Brazilian red propolis extract free and encapsulated into polymeric nanoparticles against ovarian cancer: formulation, characterisation and biological assays in 2D and 3D models

Abstract

Cancer incidence worldwide is alarming and among the cancers that affect women ovarian cancer is the most fatal. Many side effects are associated with conventional therapies and none of them are completely effective, so the development of new treatments is necessary. Brazilian red propolis extract is a natural product with complex composition and great potential for cancer treatment. However, its clinical application is harmed due to unfavourable physicochemical characteristics. To enable its application encapsulation in nanoparticles can be used.

Objectives: The aims of this work were to develop polymeric nanoparticles with Brazilian red propolis extract and compare their action with the free extract against ovarian cancer cells.

Methods: Box Behnken design was used and nanoparticles were characterised using the techniques dynamic light scattering, nanoparticle tracking analysis, transmission electron microscopy, differential scanning calorimetry and encapsulation efficiency. Activity against OVCAR-3 was also tested on 2D and 3D models.

Key findings: Nanoparticles’ sizes were ~200 nm with monomodal size distribution, negative zeta potential, spherical shape and with extract molecularly dispersed. Encapsulation efficiency was above 97% for the biomarkers chosen. Nanoparticles had greater efficacy in comparison with free propolis in OVCAR-3.

Conclusions: So far, the nanoparticles here described have the potential to be a chemotherapy treatment in the future.

polymeric nanoparticles, red propolis, ovarian cancer

Introduction

Ovarian cancer (OC) is the second most common gynaecological neoplasm and, among all types of cancer that affect women it is the most fatal, with a 5-year survival rate below 48% in metastatic stages of the disease.^[1] More than 70% of the cases are only diagnosed when they have progressed to stages III or IV, with invasive and metastatic phenotypes, which is due to the non-specificity of symptoms.^[2] Reproductive and hormonal factors such as infertility and nulliparity, as well as genetic factors such as family history and specific gene mutations and also age and overweight, can increase the risk of developing the disease.^{[3, 4]}

Ovarian tumours are of three types: epithelial carcinoma, germ cell tumour and stromal tumour. Epithelial OC is the most predominant and aggressive pathological subtype, presenting several histotypes that differ in origin, pathogenesis, molecular alterations, risk factors and prognosis.^[5]

As most ovarian cancers are diagnosed at an advanced stage, traditional treatment consists of cytoreductive surgery, followed by adjuvant chemotherapy.^[6] However, due to the toxicity of chemotherapy, patients’ quality of life dramatically decreases, as a result of the severe side effects associated with it.^[7] In addition, studies have reported the emergence of drug-resistant tumours caused by genomic instability and/or tumour microenvironment that promotes cancer growth.^[8–10] Most patients with recurrent disease tend to develop drug resistance. In addition to the drawbacks, none of the therapies have been shown to be truly effective.^[11] Worldwide the recurrence rate in patients with advanced epithelial ovarian cancer is around 80%, and relapses can occur during or after adjuvant chemotherapy.^[12]

Due to the challenges found in OC treatment, the use of pharmacotechnical strategies associated with the discovery of new therapies can improve therapeutic efficacy and quality of life of patients. Among all the pharmacological agents currently studied as chemotherapeutic drugs, opotherapeutic products are drawing attention.^[13] Propolis is a natural product used in the treatment of infections and has potential antitumor activity.^[14]

Propolis is a natural product composed of a mixture of resins, plant exudates and tree barks, added to organic secretions from the metabolism of the Apis mellifera bees, therefore classified as opotherapic. A diversity of compounds can be found in propolis such as phenolic acids or their esters, flavonoids, terpenes, aromatic aldehydes, alcohols, fatty acids and stilbenes.^{[15, 16]} This complex phytochemical composition is responsible for the multiple biological properties of propolis, such as antimicrobial, anti-inflammatory, antioxidant, antitumour and healing.^[17]

The main mechanisms of antitumour action of the different types of propolis already described involve cell cycle arrest, apoptosis and interference in metabolic pathways, including the activation of caspases, capable of activating pro-apoptotic cells, in addition to the suppression of antiapoptotic proteins. Propolis can also inhibit angiogenesis and metastasis of cancer cells.^{[18, 19]}

Brazilian red propolis extract (BRPE) produced in the state of Alagoas was identified in 2007 and classified as the 13th type of propolis and Darbergia ecastophyllum and Symphonia globulifera are its botanical origins.^[20] The complex and exclusive composition of BRPE bestows on its important biological properties since some of the compounds identified in it had not been found in other Brazilian propolis.^[21] To date, there are no reports in the literature on the antitumour activity of free or encapsulated BRPE in nanocarriers for the treatment of OC. Some researchers suggest that BRPE and some of its constituents, mainly flavonoids, act through different mechanisms and in different types of cancer. Therefore, BRPE is an interesting candidate for the development of chemotherapeutics.^{[22, 23, 19]}

Despite the diverse biological activity of natural products, their clinical application is limited due to unfavourable physicochemical properties such as low solubility and bioavailability, instability and short half-life. One of the strategies to overcome these limitations is the encapsulation in nanoparticles.^{[4, 24, 25]}

The use of nanostructured systems is constantly evolving and has been applied with great success in the health area, in drug encapsulation. Encapsulation makes it possible to modulate drug release as a function of pharmacokinetic events and, consequently, maintain its concentration at the desired location. This results in fewer administrations by reason of increased permanence of the drug at the site of action, therefore decreasing side effects. Furthermore, nanoparticles can protect the medication against early degradation.^{[26, 27]} In addition, owing to the reduced nanoparticles size and their surface properties, they can accumulate in the tumour region, much more than in normal tissues. This is a consequence of EPR effect (enhanced permeability and retention effect), in which endothelium cells in the vessels are more separated than usual and the lymphatic system is defective, thus improving nanoparticles efficacy.^[28]

Polymeric nanoparticles are versatile drug delivery systems made up of synthetic or natural polymers. The wide variety of polymers makes it possible to obtain particles with different properties, expanding its application to various diseases and routes of administration.^{[29, 30]} Among the polymers used in the preparation of polymeric nanoparticles, poly (D, L-lactic) acid (PLA) is an FDA-approved, biocompatible and biodegradable polyester. It is bioabsorbable and non-toxic, in addition to having high stability.^{[31, 32]}

The aim of this work is to produce a polymeric nanoparticle encapsulating BRPE using the design tool Box Behnken. Further, characterise this particle concerning to size, PdI, zeta potential, shape, interaction between components and test its efficacy against ovarian cancer cell OVCAR-3, comparing to the action of the free BRPE extract.

Methodology

Brazilian red propolis extract preparation

BRPE was supplied by the Laboratory of Pharmacognosy—School of Pharmaceutical Sciences—University of São Paulo. BRPE was collected in Canavieiras (Bahia) by the beekeeper’s association COAPER. Propolis samples were frozen and powdered. An hydroalcoholic solution (70% v/v) was added and submit to ultrasound (30 min) for extraction. The obtained extract was centrifuged (1500×g) and supernatant filtered (0.45 µm PTFE filter). Extraction process was repeated twice. BRPE was dried on a speed vacuum.^[33] The chromatographic profile of this extract is shown in Supplementary Material.

Encapsulation of Brazilian red propolis extract in polymeric nanocapsules

Nanoparticles were prepared by nanoprecipitation method, suitable for encapsulation of lipophilic compounds.^[13] The organic phase (acetone, ethanol, polylactic acid [PLA] [Purasorb^® PDL 02, Corbion]), Tributyrin (Sigma–Aldrich) and BRPE was poured into the aqueous phase (Poloxamer P407 [Sigma-Aldrich] and phosphate buffer [0.03 M, pH = 7.4]). The set was kept under magnetic agitation to complete organic solvent evaporation (24 h), obtaining BRPE loaded-nanocapsules (NC-BRPE).

Polymeric nanocapsule optimisation by Box Behnken design

To optimise the polymeric nanocapsules (NC), Box Behnken design (BBD), 3-factor in 3-level, low (−1), middle (0) and high (+1), was used. Independent variables chosen were polymer, surfactant and oil and dependent variables were hydrodynamic diameter (z-average) and polydispersity index (PdI). Fifteen formulations were suggested (software Minitab^®18) including a triplicate of the central point. Table 1 shows the independent variables and their limits.

Table 1

Independent variables and their limits

Independent variables	Limits
	Low (−1)	High (+1)
Polymer (mg)	30	120
Surfactant (%)	0.6	1.2
Oil (mg)	20	60

Dynamic light scattering analysis

z-Average, PdI and zeta potential of nanocapsules from BBD and optimised nanoparticles were measured by dynamic light scattering (DLS) (Zetasizer Nano ZS—Malvern Panalytical). Samples were dilute in KCl (1 mm). Stability of the optimised NC and NC-BRPE, stored at 4 °C, was analysed over 90 days.

Encapsulation efficiency of biomarkers of BRPE

A validated ultra-performance liquid chromatography coupled to mass spectrometry (UPLC-MS) method was used to quantify BRPE biomarkers vestitol (V), formononetin (F) and biochanin A (BA). Acquity UPLC H-Class (Waters), single quadrupole detector (SQ Detector 2, Waters), electrospray ionisation in the negative ion mode (ESI−) was employed. BEH C18 column (100 × 2.10 mm, 1.7 μm particle, Waters) at 40 °C was used. Mobile phase was composed by acetonitrile with 0.1% ammonium acetate (A) and ultrapure water with 0.1% ammonium acetate (B) at a flow rate of 0.4 ml/min in an elution by gradient (90% B [0 min]; 90% B → 30% B [2 min]; 30% B → 2% B [3 min]; 2% B → 2% B [6 min]; 2% B → 0% B [6.5 min]; 0% B → 0% B [8.5 min], 90% B → 10% B [9 min] and 90% B → 10% B). Mass parameters were: gas temperature: 450 °C, gas flow: 600 L/min, source temperature: 500 °C, capillary voltage: 5500 V and cone voltage: −44 V. Linear regression equations were obtained for each biomarker as follows:

$${\displaystyle \begin{array}{l}{\displaystyle V:y=87.57x-49.596,R{}^2=0.998}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle F:y=184831x+300703,R{}^2=0.996}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle BA:y=1181.9x+3029.3,R{}^2=0.995}\end{array}}$$

Encapsulation efficiency (EE) was determined by indirect method, in which 200 µl of formulation was placed in Microcon filters (Millipore^®) with ultrafiltration membrane (cutoff 10 000 g/mol) and centrifuged (10 min, 5000×g). Filtrate was diluted in methanol and biomarkers (BM) concentrations determined. Measurements were performed each 30 days, for 3 months. Equation (1) below was used to calculate EE.

$${\displaystyle <tex-math>}$$

(1)

Differential scanning calorimetry

Thermal behaviours of raw-materials (PLA, BRPE and Pluronic P407) and nanoparticles were analysed by differential scanning calorimetry (DSC; Shimadzu DSC-50). Nanoparticles previously lyophilised were hermetically sealed in an aluminium pan and heated (temperature range: 15–350 °C). Heating rate was 10 °C/min under nitrogen gas flow (3 kgf/cm²).

Nanoparticle tracking analysis

Size and nanoparticles concentration (NC-BRPE and NC) were defined by nanoparticle tracking analysis (NTA; NanoSight NS300—Malvern), red laser (642 nm) and software NTA 3.1. Nanoparticles were diluted in Milli Q water (1000×) before the analysis.

Transmission electronic microscopy

Morphology of NC-BRPE was observed through transmission electronic microscopy (TEM) (JEM—100 CXII) by negative staining method. Samples were diluted with water and stained with uranyl acetate solution (1% p/v) for 30 s. A drop of each sample was placed in copper grids, forming a thin liquid film that was dried previous to the analysis.

Cytotoxicity assay in ovarian cancer cell

Ovarian cancer cell line OVCAR-3 purchased from Rio de Janeiro Cell Bank was used in cytotoxicity studies. Cells were cultivated in complete RPMI-1640 medium supplemented (20% [v/v] foetal bovine serum, 1% [v/v] antibiotic solution [penicillin and streptomycin], 2% [m/v%] sodium bicarbonate, 4% [m/v] glucose, 10 mm HEPES and 1 mm pyruvate). Cells were incubated at 37 °C, 5% CO₂ and controlled humidity. After reach 80% confluency, cells were washed with PBS and trypsinised.

Cells (1 × 10⁵ cells/ml) were placed in a 96-well plate and incubated for 24 h in complete RPMI-1640 until the formation of a monolayer. Thus, medium was replaced for 100 µl of BRPE free or NC-BRPE at different concentrations (10–150 µg/ml of BRPE) and NC (4.5 × 10⁹ to 6.75 × 10¹⁰ NC/ml). All treatments were diluted in complete medium.

After 24 h of treatment incubation, cytotoxicity was evaluated by neutral red incorporation assay (NRA) (3-amino-7-dimethylamino-2-methylphenazine hydrochloride).^[34] Treatments were removed and cells washed with PBS 1×. So, a neutral red solution (4 µg/ml) was placed in all wells and incubated for 3 h. Absorbance was measured at 540 nm using a microplate reader (Biotek Synergy).

Cytotoxicity was also assessed by the resazurin reduction assay (RRA; 7-hydroxy-3H-phenoxazin-3-one-10-oxide).^{[35, 36]} The same conditions were followed as previously described for NRA as the number of cells seeded, concentrations of treatments tested and incubation time. A resazurin solution (25 µg/ml) was placed in all wells and incubated for 4 h. Fluorescence was measured (excitation/emission: 530/590 nm) using a microplate reader (Biotek Synergy). Results were expressed in cellular viability (%) and IC50 was defined by the method of dose-response (Prism 8 software).

3D cell culture assay

OVCAR-3 spheroids were cultivated in ultra-low attachment 96-well plates, by the method of forced floating as described by Amaral et al. (2017).^[37] Complete RPMI-1640 medium (200 µl), with 0.5% (v/v) methylcellulose and 2.5 × 10³ cell/well was added to the plates that were centrifuged (10 min at 300×g) and incubated for 72 h.

Then, spheroids were treated with different concentrations of free BRPE, NC-BRPE (66–320 µg/ml) and NC (2.27 × 10¹⁰ to 14.40 × 10¹⁰ NC/ml) for 24 h. Celltiter-Glo 3D Cell Viability solution (Promega) was added to each well (100 µl). Plates were shaken (5 min) and incubated at room temperature (25 min). Supernatants were transferred to opaque white plates and luminescence was read in a microplate reader (Biotek Synergy).

Statistical analyses

All studies were performed in triplicate and results expressed as mean standard deviation. Statistical significance was analysed by one-way ANOVA followed by Dunnett’s test. Differences were considered significant at P < 0.05. Linear regression was also used.

Results and Discussion

Polymeric nanocapsule optimisation by BBD

BBD method is a three-level design tool and can be applied to situations where there are three or more levels of interest.^[38] It is used to minimise the number of experiments and save raw-material, as well as time. Through different combinations of variables, it is possible to analyse which ones have more influence on the formulation development.^[39]

Previous to BBD a series of preliminary tests were performed to determine the independent factors and their range of concentrations (data not shown).^[40] Fifteen formulations, including a triplicate of the central point given by BBD, were produced with different amounts of polymer, oil and surfactant. Size and polydispersity index obtained are shown in Table 2.

Table 2

Independent variables and obtained responses for dependent variables in BBD

Run	Polymer (mg)	Oil (mg)	Surfactant (%)	Size (nm)	PdI
1	120	20	0.9	227.9	0.069
2	75	60	1.2	218.2	0.074
31	75	40	0.9	216.2	0.046
4	30	40	0.6	170.7	0.041
5	30	20	0.9	163.3	0.056
61	75	40	0.9	205.5	0.025
7	120	60	0.9	240.8	0.063
8	75	20	1.2	202.6	0.077
9	75	60	0.6	211.7	0.076
10	120	40	1.2	244.0	0.036
11	30	60	0.9	175.7	0.068
12	75	20	0.6	206.8	0.081
13	120	40	0.6	246.1	0.046
141	75	40	0.9	219.7	0.102
15	30	40	1.2	171.1	0.102

¹Central points.

Pareto and surface charts (Figures 1 and 2) show the influence of each component on the dependent variables. According to pareto charts (Figure 1) only the concentration of polymer influenced on the size of nanoparticles and none of the parameters influenced on PdI.

Figure 1

Pareto charts of the main effects of independent variables on dependent variables on BBD. The vertical line indicates the statistical significance (95% confidence level).

Figure 2

Response surface plot presenting the interaction between the: (A) Polymer PLA and Surfactant Poloxamer P407; (B) oil Tributyrin and Surfactant Poloxamer P407 and (C) oil Tributyrin and Polymer PLA.

It can be noted that the greater the amount of polymer, the greater is the size of the nanoparticles (Figure 2A and C). This was possibly caused by the increased viscosity of the dispersion (organic phase) and consequent decrease of PLA diffusion into the aqueous phase.^{[41, 42]} When the amount of polymer increased from 30 mg (run 4, Table 2) to 120 mg (run 13, Table 2), the nanoparticles’ size increased 75.4 nm.

Nanoparticles’ size also increases very slightly when the amount of oil increases (Figure 2B and C), which can be seen comparing runs 2 and 8 (Table 2). This behaviour was also observed by other researchers and can also be attributed to the increased viscosity of the organic phase.^{[43, 44]}

Surfactant concentration also did not have a great impact on particle size. By the comparison of runs 4 and 15 (Table 2), in which the surfactant concentration doubled from 0.6 to 1.2%, the particle size practically remained unchanged (170.7–171.1 nm).

Although the concentrations of oil and surfactant have not had greatly impact on the nanoparticle size, intermediary concentrations of them combined caused a little diminution of this feature (Figure 2B).

The procedure used to optimise multiple responses simultaneously is to consider each individual response into one global function. This is achieved by defining, a priori, the desirability function, according to what is considered inadequate or suitable for each analytical response.^[45] Due to the capacity of nanoparticles with sizes below 200 nm of being trapped inside of the tumour, owed to the permeation and retention effect (EPR), which is a passive targeting mechanism,^[46] size of the nanoparticles was set at 200 nm. Furthermore, due to the influence of the nanoparticles’ size on their biodistribution, minimum PdI was also set in BBD to guarantee the homogeneity of particles and improve the formulation stability. The effect of independent variables on the particle size can be explained by the model proposed:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{S}\mathrm{i}\mathrm{z}\mathrm{e}\left(\mathrm{n}\mathrm{m}\right)=24.2+1.016X1+2.98X2+104X3-0.00139X12}\\ -0.0218X22-29.8X32+0.00014X1X2-0.046X1X3-0.850X2X3\end{array}}$$

Where: X₁ = PLA, X₂ = Tributyrin and X₃ = Poloxamer 407.

Coefficients of determination, adjustability and variation evaluate the adequacy of the proposed model. The determination coefficient R² (0.971) of the quadratic regression model was high and indicates that only 0.0286 of the total variations were not explained by the model. R² adjusted coefficient was also high (0.920), confirming that the model was significant. A small difference was observed between determination R² and R² adjusted, indicating that the sampling number and the variables correlated were appropriate. Deviations between experimental and predicted values were low, which reflects the low coefficient of variation (0.184) and also that the experiments had high levels of reliability.^[47]

After nanocapsules optimisation, three batches were prepared according to the predicted for each compound (PLA, tributyrin and Poloxamer P407) obtaining values of size and PdI close to the optimised, with 220.0 ± 1.0 nm, PdI of 0.109 ± 0.03 and monomodal size distribution profile (Figure 3). These results indicate the reliability of the model in predicting the concentrations of the compounds.

Figure 3

Size distribution profile of the optimised nanocapsules (NC) at day 0 (red curve) and 90th after preparation (green curve).

The stability of this optimised formulation was evaluated for 90 days. Over this period, no significant difference (P > 0.05) was observed in size and PdI of the nanocapsules, and size distribution profile was similar to the day of the nanoparticle’s preparation (Figure 3). Stability of nanoparticles suspensions can be provided by different mechanisms. Electronic stability is conferred by ions that arrange around the nanoparticles forming an electric double layer that balances the attraction forces between the particles. Steric stabilisation is provided by polymers that coat the nanoparticles surface preventing their approach.^[48] Nanoparticles here described, presented electrosteric stabilisation (zeta potential of −8.6 ± 0.7 mV) in addition to steric stabilisation provided by the polymeric surfactant used, Poloxamer P407. The two mechanisms combined prevent sedimentation and agglomeration of the nanoparticles improving their stability.

BRPE was encapsulated in the optimised nanocapsules. BRPE encapsulation decreased the particle size to 177.8 ± 3.3 nm and improved the homogeneity, decreasing the PdI value (0.06 ± 0.03). This size reduction might be related to amphiphilic compounds present in BRPE that can act as surfactants, as described for other plant extracts.^[49]  Figure 4 shows the size distribution profile for day 0 and day 90th after nanoparticles preparation. Zeta potential was negative (–7.7 ± 0.3 mV) with a value similar to NC. NC-BRPE showed a narrow and monomodal size distribution profile that was not changed even after 90 days.

Figure 4

Size distribution profile of the optimised nanocapsules with Brazilian red propolis extract (NC-BRPE) at day 0 (red curve) and 90th after preparation (green curve).

Encapsulation efficiency

BRPE shows a complex and exclusive phytochemical composition including isoflavonoids (e.g. formononetin, biochanin A), isoflavones (e.g. vestitol, neovestitol) and polyprenylated benzophenones (gutifferone E and oblongifolin A) among other compounds.^[50]

Biochanin A (4ʹ-methoxy-5, 7-dihydroxy isoflavone) is a multifunctional substance with activity such as anti-inflammatory, anticancer, antioxidant, antimicrobial, neuro and hepatoprotective. As an anticancer element, biochanin A is noted for its capability to prevent cellular growth, angiogenesis and necrosis, activate cell apoptosis and arrest cell cycle.^[51] It is also considered a phytoestrogen and along with other substances within this classification, presents high lipophilicity with a logP: 3.14.^[52]

Vestitol (3-(2-hydroxy-4-methoxyphenyl)-3,4-dihydro-2H-chromen-7-ol) is an isoflavonoid of the class of hydroxyisoflavones, that also possesses various functions including anticancer, antibiotic and immunomodulatory activity. It has proven to show promising inhibitory activity against neutrophil migration.^{[52, 53]} It also presents a lipophilic nature (logP: 2.44).^[54]

Formononetin (7-hydroxy, 4ʹ-methoxy isoflavone) is a highly relevant isoflavone due to its properties such as antioxidant, anti-inflammatory, neuroprotective and osteogenic. It is also known as a potent phytoestrogen, able to bind oestrogen receptors, functioning as an agonist. It exhibits anticancer effects in several cancer lineages including breast, lung and prostate cells.^{[55, 56]} It presents a lipophilic nature too (logP: 2.96).^[57]

For the reasons presented above, these substances were chosen as biomarkers to the calculus of EE. Nanocapsules presented an EE of 97.56% ± 2.57 of biochanin A, 99.67% ± 0.07 of vestitol and 99.7% ± 0.32 of formononetin. The elevated EE can be explained by the fact that all the biomarkers are lipophilic and have high affinity for the nanocapsules’ core constituted by the oil tributyrin (logP: 2.4), a prodrug of butyrate.^[58]

A slight reduction of EE was verified after 90 days of nanocapsules preparation, but this decrease was not statistically significant (P > 0.05).

Transmission electronic microscopy

TEM is a useful technique for the characterisation of shape and size of polymeric nanoparticles.^[59]  Figure 5 shows the spherical morphology of NC-BRPE. Size corresponds to expected, with an average of 188.6 nm ± 12.7 (n = 256). The core–shell structure of the nanocapsules is also in display.

Figure 5

TEM images of the nanocapsules loading Brazilian red propolis extract (NC-BRPE).

Core–shell nanoparticles are interesting structures that have two different materials composing the inside and out.^[60] In this case, the nanoparticles are formed by an oily lipophilic core of tributyrin, in which BRPE is dispersed, limited by a polymeric shell of PLA and surfactant. Core–shell nanoparticles present advantages over nanospheres such as improvement in dispersibility of the active compound, that was crucial to the encapsulation of a high amount of BRPE and decreased toxicity, once the outside polymer layer is nontoxic and biocompatible.^[61]

Differential scanning calorimetry

Figure 6A shows the thermograms of NC and NC-BRPE and also the raw-materials PLA and Poloxamer 407.

Figure 6

Thermograms of: (A) NC, NC-BRPE, Poloxamer 407 and PLA, (B) BRPE.

Enantiomerically pure PLA is a semicrystalline polymer with a glass transition temperature (T_g) of about 55 °C, a melting temperature (T_m) of about 180 °C and a crystallisation temperature (T_c) of 149.5 °C.^[62] The presence of meso and d-lactide enantiomers causes imperfections in the crystal structure, reducing the percentage of crystallinity.^[63] PLA₅₀ made of 50% of each enantiomer (d, l lactic acid), is amorphous and exhibits T_g of approximately 40 °C and is in a glassy state at ambient temperature. The experimental T_g found for PLA was 37.22 °C, which is below standard T_g, indicating the polymer is in semicrystalline state. The T_g of PLA was not observed in the NC and NC-BRPE samples. This T_g suppression can be related to the plasticising effect of the surfactant used in the formulation. Plasticisers are non-volatile compounds that are incorporated into polymer solutions to improve mechanical properties, lowering viscosity and T_g, without modifying fundamental properties of the plasticised material. Decrease or suppression of the T_g is reached by the increased volume between the polymer chains provided by the plasticising compound, which allows more movement.^[64] In addition the plasticising effect might have contributed to the disorganisation in polymer chains and to a better encapsulation of BRPE with less expulsion of it from inside of the nanocapsules. Similar result was verified by Miranda et al. (2019).^[49]

T_m of pure Poloxamer was 57.65 °C; however, this endothermic peak was shifted to 49.95 in NC-BRPE sample and 52.66 in NC sample, what can indicate interaction between the components of the nanocapsules.^[65]

BRPE shows four endothermic peaks at 106.39 °C, 108.94 °C, 132.09 °C and 135.47 °C (Figure 6B). The first two peaks are related to the volatilisation water present in the extract and the last two refer to the melting processes of low molecular weight compounds, such as flavonoids and other phenolic compounds.^[66] In NC-BRPE thermogram, these peaks do not appear, suggesting that BRPE is molecularly dispersed in the nanocapsules.

Nanoparticle tracking analysis

Nanoparticles’ properties like size, surface area, charge, shape, hydro/lipophilicity and composition, profoundly influence their relationship with the environment and consequently their biodistribution.^[67] Therefore, the use of different techniques to determine size is necessary to guarantee the reliability of the results. Among the most used techniques for this purpose are DLS and NTA. DLS measures the scattered light intensity fluctuations caused by nanoparticles in Brownian motion.^[68] Larger particle sizes or aggregates impact the mean size causing an imprecision in the determination of size. This effect is more significant in polydisperse dispersions, underestimating the proportion of small particles in the system.^[69] NTA compensates these disadvantages of DLS because it analyses particle by particle at high resolution, simultaneously detecting particles of different sizes, which reflects in a more accurate analysis.^[70] In addition, by NTA, it is possible to obtain the nanoparticles concentration in the sample that is a potential parameter to fill the gap between in vitro characterisation and biological performance of colloids. Data of nanoparticles suspension concentration have guided different biological studies, from cellular viability to in vivo efficacy, both in pharmaceutical and environmental fields.^[71]

Figure 7 shows the size distribution profile versus concentration of nanoparticles in the suspension of both NC and NC-BRPE. In NC (Figure 6A) there was a lower concentration of nanoparticles (6.8 × 10¹¹ NC/ml) with size of 190.9 ± 9.1 nm and a span index of 0.77. As verified by DLS, the encapsulation of BRPE decreased the nanoparticles size and produced a more homogeneous formulation (Figure 7B), with size of 166.6 ± 3.7 nm, a span index of 0.38 and particle concentration of 9.57 × 10¹¹ NC-BRPE/ml.

Figure 7

Nanoparticle size distribution profile versus nanoparticle concentration of (A) NC, (B) NC-BRPE.

Cytotoxicity assay

Nanoparticles can improve the apparent solubility of natural products such as BRPE, which is a lipophilic extract and also enhance its entrance into the cells due to nanoparticles size and surface properties, improving its biological effects (Saraf et al. 2015). To evaluate the biological effect of nanoparticles on cells it is possible to use different assays including NRA and RRA.^{[72, 73]} In addition, it is necessary to select adequate controls and use more than one method to evaluate the cytotoxicity of nanostructured formulations to guarantee the reliability of the results, since nanoparticles or the carried drugs can interfere in some biological assays.^[74]

Figure 8 shows the cytotoxicity effect of BRPE free, NC-BRPE and NC on the ovarian cancer cell line OVCAR-3 evaluated by NRA. Dose-dependent cytotoxicity of BRPE free or NC-BRPE was observed in Figure 8A. Furthermore, the encapsulation of BRPE enhanced its cytotoxic effect. At a BRPE concentration of 50 µg/ml, the viability of cells treated with NC-BRPE was 48%, while the cells treated with free BRPE, at the same concentration, showed 85% of survival. The IC50 of NC-BRPE was ~2-fold lower (54.8 µg/ml, 2.6 × 10¹⁰ NC-BRPE/ml) than free BRPE (99.7 µg/ml). NC did not exhibit cytotoxicity effect on OVCAR-3 cells, until the highest concentration evaluated. Indicating that the cytotoxicity effect of NC-BRPE is due to the encapsulated extract (Figure 8B).

Figure 8

(A) Cytotoxicity of Brazilian red propolis (BRPE) free and encapsulated in nanoparticles (NC-BRPE), (B) Cytotoxicity of nanoparticles without BRPE (NC) measured by neutral red assay (NRA). *P < 0.05 compared with negative controls; ° P < 0.05 compared with BRPE.

Figure 9A shows the cytotoxicity effect of BRPE free and NC-BRPE on OVCAR-3 measured by RRA. Dose-dependent cytotoxicity of BRPE free or encapsulated was confirmed by this assay. Furthermore, along with NRA results, BRPE encapsulation increased its cytotoxicity exhibiting an IC50 ~ 1.5-fold lower (29.24 µg/ml, 1.39 × 10¹⁰ NC-BRPE/ml) than free BRPE (41.76 µg/ml).

Figure 9

(A) Cytotoxicity of Brazilian red propolis (BRPE) free and encapsulated in nanoparticles (NC-BRPE), (B) Cytotoxicity of nanoparticles without BRPE (NC) measured by resazurin reduction assay (RRA). *P < 0.05 compared with negative controls. ° P < 0.05 compared with free BRPE.

A significant cell viability reduction (P < 0.05) was observed for NC, yet the cell viability was maintained at around 75% until the highest concentration in NRA assay (Figure 8B) and until the concentration of 5.64 × 10¹⁰ NC/ml in RRA assay (Figure 9B). According to ISO 10993-5 guidance, if cell viability for the highest concentration of the sample is 70% of the control group, the material shall be considered non-cytotoxic. Thus, only at the highest concentration, in RRA assay, NC exhibited some toxicity, reducing cell viability to 63%. In literature, some papers describe the cytotoxicity of tributyrin on cancer cells.^{[58, 75]} Nevertheless, when we tested tributyrin on OVCAR-3 cells, we did not verify cytotoxicity (results not shown). However, the nanostructure (NC) in which the oil is contained may facilitate its entrance into the cell, allowing its toxic effect and reducing the cell viability in all concentrations tested.

The two different cytotoxicity methods (NRA and RRA) differ in the manner the cells interact with the dye. NRA measures uptake of the dye into lysosomes, a process that requires energy, so only living cells incorporate neutral red dye (NR) that concentrates in organelles with acidic pHs, such as lysosomes. For this reason, substances that damage lysosomes membranes or interfere with endocytosis will decrease cell capacity of absorbing NR.^{[76, 77]} By the results obtained for this test, it is possible to suggest that BRPE does not interfere or interfere less on these cells characteristics (uptake by lysosomes and endocytosis), as IC50 was twice as big in comparison with resazurin results.

In RRA, the dye is reduced by reductase enzymes present in cytosol and inside of mitochondria, so this assay measures the metabolic activity of living cells.^{[77, 78]} The IC50 results of RRA for both, BRPE and NC-BRPE, were lower in comparison with the results obtained by NRA. Taking into account the mechanism used by the cells to reduce resazurin, it is possible that BRPE may interfere/interact with the structure responsible for this conversion, once resazurin metabolisation was down-regulated. Other researchers have already determined that BRPE can decrease mitochondrial membrane potential, what is a potential damage in this structure and can also explain the decreased conversion of resazurin to resofurin.^[23] The same effect is valid for NC, once it has demonstrated some cytotoxicity in the RRA assay. Tributyrin contained in NC has also the capacity of changing the mitochondrial membrane potential,^[75] therefore it has more impact on this structure what can be detected by the RRA test. Thus, for this reason, some toxicity of NC can be verified in OVCAR-3 cells.

3D cell culture assay

In corroboration with the results from cytotoxicity 2D, NC-BRPE showed more efficacy against OVCAR-3 cells than free BRPE in 3D assay (Figure 10A). Dose-dependent cytotoxicity of BRPE free or encapsulated was also observed. According to IC50_, NC-BRPE was ~1.3-fold more efficient (137.2 µg/ml, 6.5 × 10¹⁰ NC-BRPE/ml) than free BRPE (175.1 µg/ml). As observed in the 2D assay, there was a cell viability reduction in all concentrations, except in the first one, for cells treated with NC. However, cell viability was maintained at around 75%. This reduction can be related to the effect of tributyrin, as discussed above.

Figure 10

(A) Cytotoxicity of Brazilian red propolis (BRPE) free and encapsulated in nanoparticles (NC-BRPE) in spheroids of OVCAR-3, (B) Cytotoxicity of polymeric nanocapsule without BRPE (NC) in spheroids of OVCAR-3. *P < 0.05 compared with negative controls; ° P < 0.05 compared with free BRPE.

3D assays mimic better the interactions between cells and cells and the environment, as they occur in tissues or tumours. In this configuration, cells form various layers and the access to nutrients and oxygen differs, according to the deepness in which it is encountered, but cells morphology and polarity remain the same. In 3D conformation, cells are less sensitive to active compounds, which may be caused by the reduced access to them in the medium or by pathophysiological alterations owed to hypoxia and even by cell cycle changes. Thus, this model is very relevant to screen new formulations and drugs before the in vivo experiments to confirm their antitumour potential.^{[37, 79]} Also, because of this 3D structure, the IC50 value is higher than IC50 value of 2D cell culture.^[80] Due to the small size and physical characteristics of nanoparticles, they can permeate better into the spheroid^[81] in comparison with free BRPE, what can also explain the best cytotoxic effect of NC-BRPE in this test.

Figure 11 shows the spheroids of OVCAR-3 obtained. The average diameter of negative control spheroids, without any treatment, was 287.5 ± 16.3 µm, which is a size similar to found in other works in the literature.^{[82, 83]} Spheroids of ovarian cancer cells from malignant ascites in vivo, showed a range of sizes between 60 and 400 µm.^[84]

Figure 11

Spheroids of OVCAR-3 (*) negative control; (A) Brazilian red propolis (BRPE) treatments: 1: 66 µg/ml, 2: 132 µg/ml, 3: 180 µg/ml, 4: 230 µg/ml, 5: 280 µg/ml, 6: 320 µg/ml. (B) Brazilian red propolis encapsulated (NC-BRPE) treatments: 1: 66 µg/ml (3.1 × 10¹⁰) NC-BRPE/ml, 2: 132 µg/ml (6.2 × 10¹⁰) NC-BRPE/ml, 3: 180 µg/ml (8.4 × 10¹⁰) NC-BRPE/ml, 4: 230 µg/ml (10.8 × 10¹⁰) NC-BRPE/ml, 5: 280 µg/ml (13.1 × 10¹⁰) NC-BRPE/ml, 6: 320 µg/ml (15.0 × 10¹⁰) NC-BRPE/ml.

After treatment with NC-BRPE (Figure 11B), spheroids disintegrated, consequently, the cells lost the capacity of being compacted together. This characteristic can be beneficial in the actual treatment, once the degeneration of the spheroid structure allows the entrance of nanoparticles and therefore BRPE within the spaces, with increases the contact area among cancer cells and treatment. In this scenario, NC-BRPE have the small size advantage and can permeate deeper into the tumour.

Conclusion

From the experimental design to the characterisation technics, it is possible to infer that the physicochemical characteristics of the nanoparticles produced in this work, such as their size and shape, are in the range of values indicated for passive delivery of the extract into the tumour by the EPR effect. Furthermore, we could demonstrate that the main biomarkers of BRPE were encapsulated in the nanocapsule and that NC-BRPE enhanced the antitumour activity of propolis in ovarian cancer cells, both in 2D and 3D models. So, NC-BRPE is an interesting formulation that could be an alternative or additional option for the treatment of OC.

Author Contributions

Isabela Araújo Justino carried out all the experiments, drafted the manuscript, analysed the data and interpreted results. Andréia Marincek contributed to the experiments concerning Box Behnken and their data interpretation. Iasmin R. S. Ferreira and Robson L. F. Amaral contributed to experiments concerning biological tests 2D and 3D and their data interpretation. Bianca B. Fontanezi developed the UPLC-MS method for BRPE quantification. Jairo K. Bastos and Jennyfer A. Aldana-Mejía harvested, processed and provided the BRP extract for this study. Priscyla D. Marcato supervised the study, contributed to the drafting and revision of the manuscript and also all data interpretation. All authors read and approved the final version.

Funding

This study was supported by the São Paulo Research Foundation (FAPESP; grants #2018/13465-5 and #2017/04138-8, #2016/16921-6). This study is part of the National Institute of Science and Technology in Pharmaceutical Nanotechnology: a transdisciplinary approach INCT-NANOFARMA (FAPESP grant #2014/50928-2 and CNPq grant# 465687/2014-8)

Conflict of Interest

None declared.

Ethical Statement

No ethical approval applies to this paper.

Data Availability

The data that support the findings of this study are available from the corresponding author (Marcato et al.), upon reasonable request