins is related to metastatic potential, resistance to chemotherapy and poor prognosis 3 . They are named depending on their molecular weight (Hsp60, Hsp70 and Hsp90) and among the most studied protein is Hsp90. This protein is the most prominent member of the highly abundant chaperone proteins and it is essential for folding nascent polypeptide to control the activity, stability and protein sorting 4 . Hsp90 has been identified as a promising drug target for cancer treatment, because it can stabilize and activate a variety of survival proteins to maintain cancer phenotype and help cancer cells to overcome multiple environmental stresses 5 . There is considerable interest in developing potential Hsp90 inhibitors, with a much simpler rationale, through the depletion of oncogenic Hsp90 clients 6 . The structure of this protein is composed of three major regions: an amino (N)-terminal domain with an adenosine triphosphate (ATP)-binding and hydrolyzing pocket (with ATPase activity), a middle domain involved in client protein recognition/binding, and a carboxy (C)-terminal domain 7 . Tanespimycin (17-allylamine-17-demethoxygeldanamycin, 17-AAG), the first Hsp90 inhibitor, was found to bind to the N-terminal regulatory pocket of Hsp90 and thus inhibiting its function. In Phase I clinical trials, it showed promising anticancer activity against multiple myeloma in combination with bortezomib 8,9 . However, further clinical development of 17-AAG was halted in 2010 because of poor solubility, limited bioavailability, unaccepted hepatotoxicity and the most important factor was the expiration of the patent in 2014 4,10,11 . To date, several Hsp90 inhibitors have entered clinical trials, but none of them has been approved as an anticancer agent 12 .
Another important group of proteins, topoisomerases (Topo), has also attracted attention due to their crucial role in cell survival and replication 13 . Topoisomerases are classified into two main classes: topoisomerase I and topoisomerase II with each class performing specific functions in the cell. Topoisomerase II is composed of two isoforms, α and β , which share highly similar amino acid sequence (up to 70%) 14 . Topo IIα is essential for the survival of proliferating cells and can distinguish the handedness of DNA supercoils during relaxation reactions; however, Topo IIβ is dispensable at the cellular level 15,16 . Topo II disentangle topological problems, which regulate DNA replication, transcription and chromosome segregation, as well as processes related to tumorigenesis 17 . Inhibition of Topo II activity is one of the current therapeutic protocols targeting several cancers including lung, breast, lymphomas, testicular and sarcomas 18 .
The inhibition of Topo II activity is achieved either with poisons, which interfere with the topoisomerase-DNA complex or inhibitors, which suppress the catalytic turnover. Topo II poisons are divided into two distinct classes, interfacial poison and covalent poison based on their mechanism of action 15,19 . Anticancer drugs such as etoposide, doxorubicin, mitoxantrone and bioflavonoid analogs are considered interfacial poisons that bind non-covalently to the cleavage complex at the protein-DNA interface 15,17 . On the other hand, epigallocatechin galate (EGCG) and curcumin are covalent poisons which function distal to the active site of Topo II and increase the level of enzyme-mediated DNA cleavage by altering conformation of Topo II N-terminal protein gate [20][21][22] . Despite the effectiveness of Topo II poisons as anticancer drugs, they can trigger chromosomal breaks leading to secondary leukemogenesis 14,23,24 . One solution of this side effect was the development of catalytic Topo II inhibitors such as bisdioxopiperazines which eliminate the essential enzymatic activity of Topo IIα 17 . These compounds showed modest anticancer activity but did not induce secondary malignancies. Further developments of catalytic Topo II inhibitors can introduce new classes of highly effective and relatively safe anticancer agents.
Natural products scaffolds have played a crucial rule in guiding researchers to develop efficient anticancer agents targeting proteins necessary for cancer cell survival and replication 11 . Certain classes of secondary metabolites exhibited potent anticancer activity such as terpenoids, alkaloids, and flavonoids 25 . Terpenoids are the largest and most diverse group of secondary metabolites which are divided into several subgroups including monoterpenoids, diterpenoids, sesquiterpenoids, sesterterpenes, and triterpenoids 26 . Scalarane sesterterpenoids emerged as an interesting group of terpenoids which were isolated from marine sponges and shell-less mollusks. Scalarane-type sesterpenoids are 25C-terpenoids with tetra-or penta-cyclic skeletons derived from scalarin, which was firstly isolated from the sponge Cacospongia scalaris in 1972 27 . Scalarane sesterterpenoids displayed a wide spectrum of interesting biological properties, such as antifeedant, antimicrobial, antifungal, ichthyotoxicity, antitubercular, antitumor, cytotoxicity, anti-HIV, antifouling, inhibition of platelet-aggregation, inhibition of transactivation for the nuclear hormone receptor (FXR, farnesoid X-activated receptor), stimulation of nerve growth factor synthesis, as well as anti-inflammatory activity 28 . Recent reports from our group revealed the potent cytotoxic potential of scalarane-type sesterterpenoids 29,30 . Carteriospongia (Phyllospongia) sp. sponge is a rich source of bioactive scalarane sesterterpenoids 31 . Unfortunately, many of the cytotoxic studies on sesterterpenoids evaluated their cytotoxic effect on cancer cell lines without deviling deeper into their mechanism of action 32 . Relieving the cytotoxic mechanism of action is crucial for discovering cytotoxic drug leads and fine tuning the active compounds to improve potency and reduce cytotoxicity. In the current investigation, we examined the content of EtOAc extract of Carteriospongia (Phyllospongia) sp. which led to the isolation of two new scalarane sesterterpenoids and one known tetraprenyltoluquinol-related metabolite. We evaluated the cytotoxic activity of isolates in vitro and in vivo along with their mechanism of action. The isolates caused disruption in mitochondrial membrane potential suggesting that the apoptotic effect is mediated through mitochondrial dysfunction. The cytotoxic effect of the new scalarane-type sesterterpenoid (1) on human leukemia Molt 4 cells involved induction of ER stress as demonstrated by the increase in ROS generation as well as ATF 6 cleavage and Chop expression. Compound 1 exhibited potent Hsp90 and Topoisomerase II (Topo II) catalytic inhibitory activity as determined by molecular docking and cell-free system. The treatment of Molt 4 cells with compound 1 suppressed client protein expression and accumulation of Hsp70 in the cytosolic compartment and well as reduced the transcription factors (HSF 1 and HIF 1) expression as demonstrated by Western blot and immunofluorescent analyses.
Compound 1 was obtained as colorless oil. The molecular formula of 1 was determined to be C 31 H 48 O 6 by HR-ESI-MS (m/z 523.3397 [M+ Na] + ) and 13 C NMR data (Table 1), implying eight degrees of unsaturation. IR absorptions were observed at 3448, 1717 and 1665 cm −1 , suggesting the presence of hydroxyl, saturated carbonyl and α,β-unsaturated carbonyl functionalities. Resonances due to an aldehyde carbonyl carbon (δ C 201.1), α,β-unsaturated carbonyl carbon (δ C 198.2), ester carbonyl carbon (δ C 172.1) and olefinic carbons (δ C 142.9, CH; 138.4, C) in the 13 C NMR and DEPT spectral data accounted for four double-bond equivalents, indicating a tetracyclic skeleton of 1. In the 1 H NMR data, resonances of one olefinic proton (δ H 7.06, s) and two oxygenated methines (δ H 4.86, dd, J = 11.0, 4.5 Hz; 4.22, t, J = 9.0 Hz) were observed. The planar structure and all of the 1 H and 13 C chemical shifts of 1 were elucidated by 2D NMR spectroscopic analysis, in particular 1 H-1 H COSY and HMBC experiments (Fig. 2), suggesting a characteristic scalarin-type sesterterpenoid structure. Compound 1 was found to possess a 3-hydroxybutanoyloxy at C-12, a double bond at C-16/C-17, a ketone group at C-24 and an aldehyde group at C-25. Moreover, the structure of 1 was established unambiguously, and was found to be similar to that of a known compound, dendalone 3-hydroxybutyrate 8,34 . The relative configuration of 1 was suggested to be similar to that of dendalone 3-hydroxybutyrate by comparison of the chemical shifts and coupling constants of both compounds protons and was further confirmed by NOESY correlations (Fig. 2). The 3′ R-configuration was determined by comparing the optical rotation (+ 10.7) with those of 2 (+ 5.7). The structure of 1 was thus found to possess the (4S*, 5S*, 8R*, 9R*, 10S*, 12R*, 13S*, 14S*, 18R*, 3′ R*)-configuration.
Compound 2 molecular formula was deduced as C 32 H 50 O 5 based on HR-ESI-MS (ion peak at m/z 537.3546 [M+ Na] + ) and 13 C NMR data suggesting that 2 possesses one more CH 2 group compared with 1. The NMR data of 2 (Table 1) showed similarity to those of 1 with the replacement of one 3-hydroxybutanoyloxy group at C-12 in 1 by one 3-hydropentanolyoxy group in 2. The structure of compound 2 was further confirmed by the COSY correlations from H 2 -29 to H 3 -32 and by the HMBC correlations from H-12 and H 2 -29 to C-28. Both compounds (1 and 2) were suggested to possess identical relative configuration based on the similarity in their NMR data. In order to identify the C-3′ configuration on the 3-hydroxybutanoyloxy group, the (S)-and (R)-MTPA ester derivatives, 2s and 2r, were synthesized with (− )-(R)-MTPA-Cl and (+ )-(S)-MTPA-Cl, respectively. The Δ δ-values indicated 3′ R-configuration ( Figure S19).
Apoptotic induction of these marine terpenoids via DNA damage and MMP dysfunction. The antiproliferative effect of the two new scalarane sesterterpenoids and the known tetraprenyltoluquinol-related metabolite was evaluated using MTT assay. Several cancer cell lines including leukemia K562, Molt 4, and HL 60 cells, prostate cancer LNCaP cells, colon cancer DLD-1 cells and breast cancer T-47D cells were used to evaluate the antiproliferative activity. Figure 3 (Fig. 4A,B). These results proved that the cytotoxic activity of these marine terpenoids might be mediated through mitochondrial dysfunction leading to induction of apoptosis.
The ability of Topo II to generate DNA damage in the presence of Topo II-targeting agents led to the hypothesis that an important determinant of drug sensitivity was related to the overall level of Topo II 17 . In our previous  report, 10-acetylirciformonin B, a marine sesterterpenoid analog, significantly inhibited Topo IIα activity 35 . Such findings encouraged us to evaluate the effect of the isolates on Topo II activity. In order to confirm whether the DNA damage-induced by the isolated sesterterpenoids involved the inhibition of Topo II activity, a cell-free DNA cleavage assay using an enzyme-mediated negatively supercoiled pHOT1 plasmid DNA was applied. A linear DNA strand was observed upon treating the supercoiled pHOT1 plasmid DNA with etoposide, a standard topo II poison (Lane 16). The use of compound 1 in increasing concentrations (0.08, 0.312, 1.25, 5, and 20 μ g/mL) significantly inhibited DNA relaxation by 12, 13, 17, 20 and 99%, respectively, compared with the control supercoiled DNA and resulted in the formation of supercoiled DNA products in the presence of topo IIα (Lanes 6-10). Additionally, 2 and 3 significantly inhibited DNA relaxation by 12, 23, 36, 90 and 99% (Lanes 11-15); and 7, 13, 28, 90 and 98% (Lanes 1-5), respectively. These marine terpenoids suppressed Topo II activity, resulting in the inhibition of supercoiled DNA relaxation in a dose-dependent manner (Fig. 5A). Furthermore, 1-3 inhibited topo II activities with IC 50 of 1.98, 0.37 and 0.43 μ g/mL, respectively as demonstrated by the cell-free system. To determine whether the apoptotic mechanism of these compounds affects γ H2AX (as a biomarker of DNA damage) induction, Western blot analysis was employed to examine the activation of γ H2AX. Compounds 1 and 2 significantly enhanced the expression of γ H2AX in a dose-dependent manner, but the activation of DNA damage was not observed on 3 (Fig. 5B). These results suggested that these compounds could act as potent catalytic inhibitors of Topo IIα .

2-Tetraprenil-1,4-benzochinone (3) inhibited tumor growth in in vivo human Molt 4 tumor xen-
ograft animal model. Compounds 1 and 2 induced apoptosis in several cancer cell lines; however, the amounts isolated were insufficient for further examination of their in vivo antitumor effect. On the other hand, the amount separated from 3 was sufficient to study the in vivo anti-tumor activity by evaluating its effect on tumor growth of human leukemia Molt 4 in xenograft animal model. Molt 4 (1 × 10 6 ) cells were inoculated subcutaneously at the right flank of female immunodeficient athymic mice. After 33-day of treatment, the tumor growth of Molt 4 cells was significantly suppressed under the influence of 3 (1.14 μ g/g) intraperitoneal injection. The average tumor size on day 33 in the control group was 404.63 mm 3 , whereas the average tumor size in compound 3-treated group was 212.10 mm 3 (Fig. 6A). The tumor size was significantly lower in compound 3-treated group as compared to the control group (p < 0.05) with no significant difference in the mice body weights (Fig. 6B). These results suggested that 3 exhibited anti-tumorigenic effect in vivo xenograft model.

Apoptotic effect of 12β-(3′β-hydroxybutanoyloxy)-20,24-dimethyl-24-oxo-scalara-16-en-2 5-al (1) involves the induction of ROS generation, ER stress and DNA damage in Molt 4 cells.
Oxidative stress is manifested by ROS overexpression and cannot be balanced by the available antioxidant machinery 36 . Compound 1 can damage the integrity of mitochondria, which are the major production sites  of the superoxide anion, ozone 37 . The induction of the intracellular formation of ROS by 1 was determined with a carboxyl derivative of fluorescein, carboxy-H 2 DCFDA dye using flow cytometric analysis 38 . As shown in Fig. 7A, treatment with 1 (0.0625 μ g/mL) for 0.5, 1, 2 and 3 h resulted in 2.09-, 1.51-, 1.06-and 1.01-fold increase in ROS levels, respectively, as compared with the mean fluorescence index (MFI) of the control. In addition, ROS generation could induce ER stress leading to mitochondria-related apoptosis 39,40 . To further investigate if ER stress is involved in the apoptotic effect induced by 1, Western blot analysis was used to determine the expression of ER stress-related proteins. In a time-dependent manner, 1 promoted the levels of Bip, Chop, Grp 94 as well as the activation and cleavage of ATF 6 but suppressed the levels of PERK and IRE 1 (Fig. 7B). Furthermore, the effect of 1 on the release of intracellular Ca 2+ , was evaluated using fluorescent calcium indicator, Fluo 3. The flow cytometric results showed that the treatment with 1 at different time intervals (0.5, 1, 2, 3, 6 and 18 h) induced 1.03-, 1.05-, 1.06-, 1.37-, 2.41-, and 1.06-folds increase in the intracellular Ca 2+ accumulation, respectively as compared to MFI of the control (Fig. 7C), indicating that ER stress was induced by compound 1 following the redox stress. Moreover, we noticed a significant increase of PARP, caspases-8 and − 9 cleavages in a time-dependent manner. The induction of the typical executor caspases-3 and − 7, as well as γ H2AX, was also observed using Western blot analysis (Fig. 7D). To confirm the induction of DNA damage by 1 in Molt 4 cells, a comet assay under neutral electrophoresis condition was utilized. The effect of 1 (0.0625 μ g/mL) at different time intervals (0, 3, 6 and 9 h) on   (Fig. 8B).
According to the MTT results using Molt 4 cells, the antiproliferative activity of 1 (0.01 μ g/mL, 19.1 nM) was 377 folds higher than that of 17-AAG. Recent reports indicated that Hsp90 inhibitors induced HSF 1-dependent expression of Hsp70 44 , and the genetic deletion of HSF 1 reduced the association of Hsp90 with its kinase client proteins 45,46 . It was also reported that the induction of Hsp70 is the biomarker of Hsp90 with N-terminal inhibition 3,6,7 . Aiming to understand the relation between Hsp90 function and the expression of Hsp70 and 90 clients, we treated Molt 4 cells with 1 (0.0625 μ g/mL) for different time intervals. We first observed that 1 treatment did not notably attenuate the expression of Hsp90 protein. In agreement with earlier studies [47][48][49] , the expression of Hsp70 increased in a time-dependent manner, as an established marker for HSR after Hsp90 inhibition, while surprisingly the expression of HSF 1 protein was attenuated with 1 treatment in Molt 4 cells. As expected, the suppression of Hsp90 client proteins was observed, including p70 S6k , NFκ B, Raf-1, p-GSK3β , MEK 1 and XIAP (pro-apoptotic protein), MDM 2 and Rb2 (oncoprotein), and CDK4 (cell cycle regulatory protein), HIF 1 and HSF1 (transcription factor) (Fig. 8C). The majority of Hsps accumulate in subcellular localizations that determine whether a cell is going to die or differentiate. Erythroblasts with accumulated nuclear Hsp70 continued their differentiation process during the formation of red blood cells 50 . Furthermore, we identified the localization of Hsp70 in response to 1 treatment using immunofluorescence by confocal microscope. In agreement with the Western blot results, the localization of Hsp70 was predominantly accumulated in cytosol and the accumulation increased with time (Fig. 8D).

Discussion
Clinically, there is an unmet medical need in the treatment of T-cell lymphoblastic leukemia. In comparison with other human malignancies such as colon cancer, breast cancer, prostate cancer where there has been much progress in their treatment strategies including molecular targeting therapy; patients with T-cell lymphoblastic leukemia are usually treated with conventional chemotherapy which is often associated with acute or chronic toxicities as well as a high relapse rate 51,52 . The outcome of refractory or relapsed disease remains poor and the search for more effective drugs or potential targets is urgently required. T-cell lymphoblastic leukemia Molt 4 cell line was the most sensitive cell line to treatment with the new isolated scalarane sesterterpenoids as determined with MTT assay (Fig. 3), reflecting their potential as novel anti-leukemic agents.
In this study, we investigated the therapeutic potential of scalarane sesterterpenoids (1 and 2) and a meroditerpenoid (3) in Molt 4 cell line. All compounds reduced cells viability and Topoisomerase II catalytic activity with comparable IC 50 values. Flow cytometric analysis showed that 1 induced mitochondrial dysfunction and apoptotic cell death with the lowest dose (0.0625 μ g/mL) after 24 h (Fig. 4A,B). Therefore, 1 was selected for further in depth studies to understand its apoptotic mechanism of action. It was reported that Hsp90 facilitates the proper folding of signaling proteins associated with cancer progression, tumor angiogenesis and therapy resistance by functioning as molecular chaperone, gaining attention as a target for therapeutic intervention 47,48,53 . A growing body of evidence indicates that the accumulation of unfolding/misfolding proteins by the suppression of Hsp90 function assists several stresses, including ER stress overload, the ROS over-generation, and a functional disorder of the intracellular proteins, ultimately leading to ER stress-induced apoptosis, if unfolding is overwhelming [54][55][56][57] . The unfolding protein response (UPR) is distinguished by three ER transmembrane receptors: protein kinase RNA-like ER kinase (PERK), activating transcription factor 6 (ATF6) and inositol-requiring enzyme 1 (IRE1), which involve both transcriptional and translational regulation of genes to expand the processing capacity of the ER and return this organelle to homeostasis 58,59 . If ER homeostasis is not restored, UPR promotes cell death. IRE1 is a transmembrane protein and cleaves microRNAs to control the levels of caspase family cell death proteases 60 . PERK is the major protein of the UPR and participates in ER stress-induced cell death, in part through the up-regulation of the proapoptotic CCAAT /enhancer binding protein homologous protein (Chop) expression and XIAP degradation 61 . In response to ER stress, ATF6, a transcriptional factor, is translocated to the Golgi compartment and cleaved by the action of two serine proteases, S1P and S2P. The cleaved ATF6 binds to ATF/cAMP response elements (CRE) and ER stress-response elements (ERSE-1) to activate the target genes, including Bip, GRP94 and Chop 58 . However, several lines of evidence suggested that Bip (Grp78) was not pivotal to switch the UPR on and off 58,62 . Moreover, Ca 2+ trafficking in and out of ER regulates a diversity of cell responses and signaling transduction related with stress response, modulation of transcriptional processes and development 58 . In fact, it is known that Ca 2+ release from the ER to accumulate in mitochondria could trigger a variety of signaling mechanisms to induce cell death mainly by Ca 2+ -mediated mitochondrial cell death 63 . We observed that 1 treatment first stimulated reactive oxygen species (ROS) generation, perturbed the Bip/IRE1/PERK signal pathway and activated Grp94/ATF6/Chop signal pathway implicated in ER stress (Fig. 7B). In addition, large amount of Ca 2+ release was observed which led to mitochondrial dysfunction-dependent apoptosis (Figs 2A,B and 7C).
Cumulative evidence suggests that several C-terminal Hsp90 inhibitors, as an alternatively clinical target, possess antiproliferative and apoptotic activities without eliciting heat shock response (HSR) [47][48][49] . Previous studies have indicated that activation of HSF 1 (Heat Shock Factor 1) is critical for the induction of HSR and may confer acquired resistance of Hsp90 inhibitors 64,65 . The master regulator, HSF 1, is involved with tumorigenesis and could possess potential target in cancer therapy 66,67 . Interestingly, compound 1, a specific N-terminal Hsp90 inhibitor, activated Hsp70 induction and did not elicit the HSF 1 (Fig. 8C) 68,69 . It is worth noting that a cytosolic relay of Hsp70 and 90β monitors the folding trajectory of the serotonin transporter, as the principle target of antidepressant drugs 70 . This is presumably more important that the induction of Hsp70 by 1 compartmentalized in the cytosol of cells as demonstrated by immunofluorescent analysis (Fig. 8D). Accordingly, our results showed the potential merits of the scalarane sesterterpenoids as potential candidates for future clinical trials.
Nevertheless, some limitations to this present investigation were observed. First, the extremely small isolated quantity of 1 hindered the examination of its anti-leukemia effect in xenograft animal model. It also prevented the evaluation of 1 inhibitory activity on Hsp90 with cell-free system. Second, the transcription factors (HSF 1 and HIF 1) were blocked by 1 treatment; the factor of heat shock response that may contribute to the induction of Hsp70 protein has not been explored. Therefore, the key functions of Hsp70 induction should be further explored to identify the regulatory role in apoptosis process induced by this marine sesterterpenoid derivative.
In conclusion, we found that the treatment of cancer cells with the isolated scalarane sesterterpenoids (compounds 1 and 2) and a meroditerpenoid (compound 3) induced mitochondrial dysfunction, oxidative and ER stresses leading to apoptosis. Cell-free system and computational modeling using structure-function analysis further supported that these compounds can act as potential dual topoisomerase catalytic and Hsp90 inhibitors. The results confirmed a decrease in the expression of various Hsp90 client proteins in Molt 4 cells after treatment with 1, the Hsp90 N-terminal inhibitor. Our studies provide further insights on the pro-apoptotic mechanism of these new sesterterpenoid derivatives, and their clinical potential as novel Hsp 90 and Topoisomerase II catalytic inhibitors to treat leukemia patients. Consequently, sesterterpenoids represent interesting molecular architectures which bind preferentially to proteins involved in tumorigenesis. Recent advances in sesterterpenoids synthesis 71 open new avenues in their utilization in current therapeutic regimes to increase the effectiveness and decrease doses of these regimens leading to fewer side effects.

Preparation of (R)-and (S)-MTPA Esters (2r and 2s).
Small amount of the compound 2 (2.5 mg) was divided, stored in two NMR tubes and dried under vacuum. Deuterated pyridine (0.60 mL) and (R)-MTPA-Cl (12 μ L) were added to one of the NMR tubes. The reaction NMR tubes were permitted to stand at room temperature and monitored by 400 MHz NMR every hour. After 3 h, the reaction was found to be completed, and the 1 H NMR data was obtained (400 MHz, in C 5 D 5 N). (S)-MTPA esters of 2 (2s) was obtained, and the 1 H NMR data was analyzed. Similar to 2s, (S)-MTPA-Cl (12 μ L) and deuterated pyridine (0.60 mL) were reacted at room temperature for 3 h, to afford the (R)-MTPA ester derivatives (2r), in separate experiment, and the 1  Annexin V/PI Apoptosis Assay. The externalization of phosphatidylserine (PS) and membrane integrity were quantified using an annexin V-FITC staining kit 72 . In brief, 10 6 cells were grown in 35 mm diameter plates and were labeled with annexin V-FITC (10 μ g/mL) and PI (20 μ g/mL) prior to harvesting. After labeling, all plates were washed with a binding buffer and harvested. Cells were resuspended in the binding buffer at a concentration of 2 × 10 5 cells/mL before analysis by flow cytometer FACS-Calibur (Becton-Dickinson, San Jose, CA, USA) and CellQuest software. Approximately 10,000 cells were counted for each determination.
Determination of ROS Generation, Calcium accumulation, and MMP Disruption. These assays were performed as described previously 38 . MMP disruption, calcium accumulation and ROS generation were detected with JC-1 cationic dye (5 μ g/mL), the fluorescent calcium indicator (Fluo 3, 5 mM) and the carboxy derivative of fluorescein (carboxy-H 2 DCFDA, 1.0 mM), respectively. In brief, the treated cells were labeled with a specific fluorescent dye for 30 min. After labeling, cells were washed with PBS and resuspended in PBS at a concentration of 1 × 10 6 cells/mL before analysis via flow cytometry. Assay of Topoisomerase II Catalytic Inhibitors and Poisons. The assay was performed as described previously 35,38 . Standard relaxation reaction mixtures (20 μ L) containing 50 mM Tris-HCl (pH 8.0), 10 mM MgCl 2 , 200 mM potassium glutamate, 10 mM dithiothreitol, 50 μ g/mL bovine serum albumin, 1 mM ATP, 0.3 μ g of pHOT1 plasmid DNA, two units of human topoisomerase II (Topogen, Columbus, OH, USA), and the indicated concentrations of etoposide and the compounds were incubated at 37 °C for 30 min. Reactions were terminated by adding 2 μ L of 10% SDS to facilitate trapping the enzyme in a cleavage complex, followed by the addition of 2.5 μ L of proteinase K (50 μ g/mL) to digest the bound protein (incubated at 37 °C for 15 min) and finally by adding 0.1 volume of the sample loading dye. The DNA products were analyzed via electrophoresis through vertical 2% agarose gels at 2 voltage/cm in 0.5 × TAE buffer. Gels were stained with ethidium bromide and photographed using an Eagle Eye II system (Stratagene, La Jolla, CA, USA). The quantitative analysis of the DNA Topo II activity was performed as described previously 73 . The gels were directly scanned with image analyzer, and the area representing supercoiled DNA calculated to evaluate the concentration that the compounds caused 50% inhibition (IC50) of Topo II activity.
Neutral Comet Assay for Detection of DNA Double-strand Breaks (DSBs). The assay was carried out using a CometAssayTM Kit (Trevigen, Gaithersburg, MD, USA) following the manufacturer's protocol for the neutral Comet assay. Briefly, cancer cells (2 × 10 5 cells/mL) were treated with compound 1 (0.0625 μ g/mL) at the indicated time. Cells were combined with 1% low melting point agarose at a ratio of 1:10 (v/v) and immediately 75 μ L of the mixture was pipetted onto CometSlide TM and allowed to set at 4 °C in the dark. The slides were immersed in ice-cold lysis solution (Trevigen) for 30 to 60 min. The slides were placed in a horizontal electrophoresis apparatus and electrophoresed in 1X TBE (90 mM Tris-HCl, 90 mM boric acid, and 2 mM EDTA, pH 8.0) at 20 V for 10 min. The samples were then fixed in 70% ethanol and dried before stained with 1:10,000 SYBR Green I (Trevigen) to visualize cellular DNA. The fluorescence images were analyzed using the TriTek Comet Image program to circumscribe the "head" and the "tail" regions of each comet and the integrated fluorescence values of each defined area were recorded. The comet length was measured from the trailing edge of the nucleus to the leading edge of the tail. This length was indicative of the extent of DNA damage. Calculations were averaged per replicate.
Western Blotting Analysis. Cell lysates were prepared by treating the cells for 30 min in RIPA lysis buffer, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulphate (SDS), 1 mM sodium orthovanadate, 100 μ g/mL phenylmethylsulfonyl fluoride and 30 μ g/mL aprotinin) (all chemicals were from Sigma) 38 . The lysates were centrifuged at 20,000 × g for 30 min, and the protein concentration in the supernatant was determined using a BCA protein assay kit (Pierce). Equal amounts of proteins were respectively separated by 7.5%, 10% or 12% of SDS-polyacrylamide gel electrophoresis and then were electrotransferred to a PVDF membrane. The membrane was blocked with a solution containing 5% non-fat dry milk TBST buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl and 0.1% Tween 20) for 1 h and washed with TBST buffer. The protein expressions were monitored by immunoblotting using specific antibodies. These proteins were detected by an enhanced chemiluminescence kit (Pierce).
Immunofluorescence Analysis. After treatment with the tested compound, cells were fixed with 4% paraformaldehyde in 50 mM HEPES buffer (pH 7.3) for 30 min, and permeabilized for 20 min with 0.2% Trition X-100 in PBS (pH 7.4). To prevent non-specific protein binding, cells were incubated with 5% BSA in PBS containing 0.05% Trition X-100 (T-PBS) for 1 h at room temperature. Cells were then incubated with the primary Hsp70 antibodies (1: 500) for 2 h and further with secondary antibodies (Alexa Fluor 586-conjugated goat anti-mouse IgG (H + L) (Life Technologies, Carlsbad, CA, USA) diluted at 1:1000 for 1 h at room temperature. After washing with PBS, cells were observed under a FV1000 confocal laser scanning microscope (Olympus, Tokyo, Japan).

Human Leukemia Molt 4 Cells Xenograft Animal
Model. Establishment of nude mice with xenografts was performed as described previously 38 . Six-week-old male immunodeficient athymic mice were purchased from the National Laboratory Animal and Research Center (Taipei, Taiwan). All of the animals were maintained under standard laboratory conditions (temperature 24-26 °C, 12-12 h dark-light circle) and fed with laboratory diet and water. This study was approved by the Animal Care and Treatment Committee of Kaohsiung Medical University (IACUC Permit Number 101136). All experiments were conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, and all efforts were made to minimize animal stress/distress. Molt 4 cells (1 × 10 6 ) resuspended in 0.2 mL PBS were injected s.c. into the right flank of each mouse, and tumor growth was monitored every day. Fourteen days after tumor cell injection, mice with confirmed tumor growth were randomly divided into two groups. Compound 3 (1.14 μ g/g) was intraperitoneally administered to the treatment group, and the control group received solvent only. Compound 3 was administrated every other day for 33 days. Animals were sacrificed by carbon dioxide. Tumor size was measured three times a week using calipers and tumor volumes were calculated according to the standard formula: width 2 × length/2.

Molecular Modeling Assay. The molecular docking was performed by Autodock 4.2 with Lamarckian
Genetic Algorithm 74 . The target macromolecule, Hsp90 protein (PDB ID: 1YET), was obtained from the Protein data bank (http://www.rcsb.org/pdb/home/home.do) 75 . The co-crystalized protein substrates, including ligands, water and small molecules were removed, and the Polar hydrogens and Kallman united atom charges were added to the protein for docking calculation by AutoDock Tool 1.5.4 interfaces (ADT) 76 . The ligands were optimized with MMFF94 force field by ChemBio3D software (version 11.0; Cambridge Soft Corp.). Polar hydrogens and Gasteiger charges were also added to the ligand for docking study by ADT. The Grid box calculated by AutoGrid program was centered at the activity site of Hsp90 with dimensions 56 × 56 × 56 Å grid points at spacing of 0.375Å and its size is big enough to allow the ligand move freely in the search space. All docking parameters were set to default except for the following parameter: maximum number of energy evaluation increase to 25,000,000 per run. The docking results were analyzed by ADT and shown by Accelrys Discovery Studio v3.5 client software (Accelrys Inc, San Diego, CA (2005)).

Statistics.
The results were expressed as mean ± standard deviation (SD). Comparison in each experiment was performed using an unpaired Student's t-test and a p value of less than 0.05 was considered to be statistically significant.