HTLV-1-transformed T-cell lines, MT-2, MT-4, HUT-102, C5/MJ and SLB-1, and ATL-derived T-cell lines, MT-1 and ED-40515(-), were maintained in RPMI-1640 medium (cat. no. 30264-56, Nacalai Tesque, Inc., Kyoto, Japan) supplemented with 10% fetal bovine serum (Biological Industries, Kibbutz Beit Haemek, Israel) and 1% penicillin/streptomycin (cat. no. 09367-34, Nacalai Tesque, Inc.)

A hexane extract (totally 36.2 g) of the gorgonian I. hippuris, collected July 1980, was subjected to separation first on a silica gel column and then on Sephadex LH20 (cat. no. 17-0090-03, Pharmacia, Uppsala, Sweden) column to give a crude peridinin fraction (591 mg), which was kept frozen for >30 years. Then, the fraction was passed through a Sephadex LH20 column (MeOH-CH₂Cl₂, 1-1) twice followed by silica (silica gel 60; cat. no. 107734, Merck, Darmstadt, Germany) flash column and silica HPLC (cat. no. 38005-51, Cosmosil 5SL-II, 10×250 mm; Nacalai Tesque, Inc.) (hexane-EtOAc, 1-1) to yield 9.0 mg of peridinin. Additional amount of peridinin was purified from crude fractions of another specimen of the same gorgonian, with a final total yield of 20.0 mg of peridinin. The ¹H NMR spectrum of the isolated peridinin in CDCl₃ (cat. no. 139-18001, Wako Pure Chemical Industries, Osaka, Japan) was identical to that reported previously (7).

Cell proliferation and cytotoxicity were measured by the water-soluble tetrazolium (WST)-8 assay (cat. no. 07553-44, Nacalai Tesque, Inc.). The WST-8 is taken up by viable cells and reduced to the colored formazan product by mitochondrial dehydrogenase (8). For the WST-8 assay, 100 μl (1×10⁴) cells were seeded in 96-well plates and treated with different concentrations of peridinin or dimethyl sulfoxide (DMSO) (cat. no. 13407-45, Nacalai Tesque, Inc.) for 24 h. Then, 10 μl of WST-8 was added to each well and incubated for 6 h. Absorbance was measured at 450 and 620 nm by an iMark™ microplate absorbance reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Cell viability was calculated by the percentage of treated cells relative to that of solvent controls. All experiments were performed in triplicates.

The cells were treated with different concentrations of peridinin or DMSO for 24 h and stained with the CycleTEST Plus DNA Reagent kit (cat. no. 340242, Becton-Dickinson Immunocytometry Systems, San Jose, CA, USA) for analysis of changes in the cell cycle. The stained cells were analyzed by an Epics XL flow cytometer (Beckman Coulter, Inc., Brea, CA, USA). The MultiCycle software (version 3.0) was used to calculate the percentage of cells in each cell cycle phase.

Apoptosis was assessed by the APO2.7 assay. Cells were seeded in culture plates then treated with different concentrations of peridinin or DMSO for 24 h, followed by analysis by flow cytometry after staining with phycoerythrin-conjugated APO2.7 antibody (cat. no. IM2088, Beckman Coulter, Marseille, France), which specifically detects 7A6, a 38-kDa mitochondrial membrane antigen expressed during apoptosis (9). For analysis of morphological changes in nuclei, cells stained with DNA-specific Hoechst 33342 dye (cat. no. 346-07951, Dojindo Molecular Technologies, Inc., Kumamoto, Japan) were analyzed using a Leica DMI6000 microscope (Leica Microsystems, Wetzlar, Germany). In addition, apoptosis was also assessed by monitoring cleavage of caspase-3, -8 and -9, as well as poly(ADP-ribose) polymerase (PARP) by western blot analysis.

Cells were harvested after treatment and lysed in a lysis buffer containing 62.5 mM Tris-HCl (pH 6.8) (cat. no. 35434-21, Nacalai Tesque, Inc.), 2% sodium dodecyl sulfate (cat. no. 31607-65, Nacalai Tesque, Inc.), 10% glycerol (cat. no. 17045-65, Nacalai Tesque, Inc.), 6% 2-mercaptoethanol (cat. no. 21438-82, Nacalai Tesque, Inc.) and 0.01% bromophenol blue (cat. no. 021-02911, Wako Pure Chemical Industries). The lysates were centrifuged and the supernatants were collected. The cell lysates (20 μg) were separated on sodium dodecyl sulfate-polyacrylamide gels followed by electroblotting to polyvinylidene difluoride membranes (cat. no. IPVH00010EMD, Millipore, Darmstadt, Germany). The following antibodies were used: cleaved caspase-3 (cat. no. 9664), -8 (cat. no. 9496) and -9 (cat. no. 9501), cleaved PARP (cat. no. 9541), survivin (cat. no. 2808), Bak (cat. no. 3814), Bcl-xL (cat. no. 2762), Akt (cat. no. 9272), phospho-Akt (Ser473) (cat. no. 4060), phospho-Akt (Thr308) (cat. no. 13038), phospho-IκBα (Ser32 and Ser36) (cat. no. 9246), phospho-IκB kinase (IKK)α/β (Ser176 and Ser180) (cat. no. 2694), RelA (cat. no. 8242), phospho-RelA (Ser536) (cat. no. 3033), 3-phosphoinositide-dependent protein kinase 1 (PDK1) (cat. no. 3062), phospho-PDK1 (Ser241) (cat. no. 3061), phospho-p70 S6 kinase (S6K) (Thr421 and Ser424) (cat. no. 9204), phospho-phosphatase and tensin homologue (PTEN) (Ser380) (cat. no. 9551) and phospho-extracellularly regulated kinase (Erk)1/2 (Thr202 and Tyr204) (cat. no. 4370) were all from Cell Signaling Technology, Inc. (Beverly, MA, USA); c-IAP2 (cat. no. sc-7944), cyclin D2 (cat. no. sc-593) and IκBα (cat. no. sc-371) were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA); XIAP (cat. no. M044-3) and cyclin D1 (cat. no. K0062-3) were from Medical & Biological Laboratories, Co. (Aichi, Japan); Bcl-2 (cat. no. MS-597), CDK4 (cat. no. MS-299), CDK6 (cat. no. MS-398), c-Myc (cat. no. MS-1054) and actin (cat. no. MS-1295) were from Neomarkers, Inc. (Fremont, CA, USA).

Nuclear proteins were extracted and transcription factors bound to specific DNA sequences were examined by EMSA, as described previously (10). The top strand sequences of the oligonucleotide probe or competitors were as follows: for the nuclear factor-κB (NF-κB) element of the interleukin-2 receptor α chain (IL-2Rα) gene, 5′-GATCCGGCAGGGGAATCTCCCTCTC-3′; and for the activator protein-1 (AP-1) element of the interleukin-8 (IL-8) gene, 5′-GATCGTGATGACTCAGGTT-3′. The above underlined sequences are the NF-κB and AP-1 binding sites, respectively. In competition experiments, the nuclear extracts were preincubated with 100-fold excess of unlabeled oligonucleotides for 15 min. To identify NF-κB proteins in the DNA-protein complex shown by EMSA, antibodies specific for various NF-κB family proteins, including p50 (cat. no. sc-114X), RelA (cat. no. sc-109X), c-Rel (cat. no. sc-70X), p52 (cat. no. sc-298X) and RelB (cat. no. 226X) (Santa Cruz Biotechnology Inc.) were used. These antibodies were incubated with the nuclear extracts for 45 min at room temperature before incubation with the radiolabeled probe.

Five-week-old female C.B-17/Icr-severe combined immune deficiency (SCID) mice were obtained from Kyudo, Co. (Tosu, Japan). The mice were kept in specific pathogen-free conditions and housed in cages maintained in air-conditioned rooms (temperature, 24°C; humidity, 60%) set at 12-h light/12-h dark cycles. Mice were provided with standard rodent diet (CE-2 from CLEA Japan, Inc., Tokyo, Japan) and water ad libitum. To induce malignancy, 1×10⁷ HUT-102 cells suspended in 300 μl sterile RPMI-1640 medium were inoculated subcutaneously into the postauricular region of the SCID mice, which were then divided randomly into two treatment groups (n=6/group). Peridinin was solubilized in 5.2% polyethylene glycol 400 (cat. no. 161-09065, Wako Pure Chemical Industries) and 5.2% Tween-80 (cat. no. 231181, Becton-Dickinson, Franklin Lakes, NJ, USA) at a concentration of 0.56 mg/ml, and administered intraperitoneally for 5 days/week with a 2-day rest, and the treatment was continued for 21 days, beginning on the day subsequent to cell inoculation. The control group received vehicle only, while the treated group received peridinin at dose of 8.5 mg/kg. Tumor diameter was measured weekly with a shifting caliper, and tumor volume was calculated. Mice were weighed weekly. All mice were sacrificed on day 21 when tumors did not reach the ethically allowed maximal size. Subsequently, tumors were excised and their weight was measured. This study was performed according to the Guidelines for Animal Experimentation of the University of the Ryukyus (Nishihara, Japan), and was approved by the Animal Care and Use Committee of the University of the Ryukyus (reference no. 6042).

Tumor specimens were collected from the control and peridinin-treated groups, fixed in formalin (Wako Pure Chemical Industries) solution, dehydrated through graded ethanol series (Japan Alcohol Selling Co., Tokyo, Japan) and embedded in paraffin (cat. no. 09620, Sakura Finetek Japan Co., Tokyo, Japan). The paraffin-embedded specimens of ATL tumors were stained with hematoxylin and eosin (H&E; cat. nos. 234-12 and 1159350025, Merck) and examined histologically. Analysis of DNA fragmentation by TUNEL assay was performed using a commercial kit (cat. no. 11684817910, Roche Applied Science, Penzberg, Germany). Cells were examined under a light microscope (Axioskop 2 Plus) with an Achroplan 40x/0.65 lens (both from Zeiss, Hallbergmoos, Germany). Images were acquired with an AxioCam 503 color and AxioVision LE64 software (Zeiss).

Serum concentrations of human soluble IL-2R (sIL-2R; cat. no. DR2A00, R&D Systems, Inc., Minneapolis, MN, USA) and human soluble cluster of differentiation 30 (sCD30; cat. no. RBMS240R, BioVendor Inc., Brno, Czech Republic) were measured by enzyme-linked immunosorbent assay (ELISA), according to the protocol supplied by the manufacturer.

All values are expressed as the mean ± standard deviation (SD). Differences between the control and treated groups were tested for statistical significance by the unpaired t-test. A P-value <0.05 denoted the presence of statistically significant difference.

The suppressive effect of peridinin on the survival of HTLV-1-transformed T-cell lines MT-2, MT-4, HUT-102, C5/MJ, SLB-1, and patient-derived ATL cell lines MT-1 and ED-40515(-) was confirmed by WST-8 assay. Furthermore, peridinin dose-dependently suppressed the viability of HTLV-1-infected T-cell lines (Fig. 2). The IC₅₀ values of peridinin on HTLV-1-infected T-cell lines as estimated by the WST-8 assay were 0.71–5.38 μM (Table I).

Disturbance of the cell cycle is one of therapeutic targets for development of new anticancer drugs. The MT-2 and HUT-102 cell lines were treated with different concentrations of peridinin for 24 h and cell cycle was analyzed by propidium iodide staining and DNA content analysis by flow cytometry. Low-dose peridinin (1.25, 2.5 and 5.0 μM) significantly reduced the proportion of cells in the S phase but increased those of cells in the G₁ phase population (Fig. 3A and B). These results suggest that peridinin causes G₁ cell cycle arrest in HTLV-1-infected T-cell lines. The increase in G₁ cell cycle arrest was peridinin dose- and time-dependent (Fig. 3). On the other hand, the percentages of cells in the sub-G₁ phase of the high-dose group (10 μM) (MT-2, 28.4%; HUT-102, 21.3%) were significantly higher than those of low-dose groups and control (MT-2, 3.1%; HUT-102, 6.9%) at 24 h (Fig. 3A). Furthermore, peridinin at high dose markedly increased the percentage of apoptotic cells of sub-G₁ population and this effect was time-dependent (Fig. 3C).

MT-2 and HUT-102 cells were treated with different concentrations of peridinin for 24 h and apoptosis was assessed by APO2.7 staining. There was no significant difference in rates of apoptosis between the low-dose groups (1.25, 2.5 and 5.0 μM) and negative control at 24 h, while high-dose peridinin (10 μM) induced significant apoptosis of MT-2 and HUT-102 cells (Fig. 4A). As shown in Fig. 4B, by Hoechst-stained MT-2 cells showed distinct morphological features of chromatin condensation and fragmented nuclei in the presence of 10 μM peridinin. In another method for apoptosis based on caspase-3, -8, -9 and PARP, increases in cleaved caspase-3, -8, -9 and PARP expression levels were detected in MT-2 cells treated with 10 μM peridinin (Fig. 4C).

Apoptosis is regulated by a balance between pro-apoptotic and anti-apoptotic proteins. The Bcl-2 family members play major roles in cell survival and apoptosis (11). For this reason, we measured the effects of peridinin on the expression of Bcl-2, Bcl-xL and Bak by western blot analysis. Peridinin inhibited the expression of Bcl-2 in a dose-dependent manner, but not that of Bcl-xL and Bak. Inhibitor of apoptosis (IAP) proteins, including XIAP, c-IAP2 and survivin directly bind to activated caspase-3, -7 and -9, and inhibit their activities (12). As shown in Fig. 5, peridinin reduced the protein levels of survivin and XIAP, but not c-IAP2, in cultured MT-2 cells and the effect was dose-dependent. These results suggest that peridinin inhibits cell survival and induces apoptosis through the regulation of Bcl-2 family members and IAPs.

Cell cycle progression is governed by a family of cyclins and CDKs, and cyclin D/CDK4/CDK6 is a critical determinant of progression through G₁ phase of the cell cycle (13). We next evaluated the effects of peridinin on cyclins and CDKs involved in cell cycle arrest in MT-2 cells. As shown in Fig. 5, peridinin significantly decreased cyclin D1, cyclin D2, CDK4 and CDK6 levels in MT-2 cells.

Proto-oncogene c-myc encodes a basic loop-helix-loop zipper transcription factor that plays crucial roles in cell proliferation, apoptosis, differentiation and metabolism (14,15). c-Myc can activate the expression of many downstream cell cycle regulators, such as cyclin D1, cyclin D2 and CDK4 (14–16). Peridinin downregulated c-Myc protein levels in MT-2 cells (Fig. 5), suggesting that c-Myc served as an upstream target in the peridinin-mediated blockade of cell cycle progression.

NF-κB encompases a family of transcription factors that are involved in numerous biological processes including cell growth and survival. The NF-κB proteins are usually sequestered in the cytoplasm by a family of inhibitors including IκBα. Activation of NF-κB occurs via phosphorylation of IκBα at Ser32 and Ser36. This is followed by proteasome-mediated degradation resulting in release and nuclear translocation of active NF-κB, where it regulates the expression of several pro-survival proteins and cell cycle regulatory molecules (17). Interestingly, many genes downregulated by peridinin (cyclin D1, cyclin D2, CDK4, CDK6, c-Myc, Bcl-2, XIAP and survivin) are also regulated by NF-κB (18–25). Culture of MT-2 cells in the presence of peridinin resulted in a significant dose-dependent inhibition of IκBα protein phosphorylation (Fig. 6A). Pursuing further this line of experiments, we examined, using the EMSA, the binding of NF-κB family proteins to the NF-κB sequence of the IL-2Rα gene. To validate the relevance of the visualized bands with regard to the presence and specificity of the NF-κB-DNA complex, we incubated nuclear extracts from MT-2 cells with antibodies specific for NF-κB family proteins as well as to an excess of unlabeled competitors NF-κB and AP-1 sequences. The band completely disappeared by excess unlabeled NF-κB sequence, but not by AP-1 sequence, and was affected by preincubation with antibodies specific for p50, RelA, c-Rel, p52 and RelB, indicating that this band contains the specific DNA-NF-κB subunit complex (Fig. 6B, right panel). Treatment with peridinin completely abolished this binding process (Fig. 6B, left panel), indicating that peridinin inhibits the NF-κB pathway. This inhibition can result from upstream signaling events or from direct inhibition of one or more stages of the NF-κB pathway. This latter possibility was assessed by determining the types of proteins along the NF-κB pathway that are direct targets for peridinin.

Attenuation of phosphorylation of IκBα (Fig. 6A) suggested that the upstream kinase IKKβ was the likely site of the inhibitory action of peridinin. To further explore this tenable conclusion, we examined IKKβ phosphorylation. The results showed that peridinin suppressed the phosphorylation of IKKβ (Fig. 6A).

RelA is phosphorylated at Ser536 by a variety of kinases through various signaling pathways and such phosphorylation enhances RelA transactivation potential (26). As shown in Fig. 6A, peridinin significantly inhibited phosphorylation of RelA in MT-2 cells. Akt is a component of an essential pathway for cell survival and growth during carcinogenesis, and it can also phosphorylate RelA at Ser536 through an IKKα- or IKKβ-dependent mechanism (26). Peridinin induced dose-dependent inhibition of Akt phosphorylation at Ser473 and Thr308 in MT-2 cells (Fig. 6C). Furthermore, peridinin suppressed the phosphorylation of IKKα and IKKβ (Fig. 6A). Considered together, the above results suggest that peridinin inhibits RelA phosphorylation by inhibiting Akt and IKK activation.

PDK1 is an immediate downstream mediator of phosphoinositide 3-kinase (PI3K) and is activated upon phosphorylation at Ser241 (27). Activated PDK1 controls cell proliferation and survival by further activating its downstream cAMP-dependent, cGMP-dependent, protein kinase C family of protein kinases, including Akt (28) and S6K (29). We next examined the effect of peridinin on PDK1 protein. Peridinin reduced phosphorylation and protein expression of PDK1 (Fig. 6C) and induced significant inhibition of S6K phosphorylation.

The tumor suppressor PTEN directly dephosphorylates phosphatidylinositol 3,4,5-triphosphate (PIP3) to PI(4,5)P2 and opposes the PI3K/Akt signaling (30). The oncogenicity of abnormal PI3K/Akt signaling is emphasized by the observation that deactivating mutations of the gene encoding PTEN are among the most frequently occurring in human malignancy (30). However, ATL cells do not harbor genetic alterations in PTEN but express high levels of PTEN that is highly phosphorylated through loss of tumor suppressor N-myc downstream-regulated gene 2 (31). It is believed that phosphorylation keeps PTEN in an inactive form in the cytoplasm (32). To evaluate whether peridinin increases PTEN activity by dephosphorylation, the phosphorylation status of PTEN was examined. However, peridinin did not affect PTEN phosphorylation (Fig. 6C). Erk pathway is also involved in HTLV-1 transforming protein Tax-mediated apoptosis protection (33), but peridinin did not inhibit Erk phosphorylation (Fig. 6C).

To determine the antitumor activity of peridinin in vivo, HUT-102 cells were subcutaneously injected into the postauricular region of the SCID mice. Mice were intraperitoneally injected with vehicle or peridinin at 8.5 mg/kg 5 days a week over a period of 21 days. Tumorigenesis was noted in all mice transplanted with HUT-102 cells (Fig. 7A), but none showed signs of discomfort and/or suffering until euthanasia. Mice treated with peridinin showed significantly inhibited tumor growth compared with the control group (P<0.005). Furthermore, the weight of the excised tumors was lower after treatment with peridinin compared with those of untreated mice, albeit statistically insignificant (Fig. 7B and C). In addition, mice seemed to tolerate treatment with peridinin without overt signs of toxicity or significant loss of body weight, similar to the vehicle-treated group (Fig. 7D).

Tumor sections from the treated mice were further examined by H&E staining. Apoptosis of tumor cells was noted in the peridinin-treated group, which was characterized by cytoplasmic condensation, chromatin hyperchromatism and condensation, and nuclear fragmentation (Fig. 7E, left panel). TUNEL staining also demonstrated the presence of abundant apoptotic cells in the tumors of the peridinin-treated group (Fig. 7E, right panel). These findings suggest that the observed inhibition of tumor growth in the peridinin-treated tumors in vivo was mainly due to increased apoptosis.

Finally, to validate the results of the in vivo xenograft model, the effect of peridinin on the surrogate tumor markers sIL-2R (34) and sCD30 (35) by ELISA was investigated in serum samples. Compared with the control group, treatment of mice with peridinin significantly reduced serum sIL-2R and sCD30 levels (Fig. 7F).