Preparation, Characterization and Optimization of Etoposide-Loaded Gold Nanoparticles Based on Chemical Reduction Method

In recent years, observed focus greatly on gold nanoparticles synthesis due to its unique properties and tremendous applicability. In most of these researches, the citrate reduction method has been adopted. The aim of this study was to prepare and optimize monodisperse ultrafine particles by addition of reducing agent to gold salt, as a result of seed mediated growth mechanism. In this research, gold nanoparticles suspension (G) was prepared by traditional standard Turkevich method and optimized by studying different variables such as reactants concentrations, preparation temperature and stirring rate on controlling size and uniformity of nanoparticles through preparing twenty formulas (G1-G20). Subsequently, the selected formula that prepared from the best tested condition was further optimized by preparing it using inverse method via the addition of gold salt to the reducing agent in opposite to the previous traditional method (G21). The optimized gold nanoparticles were characterized by SEM, EDX, TEM and zeta potential. The obtained results indicated that (G21) with reactants concentrations of 0.5mM and 10mM for HAuCl4.3H2O and trisodium citrate dihydrate respectively, 65°C of preparation temperature and 1500rpm of stirring rate was chosen as an optimized formula according to AFM provided gold nanoparticles with smoother surface, smaller size (average 8.75nm) with more uniform size distribution (7.32%) as well as short over all preparation time (27minutes). In addition to that all results of SEM, EDX and TEM indicated uniform spherical shape with zeta potential of -47.87. Gelatin was used to enhance the colloidal stability. Three formulas of etoposide-loaded gold nanoparticles were prepared using different initial loading of etoposide. Formula (F1) was prepared from 5mg/ml etoposide solution which gave higher loading efficiency (91.73%) and in-vitro released with controlled manner (better retardation at physiological pH ~ 7.4 and greater release at lysosomal pH ~ 4.6) as well as AFM results provided gold nanoparticles with smoother surface, average particles size (22.74nm) with uniform size distribution (20.73%). In conclusion, inversed method is promising for the preparation of gold nanoparticles with high monodispersity.


Introduction
The most emerging branch in pharmaceutical sciences known as "pharmaceutical nanotechnology" presents new tools, opportunities and scope, which are expected to have significant applications in disease diagnostics and therapeutics. Various researches are currently working on nanoparticles especially in metal nanoparticles and further more specifically in gold nanoparticles (1) . Recently metal nanoparticles are universally considered as promising multifunctional platforms for wide purposes particularly in the case of cancer and gene diseases. Since, gold nanoparticles reveal enormous potential in drug delivery as carrier for spatial and temporal delivery of bioactive and diagnostic (2) .
Gold nanoparticles has gained increasing importance in pharmaceutical nanotechnology owing to their special features, such as tunable their size and shape, strong absorption and scattering properties, easy surface functionalization, simple set up for synthesizing and high biological stability and compatibility (3) . Thus, a lot of volume of researches focuses on synthesis techniques of gold nanoparticles, including chemical, thermal, electrochemical, sonochemical and photochemical reduction pathways. Among all these methods, the citrate reduction; pioneered by John Turkevich in 1951, is one of the most broadly used techniques (4; 5) .
Until today, the protocol of original synthesis was numerous times modified allowing synthesis of nanoparticles in wide sizes range. The monodispersed gold nanoparticles with desired size and morphology can be obtained depending on the seed mediated growth mechanism. In this classical technique, a trisodium citrate dihydrate solution was quickly added to a preheated solution of hydrogen tetrachloroaurate trihydrate. Since, citrate acts as poor reducing and stabilizing agent. In contrast, the inversed method designed by sirvaraman et al. comprises change the order of reactants addition leads to formation of stronger reducing and stabilizing agent (dicarboxy acetone) by induced thermal oxidation of trisodium citrate dihydrate (6; 7) .
The aim of the present study was to develop gold nanoparticles with high size and shape uniformity in order to improve their ability as drug carrier.

Methods Preparation of gold nanoparticles by standard turkevich method
Gold nanoparticles suspension (G) was prepared by standard chemical reduction method; proposed by Turkevich. The gold (HAuCl4.3H2O) solution used as precursor while the citrate (Na3C6H5O7.2H2O) solution used as a mild reducing, pH mediating, stabilizing and capping agent, as demonstrated in Table 1.
A volume of 1ml of gold solution (50mM) was diluted with 198ml of preheated deionized water (DW) in a tightly closed conical flask on hot plate magnetic stirrer. The solution was gently stirred manually for 15minuts using glass rod stirrer at 75°C. Then after, 1ml of citrate solution (500Mm) was quickly transferred (the final concentration was 2.5mM) with continuous stirring at a constant temperature until a characteristic red color was observed, indicating for the formation of gold nanoparticles. Subsequently, the heating was switched off and the suspension was allowed to cool at room temperature with continuous stirring, then filtered using 0.1μm syringe-filter to remove any possible aggregates or dust impurities and stored in refrigerator at 2-8°C in a tightly closed dark container coated by aluminum foil, in order to quench any further reaction might be occurred due to light or temperature effects, prior to further characterization (5)(6)(7)(8) .

Characterization of gold nanoparticles (G) qualitative dispersibility measurements
The simplest inexpensive approaches that confirm the presence of spherical gold nanoparticles are:

Visual appearance
The sample was visually inspected for color, clarity, transparency, glossiness, and homogenecity.

Laser pointer test
The test was done by shining a laser pointer light into the sample, which put in a glass test tube.

Salt test
NaCl solution 10%w/v was added drop by drop to a test tube containing 3ml of sample. The electrolytes change the optical properties of gold nanoparticles significantly (9) .

UV-Visible spectrum analysis
UV-Visible analysis is a specialized technique confirms presence of gold nanoparticles. The shape of absorption spectrum varies in different shapes and sizes of gold nanoparticles, usually gold nanospheres give a single absorption peak between 515-550nm depending on their particles diameter. Gold nanoparticles with shape rather than spherical gave another peak greater than 600nm. The relative uniformity of nanoparticles can be gauged by the full width of half maximum (FWHM) (9; 10) .

Atomic Force Microscopy Measurement (AFM)
The AFM analysis allows for three dimensions (3D) characterization of gold nanoparticles with sub-nanometer resolution provides both qualitative and quantitative information about morphology, size, and size distribution. The sample was prepared by dropletevaporation method; 100μl of suspension was diluted with 900μl of DW and sonicated for 15minutes, then filtered using 0.1μm syringe-filter. A droplet of sample was deposited on silica glass slide and leave to dry overnight at room temperature in a light and dust protected environment (10; 11) .

Optimization of gold nanoparticles
Different preparation parameters (Table 1) were studied to produce high quality gold nanoparticles with desired morphology, size, and size distribution, based on AFM analysis.

Effect of HAuCl4.3H2O concentration
Five suspensions (G1-5) were prepared by classical citrate reduction method, using different final molar concentration of HAuCl4.3H2O (0.0625, 0.125, 0.5, 1, and 2mM, respectively), at a fixed citrate concentration, and compared to standard suspension (G) with 0.25Mm HAuCl4.3H2O to choose the best one for further studies.

Effect of preparation temperature
Five suspensions (G11-15) were prepared, under different temperature (room temperature, 45°C, 55°C, 65°C, and 85°C respectively), and compared with selected suspension prepared at 75°C to choose the best temperature for further studies.

Effect of stirring rate
Five suspensions (G16-19) were prepared, under different stirring rate (500, 750, 1000, 1250, and 1500 rpm respectively), and compared with selected suspension prepared under gentile stirring by glass rod stirrer to choose the best stirring rate for further studies.

Effect of reactants addition order
G21 suspension with the desired final molar concentration of HAuCl4.3H2O (0.5mM) and Na3C6H5O7.2H2O (10mM), under the selected preparation temperature (65°C) and stirring rate (1500 rpm), were prepared by inversed the classical method, corresponds to addition of small volume (2ml) of highly concentrated (50mM) gold solution to large volume (198ml) of diluted citrate solution in order to study the effect of change the order of reactants addition on the morphology, size, and distribution to choose the optimized gold nanoparticles suspension (7; 12) .

Characterization of the optimized gold nanoparticles for colloidal stability, elemental composition, surface morphology and particles, shape, size and distribution Zeta potential measurement
Zeta potential has important indicant about dispersion stability in water. Typically, gold nanoparticles with zeta potential greater than +20mV or less than -20mV have sufficient electrostatic repulsion between neighboring correspondingly charged particles to remain stable in their suspension. The sample was prepared by diluted 100μl of suspension with 10ml of DW then filtered using 0.1μl syringe-filter (13) .

Scanning Electron Microscopy (SEM)
SEM technique gives detailed high resolution 3D image, has high accuracy and reliability for determining the size, size distribution and morphology of gold nanoparticles. The sample was prepared by drop-casting method; One drop of suspension was deposited on silica glass slide and leave to dry at room temperature in a light and dust protected environment, then another drop of sample deposited and continue until dense dry layer was formed, then scanned at different magnification (14; 15) .

Energy Dispersive X-Ray Spectroscopy (EDX)
EDX provides elemental identification and quantitative compositional information for gold nanoparticles. The sample was prepared by dropcasting method, as mentioned previously (16) .

Transmission Electron Microscopy (TEM)
TEM provides a much magnified high resolution two direction (2D) image, which interprets the exact shape and size of gold nanoparticles. One drop of sample was placed on a 400 mesh carbon coated copper grid by an ultramicrometry process and scanned at different magnification (17; 18) .

Functionalization of gold nanoparticles
The optimized gold nanoparticles were functionalized by bio-conjugation method, using gelatin as a biocompatible polymer. Firstly, gelatin solution was prepared by mixing 300mg of polymer with 10ml of deionized water for 3hours using magnetic stirrer at 1500rpm. Then after, the polymeric-gold nanoparticles suspension was prepared by mixing 1ml of gelatin solution with 10ml of the optimized gold nanoparticles suspension, in a tightly closed conical flask, for 1hour at room temperature and 1500rpm; and sonicated for 6hours. Then filtered using 0.1μm syringe-filter to remove the unconjugated polymer and any possible aggregates, and stored in refrigerator at 2-8°C in a tightly closed dark container coated by aluminum foil (10,19) .
AFM measurement was used to infer the binding of polymer layer on the surface of gold nanoparticles and conclude about the stability of bio-conjugate as well as to evaluate the effect of polymer on the gold nanoparticles growth and morphology.

Incorporation of etoposide
Firstly the stock solution of etoposide was prepared by dissolving 0.15gm of etoposide powder in 0.5ml of dimethyl sulfoxide ( (DMSO) and completed to 10ml by buffer solution of (pH 10.5), sonicated for about 3hours at room temperature till a clear solution was obtained.
Etoposide-loaded gold nanoparticles suspensions were prepared by adding 1ml of etoposide solution at a calculated concentration (5, 10 or 15mg/ml for F1-F3, respectively) to 9ml of polymeric gold nanoparticles suspension, in a tightly closed conical flask, and mixed for 1hour on a hot plate magnetic stirrer at 55°C and 1500rpm, then sonicated for 5hours until achieved a clear transparent suspension, and stored in refrigerator at 2-8°C in a tightly closed dark container coated by aluminum foil, prior to further characterization (19,20) .
AFM measurement was used to evaluate the effect of initial etoposide loading on nanoparticles size, distribution and morphology. Evaluation of etoposide loading efficiency 0.5ml of etoposide-loaded gold nanoparticles suspension was centrifuged with Amicon filter of 3KDa/MWCO, at 6000rpm for 20minutes. The amount of free etoposide was determined by UV-Visible measurement, and then calculated the etoposide loading efficiency by the following equation: Where LE% is the percentage of drug loading efficiency, D0 is the total amount of drug fed in formula and Df is the amount of free drug leak through Amicon filter (21,22) .

Evaluation of in-vitro etoposide release
2ml of etoposide-loaded gold nanoparticles suspension was added into cellulose diffusion dialysis tubing of 12KDa/MWCO; both ends were tied, and immerged in a dissolution jar filled with 500ml of dissolution media at a constant temperature (37±1°C) and rotated at 100rpm. Samples of 3ml were withdrawn at predetermined time intervals of 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 5, 10, 15, 20 and 24hours; and replaced each time immediately with equal volume of the fresh dissolution media to maintain constant volume. The samples were analyzed for etoposide content using UV-Visible spectrophotometer. The percentage of drug released was calculated by the following equation: Where D% is the percentage of drug release at time t, Dt is the amount of drug release at certain time interval and D0 is the total amount of drug fed in formula.
The pH-responsive capacity of etoposideloaded gold nanoparticles was investigated using two different dissolution media (phosphate buffer pH 7.4 and acetate buffer pH 4.6) (23,24) .

Results and Discussion
Citrate reduction method was a more straightforward procedure for gold nanoparticles preparation, as a result of seed mediated growth mechanism; illustrated in Figure 1.

Figure 1. Gold nanoparticles formation by seed-mediated growth mechanism (12)
Citrate has three carboxyl groups, which can be protonated/deprotonated leading to an acid base buffer. In addition, citrate can be irreversibly oxidized to faster reducing agent (dicarboxy acetone). Citrate was interplay many chemical and physicochemical processes which initiate and also terminate of seed particles formation. While gold ions Au +3 exist in different chemical species which can be reversibly converted to each other as shown in the following reaction:  (25) In first step of preparation; immediately after the addition of reducer, gold solution was discoloring instantly from faint yellow to colorless solution. This result was emerged due to a partial fast reduction of Au +3 and formation a small cluster from neutral gold monomers Au, which initiate formation of seeds in the next step. As well as, initial pH (3.3) increased to about 6.5 upon addition of citrate solution, which shift the gold reaction equilibrium toward the less reactive species [AuCl3-X(OH)1+X] that terminate uncontrolled excessive formation of seed particles. The balance between these two processes determines the final number of particles formed (26) .
After about five seconds, a light gray color starts to appear quickly because the solution became supersaturated with Au clusters. Then the solution color was turned into dark gray-bluish within about thirty seconds and gradually changed to purple. This result indicated the formation of stable seed particles with size more than 1.5nm (consist about 840 Au atoms). Subsequently, the remaining Au +3 attached as co-ions in an electronic double layer of seeds, leaded to change their electronic properties (27,28) .
At the last step of preparation; after about 28 minutes, a red color was beginning to appear. This step comprises the slow reduction of Au +3 until a full consumed of the precursor whereby the generated Au exclusively grew on the top of seed particle's surfaces (29) .

Qualitative dispersibility measurements
Gold nanoparticles suspension (G) was clear, transparent, shiny, and homogeneous, with a characteristic red color. A visible red laser beam path was observed through suspension, which confirm the presence of suspended spherical gold nanoparticles; that was scattering the shorter wavelength light by Willis-Tyndall effect. The brightness of suspension indicated for particles distribution uniformity. As well as it gave a positive salt test; Sodium ions neutralized the surface charge of nanoparticles and subsequently, lose their mutual electrostatic repulsion and started to aggregate for larger nanoparticles, as a result their color changed from red to blue-purple. Upon addition an excess amount of NaCl, a colorless solution with dark precipitate was observed, which mean absent of suspended nanoparticles by overcoming the colloidal stability (9) .

UV-visible spectrum analysis
UV-Visible curve (Figure 2) of gold nanoparticles (G) had a single peak with maximum absorption at 525nm and high size-sensitive regions confirmed the successful formation of spherical gold nanoparticles with size range of 20-30nm. The curve revealed a low shape-sensitive region with narrow FWHM indicated for particles distribution uniformity (18,30) .

Optimization of gold nanoparticles
In this study, different variables had been studied for their effect on the prepared suspensions, in order to select the optimum gold nanoparticles that achieve the aim of our study.

Effect of HAuCl4.3H2O concentration
Gold nanoparticles G2, G, and G3 with HAuCl4.3H2O concentration 0.125, 0.25 and 0.5mM respectively, showed a significant decreased (p<0.05) in final nanoparticles size ( Figure 5) and overall reaction time with increased HAuCl4.3H2O concentration, while size distribution ( Figure 6) was not significantly affected (p>0.05) with this molar concentration limit. These results occurred because of HAuCl4.3H2O concentration determined pH value of precursor solution and thereby ratio of gold chemical species during seed particles formation. G3 with higher HAuCl4.3H2O concentration (0.5mM) and lower pH value (3), larger percentage of higher redox potential species [AuCl4] -1 was available to provide more Au and form more seeds resulting in smaller final nanoparticles (average 19.63nm) with better size distribution (23.95%). As well as shorter overall reaction time (23minutes), indicated that larger total surface area of seeds facilitated the kinetic of growth mechanism. G1 with very low HAuCl4.3H2O concentration (0.0625mM) and initial pH value (4), gave larger nanoparticles with wide distribution, because the initial reactive species [AuCl4] -1 was very low to trigger fast formation of seeds leaded to formation of less particles with uncontrolled growth (12; 31) .
G4 with 1mM HAuCl4.3H2O (initial pH=2.4), A black precipitate was observing within few hours after preparation, which due to the buffering capacity of citrate was not sufficient to stop particles growth also the ionic strength which subsequently decreased the colloidal stability by increased the attractive van-der-walls force between particles. For gold nanoparticles G5 with 2mM HAuCl4.3H2O (initial pH=1.5), the reduction of Au +3 was observed as converting yellow color of diluted gold solution to colorless, but only macro-sized particles were formed and precipitated at lab glass walls. This observation was explained by the effect of ionic strength that was vanished the colloidal stability. At higher concentration of HAuCl4.3H2O more than 2mM (initial pH<1.5), no reduction was observed and solution still has faint yellow color after addition of reducer. This happened due to the concentration of hydrogen ions H + exceeded the buffer capacity of citrate; which became full protonated by mixing with HAuCl4.3H2O. First step of seed-mediated growth mechanism required at least one deprotonated carboxyl group (26; 27; 32) .
G3 with best HAuCl4.3H2O concentration (0.5mM) gave a better morphology, size, and size distribution was chosen for further optimization studies.

Effect of Na3C6H5O7.2H2O concentration
Gold nanoparticles G3 and G7-9 with sufficient concentration of Na3C6H5O7.2H2O (0.5-10mM) showed a significant decreased (p<0.05) of particles size distribution (Figure 8) by increased citrate concentration, while nanoparticles size (Figure 7) remained almost non-significant affected (p>0.05) within this certain concentration limit. These results were happened due to the kinetic of the protonation equilibrium of citrate solution and the final neutral pH (about 6.5), which approximately the same irrespective to molar excess. G9 has better size distribution (21.70%) indicated that the balance between gold and citrate concentration was reached (19) . G6 and G10 with very low (0.25mM) and very high (20mM) citrate concentrations respectively, gave larger nanoparticles (average 34.51nm and 38.09nm) with wide distribution (29.23% and 34.66%) respectively. The reasons for these results were the citrate concentration in G6 might still be sufficient to reduce Au +3 but not to shift gold complex equilibrium from [AuCl4] -1 to less reactive hydrolyzed species, therefore seeds which still grew to reach colloidal stability. In contrast, in G10 the colloidal stability of seeds decreased due to increase the ionic strength leaded to larger final particles size with non-uniform distribution (25; 33) .
G9 with best Na3C6H5O7.2H2O concentration (10mM) gave a better morphology, size, and size distribution was chosen for further optimization studies.

Effect of preparation temperature
AFM analysis (Figures 9-10) appeared a complex relationship between the preparation temperatures and gold nanoparticles size reduction. Firstly, the gold nanoparticles size and their distribution significantly decreased (p<0.05) with increased the preparation temperature (from room temperature to 65°C), then the nanoparticles size and their distribution significantly increased (p<0.05) by increased the preparation temperature from 65-85°C. This nonlinear temperature dependence happened due to the time frame of seeds formation became shorter by increased temperature, while the reactive [AuCl4] -1 was faster reduced during an available time, also the elevated temperature promote shift of gold equilibrium toward more hydroxylated species [Au(OH)4] -1 . Because of these opposing tendencies, amount of Au available for seed formation would vary with temperature (12) . G14 prepared at 65°C gave smallest size (average 13.24nm) with best distribution (18.90%) means at this temperature the balance between reducing and stabilizing activity of citrate was occurred. G11 prepared at room temperature gave larger nanoparticles (average 41.17nm) with bad distribution (42.63%), because at this temperature the reaction takes about 2days, without formation of dicarboxy acetone, thereby very slow reduction of Au +3 at initial phase of preparation, leaded to produce too few Au monomers for seeds formation and lastly larger particles was formed with uncontrolled growth mechanism (7; 34; 35) .
G14 prepared under best temperature 65°C gave a better morphology, size, and size distribution was chosen for further optimization studies.

Effect of stirring rate
The gold nanoparticles size ( Figure 11) was not significantly affected (p>0.05) by changed the stirring rate, while their size distribution ( Figure 12) as well as the overall reaction time were decreased significantly (p<0.05) with increased the stirring rate (36) . G20 with vigorous stirring rate of 1500rpm gave a best average particles size (13.05nm), and their distribution (12.47%) was chosen for further optimization studies. Effect of reactants addition order G21 was prepared by inversed citrate reduction method and determined the effect of change the reactants addition order via comparing with G20, which prepared by the classical method. AFM analysis (Figure 13-14) indicated that the morphology, particles size, and distribution were improved significantly (p<0.05) when prepared the gold nanoparticles by inversed method. This improvement attributed to formation of dicarboxy acetone via thermal oxidation of citrate during preheating of diluted reducer solution, which acts as a kinetically faster reducer and strong stabilizer during initial phase of reaction. In addition, HAuCl4.3H2O concentration was very high (50Mm) in the inverse method this provided presence of more [AuCl4] -1 during seeds formation. Subsequently, more Au monomers were available and more seeds formed, which leaded to produce smaller gold nanoparticals (average 8.75nm) with more uniform size distribution (7.32%) as well as short over all preparation time (27minutes) (7; 25; 37) .
From all these studies, G21 considered as the optimized gold nanoparticles suspension.

Zeta potential measurement
Zeta potential is an important pointer of steadiness the gold colloidal dispersion. For the optimized suspension G21; with nanoparticles which small enough, a highly negative zeta potential of -47.87 ( Figure 15) gave a steadiness, therefore the suspension would be struggle against aggregation due to a strange repellent force between particles. The reason behind negative charge was the citrate ions at the surface of nanopaticles (38) .

Scanning electron microscopy
The optimized suspension G21 was examined by SEM at different magnifications and the nanoparticles only appeared at higher magnification 50Kx with spatial resolution of 500nm.
SEM image ( Figure 16) revealed gold nanoparticles with uniform shiny spherical shape and smooth surface that consistently dispersed on the surface (18) . nanoparticles size measured by SEM (average 8.96nm) was not significantly differed (p>0.05) than the data of AFM analysis (average 8.75nm).
The comparison of particles size distribution measured by AFM and SEM indicated that wider distribution (10.79%) when measured via SEM than AFM (7.32%), this result was most likely artifacts formed during SEM sample preparation (drying at room temperature take about 1week) and do not represent actual properties of gold nanoparticles G21.

Figure 16. SEM images of the optimized gold nanoparticles G21
Energy dispersive x-ray spectroscopy EDX measurements considered as good qualitative analysis for different materials due to the number of bremstrahlung x-rays varies with mean atomic number the background counts. Heavy elements as gold (atomic number Z=79) quantify easily by comparing the intensity of their peak. Light elements; oxygen (Z=6), carbon (Z=8), sodium (Z=11) and chloride (Z=17) are difficult to quantify by this method. Hydrogen (Z=1) has too low energy to be detected by EDX and do not have characteristic x-ray peak (39) .
EDX analysis (Figure 17) of the optimized suspension G21 gave an optical absorption band peak at approximately 2.2KeV with high intensity, which was the typical of metallic gold nanocrystallites (16) . Elemental mapping analysis showed extreme distribution of metallic gold in the EDX micrograph image (Figure 18).

Transmission electron microscopy
TEM 2D image (Figure 19) of the optimized suspension G21, revealed gold nanoparticles with uniform spherical shape and average particles size 8.81nm. This data was not significantly (p>0.05) differed from the data of AFM and SEM analysis (average 8.75nm and 8.96nm respectively).

Functionalization of gold nanoparticles
The phenomenon for polymer to enhance the stability of gold nanoparticles depended on the steric repulsion force between polymer moieties, once polymer attached to the outer surface of gold nanoparticles provided enough repulsive potential that can stabilize the dispersion (40,41) . Figures 20 and 21 demonstrated the AFM analysis of the functionalized gold nanoparticles. Increased the particles size of polymeric gold nanoparticles compared to the free gold nanoparticles, with uniform particles size distribution, indicated that gold nanoparticles were indeed conjugated with polymer through coordination bonds (42) .

Incorporation of etoposide
Three formulas of etoposide-loaded gold nanoparticles were prepared by incorporation technique. Polymeric gold nanoparticles with ultra nano-size can give a high surface area for etoposide incorporation. These nanoparticles had a high loading capacity besides the presence of surface active groups (hydroxyl, carboxyl or amide) might the formation of bonds with etoposide per unit area (43,44) .
The effect of etoposide loading was evaluated using three different concentrations of etoposide solution (5, 10, 15mg/ml) for formulas F1-F3 respectively. Etoposide solutions of concentration above 5mg/ml at pH 10.5 became cloudy within about thirty minutes after preparation, as well as the aqueous etoposide solution became less stable at pH more than 5 (45,46) , therefore it must be freshly prepared through the incorporation process.
The etoposide loading efficiency was significantly (p<0.05) decreased with increased the initial loading of etoposide, as shown in Figure 22. F2 and F3 were rejected from experiment. Figures 23-24 demonstrated the AFM analysis data of formula F1. The etoposide dissolution profile showed a better etoposide retardation at phosphate buffer pH 7.4 and greater release rate with controlled manner at acetate buffer pH 4.6, as shown in Figure 25 (19,47) .

Conclusion
In conclusion, the citrate reduction method was applied due to its feasibility, scalability and reproducibility. Molar concentration of reactants, preparation temperature and stirring rate had a considerable role in the controlling size and uniformity of gold nanoparticles. As well as, inversed method is seemed to be promising for the preparation of gold nanoparticles with high quality in one step; it can achieve that gold nanoparticles were greatly improved their morphology, size and monodispersity compared to nanoparticles prepared via classical method. nanotherapeutic delivery system based on gelatin-coated gold nanoparticles: formulation, drug release, and multimodal imaging of cellular