Development of Bergamot nanoemulgel using central composite design and its evaluation for antimicrobial activity

Koushik Yetukuri; Marakanam Srinivasan Umashankar; S.T.V. Raghavamma

doi:10.7324/JAPS.2025.236869

Abstract

This study aims to develop sophisticated mathematical models to describe the behavior of a Bergamot oil (BO) nanoemulgel formulation and optimize its physicochemical and biological properties. Conducted in a controlled laboratory setting, the formulation parameters were optimized using a central composite design as part of a response surface methodology approach. The interactions between Tween 80 concentration and homogenization time were explored to achieve an ideal nanoemulsion. Differential scanning calorimetry and attenuated total reflectance-Fourier transform infrared analyses confirmed the successful encapsulation of BO and evaluated its thermal stability. Statistical evaluations through regression modeling and ANOVA indicated significant effects of formulation variables on particle size and zeta potential. The optimized nanoemulgel exhibited a particle size of 116.60 nm, a polydispersity index of 18.9%, and a zeta potential of −26.9 mV, with a high desirability score of 0.938. Kinetic stability studies demonstrated excellent thermodynamic resilience with no signs of phase separation. The antibacterial activity of the optimized BO nanoemulgel showed prominent zones of inhibition, measuring 23.9 ± 0.5 mm against Staphylococcus aureus [Microbial Type Culture Collection and Gene Bank (MTCC) 737], 22.98 ± 0.4 mm against Pseudomonas aeruginosa (MTCC 1035), 21.6 ± 0.3 mm against Bacillus subtilis (MTCC 441), and 20.1 ± 0.2 mm against Escherichia coli (MTCC 443), suggesting broad-spectrum antimicrobial potential. These findings reinforce the utility of BO nanoemulgel as a promising candidate for applications in pharmaceutical and personal care formulations, offering both enhanced drug delivery and effective antibacterial performance.

Keyword: Bergamot oil response surface methodology central composite design nanoemulgel antibacterial activity

Citation:

Umashankar MS, Yetukuri K, Raghavamma STV. Development of Bergamot nanoemulgel using central composite design and its evaluation for antimicrobial activity. J Appl Pharm Sci. 2025;15(11):123-135. http://doi.org/10.7324/JAPS.2025.236869

Copyright: © The Author(s). This is an open-access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

INTRODUCTION

Bergamot oil (BO) is a volatile essential oil extracted from the peel of the bergamot orange (Citrus bergamia), a fruit primarily cultivated in the Mediterranean region. Because of its refreshing citrus aroma, BO finds extensive use in aromatherapy, perfumery, and flavoring industries. It is rich in bioactive compounds, such as linalool, linalyl acetate, limonene, and bergapten, which contribute to its wide range of biological activities [1]. In traditional and alternative medicine, BO is employed for its mood-enhancing, anxiolytic, anti-inflammatory, and particularly antimicrobial effects. Several studies have demonstrated the broad-spectrum antimicrobial activity of bergamot essential oil, with proven efficacy against Gram-positive and Gram-negative bacteria, as well as fungi, supporting its potential in topical antimicrobial formulations [2]. Despite its therapeutic potential, the application of BO in pharmaceutical and cosmetic formulations faces several challenges. BO is highly susceptible to oxidative degradation, photoinstability, and evaporation losses, which compromise its efficacy and shelf life. Furthermore, direct application can cause skin irritation and photosensitivity due to the presence of furanocoumarins. Therefore, innovative delivery systems are required to enhance their stability, control their release, and improve dermal tolerability [3].

Nanoemulsions, characterized by their small droplet size (typically <200 nm), high kinetic stability, and improved solubilization capacity for lipophilic compounds, offer a promising strategy for the effective delivery of essential oils such as BO [4]. This method is especially effective in enhancing antimicrobial activity by improving the penetration of the active components into microbial cells [5]. Nanoemulsification not only protects volatile components from degradation but also enhances skin penetration and bioavailability, thereby maximizing therapeutic outcomes. The application of response surface methodology (RSM), particularly using a central composite design (CCD), allows systematic optimization of these critical factors by studying their individual and interactive effects on the desired responses [6,7]. This statistical approach reduces the number of experimental runs required while generating robust predictive models that facilitate formulation development under the quality-by-design framework [8]. Furthermore, converting a nanoemulsion into a nanoemulgel by incorporating a suitable gelling agent not only improves the ease of topical application but also enhances the residence time on the skin, providing sustained drug release and prolonged therapeutic effects. Nanoemulgels offer better patient compliance, controlled drug release, and improved site-specific delivery compared to conventional gels or creams. For antimicrobial assessment, the agar well diffusion method was selected due to its simplicity, reproducibility, and ability to visually compare the efficacy of different formulations.

Therefore, the present study aimed to develop, optimize, and characterize a BO-loaded nanoemulgel using an RSM-based CCD design for topical antibacterial applications. The study investigates the effects of formulation variables on critical quality attributes such as droplet size and polydispersity index (PDI) and evaluates the antibacterial efficacy of the optimized nanoemulgel against selected pathogenic bacterial strains. This approach seeks to establish a stable, efficient, and therapeutically effective formulation of BO for future pharmaceutical and cosmetic applications.

MATERIALS AND METHODS

Materials

BO was procured from Avi Naturals, Delhi, India. Tween 80, Carbopol 934, benzyl alcohol, and sodium hydroxide (NaOH) were purchased from Loba Chemie Pvt. Ltd., Mumbai, India. The bacterial strains Staphylococcus aureus [Microbial Type Culture Collection and Gene Bank (MTCC) 737], Pseudomonas aeruginosa (MTCC 1035), Bacillus subtilis (MTCC 441), and Escherichia coli (MTCC 443) were obtained from the Microbial Type Culture Collection and Gene Bank (MTCC), Chandigarh, India.

Preparation of nanoemulsion loaded with BO

The nanoemulsion formulation was prepared using the high-speed homogenization technique. Initially, the aqueous phase comprising distilled water (80%–90%) and Tween 80 (2%–6%) was prepared separately from the oil phase containing 10% BO. Both phases were brought to room temperature (25°C ± 2°C) before mixing. The oil phase was gradually added to the aqueous phase under continuous stirring using a magnetic stirrer. This gradual addition of the oil phase into the aqueous phase ensures proper dispersion and forms a pre-emulsion. The pre-emulsion was then subjected to high-speed homogenization at 12,000 rpm for 10 minutes using a high-speed homogenizer (IKA T25 Digital Ultra-Turrax). The homogenization process involves high shear forces that break down the oil droplets and uniformly disperse them within the aqueous phase, resulting in a stable nanoemulsion. Throughout the process, the temperature was maintained at ambient conditions (25°C ± 2°C) to prevent thermal degradation. This method ensures the formation of a stable nanoemulsion with a small droplet size and improved bioavailability of the BO [9].

Optimization of formulation using CCD

To optimize the BO nanoemulsion, a study involving 13 experimental runs was conducted using a 2-variable, 3-level CCD developed by DoE software (7.1.6 Version, Stat-Ease, Inc., Minneapolis, MN). Within this design matrix, the independent variables included the amount of the surfactant Tween 80 (X₁) and the homogenization time (X₂). These variables were coded as −1 (low) and +1 (high) to signify their respective values. The dependent variables in this study encompassed two key parameters: particle size (Y₁) and zeta potential (Y₂). The design matrix comprised 13 experimental runs as shown in Table 1, and the software derived nonlinear quadratic model equations to represent the experimental design [10].

Table 1. Different sets of experiments for 13 trials of formulations with coded and actual values.

[Click here to view]

$Y = A 0 + A 1 X_{1} + A 2 X_{2} + A 3 X_{1} X_{2} + A 4 X_{1}^{2} + A 5 X_{2}^{2}$

Where, Y = response and A0 = the intercept indicates the arithmetic combination of all 13 experiments. A1, A2, A3, A4, A5 = the regression coefficient represents the response variables of Y. The outcomes obtained from preparing the 13 formulations were meticulously recorded and integrated into the experimental design.

CHARACTERIZATION OF NANOEMULSION

Particle size and PDI

The BO nanoemulsion was evaluated for particle size and PDI analysis using Malvern Zetasizer Nano ZS 90 equipment (Malvern Instruments, United Kingdom) via photon correlation spectroscopy. Measurements were taken at a controlled temperature of 25°C in disposable polystyrene cells [11].

Zeta potential of nanoemulsion

The zeta potential of the prepared BO nanoemulsion was measured using photon correlation spectroscopy with a Malvern Zetasizer Nano ZS 90 instrument (Malvern Instruments, United Kingdom). The measurements were carried out at a constant temperature of 25°C using omega cuvettes [12].

Heating cooling cycle

The sample underwent storage at a high temperature of 40°C for a period of 48 hours, followed by subsequent incubation at an even higher temperature of 48°C for an additional 48 hours, with these three repeated heating-cooling cycles. Subsequently, samples of nanoemulsions were examined for an indication of phase separation or precipitation [13].

Freeze-thaw cycle

BO nanoemulsion was exposed to a rigorous freeze-thaw cycle, involving freezing at −20°C for 24 hours followed by thawing at approximately 25°C room temperature. This process was repeated three times to simulate potential temperature fluctuations encountered during storage or transportation. Subsequently, the nanoemulsion underwent centrifugation analysis to scrutinize for any indications of phase separation [14].

Stability index

The BO nanoemulsion underwent three successive freeze-thaw cycles to assess its stability under rigorous conditions. Subsequently, the stability index of the BO nanoemulsion was calculated through the provided formula [15].

$Stability index of NE = \frac{Original globule size - Change in globule size}{Original globule size} \times 100$

Kinetic stability

To assess the macroscopic stability properties, the optimized BO nanoemulsion underwent a centrifugation process. The final formulations were subjected to centrifugation at speeds of 1,000, 2,000, and 3,000 rpm for 15 minutes each. Following centrifugation, the kinetic stability of the BO nanoemulsion was assessed by comparing its characteristics before and after the centrifugation cycle [16].

Scanning electron microscopy analysis

Scanning electron microscopy (SEM) was utilized to investigate the surface morphology and structural features of the nanoemulsion droplets within the optimized nanoemulgel formulation. A small amount of the sample was carefully mounted onto an aluminum stub using a double-sided adhesive carbon tape. The sample was then air-dried under a vacuum to remove residual moisture and sputter-coated with a thin layer of gold using a sputter coater to enhance conductivity and prevent charging during imaging. SEM analysis was performed using a high-resolution microscope (Zeiss) at an accelerating voltage of 5 kV to achieve optimal imaging conditions. The nanoemulsion droplets were observed at various magnifications to assess their size, shape, and surface characteristics. The globules exhibited spherical morphology with smooth surfaces, indicating successful formulation of the nanoemulsion. Image analysis software was used to evaluate particle uniformity and size distribution, which supported the data obtained from particle size analysis and confirmed the structural stability of the formulation [17].

Preparation of BO nanoemulgel

The nanoemulgel was formulated by incorporating the optimized BO nanoemulsion into a gel base prepared using Carbopol 934. Initially, Carbopol 934 was dispersed in a specified quantity of distilled water under magnetic stirring until a homogenous gel base was formed, typically left to hydrate for 24 hours to ensure complete swelling. The pre-formulated nanoemulsion containing BO, Tween 80, and the aqueous phase was then slowly added dropwise into the hydrated Carbopol gel under continuous mechanical stirring to ensure uniform distribution of the oil droplets within the gel matrix. Benzyl alcohol (1% w/w) was incorporated as a preservative to prevent microbial growth. Finally, the pH of the nanoemulgel was adjusted to the skin-friendly range of 5.5–6.5 using 1 M NaOH dropwise, while monitoring with a calibrated pH meter. The resulting nanoemulgel was evaluated for homogeneity, consistency, and absence of phase separation before subjecting it to further characterization [18].

EVALUATION OF NANOEMULGEL

Viscosity

The viscosity of the optimized BO nanoemulgel formulation was measured with a Brookfield viscometer at room temperature. The ultra low adapter (ULA) S00 spindle was adjusted to 4 rpm, and the sample was placed in the ULA cylinder. A torque level of 10 was set during the viscosity measurement [19].

Spreadability

The spreadability of the BO nanoemulgel was assessed 72 hours post-formulation by analyzing the spread diameter of the gel between two glass plates for a duration of 1 minute. Initially, a 400 mg sample of nanoemulgel was carefully weighed and placed on a glass plate featuring a pre-marked 1 cm circle. Following this, a second glass plate was positioned over the sample. As pressure increased on the upper plate, the subsequent expansion of the gel’s diameter was observed and quantified using the provided equation [20].

$S = m \times \frac{l}{t}$

Whereas S = spreadability of BO nanoemulgel, m = weight of BO nanoemulgel placed on an upper glass plate (gm), 1 = length of an upper glass plate (cm), and T = time taken for BO nanoemulgel spreading (seconds).

Extrudability

The extrudability test was conducted using a hardness tester. A 5 g portion of the nanoemulsion gel was loaded into aluminum collapsible tubes, and a plunger was employed to secure the tube in place [21]. Subsequently, pressure equivalent to 1 g/cm² was applied to the tube for a duration of 30 seconds. The amount of nanoemulsion gel extruded from the tube under this pressure was measured. This cycle was repeated three times to ensure the reliability and consistency of results, allowing for a comprehensive assessment of the gel’s extrudability characteristics [22].

Swelling index

The swelling index of the prepared nanoemulsion gel was assessed through a systematic procedure. Initially, 1 g of the gel was placed on porous aluminum foil and then individually positioned in 50 ml beakers containing 10 ml of 0.1 N NaOH solution. Subsequently, the samples underwent a swelling process within the NaOH solution for varying time intervals. After each designated interval, the samples were carefully retrieved from the beakers and allowed to dry thoroughly. Upon complete drying, the samples were reweighed to ascertain their final weight post-swelling [23].The swelling index was determined through the following equation.

$Swelling Index (SW) % = \frac{[(Wt - Wo)]}{Wo} \times 100$

Where Wo = initial weight of the gel before swelling and Wt = final weight of the gel after swelling time t.

pH determination

The pH of the optimized formulation was analyzed with a calibrated digital pH meter 335 from Systronics Pvt. Ltd., India. The pH meter’s probe is immersed in the film-forming solution under examination, and the pH reading is recorded [24].

Attenuated total reflectance-Fourier transform infrared analysis

Attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy was employed to assess the compatibility between the active components (BO) and the excipients used in the optimized nanoemulgel formulation. This technique helps detect any significant physicochemical interactions that may occur during formulation. The infrared spectra of both pure BO and the BO-loaded nanoemulgel were recorded using an ATR-FTIR spectrophotometer (Bruker or equivalent model). A small quantity of each sample was directly placed on the ATR crystal, and spectra were scanned over a wavelength range of 4,000–400 cm?¹ at a resolution of 4 cm?¹. Each sample was scanned for 16 accumulations to improve the signal-to-noise ratio. The characteristic peaks corresponding to functional groups of the BO were compared with those in the nanoemulgel formulation. Shifts in peak positions and changes in intensity—particularly lower intensities—suggested successful encapsulation of the oil and confirmed the absence of major incompatibilities between the drug and excipients [25,26].

Differential scanning calorimetry analysis

Differential scanning calorimetry (DSC) was performed to evaluate the thermal behavior and physical interactions of the active constituents in pure BO and the BO-loaded nanoemulgel. The purpose of this analysis was to determine the melting point, assess the thermal stability, and detect any potential interactions between BO and excipients that might influence the formulation’s performance. The DSC thermograms were recorded over a temperature range of 0°C–300°C at a constant heating rate, typically 10°C/min, under a nitrogen atmosphere to prevent oxidative degradation [27–30].

Antimicrobial activity

The antimicrobial activity of the optimized BO nanoemulsion formulations was assessed using the agar well diffusion method on Mueller−Hinton agar, with all tests performed in triplicate [31]. To evaluate broad-spectrum antibacterial efficacy, four clinically relevant bacterial strains commonly associated with skin and wound infections were selected: S. aureus (MTCC 737) and B. subtilis (MTCC 441) representing Gram-positive bacteria and P. aeruginosa (MTCC 1035) and E. coli (MTCC 443) representing Gram-negative bacteria. These organisms were cultured in nutrient broth for 24 hours to achieve a turbidity of approximately 10? CFU/ml [31–33]. Each culture was then mixed with molten soft agar (maintained at 45°C–50°C), poured evenly over the surface of base agar plates, and allowed to solidify under laminar airflow. Wells of 6 mm diameter were created aseptically in the agar plates, and 0.15?ml of test formulations, dissolved in Dimethyl Sulfoxide (DMSO) at concentrations ranging from 0.020 to 10?mg/ml, were added into the wells. Ciprofloxacin was used as the positive control, while DMSO served as the negative control. Additionally, a blank nanoemulgel formulation containing all the excipients (Tween 80, Carbopol 934, and distilled water), but excluding BO was prepared and tested under identical conditions to determine any inherent antibacterial effects of the base formulation. Plates were allowed to stand for 15 minutes at room temperature before being incubated at 37°C for 24 hours. After incubation, the diameters of the zones of inhibition were measured using a digital antibiotic zone reader (Optics Technology, India) to quantify antibacterial activity.

RESULTS

RSM was employed to optimize the BO-loaded nanoemulsion, utilizing CCD for optimization of the BO nanoemulsion. Perturbation plots were utilized to ascertain which factors had the most significant impact on the responses. The influence of independent factors on the dependent variables (responses) was thoroughly assessed utilizing 3D response surface plots and contour plots. To validate the model fitting for the datasets, an ANOVA was conducted for individual responses using DoE software [34,35]. This statistical analysis confirmed the appropriateness of the model for representing the experimental data accurately. Thirteen formulations were meticulously prepared and input into the DoE software (Table 2). Among these formulations, the particle size (Y₁) of run 9 stood out as the smallest, measuring 100.7. Conversely, the particle size of run 1 was found to be the highest among all 13 formulations, measuring 240.7 nm. Regarding zeta potential (Y₂), run 1 displayed the lowest value at −12.8, while run 9 exhibited the highest value at −29.8.

Table 2. Design of experiment for CCD method with 13 trial runs with results of responses.

[Click here to view]

Analysis of the responses

Response Y₁ (particle size)

The examination of various statistical parameters, including p-value, sum of squares, mean squares, residual sum of squares, adequate precision, predicted R², F-value, adjusted R², and predicted residual error sum of square (PRESS) values [36], suggests that the quadratic model adequately fits the Y₁ response. The model F-value for Y₁, registering at 55.02, indicates the statistical significance of the model. Such a high F-value occurring with only a 0.01% probability due to random variation underscores the model’s validity. The presence of p-values < 0.0500 further confirms the statistical significance of model terms. According to this ANOVA analysis, it was determined that the independent factors influencing the Y₁ response are homogenization time and surfactant concentration. Additionally, the lack of fit F-value for the Y₁ response was calculated at 0.8698. By coded variables, the quadratic equation for the Y₁ response was given in the formula.

Y₁ = 158.96 − 28.54 X₁ − 40.55 X₂ − 10.02 X₁ X₂+ 10.44 X₁² + 8.64 X₂²

This equation outlines the quantitative impact of independent factors X₁ and X₂ on the response Y₂, where the main effects are identified as the primary influencers of the response [37]. Additionally, the interaction terms, including X₁, X₂, X₁X₂, X₁², and X₂², illustrate a non-linear relationship between the Y₁ response and the variables when changed simultaneously. The perturbation plot serves as a tool to evaluate the influence of various factors on the response. Concerning response Y₁, both factor X₁ and factor X₂ display shallow slopes with slight curvature, indicating that the concentration of surfactant and homogenization time are the most significant factors impacting the nanoemulsion. Increasing the concentration of these factors leads to a reduction in particle size (Fig. 1A). Moreover, the response surface plot depicts the relationship between the independent and dependent factors. Analysis of the interaction between factors X₁ and X₂ on particle size response indicates that both homogenization time and surfactant concentration interact with the particle size of the formulation. Elevating the concentration of both factors results in a decrease in the particle size of the formulation (Fig. 2A).

Figure 1. Perturbation plots showing the individual effect of independent variables on the responses: (A) The influence of Tween 80 concentration (X₁) and homogenization time (X₂) on particle size (Y₁). The curves illustrate that both variables exhibit moderate curvature, indicating their significant but non-linear influence on the particle size. (B) The influence of Tween 80 concentration (X₁) and homogenization time (X₂) on zeta potential (Y₂). Both factors exhibit a consistent negative slope, indicating an increase in the negative charge (zeta potential) with increasing levels of the variables.

[Click here to view]

Figure 2. 3D response surface plots showing the interaction effect of independent variables on the responses: (A) Interaction effect of Tween 80 concentration (X₁) and homogenization time (X₂) on particle size (Y₁). Lower particle size was observed at higher Tween 80 levels and intermediate homogenization time. (B) Interaction effect of Tween 80 concentration (X₁) and homogenization time (X₂) on zeta potential (Y₂). A more negative zeta potential was observed with increasing homogenization time and surfactant concentration, indicating enhanced stability of the formulation.

[Click here to view]

Response Y₂ (zetapotential)

The PRESS values support the suitability of the quadratic model for the Y₂ response. The F-value for Y₂, measuring 132.64, signifies the statistical significance. Additionally, p-values < 0.0500 suggest that the significance of model terms. According to the ANOVA analysis, homogenization time and surfactant concentration are identified as the independent variables influencing the Y₂ response. Furthermore, the lack of fit value for Y₂ was calculated at 0.4415. By coded variables the quadratic equation for the response Y₂was shown in the equation Y₂= −25.52 − 3.98 X₁− 4.81 X₂+ 0.6500 X₁ X₂+ 1.79 X₁² + 1.72 X₂²

This equation provides insights into the quantitative effects of the two independent factors, X₁ and X₂, on the response Y₂. Notably, the main effects of X₁ and X₂ are identified as primary influencers of the response. Additionally, the interaction terms, including X₁, X₂, X₁X₂, X₁², and X₂², highlight the non-linear bond between the response Y₂ and the variables when changed simultaneously. The negative sign associated with X₁X₂, X₁², and X₂² indicates an incompatible effect on the response Y₂. For the response Y₂, both Factor X₁ and Factor X₂ exhibit positive slopes, suggesting an interaction between the concentration of surfactant and homogenization time on the zeta potential. The positive slope indicates that increasing the concentration of these two variables leads to an interaction in the zeta potential as depicted in Figure 1B, the response surface plot illustrates the relationship between factors X₁ and X₂ concerning the zeta potential. The plot demonstrates that increasing the concentration of both factors results in significant changes in the zeta potential of the compound as observed in Figure 2B.

Overlay plot

The overlay plot suggested one optimal solution out of six, with a desirability score of 0.938. This solution was achieved by maintaining the BO concentration at 10%, the Tween 80 concentration at 50% (X₁), and a homogenization time of 27.1005 minutes (X₂). Under these conditions, the probability of obtaining a particle size of 109.31 nm (Y₁) and a zeta potential of −29.8 mV (Y₂) was high. The desirability score of 0.938 indicates that the formulation parameters were optimized to near-ideal conditions in terms of particle size and zeta potential. While this suggests promising in vitro effectiveness and stability of the formulation, these findings require further validation through in vivo studies to confirm the formulation’s safety and performance in clinical settings.