Optimizing a Novel Au-Grafted Lipid Nanoparticle Through Chelation Chemistry for High Photothermal Biologic Activity

Abstract

Gold nanoparticles through nucleation of Au clusters have been extensively studied. However, due to low potency, prolonged tissue retention, and irreversible accumulation, the safety considerations have limited their therapeutic and diagnostic applications. Novel gold nanostructures with retained physical properties and higher biodegradability could be prepared by alternative approaches. Previously, a lipid nanoparticle (LNP) platform carrying gadolinium (Gd³⁺) has been reported to eliminate through the biliary without accumulation in the liver or kidney within 24 h. Inspired by this discovery, we investigated a new approach of forming gold nanoparticles using preformed LNPs grafting diethylenetriamine-pentaacetic acid as a chelating agent. Tiny Au nanoparticles are formed by simply mixing Au³⁺ with preformed diethylenetriamine-pentaacetic acid—LNP. The Au³⁺ associates stably to these LNPs after a systematic optimization. The Au-grafted LNPs are scalable and showed excellent photothermal effects when subjected to near-infrared light irradiation. They exhibit enhanced light-induced tumor cell killing at higher efficiency, compared with that of classical gold nanoparticles (citrated reduced). Given an additional small dose (2 Gy) of gamma irradiation, Au-grafted LNP could produce synergistic photothermal and radiotherapeutic effects under reduced light dose. The simple and adaptive nanoparticle design may enhance the margin of safety of gold nanoparticles in the treatment of cancers and other diseases.

Introduction

Gold in metallic state is inert, and colloidal gold is often used as a tracer for molecular and cellular studies because of its unique photophysical properties for tracing single-molecule localization and trafficking within the cells and tissues.^1,2 The nanoparticulate gold nanoparticles of about 10-30 nm were commonly synthesized using classical Turkevich method from gold (III) chloride interaction with sodium citrate via nucleation reactions. Addition of other reducing agents to gold nucleation reactions has been investigated in great detail with respect to size, charge, and stability and means to reduce aggregations in suspension.³ Some additional chemistry has been developed to modify the gold nanoparticle surface to allow gold particle surface modification via thio-linkages using polyethylene glycol (PEG) polymers with surface hydration property to reduce gold particle aggregation.^4,5 Gold-thiol chemistry also allows conjugation of gold nanoparticles to DNA aptamer, peptide, and carbohydrate for target-selective gold particle docking (binding).^6-9 Also, drug-gold particle conjugates have been reported with other innovative chemistry.^10,11

With available gold nanoparticles, a number of biomedical applications exploited the photothermal properties. Some reports demonstrated the potential of gold nanoparticles in heat-activated cell killing for use in photothermal therapy (PTT),¹² whereas others demonstrated gold particle—mediated DNA damage due to the enhancement of cell’s sensitivity to gamma or X-ray irradiation for use as an adjuvant to radiation therapy (RT).¹³ These reports based on cell studies have provided promising results.

However, in animal model studies, long resident time and slow elimination of gold nanoparticles in tissues become a major challenge.¹⁴ With a single intravenous IV dose of 20 nm citrate-gold nanoparticle (~0.7 mg Au/kg) in rats, the liver retained significant amount of gold particles at day 28 (3920 ± 592 ng/g; ~1/3 retention compared with 13,205 ± 1291 ng/g at 30 min).¹⁵ The high gold concentrations in the liver may be related to mild anemia and spleen atrophy detected at day 28. In addition, gold nanoparticles (d = 8-37 nm) (8 mg/kg/wk) injected intraperitoneally were reported to induce severe sickness in mice, and most of these mice died within 21 days.¹⁶ Histopathological analysis revealed that the accumulation of gold in mouse tissues caused an increase of Kupffer cells in the liver, loss of structural integrity in the lungs, and diffused white pulp in the spleen. Collectively, the data suggest that nonpreferential uptake and retention of gold nanoparticles have led to systemic toxicity. Inability of clearance of gold nanoparticles limits the therapeutic potential of otherwise attractive photo-thermal and radiation-adjuvant property of the gold nanoparticles.

Recently, we have reported a biocompatible and biodegradable lipid nanoparticle (LNP) expressing gadolinium (lanthanide) that exhibits limited liver uptake, and a majority of the particles are cleared within 24 h. These LNPs contained diethylenetriamine-pentaacetic acid (DTPA) with negative tetravalency that allows stable association of gadolinium with positive trivalency (Gd³⁺).¹⁷ When given to rats, a majority of these particles (d = 60~70 nm) were cleared from the blood through the bile and eliminated in feces within 24 h. If DTPA particles could stably chelate Au³⁺ and exhibit photodynamic response, it could overcome long resident time of gold nanoparticles that often prevents clinical translation because of toxicity concern. Binding onto these LNPs could promote clearance of Au from the body and minimize systemic toxicity.

In this report, we evaluated the interaction of Au and DTPA-conjugated LNPs and formation of stable Au-grafted LNPs (Au-LNPs) based on the Gd-LNP platform, which has been proven to be safe in preclinical studies. We found that Au is stably loaded on the DTPA chelate that is stably inserted into lipid excipients in the LNPs. The Au-LNPs are scalable and stable in suspension. Furthermore, Au-LNPs showed excellent photothermal effects and photothermal stability and exhibit enhanced light-induced tumor cell killing at much higher potency than that of classical citrate-coated gold nanoparticles.

Materials and Methods

Materials

1,2-Distearoyl-sn-glycero-3-phosphocholine (DSPC) and 1,2-distearoyl-sn-Glycero-3-phosphoethanolamine-N-DTPA (DSPE-DTPA) were purchased from Avanti Polar Lipids (Alabaster, AL). 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[poly (ethylene glycol) 2000] (mPEG-DSPE) was obtained from Genzyme Pharmaceuticals (Cambridge, MA). Gold (III) chloride trihydrate (HAuCl4•3H₂O) and sodium citrate were purchased from Sigma Chemical Co. (St. Louis, MO). The RPMI 1640 medium, fetal bovine serum (FBS), and trypsin were purchased from Invitrogen Corporation. Alamar Blue obtained from Sigma was used as a 10% solution unless otherwise stated. Other ingredients were of analytical grade or higher.

Preparation and Characterization of Au-LNPs

The Au-DTPA LNPs were prepared by mixing LNPs with HAuCl₄•3H₂O. DSPC, DTPA-DSPE, and mPEG-DSPE (18:5:2 molar ratio) were dissolved in chloroform/ethanol, dried into a thin film under N₂, and placed in vacuum overnight. The phosphate-buffered saline (PBS, pH 7.4) was added to the film and either sonicated or extruded through 50 nm polycarbonic filter at 60°C to obtain the LNPs. To prepare Au-bound LNPs, the LNPs in suspension were mixed with Au³⁺ (1:1 DTPA-PE:Au³⁺ mole ratio) for at least 6 h. The Au-LNP particle diameter was determined by laser light scattering using a Nicomp 380/ZLS zeta potential analyzer (Particle Sizing System) and expressed as mean ± SD. A transmission electron microscope (HT7700, Hitachi, Japan) image was detected after negative staining with uranyl acetate (UA, 1%, w/v). The stability of Au-LNPs was studied by checking the mean particle size of the samples at different conditions. Each experiment was independently performed at least 3 times. Citrate-stabilized gold nanoparticles were synthesized using the Turkevich reduction method.¹⁸

Photothermal Measurement for Au-LNPs

To study the photothermal properties of Au-LNPs, Au-LNP aqueous solutions with different concentrations were irradiated with 808-nm laser at different power density for different times. The temperature of the solution samples for photothermal conversion measurement was recorded by an IR thermal camera (E50, Arlington, VA).

Cell Culture

MDA-MB-231 cell line was obtained from the American Type Culture Collection (ATCC, Middlesex, UK). The MDA-MB-231 cells were grown in normal RPMI 1640 medium containing 10% FBS. The cells were maintained in a humidified cell incubator at 37° C with 95% humidity and 5% CO₂.

Photothermal Effects

Cell viability in terms of mitochondrial integrity was measured by the Alamar Blue assay. MDA-MB-231 cells were seeded onto 96-well plates at a density of 2 × 10³ cells per well and cultured for 24 h. Various concentrations of Au-LNPs were added into each well. After being cultured for 4 h, the irradiation groups were exposed to 3.0 W/cm² light irradiation (heat lamp) for 2 min followed by cultivation for another 5 days. The cells were then washed by PBS twice and then treated with Alamar Blue solution. The plates were again incubated for 4 h at 37°C. Fluorescence readings of the wells were recorded by a PerkinElmer 1420 VICTOR 3V multi-well plate reader (PerkinElmer, Waltham, MA) with excitation at 530 nm and emission at 580 nm.

Synergistic Photothermal Therapy and Radiotherapy

A total of 2 × 10³ MDA-MB-231 cells were seeded in a 96-well plate and cultured overnight. Various concentrations of Au-LNPs were added into each well. After 4 h of incubation, the irradiation groups were exposed to either 3.0 W/cm² light irradiation for 1 min (light bulb), or 2 Gy irradiation dose (Cobalt-60 Gamma-Ray Irradiator), or both, followed by cultivation for another 7 days. The cell viability was then measured by Alamar Blue assay.

Statistical Analysis

Data are presented as the mean ± SD. Statistical analysis was performed based on unpaired Student's t-tests (2-sided) or one-way ANOVA with the SigmaPlot software (Systat, San Jose, CA).

Results

Interactions of Au and DTPA Expressed in Lipid Nanoparticles

We first evaluate the role of reaction order and DTPA/Au ratio on nanoparticle formation. Three different orders used are schematically presented in Scheme 1. The first method (A) is mixing Au³⁺ ions with DTPA-DSPE (molar ratio 1:1) to allow chelation in the solution. The transparent solution of DTPA-DSPE turned to gray suspension after the addition of Au³⁺ ions. Then the Au/DTPA-DSPE suspension was mixed with DSPC/DSPE-PEG lipid suspension to form Au LNPs. This product, designated as Au-LNPs1, exhibited gray-to-whitish color in suspension. However, after overnight storage, black precipitates appeared in solution of Au-LNPs1, suggesting dissociation of gold or gold aggregates from Au-LNPs1. A second method (B) directly adds Au³⁺ ions in solution to the DSPC/DSPE-PEG/DTPA-DSPE lipid solution (molar ratio of DTPA-DSPE/Au³⁺ ions 1:1). Then, the mixture was subjected to size-reduction by sonication to form Au-LNPs (Au-LNPs2). The resulting purple-colored Au-LNPs2 exhibited a mean diameter of ~30 nm. As the color of gold nanoparticles, prepared by the nucleation method, corresponds to the size of gold particles,¹⁹ it is possible that the purple color might be a result from the gold aggregation. The Au-LNPs2 was also found to be unstable, and precipitants were found at the bottom of the purple suspension. Thus, a third method (C) was deployed. Au³⁺ ions in the solution were added to preformed LNPs containing DTPA-DSPE (with DSPC and mPEG-DSPE); (the preformed LNPs were reported previously¹⁷). After 1.5 h incubation of this mixture in suspension at 25°C, the LNPs (Au-LNPs3) turned from yellow to green. Compared with the gold nanoparticles prepared from the method A or B, Au-LNPs3 show good stability and distinct color of the solution. In addition, the final method C is simple and could be easily scalable. Therefore, we chose Au-LNPs3 for further optimization studies. A schematic representation of the formation of Au-LNPs3 is shown in Scheme 2.

Scheme 1.

Preparation orders and appearance of differently formed Au-LNP. Au³⁺ ions in the solution were reacted with DTPA-DSPE before addition into the DSPC/DSPE-PEG lipid solution (method a), directly added into the DSPC/DSPE-PEG/DTPA-DSPE lipid in suspension (method b), or added into the preformed lipid nanoparticles made from DSPC/DSPE-PEG/DTPA-DSPE lipids (method c). The final products were monitored for color and suspension behavior at 25°C for 24 h. The molar ratio of DSPC/DSPE-PEG2000/DTPA-DSPE in all preparations was 18:2:5. Au-LNPs1 and Au-LNPs2 readily aggregate. Au-LNPs3 demonstrates good suspension and chromogenic stability.

Scheme 2.

The schematic representation of the formation of Au-LNPs3.

Optimizing Composition and Methods to Prepare Gold-Grafted Lipid Nanoparticles Au-LNPs3

In initial studies, the stability, color appearance, and particle size of Au-LNPs3 were closely related to the DSPE-DTPA-to-Au³⁺ molar ratio and lipid concentration in preformed DTPA-LNPs. With the total lipid concentration maintained at 7.75 mM, the color of solution, the maximum absorption wavelength, and particle size distribution of Au-LNPs3 are closely related to the DSPE-DTPA/Au³⁺ molar ratios (Fig. 1). The color changed from transparent to pink when decreasing the DSPE-DTPA/Au³⁺ molar ratio from 10 to 5 (Fig. 1a). When the DSPE-DTPA/Au³⁺ molar ratio decreased to 1, the color of the aqueous solution became green. When the DSPE-DTPA/Au³⁺ molar ratio reached 0.2, the preformed LNPs were unable to provide adequate stabilization capacity for gold nanoparticles and solid gold could be found. The maximum absorption wavelengths of the Au-LNPs3 increased from about 530 nm to 650 nm when the DSPE-DTPA/Au³⁺ molar ratio decreased from 10 to 0.5 (Fig. 1b).

Figure 1.

Effects of the DSPE-DTPA/Au³⁺ molar ratio on the properties of Au-LNPs3. The color (a), the maximum absorption wavelength (b), and the particle size distribution (c) of Au-LNPs3 changed as the DSPE-DTPA/Au³⁺ molar ratio changed.

The particle size distribution of Au-LNPs3 varied with increasing DSPE-DTPA/Au³⁺ molar ratio. Using the dynamic light scattering method, 2 populations of Au-LNPs3 were found, similar to that of blank LNPs, albeit, at much smaller size distribution. Two-thirds of the particle size exhibited at 16.8 nm, and one-third of the particle size was about 31.7 nm. The Au³⁺ solution was also tested for particle size detection, but no readout could be obtained. Compared to the size distribution spectra of blank LNPs, Au-LNPs3 showed larger but different particle size and size distribution (Fig. 1c), suggesting the apparent size increase is due to Au associated to and grafted on the LNPs. With an increase of the DSPE-DTPA/Au³⁺ molar ratio, the percentage of the size in the range of 11.0-11.4 nm decreased. When the DSPE-DTPA/Au³⁺ molar ratio was higher than 2, particles size in the range of 100-250 nm was found, indicating the aggregation of the Au-LNPs3 due to the strong reduction of Au³⁺ by excessive lipid. Based on the data collected with varying DSPE-DTPA/Au³⁺, the optimal DSPE-DTPA/Au³⁺ ratio range was within 0.5~1.5.

We next determined the effects of initial lipid concentration on the formation of Au-LNPs3. We found that the particle size distribution vary with the total lipid concentrations during the Au addition to form Au-LNPs3. As shown in Figure 2 at a fixed DSPE-DTPA/Au³⁺ molar ratios of 0.5, 1, or 5, the proportion of particles with larger size appeared to increase with lower lipid concentration. Thus, to obtain smaller particle size, total lipid concentration should be 1 mM or higher at 25°C to avoid the formation of large aggregates.

Figure 2.

Effects of the initial lipid concentration on the particle size distribution of Au-LNPs3. The particle size distribution varied with the total lipid concentrations when the DSPE-DTPA/Au³⁺ molar ratios were fixed at 0.5 (a), 1 (b), and 5 (c).

Through observation of the color of mixed solution of LNPs and Au³⁺, we found that the color turned from yellow to blue at about 1 h incubation time at 25°C room temperature, and the color turned from blue to green in about 1.5 h. The color continued to darken, suggesting larger aggregate formation up to about 6 h. At DTPA-DSPE/Au³⁺ molar ratio of 1 and lipid concentration of 7.75 mM, we found time-dependent effects on the particle size distribution of Au-LNPs3 at 25°C (Fig. 3). It was reported that the formation of gold nanoparticles by the reduction of Au(III) complex ions is a very complex process, and the reaction mechanism constitutes several steps.²⁰ We speculate that the formation of the Au-LNPs3 is a multistage dynamic process. Thus, the particle size distribution of Au-LNPs3 did not seem to follow a particular trend over the reaction time. To provide sufficient time to complete Au³⁺ and LNP interactions, each mixture at 25°C was allowed to react overnight.

Figure 3.

The effects of the reaction time on particle size distribution of Au-LNPs3. Au-LNPs3 were prepared with DSPE-DTPA/Au³⁺ molar ratio of 1 and lipid concentration of 7.75 mM.

We further investigated the effect of the temperature on the characteristic of Au-LNPs3. Increasing the temperature from 25°C to 45°C appeared to accelerate the formation of particles as indicated by color change. However, the color of the suspension turned from yellow to purple, which suggests Au dissociation from the nanoparticles or aggregation of the gold nanoparticles. When we kept the mixture of LNPs and Au³⁺ in the 4°C, the reaction was much slower than higher temperature. Only slight green (an indicator of Au-LNPs3 formation) was detectable only after 2-week incubation. Based on these data, we found that DTPA-DSPE-to-Au³⁺ optimal ratio is 1:1, and total lipid concentration is 7.75 at 25°C for 12-18 h to provide Au-LNPs3 preparation in a reproducible manner. Au-LNPs3 showed superior stability. There are no obvious changes in the particle size of Au-LNPs3 even in the presence of 10% FBS for 48 h (Fig. 4). Therefore, this optimal condition is used for further studies.

Figure 4.

The stability of Au-LNPs3 in RPMI 1640 + 10% FBS medium measured by the changes of the particle sizes. Au-LNPs3 were prepared with DSPE-DTPA/Au³⁺ molar ratio of 1 and lipid concentration of 7.75 mM. The particle size was shown as Z-average hydrodynamic diameter.

Characterization of Au-LNPs3 Morphology and Photothermal Properties

To evaluate the morphology of Au-LNPs3 and compare to typical citrate stabilized gold nanoparticles, we analyzed these particles with electron microscopy (Fig. 5). To improve contrast resolution of lipids in electron micrograph, we stained LNPs with 1% UA. As shown in Figure 5, we found that LNPs exhibit an elongated and rice-like shape with dimensions of ~25 × 15 nm (Fig. 5a). Blank LNPs without staining (Fig. 5b) did not show any discernible particle. Au-LNPs3 stained with 1% UA showed size and shape similar to those of LNPs (Fig. 5c), indicating that the chelation of Au did not change the morphologies of LNPs. Small black dots could be clearly seen in the image of Au-LNPs3 without staining (Fig. 5d), which could be attributed to the formation of the ultra-small gold nanoparticles bound on the LNPs. The morphologies of the Au-LNPs3 at different DTPA-DSPE/Au³⁺ molar ratios were also investigated. Because of low Au concentration, it is difficult to find gold dots when the DTPA-DSPE/Au³⁺ molar ratio was higher than 1. Interestingly, Au-LNPs3 contained snowflake-shaped gold nanostructure when DTPA-DSPE/Au³⁺ molar ratio was 0.5 (Fig. 5e). This could be due to the insufficiency of the DTPA-DSPE in LNPs to stabilize ultra-small gold nanoparticles. The shape of the Au-LNPs3 at DTPA-DSPE/Au³⁺ molar ratio 1 was entirely different from that of citrate-coated gold nanoparticles (Fig. 5f).

Figure 5.

TEM images of the nanoparticles. (a) Blank LNPs stained with 1% UA. (b) Blank LNPs without staining. (c) Au-LNPs3 with DTPA-PE/Au³⁺ molar ratio of 1 stained with 1% UA. (d) Au-LNPs3 with DTPA-PE/Au³⁺ molar ratio of 1 without staining. (e) Au-LNPs3 with DTPA-PE/Au³⁺ molar ratio of 0.5 stained with 1% UA. (f) Citrate-coated gold nanoparticles without staining.

To evaluate the photothermal properties of Au-LNPs3, the temperature changes of each solution were recorded at various Au concentrations (0, 0.15, 0.5, and 1.5 mM) under near infrared (NIR) laser irradiation (808 nm, 0.3~0.9 W/cm²). As illustrated in Figure 6a, it was observed that with increase of Au-LNPs3 concentration, the solution temperatures increases greatly. After 4 min laser irradiation (0.9 W), the solution temperature of Au-LNPs3 (1.5 mM) shows an increase of around 25°C, suggesting good photo-thermal effect of Au-LNPs3. In contrast, PBS showed only a slight temperature increase of 4°C. Moreover, the temperature changes under different power intensity (0.3, 0.5, 0.7, and 0.9 W/cm²) were measured, as well at a concentration of 1.5 mM (Fig. 6b). It is evident that the increase of NIR light intensity led to faster increase of Au-LNPs3 suspension’s temperature.

Figure 6.

Photothermal heating curves for Au-LNPs3 suspension under NIR laser irradiation (808 nm, 0.9 W/cm²) at different concentrations (a) or under NIR laser irradiation (808 nm, 1.5 mM) at different laser power densities (b). (c) Photothermal stability of Au-LNPs3 aqueous dispersion under irradiation for alternate 5 ON/OFF cycles (808 nm, 0.9 W/cm²). (d) Comparison of the photothermal heating curves for Au-LNPs3 and the citrate-stabilized gold nanoparticle aqueous solution under NIR laser irradiation (808 nm, 0.9 W/cm², 1.5 mM).

We further evaluated the photothermal stability of Au-LNPs3, under NIR laser irradiation (808 nm, 0.9 W/cm²), for 5 alternate ON/OFF cycles (Fig. 6c). The temperature changes of Au-LNPs3 (1.5 mM) were consistent in each cycle, indicating the excellent photothermal stability of Au-LNPs3. We also evaluated the photo-thermal effect of citrate-stabilized gold nanoparticles. Au-LNPs3 demonstrated excellent photothermal effects compared with citrate-stabilized gold nanoparticles (Fig. 6d).

Effects of Au-LNPs3 Mediated Breast Cancer Cell Sensitivity to Infrared and Gamma Irradiation

The photothermal effects of Au-LNPs3 were investigated in MDA-MB-231 human breast cancer cells. Cell viabilities were detected with or without light irradiation (3.0 W/cm², 2 min) after 5 days of treatment. When Au concentrations were initially fixed at 16 μM, both Au-LNPs3 and citrate-stabilized gold nanoparticles showed negligible toxicity toward MDA-MB-231 cells (Fig. 7). We then performed a dose-dependent study to explore the ability of Au-LNPs3 to enhance tumor cell killing under light exposure. With light irradiation, the citrate-gold—treated group showed no cell growth inhibition with over 90% cell viability. However, Au-LNPs3 with 2 min light irradiation could significantly inhibit cancer cell growth, and the cell viability was lower than 10% with an equivalent 15.6 μM Au concentration.

Figure 7.

Photothermal effects of Au-LNPs3 in MDA-MB-231 cells. Cells were treated with different concentrations of Au-LNPs3 or citrate-stabilized gold nanoparticles. After being cultured for 4 h, the irradiation groups were exposed to 3.0 W/cm² light irradiation for 2 min, followed by cultivation for another 5 d.

The synergistic photothermal and radiotherapeutic effects of Au-LNPs3 were then investigated. Cells treated with Au-LNPs3 were exposed to a very low dose of irradiation including 3.0 W/cm² light irradiation for 1 min and 2 Gy gamma irradiation, followed by cultivation for another 7 days for observation of the high-energy, irradiation-induced cell death. Compared with cells treated with Au-LNPs3 for 5 days, prolonging the incubation time to 7 days could inhibit certain cell proliferation to about 80%, suggesting the potential long-term toxicity of gold nanoparticles (Fig. 8). Cells after treatment with Au-LNPs3 exposed to either 3.0 W/cm² light irradiation for 1 min or 2 Gy gamma irradiation showed no obvious cell proliferation inhibition effects. The combination of light and gamma irradiation could significantly increase the anticancer effects of Au-LNPs3, indicating the synergistic photothermal and radiotherapeutic effects of Au-LNPs3.

Figure 8.

Synergistic photothermal and radiotherapeutic effects of Au-LNPs3 in MDA-MB-231 cells. Cells were treated with different concentrations of Au-LNPs3. After being cultured for 4 h, the irradiation groups were exposed to either 3.0 W/cm² light irradiation for 1 min, or 2 Gy gamma irradiation, or both, followed by cultivation for another 7 d.

Discussion

Gold nanoparticles have been widely studied for biomedical application because of their intrinsic optical, electronic, and physico-chemical properties.²¹ Gold nanoparticles have the ability to absorb light in the visible or NIR region and generate heat.²² They also have the ability to absorb X-rays and effectively enhance the radiation cytotoxicity.²³ For cancer treatment, gold nanoparticles have been extensively studied for PTT to cause direct antitumor effects via photothermal ablation or for a combination of hyperthermia and RT by inducing mild hyperthermia that sensitizes cells to RT. The optical and electrical properties of gold nanoparticles are closely related with size, thickness, and shape. Au nanorods (~50 nm) and nanoshells (~130 nm) with longitudinal plasmon bands (LSPR) in the NIR region (>800 nm) demonstrate good photo-thermal and radiation therapeutic effects.²⁴ The size of spherical solid Au nanoparticles larger than 50 nm is more suitable for PTT because of strong NIR absorption.²⁵ However, particles with larger size are likely to accumulate in the liver and lung and prove difficult for elimination from the body through excretion, which may lead to long-term toxicity and immunogenic response.^14,26 Mice exposed to commercially available 15-nm gold nanoparticles developed granulomas in the liver, which transiently increased the proinflammatory cytokine interleukin-18 in serum level.²⁷ Currently, due to the safety concern, no gold nanoparticle products are available for clinical application.^2,28 The high degree of tissue retention of gold nanoparticles, regardless of the size and shape, and long-term safety on human health and reproduction concerns have prevented the otherwise attractive physical and biological properties of gold nanoparticles to enhance tumoricidal activities in vivo.²⁹

To reduce tissue retention of gold particles, a number of reports have attempted to reduce self-aggregation and surface modification to enhance particle clearance from the tissues while retaining their photodynamic properties for application in photothermal therapy. In one report, 15-nm spherical Au nanoparticles are mixed with PEG shell to form stable particles at physiological pH; these particles were able to absorb light in the visible region. Under acidic condition, the Au nanoparticles containing lipoid acids caused aggregation-dependent red shift into absorption optimum peak at the NIR region.³⁰ However, the irreversible aggregation might cause severe toxicity reported with these particles. Other 4-5 nm spherical Au nanoparticles with pH- or thermal-responsive reversible aggregation properties were also developed.^31,32 But their reversible aggregation properties under the physiological conditions were not studied. In addition, small Au nanoparticles (<7 nm) may undergo rapid renal clearance; however, these smaller particles are unable to accumulate in the tumor, thus limiting their potential for use as photothermal cancer therapy.

Another approach to overcoming this limitation is to generate hybrid systems by encapsulating small Au nanoparticles in larger nanoparticle platforms such as liposomes³³ and silica nanoparticles.³⁴ These larger nanoparticle platforms can also provide loading space for other imaging or therapeutic agents which could provide an opportunity for multimodal treatment. However, the difficulty of large-scale preparation, and their storage stability, may impede the clinical translation of these hybrid systems. In our laboratory, LNPs based on similar compositions were produced at a large greater than 1 L scale, and they have been proved to be safe in multiple preclinical species including in monkeys.^35,36 Au-LNPs developed similarly could be prepared through a very simple process with easy scalability (Scheme 1). Within the Au-LNP hybrid system, the ultra-small Au nanoparticles could bind on the LNPs without changing the morphologies of LNPs (Fig. 5). The Au-LNPs3 showed excellent photothermal effects and photothermal stability (Figs. 6a-6c). The photothermal effect of Au-LNPs was proved in cancer cells (Fig. 7). Therefore, Au-LNPs3 might overcome the size dilemma of gold nanoparticles and show great potential for further development for clinical use.

Several techniques such as microwave, radiofrequency, laser, or ultrasound are usually implemented in the clinic for delivering heat to generate hyperthermia. However, the difficulty of heating deep tumors impeded the clinical implementation of hyperthermia. The optimal wavelength for best tissue penetration is in the NIR region or further. Spherical Au nanoparticles absorb UV and visible light but do not absorb NIR light significantly. So, it is generally a poor choice to use small spherical Au nanoparticles for PTT. Gold nanoshells and nanorods with large particle sizes were found to have strong absorbance at ~800 nm and were used to treat subcutaneous tumors in mice.^22,37 However, gold nanoshells and nanorods with a particle size larger than 50 nm showed poor tumor penetration. Consistent with previous reports, the gold nanoparticles (produced by the citrate nucleation method) showed poor photothermal effects under 808-nm laser irradiation (Fig. 6d). By contrast, Au-LNPs3 showed excellent photothermal effects and photothermal stability under 808-nm laser irradiation (Figs. 6a-6c). The results suggested that Au-LNPs3 integrate the advantages of LNPs and gold, which allow for a good tumor penetration/accumulation (mediated by LNPs) and absorb NIR light to generate hyperthermia for tumor-killing effects.

Several studies have demonstrated that the sensitivity of radio (or gamma irradiation) therapy (RT) could be enhanced by hyperthermia. Gold nanoparticles with unique optical properties have been studied extensively in combined PTT and RT for synergistic effects. It was reported that gold nanoparticles have the ability to reduce the dose of X-rays and increase sensitivity to radiation in a very radioresistant subcutaneous squamous cell carcinoma (SCCVII) in mice during radiotherapy.³⁸ Au-LNPs3 exposed to a very low dose of irradiation, including 3.0 W/cm² light irradiation for 1 min and 2 Gy gamma irradiation, could significantly increase the anticancer effects (Fig. 8). The results suggest the potential of Au-LNPs3 to be used in combined PTT and RT. Although the exact mechanism of synergy is not clear, it is possible that the mild hyperthermia may reduce the radioresistant hypoxic cell death, leading to enhanced cell toxicity to gamma irradiation.³⁹

Chelation therapy based on a metal chelator such as disodium ethylenediaminetetraacetic acid (EDTA) and an oral iron chelator FBS0701 has been used to treat heavy metal ion poisoning, neurodegeneration, cancer,⁴⁰ and cardiovascular diseases.⁴¹ Some natural and synthetic antioxidants which possess metal-chelating properties have been used in disease prevention and clinical recovery against heavy metal intoxication.⁴² Based on the proven property of LNPs containing DTPA to chelate Gd³⁺ and the DTPA-LNPs ability to clear from the blood and liver within 24 h,¹⁷ we used LNPs in this study to form an optimal Au³⁺ chelated to LNPs (referred to as Au-LNPs3 particles). Although the ionic form of gold (Au³⁺) without chelation is toxic, the nanoparticulate form of gold has been shown to reduce nonspecific toxicity. The optimized formulation Au-LNPs3 can provide a stable, reproducible, and potentially scalable product. In addition, the approach of producing LNPs with DTPA expressed on the surface may be useful to chelate gold (III) and other metals to detoxify metal toxicity in humans.

Conclusion and Future Direction

Au/LNPs were synthesized by simply mixing Au³⁺ with preformed LNPs with DTPA. As the LNPs loaded with Gd³⁺ have been reported to clear from the liver and blood within 1 day, we expect that the Au-loaded LNPs or Au-LNPs3 would likely be cleared in similar manner, thus overcoming high degree of Au nanoparticle retention (for months to years). The properties of Au-LNPs3 were closely related with the DSPE-DTPA/Au³⁺ molar ratio and the lipid concentration during reaction. Ultra-small gold nanoparticles bound on the LNPs or Au-LNPs3 (~25 × 15 nm) could be prepared reproducibly with DTPA-DSPE/Au³⁺ molar ratio of 1 and lipid concentration of 7.75 mM at 25° °C for 18 h. Au-LNPs3 demonstrate excellent stability and are highly scalable. Au-LNPs3 exhibit enhanced light irradiation—induced killing of tumor cells compared with that of classical gold nanoparticles (prepared with citrate) and show synergistic photothermal and radiotherapeutic effects. The simple and adaptive nanoparticle design as well as increased safety margin of these gold nanoparticles may have great potential for the treatment of cancers and other diseases.