- Email: [email protected]
Synthesis of upconversion nanoparticles conjugated with graphene oxide quantum dots and their use against cancer cell imaging and photodynamic therapy
G Model BIOS 9091 1–7
BIOS : 9091
Model5G
pp:17ðcol:fig: : NILÞ Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Synthesis of upconversion nanoparticles conjugated with graphene oxide quantum dots and their use against cancer cell imaging and photodynamic therapy Seung Yoo Choi a, Seung Hoon Baek a, Sung-Jin Chang a, Yohan Song b, Rafia Rafique a, Kang Taek Lee b, Tae Jung Park a,n a b
Department of Chemistry, Chung-Ang University, 84 Heukseok-ro, Dongjak-gu, Seoul 06974, Republic of Korea Department of Chemistry, Gwangju Institute of Science and Technology, 123 Cheomdan-gwagi-ro, Buk-gu, Gwangju 61005, Republic of Korea
art ic l e i nf o
a b s t r a c t
Article history: Received 3 June 2016 Received in revised form 26 August 2016 Accepted 27 August 2016
Multifunctional nanocomposite has a huge potential for cell imaging, drug delivery, and improving therapeutic effect with less side effects. To date, diverse approaches have been demonstrated to endow a single nanostructure with multifunctionality. Herein, we report the synthesis and application of coreshell nanoparticles composed with upconversion nanoparticle (UCNP) as a core and a graphene oxide quantum dot (GOQD) as a shell. The UCNP was prepared and applied for imaging-guided analyses of upconversion luminescence. GOQD was prepared and employed as promising drug delivery vehicles to improve anti-tumor therapy effect in this study. Unique properties of UCNPs and GOQDs were incorporated into a single nanostructure to provide desirable functions for cell imaging and drug delivery. In addition, hypocrellin A (HA) was loaded on GOQDs for photo-dynamic therapy (PDT). HA, a commonly used chemotherapy drug and a photo-sensitizer, was conjugated with GOQD by π-π interaction and loaded on PEGylated UCNP without complicated synthetic process, which can break structure of HA. Applying these core-shell nanoparticles to MTT assay, we demonstrated that the UCNPs with GOQD shell loaded with HA could be excellent candidates as multifunctional agents for cell imaging, drug delivery and cell therapy. & 2016 Elsevier B.V. All rights reserved.
Keywords: Upconversion nanoparticle Graphene oxide quantum dot Hypocrellin A Photodynamic therapy Photoluminescence
1. Introduction Lanthanide-doped upconversion nanoparticles (UCNPs) have been attracted a great interest in various fields because of their unique optical properties reported previously (Chen et al., 2014). Thus, these nanostructures have a specific value for biological applications in various therapeutic fields such as in vivo imaging and photodynamic cancer therapies (Shen et al., 2013). In other words, lanthanide-doped UCNPs with ladder-like energy level structures have an ability to convert low energy excitation to highenergy emission via two or more photon absorption or energy transfer. Upconversion refers to sequential absorption of multiphotons leading to the emission shorter than the excited wavelength, which is anti-stokes type emission (Junwei et al., 2008). UCNPs have been differentiated from others in terms of very little auto-fluorescence and scattering on multiphoton absorption. Consequentially, they have a low background signal and high n
Corresponding author. E-mail address: [email protected] (T.J. Park).
signal-to-noise ratio offering outstanding advantages in application of therapeutic fields. Therefore, UCNPs have become efficient candidates for the analysis of biological and environmental samples due to their bio-compatibility, especially for in vivo and in vitro fluorescence imaging (Haase and Schafer, 2011; Park et al., 2015; Seo et al., 2015). The similar atomic size and chemical properties of lanthanide elements make them more biocompatible and stable among other elements. Photodynamic therapy (PDT) is a clinical tumor treatment using nontoxic light-sensitive agents that are toxic to target malignant and other diseased cells (Idris et al., 2012). When photosensitizing agents are exposed to a particular type of light, they produce cytotoxic reactive oxygen species (ROS). ROS leads nearby cells to death with minimal invasiveness, toxicity and side effect (Sharman et al., 2000; Martin and Barrett, 2002; Chatterjee et al., 2008). PDT requires photosensitizer (PS), light source, and oxygen. PS helps production of singlet oxygen by absorbing ultraviolet or visible region and transferring it to adjacent molecules (Konan et al., 2002). The PS transforms tumor cells with excited singlet state by intersystem crossing as a result of an irradiation at a
http://dx.doi.org/10.1016/j.bios.2016.08.094 0956-5663/& 2016 Elsevier B.V. All rights reserved.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
2
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
specific wavelength of light. The excited triplet state can transfer hydrogen atom or electron to form radicals interacting with oxygen. These radicals produce oxygenated products like hydrogen peroxide, super oxygen, and hydroxyl ion or can transfer from oxygen to singlet oxygen, which is highly reactive. As a result, the ROS leads tumor cell to death. Hypocrellin A (HA) is one of PSs isolated from natural fungus, Hypocrella bambusae Sacc., which was discovered in China (Zhang et al., 1998). HA has low toxicity in an inactivated form (Li et al., 2015) and can be metabolized in vivo. Thus, HA can be used in some capillary disease treatment due to higher PDT effect under presence of singlet oxygen than many widely used PSs (Diwu et al., 1990; Shan et al., 2013). Graphene oxide quantum dot (GOQD) is a class of carbon-based dots. The GOQD has received tremendous attention, due to its outstanding advantages such as low toxicity, low cost, excellent biocompatibility, resistance to photobleaching and ease of fabrication (Dong et al., 2012, 2013; Jiang et al., 2015; Lu et al., 2015; Zhang et al., 2016). Both GOQD and HA can easily form the HA/ GOQD complexes through π-π stacking due to same planar sp2 structures (Thomas et al., 2010; Jang et al., 2015) without complicated synthetic process which can break the structure of PS (Guo et al., 2013). Other materials such as organic dyes and semiconductor quantum dots (QDs) have been used as biological luminescent labels, but they have certain limitations. For example, organic dyes exhibit rapid photobleaching, and QDs are less chemically stable, inherently toxic and show fluorescence intermittence. Hence, NIRto-visible UCNP seems to be a promising alternative fluorescent material for bio-detection based on their advantages such as high chemical stability, low toxicity, economic instrument cost, and tunable optical properties. In this study, we report ytterbium and erbium ions-doped sodium yttrium fluoride (NaYF4:Yb3 þ , Er3 þ ), the most efficient NIR-to-visible UCNP in solid-state materials doped with rare-earth ions. NaYF4:Yb3 þ , Er3 þ UCNP has been synthesized and analyzed their unique advantages. We designed multifunctional nanostructure composed of UCNPs, GOQDs and HA for cell imaging drug delivery and PDT of cancer cell (Scheme 1). First, ytterbium and erbium ions-doped sodium yttrium fluoride (NaYF4:Yb3 þ , Er3 þ ) nanoparticles were synthesized as UCNP. Polyethylene glycol (PEG) was coated on the surface of the UCNPs additionally for bio-compatibility. The PEGylation was performed to make UCNP more effective in disease
monitoring and drug delivery system. Next, we attached GOQDs to the PEG-coated UCNPs for drug delivery carrier. Finally, multifunctional material, GOQD-attached UCNP were synthesized with a great potential for both upconversion luminescence (UCL) imaging and PDT in cancer diagnostics. Furthermore, HA as a PS was loaded on the surface of the GOQD-attached UCNP (Oleinick et al., 2002; Zhou et al., 2015). This drug-delivery system was proposed to minimize side effect, achieve a delivery system for PDT and help in vitro imaging (Zheng et al., 2013). These nanocomposites were designed due to their intrinsic properties for drug-delivery from GOQD, therapy from HA and imaging capability from UCL.
2. Experimental 2.1. Materials Yttrium nitrate hexahydrate, ytterbium chloride hexahydrate, erbium nitrate pentahydrate, sodium fluoride, and nitric acid were purchased from Sigma-Aldrich (St Louis, MO, USA). Sodium citrate was obtained from APS Biotech (Seoul, Korea). Ethyl alcohol was acquired from Emsure (Billerca, MA, USA). Cetyltrimethylammonium bromide (CTAB) was purchased from Daejung (Seoul, Korea). All solutions were prepared with deionized (DI) water (Direct-Qs Water Purification System, Millipore, Billerica, MA, USA). Graphite nanoparticle (GNP) (93%, 3–4 nm) was obtained from SkySpring Nanomaterials (Houston, TX, USA). Sulfuric acid (95.0%) purchased from Junsei (Tokyo, Japan) and Nitric acid (60.0%) obtained from Samchun (Seoul, Korea). HA ( 477%) was purchased from abcam (Cambridge, UK). 2.2. Instruments and characterizations The UCL spectra under 980 nm continuous wave laser excitation were analyzed using Shamrock Spectrograph (Shamrock 303i, Andor Technology, Belfast, Ireland). The absorbance was analyzed by UV/Visible/NIR spectrophotometer (V-670, Jasco, Tokyo, Japan). Their fluorescence properties were also analyzed using the Multimode microplate reader (BioTek Synergy H1, Winooski, VT, USA), absorbance on methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay. The particle size was determined by transmission electron microscopy (TEM, Tecnai G2 F30 S-TWIN, FEI, Hillsboro, OR, USA) and particle size analyzer (Otsuka ELSZ-1000, Osaka,
Scheme 1. The action procedure of HA/GOQD/UCNP nanoparticle.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Japan). Additionally, the crystal structure of UCNP was analyzed using X-ray powder diffractometer (XRD, Bruker-AXS New D8Advance, Bruker, Berlin, Germany) with Cu-radiation from Cu-Kα X-ray source (λ ¼1.54056 Å) ranging from 5°≦2θ≦70°. PEGylation of UCNP was checked by thermal gravimetric analysis (TGA, N-1000, Scinco, Seoul, Korea). For HA/GOQD/UCNP imaging in the HeLa cells, inverted microscope (IX73, Olympus, Tokyo, Japan) was composed of a 980nm NIR diode laser (P161-600-980A, EM4), an electron multiplying charge-coupled device (EMCCD) camera (iXON3, Andor Technology) and a live-cell incubation chamber (TC-L-10, Live Cell Instrument, Seoul, Korea) on the stage. The cell samples on the cover glass bottom dish were mounted on the microscope stage in the incubation chamber which provided optimal conditions (37 °C and 5% CO2) for live-cell imaging. Typical power density of illumination was 21.75 mW/cm2. 2.3. Synthesis of therapeutic probe HA/GOQD/UCNP 2.3.1. Synthesis and surface modification NaYF4:Yb3 þ ,Er3 þ UCNP and PEGylation (The first cycle of UCNP synthesis, “One-cycle”) NaYF4:Yb3 þ , Er3 þ UCNP was synthesized according to the previously reported procedure (Sun et al., 2007; Chen et al., 2013). 3.6 mL of 0.2 M Y(NO3)3, 3 mL of 0.1 M YbCl3, 0.3 mL of 0.1 M Er (NO3)3, and 6 mL of 0.1 M citrate sodium were vigorously mixed at room temperature. Then, 6.3 mL DI water, 45 mL ethanol, and 0.3 g CTAB was subsequently added to the solution. Next, 18 mL of 1.0 M NaF was added dropwise and the solution was incubated with vigorous stirring for 2 h at room temperature. Then, 3 mL of HNO3 was mixed in the solution. Mixture solution was autoclaved in the teflon liner-packed autoclave vessel sized with 23 mL, which was heated to 180 °C for 4 h. Reaction solution was cooled in the air condition and then the synthesized cubic phase of UCNP was prepared (Fig. 1A). The UCNPs were cooled down to room temperature and then separated by centrifugation at 3,500 rpm for 10 min. (The second cycle of UCNP synthesis, “two-step”). As-prepared cubic phases of UCNPs were dispersed in DI water as seed materials to synthesize cubic-hexagonal phases of UCNPs. The UCNPs were dried off completely at 60 °C in dry air oven. The prepared UCNPs were finalized with overall reaction one more time from a previous synthetic procedure. As a result, the cubichexagonal phase of UCNP was developed. UCNP was further proceeded for surface modification and
3
functionalization with amine groups (Zhang at al., 2007). The mixture solution of diethylene glycol (DEG) and poly(allylamine) (PAAm) were prepared in the neck flask, and heated under the N2 gas environment while keeping the temperature at 110 °C with reflux. Then, UCNP was mixed in 10 mL of DI water and injected rapidly into the heated mixture. Reaction was proceeded at 240 °C for another half an hour. When reaction was finished, the reactant was cooled to room temperature and a solution of 0.1 M of hydrochloric acid was injected. The final product was separated by centrifugation and washed three times with DI water. Amine-functionalized UCNP (20 mg) was dissolved in 20 mL 2(N-morpholino)ethanesulfonic acid (MES) buffer (0.1 M, pH 6.0), then sonicated for 30 min (100 W). 20 mg of amine-PEG (polyethylene glycol 2-aminoethyl ether acetic acid) was added into asprepared UCNP solution and sonicated for 10 min (100 W). Subsequently, 6 mL of aqueous EDC (4 mg/mL) was added in the solution, mixed and stirred for 24 h at room temperature. The resulting PEGylated UCNP was centrifuged at 3,500 rpm for 10 min and washed with DI water for several times. 2.3.2. Preparation of the GOQD and binding of GOQD on UCNPs using EDC coupling 1 g of graphite nanoparticles (GNPs) was pre-oxidized in c-H2SO4 (60 mL) and HNO3 (20 mL) for 3 h in mild sonication (100 W) (Thomas et al., 2015; Kuo et al., 2016). Next, the mixture was stirred at 100 °C under a refluxing condition for 14 h. After that, the mixture was cooled down to 25 °C and diluted with 150 mL of DI water in the ice bath. The suspension was centrifuged to remove the aggregates of GOQD at 14,000 rpm for 30 min and the supernatant was separated. Then, supernatant solution was dialyzed for 24 h to remove the acidic ions, and the pH of the aqueous solution was adjusted to 7.0 with 1.0 M NaOH. The final solution was dialyzed to remove the salt for 7 days at room temperature (Dong et al., 2012). PEGylated UCNP (10 mg) was dissolved in 10 mL MES buffer (0.1 M, pH 6.0) and then sonicated for 30 min. GOQD (OD220 ¼ 1.0, 10 mL) was added into that solution and stirred for 10 min. Subsequently, 4 mL aqueous solution of EDC (4 mg/mL) was added dropwise into mixture solution and stirred for 24 h at 25 °C. The resulting GOQD/UCNP was centrifuged at 3,500 rpm for 10 min and washed with DI water. 2.3.3. HA loading through π-π stacking and releasing Lastly, the conjugation of GOQD/UCNPs with HA was performed by dispersing in N,N-dimethyl-formamide (DMF) as HA dissolves
Fig. 1. (A) X-ray powder diffraction (XRD) pattern with the lower side about cubic phase (PDF no. 01-077-2042) and the upper side of cubic-hexagonal (PDF no. 00-0270689) phase of NaYF4:Yb3 þ ,Er3 þ nanoparticles. The well-defined peaks indicated the high crystallinity of the synthesized material. (B) Photoluminescence spectroscopy represented energy transfer and multiphoton relaxation in cubic and cubic-hexagonal phase of NaYF4:Yb3 þ ,Er3 þ nanoparticles. Er3 þ emitted photon at 550 and 660 nm under excitation of 980 nm (300 mW).
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
in polar aprotic solvents. In detail, 100 mL of 15 mM HA in DMF was dissolved in 10 mL DI water. Then, 8 mg GOQD/UCNP was dispersed in 10 mL ultra-pure water using sonication (100 W) for 10 min. These aqueous suspensions were mixed well with slow stirring at room temperature for 24 h to form HA/GOQD/UCNP complex. 2.4. Cell culture and cellular uptake The HeLa cells were seeded in cell culture plate at the density of 1.5 105 cells/mL in DMEM. Shortly thereafter, the cells were cultivated with 10% Fetal bovine serum (FBS) and 1% antibiotic antimycotic (AAM) (5.1% CO2, 37 °C) for 24 h. HA/GOQD/UCNP was treated in medium (each concentration calculated by HA) on the plate. After incubation for 20 min, cells were washed with 1X phosphate-buffered saline (PBS, pH 7.4) for four times and monitored using microscopy through bright filter and 980-nm filter. The cell viability was checked through MTT assay (Hu et al., 2012). HeLa cells were seeded in 96-well plate at the density of 8 103 cells/well. The cells were divided into four groups, and each group was treated with HA/GOQD/UCNP for 20 min and then washed with PBS for four times. After washing, following conditions were applied to each group to determine cell viability; (1) they were placed in a dark incubator for 12 h, (2) irradiated by 460-nm LED for 20 min and placed in a dark incubator for 12 h. 2.5. UCL imaging in vitro 1 105 cells/mL of HeLa cells were seeded on the cover glass bottom dish (SPL, Korea), and cultured at 37 °C in a humidified 5% CO2 incubator for 24 h. Then the medium was changed with 1 mL of HA/GOQD/UCNP containing medium (5 μg/mL). After additional incubation for 30 min, the cell samples were washed and used for live-cell imaging. The image of HA/GOQD/UCNP-treated HeLa cell was recorded by excitation at 980 nm with CW laser diode.
3. Results and discussion 3.1. Preparation and characterization of UCNP The crystal structure of NaYF4:Yb3 þ , Er3 þ UCNP was examined by XRD. Fig. 1A shows XRD spectra of UCNPs synthesized after one cycle process (Left and middle panel, (1)) and two-step process (Left and top panel, (2)). The XRD peaks of cubic and hexagonal phases were assigned with PDF no. 01-077-2042 (Fig. 1A, Left and bottom panel, blue lines) and PDF no. 00-027-0689 (Fig. 1A, Left and bottom panel, red lines), respectively. Both UCNP samples after one-cycle and two-step synthesis dominantly exhibit cubic phase. However, XRD peaks of hexagonal phase are discernible in XRD spectrum of the UCNPs after two-step synthesis, suggesting that some of the cubic phase UCNPs has been transformed into hexagonal phase during the second synthetic process. It is worth noting that phase transition from cubic to hexagonal structure cannot be achieved by one-cycle synthesis but only through the two-step synthesis (Fig. S1). The UCL of one-cycle and two-step synthesis dominantly exhibits cubic phase. Fig. 1B shows UCL spectra of two types of UCNP samples (1 mg/mL in DI water), one produced by one-cycle synthesis (black trace) and the other produced by two-step synthesis (red trace). Cubic-hexagonal phase of UCNPs exhibit about 10 times higher intensity of UCL than cubic phase of UCNPs. This result confirms that a part of cubic phase UCNPs has changed into hexagonal phase UCNPs (Park et al., 2015). Therefore, it can be suggested that our UCNP samples after two-step synthetic process are composed of two types of UCNPs, one is cubic phase, and the
other is hexagonal phase. Moreover, UCL characteristics of the UCNP samples after two-step synthesis are dominated by those of the hexagonal phase UCNPs. Fig. 1B depicts that Yb3 þ ion absorbs 980 nm (300 mW, 1 mg/mL in DI water) excited photon and then transfers its energy next to the Er3 þ ion which acts as an activator. Upconversion fluorescence spectra of NaYF4 nanoparticles appears green and red emissions of 550-nm and 660-nm regions, respectively (Haase and Schafer, 2011; Park et al., 2015). The synthesized cubic-hexagonal phases of UCNP have higher emission intensity as compared to cubic phase of UCNP. To verify the crystallinity and the presence of the GOQD and UCNP, NaYF4:Yb3 þ , Er3 þ UCNP, and PEGylated UCNP were characterized with a transmission electron microscope (TEM) as shown in Fig. 2 and S1. The results show that the particle size of GOQD was 5 nm (Fig. S2A). Fig. S2B shows the diffraction pattern image (inset) and the size of cubic-hexagonal phase of UCNPs which was around 105.057 16.79 nm. TEM imaging was also used to compare the UCNP size with phase transition and PEGylation of UCNPs (Fig. 2C and Fig. S2C). The average diameter of cubic phase UCNPs and PEGylated UCNPs were 77.02 7 15.02 nm and 112.26 720.64 nm, respectively (Fig. 2D). Moreover, PEGylation of UCNPs was also confirmed by using TGA (Fig. 2E). The red line curve indicated that PEGylated UCNPs loss their weight with increasing temperature and 1.0 wt% weight loss was observed below 200 °C. The similar trend was observed for amine-functionalized UCNPs (black line) and no weight loss occurred below 200 °C. The results were analyzed below 200 °C because PEG has a boiling point below this range. 3.2. Binding of GOQD and HA Fig. S3A shows that HA and GOQD did not have photoluminescence at 980 nm. The concentration of GOQD on individual UCNPs needed to be optimized as GOQDs-attached UCNP had low UCL efficiency. From several values of OD220, the optimum values for UCL of GOQD-attached PEG-coated UCNP were determined to be 0.1, 0.3, 0.5, 0.8 and 1.0 (Fig. S3B). Fig. S3B was shown OD220 value 1.0 of GOQD have a minor influence on UCL. The GOQD also had a wide HA binding sight on the surface of GOQD/UCNP. Therefore, we prepared the complex of GOQD-attached UCNP with OD220 value 1.0 of GOQD based on core-shell nanoparticles. HA bound with GOQDs on the surface of PEG-coated UCNPs via π-π interaction to examine PDT activity. The fourier transform infrared (FTIR) spectroscopy was used to characterize the chemical bonds and compositions of as-prepared samples. The change of the oxygen functional groups from GOQD can be reflected in FTIR spectroscopy, as shown in Fig. S4. In the FTIR spectra of PEGylated UCNP and GOQD/UCNP composite, a broad absorption band appears in the range of 3000–3700 cm 1 due to the stretching vibration of -OH band and with primary and secondary N–H groups that also absorb in this range at 3300–3500 cm 1. For a neat liquid, the N–H peak is typically weaker than O–H. The carboxyl group -OH stretch appears as a very broad band in the region 3300–2500 cm 1 while the N–H stretches of primary amines are two bands from 3300 to 3500 and 3250–3330 cm 1. Compared with PEGylated UCNP (black trace), the intensity of all absorption peaks related to oxygen functional groups about epoxy and carboxyl group in GOQD/UCNP (red trace) significantly decreased. The peaks at 1052, 1430 and 1625 cm 1 are appeared due to the binding of GOQD with UCNP, but some absorption peaks at 799 cm 1, and 1262 cm 1 still exists which indicates the presence of part of the epoxy and -OH band of carboxyl group. The absorption observed below 799 cm 1 is also due to primary and secondary N-H groups from PEGylation. A peak is observed in red trace at 1625 cm 1 due to the stretching vibration of C˭O band in a carboxyl group. The weak peak at 1430 cm 1 corresponds to be
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
5
Fig. 2. Transmission electron microscopy (TEM) images of (A) cubic phase of UCNPs, (B) cubic-hexagonal phase of UCNPs, and (C) PEGylated UCNPs. (D) Size analysis of each UCNP through TEM imagings using Gatan program. (E) Thermal gravity analysis (TGA) curve of different nanoparticles about surface-modified UCNPs with amine group and PEGylated UCNPs. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
aromatic skeletal of C˭C band stretching vibration. The peak at 1052 cm 1 can be attributed to the vibration adsorption of C–O–C band. The FTIR spectra confirmed the binding of GOQD due to the presence of a band related to oxygen functional groups for epoxy and carboxyl group on the GOQD/UCNP. HA/GOQD/UCNP complex was synthesized and irradiated by 460 nm LED. Then, HA initiated the activity of ROS to attack on the cells. In order to understand the effect of binding of GOQD and HA with UCNPs, we investigated the surface charge difference, absorption and UCL in a visible range. As shown in Fig. 3A, the charge transition was monitored by zeta potential measurement, and it was found that positive charges were measured from UCNPs to PEGylation of UCNPs, but negative charges were measured after binding of GOQDs with PEGylated UCNPs. Absorbance spectrum of
HA (Fig. S5A, black trace) exhibited a peak at 460 nm. Absorbance spectra of HA/GOQD (Fig. S5A, green trace) and HA/GOQD/UCNP (Fig. S5A, red trace) also showed a peak at 460 nm. GOQD/UCNP (Fig. S5A, blue trace) showed no peak at 460 nm. From the absorption spectrum, it was confirmed that HA was successfully immobilized on the GOQD/UCNP. At the fixed excitation on 460 nm, fluorescence emission of synthesized products was also investigated (Fig. S5B). HA (Fig. S5B, black trace), HA/GOQD (Fig. S5B, green trace) and HA/GOQD/UCNP (Fig. S5B, red trace) exhibited strong emission at the same wavelength of 600 nm, indicating the binding of HA on the surface of nanoparticles. However, GOQD/UCNP (Fig. S5B, blue trace) showed no emission at 600 nm, indicating the absence of HA on the surface. Additionally, Fig. 3B represents the luminescence spectrum of HA/GOQD/UCNP,
Fig. 3. (A) Zeta-Potential of phase transition of UCNP, amine-functionalized UCNP, PEGylated UCNP, and GOQD/UCNP dispersed in DI water and HA/GOQD/UCNP in DMSO. (B) Fluorescence emission spectrum for HA/GOQD/UCNP compared with HA/GOQD, and GOQD/UCNP excited over 980 nm.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
Fig. 5. Fluorescence microscopic images of (A) HeLa cells incubated with HA/GOQD/UCNP, (B) HeLa cells illuminated with 980 nm, and (C) its merged image indicated binding of HeLa cells with UCNP (red).
HA/GOQD and GOQD/UCNP. The GOQD/UCNP (blue line) and HA/ GOQD/UCNP (red line) showed good luminescence at 550 nm and 660 nm compared to HA/GOQD. HA/GOQD/UCNP supposed to show the highest intensity because of the presence of HA, an excellent PS, on the surface. 3.3. In vitro cytotoxicity The HeLa cells were seeded in 96-well plate at the density of 8 103 cells/well in DMEM and cultured at 37 °C in a humidified 5% CO2 incubator for 24 h. The medium contained different concentrations of HA, HA/GOQD, GOQD/UCNP, and HA/GOQD/UCNP. The concentrations were ranged from 0.1 to 10 μg/mL (each concentration level was calculated by HA). Then prepared concentrations were applied on the plate containing HeLa cells. After incubation for 20 min, cells were washed with 1X PBS four times and examined cell viability test through MTT assay (Hu et al., 2012). The inhibitory effect of different doses of particles on cell growth was determined by MTT assay. The cells were divided into two groups for cell viability analysis based on irradiation and nonirradiation of 460 nm. For this purpose, one plate was irradiated at 460 nm for 20 min to activate HA, and another plate was not radiated, at the same time. Fig. S6 describes the percentage of HeLa cell viability after exposure to different concentrations (0.1, 0.5, 1, 5, 10 μg/mL) of HA, HA/GOQD, GOQD/UCNP and HA/GOQD/UCNP. The viability (%) of HeLa cells gradually declined as HA/GOQD/ UCNP was activated with irradiation of 460 nm. The morphology of HA/GOQD/UCNP (5 μg/mL) treated to HeLa
cells was damaged by activated HA (Fig. 4). The cell viability was 52% in irradiation of 460-nm as the non-irradiated cells showed 92%. The dose-dependent cytotoxicity was observed in both groups of HeLa cells treated with different materials. The main reason behind this phenomenon is the presence of cytotoxic agent, i.e. singlet oxygen. This singlet oxygen can cause oxidative damage to nearby cells and ultimately cause cell death. The results of different concentrations of HA confirmed that it had an insignificant effect on cell viability percentage, whether radiated or non-irradiated. Therefore, the use of HA is recommendable due to a lower systemic toxicity and little side effects. However, this drug showed some limitations such as self-aggregation in medium. The selfaggregation causes quenching of energy transfer steps before production of singlet oxygen and can damage the cells (Indrajit et al., 2003). To avoid this problem, we performed the dispersion of HA in polar solvent for more time to prevent aggregation and quenching effect. 3.4. UCL imaging in vitro HA/GOQD/UCNP was taken up by tumor cells and improved UCL signal properties. The cellular uptake of HA/GOQD/UCNP and intracellular fluorescence images were shown in Fig. 5. The HeLa cells can be detected and imaged when treated with 5 μg/mL of HA/GOQD/UCNP due to illumination at 980 nm. These results indicated that HA/GOQD/UCNP proved to be an effective probe for bioimaging.
4. Conclusion
Fig. 4. Irradiation effect of HA/GOQD/UCNP compared with control particles (HA, HA/GOQD, and GOQD/UCNP) on growth of HeLa cells after 12 h treatment. Cell cytotoxicity determined by MTT assay and expressed as percent of live cells (n¼ 4) versus the control group.
It is concluded that synthesized NaYF4:Yb3 þ , Er3 þ UCNP via solid-state method has nice solubility through PEGylation. The GOQD was bound onto the surface of UCNP by coupling chemistry with EDC for an efficient drug delivery. Then photosensitizer, HA, was bound on the surface of prepared nanoparticles through π-π stacking interaction. The HA/GOQD/UCNP applied photodynamic effects by cellular uptake and production of singlet oxygen under irradiation with better relative potency than HA/GOQD, and added cell image portion. Non-irradiation of the cells with HA/GOQD/ UCNP preserved morphology of cancer cells as compared to the cells radiated with 460 nm. These results indicated that HA/GOQD/ UCNP is an effective probe for cancer cells detection and therapy. It can be suggested that the synthesized nanoparticles can be applied to in vivo imaging for detecting and preventing diseases.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Acknowledgments This work supported by a Grant of the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (HI16C1553) and the Korea Food Research Institute (Project no. E0152200).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2016.08.094.
References Chatterjee, D.K., Fong, L.S., Zhang, Y., 2008. Adv. Drug. Deliv. Rev. 60, 1627–1637. Chen, G., Qiu, H., Prasad, P.N., Chen, X., 2014. Chem. Rev. 114, 5161–5214. Chen, H., Yuan, F., Wang, S., Xu, J., Zhang, Y., Wang, L., 2013. Biosens. Bioelectron. 48, 19–25. Diwu, Z., Lown, J.W., 1990. Photochem. Photobiol. 52, 609–616. Dong, Y., Chen, C., Zheng, X., Gao, L., Cui, Z., Yang, H., Guo, C., Chi, Y., Li, C.M., 2012. J. Mater. Chem. 22, 8764–8766. Dong, Y., Chen, C., Pang, H., Yang, H.B., Guo, C., Shao, J., Chi, Y., Li, C.M., Yu, T., 2013. Angew. Chem. Int. Ed. 52, 7800–7804. Guo, C.X., Dong, Y., Yang, H.B., Li, C.M., 2013. Adv. Energy Mater. 3, 997–1003. Haase, M., Schafer, H., 2011. Angew. Chem. Int. Ed. 50, 5808–5829. Hu, Z., Zhang, C., Huang, Y., Sun, S., Guan, W., Yao, Y., 2012. Chem. Biol. Interact. 195, 86–94. Idris, N.M., Gnanasammandhan, M.K., Zhang, J., Ho, P.C., Mahendran, R., Zhang, Y.,
7
2012. Nat. Med. 18, 1580–1585. Indrajit, R., Tymish, Y.O., Haridas, E.P., Earl, J.B., Allan, R.O., Janet, M., Thomas, J.D., Paras, N.P., 2003. J. Am. Chem. Soc. 125, 7860–7865. Jang, M.H., Ha, H.D., Lee, E.S., Liu, F., Kim, Y.H., Seo, T.S., 2015. Small 11, 3773–3781. Jiang, D., Chen, Y., Li, N., Li, W., Wang, Z., Zhu, J., Zhang, H., Liu, B., Xu, S., 2015. PLoS One 10, e0144906. Junwei, Z., Yajuan, S., Xianggui, K., Lijin, T., Yu, W., Langping, T., Jialong, Z., Hong, Z., 2008. J. Phys. Chem. B 112, 15666–15672. Konan, Y.N., Gurny, R., Allémann, E., 2002. J. Photochem. Photobiol. B 66, 89–106. Kuo, N.-J., Chen, Y.-S., Wu, C.-W., Huang, C.Y., Chan, Y.-H., Chen, I.P., 2016. Sci. Rep. 6, 30426. Li, Y., Dong, H., Li, Y., Shi, D., 2015. Int. J. Nanomed. 10, 2451–2459. Lu, Z., Chen, X., Wang, Y., Zheng, X., Li, C.M., 2015. Microchim. Acta 182, 571–578. Martin, K.R., Barrett, J.C., 2002. Hum. Exp. Toxicol. 21, 71–75. Oleinick, N.L., Morris, R.L., Belichenko, I., 2002. Photochem. Photobiol. Sci. 1, 1–21. Park, Y.I., Lee, K.T., Suh, Y.D., Hyeon, T., 2015. Chem. Soc. Rev. 44, 1302–1317. Seo, J.M., Kim, E.B., Hyun, M.S., Kim, B.B., Park, T.J., 2015. Colloid Surf. B 135, 27–34. Shan, J., Liangjun, Z., ZhanJun, G., Gan, T., Liang, Y., Wenlu, R., Wenyan, Y., Xiaodong, L., Xiao, Z., Zhongbo, H., Yuliang, Z., 2013. Nanoscale 5, 11910–11918. Sharman, W.M., Allen, C.M., van Lier, J.E., 2000. Method Enzym. 319, 376–400. Shen, J., Zhao, L., Han, G., 2013. Adv. Drug. Deliv. Rev. 65, 744–755. Sun, Y., Chen, Y., Tian, L., Yu, Y., Kong, X., Zhao, J., 2007. Nanotechnology 18, 275609. Thomas, F.H., Christopher, A., Neal, T.S., Michael, A.W., Daniel, T.B., Alan, K.S., 2010. J. Am. Chem. Soc. 16, 5735–5742. Thomas, H.R., Marsden, A.J., Walker, M., Wilson, N.R., Rourke, J., 2015. Angew. Chem. Int. Ed. 53, 7613–7618. Zhang, F., Liu, F., Wang, C., Xin, X., Liu, J., Guo, S., Zhang, J., 2016. ACS Appl. Mater. Interfaces 8, 2104–2110. Zhang, J., Cao, E., Li, J., Zhang, T., Ma, W., 1998. J. Photochem. Photobiol. B 43, 106–111. Zhang, T., Ge, J., Hu, Y., Yin, Y., 2007. Nano Lett. 7, 3203–3207. Zheng, X.T., He, H.L., Li, C.M., 2013. RSC Adv. 3, 24853–24857. Zhou, L., Zhou, L., Ge, X., Zhou, J., Wei, S., Shen, J., 2015. Chem. Commun. 51, 421–424.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
BIOS : 9091
Model5G
pp:17ðcol:fig: : NILÞ Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Synthesis of upconversion nanoparticles conjugated with graphene oxide quantum dots and their use against cancer cell imaging and photodynamic therapy Seung Yoo Choi a, Seung Hoon Baek a, Sung-Jin Chang a, Yohan Song b, Rafia Rafique a, Kang Taek Lee b, Tae Jung Park a,n a b
Department of Chemistry, Chung-Ang University, 84 Heukseok-ro, Dongjak-gu, Seoul 06974, Republic of Korea Department of Chemistry, Gwangju Institute of Science and Technology, 123 Cheomdan-gwagi-ro, Buk-gu, Gwangju 61005, Republic of Korea
art ic l e i nf o
a b s t r a c t
Article history: Received 3 June 2016 Received in revised form 26 August 2016 Accepted 27 August 2016
Multifunctional nanocomposite has a huge potential for cell imaging, drug delivery, and improving therapeutic effect with less side effects. To date, diverse approaches have been demonstrated to endow a single nanostructure with multifunctionality. Herein, we report the synthesis and application of coreshell nanoparticles composed with upconversion nanoparticle (UCNP) as a core and a graphene oxide quantum dot (GOQD) as a shell. The UCNP was prepared and applied for imaging-guided analyses of upconversion luminescence. GOQD was prepared and employed as promising drug delivery vehicles to improve anti-tumor therapy effect in this study. Unique properties of UCNPs and GOQDs were incorporated into a single nanostructure to provide desirable functions for cell imaging and drug delivery. In addition, hypocrellin A (HA) was loaded on GOQDs for photo-dynamic therapy (PDT). HA, a commonly used chemotherapy drug and a photo-sensitizer, was conjugated with GOQD by π-π interaction and loaded on PEGylated UCNP without complicated synthetic process, which can break structure of HA. Applying these core-shell nanoparticles to MTT assay, we demonstrated that the UCNPs with GOQD shell loaded with HA could be excellent candidates as multifunctional agents for cell imaging, drug delivery and cell therapy. & 2016 Elsevier B.V. All rights reserved.
Keywords: Upconversion nanoparticle Graphene oxide quantum dot Hypocrellin A Photodynamic therapy Photoluminescence
1. Introduction Lanthanide-doped upconversion nanoparticles (UCNPs) have been attracted a great interest in various fields because of their unique optical properties reported previously (Chen et al., 2014). Thus, these nanostructures have a specific value for biological applications in various therapeutic fields such as in vivo imaging and photodynamic cancer therapies (Shen et al., 2013). In other words, lanthanide-doped UCNPs with ladder-like energy level structures have an ability to convert low energy excitation to highenergy emission via two or more photon absorption or energy transfer. Upconversion refers to sequential absorption of multiphotons leading to the emission shorter than the excited wavelength, which is anti-stokes type emission (Junwei et al., 2008). UCNPs have been differentiated from others in terms of very little auto-fluorescence and scattering on multiphoton absorption. Consequentially, they have a low background signal and high n
Corresponding author. E-mail address: [email protected] (T.J. Park).
signal-to-noise ratio offering outstanding advantages in application of therapeutic fields. Therefore, UCNPs have become efficient candidates for the analysis of biological and environmental samples due to their bio-compatibility, especially for in vivo and in vitro fluorescence imaging (Haase and Schafer, 2011; Park et al., 2015; Seo et al., 2015). The similar atomic size and chemical properties of lanthanide elements make them more biocompatible and stable among other elements. Photodynamic therapy (PDT) is a clinical tumor treatment using nontoxic light-sensitive agents that are toxic to target malignant and other diseased cells (Idris et al., 2012). When photosensitizing agents are exposed to a particular type of light, they produce cytotoxic reactive oxygen species (ROS). ROS leads nearby cells to death with minimal invasiveness, toxicity and side effect (Sharman et al., 2000; Martin and Barrett, 2002; Chatterjee et al., 2008). PDT requires photosensitizer (PS), light source, and oxygen. PS helps production of singlet oxygen by absorbing ultraviolet or visible region and transferring it to adjacent molecules (Konan et al., 2002). The PS transforms tumor cells with excited singlet state by intersystem crossing as a result of an irradiation at a
http://dx.doi.org/10.1016/j.bios.2016.08.094 0956-5663/& 2016 Elsevier B.V. All rights reserved.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
2
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
specific wavelength of light. The excited triplet state can transfer hydrogen atom or electron to form radicals interacting with oxygen. These radicals produce oxygenated products like hydrogen peroxide, super oxygen, and hydroxyl ion or can transfer from oxygen to singlet oxygen, which is highly reactive. As a result, the ROS leads tumor cell to death. Hypocrellin A (HA) is one of PSs isolated from natural fungus, Hypocrella bambusae Sacc., which was discovered in China (Zhang et al., 1998). HA has low toxicity in an inactivated form (Li et al., 2015) and can be metabolized in vivo. Thus, HA can be used in some capillary disease treatment due to higher PDT effect under presence of singlet oxygen than many widely used PSs (Diwu et al., 1990; Shan et al., 2013). Graphene oxide quantum dot (GOQD) is a class of carbon-based dots. The GOQD has received tremendous attention, due to its outstanding advantages such as low toxicity, low cost, excellent biocompatibility, resistance to photobleaching and ease of fabrication (Dong et al., 2012, 2013; Jiang et al., 2015; Lu et al., 2015; Zhang et al., 2016). Both GOQD and HA can easily form the HA/ GOQD complexes through π-π stacking due to same planar sp2 structures (Thomas et al., 2010; Jang et al., 2015) without complicated synthetic process which can break the structure of PS (Guo et al., 2013). Other materials such as organic dyes and semiconductor quantum dots (QDs) have been used as biological luminescent labels, but they have certain limitations. For example, organic dyes exhibit rapid photobleaching, and QDs are less chemically stable, inherently toxic and show fluorescence intermittence. Hence, NIRto-visible UCNP seems to be a promising alternative fluorescent material for bio-detection based on their advantages such as high chemical stability, low toxicity, economic instrument cost, and tunable optical properties. In this study, we report ytterbium and erbium ions-doped sodium yttrium fluoride (NaYF4:Yb3 þ , Er3 þ ), the most efficient NIR-to-visible UCNP in solid-state materials doped with rare-earth ions. NaYF4:Yb3 þ , Er3 þ UCNP has been synthesized and analyzed their unique advantages. We designed multifunctional nanostructure composed of UCNPs, GOQDs and HA for cell imaging drug delivery and PDT of cancer cell (Scheme 1). First, ytterbium and erbium ions-doped sodium yttrium fluoride (NaYF4:Yb3 þ , Er3 þ ) nanoparticles were synthesized as UCNP. Polyethylene glycol (PEG) was coated on the surface of the UCNPs additionally for bio-compatibility. The PEGylation was performed to make UCNP more effective in disease
monitoring and drug delivery system. Next, we attached GOQDs to the PEG-coated UCNPs for drug delivery carrier. Finally, multifunctional material, GOQD-attached UCNP were synthesized with a great potential for both upconversion luminescence (UCL) imaging and PDT in cancer diagnostics. Furthermore, HA as a PS was loaded on the surface of the GOQD-attached UCNP (Oleinick et al., 2002; Zhou et al., 2015). This drug-delivery system was proposed to minimize side effect, achieve a delivery system for PDT and help in vitro imaging (Zheng et al., 2013). These nanocomposites were designed due to their intrinsic properties for drug-delivery from GOQD, therapy from HA and imaging capability from UCL.
2. Experimental 2.1. Materials Yttrium nitrate hexahydrate, ytterbium chloride hexahydrate, erbium nitrate pentahydrate, sodium fluoride, and nitric acid were purchased from Sigma-Aldrich (St Louis, MO, USA). Sodium citrate was obtained from APS Biotech (Seoul, Korea). Ethyl alcohol was acquired from Emsure (Billerca, MA, USA). Cetyltrimethylammonium bromide (CTAB) was purchased from Daejung (Seoul, Korea). All solutions were prepared with deionized (DI) water (Direct-Qs Water Purification System, Millipore, Billerica, MA, USA). Graphite nanoparticle (GNP) (93%, 3–4 nm) was obtained from SkySpring Nanomaterials (Houston, TX, USA). Sulfuric acid (95.0%) purchased from Junsei (Tokyo, Japan) and Nitric acid (60.0%) obtained from Samchun (Seoul, Korea). HA ( 477%) was purchased from abcam (Cambridge, UK). 2.2. Instruments and characterizations The UCL spectra under 980 nm continuous wave laser excitation were analyzed using Shamrock Spectrograph (Shamrock 303i, Andor Technology, Belfast, Ireland). The absorbance was analyzed by UV/Visible/NIR spectrophotometer (V-670, Jasco, Tokyo, Japan). Their fluorescence properties were also analyzed using the Multimode microplate reader (BioTek Synergy H1, Winooski, VT, USA), absorbance on methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay. The particle size was determined by transmission electron microscopy (TEM, Tecnai G2 F30 S-TWIN, FEI, Hillsboro, OR, USA) and particle size analyzer (Otsuka ELSZ-1000, Osaka,
Scheme 1. The action procedure of HA/GOQD/UCNP nanoparticle.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Japan). Additionally, the crystal structure of UCNP was analyzed using X-ray powder diffractometer (XRD, Bruker-AXS New D8Advance, Bruker, Berlin, Germany) with Cu-radiation from Cu-Kα X-ray source (λ ¼1.54056 Å) ranging from 5°≦2θ≦70°. PEGylation of UCNP was checked by thermal gravimetric analysis (TGA, N-1000, Scinco, Seoul, Korea). For HA/GOQD/UCNP imaging in the HeLa cells, inverted microscope (IX73, Olympus, Tokyo, Japan) was composed of a 980nm NIR diode laser (P161-600-980A, EM4), an electron multiplying charge-coupled device (EMCCD) camera (iXON3, Andor Technology) and a live-cell incubation chamber (TC-L-10, Live Cell Instrument, Seoul, Korea) on the stage. The cell samples on the cover glass bottom dish were mounted on the microscope stage in the incubation chamber which provided optimal conditions (37 °C and 5% CO2) for live-cell imaging. Typical power density of illumination was 21.75 mW/cm2. 2.3. Synthesis of therapeutic probe HA/GOQD/UCNP 2.3.1. Synthesis and surface modification NaYF4:Yb3 þ ,Er3 þ UCNP and PEGylation (The first cycle of UCNP synthesis, “One-cycle”) NaYF4:Yb3 þ , Er3 þ UCNP was synthesized according to the previously reported procedure (Sun et al., 2007; Chen et al., 2013). 3.6 mL of 0.2 M Y(NO3)3, 3 mL of 0.1 M YbCl3, 0.3 mL of 0.1 M Er (NO3)3, and 6 mL of 0.1 M citrate sodium were vigorously mixed at room temperature. Then, 6.3 mL DI water, 45 mL ethanol, and 0.3 g CTAB was subsequently added to the solution. Next, 18 mL of 1.0 M NaF was added dropwise and the solution was incubated with vigorous stirring for 2 h at room temperature. Then, 3 mL of HNO3 was mixed in the solution. Mixture solution was autoclaved in the teflon liner-packed autoclave vessel sized with 23 mL, which was heated to 180 °C for 4 h. Reaction solution was cooled in the air condition and then the synthesized cubic phase of UCNP was prepared (Fig. 1A). The UCNPs were cooled down to room temperature and then separated by centrifugation at 3,500 rpm for 10 min. (The second cycle of UCNP synthesis, “two-step”). As-prepared cubic phases of UCNPs were dispersed in DI water as seed materials to synthesize cubic-hexagonal phases of UCNPs. The UCNPs were dried off completely at 60 °C in dry air oven. The prepared UCNPs were finalized with overall reaction one more time from a previous synthetic procedure. As a result, the cubichexagonal phase of UCNP was developed. UCNP was further proceeded for surface modification and
3
functionalization with amine groups (Zhang at al., 2007). The mixture solution of diethylene glycol (DEG) and poly(allylamine) (PAAm) were prepared in the neck flask, and heated under the N2 gas environment while keeping the temperature at 110 °C with reflux. Then, UCNP was mixed in 10 mL of DI water and injected rapidly into the heated mixture. Reaction was proceeded at 240 °C for another half an hour. When reaction was finished, the reactant was cooled to room temperature and a solution of 0.1 M of hydrochloric acid was injected. The final product was separated by centrifugation and washed three times with DI water. Amine-functionalized UCNP (20 mg) was dissolved in 20 mL 2(N-morpholino)ethanesulfonic acid (MES) buffer (0.1 M, pH 6.0), then sonicated for 30 min (100 W). 20 mg of amine-PEG (polyethylene glycol 2-aminoethyl ether acetic acid) was added into asprepared UCNP solution and sonicated for 10 min (100 W). Subsequently, 6 mL of aqueous EDC (4 mg/mL) was added in the solution, mixed and stirred for 24 h at room temperature. The resulting PEGylated UCNP was centrifuged at 3,500 rpm for 10 min and washed with DI water for several times. 2.3.2. Preparation of the GOQD and binding of GOQD on UCNPs using EDC coupling 1 g of graphite nanoparticles (GNPs) was pre-oxidized in c-H2SO4 (60 mL) and HNO3 (20 mL) for 3 h in mild sonication (100 W) (Thomas et al., 2015; Kuo et al., 2016). Next, the mixture was stirred at 100 °C under a refluxing condition for 14 h. After that, the mixture was cooled down to 25 °C and diluted with 150 mL of DI water in the ice bath. The suspension was centrifuged to remove the aggregates of GOQD at 14,000 rpm for 30 min and the supernatant was separated. Then, supernatant solution was dialyzed for 24 h to remove the acidic ions, and the pH of the aqueous solution was adjusted to 7.0 with 1.0 M NaOH. The final solution was dialyzed to remove the salt for 7 days at room temperature (Dong et al., 2012). PEGylated UCNP (10 mg) was dissolved in 10 mL MES buffer (0.1 M, pH 6.0) and then sonicated for 30 min. GOQD (OD220 ¼ 1.0, 10 mL) was added into that solution and stirred for 10 min. Subsequently, 4 mL aqueous solution of EDC (4 mg/mL) was added dropwise into mixture solution and stirred for 24 h at 25 °C. The resulting GOQD/UCNP was centrifuged at 3,500 rpm for 10 min and washed with DI water. 2.3.3. HA loading through π-π stacking and releasing Lastly, the conjugation of GOQD/UCNPs with HA was performed by dispersing in N,N-dimethyl-formamide (DMF) as HA dissolves
Fig. 1. (A) X-ray powder diffraction (XRD) pattern with the lower side about cubic phase (PDF no. 01-077-2042) and the upper side of cubic-hexagonal (PDF no. 00-0270689) phase of NaYF4:Yb3 þ ,Er3 þ nanoparticles. The well-defined peaks indicated the high crystallinity of the synthesized material. (B) Photoluminescence spectroscopy represented energy transfer and multiphoton relaxation in cubic and cubic-hexagonal phase of NaYF4:Yb3 þ ,Er3 þ nanoparticles. Er3 þ emitted photon at 550 and 660 nm under excitation of 980 nm (300 mW).
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
4
in polar aprotic solvents. In detail, 100 mL of 15 mM HA in DMF was dissolved in 10 mL DI water. Then, 8 mg GOQD/UCNP was dispersed in 10 mL ultra-pure water using sonication (100 W) for 10 min. These aqueous suspensions were mixed well with slow stirring at room temperature for 24 h to form HA/GOQD/UCNP complex. 2.4. Cell culture and cellular uptake The HeLa cells were seeded in cell culture plate at the density of 1.5 105 cells/mL in DMEM. Shortly thereafter, the cells were cultivated with 10% Fetal bovine serum (FBS) and 1% antibiotic antimycotic (AAM) (5.1% CO2, 37 °C) for 24 h. HA/GOQD/UCNP was treated in medium (each concentration calculated by HA) on the plate. After incubation for 20 min, cells were washed with 1X phosphate-buffered saline (PBS, pH 7.4) for four times and monitored using microscopy through bright filter and 980-nm filter. The cell viability was checked through MTT assay (Hu et al., 2012). HeLa cells were seeded in 96-well plate at the density of 8 103 cells/well. The cells were divided into four groups, and each group was treated with HA/GOQD/UCNP for 20 min and then washed with PBS for four times. After washing, following conditions were applied to each group to determine cell viability; (1) they were placed in a dark incubator for 12 h, (2) irradiated by 460-nm LED for 20 min and placed in a dark incubator for 12 h. 2.5. UCL imaging in vitro 1 105 cells/mL of HeLa cells were seeded on the cover glass bottom dish (SPL, Korea), and cultured at 37 °C in a humidified 5% CO2 incubator for 24 h. Then the medium was changed with 1 mL of HA/GOQD/UCNP containing medium (5 μg/mL). After additional incubation for 30 min, the cell samples were washed and used for live-cell imaging. The image of HA/GOQD/UCNP-treated HeLa cell was recorded by excitation at 980 nm with CW laser diode.
3. Results and discussion 3.1. Preparation and characterization of UCNP The crystal structure of NaYF4:Yb3 þ , Er3 þ UCNP was examined by XRD. Fig. 1A shows XRD spectra of UCNPs synthesized after one cycle process (Left and middle panel, (1)) and two-step process (Left and top panel, (2)). The XRD peaks of cubic and hexagonal phases were assigned with PDF no. 01-077-2042 (Fig. 1A, Left and bottom panel, blue lines) and PDF no. 00-027-0689 (Fig. 1A, Left and bottom panel, red lines), respectively. Both UCNP samples after one-cycle and two-step synthesis dominantly exhibit cubic phase. However, XRD peaks of hexagonal phase are discernible in XRD spectrum of the UCNPs after two-step synthesis, suggesting that some of the cubic phase UCNPs has been transformed into hexagonal phase during the second synthetic process. It is worth noting that phase transition from cubic to hexagonal structure cannot be achieved by one-cycle synthesis but only through the two-step synthesis (Fig. S1). The UCL of one-cycle and two-step synthesis dominantly exhibits cubic phase. Fig. 1B shows UCL spectra of two types of UCNP samples (1 mg/mL in DI water), one produced by one-cycle synthesis (black trace) and the other produced by two-step synthesis (red trace). Cubic-hexagonal phase of UCNPs exhibit about 10 times higher intensity of UCL than cubic phase of UCNPs. This result confirms that a part of cubic phase UCNPs has changed into hexagonal phase UCNPs (Park et al., 2015). Therefore, it can be suggested that our UCNP samples after two-step synthetic process are composed of two types of UCNPs, one is cubic phase, and the
other is hexagonal phase. Moreover, UCL characteristics of the UCNP samples after two-step synthesis are dominated by those of the hexagonal phase UCNPs. Fig. 1B depicts that Yb3 þ ion absorbs 980 nm (300 mW, 1 mg/mL in DI water) excited photon and then transfers its energy next to the Er3 þ ion which acts as an activator. Upconversion fluorescence spectra of NaYF4 nanoparticles appears green and red emissions of 550-nm and 660-nm regions, respectively (Haase and Schafer, 2011; Park et al., 2015). The synthesized cubic-hexagonal phases of UCNP have higher emission intensity as compared to cubic phase of UCNP. To verify the crystallinity and the presence of the GOQD and UCNP, NaYF4:Yb3 þ , Er3 þ UCNP, and PEGylated UCNP were characterized with a transmission electron microscope (TEM) as shown in Fig. 2 and S1. The results show that the particle size of GOQD was 5 nm (Fig. S2A). Fig. S2B shows the diffraction pattern image (inset) and the size of cubic-hexagonal phase of UCNPs which was around 105.057 16.79 nm. TEM imaging was also used to compare the UCNP size with phase transition and PEGylation of UCNPs (Fig. 2C and Fig. S2C). The average diameter of cubic phase UCNPs and PEGylated UCNPs were 77.02 7 15.02 nm and 112.26 720.64 nm, respectively (Fig. 2D). Moreover, PEGylation of UCNPs was also confirmed by using TGA (Fig. 2E). The red line curve indicated that PEGylated UCNPs loss their weight with increasing temperature and 1.0 wt% weight loss was observed below 200 °C. The similar trend was observed for amine-functionalized UCNPs (black line) and no weight loss occurred below 200 °C. The results were analyzed below 200 °C because PEG has a boiling point below this range. 3.2. Binding of GOQD and HA Fig. S3A shows that HA and GOQD did not have photoluminescence at 980 nm. The concentration of GOQD on individual UCNPs needed to be optimized as GOQDs-attached UCNP had low UCL efficiency. From several values of OD220, the optimum values for UCL of GOQD-attached PEG-coated UCNP were determined to be 0.1, 0.3, 0.5, 0.8 and 1.0 (Fig. S3B). Fig. S3B was shown OD220 value 1.0 of GOQD have a minor influence on UCL. The GOQD also had a wide HA binding sight on the surface of GOQD/UCNP. Therefore, we prepared the complex of GOQD-attached UCNP with OD220 value 1.0 of GOQD based on core-shell nanoparticles. HA bound with GOQDs on the surface of PEG-coated UCNPs via π-π interaction to examine PDT activity. The fourier transform infrared (FTIR) spectroscopy was used to characterize the chemical bonds and compositions of as-prepared samples. The change of the oxygen functional groups from GOQD can be reflected in FTIR spectroscopy, as shown in Fig. S4. In the FTIR spectra of PEGylated UCNP and GOQD/UCNP composite, a broad absorption band appears in the range of 3000–3700 cm 1 due to the stretching vibration of -OH band and with primary and secondary N–H groups that also absorb in this range at 3300–3500 cm 1. For a neat liquid, the N–H peak is typically weaker than O–H. The carboxyl group -OH stretch appears as a very broad band in the region 3300–2500 cm 1 while the N–H stretches of primary amines are two bands from 3300 to 3500 and 3250–3330 cm 1. Compared with PEGylated UCNP (black trace), the intensity of all absorption peaks related to oxygen functional groups about epoxy and carboxyl group in GOQD/UCNP (red trace) significantly decreased. The peaks at 1052, 1430 and 1625 cm 1 are appeared due to the binding of GOQD with UCNP, but some absorption peaks at 799 cm 1, and 1262 cm 1 still exists which indicates the presence of part of the epoxy and -OH band of carboxyl group. The absorption observed below 799 cm 1 is also due to primary and secondary N-H groups from PEGylation. A peak is observed in red trace at 1625 cm 1 due to the stretching vibration of C˭O band in a carboxyl group. The weak peak at 1430 cm 1 corresponds to be
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
5
Fig. 2. Transmission electron microscopy (TEM) images of (A) cubic phase of UCNPs, (B) cubic-hexagonal phase of UCNPs, and (C) PEGylated UCNPs. (D) Size analysis of each UCNP through TEM imagings using Gatan program. (E) Thermal gravity analysis (TGA) curve of different nanoparticles about surface-modified UCNPs with amine group and PEGylated UCNPs. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
aromatic skeletal of C˭C band stretching vibration. The peak at 1052 cm 1 can be attributed to the vibration adsorption of C–O–C band. The FTIR spectra confirmed the binding of GOQD due to the presence of a band related to oxygen functional groups for epoxy and carboxyl group on the GOQD/UCNP. HA/GOQD/UCNP complex was synthesized and irradiated by 460 nm LED. Then, HA initiated the activity of ROS to attack on the cells. In order to understand the effect of binding of GOQD and HA with UCNPs, we investigated the surface charge difference, absorption and UCL in a visible range. As shown in Fig. 3A, the charge transition was monitored by zeta potential measurement, and it was found that positive charges were measured from UCNPs to PEGylation of UCNPs, but negative charges were measured after binding of GOQDs with PEGylated UCNPs. Absorbance spectrum of
HA (Fig. S5A, black trace) exhibited a peak at 460 nm. Absorbance spectra of HA/GOQD (Fig. S5A, green trace) and HA/GOQD/UCNP (Fig. S5A, red trace) also showed a peak at 460 nm. GOQD/UCNP (Fig. S5A, blue trace) showed no peak at 460 nm. From the absorption spectrum, it was confirmed that HA was successfully immobilized on the GOQD/UCNP. At the fixed excitation on 460 nm, fluorescence emission of synthesized products was also investigated (Fig. S5B). HA (Fig. S5B, black trace), HA/GOQD (Fig. S5B, green trace) and HA/GOQD/UCNP (Fig. S5B, red trace) exhibited strong emission at the same wavelength of 600 nm, indicating the binding of HA on the surface of nanoparticles. However, GOQD/UCNP (Fig. S5B, blue trace) showed no emission at 600 nm, indicating the absence of HA on the surface. Additionally, Fig. 3B represents the luminescence spectrum of HA/GOQD/UCNP,
Fig. 3. (A) Zeta-Potential of phase transition of UCNP, amine-functionalized UCNP, PEGylated UCNP, and GOQD/UCNP dispersed in DI water and HA/GOQD/UCNP in DMSO. (B) Fluorescence emission spectrum for HA/GOQD/UCNP compared with HA/GOQD, and GOQD/UCNP excited over 980 nm.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
Fig. 5. Fluorescence microscopic images of (A) HeLa cells incubated with HA/GOQD/UCNP, (B) HeLa cells illuminated with 980 nm, and (C) its merged image indicated binding of HeLa cells with UCNP (red).
HA/GOQD and GOQD/UCNP. The GOQD/UCNP (blue line) and HA/ GOQD/UCNP (red line) showed good luminescence at 550 nm and 660 nm compared to HA/GOQD. HA/GOQD/UCNP supposed to show the highest intensity because of the presence of HA, an excellent PS, on the surface. 3.3. In vitro cytotoxicity The HeLa cells were seeded in 96-well plate at the density of 8 103 cells/well in DMEM and cultured at 37 °C in a humidified 5% CO2 incubator for 24 h. The medium contained different concentrations of HA, HA/GOQD, GOQD/UCNP, and HA/GOQD/UCNP. The concentrations were ranged from 0.1 to 10 μg/mL (each concentration level was calculated by HA). Then prepared concentrations were applied on the plate containing HeLa cells. After incubation for 20 min, cells were washed with 1X PBS four times and examined cell viability test through MTT assay (Hu et al., 2012). The inhibitory effect of different doses of particles on cell growth was determined by MTT assay. The cells were divided into two groups for cell viability analysis based on irradiation and nonirradiation of 460 nm. For this purpose, one plate was irradiated at 460 nm for 20 min to activate HA, and another plate was not radiated, at the same time. Fig. S6 describes the percentage of HeLa cell viability after exposure to different concentrations (0.1, 0.5, 1, 5, 10 μg/mL) of HA, HA/GOQD, GOQD/UCNP and HA/GOQD/UCNP. The viability (%) of HeLa cells gradually declined as HA/GOQD/ UCNP was activated with irradiation of 460 nm. The morphology of HA/GOQD/UCNP (5 μg/mL) treated to HeLa
cells was damaged by activated HA (Fig. 4). The cell viability was 52% in irradiation of 460-nm as the non-irradiated cells showed 92%. The dose-dependent cytotoxicity was observed in both groups of HeLa cells treated with different materials. The main reason behind this phenomenon is the presence of cytotoxic agent, i.e. singlet oxygen. This singlet oxygen can cause oxidative damage to nearby cells and ultimately cause cell death. The results of different concentrations of HA confirmed that it had an insignificant effect on cell viability percentage, whether radiated or non-irradiated. Therefore, the use of HA is recommendable due to a lower systemic toxicity and little side effects. However, this drug showed some limitations such as self-aggregation in medium. The selfaggregation causes quenching of energy transfer steps before production of singlet oxygen and can damage the cells (Indrajit et al., 2003). To avoid this problem, we performed the dispersion of HA in polar solvent for more time to prevent aggregation and quenching effect. 3.4. UCL imaging in vitro HA/GOQD/UCNP was taken up by tumor cells and improved UCL signal properties. The cellular uptake of HA/GOQD/UCNP and intracellular fluorescence images were shown in Fig. 5. The HeLa cells can be detected and imaged when treated with 5 μg/mL of HA/GOQD/UCNP due to illumination at 980 nm. These results indicated that HA/GOQD/UCNP proved to be an effective probe for bioimaging.
4. Conclusion
Fig. 4. Irradiation effect of HA/GOQD/UCNP compared with control particles (HA, HA/GOQD, and GOQD/UCNP) on growth of HeLa cells after 12 h treatment. Cell cytotoxicity determined by MTT assay and expressed as percent of live cells (n¼ 4) versus the control group.
It is concluded that synthesized NaYF4:Yb3 þ , Er3 þ UCNP via solid-state method has nice solubility through PEGylation. The GOQD was bound onto the surface of UCNP by coupling chemistry with EDC for an efficient drug delivery. Then photosensitizer, HA, was bound on the surface of prepared nanoparticles through π-π stacking interaction. The HA/GOQD/UCNP applied photodynamic effects by cellular uptake and production of singlet oxygen under irradiation with better relative potency than HA/GOQD, and added cell image portion. Non-irradiation of the cells with HA/GOQD/ UCNP preserved morphology of cancer cells as compared to the cells radiated with 460 nm. These results indicated that HA/GOQD/ UCNP is an effective probe for cancer cells detection and therapy. It can be suggested that the synthesized nanoparticles can be applied to in vivo imaging for detecting and preventing diseases.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i
S.Y. Choi et al. / Biosensors and Bioelectronics ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Acknowledgments This work supported by a Grant of the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (HI16C1553) and the Korea Food Research Institute (Project no. E0152200).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2016.08.094.
References Chatterjee, D.K., Fong, L.S., Zhang, Y., 2008. Adv. Drug. Deliv. Rev. 60, 1627–1637. Chen, G., Qiu, H., Prasad, P.N., Chen, X., 2014. Chem. Rev. 114, 5161–5214. Chen, H., Yuan, F., Wang, S., Xu, J., Zhang, Y., Wang, L., 2013. Biosens. Bioelectron. 48, 19–25. Diwu, Z., Lown, J.W., 1990. Photochem. Photobiol. 52, 609–616. Dong, Y., Chen, C., Zheng, X., Gao, L., Cui, Z., Yang, H., Guo, C., Chi, Y., Li, C.M., 2012. J. Mater. Chem. 22, 8764–8766. Dong, Y., Chen, C., Pang, H., Yang, H.B., Guo, C., Shao, J., Chi, Y., Li, C.M., Yu, T., 2013. Angew. Chem. Int. Ed. 52, 7800–7804. Guo, C.X., Dong, Y., Yang, H.B., Li, C.M., 2013. Adv. Energy Mater. 3, 997–1003. Haase, M., Schafer, H., 2011. Angew. Chem. Int. Ed. 50, 5808–5829. Hu, Z., Zhang, C., Huang, Y., Sun, S., Guan, W., Yao, Y., 2012. Chem. Biol. Interact. 195, 86–94. Idris, N.M., Gnanasammandhan, M.K., Zhang, J., Ho, P.C., Mahendran, R., Zhang, Y.,
7
2012. Nat. Med. 18, 1580–1585. Indrajit, R., Tymish, Y.O., Haridas, E.P., Earl, J.B., Allan, R.O., Janet, M., Thomas, J.D., Paras, N.P., 2003. J. Am. Chem. Soc. 125, 7860–7865. Jang, M.H., Ha, H.D., Lee, E.S., Liu, F., Kim, Y.H., Seo, T.S., 2015. Small 11, 3773–3781. Jiang, D., Chen, Y., Li, N., Li, W., Wang, Z., Zhu, J., Zhang, H., Liu, B., Xu, S., 2015. PLoS One 10, e0144906. Junwei, Z., Yajuan, S., Xianggui, K., Lijin, T., Yu, W., Langping, T., Jialong, Z., Hong, Z., 2008. J. Phys. Chem. B 112, 15666–15672. Konan, Y.N., Gurny, R., Allémann, E., 2002. J. Photochem. Photobiol. B 66, 89–106. Kuo, N.-J., Chen, Y.-S., Wu, C.-W., Huang, C.Y., Chan, Y.-H., Chen, I.P., 2016. Sci. Rep. 6, 30426. Li, Y., Dong, H., Li, Y., Shi, D., 2015. Int. J. Nanomed. 10, 2451–2459. Lu, Z., Chen, X., Wang, Y., Zheng, X., Li, C.M., 2015. Microchim. Acta 182, 571–578. Martin, K.R., Barrett, J.C., 2002. Hum. Exp. Toxicol. 21, 71–75. Oleinick, N.L., Morris, R.L., Belichenko, I., 2002. Photochem. Photobiol. Sci. 1, 1–21. Park, Y.I., Lee, K.T., Suh, Y.D., Hyeon, T., 2015. Chem. Soc. Rev. 44, 1302–1317. Seo, J.M., Kim, E.B., Hyun, M.S., Kim, B.B., Park, T.J., 2015. Colloid Surf. B 135, 27–34. Shan, J., Liangjun, Z., ZhanJun, G., Gan, T., Liang, Y., Wenlu, R., Wenyan, Y., Xiaodong, L., Xiao, Z., Zhongbo, H., Yuliang, Z., 2013. Nanoscale 5, 11910–11918. Sharman, W.M., Allen, C.M., van Lier, J.E., 2000. Method Enzym. 319, 376–400. Shen, J., Zhao, L., Han, G., 2013. Adv. Drug. Deliv. Rev. 65, 744–755. Sun, Y., Chen, Y., Tian, L., Yu, Y., Kong, X., Zhao, J., 2007. Nanotechnology 18, 275609. Thomas, F.H., Christopher, A., Neal, T.S., Michael, A.W., Daniel, T.B., Alan, K.S., 2010. J. Am. Chem. Soc. 16, 5735–5742. Thomas, H.R., Marsden, A.J., Walker, M., Wilson, N.R., Rourke, J., 2015. Angew. Chem. Int. Ed. 53, 7613–7618. Zhang, F., Liu, F., Wang, C., Xin, X., Liu, J., Guo, S., Zhang, J., 2016. ACS Appl. Mater. Interfaces 8, 2104–2110. Zhang, J., Cao, E., Li, J., Zhang, T., Ma, W., 1998. J. Photochem. Photobiol. B 43, 106–111. Zhang, T., Ge, J., Hu, Y., Yin, Y., 2007. Nano Lett. 7, 3203–3207. Zheng, X.T., He, H.L., Li, C.M., 2013. RSC Adv. 3, 24853–24857. Zhou, L., Zhou, L., Ge, X., Zhou, J., Wei, S., Shen, J., 2015. Chem. Commun. 51, 421–424.
Please cite this article as: Choi, S.Y., et al., Biosensors and Bioelectronics (2016), http://dx.doi.org/10.1016/j.bios.2016.08.094i