Introduction

Signal pathways are responsible for most biological processes such as cell growth, differentiation, migration and apoptosis. Abnormal expression of proteins corresponding to specific signal pathway is closely related with many serious diseases, especially cancer1,2,3. In recent decades, advances in genomics, proteomics, chemistry and protein engineering coalesce to accelerate the development of targeted anticancer drugs4,5,6. In these studies, small molecule drug-antibody conjugates that target specific proteins to retard cancer cell signal transduction to arrest tumor growth are most favored as a result of relatively low side effect and strong selectivity. However most efforts are made to inhibit protein activity temporarily rather than inactivate them permanently. In the latter case, the protein remains inactive after the inhibitor diffuses away and thus enhanced drug potency can be achieved.

Photodynamic therapy (PDT) has been widely used for various cancer treatments to kill tumor cells7. The cytotoxic agents, reactive oxygen species (ROS), are generated from the photosensitizer excited by appropriate light exposure. With the interactions between ROS and biomacromolecules, tumor cell apoptosis and necrosis can be initiated. Nevertheless, the surrounding healthy tissues are damaged at the same time. To minimize the side effects of conventional PDT, molecule-targeted PDT systems have been developed, yet with limited success8,9,10,11,12. In these systems, proteins targeted by antibodies or small molecules can be selectively inactivated by ROS generated by photosentizers under light, without affecting the surrounding biomolecules. These systems combine the benefits of targeting and inactivating and exhibit high spatial and temporal resolution in a noninvasive manner. However, these systems have two limitations: 1) proteomic or genetic modification is always needed9; 2) intracellular targets cannot be accessed by antibodies without microinject13. To the best of our knowledge, although selective photo-inactivation of proteins was studied8,9,13,14,15,16,17, selective inactivation of intracellar signal protein for cell growth inhibition and cancer treatment is seldom reported.

With the aim to develop a new molecule-targeted PDT system that can selectively target and kill the intracellular signal protein that tumor relies on efficiently with very low side effect, in this work, we designed multifunctional conjugated polymer-drug conjugates (PTD and PTDP, see their chemical structures in Figure 1A ). Small molecule drug was conjugated to polymer side chain for intracellular signal protein targeting. The mean particle sizes of PTD and PTDP are 31 and 83 nm from dynamic light scattering (DLS) experiments, respectively (as shown in Figure 1A ), which is in favor of the endocytosis18. With light irradiation, the conjugated polymer can sensitize oxygen to produce ROS12,17,19,20,21,22,23 that specifically inactivate the targeted protein and thus selectively inhibit the growth of tumor cells. The fluorescent properties of these conjugates can also serve to trace the cellular uptake and localization at different time points by fluorescence imaging. To the best of our knowledge, this is the first polymer/drug/photosensitizer conjugate by the special design and synthesis for intracellular molecule-targeted photodynamic therapy that combines intracellular targeting of a particular protein (estrogen receptor) and its photoinduced inactivation8,24.

Figure 1
figure 1

(A) Chemical structures of conjugated polymer-drug conjugate PTD and PTDP and dynamic light scattering analysis (DLS) of their aggregates in aqueous solution. (B) Schematic mechanism of PTDP for selective targeting and inactivation of intracellular estregen signal pathway protein.

Full size image

Results

The mechanism of our new conjugated polymer-drug conjugates (PTD and PTDP) for selective targeting and inactivation of intracellular signal protein is shown in Figure 1B . The estrogen receptor α (ERα) can be activated by estrogen, which mediates nuclei receptor signal pathways and influences cell growth and proliferation. Estrogen-mediated growth of human tumors (such as breast tumor) can be inhibited by inactivating ERα using antiestrogen drugs24,25. Tamoxifen (TAM), the most widely used estrogen receptor modulator26,27,28, was linked to side chains of PTD and PTDP via an oligo-ethyleneglycol (OEG) linker. It was reported that the permanently charged tamoxifen derivatives can decrease side effects, display good affinity to ERα and still modulate ERα-mediated transcription28. It is expected that similar modification of tamoxifen to PTD and PTDP would not deprive the activity of tamoxifen. Under white light irradiation, the conjugated polymer PTD can sensitize oxygen to produce ROS that specifically inactivated the targeted ERα protein, retard the estrogen signal pathway and selectively inhibit the growth of the signal pathway relied breast tumor cells. Covalent attachment of porphyrin moieties to the light harvesting backbone of PTD yields conjugate PTDP, which constrains interchromophore distances for optimizing energy transfer (ET) from polythiophene to porphyrin. This design can increase the ROS generation efficiency and reduce light intensity and polymer concentration requirements12,22. Furthermore, the fluorescent properties of the conjugated polymers can also serve to trace the cellular uptake and localization at different time points by fluorescence imaging.

Figure 2 shows the synthesis of PTD, PTDP and a cationic polythiophene derivative PTN using OEG modified thiophene monomer. Reaction of 2-(thiophen-3-yl)ethanol (1) and 4-methylbenzene-1-sulfonyl chloride in the presence of pyridine in CH2Cl2 affords 2-(thiophen-3-yl)ethyl 4-methylbenzenesulfonate (2) in 95% yield. Treatment of 2 with sodium hydride and tetraethylene glycol in anhydrous THF provides compound 3 in 20% yield. Compounds 4 and 6 were prepared by similar procedure as compound 2 in 80% and 95% yields, respectively. Compound 5 was obtained by similar procedure as compound 3 in 50% yield. Treatments of 6 with LiBr and NaI in acetone provide monomer 7 and 8 in 40% and 85% yields, respectively. Monomers 3 and 8 undergo oxidative copolymerization under nitrogen in the presence of FeCl3 to give polythiophene derivative 9 in 39% yield. Subsequent conversion to cationic polymer PTN is accomplished in 94% yield by reaction with excess trimethylamine in methanol. Modification of polymer 9 with tamoxifen affords PTD in 95% yield. The actual tamoxifen content was determined to be ~60% from 1H NMR spectra. Oxidative copolymerization of monomers 3, 7 and 1012 under nitrogen in the presence of FeCl3 gives polythiophene derivative 11, followed by modification with tamoxifen and purification by dialysis in water to afford PTDP in 36% yield. The actual tamoxifen and porphyrin contents were determined to be ~47% and 3.5%, respectively from1H NMR spectra. Due to the amphiphilic character, PTD and PTDP form aggregates in water. By DLS experiments, the mean particle sizes of PTD and PTDP are 31 and 83 nm, respectively (as shown in Figure 1A ), which are in favor of the endocytosis18.

Figure 2
figure 2

Synthetic routes of PTN, PTD and PTDP.

Full size image

The UV-vis absorption spectrum of control polymer PTN exhibits maximum peak at 410 nm, while PTD has an additional peak below 300 nm that originates from tamoxifen. The UV-vis absorption spectrum of PTD shows a maximum peak at 423 nm in water with a quantum yield of 6% with quinine bisulfate as the standard. The fluorescent property of PTD can be used for tracing the cellular uptake and localization. The absorption spectrum of PTDP exhibits a peak below 300 nm, a sharp peak at 420 nm and Q bands between 520 nm and 670 nm, which corresponds to tamoxifen, the polymer backbone and porphyrin units. For monomer 10, the absorption spectrum exhibits a Soret band at 416 nm and Q bands between 520 nm and 670 nm; the emission spectrum shows a maximum peak at 653 nm ( Figure 3A ) with an excitation of 420 nm and there is very low fluorescence excited by 450 nm and 470 nm. Excitation of PTDP at 450 or 470 nm, where the porphyrin units do not exhibit absorption, leads to emission with peaks at 578 nm and 658 nm. The peak at 658 nm demonstrates efficient energy transfer from polythiophene units to the porphyrin sites ( Figure 3B ). To verify the ROS producing abilities by PTD and PTDP, 2,7-dichloriflurescin-diacetate (DCFH-DA) was used for semi-quantitative analysis. After conversion of DCFH-DA into 2,7-dichloriflurescin (DCFH), DCFH can transform into highly fluorescent 2,7dichlorofluorescein (DCF) in the presence of ROS22. In these experiments, the concentrations of PTD and PTDP were 5.0 μM in repeat units (RUs) and the concentration of monomer 10 (0.2 μM) matched that of the porphyrin units in the PTDP solutions was also measured for comparison. As shown in Figure 4 , linear relations between the fluorescent intensity of DCF at 530 nm and irradiation time were observed for monomer 10, PTD and PTDP, respectively. The slope of PTDP is 2 times larger than that of PTD and 8 times larger than that of monomer 10. Thus PTDP with porphyrin moieties as side chains thus significantly increases the ROS generation relative to the isolated PTD and monomer 10 due to the efficient energy transfer from the polythiophene backbone to the porphyrin sites.

Figure 3
figure 3

Normalized UV-vis absorption and fluorescent emission spectra of monomer 10 (A) and PTDP (B) with excitations of 420 nm, 450 nm and 470 nm.

[PTDP] = 5.0 μM in RUs, [10] = 0.2 μM.

Full size image
Figure 4
figure 4

DCF emission intensity at 530 nm as a function of light irradiation time in aqueous solution in the presence of monomer 10, PTD and PTDP.

[PTD] = [PTDP] = 5.0 μM in RUs, [10] = 0.2. The fluence rate of white light (400–800 nm) is 10 mW·cm−2.

Full size image

Discussion

PTD was firstly studied for cell imaging. The distribution of PTD in ERα positive cell line MCF-7 was traced using fluorescence microscope. CLSM study showed the cellular uptake and localization of PTD at different time points ( Figure 5A ). After 2 h incubation with PTD, significant uptake in MCF-7 cells was observed and PTD mainly located the same place with lysosome stain Lyso Traker DND 99, which revealed the endocytosis process of PTD. After 18 h, more polymers escaped from the lysosome, which is in accordance with the reports that polythiophene are usually dispersed in cytoplasm nonspecifically. PTD was further tested on the ERα positive cell line MCF-7 to study the influence of cell proliferation and thus verify the targeting effect. A standard assay was used in which the conversion of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium hydrobromide) into formazan is related to mitochondrial activity and thereby cell viability. Control experiment indicates that PTN shows low toxicity to MCF-7 cells with the concentration up to 100 μM. For PTD, the cell viability of MCF-7 tumor cells decrease in a concentration dependent manner ( Figure 5B ). The different concentrations of tamoxifen in PTD can be calculated ( Figure 5C ). At the same effective concentration of tamoxifen, polymer-tamoxifen conjugate (PTD) exhibits significantly enhanced cell growth inhibition in comparison with tamoxifen itself. Two possible reasons may result in the enhanced cell growth inhibition. Firstly, the polymer–drug system with an appropriate size in nanometer scale (31 nm) that can be endocytosed and enriched in tumor cells. Secondly, local high concentration of tamoxifen in polymer can enhance its binding ability to target protein, ERα.

Figure 5
figure 5

(A) The cellular location of PTD at different incubation time: 2 h (upper) and 18 h (lower). The false color of PTD is green; the false color of Lyso Tracker DND 99 (lysosome stain) is red; their merge is yellow. (B, C) Dose-response curves for cell viability of MCF-7 Cells treated with tamoxifen or PTD as functions of PTD molar concentration (A) or mass concentration of effective TAM (B). (D) Reverse transcriptase-PCR analysis of mRNA. Images of DNA gel electrophoresis after reverse transcription-PCR reaction. After seeded for 12 h, MCF-7 cells are incubated under dark, group 1: without addition for 34 h; group 2: with PTD for 34 h; group 3: without addition for 24 h and then with 10 nM 17-β-estrodiol for 10 h; group 4: with PTD for 24 h and then 10 nM 17-β-estrodiol added for 10 h. The housekeeping gene GAPDH is used as the internal reference. (E) Semi-quantitative analysis of mRNA expression. Red: group 1 as control and group 2 as treated; green: group 3 as control and group 4 as treated. Error bars correspond to standard deviations from three separate measurements.

Full size image

ERα-mediated transcription was analyzed to further confirm the efficient targeting effect of PTD. Tamoxifen inhibits the regulation of specific gene by estrogen and the same inhibition of PTD was expected. We know that the binding effect of 17-β-estrodiol to ERα is about 1000 times more powerful than tamoxifen and the tamoxifen-induced mRNA expression should be reversed with incubation with 17-β-estrodiol. As shown in Figure 5D , expression of estrodiol-responsive gene pS2 and PDZK1 was found to be repressed by PTD. The mRNA expression of the three genes is about two thirds relative to the control. It is noted that the mRNA expression of pS2 and PDZK1 was recovered after the addition of 17-β-estrodiol 10 hours before cell lysis ( Figure 5E ). The reversible suppression of mRNA transcription confirms that PTD can target ERα and inhibit estrogen-induced activity of ERα.

As shown in Figure 4 , PTD possess the ability to sensitize oxygen to generate ROS under light irradiation due to its conjugated backbone. We subsequently studied whether PTD could inactive the targeted ERα and thus enhance the growth inhibition of ERα positive cells. Cytotoxicity experiments on PTD, PTN and tamoxifen were carried out with white light irradiation at fluence rates of 2 mW ·cm−2 for 30 min (3.6 J·cm−2) and 8 mW·cm−2 for 30 min (14.4 J·cm−2). The control experiments show that PTN and tamoxifen exhibit no further cytotoxicity under white light irradiation even at the fluence rate of 8 mW·cm−2 ( Figure 6A and 6B ). As shown in Figure 6C , PTD exhibits little further cytotoxicity under white light irradiation at a fluence rate of 2 mW, nevertheless it displays significantly enhanced cytotoxicity by irradiated at the fluence rate of 8 mW·cm−2. However, the growth of the MCF-7 cells is affected at higher fluence rate irradiation (8 mW·cm−2). Since the PTDP with porphyrin moieties as side chains can significantly increase the ROS generation efficiency relative to PTD, MCF-7 cells treated with PTDP displays quite remarkable sensitivity to light irradiation as expected. As shown in Figure 6D , the inhibition of cell proliferation under light irradiation at a fluence rate of 2 mW·cm−2 was greatly enhanced than that under dark. We summarized the cell viability of PTD and PTDP in different conditions ( Table 1 ), which shows that PTDP exhibits better inhibition activity to cell growth even at lower concentration and light fluence rate in comparison to PTD. It is noted that PTDP shows little cytotoxicity towards the ERα negative MDA-MB-231 cells under white light irradiation at a fluence rate of 2 mW·cm−2 ( Figure 6E ). In this case, PTDP diffuses in the cytoplasm randomly and the generated ROS spreads quickly and may only cause weak collateral damage. Thus, cell growth inhibition was achieved by selectively retarding the intracellar signal transduction pathway using polymer-drug conjugates. We further studied whether 17-β-estrodiol was able to reverse the irradiation-induced growth inhibition, since excess 17-β-estrodiol can reduce the target effect of tamoxifen and protect the ERα from ROS damaging. As shown in Figure 6F , the growth of MCF-7 cells was inhibited after PTDP treatment and the inhibition was reversed when 17-β-estrodiol was added to the culture medium 2 h before light irradiation. The selective inactivation of intracellar signal protein for cell growth inhibition will contribute to the development of molecular targeted photodynamic therapy and new strategy of tumor treatment.

Table 1 Comparison of PTD and PTDP for inhibition activity to cell growth
Full size table
Figure 6
figure 6

Dose-response curves for cell viability of MCF-7 cells treated with TAM (A), PTN (B) and PTD (C) by using a typical MTT assay under light irradiation or in the dark.

Dose-response curves for cell viability of MCF-7 cells (D) and MDA-MB-231 cells (E) treated with PTDP by using a typical MTT assay under light irradiation or in the dark. (F) Cell viability of MCF-7 cells treated with PTDP in the dark, under light irradiation or under light irradiation with 17-β-estrodiol added. The the 17-β-estrodiol with identical concentration to PTDP was added 2h before irradiation. Cells were irradiated with white light (400–800 nm). The concentrations of polymer are in RUs. Error bars correspond to standard deviations from three separate measurements.

Full size image

Methods

Materials and instruments

Tamoxifen was purchased from Beijing Zhongsheng Huateng technology (Beijing, China) and confirmed by 1H NMR. 17-β-estrodiol was purchased from Aladdin. Other chemicals were purchased from Acros, Aldrich Chemical Company or Alfa-Aesar and used without purification. All organic solvents were purchased from Beijing Chemical Works and used as received. MCF-7 and MDA-MB-231 cell lines were obtained from cell culture center of Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences (Beijing, China). Fetal bovine serum (FBS) was purchased from Sijiqing Biological Engineering Materials (Hangzhou, China). Dulbecco's modified Eagle medium (DMEM) and L-15 medium (Leibovitz) were purchased from HyClone/Thermofisher (Beijing, China). RNAprep pure Cell/Bacteria Kit and Quant Script RT Kit were purchased from Tiangen Biotech (Beijing, China). The absorbance for MTT analysis was recorded on a microplate reader (BIO-TEK Synergy HT, USA) at a wavelength of 570 nm. Water used in cell culture was purified by a Millipore filtration system. The1H NMR and 13C NMR spectra were recorded on a Bruker Avance 400 MHz spectrometer. Mass spectra were recorded on a Waters GCT spectrometer for high resolution mass spectra (HRMS), a SHIMADZU LCMS-2010 spectrometer for ESI. Elemental analyses were carried out with a Flash EA1112 instrument. UV-Vis absorption spectra were taken on a JASCO V-550 spectrophotometer. Fluorescence spectra were measured on a Hitachi F-4500 fluorometer equipped with a xenon lamp excitation source. The images of gel electrophoresis were taken by a Bio-Rad Molecular Imager ChemiDoc XRS system and quantitative analysis data were obtained using Quantity One software (Version 4.6.5). Confocal laser scanning microscopy (CLSM) characterization was conducted with a confocal laser scanning biological microscope (FV1000-IX81, Olympus, Japan). The white light source (400–800 nm) was provided by a metal halogen lamp (MVL-210, Mejiro Genossen, Japan). The intensity of the incident beam was determined by a radiometer (Photoelectric Instrument Factory (Beijing Normal University). The size of polymer systems was measured on a Nano ZS (ZEN3600) system.

Synthesis of 2-(thiophen-3-yl)ethyl 4-methylbenzenesulfonate (2)

To dichloromethane (40 mL), 2-(thiophen-3-yl)ethanol (5.1 g, 40 mmol) and pyridine (4.0 g, 50 mmol) were added and the mixture was kept in 0°C. 4-methylbenzene-1-sulfonyl chloride (9.5 g, 50 mmol) in dichloromethane (20 mL) was then added slowly. The mixture was stirred for 24 h under room temperature. After that, the mixture was washed with excess HCl solution and then dried over MgSO4. The solvent was removed and the residue was purified by silica gel column chromatography using petroleum ether/ethyl acetate (10:1) as the eluent to afford a white solid (10.7 g, 95%). 1H NMR (400 MHz, CDCl3, δ): 7.74 (d, 2H), 7.32(d, 2H), 7.24(dd, 1H), 6.98 (d, 1H), 6.88 (d, 1H), 4.23 (t, 2H), 3.01 (t, 2H), 2.46 (s, 3H).

Synthesis of 3-(2-(2-(2-(2-(2-hydroxyethoxy)ethoxy)ethoxy)ethoxy)ethylthiophene (3)

To a solution of sodium hydride (60% in oil dispersion, 3.0 g, 75 mmol) in anhydrous THF (60 ml) was added tetraethylene glycol (14.5 g, 75 mmol) drop by drop. After stirring at room temperature for 30 min, compound 2 (7.0 g, 25 mmol) in anhydrous THF was added to the mixture and allowed to react at 40°C for 12 h. The resulting mixture was quenched by distilled water and extracted with CHCl3 three times. The combined organic layer was washed with water. Then the solvent was removed and the residue was purified by silica gel chromatography using petroleum ether/ dichloromethane/ethyl acetate/dimethoxyethylene (10:10:10:1) as the eluent to afford a colorless oil (1.5 g, 20%). 1H NMR (400 MHz, CDCl3, δ): 7.21 (dd, 1H), 7.00 (d, 1H), 6.95 (d, 1H), 3.56-3.72 (m, 18H), 2.90 (t, 2H). 13C NMR (100 MHz, CDCl3, δ): 139.1, 128.4, 125.1, 121.1, 72.51, 71.40, 70.53, 70.48, 70.22, 70.14, 61.61, 30.52. HRMS (ESI) m/z: [M+H]+ calcd. 305.1417; found 305.1411.

Synthesis of 14-(thiophen-3-yl)-3,6,9,12-tetraoxatetradecyl 4-methylbenzenesulfonate (4)

To dichloromethane (30 mL), compounds 3 (2.0 g, 6.5 mmol) and pyridine (2.0 g, 26 mmol) were added and the mixture was kept in 0°C. 4-methylbenzene-1-sulfonyl chloride (2.5 g, 13 mmol) in dichloromethane (20 mL) was then added slowly. The mixture was stirred for 48 h under room temperature. After that, the mixture was washed with excess HCl solution and then dried over Na2SO4. The solvent was removed and the residue was purified by silica gel column chromatography using petroleum ether/ethyl acetate/dimethoxyethylene (30:10:1) as the eluent to afford a colorless oil (2.4 g, 80%). 1H NMR (400 MHz, CDCl3, δ): 7.77 (d, 2H), 7.32 (d, 2H), 7.22 (dd, 1H), 7.01 (d, 1H), 6.96 (d, 1H), 4.14 (t, 2H), 3.56–3.72 (m, 16H), 2.89 (t, 2H), 2.42(s, 3H). 13C NMR (100 MHz, CDCl3, δ):144.9, 139.3, 133.1, 129.9, 128.6, 128.1, 125.3, 121.2, 71.58, 70.87, 70.71, 70.65, 70.34, 69.35, 68.79, 30.76, 21.77. ESI-MS m/z: [M+Na]+ calcd. 481.1; found 481.2. C21H30O7S2: calcd. C 55.00, H 6.59; found C 54.84, H 6.58.

Synthesis of compound 5

To a solution of sodium hydride (60% in oil dispersion, 200 mg, 5 mmol) in anhydrous THF (30 mL) was added tetraethylene glycol (1.4 g, 7.5 mmol) drop by drop. After stirring at room temperature for 30 min, compound 4 (1.1 g, 2.5 mmol) in anhydrous THF was added to the mixture and allowed to react at 40°C for 12 h. The resulting mixture was quenched by distilled water and extracted with CHCl3 three times. The combined organic layer was washed with water. Then the solvent was removed and the residue was purified by silica gel chromatography using petroleum ether/ethyl acetate/methanol/dimethoxyethylene (10:30:1:1) as the eluent to afford a colorless oil (600 mg, 50%). 1H NMR (400 MHz, CDCl3, δ): 7.24 (dd, 1H), 7.01 (d, 1H), 6.96 (d, 1H), 3.55–3.72 (m, 34H), 2.92 (t, 2H) 2.51 (br, 1H). 13C NMR (100 MHz, CDCl3, δ): 139.3, 128.6, 125.4, 121.2, 72.64, 71.60, 70.70, 70.46, 70.35, 69.59, 61.86, 30.76. HRMS (ESI) m/z: [M+H]+ calcd. 481.2466; found 481.2462.

Synthesis of compound 6

To dichloromethane (8 mL), compounds 5 (500 mg, 1.0 mmol) and pyridine (420 mg, 5.2 mmol) were added and the mixture was kept in 0°C. 4-methylbenzene-1-sulfonyl chloride (570 mg, 3 mmol) in dichloromethane (2 mL) was then added slowly. The mixture was stirred for 48 h under room temperature. After that, the mixture was neutralized with HCl solution and then dried over Na2SO4. The solvent was removed and the residue was purified by silica gel column chromatography using petroleum ether/ dichloromethane/ethyl acetate/dimethoxyethylene (10:10:10:1) as the eluent to afford a colorless oil (630 mg 95%). 1H NMR (400 MHz, CDCl3, δ): 7.79 (d, 2H), 7.34 (d, 2H), 7.24 (dd, 1H), 7.03 (d, 1H), 6.98 (d, 1H), 4.16 (t, 2H), 3.55–3.72 (m, 32H), 2.93 (t, 2H), 2.45 (s, 3H). 13C NMR (100 MHz, CDCl3, δ): 144.9, 139.3, 133.1, 129.9, 128.6, 128.1, 125.3, 121.2, 71.58, 70.87, 70.69, 70.34, 69.36, 68.79, 30.76, 21.76. ESI-MS m/z: [M+K]+ calcd. 673.2; found 673.3. C29H46O11S2: calcd. C 54.87, H 7.30; found C 54.68, H 7.26.

Synthesis of compound 7

To a solution of compound 6 (516 mg, 0.8 mmol) in anhydrous acetone (25 mL) was added LiBr·H2O (1.6 g, 1.6 mmol) under nitrogen. The mixture was stirred and reflux at 75°C for 12 h. The mixture was concentrated under vacuum and residue was purified by silica gel column chromatography using petroleum ether/ dichloromethane/ethyl acetate/dimethoxyethylene (10:10:10:1) as the eluent to afford a colorless oil (200 mg, 40%). 1H NMR (400 MHz, CDCl3, δ): 7.24 (dd, 1H), 7.02 (d, 1H), 6.97 (d, 1H), 3.81 (t, 2H), 3.55–3.72 (m, 34H), 3.47 (t, 2H), 2.92 (t, 2H). 13C NMR (100 MHz, CDCl3, δ): 139.3, 128.5, 125.2, 121.1, 71.50, 71.25, 70.70, 70.63, 70.28, 30.70, 30.30. ESI-MS m/z: [M+NH4]+ calcd. 560.2; found 560.3.

Synthesis of compound 8

To a solution of compound 6 (530 mg, 0.8 mmol) in anhydrous acetone (40 mL) was added NaI (1.2 g, 8 mmol) under Nitrogen. The mixture was stirred and refluxed at 75°C for 12 h. The mixture was concentrated under vacuum. Water was added and the mixture was extracted with CH2Cl2. The combined organic layer was dried over Na2SO4. The solvent was removed and the residue was purified by silica gel column chromatography using petroleum ether/ dichloromethane/ethyl acetate/dimethoxyethylene (10:10:10:1) as the eluent to afford a colorless oil (420 mg, 85%). 1H NMR (100 MHz, CDCl3, δ): 7.24 (dd, 1H), 7.02 (d, 1H), 6.97 (d, 1H), 3.74 (t, 2H), 3.56–3.72 (m, 30H), 3.25 (t, 2H), 2.92 (t, 2H). 13C NMR (400 MHz, CDCl3, δ): 139.3, 128.6, 125.3, 121.3, 72.09, 71.60, 70.71, 70.35, 30.76. HRMS (ESI) m/z: [M+H]+ calcd. 591.1483; found 591.1479.

Synthesis of polymer 9

A suspension of anhydrous FeCl3 (100 mg, 0.6 mmol) in CHCl3 (15 mL) was stirred for 30 min at room temperature under nitrogen. To this suspension was added a solution of monomer 3 (20 mg, 0.067 mmol) and monomer 8 (44 mg, 0.075 mmol) in CHCl3 (10 mL) and the resulting solution was stirred for 2 days at room temperature. The solvent was removed gently and the residue was dissolved with methanol. The solution that phosphorus pentoxide directly dissolved in water was added. After stirring for a while, the supernatant was collected and the precipitate was washed repeatedly by methanol. The combined solution was dried and the residue was dissolved in DMSO/H 2 O (1:10). Then the solution was dialyzed through a membrane with a molecular weight cutoff of 3500 for 3 days to yield a red sticky substance (25 mg, 39%). 1H NMR (400 MHz, CDCl3, δ): 6.8–7.2 (br), 3.75 (t), 3.4–3.74 (br), 3.25 (t), 3.10 (br), 2.80–3.00 (br).

Synthesis of PTN

To a solution of PTI (7 mg) in THF (1 mL), trimethylamine (1 mL, 33% in methanol, 4.2 mmol) was added. The resulted solution was stirred for 72 h at 60°C with trimethylamine extra added due to the not well-sealed system. The solution was removed under vacuum. The residue was dissolved in water and then dialyzed through a membrane with a molecular weight cutoff of 3500 for 3 days to yield a red sticky substance (7 mg, 94%). 1H NMR (400 MHz, CDCl3, δ): 6.8–7.2 (br), 4.4 (br), 3.93 (br), 3.88(br), 3.75 (m), 3.43–3.74 (br), 3.30 (s), 3.03–3.23 (br), 2.50–2.80 (br).

Synthesis of PTD

To a solution of PTI (15 mg) in methanol/CHCl3 (1:1, 3 mL), tamoxifen (125 mg, 0.33 mmol) was added. The resulted solution was stirred for 48 h at 72°C with solvent extra added. The solution was removed under vacuum and the residue was dissolved in little CHCl3. The solution was precipitated into petroleum ether repeatedly to remove the unreacted tamoxifen and yield a red sticky substance (21 mg, 95%).1H NMR (400 MHz, CDCl3, δ): 7.31 (m, 2H), 7.10–7.25 (m, 9H), 6.79 (d, 2H), 6.54 (d, 2H), 4.31 (br, 2H), 3.9–4.2 (br, 6H), 3.50–3.85 (br, 40H), 3.42 (br, 6H), 2.80–3.20 (br, 3H), 2.43 (q, 2H), 0.90 (t, 3H).

Synthesis of PTDP

A suspension of anhydrous FeCl3 (100 mg, 0.6 mmol) in CHCl3 (10 mL) was stirred for 30 min at room temperature under nitrogen. To this suspension was added a solution of monomer 3 (8 mg, 0.03 mmol), monomer 7 (22 mg, 0.04 mmol) and monomer 10 (4 mg, 0.004 mmol) in CHCl3 (10 mL) and the resulting solution was stirred for 2 days at room temperature. The solvent was removed gently and the residue was dissolved with methanol and CHCl3. The solution that directly dissolves phosphorus pentoxide in water was added. After stirring for a while, the organic layer was collected and the precipitate was treated three times repeatedly. The combined solution was dried under vacuum and the residue was added to a solution (methanol:CHCl3 = 1:1, 3 ml) of tamoxifen (500 mg, 1.3 mmol) and KI (45 mg, 0.3 mmol). The resulted solution was stirred for 48 h at 72°C with solvent extra added. The solution was removed under vacuum and the residue was dissolved in little CHCl3. The solution was precipitated into petroleum ether repeatedly to remove the unreacted tamoxifen, then dissolved in water and dialyzed through a membrane with a molecular weight cutoff of 3500 for 3 days to yield a red sticky substance (18 mg, 36%). 1H NMR (400 MHz, CDCl3, δ): 8.83 (br, 8H), 8.13 (br, 8H), 7.52 (br, 6H), 7.10–7.40 (m, 162H), 6.78 (d, 26H), 6.54 (d, 26H), 4.31 (br, 28H), 3.9–4.2 (br, 80H), 3.20–3.85 (br, 700H), 2.80–3.20 (br, 52H), 2.70 (s, 9H), 2.44 (t, 26H), 1.70–2.10 (br, 8H), 0.90 (t, 39H), −2.76 (s, 2H).

Cell culture

MCF-7 cells were cultured in DMEM and MDA-MB-231 cells were cultured in L-15 medium, supplemented with 10% FBS at 37°C in a humidified atmosphere containing 5% CO2.

In vitro imaging and localization

MCF-7 cells were seeded in 35 mm culture plates at a density of approximately 10% per plate for 24 h and then the culture medium 2 mL with 5 μM PTD replaced. After the cells were further culture for 1 h or 17 h, Lyso Traker DND 99 was added with a final concentration of 200 nM. After further culture for 1h, culture medium was discarded, the plate was washed once by PBS and fresh DMEM medium was added. Then the sample was characterized using CLSM. The wavelength of stimulating laser of PTD is 404 nm and that of Lyso Traker DND 99 is 559 nm. The false color of PTD is green and the false color of Lyso Traker DND 99 is red.

ROS measurements

The final concentration of DCFH was 40 μM. To 1.0 mL of the activated DCFH solution were added compounds. The fluorescence spectra were measured after the specimens were irradiated with white light (10 mW·cm−2). Fluorescence spectra of DCF solution was recorded in 498–700 nm emission range with the excitation wavelength of 488 nm.

Reverse transcription PCR

MCF-7 cells in one group were treated with or without PTD for 34 h under dark. In another group, after cells were treated with or without PTD for 24 h, 17-β-estrodiol was added to the culture medium and the incubation was continued for 10 h. Total cellular RNA was extracted from cells with RNAprep pure Cell/Bacteria Kit and quantified by UV absorbance spectroscopy. The reverse transcription reaction was performed using the QuantScript RT Kit in a final volume of 20 μL containing 2 μg total RNA. After incubation at 37°C for 60 minutes, the reverse transcription was terminated. The 25 μL reaction mixture contained 2 μL cDNA as template. Primers used for GAPDH amplification were: (sense, 5′-ACAGTCCATGCCATCACTGCC-3′; reverse, 5′-GCCTGCTTCACCACCTTCTTG-3′; 266 bp). Primers for pS2 amplification were: (sense, 5′-ATACCATCGACGTCCCTCCA-3′; reverse, 5′-AAGCGTGTCTGAGGTGTCCG-3′; 147 bp). Primers used for PDZK1 amplification were: (sense, 5′-GAATCCAGAGCAGTGGGAAG-3′; reverse, 5′-AGGGTGTCAAGTGGATCAG-3′; 119 bp). Primers for eEF1A1 amplification were: (sense, 5′-TCGGGCAAGTCCACCACTAC-3′; reverse, 5′-GCACAGTCAGCCTGAGATGTC-3′; 284 bp). Amplification cycles were: 95°C for 4 min, the 30 cycles at 95°C for 30 s, 60°C for 30 s, 72°C for 30 s; followed by 72°C for 10 min. Amplification products with same volume were loaded on 3.5% agarose gel stained with Goldview DNA dye. Gels were run at 100 mA for 26 min and visualized in Bio-Rad Molecular Imager ChemiDoc XRS system. DNA bands were quantified by Quantity One software.

Cell growth inhibition assay

The stock solutions of tamoxifen, 17-β-estrodiol and 10 were prepared in ethanol. The final concentration of ethanol in the culture medium is below 0.1%, which hardly induce any cytotoxicity. All the polymers were directly dissolved in sterile water. In all the experiments, MCF-7 cells were seeded in 96-well U-bottom plates at a density of 7×103 cells/well while MDA-MB-231 cells at a density of 1.5×104 cells/well. After 12 h, cells were incubated with various concentrations of compounds in fresh medium. Followed by 48 h incubation, MTT (5 mg mL−1 in water, 10 μL/well) was added to the wells by incubation at 37°C for 4 h. The supernatant was removed and 100 μL DMSO per well was added to dissolve the produced formazan. After shaking the plates for 10 min, absorbance values of the wells were recorded with a microplate reader at 570 nm.

In the experiments of PDT, PDT treatment was performed 30 hours before MTT was added by using white light source at fluence rates of 2 mW·cm−2 or 8 mW·cm−2 for 30 min. In the experiments to test the reverse ability to PDT-induced growth inhibition, 17-β-estrodiol was added into the culture medium containing PTDP 2 hours before PDT treatment, with the concentration equal to PTDP. The control group was carried out with equal water added. All the data have been corrected with the data from zero poles.