Iodinated Cyanine Dyes for Fast Near-Infrared-Guided Deep Tissue

Jun 27, 2019 - (1−6) Photothermal therapy (PTT) and photodynamic therapy (PDT) are gaining ..... The Supporting Information is available free of cha...
1 downloads 0 Views 6MB Size
Research Article www.acsami.org

Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Iodinated Cyanine Dyes for Fast Near-Infrared-Guided Deep Tissue Synergistic Phototherapy Jie Cao,*,† Jinnan Chi,† Junfei Xia,‡ Yanru Zhang,§ Shangcong Han,† and Yong Sun*,† †

Department of Pharmaceutics, School of Pharmacy, and §Department of Medicinal Chemistry, School of Pharmacy, Qingdao University, Qingdao 266021, China ‡ Department of Bioengineering, Northeastern University, Boston, Massachusetts 02115, United States

Downloaded via GUILFORD COLG on July 18, 2019 at 10:18:07 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Phototheranostics, which combines deep tissue imaging and phototherapy [photodynamic therapy (PDT) and/or photothermal therapy (PTT)] via light irradiation, is a promising strategy to treat tumors. Near-infrared (NIR) cyanine dyes are researched as potential phototheranostics reagents for their excellent photophysical properties. However, the low singlet oxygen generation efficiency of cyanine dyes often leads to inadequate therapeutic efficacy for tumors. Herein, we modified an indocyanine green derivative Cy7 with heavy atom iodine to form a novel NIR dye CyI to improve the reactive oxygen species (ROS) production and heat generation while, at the same time, maintain their fluorescence characteristics for in vivo noninvasive imaging. More importantly, in vitro and in vivo therapeutic results illustrated that CyI could quickly and simultaneously generate enhanced ROS and heat to induce more cancer cell apoptosis and higher inhibition rates in deep HepG2 tumors than other noniodinated NIR dyes upon NIR irradiation. Besides, low toxicity of the resulted iodinated NIR dyes was confirmed by in vivo biodistribution and acute toxicity. Results indicate that this low toxic NIR dye could be an ideal phototheranostics agent for deep tumor treatments. Our study presents a novel approach to achieve the fastsynergistic PDT/PTT treatment in deep tissues. KEYWORDS: Iodinated cyanine dyes, photodynamic therapy, photothermal therapy, NIR imaging, enhanced ROS production



INTRODUCTION Phototheranostics, which synergistically combines deep tissue fluorescence imaging with phototherapy, has showed great potential in addressing inherent complex challenges in cancer treatment.1−6 Photothermal therapy (PTT) and photodynamic therapy (PDT) are gaining worldwide research interests for treating cancer owing to their excellent spatial specificity, noninvasiveness, and minimal side effect to normal tissues over traditional chemotherapy and radiotherapy.7−9 PTT takes advantage of photothermic agents which can convert certain wavelength of light into heat to elevate local environment temperature and subsequently kill cancerous cells.10,11 PDT is an efficient tumor therapy that combines photosensitizers (PSs) and excitation light at a specific wavelength to produce light toxic effects.12 Upon illumination, PSs could produce reactive oxygen species (ROS), especially singlet oxygen (1O2) to damage or destroy surrounding cells.13 Although phototherapy has demonstrated some potential for treating tumors, a single phototherapy modality alone has shown limited in vivo therapeutic efficacy, especially in large and deep tumors. Because ROS and heat are two main effectors of PDT and PTT, the concurrent generation of the two effectors promises to overcome these drawbacks. Meanwhile, the deep tissue imaging technique enables noninvasively monitoring biodistribution and identifying tumors.14−17 For clinical use, ideal © XXXX American Chemical Society

PDT, PTT, and imaging reagents need to display high nearinfrared (NIR) absorbance (650−900 nm), with the absorption coefficients of tissue pigments and hemoglobin being relatively low, thus allowing for deep tissue penetration and minimizing interference from autofluorescence of tissue. Recently, many NIR-absorbing materials have been synthesized as effective imaging or therapeutic agents in multimodal tumor phototheranostics, such as silver nanoparticles,18,19 oxide nanoparticles,20−22 gold nanostructures,10,23,24 metal sulfide PS,25,26 and carbon derivatives.27 However, as is well known, existing inorganic phototheranostics agents usually have some disadvantages such as nonbiodegradability and long retention time that hindered their clinical application. On the contrary to inorganic nanoparticles, NIR absorbing organic dyes, with merits of bypassing the disadvantage of metal-ion induced potential long-term toxicity and superior biocompatibility, have been become attractive clinical phototheranostics.28 Up to now, several NIR dyes have been developed for synergistic PDT and PTT, such as phthalocyanine molecules29,30 and cyanine dyes,31−34 and they have significantly Received: May 3, 2019 Accepted: June 27, 2019 Published: June 27, 2019 A

DOI: 10.1021/acsami.9b07694 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 1. (A) Synthetic scheme of CyI (reagents and conditions: (i) SOCl2/MeOH/refluxed; (ii) KI/acetone; (iii) NaNO2, SnCl2·2H2O/HCl/ H2O/−10 °C; (iv) 3-methyl-2-butanone/CH3COOH; (v) CH3CN/refluxed; (vi) LiOH/MeOH/H2O/40 °C; (vii) Glutacondianil hydrochloride/ acetic anhydride/DIEA/DCM); (B) Molecular structures of Cy7 acid and Cypate; (C) photos of CyI in DMF under ambient light and NIR light, respectively; (D) Absorption spectra of Cy7, Cypate, and CyI; (E) Normalized fluorescence spectra of Cy7, Cypate, and CyI; (F) 1O2 generation by PBS, Cypate, Cy7, and CyI under different NIR light power densities (0.3, 0.96, or 1.6 W/cm2, 1 min); (G) 1O2 generation by CyI plus NIR light (0.96 W/cm2) for different times; (H) Photothermal response of CyI aqueous solution with laser on for 300 s and laser off for 350 s. Laser wavelength: 808 nm; laser power: 0.96 W/cm2. (I) Plot of linear time against −ln θ calculated from the laser-off period of (H). (J) Temperature change curve of PBS, Cypate, Cy7, and CyI aqueous solution upon NIR light (0.96 W/cm2); (K) Temperature profiles of CyI aqueous solution upon NIR light at different power densities (0.3, 0.96, or 1.6 W/cm2).

deep tissue combinatorial phototherapy. In vitro and in vivo photo-to-photodynamic and photo-to-photothermal conversion efficiency were investigated. Furthermore, the in vivo pharmacokinetics and safety evaluation of the iodinated NIR dyes were carried out. Our study aimed at providing a promising strategy to achieve the fastsynergistic PDT/PTT treatment in deep tissues.

improved therapeutic efficacy. For example, indocyanine green (ICG), the exclusive FDA-approved NIR dye used in clinical application, has been explored as a potential phototheranostics agent.31,32 However, the low singlet oxygen generation efficiency of cyanine dyes often leads to inadequate therapy efficacy for tumors. In PDT, singlet oxygen is formed when molecular oxygen at the ground state absorbs energy from the PS at the triplet state (3PS*) excited by light.35,36 The electrons of PS will rise to a higher energy level (1PS*) upon light irradiation. The photoexcited singlet state of PS (1PS*) may form excited triplet states of PS (3PS*) through intersystem crossing (ISC); therefore, the efficiency of ISC will directly influence 1O2 generation. As far as we are concerned, the presence of heavy atoms could promote the rate of ISC between singlet and triplet states by spin−orbit coupling.36 Therefore, introduction of heavy metal ions to PSs could improve singlet oxygen generation.37−41 Consequently, heavy atom-modified cyanine dyes are considered as potential enhanced PDT PSs. Atchison et al.42 have constructed iodinated IR783 for efficient NIR PDT, but there are still several issues that remains to be considered before their future clinical application, such as (i) long-term biotoxicity, (ii) biodegradability and body clearance, (iii) photo-to-photothermal conversion efficiency, and (iv) PTT/PDT synergistic efficiencies for deep tissue cancer treatment. Hence, in this study, we prepared iodinated derivatives of Cy7, which we named them CyI, and for the first time, developed them for NIR-guided



RESULTS AND DISCUSSION Synthesis of CyI. Cy7 is a derivative of ICG, with λmax at 749 nm and Em at 776 nm. Its molecular structure is shown in Figure 1B. It is well known that the quantum yields of singlet oxygen and fluorescence intensity are interdependent, as evidenced by increased production of ROS usually leads to decreased fluorescence intensity.42,43 Hence, their photophysical characteristics are dependent on the position and the number of iodine atoms. Herein, we prepared monoiodinated derivative of Cy7 (Figure 1A) and the detailed synthetic protocol was in the Supporting Information. The retention time of the desired product was determined to be 15.288 min according to an analytical high-performance liquid chromatography and the molecular weight to be 776.69 (MW: 776.5) according to the MALDI-TOF MS profile (Figure S1). The 1H NMR spectra of intermediate products (c, f, g, h) and CyI is shown in Figure S2. The absorption and fluorescence spectra of CyI were obtained (Figure 1D,E). For comparison, we also recorded B

DOI: 10.1021/acsami.9b07694 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 2. Cell viability of Cypate, Cy7, and CyI in HepG2 cells (A) and L-02 cells (B) with different concentrations (12.5, 25, 50, 100, and 200 μM); (C) Mean plasma concentration profile after iv administration of 3 mg/kg CyI (n = 4); (D) Tissue distribution of CyI in vivo (n = 4); (E) In vivo dynamics of CyI in normal mice; (F) NIR images of main organs excised from mice 2 h (F1) and 12 h (F2) post intravenous injection of CyI; (G) H&E stained main organs excised from mice 1 week post-injection of CyI at different dosages (image magnification is 200×).

the absorption and fluorescence spectra of Cy7 and a hydrophobic NIR dye Cypate (Abs: 780 nm, Em: 804 nm), as well as an ICG derivative, which we previously used for NIR imaging. The corresponding molecular structure is shown in Figure 1B. As shown in Figure 1D,E, both absorption and fluorescence spectra of CyI were located within the NIR spectral region. Introduction of iodine to the indole rings caused minor changes in the shape of the absorption bands, which demonstrated that introduction of an electron-withdrawing group (I) would not affect the advantageous absorption characteristics of Cy7. The fluorescence property of CyI was examined after excitation at 750 nm in dimethylformamide (DMF) and Cy7 in aqueous formulated solutions. Both CyI and Cy7 exhibited fluorescence spectra with Stoke shifts in the range of 27−62 nm (Figure 1E). The fluorescence quantum yield of the monoiodinated dye CyI was calculated to be 48% according to eq S1 (Supporting Information). Although it has a significant decrease when compared to Cy7 (QY: 60%), the fluorescence intensity was much higher than Cypate (QY: 13%).44 Decreased fluorescence intensity would indicate that when an iodine atom is substituted on the indole ring of indocyanine dyes, it triggers a large heavy atom effect which may result in an increased level of singlet oxygen, depending on other possible competing photophysical pathways. To determine the potential of CyI for both photo-tophotothermal and photo-to-photodynamic ability, ROS and heat generation upon NIR laser irradiation were assessed. Also, we compared PDT and PTT efficiencies of CyI with both hydrophilic dye Cy7 and hydrophobic dye Cypate. Generation of singlet oxygen was indicated by change of fluorescence intensity of SOSG, which is a highly selective indicator of 1 O2.45 Phosphate buffered saline (PBS) and water solution of

NIR dyes (Cypate, Cy7 and CyI) were treated with a laser diode at 808 nm and set at different power densities for 1 min. Figure 1F shows fluorescence intensity of each solution at different laser power density. It should be mentioned that we chose the 808 nm laser diode rather than other wavelengths, where the dyes have better absorption for the following reason: we have compared the fluorescence, photo-to-photodynamic, and photo-to-photothermal conversion efficiency of CyI under 765 and 808 nm at the same power density, alternately (Figure S3). The results showed although the fluorescence intensity and 1O2 generation of CyI upon 765 nm irradiation are stronger than that of 808 nm irradiation, the heat generation of CyI under 765 nm was relatively slower and weaker than that of 808 nm. The temperature profile of CyI could reach at 46.2 °C within 1 min upon 808 nm irradiation (Figure S3C). However, under 765 nm laser, CyI only reached at 36.5 °C after 1 min irradiation, which could not be used for PTT. This phenomenon was consistent with the results reported previously46 and could be explained that when a molecule absorbs light, a majority of the energy converts to photons in the form of fluorescence or heat. Because of conservation of energy, higher fluorescence quantum yield corresponds to lower photothermal conversion efficiency and vice versa. As shown in Figure 1F, all NIR dyes generated 1O2 upon 808 nm NIR irradiation, and remarkably, CyI generated much stronger (4.5−7.5-fold) SOSG fluorescence intensity compared with Cypate and Cy7. According to eq S2 (Supporting Information), 1O2 quantum yield (ΦΔ) of CyI was calculated to be 0.75. Figure 1G showed that 1O2 was rapidly produced in the first 1 min and maintained steady state within 6 min. Therefore, enough singlet oxygen could be generated upon 1 min irradiation. It should be mentioned that hydrophobic NIR dyes (Cypate and CyI) in aqueous solution can easily form the C

DOI: 10.1021/acsami.9b07694 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 3. (A) Confocal images of ROS generation in HepG2 cells after various treatments: untreated cells as control; cells treated with Cypate, Cy7 or CyI at different laser densities for 1 min, and CyI-treated cells cover with 1 cm pork tissue plus 1 min of laser irradiation (0.96 W/cm2), scale bar is 30 μm; (B) Mean fluorescence of ROS production in HepG2 cells after various treatments; (C) Temperature changes of HepG2 cells treated with different NIR dyes upon NIR irradiation (0.3 or 0.96 W/cm2). Untreated cells and cells only treated with NIR light were as control. The data were represented as mean ± SD, where * denotes P < 0.05, and ** denotes P < 0.01.

H-aggregate, thus resulting in reduced PDT efficiency. The improvement of hydrophilicity of CyI still needs further investigation. Raising and maintaining an elevated temperature (≥40−43 °C) localized in the tumor play a vital role for efficient PTT.29,47 First, we evaluated the photothermal conversion capability of CyI according to eq S3. Only hA is unknown for the calculation. The graph of ΔT versus time and time versus −ln θ is shown in Figure 1H,J, and the obtained slope was hA. Replacing the hA value into eq S3, the photothermal conversion efficiency of CyI was calculated to be 49.15%, much higher than commonly used PTT agent Au nanorods (22%). Then, to explore photothermal properties of CyI upon activation with NIR light, we evaluated the temperature changes during 8 min laser exposure (0.96 W/cm2), and compared the results with that of Cy7 and Cypate. As shown in Figure 1J, the temperature profile of CyI had a quick rising phase and peaked at 58 °C within 2.5 min of irradiation. Under the same experimental conditions, Cy7 and Cypate exhibited similar temperature rising profiles but with relatively longer time to reach the peak (∼4.5 min) and lower peak temperatures at 44.2 and 49.3 °C, respectively. In contrast, PBS control peaked at 27 °C and showed no such photothermal effect, which is a clear evidence that CyI possess sufficient photo-to-photothermal efficiency to damage cancer cells. CyI exhibiting higher peak temperature over Cypate and Cy7 may be attributed to the introduction of the iodine atom, which had positive influence on the photothermal effect. We hypothesized that similar to ICG, CyI could be in part converted to the generation of 1O2 while simultaneously transited into heat exposed to NIR light, resulting in dual PDT/PTT behavior. However, it should be mentioned that, such as Cypate and Cy7, photodecomposition of CyI took place through the self-sensitized photo oxidation process and caused temperature decrease after reaching a peak. Further experiments were carried out at different laser densities (0.3−1.6 W/cm2), and the temperature profiles of

CyI in Figure 1K indicated when the laser density was below 0.3 W/cm2, no photothermal effect was detected. Consequently, we can envision rapid transition from single PDT to synergistic PDT/PTT by simply changing the power density of laser. In summary, this technology platform sheds light upon future clinical application of CyI in combinatorial PDT and PTT in cancer therapy. Safety Evaluation of CyI. In Vitro Toxicity of CyI. The safety profile of NIR dyes after introduction of the iodine atom needs to be investigated before their clinical use. Herein, the cytotoxicity of CyI in both normal cells and cancer cells was first investigated using MTT assay. The results shown in Figures 2A,B and S4 indicated that similar as Cypate and Cy7, CyI exhibited no apparent cytotoxicity on either cancer cells (HepG2) or normal cells (L-02, MC3T3) after 48 h incubation. The proof of little or noncytotoxicity of CyI demonstrated that the introduction of the iodine atom to NIR dyes did not render detectable toxicity. In Vivo Toxicity of CyI. Before investigating the toxicity of CyI in vivo, the pharmacokinetic and biodistribution of CyI in normal mice was first studied. The mean plasma concentration profile in Figure 2C showed that the CyI concentration was basically not detectable in plasma after 4 h, indicating CyI could be fast cleared from plasma. Quantitative analysis of tissue distribution results (Figure 2D) showed that CyI first gathered in liver and peaked at 2 h. After 4 h, CyI mainly accumulated in intestines, attaining a maximum at about 8 h and gradually decreasing after 12 h. After 24 h, the dyes were almost cleared from the mice. The results demonstrated that CyI was mostly transferred from liver to gastrointestinal via hepato−enteric circulation and cleared by intestinal and partially by the renal pathway. To further investigate the metabolic process and in vivo imaging ability of CyI, the dynamic and tissue distribution of CyI was monitored by the NIR imaging system (Figure 2E). As shown, CyI dyes spread all over the mice, and majority of them enriched at the liver and kidney 0.5 h after the injection. D

DOI: 10.1021/acsami.9b07694 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

Figure 4. (A) Cell viability under various CyI concentration plus NIR irradiation (0.3 or 0.96 W/cm2, 1 min); (B) Cell viability under different treatments: control (no dye or irradiation), NIR irradiation only (808 nm, 0.96 W/cm2), Cypate or Cy7 plus NIR irradiation (808 nm, 0.96 W/ cm2), CyI plus NIR irradiation (808 nm, 0.3 W/cm2, PDT only), CyI plus NIR irradiation (808 nm, 0.96 W/cm2, PDT/PTT), CyI plus NIR irradiation (808 nm, 0.96 W/cm2, PDT/PTT) plus pork tissue; (C) Flow cytometry results of cell apoptosis under different CyI concentrations upon NIR irradiation with or without pork tissue; (D) Corresponding column diagram of the flow cytometry results; (E) Fluorescence images of HepG2 cells stained with calcein AM and PI under different treatments (scale bar, 30 μm).

intracellular ROS and temperature were measured. The ROS level generated in CyI-treated cancer cells was assessed using DCFH-DA, a ROS-detecting fluorescent probe. As shown in Figure 3A, Cypate- or Cy7-treated cancer cells under 0.96 W/ cm2 laser irradiation showed weak green fluorescence, while cells treated with CyI showed stronger fluorescence due to more activated H2DCFDA, indicating a high level of intracellular ROS generation. With higher laser power, CyItreated cells showed much brighter fluorescence. The deep tissue mimic group under 0.96 W/cm2 laser irradiation also showed strong fluorescence. No or little ROS was generated in the cells without NIR dyes, regardless whether NIR irradiation is performed. In addition, the average fluorescence intensity of NIR dyes-treated cells, as shown in Figure 3B, is consistent with solution studies (Figure 1F), which demonstrated the obvious photodynamic effect of CyI in HepG2 cells. The temperature profiles of the cells pretreated with the abovementioned NIR dyes upon NIR irradiation were represented in Figure 3C. As shown, with a starting temperature of 26 °C, the highest temperature of CyI-treated cell pellets (both non-deep tissue mimic and deep tissue mimic group, laser density: 0.96 W/cm2) was 48.2 °C (dark yellow curve) and 46.9 °C (navy curve), much higher that of Cypateand Cy7-treated cell pellets (35.7 °C, (magenta curve), 34.8 °C, (dark cyan curve), respectively. The results further indicated that iodinated NIR dyes have high photon-to-heat conversion efficiency for an efficient PTT. In contrast, the

Then, the bright signals in the liver and gastrointestinal system inferred that the dye was predominantly cleared through the liver−gastrointestinal pathway. The ex vivo images of main organs in Figure 2F showed that CyI mainly accumulated in liver and kidney after 2 h (Figure 2F1), and then transferred to the intestines after 12 h (Figure 2F2). The results were consistent with pharmacokinetic and in vivo dynamic results, further indicating CyI was primarily excreted by the liver− gastrointestinal system and secondarily by the renal pathway. In vivo toxicity and acute toxicity of CyI were explored by conventional procedures, as described in Safety Evaluation of CyI section in Supporting Information. Results present that the mice administered with CyI did not show appreciable abnormal response behavior with 1-month post-injection compared to controls and had zero death rate even at the group that received the highest dosage (10 mg/kg). Figure 2G shows H&E stained main organ sections from mice 1 week after CyI treatment. As shown, compared with control groups injected with PBS, no obvious histological difference was detected in the CyI group at normal dosage (