Endoplasmic Reticulum-Localized Two-Photon-Absorbing Boron

Subscriber access provided by UNIV OF DURHAM

Endoplasmic Reticulum-Localized Two-Photon-Absorbing Boron Dipyrromethenes as Advanced Photosensitizers for Photodynamic Therapy Yimin Zhou, Ying-Kit Cheung, MA Chao, Shirui Zhao, Di Gao, PuiChi Lo, Wing-Ping Fong, Kam Sing Wong, and Dennis K. P. Ng J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b01907 • Publication Date (Web): 22 Apr 2018 Downloaded from http://pubs.acs.org on April 22, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Endoplasmic

Reticulum-Localized

Two-Photon-Absorbing

Boron

Dipyrromethenes as Advanced Photosensitizers for Photodynamic Therapy

Yimin Zhou,† Ying-Kit Cheung,‡ Chao Ma,§ Shirui Zhao,† Di Gao,¶ Pui-Chi Lo,*,¶ Wing-Ping Fong,‡ Kam Sing Wong,§ and Dennis K. P. Ng*,†

†

Department of Chemistry, The Chinese University of Hong Kong, Shatin, N. T., Hong

Kong, China ‡

School of Life Sciences, The Chinese University of Hong Kong, Shatin, N. T., Hong Kong,

China §

Department of Physics, The Hong Kong University of Science and Technology, Clear

Water Bay, Kowloon, Hong Kong, China ¶

Department of Biomedical Sciences, City University of Hong Kong, Tat Chee Avenue,

Kowloon, Hong Kong, China

ABSTRACT: Two advanced boron dipyrromethene (BODIPY) based photosensitizers have been synthesized and characterized. With a glibenclamide analogous moiety, these compounds can localize in the endoplasmic reticulum (ER) of HeLa human cervical carcinoma cells and HepG2 human hepatocarcinoma cells. The BODIPY π skeleton is conjugated with two styryl or carbazolylethenyl groups, which can substantially red-shift the Q-band absorption and fluorescence emission, and impart two-photon absorption (TPA) property to the chromophores. The TPA cross section of the carbazole-containing analogue reaches a value of 453 GM at 1010 nm. These compounds also behave as singlet oxygen generators with high photostability. Upon irradiation at λ > 610 nm, these photosensitizers

1 Environment ACS Paragon Plus

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cause photocytotoxicity to these two cell lines with IC50 values down to 0.09 µM, for which the cell death is triggered mainly by ER stress. The two-photon photodynamic activity of the distyryl derivative upon excitation at λ = 800 nm has also been demonstrated.

INTRODUCTION Photodynamic therapy (PDT) has been regarded as a promising treatment option for various kinds of cancers.1 It is based on light-induced activation of photosensitizers to generate cytotoxic reactive oxygen species (ROS), leading to cell death and tumor ablation. The photosensitizers thus play a key role in determining the therapeutic outcome. In particular, their tumor selectivity and ROS generation efficiency in the tumor microenvironment are two of the most important characteristics. In addition, it has been revealed that the subcellular localization of photosensitizers has profound influence on the signalling pathways and mode of cell death initiated by PDT.2 Mitochondria have long been recognized as important subcellular targets in PDT to induce intrinsic and extrinsic cell death.3,4 By contrast, despite endoplasmic reticulum (ER) plays an important role in the assembly and transport of proteins, and the accumulation of unfolded proteins in the ER would trigger ER stress response which could be a target for the development of chemotherapeutic agents,5-7 PDT with ER-localized photosensitizers remains little studied.8-11 In fact, many stimuli can induce stress in the ER resulting in apoptosis through the unfolded protein response and Ca2+ signalling mechanism.12-14 It is believed that the ROS generated at the ER during PDT can induce ER stress, and such photosensitizers can enhance the anticancer efficacy through multiple cellkilling pathways. We report herein two novel boron dipyrromethene (BODIPY)-based photosensitizers which can localize in the ER and exhibit substantial two-photon absorption (TPA). BODIPY derivatives have been widely used for bioimaging due to their excellent spectroscopic and photophysical properties, as well as high environmental stability.15,16 By


Page 2 of 34

Page 3 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


introducing two heavy halogen atoms at the meso positions, BODIPYs can generate singlet oxygen effectively through the heavy-atom effect, and therefore function as efficient photosensitizers for PDT.17 With a view to targeting the ER of cancer cells, we have introduced a glibenclamide-related moiety to the BODIPY core to bind with the ATPsensitive K+ channels which are widely found in the ER.18 The TPA property enables the compounds to be excited at longer wavelengths, which can facilitate the light penetration into tissues.19,20 With these desirable features, these compounds could serve as multifunctional photosensitizers which are highly promising for advanced PDT.

RESULTS AND DISCUSSION The synthetic scheme used to prepare these compounds is shown in Scheme 1. Bromination of the previously described BODIPY 121 with N-bromosuccinimide (NBS) gave the dibromo BODIPY 2, which underwent Knoevenagel condensation with the aryl aldehyde 322 or 423 to afford the corresponding condensed products 5 and 6, respectively. The triethylene glycol chain was introduced to enhance the aqueous solubility, biocompatibility, and cellular uptake of the dyes, while the electron-donating carbazole group was added with a view to improving the TPA property.24 These compounds were then hydrolyzed with NaOH in a mixture of tetrahydrofuran (THF) and water to give the carboxy BODIPYs 7 and 8, and subsequently coupled with 4-(2-aminoethyl)-N-(cyclohexylcarbamoyl)benzenesulfonamide (9),25 which is a glibenclamide analogue, in the presence of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDCl) and N,N-dimethylaminopyridine (DMAP) to give the target compounds 10 and 11. Both BODIPY derivatives were highly soluble in common organic solvents and could be purified readily by column chromatography on silica gel. All the new compounds were characterized with various spectroscopic methods.



Scheme 1. Synthesis of the ER-localized BODIPYs 10 and 11.

The electronic absorption spectra of these two compounds were recorded in phosphate buffered saline (PBS) in the presence of a small amount of Tween 80 (0.3% v/v) and N,Ndimethylformamide (DMF) (1% v/v) (Figure 1a) and the data are compiled in Table 1. The Q-band of 11 was significantly red-shifted (by 39 nm) compared with that of 10, which can be attributed to the strong intramolecular charge transfer effect arising from the two electrondonating carbazole units. Compound 11 also displayed an additional band at 513 nm assignable to these moieties.24 Upon excitation at 610 nm, both compounds showed a


Page 4 of 34

Page 5 of 34

fluorescence emission (Figure 1b). Compound 11 exhibited a larger Stokes shift (844 cm-1 vs. 497 cm-1 for 10), giving the emission maximum at 753 nm, but its fluorescence quantum yield (Фf = 0.17) was significantly lower than that of 10 (Фf = 0.32). The TPA properties of these two compounds were then studied using a two-photon excited fluorescence (TPEF) method in PBS in the range of 800-1100 nm with Rhodamine B as reference. As show in Figure 2, the carbazole-conjugated BODIPY 11 exhibited a higher TPA cross section across virtually the whole region. The maximum values were calculated to be 373 GM (at 800 nm) for 10 and 453 GM (at 1010 nm) for 11, which were significantly higher than those of other BODIPY-based TPA dyes (45-300 GM).26-28 All these spectral data are also summarized in Table 1. (b)

1.0

10 11

0.8 0.6 0.4 0.2 0.0 300

400

500

600

700

800

Fluorescence intensity (a. u.)

(a)

Absorbance (a. u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.0

10 11

0.8 0.6 0.4 0.2 0.0

640

680

Wavelength (nm)

720

760

800

840

Wavelength (nm)

Figure 1. Normalized (a) electronic absorption and (b) fluorescence spectra of 10 and 11 in PBS (pH = 7.4 with 0.3% v/v Tween 80 and 1% v/v DMF).



Table 1. One-photon and two-photon absorption and fluorescence emission data for BODIPYs 10 and 11 BODIPY

λem (nm)a,b Фfa,c Ф∆d,e

λabs (nm) (log ε)a

σ2

(GM)max

[λex

(nm)]a,f

316 (4.56), 382 (4.76), 442 (4.28), 617 (4.66), 692

10

0.32 0.11

373 [800]

0.17 0.11

453 [1010]

669 (5.06) 336 (4.77), 405 (4.48), 11 a

513 (4.45), 708 (4.94)

753

In PBS with 0.3% v/v Tween 80 and 1% v/v DMF.

b

Excited at 610 nm. c Fluorescence

quantum yield (ФF) with reference to ZnPc (ФF = 0.28 in DMF). d In DMF. e Singlet oxygen quantum yield (Ф∆) with reference to ZnPc (Ф∆ = 0.56 in DMF). f GM = 10-50 cm4 s photon-1 molecule-1.

TPA cross section (GM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 34

500 400

10 11

300 200 100 0 800 850 900 950 1000 1050 1100

Wavelength (nm)

Figure 2. TPA spectra of 10 and 11 in PBS (pH = 7.4 with 0.3% v/v Tween 80 and 1% v/v DMF).

The singlet oxygen generation efficiency of these compounds was also studied in DMF and in PBS, as reflected by the rate of decay of the singlet-oxygen quencher 1,3diphenylisobenzofuran (DPBF). As shown in Figure S1 in the Supporting Information, both


Page 7 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


compounds could induce photodegradation of DPBF with a comparable efficiency in DMF, but the efficiency was significantly lower than that of the unsubstituted zinc(II) phthalocyanine (ZnPc) used as the reference. Their singlet oxygen quantum yields (Ф∆) were calculated to be 0.11 with reference to ZnPc (Ф∆ = 0.56) (Table 1). Their efficiency was also comparable in PBS, which was higher than that of methylene blue used as the reference (Figure S2 in the Supporting Information). As the singlet oxygen quantum yield of methylene blue in this solvent system has not been reported, the corresponding values for 10 and 11 were not determined. The in vitro photodynamic activity of 10 and 11 was then investigated against HeLa human cervical carcinoma cells and HepG2 human hepatocarcinoma cells. The light source consisted of a 300 W halogen lamp, a water tank for cooling, and a color glass filter cut-on at λ = 610 nm. Although the Q-band absorptions of 10 and 11 appear at different positions (Table 1), they are embedded by the spectrum of the light source (Figure S3 in the Supporting Information). Hence, both compounds should be excited by the very similar number of photons. Upon irradiation with this source (λ > 610 nm), which has a fluence rate of 40 mW cm-2, for 20 min giving a total fluence of 48 J cm-2, the effect of drug dose was determined as shown in Figure 3. The IC50 values, defined as the dye concentrations required to kill 50% of the cells, are summarized in Table 2. It can be seen that both compounds are essentially noncytotoxic in the absence of light. However, upon illumination with this light source, they become cytotoxic and 10 is much more potent than 11 by ca. 35-fold. The IC50 values of 10 are comparable with those of the tris(triethylene glycol)-substituted analogue reported by us previously,29 showing that the glibenclamide analogous moiety does not significantly affect the photocytotoxicity. As both compounds have similar singlet oxygen generation efficiency, their difference in IC50 values could be attributed to their difference in cellular uptake.



Table 2 . IC50 (µM) values of BODIPYs 10 and 11 BODIPY

For HeLa cells

For HepG2 cells

10

0.09 ± 0.01

0.16 ± 0.01

11

3.2 ± 0.3

5.2 ± 0.5

(a)

Cell viability (%)

120 100 80

HeLa in dark HepG2 in dark HeLa with light HepG2 with light

60 40 20 0 0.0

0.1

0.2

0.3

0.4

[10] (µ µM)

(b) 120

Cell viability (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 34

100 80 60 40 HeLa in dark HepG2 in dark HeLa with light HepG2 with light

20 0 0

2

4

6

8

[11] (µ µ M)

Figure 3. Cytotoxicities of (a) 10 and (b) 11 against HeLa (squares) and HepG2 (circles) cells in the absence (closed symbols) and presence (open symbols) of light (λ > 610 nm, 40 mW cm-2, 48 J cm-2). Data are expressed as mean value ± standard error of the mean value (SEM) of three independent experiments, each performed in quadruplicate.


Page 9 of 34

To provide evidence to support this hypothesis, the intracellular fluorescence intensities of these two compounds in HeLa and HepG2 cells were measured and compared using flow cytometry, which could shed light on their cellular uptake. As shown in Figure 4, the fluorescence intensity of 10 was significantly higher than that of 11 (12.8-fold for HeLa cells and 4.5-fold for HepG2 cells) despite a shorter incubation time, which indicated that 10 was taken up preferentially by these two cell lines compared with 11. It seems that the extended π skeleton constructed by the larger carbazole units of 11 hinders the uptake process.

Fluorescence intensity (a.u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


80000

10 11

60000 40000 20000 0

HeLa

HepG2

Figure 4. Comparison of the relative intracellular fluorescence intensities of 10 and 11 in HeLa and HepG2 cells as determined by flow cytometry. The incubation time was 2 h for 10 and 4 h for 11. Data are expressed as mean value ± standard deviation of three independent experiments.

To reveal if these photosensitizers were stable during the photodynamic treatment, we monitored their absorption spectra in PBS under the same conditions over 25 min. It was found that the spectra remained essentially unchanged, and the Q-band absorbance did not change with time (Figures S4 and S5 in the Supporting Information). The results clearly showed that both 10 and 11 were stable even upon illumination. It is worth noting that the absorbance at 417 nm also did not vary along with time, showing that in the study of singlet



Page 10 of 34

oxygen generation, the decrease in absorbance at that position was solely due to the decay of DPBF. Apart from the effect of drug dose, we also examined the effect of light dose on the cell viability for 10 against HepG2 cells. At a concentration of 0.19 µM, the cell viability decreased almost monotonically with the irradiation time and reached a value of ca. 50% after irradiation for 20 min (with a total fluence of 48 J cm-2) (Figure S6 in the Supporting Information). The results are in good agreement with those of the drug-dose study as described above (Figure 3a). Apparently, there was no a threshold level of light required for the photocytotoxicity. The two-photon photodynamic activity of 10 was also examined using a 800 nm laser equipped in a Zeiss laser scanning microscope. After the HeLa cells were incubated with 10 for 2 h followed by two-photon irradiation for 10 min, they were kept in a chamber with 5% CO2 at 37 °C for 3 h to allow the occurrence of cell death. After 3 h post-irradiation, a PBS solution of propidium iodide (PI) was added to distinguish the dead and viable cells. PI cannot pass through the plasma membrane of the viable cells, while it can stain the late apoptotic and necrotic cells. After the removal of PI, the cells were imaged. This method is commonly

used

to

demonstrate

the

in

vitro

potency

of

two-photon-activated

photosensitizers.30,31 It can be seen clearly in Figure 5 that the cells after two-photon PDT treatment showed a red fluorescence signal in the PI channel, indicating the occurrence of cell death. Without light irradiation, the fluorescence signal due to PI could hardly be observed. Moreover, the bright field image also showed significant cell shrinkage upon treatment with 10 and irradiation. After two-photon irradiation, the fluorescence intensity of 10 was diminished as the dye leaked out from the cells when the cells were dying. As mentioned above, this compound could generate singlet oxygen upon one-photon excitation. It is believed that this ROS was also generated under a two-photon excitation condition as the


Page 11 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


same excited states were involved, and this ROS was responsible for the observed cellular damage.

Figure 5. Fluorescence images of HeLa cells after incubation with 10 (2 µM) for 2 h in the presence or absence of two-photon irradiation (800 nm, 40 mW) for 10 min. The PI was excited at 488 nm and its fluorescence was monitored at 550-700 nm, while 10 was excited at 633 nm and its fluorescence was monitored at 640-735 nm. The red color shown in the PI channel indicates the dead cells.

The subcellular localization of these compounds was further examined with confocal microscopy. As shown in Figures 6 and 7, the merged images in panels (d) and (h) with ERTracker Green indicate selective localization of 10 and 11 at the ER in both the cell lines. The studies of subcellular localization using Mito-Tracker Green and Lyso-Tracker Green showed no obvious mitochondrial and lysosomal localization (Figures S7-S10 in the Supporting Information). BODIPY 10 showed a much higher fluorescence intensity in both the cell lines compared with 11, which could be attributed to its higher fluorescence quantum yield (Table 1) and cellular uptake (Figure 4). To reveal the role of the glibenclamide-derived moiety, the



subcellular localization of the non-glibenclamide-containing analogues 5 and 6 was also studied under the same conditions. It was found that they did not exhibit preferential localization in any of the above subcellular organelles (Figure S11-S14 in the Supporting Information). The results clearly indicated that the ER-targeting effect of 10 and 11 should be due to the glibenclamide-derived moiety. Interestingly, the fluorescence images of the cells could also be obtained under two-photon excitation (at 800 nm), though the intensity was weaker compared to the images obtained upon one-photon excitation (at 633 nm) (see Figure S15 in the Supporting Information for the images of 10 in HeLa cells).

Figure 6. Visualization of the intracellular fluorescence of 10 (2 µM) after 2 h incubation and ER-Tracker Green (0.2 µM) after 15 min incubation in HeLa and HepG2 cells: panels (a, e) show the fluorescence of 10; panels (b, f) show the fluorescence of ER-Tracker Green; panels (c, g) show the bright field images, and panels (d, h) are the corresponding superimposed images. Scare bar = 25 mm.


Page 12 of 34

Page 13 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. Visualization of the intracellular fluorescence of 11 (4 µM) after 4 h incubation and ER-Tracker Green (0.2 µM) after 15 min incubation in HeLa and HepG2 cells: panels (a, e) show the fluorescence of 11; panels (b, f) show the fluorescence of ER-Tracker Green; panels (c, g) show the bright field images, and panels (d, h) are the corresponding superimposed images. Scare bar = 25 mm.

It has been reported that subcellular localization of photosensitizers could strongly affect the cell death pathways.2 As these two BODIPYs showed high affinity to the ER, we focused on the ER stress induced by the PDT action of the more potent 10 on HepG2 cells. As ER stress is associated with the generation and accumulation of ROS, a state commonly regarded as oxidative stress,32 we measured the production of ROS inside HepG2 cells using 2’,7’dichlorodihydrofluorescein diacetate (H2DCFDA) as a probe.33 While the intracellular ROS level remained low and unchanged in the absence of light, there was a trend of increase in ROS generation with increasing the concentration of 10 (Figure 8). At a concentration of 0.8 µM, The ROS level was about 9-fold higher than that of the dark control.



1000

***

800

ROS level (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 34

600 ***

400

Light Dark

***

200 0 0.0

* ***

0.2

0.4

0.6

0.8

[10] (µ µM) Figure 8. Intracellular ROS production induced by 10. HepG2 cells were incubated with different concentrations of 10 (0.05, 0.1, 0.2, 0.4, and 0.8 µM) for 2 h. The intracellular ROS levels, as reflected by the fluorescence of H2DCFDA (10 µM), after the treatment with 10 with or without light illumination (λ > 610 nm) are shown as the means with SEM of three independent experiments. *p

Endoplasmic Reticulum-Localized Two-Photon-Absorbing Boron

Recommend Documents