Antitumor Function and Mechanism of Phycoerythrin from Porphyra Haitanensis

The anti-tumor eff ect of R-Phycoerythrin (R-PE) from Porphyra haitanensis was studied using cell line HeLa as an in vitro model and Sarcoma-180 (S180) tumor-bearing mice as an in vivo model. The results showed that the combination treatment of R-PE and photodynamic therapy PDT) signifi cantly inhibited the growth of HeLa cells up to 81.5%, with a fair dose-eff ect relationship, but did not inhibit endothelial cells. The annexin v-fi tc/PI fl uorescence staining experiments demonstrated that at doses between 0~60μg/mL, apoptosis cells and later stage apoptosis cells or necrosis cells increased signifi cantly as the R-PE dosage increased. DNA electrophoresis showed that after R-PE+PDT treatment of HeLa cells for 24 hours, a light " smear " band between 100~400bp appeared to indicate the degradation of genomic DNA. The QRT-PCR results showed that R-PE+PDT treatment increased caspase-3 and caspase-10 gene expression and decreased the Bcl-2 gene expression level signifi cantly as the R-PE dose increased, i mplying that R-PE promoted HeLa cell a poptosis. Compared with untreated S180 tumor-bearing mice, R-PE injection signifi cantly inhibited the growth of S180 in tumor-bearing mice up to 41.3% at a dose of 300mg·kg-1. Simultaneously, the signifi cant increase of superoxide dismutase (SOD) activity in serum (p < 0.01) and the decrease of the malondialdehyde (MDA) level in liver suggests that R-PE improved the anti-oxidant ability of the S180 t umor-bearing mice, which may related to its antitumor eff ect. In addition, the R-PE caused a significant increase (p < 0.05) in the spleen index and thymus index, and a significant increase (p < 0.01) in lymphocyte proliferation, NK cell kill activity and the TNF-α level in the serum of S180 tumor-bearing mice. These results strongly suggest that the antitumor eff ect of R-PE from Porphyra haitanensis functioned by increa sing the immunity and antioxidant ability of S180 tumor-bearing mice, promoting apoptosis by increasing protease gene expression and TNF-α secretion.


INTRODUCTION
Phycoerythrin (PE) is an oligomeric protein formed in a (αβ) 6 hexamer in an annular architecture.The two diff erent subunits, α and β, have molecular weights of 17 to 21 KDa..Each α subunit contains 2 phycoerthrobilin (PEB) chromophores which link to Cys82 and Cys139 by a thioester bond.Each β subunit contains two PEBs linking to Cys82 and Cys158 and one phycourobilin (PUB) linking to two cysteines, Cys50 and Cys61.The PE hexamers can pile up to form rods which allow the pigment molecules to play an important role in light absorption, storage and transmission.The PE purifi ed from Porphyra haitanensis belongs to R type PE (R-PE), with characteristic absorption peak of 498 nm from its linked PUB which allows it to play a photosensitive role (1-4).
Photodynamic therapy (PDT), the activation of a tumorlocalized photosensitizer by light, is generally applied as a single modality for the treatment of a variety of solid tumors, and has received regulatory approval.Its dominant mechanism of action is the local generation of cytotoxic singlet oxygen, which causes the destruction of tumor cells and damage to the tumor microvasculature.PE is an attractive photosensitizer because of its good fl uorescence characteristics (2 PEB, 1 PUB) (5)(6)(7)(8); it has drawn considerable attention because of its low phototoxicity.Ling Hu et al. (9) studied the survival rate of the normal liver cell line HL7702 and liver cancer cell line HepG2 after treatment with 25~200 μg/mL of the β subunit of PE.Comparison of PE in four diff erent treatments, PE subunit alone (PE-sub), PE subunit with liposome package (PE-sub-lip), PE subunit with laser irradiation (PE-sub-PDT) and PE subunit with liposome package and laser irritation (PE-sub-lip-PDT), demonstrated that the use of the liposome carrier increased the PE-sub accumulation in the cells and enhanced its PDT eff ect on HepG2.
Caspase is a special family of proteases which play a central role in programmed cell apoptosis (10).Caspases can degrade anti-apoptotic protein and induce the release of mitochondrial cytochrome C. At the same time, caspase-mediated degradation can weaken the function of DNAase inhibitor ICAD (inhibitor of caspase DNAase), thus increasing intracellular Ca 2+ concentration, activating DNAase and encouraging the fragmentation of DNA as depicted in Fig. 1 (11)(12)(13)(14).
Fig. 1.Graphic summary of events after global brain ischemia during reperfusion.The activation of major, caspasedependant apoptosis pathways, the possible cross-talk between necrosis-inducing cytoplasmic-lysosomal proteases and their possible interaction with Bcl-2 and caspase protein families are depicted (12).
The objective of this investigation was to study the in vivo and in vitro anti-tumor eff ect of R-PE from Porphyra haitanensis.The results showed that R-PE had a strong in vivo and in vitro anti-tumor eff ect, and that PDT can improve the in vitro anti-tumor eff ect of R-PE.These results may provide a basis for developing the PE into a photosensitive drug for cancer therapy.

MATERIALS AND METHODS
PE from Porphyra haitanenis (R-PE), with food grade purity (A565/A280>0.7)and drug class purity (A565/A280>2.0)was obtained in the laboratory.The cervical carcinoma cell line HeLa, mouse ascitic tumor cell line S180 and KM mice (qualifi ed number: SCXKG Fujian 2004-0001) were obtained from the animal center of Xiamen University, Xiamen, China.Concanavalin (ConA) and MTT was purchased from Sigma (St Louis, MO, USA).

Extraction and purifi cation of phycoerythrin
Fresh Porphyra haitanenis was washed with 1 M PBS buff er (pH 6.8).One volume of washed algal mass was resuspended in fi fteen volumes of the same buff er and subjected to repeated freeze-thaw cycles of -20 ℃ and 4 ℃ for the release of PE.Purification was done by two-step ammounium sulphate precipitation (25% and 55%) followed by chromatography using a gel sephadex G-100 column (60cm×2cm) pre-equilibrated and eluted with 10mM PBS buff er (pH 7.0).The fl ow rate was maintained at 45ml•h -1 .Elute was collected and run through a DEAE-Cellulose ion exchange column (30cm×2cm) preequilibrated with 20mM PBS buff er (pH6.0).The R-PE rich fraction was eluted at 0.2M NaCl.The R-PE was stored at -20 ℃.At each stage of purifi cation the R-PE purity and molecular weight were determined by 12% SDS-PAGE (Fig. 1).The R-PE purity measured as the ratio of OD565/OD280 was also determined at each stage of purifi cation (Table 2).The UV-vis overlay spectrum for purifi ed R-PE is shown in Fig. 2.

Cell culture
Cervical carcinoma line HeLa was routinely grown in Dulbecco's Modifi ed Eagle's medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin and kept in an atmosphere of 95% air and 5% CO 2 in a 37 ℃ humidifi ed incubator.

PDT
In total, 5×10 4 /mL cells were inoculated in 96-well plates with serum-free DMEM and treated with 0, 3.5, 7, 15, 30 or 60μg/ mL R-PE for 4h before laser irradiation.The PDT excitation wavelength was 632.8 nm and energy intensity was 12.5 J/ cm 2 laser light; the two kinds of cells were cultured in DMEM containing 10% serum for 24h and 48h, respectively.Each concentration was repeated 6 times, along with a blank control (without R-PE or laser light irradiation) and a positive control group (treated with 20μg/mL 5-fl uorouracil).The survival rate of the PDT-treated HeLa cell lines was analyzed using the 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay.
Cell survival rate (%)=OD experimental group /OD blank group ×100% 95 °C for 30s, 60 °C for 30s, followed by a fi nal elongation at 74 °C for 5 min.As an internal control, a crab beta-actin cDNA fragment was amplifi ed with primers beta-actin-F and betaactin-R (Table 1) and the same PCR amplifi cation conditions.

3. Anti-tumor activity of R-PE on S180 tumor-bearing mice
2.3.1.Model establishment of S180 sarcoma S180 tumor cells in mice were reproduced by ascites serial passage 3 times, 7 days per passage.S180 tumor cells were extracted from the last tumor-bearing mice under sterile conditions.The cell concentration was adjusted to 1.4×10 7 /mL with cold sterile saline, then 0.2 mL cell suspension fl uid was inoculated in the left fore oxter of the mice.

Experimental design
Male mice (28±2 g/mouse) were used in the experiments.Seventy two mice were arranged into six groups (n = 12): The blank group served as a normal control and received distilled water at level of 0.2 ml/day for 9 days.The model group served as an S180 sarcoma group and received distilled water at level of 0.2 ml/day for 9 days.The positive drug group served as an S180 sarcoma group and received 5-FU (20 mg•Kg -1 bw•d -1 ) (15) for 9 days.The R-PE low-dose group served as an S180 sarcoma group and received R-PE (100 mg•Kg -1 bw•d -1 ).The R-PE middle-dose group served as an S180 sarcoma group and received R-PE (200 mg•Kg - 1 bw•d -1 ).The R-PE high-dose group served as an S180 sarcoma group and received R-PE (300 mg•Kg -1 bw•d -1 ).The mice were weighed every other day.At the end of the 9-day treatment, mice were sacrifi ced.Blood was collected using the eye socket and centrifuged at 1000×g for 10min.Serum was collected for antioxidative experiments and analyzed for TNF-αcontent.Liver and spleen were excised and used for antioxidative and cellular immunity experiments.

Tumor inhibitory rate and immune organ index
The tumor inhibitory rate and the immune organ index were calculated with the following formulae: Tumor inhibitory rate (%) = tumor weight of text group / tumor weight of blank group×100% Immune organ index (mg/10g) = weight of organs(mg) / weight of mouse(10g)×100%

TABLE 1
Nucleotide sequence of primers primer Sequence(5'-3') Cell apoptotic analysis by Annexin V-FITC / PI double fl uorescent At the exponential phase stage, 5×10 5 /mL cells were inoculated in 24-well plates with serum-free DMEM and treated with 0, 15, 30 or 60 μg/mL R-PE for 4h before laser irradiation.The PDT excitation wavelength was 632.8nm and energy intensity was 12.5 J/cm 2 laser light.The negative control and positive control groups were set up with treatments of 15, 30 or 60 μg/mL R-PE and with 20 μg/mL 5-fl uorouracil.The cells were cultured in DMEM containing 10% serum for 24h.After the cells were washed with cold 10 mM PBS, Annexin V-FITC / PI double fl uorescence dyeing liquid was added and the plates were placed in the dark for 15 min.The morphology of the HeLa cells was examined under a fl uorescence microscope.

RNA isolation and cDNA synthesis
Total RNA was extracted from HeLa cells treated with 0, 15, 30 or 60 μg/mL R-PE for 4h before laser irradiation, using an RNAiso plus kit according to the manufacturer's instructions.Each concentration was repeated 3 times.The extracted RNA was treated with RNase-Free DNase to remove contaminating DNA, and cDNA was synthesized using the PrimeScript RT Reagent Kit following the manufacturer's instructions.

mRNA expression of caspase and Bcl-2
Caspase and Bcl-2 cDNA were amplifi ed from the extracted total RNA (2 μg) using primers caspase-F, caspase-R and Bcl-2-F, Bcl-2-R, respectively (Table 1), under the following conditions: pre-denaturation at 95 °C for 2 min, 30 cycles of SOD enzyme activity and the MDA content in liver and serum were determined using the commercial kit according to the manufacturer's instructions.

Splenic lymphocyte proliferation capacity of tumor-bearing mice
Spleen cells were prepared in sterile conditions and were diluted to 2×10 6 /mL with the complete culture medium.The cell suspension liquid was inoculated in 96-well plates, 200 μL per well, 4 wells with each treatment group; three of them with 5μL 200 mg/ml ConA (fi nal concentration 5 mg/l) and one well set as blank control with 5μL of the same complete culture medium.The plates were incubated for 3d. 100 μL of supernatant liquid per well was taken 4h before the end of the incubation.After adding 20 μL of MTT solution (5mg/ml), the plates were incubated for 4 more hours.The proliferation capacity of spleen lymphocytes was analyzed using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (16,17).

Natural killing activity of NK cells
Spleen cells were prepared in sterile conditions as the eff ector cell and were diluted to 2×10 7 /mL with the complete culture medium.HeLa cells were prepared as the target cell and were diluted to 2×10 5 /mL with the complete culture medium.100 μL of the target cell suspension liquid was inoculated in 96well plates and 4 wells for each treatment group and cultivated for 8h.The isopycnic eff ector cells were added in three wells per group, adjusting the ratio of eff ector cells to target cells to 100 to 1.The remaining wells were set as two control groups of the target cell and the eff ector cell wells.The cells were incubated in an incubator at 37 °C with 5% CO 2 for 4h.Cell survival rate was determined by MTT assay(18) using the following formula.
Natural killing activity of NK cells (%)=[OD target cell group -(OD target cell treated with eff ector cell group -OD eff ector cell group )]/ OD target cell group 2.3.7.TNF-α secretion capacity of the spleen cells TNF-α secretion capacity of the spleen cells in mice was determined using enzyme-linked immunosorbent assay (ELISA) according to manufacturer instructions.

Histopathology
The tumor specimens were examined microscopically after routine preparation (19).The tumor-bearing mice treated with 5-FU and diff erent concentrations of R-PE were sacrifi ced 9 days after establishment of the S180 sarcoma model.Tumors were fi xed using 10% neutral buff ered formalin solution for 24h and then dehydrated in a series of diff erent concentrations of ethanol, infi ltrated with xylene and paraffi n and embedded in paraplast.
The specimens were serially sectioned (6 μm thickness) with a microtome and stained on microscope slides using haematoxylin and eosin (H&E).The stained slides were observed and photographed using a fl uorescence microscope (20).

Statistical analysis
Data were expressed as mean ± 1 SD for at least three experiments.Student's t-test was used to compare data.The data with P<0.05 were considered to be statistically signifi cant.

Extraction and purifi cation of R-PE
The purity and homogeneity of the R-PE extracted from Porphyra haitanensis were examined using spectroscopic methods and denaturing gel electrophoresis (Table 2 and Figs. 1 and 2).The purifi ed protein had three subunits, α, β and γ, corresponding to molecular weights of 18, 23 and 33KDa, and had its absorbance maximum at 565nm wavelength.

R-PE PDT eff ect on HeLa cells
In order to study R-PE as a photosensitizer on HeLa cells and its cell toxicity in normal cells, we treated cells with diff erent purity and concentrations of R-PE for 4h and irradiated them with 12.5 J/cm2 of HE-NE laser.As shown in Table 3, at a concentration of 60 μg/mL and purity ratio of 3.3, the combination of 4h incubation with the R-PE and laser irradiation resulted in cytotoxicity, where 18.6% ±0.59% of the HeLa cells survived, which is 34% ±0.32% (n=6, p<0.01) less than the same treatment but without laser irradiation and 1.5% ±0.12% less than 5-FU group.The half lethal dose of R-PE (purity 2.45) for the R-PE+PDT treatment was approximately 15 μg/mL.At this concentration the survival rate of HeLa reached 73.5% ±1.63% without laser irradiation.Cell death induced by R-PE on normal endothelial cells after R-PE+PDT treatment was examined at 3.5-60 μg/mL of R-PE.The results showed that R-PE has low cell toxicity on normal cells and good tumor-targeting characteristics.
As shown in Table 4, when R-PE was incubated with the cells during rinsing with PBS and cells were then irradiated with laser, no cytotoxicity was observed.These results revealed that the laser irradiation (12.5 J/cm2) alone produced no cytotoxicity.

Apoptotic analysis of R-PE+PDT-treated cells
To investigate the R-PE+PDT eff ect on cell apoptosis, HeLa cells treated with R-PE alone and the combination of PDT and diff erent concentrations of R-PE (15, 30 or 60 μg/mL) were harvested for morphological observation.As shown in Fig. 3, the blank group of HeLa cells showed tight connections (Figs 3a, 3e); after R-PE treatment the cells became more separated and loosely arranged with increasing concentration of R-PE (Figs. 3b, 3c, 3d).The combination treatments of R-PE and PDT changed HeLa cells into oval cells in suspension.A large number of cells lysed (Fig. 3h) after the treatment with the highest concentration of R-PE (60 μg/mL) and PDT.
For morphological observation of the PDT-treated cells, the cells were stained with annexin V-FITC and propidium iodide and examined under a fl uorescence microscope.The green fl uorescent spots represent viable apoptotic cells and the red fl uorescent spots represent non-viable apoptotic cells.As shown in Fig. 4, the non-viable apoptotic cells increased as the concentration of the R-PE increased (Fig. 4b, 4c, 4d), and the additional PDT treatment signifi cantly enhanced the eff ect.

Expression of caspase-3, casapse-10 and Bcl-2 mRNA
Expressions of Caspase-3, Caspase-10 and Bcl-2 mRNA were determined using RT-PCR.As shown in Fig. 5, with increasing concentrations of R-PE Caspase-3 and the Caspase-10 mRNA transcription increased.In contrast, Bcl-2 mRNA transcription decreased as the concentration of R-PE increased.
A quantitative real-time PCR assay was carried out to determine the transcript levels of Caspase-3, Caspase-10 and Bcl-2 in HeLa cells after R-PE+PDT treatment.Caspase-3 mRNA expression increased with increasing concentration of R-PE.At the concentration of 60 μg/mL R-PE Caspase-3 mRNA increased 22.1-fold in comparison to the control (Fig. 6A).However, the transcript levels of Caspase-10 mRNA were unchanged until the concentration of R-PE reached 60μg/mL, where the Caspase-10 mRNA expression increased 8.45-fold in comparison to the control (Fig. 6C).In contrast, Bcl-2 mRNA decreased signifi cantly with increasing concentration of R-PE.At the concentration of 60μg/mL, Bcl-2 mRNA expression decreased 0.15-fold compared to the control (Fig. 6B).All these results were consistent with results shown in Fig. 5.

Inhibitory eff ect of R-PE on S180 tumor-burdened mice
As shown in Table 5, the inhibition eff ect of R-PE on S180 had an obviouse dose eff ect.The R-PE in the high dose group (300mg•Kg -1 bw•d -1 ) showed much greater tumor inhibition rate, 41.3%±1.22%(n=12, p<0.01), than the 7.89±0.44 in the 5-FU

Histopathology of the S180 sarcoma sections of the mice
As shown in Fig. 7, the tumor cells in the model group were mononucleated or multinucleated in a large dense patch.
Tumor giant cells and the nucleus were hyperchromatic, nuclear division was easy to see and small tumor cells were observed in the cytoplasm .The tumor infi ltration of the soft tissue was deep, and less fi brous tissues were observed in the tu mor tissue.In contrast, the tumor cells in R-PE treatment group were arranged loosely, and tumor giant cells decreased signifi cantly.Very little tumor infi ltration of tumor biopsies was observed.The fibrous tissue increased significantly as the concentration of R-PE increased.A large number of lymphocytes and macrophages were observed to infi ltrate into the tumor tissue to kill the tumor cells.Therefore, R-PE showed signifi cant antitumor eff ect and obviously had a dose eff ect.(24,25).On the contrary, Bcl-2 protein can inhibit cell apoptosis by reducing membrane permeability of the mitochondria and preventing the release of cytochrome C (26).In vitro characterization studies revealed that exposure with R-PE to 632.8 nm laser light produced significant cytotoxicity on HeLa cells in a concentrationdependent manner.QRT-PCR studies showed that R-PE promoted the mRNA expression of caspase-3 and caspase-10, but inhibited the mRNA expression of Bcl-2 in a concentrationdependent manner.
It is generally acknowledged that cellular immunity plays a leading role in anti-tumor immunity.Cellular immunity is mediated by sensitized T cells and lymphatic factor; its characteristic is the infl ammation caused by cell infi ltrating, and lymphocyte T cells, macrophages, and the mechanism of NK cells are the important eff ect cells (27).In vivo, R-PE can improve lymphocyte proliferation capacity, the kill activity of NK cells of tumor-bearing mice and inhibit tumor growth.
Tumor necrosis factor a (TNF-a) secreted by mononuclear macrophages can combine with TNFR I to induce cell apoptosis by the death structure domain of the cell.It can also increase the membrane surface molecules of cell apoptosis (FAS) to express and through the FAS/FASL pathway induce the cell apoptosis.Meanwhile, TNF-a can inhibit the angiogenesis of the tumor and obviously increased the kill activity of the LAK cells, tumor infi ltrating lymphocyte cells and NK cells (28).The results from this study demonstrated that the anti-tumor effect of R-PE resulted through its anti-oxidative activity, improving apoptosis protease gene expression, immunity enhancement and promoting the secretion of the TNF-α.Therefore R-PE could be a potential drug in the prevention of cancer.
fl uorescence properties are a consequence of interactions between covalently bound bilin chromophores and the apoprotein(21).Phycoerythrin was found to be sensitive to heat and light in aqueous solution and has strong fl uorescence properties in vitro.Under laser irradiation(22), R-PE can absorb and transmit the light to the surrounding oxygen and produce singlet and reactive oxygen species components.These light reaction products can kill cancer cells(23).Antitumor drugs screening by cancer cells has been widely adopted.In vitro experiments showed that R-PE has a signifi cant inhibition on HeLa cells; this eff ect is obviously dose dependent and additional PDT treatment can obviously improve the eff ect.Cell apoptosis is a kind of energy dependence and programmed cell death.Caspase, a member of the aspartic acid specifi c homocysteine protease (caspase) family, can degrade the antiapoptotic protein and induce the release of cytochrome C of mitocondria.Caspase can also reduce the inhibitory eff ect of DNAase by ICAD (inhabitor of caspase DNAase) and improve the [Ca 2+ ] density of cells and activate the DNases, thus inducing cell apoptosis

TABLE 2
Total protein concentration and spectroscopic purity of Porphyra haitanensis R-PE at each stage of purifi cation

TABLE 3
PDT effect of R-PE on HeLa cell line.n=6.Mean ± SD.The blank group was the control.*P<0.01 vs. control