Biosynthesis and characterization of gold nanoparticles using Brazilian red propolis and evaluation of its antimicrobial and anticancer activities

Gold nanoparticles (AuNPs) are highlighted due to their low toxicity, compatibility with the human body, high surface area to volume ratio, and surfaces that can be easily modified with ligands. Biosynthesis of AuNPs using plant extract is considered a simple, low-cost, and eco-friendly approach. Brazilian Red Propolis (BRP), a product of bees, exhibits anti-inflammatory, anti-tumor, antioxidant, and antimicrobial activities. Here, we described the biosynthesis of AuNPs using BRP extract (AuNPextract) and its fractions (AuNPhexane, AuNPdichloromethane, AuNPethyl acetate) and evaluated their structural properties and their potential against microorganisms and cancer cells. AuNPs showed a surface plasmon resonance (SPR) band at 535 nm. The sizes and morphologies were influenced by the BRP sample used in the reaction. FTIR and TGA revealed the involvement of bioactive compounds from BRP extract or its fractions in the synthesis and stabilization of AuNPs. AuNPdichloromethane and AuNPhexane exhibited antimicrobial activities against all strains tested, showing their efficacy as antimicrobial agents to treat infectious diseases. AuNPs showed dose-dependent cytotoxic activity both in T24 and PC-3 cells. AuNPdichloromethane and AuNPextract exhibited the highest in vitro cytotoxic effect. Also, the cytotoxicity of biogenic nanoparticles was induced by mechanisms associated with apoptosis. The results highlight a potential low-cost green method using Brazilian red propolis to synthesize AuNPs, which demonstrated significant biological properties.

Metallic nanoparticles can be considered one of the most versatile types of nanoparticles due to their applications in chemistry, electronics, medicine, and pharmaceutical sciences 1 . Among them, the gold nanoparticles (AuNPs) stand out for their advantages such as biocompatibility, tunable optical properties, and easily changed surface chemistry 2,3 . Because of these unique physical-chemical properties, the AuNPs are widely used as carriers of drugs and molecules to improve the diagnosis and treatment of diseases 4,5 .
The synthesis of AuNPs through chemical and physical routes has been already well-established. However, these pathways generally use toxic substances and non-polar solvents, which generate hazardous impacts for the environment and requires various steps of product purification, resulting in an expensive process 6 .
For overcoming the challenges related to conventional methods, a biosynthetic route has been proposed in the literature 7 . The green synthesis uses natural compounds from plants or microorganisms (e.g., fungi, bacteria, algae) as precursors of the reaction of gold ions reduction 8,9 . Biosynthesis is considered a simple, low-cost, and eco-friendly approach since it uses non-toxic solvents, such as water 10 . The production of metallic nanoparticles using natural sources has already been reported in the literature, showing it is a potential synthetic route that should be explored 11,12 .
Plant extracts are complex mixtures providing a rich arsenal of molecules with high redox potential 13 , such as flavanones, flavones, flavonols and chalcones, fatty acids, amino acids, terpenoids, aldehydes, and alcohols 14 . Biosynthesis of gold nanoparticles. Au(III) has a high reduction potential and can be reduced by phenolic compounds from the natural extracts 42 . In this study, we produced gold nanoparticles (AuNPs) with BRP extract (AuNP extract ) and its fractions (AuNP hexane, AuNP dichloromethane , AuNP ethyl acetate ) using the green synthesis method. In order to increase the solubility of the compounds present in the extract, the pH of the mixture of gold solution with the propolis extract was adjusted to 7.0. Moreover, this pH alteration is also involved in the activation of phytochemical compounds, which facilitates the donation of electrons to the metal, reducing Au 3+ to Au 042 .
The formation of AuNPs was confirmed by the presence of the Surface Plasmon Resonance (SPR) band. This band occurs on the surface of certain metals on a nanometer scale 43 . Both the extract and its fractions (hexane, dichloromethane and ethyl acetate) produced AuNPs, showing a prominent peak at a range of 523-541 nm (Fig. 2). The optimal extract or fractions concentration to produce AuNPs was also investigated. The high formation of AuNPs was obtained using 200 μg mL −1 of the extract or its fractions. Moreover, we verified a wider SPR band and turbidity signs in the AuNPs dispersion when the red propolis extract concentration was increased twofold (400 μg mL −1 ) (Fig. 2a).
In consonance with Gatea et al. 30 and Roy et al. 29 , we also observed an increase in the absorbance values related to the time with a redshift of the SPR band ( Fig. 2b-e). All formulations exhibited a similar absorption peaks profile, indicating a rapid growth of particles. After some time (~ 2 h), the saturation was reached, indicating the formation of stable nanoparticles 29 . The acetate and dichloromethane fractions were more efficient to produce AuNPs since the intensity of SPR bands was higher than the others 44 . This result can be associated with the higher polarities of dichloromethane and ethyl acetate in comparison with hexane used in the extraction process 45 . The color of the samples changed from pale yellow to dark red or purple (Fig. 2). The different colors of AuNPs-from light pink to dark red-are dependent on the size, shape, and structural characteristics of these nanoparticles 42 . Morphology and diameter distribution. The morphologies and diameter distributions of AuNPs formed by three fractions and crude extract of BRP were investigated using the transmission electron microscopy (TEM) analysis. Figure 3a,c exhibits AuNP extract and AuNP ethyl acetate with mostly spherical shapes, whereas the AuNP dichloromethane and AuNP hexane showed a variety of shapes (Fig. 3e,g). In general, the biosynthetic route produces nanoparticles with different morphology and size as a result of the chemical composition of the extract 46,47 . Hexane fraction is rich in benzophenones (Guttiferone E; Oblongifolin A), while ethyl acetate fraction is rich in flavonoids and isoflavonoids such as liquiritigenin, formononetin, vestitol, and neovestitol. BRP crude extract and dichloromethane fraction present a similar phytochemical profile ( Table 1).
The average size, measured by TEM, was in the range of 8-15 nm for all gold nanoparticles showing a narrow particle size distribution (Fig. 3b,d,f,h). The smaller nanoparticles showed spherical shapes, whereas the larger particles exhibited several geometrical forms, such as triangles, pentagons, hexagons, and rods ( Fig. 4a,b). Similar results were reported by Smitha et al. 48 , Philip et al. 49 , and Gosh et al. 50 , who verified that nanoparticles size with different geometries were larger than those with spherical morphology due to low quantities of the  49 . The Energy Dispersive X-ray Spectroscopy (EDXS) was performed to gather information about the chemical composition of samples for elements with atomic numbers (Z) > 3. All AuNPs spectra showed an absorption band peak of approximately 2.2 keV, which is characteristic of gold absorption 5,51 . The HR-TEM images ( Fig. 5a-d) of all AuNPs demonstrated highly ordered planar spacing consistent with the internal spacing of the gold plane. The AuNP hexane showed inter-planer spacing of 1.9 Å, whereas the AuNP ethyl acetate and AuNP dichloromethane exhibited 2.2 Å of d-spacing, which is in agreement with the (200) and (111) lattice of face-centered cubic (fcc) gold (JCPDS card N o 04-0784) 52,53 . A crystalline structure of the biosynthesized gold nanoparticles was also evidenced by the SAED pattern ( Fourier-transform infrared spectroscopy (FTIR) and Thermal gravimetric analysis (TGA). The relation of phytochemicals compounds involved in the biosynthesis and stabilization of AuNPs was evaluated by FTIR and TGA. Table 2 shows the main peaks in the FTIR spectra of BRP extract or its fraction, as well as the AuNPs produced.
The peaks around 1700 cm −1 are distinctive in hexane, and the dichloromethane fractions spectra are associated with carbonyl groups (C=O). These signals can be related to the major compounds in these fractions such  www.nature.com/scientificreports/ as prenylated benzophenones 55 and phenolic compounds 56 . The peaks at 1600 cm −1 are usually correlated to the stretching vibration of carboxylate anion -COO − , probably due to the oxidation of polyphenols during Au +3 reduction 10,57 . These bands show large amounts of alcohol or phenol in plant extracts. The bands around 3400 cm −1 in the AuNPs dispersions were smaller than the extract FTIR spectra (Fig. 6), suggesting the oxidation of hydroxyl groups to carbonyl groups. It indicates that the OH groups present in BRP extract or fractions are the main compounds involved in the reduction of Au ions 58 . The absorption peaks measured at 1500 cm −1 are probably due to free NH groups present in proteins. The reduction of this band intensity after the formation of AuNPs indicates that the proteins in the extract were also used for capping AuNPs, improving its stability 59 . Furthermore, decreased peak intensity at 800 cm −1 suggests the binding between the C-H group of phenolic acids and AuNPs 60 .
These results explain the role of chemical compounds of the BRP extract in reducing Au +3 and stabilizing AuNPs 65 . Also, FTIR spectra results support the idea that biosynthesized nanoparticles are surrounded by a thin layer of phytomolecules including polyphenols, such as flavonoids and tannins, in addition to terpenoids and proteins 58 . The BRP extract FTIR spectra show intense bands in 3436 cm −1 and 2922 cm −1 , before and after the Au ions reduction (Fig. 6), which corresponds to free O-H bonds and CH 2 stretching vibrations, respectively 30,59 .
The TGA graphs of the BRP extract, fractions, and AuNPs (Fig. 7) showed a steady weight loss in the temperature range of 150-600 °C with a total weight loss of up to 800 °C. Table 3 shows the percentage of weight loss of the AuNPs. The percentage differences in the weight loss shown in the table are related to the difference in organic composition present on the AuNP's surface. It is expected that the weight loss between 100 °C and 200 °C refers to the evaporation of adsorbed water of the capping extract 8 . The largest weight loss observed in the range of 250 to 500 °C may be a result of the thin layer burning of organic material surrounding the nanoparticles 59 . Also, it is supposed that after 400 °C occurs the degradation of resistant aromatic compounds 62 . At the final of the  BRP crude extract and its fractions exhibited the highest antimicrobial activity. In fact, the loss of activity of biosynthesized nanomaterials has been reported in the literature 30,68 . This difference might be explained by the oxidation of some active compounds of plant extracts during the reaction with metals. Several mechanisms have been proposed for the reduction of metal ions using plant extracts. Several authors suggest that the phenolic compounds are the main reducing and capping agents involved in the synthesis of metallic nanoparticles 69 . It would justify the lower antimicrobial activity of AuNPs, although they present phenolic compounds.  The difference between the antimicrobial activity of each biosynthesized AuNPs can be explained by the phytochemical composition of each extract and fraction 70 . According to a bio-guided study of Brazilian red propolis extract and its fractions, the crude extract and hexane fraction showed the highest antimicrobial activity against microorganisms, including S. aureus and E. coli. This activity can be related to the presence of medicarpin, elemicin, and vestitol 71,72 . Therefore, the medicarpin and vestitol we quantified in the BRP crude extract might have contributed to the antimicrobial effect of the BRP extract and AuNP dichloromethane .
Furthermore, benzophenones (Guttiferone E and Oblongifolin B) are present in low concentrations in the BRP crude extract and the dichloromethane fraction, and in high amounts in the hexane extract. Benzophenones are described as effective compounds against bacteria and fungi 39,73,74 . Also, the activity of AuNP hexane and AuNP dichloromethane can be related to the presence of these molecules on the surface of nanoparticles. On the other hand, AuNP extract and AuNP ethyl acetate showed low or no antimicrobial activity that can be associated with the low amounts of medicarpin, vestitol, and benzophenones in these extract and fraction. Moreover, ethyl acetate fraction did not present benzophenones showing the lower antimicrobial activity.
Additionally, the nanoparticle shape can influence its interaction with microorganisms and, consequently, affect its antimicrobial activity 75,76 . Thus, the heterogeneous gold nanoparticle shapes observed in the AuNP dichloromethane and AuNP hexane dispersion may have influenced its antimicrobial activity.
Cytotoxic activity. The in vitro cytotoxic activity of BRP extract, fractions, and biosynthesized AuNPs were evaluated in bladder cancer cells (T24) and prostate cancer cells (PC-3) using the Resazurin assay. The resazurin assay provides a simple, non-toxic and sensitive measurement for the viability of mammalian cells 77,78 .
Dose-dependent cytotoxic activity was observed in all samples (Fig. 8). The BRP crude extract and its fraction showed a high cytotoxic effect (Table 5). According to the U.S. National Cancer Institute, extracts with IC 50 up to 30 µg mL −1 are classified as potent cytotoxic agents 79,80 . Moreover, BRP crude extract and dichloromethane fraction exhibited the highest cytotoxic effect (Table 5). BRP crude extract and dichloromethane fraction present similar phytochemical composition such as formononetin, liquiritigenin, and medicarpin in high concentration, which are known to exhibit cytotoxic activity in different cancer cells (Table 1)  Although AuNP extract and AuNP dichloromethane presented the same cytotoxic profile as their respective precursors (BRP crude extract and its fractions), the AuNP extract exhibited the highest cytotoxic activity among the nanoparticles ( Fig. 8 and Table 5). The hexane fraction displayed the highest in vitro antitumor effects, with IC 50 values around 12 µg mL −1 (Table 5), probably due to the presence of benzophenones (Table 1). Novak et al. 85 reported that a red propolis fraction induced tumor regression of melanoma xenografts in mice and related this to a benzophenone's activity. On the other hand, the low cytotoxicity of AuNP hexane (IC 50 of 59.5 µg mL −1 for T24 and 89 µg mL −1 for PC-3) might be explained by the participation of the benzophenones in the AuNP formation.
Ethyl acetate fraction and AuNP ethyl acetate showed the least activity among the samples tested. These results could be related to the absence of compounds such as medicarpin and the benzophenones (present in extract and hexane fraction), which could have an important role in the anti-cancer activity and the synergism effect between the molecules. Some studies proposed that the anti-proliferative and cytotoxic effects of propolis against cancer cells might be correlated to the synergism between properties of several compounds and may not exclusively due to the concentration of a specific molecule 21 .      50 values between PC-3 and T24 cells can be related to the malignant degree of the cells. T24 cell is derived from transitional cell carcinoma grade II, whereas the PC-3 cell line is derived from bone metastasis from grade IV adenocarcinoma. Thus, the PC-3 cell is more malignant and resistant to treatment 86 . Similar results were verified by Carvalho et al. 87 . Although all AuNPs were more cytotoxic for T24 cells, some nanoparticles showed high cytotoxicity to PC-3 cells decreasing the cell viability up to 10% in the highest concentration (Fig. 8).
Biogenic gold nanoparticles presented higher IC 50 than their respectively extract or fraction precursors. These results can be explained by the mechanism of gold nanoparticles production, since some compounds of extract or fraction were used to reduce Au 3+ to form the metallic nanoparticles in the green synthesis 29 . Hence, after nanoparticles preparation some active molecules of extract or fractions may lose its biological function. Owing to the complexity of propolis composition, future studies to determine the specific compounds from BRP that are responsible for gold nanoparticles reduction are demanded.
Although biosynthesized gold nanoparticles did not display a better cytotoxic effect than their precursors, exhibiting only residual cytotoxicity from BRP active compounds, they still constitute a great advantage compared with gold nanoparticles synthetized by chemical methods. El Domany et al. 88 described that biosynthesized AuNPs were more cytotoxic than chemically synthesized AuNPs in studies with PC-3, HCT116, and HepG2 tumor cells. Besides, the easy modification of the allows their association with other antitumor drugs and molecules that can target specifics cells, such as cancer cells 89 . Researchers have been reported the beneficial effects of the combination between anticancer agents and polyphenols 90,91 , as well as compounds found in BRP 92-94 . Qiao et al. 95 showed that green tea catechins associated with antitumor drugs are more effective than monotherapy.
Adenosine triphosphate (ATP) bioluminescence assay. Cell culture viability was determined through quantification of the luminescent signal, produced by transformation of luciferin by luciferase as a function of intracellular ATP concentration 96 . ATP is the primary source of cellular energy, and its concentration is related to the number of living cells 97 . Luminescent ATP detection assay are robust and more sensitive than MTT or similar assays once the ATP generation depletes as soon as the cell dies 98,99 .
The results of ATP tests in PC-3 cells showed that biosynthesized AuNPs and its precursors (crude BRP extract and its fractions) present the same cytotoxic profile obtained by resazurin colorimetric determination (Fig. 9). This strain was chosen for the experiment because it is considered more resistant than T24 cells. Thus, as obtained in the resazurin assay, the hexane fraction was the most cytotoxic between the extract and fractions. Among the nanoparticles, AuNP extract was the most effective in reducing cell viability, followed by AuNP dichloromethane . Therefore, these results confirm the cytotoxic profile previously verified by the resazurin assay.
Flow cytometry. The cell death mechanism by flow cytometry assays demonstrated that BRP crude extract and its fractions, and biosynthesized AuNPs induced the death mainly by apoptosis. The percentage of apoptosis ranged between 44 and 66% related to the total of dead cells for all treatments (Fig. 10). Begnini et al. 100 described BRP induced apoptosis in 5637 cells by molecular ways related to the P53 and Bax/Bcl-2. Another This preliminary evidence has proven that biosynthesized gold nanoparticles using BRP displayed cytotoxic effects in cancer cells and can be a promising alternative therapeutic agent in cancer treatment.

Conclusion
The Brazilian Red Propolis hydroethanolic extract and its fractions showed a great potential to produce gold nanoparticles with size range of 8-15 nm. Due to the specific composition of the extract or its fraction, AuNPs with different morphology were produced. Spherical AuNPs were obtained using BRP crude extract and ethyl acetate fraction, while dichloromethane and hexane fractions produced AuNPs with different shapes. All AuNPs showed a crystalline structure with a face-centered cubic (fcc) lattice. FTIR spectroscopy suggested the attachment of bioactive compounds from the Brazilian red propolis extract or its fraction on the surface of AuNPs. These results were confirmed by antimicrobial and cytotoxic activity of AuNPs produced. AuNP extract showed antifungal activity and high cytotoxicity at low concentrations in bladder and prostate cancer cells. AuNPs obtained with dichloromethane and hexane fractions displayed high antibacterial and antifungal activities and cytotoxicity in both cells studied. On the other hand, the nanoparticles prepared with ethyl acetate fraction did not show antimicrobial activity and exhibited the lowest in vitro cytotoxic effect in the cells evaluated. The different phytochemical profiles of BRP extract and its fractions enabled the preparation of gold nanoparticles with distinct properties and interesting for several applications. Therefore, these in vitro results demonstrate the promising therapeutic applications of the biogenic gold nanoparticles in nanomedicine.

Materials and methods
Materials. Gold   www.nature.com/scientificreports/ Biosynthesis of gold nanoparticles. AuNPs were synthesized by mixing a stock solution of tetrachloroauric acid trihydrate (0.5 mM) with BRP crude extract solution (200 μg/mL) or its fractions solution (hexane, acetate and dichloromethane), followed by NaOH addition until pH 7.0. The mixture was stirred for 1 h at optimized temperature. The gold nanoparticles formation was observed by color changed from pale yellow to dark red.
Characterization of gold nanoparticles. The formation of gold nanoparticles was verified by the presence of a surface plasmon resonance (SPR) band, with maximum absorption between 500 and 550 nm, using UV-Vis spectroscopy. The spectra were collected at different times of formation using a UV-Vis spectrophotometer (Implen, Munich, Germany). AuNPs concentration used in the biological analysis was performed by Nanoparticle tracking analysis (NTA) using a Nanosight NS300 with 488 nm laser (Malvern Instruments). The morphology, shape, size and elemental composition of AuNPs were analyzed by Transmission Electron Microscopy (TEM) and high-resolution TEM FEI TECNAI G 2 F20 coupled with Energy-dispersive X-ray Spectroscopy (EDXS) (Thermo Fisher Scientific, USA), operating a beam voltage of 200 keV. The diameter distribution was obtained using the Image J (NIH, USA) software using the TEM images. The crystalline structure of AuNPs was assessed by Selected Area Electron Diffraction (SAED). For Fourier-transform infrared spectroscopy (FTIR) and Thermal gravimetric analysis (TGA), biosynthesized AuNPs dispersion was centrifuged at 15,000 rpm for 60 min at 4 °C. AuNPs pellet was washed three times with deionized water with 50% ethanol to remove the excess of the extract from the AuNPs solution. The samples were lyophilized, and AuNPs powder or the BRP crude extract were mixed with potassium bromide (KBr) to obtain pellets. FTIR spectra were recorded using an IRTracer-100 (Shimadzu, Kyoto, JA) in the range of 4500 to 500 cm −1 with a resolution of 2 cm −1 . TGA was carried out using an SDT Q 600 thermal analyzer (TA Instruments, New Castle, DE, USA). The samples were placed in platinum pans and heated under an inert atmosphere with a rate of 10 °C/min at 900 °C.
Antibacterial and antifungal activities. Antibacterial 101 . The AuNPs samples were diluted in Brain Infusion Broth (Difco) and different concentrations of the samples (7.3 × 10 6 to 7.5 × 10 9 AuNP/ mL) were added to each well. Then, the microorganism (5 × 10 5 CFU/mL) was added to all wells. Tetracycline (0.0115 µg mL −1 to 5.9 µg mL −1 ) was used as a positive control for bacteria and Amphotericin B (0.031 µg mL −1 to 16 µg mL −1 ) was used as a positive control for yeast. The plates were incubated for 24 h at 37 °C. Afterward, a resazurin solution (0.02%) was added to determine the microorganism viability. Before Resazurin addition, an aliquot of the inoculum was removed from each well and seeded on BHI agar supplemented with 5% sheep blood for the MBC test. The plates were incubated as previously described. MBC was defined as the lowest concentration of the sample without microbial growth.
Cytotoxicity assay. The cytotoxic activity of biosynthetic gold nanoparticles was evaluated in human urologic cancer cell lines. T24 bladder cancer and PC-3 prostate cancer cell lines were obtained from the Rio de Janeiro Cell Bank (BCRJ). T24 cell was cultured in RPMI medium (Sigma-Aldrich, USA) and PC-3 cell was cultured in DMEM-Dulbecco's Modified Eagle Medium (Sigma-Aldrich, USA), both containing 1% antibiotic (Penicillin-Streptomycin-Sigma-Aldrich, USA) supplemented with 10% fetal bovine serum (SBF-GIBCO, Thermo Fisher Scientific, USA). These cells were maintained in continuous culture under a humid atmosphere at 37 °C and 5% CO 2 . T24 or PC-3 cells were seeded in 96-well plates (2 × 10 4 cells/well) and treated with different concentrations of biosynthesized AuNPs for 24 h. Afterward, the medium containing nanoparticles was removed and the cells were washed with PBS. Subsequently, a Resazurin solution (25 µg mL −1 ) was added to each well and incubated for an additional 4 h. Then, fluorescence was evaluated on a Microplate reader (Synergy HTX Multi-Mode Microplate Reader, Biotek) using 530 nm and 590 nm as the excitation and emission wavelengths respectively. The negative control received only medium and the positive control was treated with DMSO 1%. IC 50 values were determined using the Graph Pad Prism version 8 software (GraphPad Software Inc. San Diego CA, USA).
Adenosine triphosphate (ATP) bioluminescence assay. Cell viability was assessed using the CellTiter-Glo Assay kit (Promega Corporation). The kit reagents were prepared according to the manufacturer's protocol. PC-3 cells were plated onto a 96-well plate at a density of 20,000 cells per well. The CellTiter-Glo reagent was added to the washed cells 24 h after the treatments. The plates were shaken for 10 min at room temperature, followed by luminescence measurement using absorbance/fluorescence plate reader (Synergy HTX Multi-Mode Microplate Reader, BioTek).
Flow cytometry. The percentage of apoptosis was determined by flow cytometry using FITC Annexin V (emission) as the apoptosis marker and FVD eFluor 450 (emission) as the viability marker. PC-3 cells were seeded in 6-well plates (6 × 10 5 cell/well) and exposed to a concentration (IC 50 values) of BRP crude extract and fractions and biosynthesized AuNPs. After 24 h, the cells were washed twice with phosphate-buffered saline