Doxorubicin loaded gold nanorods: a multifunctional chemo-photothermal nano-platform for cancer management

One of the limitations associated with cancer treatment is low efficacy and high dose-related side effects of anticancer drugs. The purpose of the current study was to fabricate biocompatible multifunctional drug loaded nano-moieties for co-therapy (chemo-photothermal therapy) with maximum efficiency and minimum side effects. Herein, we report in vitro anticancerous effects of doxorubicin (DOX) loaded on polyelectrolyte-poly (sodium-4-styrenesulfonate) coated Gold nanorods (PSS-GNRs) with and without NIR laser (808 nm, power density = 1.5 W/cm for 2 min) exposure. Drug loading capacity of PSS-GNRs was about 76% with drug loading content of 3.2 mg DOX/mL. Cumulative DOX release significantly increased after laser exposure (1.5


Introduction
Regardless of enormous advances in medical research, cancer is still the second leading cause of death worldwide and has killed 8.8 million peoples in 2015 [1]. Currently, there are number of treatment modalities, including chemotherapy, immunotherapy, targeted therapy, radiations, and surgery [2]. Among these, chemotherapy is the most commonly used method for treatment of various cancers [3]. However, use of conventional chemotherapeutic agents in cancer treatment is limited, due to several unwanted characteristics of poor solubility, broad bioavailability range, narrow therapeutic index, rapid elimination from systemic circulation, unselective site of action after oral/intravenous administration, and cytotoxic effects on normal tissues [4]. Such problems can be overcome by advanced drug delivery systems which offer carrier systems to hold sufficient amount of drug, with prolonged circulation time sustained drug release within tumor tissues and distinctive accumulation behavior at the tumor site [5]. Nanocarriers can improve 3 pharmacological properties of free drugs that contribute towards enhanced therapeutic efficacy in physiological environment [6].
Designing of multi-functional tumor targeting and therapeutic agent nanocarriers, provides a multimodal and effective approach such as significant absorption or scattering in the visible and near-infrared (NIR) regions, tunable aspect ratio, biocompatibility, fluorescence properties and ease of biofunctionalization make them ideal nanocarriers in biomedical applications [8]. Cetyl Trimethyl Ammonium Bromide (CTAB) is indispensable for synthesizing particles to direct the growth and stabilize the shape of gold nanorods but CTAB coated GNRs cannot be used in applications with the cells because of their high cytotoxicity [9]. Different polymers can be used to coat GNRs, to enhance their biocompatibility and dispersion at physiological pH. The positive CTAB layer on GNRs surface facilitates electrostatic adsorption of anionic compounds such as polysodium 4-styrenesulfonate (PSS) which ultimately facilitate electrostatic interaction with cationic anticancerous drugs like doxorubicin (DOX) [10,11]. DOX, an anticancer drug is extensively used in the management of different tumors [12]. Antitumor activity of the drug is exerted by interaction with DNA replication [13]. High dose regimens of DOX are associated with sever cardiotoxicity and bone marrow suppression. Different strategies are being used to encapsulate the drug to minimize its side effects; however this decreased the chemotherapeutic effectiveness [14]. To address these issues, advanced synergistic therapies such as combination of chemotherapy and photothermal therapy have been applied to enhance the overall therapeutic efficacy. This includes gold nanorods capped magnetic core, silica nanorattle gold shells and DNA based platform loaded with GNRs and DOX [15][16][17].
In the current study, DOX conjugated with PSS coated GNRs are designed for photothermal therapy. We hypothesized that our designed multimodal system will ruin tumor cells after NIR exposure (808 nm) via hyperthermia induced apoptosis and necrosis.

DOX loaded GNRs nano-complex fabrication
The prepared GNRs suspension has a surplus of cytotoxic CTAB, which was removed by repetitive cycles of centrifugation and re-dispersion. A CTAB bilayer remained non-covalently conjugated onto the GNRs surface to maintain the stability of the final product. The longitudinal localized plasmon resonance (LSPR) and the transverse plasmon resonance (TSPR) of prepared GNRs was found to be 780 and 526 nm, respectively ( Figure 1a). TEM images display monodispersed rods with aspect ratio 4.2 (Figure 1b & 1c). The biocompatible GNRs were prepared by PSS coating on their surface. The LSPR peak of the PSS coated GNRs was slightly red shifted to 783 nm ( Figure 1c). The shift in LSPR peak after PSS coating is due to the side by side assembly of the PSS-GNRs.
Absorption spectra confirmed the successful loading of chemotherapeutic drug (DOX) on PSS coated GNRs (Figure 1d). The polyelectrolyte coating allowed GNRs to easily interact with the surrounding environment, consequently the wavelength of LSPR of GNRs perceptively responded to the refractive index change caused by molecular adsorption. The conjugation of DOX onto the PSS-GNRs surface resulted in red shift of the LSPR band, while the TSPR peak remained same. Increased local refractive index around GNRs due to adsorption of DOX might lead to stronger the Columbic restoring force and red shift of the LSPR peak [18]. The percentage yield of the DOX-PSS-GNRs was found to be 81.2% ±0.21. To confirm the surface chemistry modification of GNRs the zeta-potential of CTAB-GNRs, PSS-GNRs and DOX-PSS-GNRs were investigated. The zeta potential of unrefined GNRs was measured to be +60 ±0.2 mV which decreased to +42 ±0.1 mV after removal of excess CTAB (two rounds of centrifugation and re-dispersion). A negative zeta-potential of -40 ±0.3 mV was recorded after successful coating of PSS on GNRs surface and again positive (39.3 ±0.6 mV) of DOX-PSS-GNRs confirmed chemistry changes to the GNRs surface (Figure 1e). Each bar represents the mean value ±SEM of triplicates.

Drug loading efficiency (DLE)
The loading efficacy of DOX on the PSS-GNRs was measured systematically using a standard curve of absorption of DOX (at 490 nm) by changing the concentration of DOX against the fixed concentration of PSS-GNRs (40 µg/mL). Molar absorptivity was recorded to be 0.87. Drug loading capacity of PSS-GNRs was about 76% with drug loading content of 3.2 mg DOX/mL of GNRs.

Photothermal stability of PSS-GNRs
Optical characterization of PSS-GNRs showed the LSPR peak of GNRs strongly depends on their aspect ratio, therefore, the LSPR peak position is an excellent indicator for any shape changes of GNRs. Aqueous solution of PSS-GNRs after laser exposure for 2 min (power density = 1.5 W/cm 2 ) remain stable, the LSPR peak shifted approximately 4 nm ( Figure 2). Stability of PSS-GNRs after NIR laser exposure was good enough for photothermal therapy.

In vitro DOX release after NIR exposure
Drug release from PSS-GNRs can easily be controlled with NIR irradiation. The cumulative DOX release almost doubled after laser exposure (1.5 W/cm 2 ) compared to non-irradiated samples ( Figure 3). Enhanced drug release stimulated by laser (808 nm) may possibly because of the heat generated by the nanomaterial. DOX release from irradiated samples was significantly reduced after 5 hrs as extracellular tissues of tumors and intracellular lysosomes have acidic environment (pH 5-6), so DOX release experiment was conducted at 5.6 pH [19].

PSS-GNRs nano complex biocompatibility
Dose dependent biocompatibility and cytotoxicity efficiency of the nanocarriers was recorded invitro. The efficiency of the GNRs in mediating cytotoxicity against HepG2 (carcinogenic) and 3T3 (non-carcinogenic) cells was evaluated. Cells were treated for 12 h with PSS-GNR and analyzed using the MTT assay. Figure 4 showed that exposure of cells with PSS-GNR had no significant reduction in the cells viability compared to control cells as viability remained higher than 88%, at the concentration of 500 µg/mL for HepG2 cells and 1000 µg/mL for 3T3 cells ( Figure 4a). If nanoparticles interact with RBCs in the blood stream they can cause hemolysis.
Therefore hemolytic properties and interaction with RBCs are the main parameters for the biocompatibility of nanocarriers [19]. Analysis of hemoglobin released from RBCs after incu-

Cell Inhibition after NIR exposure of PSS-GNR-DOX complexes
Optically triggered drug release from PSS-GNR-DOX by NIR laser (808 nm) exposure at an output power density of 1.5 W/cm 2 on HepG2 cells was studied. DOX release from PSS-GNR-DOX was increased significantly (p<0.05) after 2 min of NIR irradiation ( Figure 5). HepG2 cells were treated with free DOX and DOX-PSS-GNRs, either irradiated with NIR laser or not exposed to NIR light. Dose-dependent cytotoxicity was observed in all study groups. About 84% b a of cells were killed by free DOX and 65% by DOX-PSS-GNRs at equivalent DOX concentration of 10 μg/mL ( Figure 5). This showed that, free DOX was more toxic as compared to DOX conjugated to a nano-carrier at the same drug concentration. Similar findings were reported by Zhang and coworkers [20]. The high cytotoxic effect might exhibited by free DOX is due to higher availability of drug to the cells after the cell uptake whereas; the decreased cytotoxicity with DOX-PSS-GNRs is because of delayed drug release inside cells [19]. PSSGNRs nanocomplex has promising potential as biocompatible nanocarriers for drug loading and delivery in cancer therapy. Previous report showed DOX loaded tiopronin coated gold nanoparticles (Au-TIOP-DOX) had a better effect in killing cancer cells than free DOX [21]. The cytotoxic efficiency of the DOX-loaded PAA-PEG-GNRs was found to be similar to free DOX and improved with an increase in their concentrations [10]. In a previous study GNR-DOX-cRGD, viability was significantly decreased down to 57%, whereas free DOX demonstrated the highest level of cytotoxicity (41% of control) in U87MG cells [22]. We found that DOX-PSS-GNRs complexes killed more cancer cells (93%) after NIR laser irradiation ( Figure 5). Higher cytotoxicity of complex is due to the enhanced release of drug upon NIR laser irradiation. The IC50 of complex (PSS-GNRs-DOX) was 7.99 ±0.0032 µg/mL and for PSS-GNRs-DOX with laser irradiation was 3.12 ±0.0906 µg/mL. The IC50 of free DOX and DOX with laser exposure was 3.999 ± 0.04211 and 4.41 ± 0.0037 µg/mL, respectively. Previously, Au-HNS-EGFR-DOX are reported to have significant antiproliferaitve activity against lung cancer cells when irradiated with NIR laser (125 mW/cm 2 , 25 s) in contrast to non-exposed cells [23]. Free DOX showed no significant change on viability, for both with and without laser. This indicates that increased cell death upon NIR laser irradiation might be attributed to the presence of gold nano-carrier.
Without laser treatment low drug release was observed from nano-complex, showing that drug 11 release was turned off without laser illumination. Laser (808 nm) triggered DOX release was recorded using the same laser treatment at different time intervals (2, 3, and 4 hrs) in which drug release was improved in time dependent fashion. Below 10% of DOX was released within 4 h from PSS-GNR-DOX without NIR exposure, using the same experimental conditions ( Figure 5).
Drug release from the nano-complex (PSS-GNR-DOX) might be easily turned "on" and "off" by NIR laser exposure. The NIR laser irradiation cause melting of PSS that would lead to decreased stability and enhanced drug diffusion coefficient. No drastic change was observed in temperature of the solution after NIR irradiation. Mild NIR irradiation might cause a rapid and localized raised in the temperature on the GNRs surface, which leads to melt the PSS coating and accelerate drug release. Our study suggests that co-therapy based on combined chemo and photothermal treatment using nano-carriers is a best choice for cancer management. Low dosage treatment minimize side effect and maximize therapeutic efficiency of drug [19]. 13

Gold nanorods synthesis
GNRs were synthesized through seed-mediated growth method with slight modification [24].
Gold seed particles were synthesized by adding 250 μL of 10 mM HAuCl4·3H2O to 10 mL

PSS coating of GNRs
A reported method by Venkatesan was used, with slight modification, for PSS coating on GNRs [25]. Prepared GNRs (2 mL, 40 µg/mL) were centrifuged at 12000 g for 10 min and the pellet was re-dispersed in 2 mL of deionized water. GNRs solution was added drop-wise to 2 mL of PSS (2 mg/mL in 8 mM NaCl). For maximum adsorption, the solution was kept on stirring at room temperature for 2 h. Excess polymer (supernatant fraction) was removed by centrifugation (12000g for 10 min). PSS-stabilized GNRs were re-suspended in 2 mL deionized water and stored at 4°C.

Doxorubicin loaded PSS-GNRs
PSS-GNRs (40 µg/mL, 2 mL) were added to an aqueous solution of DOX at a final concentration of 10 µg/mL and was stirred overnight in the dark at room temperature. Excess drug was removed by centrifugation at 12,000g for 10 min and pellet was re-dispersed in 2 mL deionized water. UV-Vis spectra of DOX loaded GNRs were scanned at a wavelength range of 14 400-1100 nm. Surface charge distribution of DOX loaded PSS-GNRs conjugate, at a different level, was determined by the zeta potential analyzer (Zetasizer Nano ZS90 DLS system Malvern Instruments Ltd., England).

Percentage yield
Nanoparticles were collected and weighed accurately. The percentage (%) yield was then calculated using formula given below [26] % yield = Mass of nanoparticles obtained Total weight of drug and polymer X100

Drug loading efficiency (DLE)
In order to calculate the drug loading efficiency, a known quantity of DOX was mixed with an

Photothermal stability of PSS-GNRs
The photothermal stability of PSS-GNRs was recorded by irradiating conjugate with NIR laser (power density= 1.5 W/cm 2 ) for 2 min. After laser treatment stability of the sample was analyzed by UV-Vis spectroscopy.

In vitro drug release by NIR exposure
Near infrared (NIR) triggered drug release from PSS-GNRs was recorded in 10 mM phosphate buffer saline (PBS) (pH 5.6 at 37°C). A continuous wave 808 nm NIR laser (Ti-Sapphire, Spectra Physics CA 95054, USA) was used. DOX-PSS-GNRs (40 µg/mL, 2 mL) were dispersed in 10 mL of PBS followed by NIR laser irradiated at an output power of 1.5 W/cm 2 for 2 min and 800 μL of the resulting solutions was taken out for analysis. Exposed media was centrifuged at 12,000g for 10 min. Amount of DOX released from PSS-GNRs in the supernatant was determined by fluorescence measurement(Biotek synergy H4 multi-mode plate reader).

In vitro cytoxicity assays
The in vitro cytotoxicity of PSS-GNRs was recorded using 3T3 and HepG2 cells. Cells were seeded in 96 well plates (4x10 3 cells per well). After 24 h of incubation, cells were exposed to different concentrations of PSS coated GNRs. Cells were allowed to incubate at 37°C for an additional 24 h. Viability was recorded by the MTT assay [27].

Hemolysis Assay
All human blood samples in this study were from healthy volunteers and used with Institutional Review Board (IRB) bioethics approval. The hemolysis assay was carried out according to the protocol from National Cancer Institute (NCI).
Whole blood (5 ml) from two healthy human donors was drawn directly into K2-EDTA-coated tubes to prevent coagulation. Blood collection was performed by a trained phlebotomist in order to minimize the risk to the donor. A written informed consent was obtained from each donor prior to the blood drawn.
In the 5 mL of blood 15 mL of sterilized phosphate buffer saline (PBS) was added and after slow agitation tubes were centrifuged at 500×g for 10 min. Supernatant containing plasma was aspirated and the buffy coat was washed thrice and diluted with normal saline to a 50% packed 16 cell volume (hematocrit) adjusted at pH 7.4 and stored at 4°C. Different concentrations of PSS-GNRs (100 μL each) were incubated with 100 μL of RBCs suspension at 37°C in CO2 incubator for 4 h. 0.2% Triton ×100 was used as positive control and PBS was taken as negative control [28]. After incubation, 50 μL of 2.5% glutaraldehyde was added to the sample in order to stop the process of haemolysis and centrifuged at 1000×g for 10 min. Hemoglobin release was monitored at 562 nm using a microplate reader (Platos R496, Austria) by transferring supernatant to 96 well plate. Percentage hemolysis was recorded using following formula Percent Hemolysis = Sample absorbance −negative control absorbance ×100 Positive control absorbance −negative control absorbance

Cell Inhibition after Photothermal Treatment
The HepG2 cells were seeded into 96-well plates (5×10 3 per well) and incubated for 24 h before the adding the different concentrations of PSS-GNRs, free DOX and PSS-GNRs-DOX conjugate. The treated cells were incubated for 12 h for proper uptake before laser irradiation.
After that cells were illuminated by 808 nm NIR laser (power density =1.5 W/cm 2 for 2 min) and incubated at 37 o C for 24 h. The MTT assay was performed to record cell inhibition.