Empty nano and micro‐structured lipid carriers of virgin coconut oil for skin moisturisation

Abstract

Virgin coconut oil (VCO) is the finest grade of coconut oil, rich in phenolic content, antioxidant activity and contains medium chain triglycerides (MCTs). In this work formulation, characterisation and penetration of VCO‐solid lipid particles (VCO‐SLP) have been studied. VCO‐SLP were prepared using ultrasonication of molten stearic acid and VCO in an aqueous solution. The electron microscopy imaging revealed that VCO‐SLP were solid and spherical in shape. Ultrasonication was performed at several power intensities which resulted in particle sizes of VCO‐SLP ranged from 0.608 ± 0.002 µm to 44.265 ± 1.870 µm. The particle size was directly proportional to the applied power intensity of ultrasonication. The zeta potential values of the particles were from −43.2 ± 0.28 mV to −47.5 ± 0.42 mV showing good stability. The cumulative permeation for the smallest sized VCO‐SLP (0.608 µm) was 3.83 ± 0.01 µg/cm² whereas for larger carriers it was reduced (3.59 ± 0.02 µg/cm²). It is concluded that SLP have the potential to be exploited as a micro/nano scale cosmeceutical carrying vehicle for improved dermal delivery of VCO.

1 Introduction

Proper cleaning and moisturising are considered as basic processes to keep the (human) skin in good condition [1, 2]. The aim of moisturising is to restore skin to its natural protective condition. Moisturisers are cosmeceuticals that could penetrate and visibly alter the skin structure and function. They are topically applied compounds that consist of multiple components including occlusive ingredients, emollients and humectants [3, 4]. Occlusive (moisturising) ingredients are oily substances which retain the skin moisture by forming epicutaneous greasy film that acts as a physical barrier to prevent water loss. The skin hydration will be increased due to the decline in evaporation.

In the stratum corneum (SC), there are no phospholipids. The phospholipids from the keratinocytes of the viable layers are broken down by phospholipases in the lower SC [5, 6, 7]. This results in the biosynthesis of fatty acids, which are not only necessary for the development of a functional SC barrier but also plays a vital role in making the pH acidic [5]. The entrapment of moisture in SC is good enough to keep the skin soft, supple and flexible. Fatty acids help in retaining the moisture level of the skin. The enzymatic breakdown of triglycerides by indigenous microbes in the skin also adds to the free fatty acid accumulation [8]. Solid lipid nanoparticles (SLN) system was developed in the cosmetic industries to deliver lipophilic bioactive compounds such as grapeseed oil and manuka oil [9, 10]. SLN are characterised as lipid particles in a solid physical state in the sub‐micron and predominantly nanometer size range. The solid particles that have a mean size of 50–1000 nm conserve occlusive effects [11, 12]. They also exhibit excellent hydration and controlled (prolonged) release properties [13]. Solid lipid particles (SLP) are becoming more popular due to their ease in characterisation and for the fact that they contain a high range of well‐defined (tolerated) surfactant molecules. Moreover, they can be developed for several administration routes. SLP system can be defined as the encapsulation technique of the lipophilic active ingredient within a solid lipid matrix [14, 15]. The encapsulation strategy yields SLN or solid lipid microparticles depending upon the acquired size. SLP could also find applications in a number of unmapped terrain such as delivery of drugs against virulent diseases. Besides, they are also used in cosmetic industries [16].

VCO is the highest quality coconut oil that can be obtained from coconut fruit. VCO shares the similar chemical properties as coconut oil with added benefits, that is, higher phenolic content and antioxidant activity. Ferulic acids and p‐coumaric are major substances that contribute to antioxidant properties of VCO [17]. It is already being extensively used in tropical areas as homeo medicine and is traditionally used to improve skin health [18]. However, there is a limited literature available on the performance of VCO exploited as a topical cosmetic product. Since the topical administration of bioactive compounds often results in poor absorption and limited bioavailability therefore studies on VCO‐based topical cosmaceutical is necessary. The externally applied cosmaceuticals (drugs/skin care products) may be targeted either to remain at the epidermis, such as in parasitic treatments, or could penetrate the dermis and reach the blood stream. The aforementioned fates are dependent on the size of the particles of these chemicals and/or drugs since size plays a vital role in determining if the particles would be absorbed or remain superficial [19]. To ensure skin moisture to be long lasting, the active ingredients should remain at the epidermis layer. The skin layer (SC) represents the main physical barrier and the substance permeating across or diffusing through it, is the rate limiting step [20]. Any material entering the skin can either penetrate through the skin appendages (hair follicles, sweat glands) or through the SC and the underlying layers [21].

There are many cosmeceutical products that are incorporated with delivery systems to enhance the performance of the products [13, 16, 17]. SLP is one of the delivery systems that can be used in the cosmeceutical and pharmaceutical industries to enhance penetration and control the release of active ingredients [17, 18]. In this work, SLP were studied as potential carrier scaffolds for VCO. Particle size, zeta potential, entrapment efficiency (EE) and penetration study are investigated as indicators of product quality, stability and efficacy.

2 Materials and methods

2.1 Chemicals

Stearic acid (n‐octadecanoic acid) 95%, Soy lecithin, Tween™ 20, Tween™ 60, Tween™ 80, Sephadex G‐50 and Phosphotungstic acid, were obtained from Sigma‐Aldrich (Selangor, Malaysia). VCO was produced by Institute of Bioproduct Development (Universiti Teknologi Malaysia, Malaysia). Single‐use syringe filter (0.2 µm) were purchased from Sartorius, Malaysia. Distilled water was obtained on site from a Sartorius Arium 611 water system.

2.2 Preparation of VCO‐SLP

VCO‐SLP were prepared by the ultrasonication method based on hydrophilic‐lipophlic balance value. These values are related to their solubility both in oil and water [22]. The lipid phase, stearic acid (10%) and 5% (w/w) virgin coconut oil (VCO) was melted in a double boiled beaker and dispersed in a warm aqueous solution with the addition of emulsifier. The weight percentage of the surfactants used was 2.5% of the total VCO‐SLP formulation. The amount of surfactants blend used was 29 and 71% for Soy lecithin and Tween 80, respectively. A pre‐emulsion was obtained using a high speed stirrer (IKA Ultra Turrax^® T25) at 12,000 rpm for 2 min. The particle size was narrowed down using a probe sonicator (Fisher Scientific Sonic Dismembrator Model 500) at different power intensities (60%, 70%, 80% and 90%) while ultrasonication time was maintained at 180 s for each process. The emulsion was cooled at room temperature to obtain lipid particle dispersions. All samples were stored at 4°C in a refrigerator.

2.3 Measurement of particle size distribution

The VCO‐SLP samples were analysed for particle size distribution (volume weighted mean) measurement using Mastersizer 2000S (Malvern Instruments, UK). The samples were added to the sample dispersion unit and stirred to minimise interparticle interactions. The laser obscuration range was maintained between 10 and 20% [23].

2.4 Measurement of zeta potential

The zeta potential of the samples was measured using Zetasizer Nano Z (Malvern Instrument, UK). The samples were diluted with distilled water and the measurements were performed at 25°C.

2.5 Determination of VCO‐SLP morphology

The morphology of the VCO‐SLP was determined via imaging in a transmission electron microscope (TEM) (JEM 2100 TEM). A volume of 10 µl VCO‐SLP dispersion was placed on a 300 mesh copper (carbon‐coated) grid (Ted Pella, USA). The excess fluid was removed with a piece of filter paper [24]. After 5 minutes, a drop of 2% phosphotungstic acid was added to the grid for staining. The grid was dried at room temperature before mounting into the TEM for imaging.

2.6 Entrapment efficiency of VCO‐SLP

EE of VCO‐SLP samples was calculated by the method already reported by Zhang et al. [25], with some modifications. The entrapped and un‐entrapped VCO were separated using gel chromatography through a 10 mm × 130 mm column packed with Sephadex G‐50, at a flow rate of 2.0 ml/min. The VCO‐SLP suspension was diluted with distilled water in a ratio of 1:5. Then 1 ml of solution was pipetted into the column. The collected sample was diluted with ethanol (1:1) and sonicated for 20 min to break the particles and later on filtered through a 2 µm pore size filter.

The particles were evaluated by determining the amount of entrapped ferulic acid in SLP using HPLC (Waters, USA). The column used was Synergy hydro‐RP 80A (particle size: 250 mm × 4.60 mm × 4 µm) and it was thermostated to 25°C using a column temperature control module. For analysis, elution was carried out with water/acetonitrile/acetic acid (80:20:0.25 v/v) as the mobile phase. The solvent was allowed to flow at a rate of 1.5 ml/min. The concentration of ferulic acid in the VCO in the suspension (n ₁) and SLP (n ₂) were assayed by the HPLC detector at 321 nm (1). The percentage of EE was obtained from the following [25]

$$\mathrm{Entrapment}\mathrm{efficiency}(\text{\%},w/w)=\frac{n_2}{n_1}\times 100$$ (1)

where; n ₁ = total concentration of ferulic acid in the VCO (total amount of VCO in the starting solution), n ₂ = concentration of ferulic acid in encapsulated VCO.

2.7 Skin penetration study

University Kebangsan Malaysia Animal Ethics Committee approved the studies (experiment) done on animal skin. The experiments were conducted in an automatic transdermal diffusion system (PermeGear, Germany) that is composed of six horizontal diffusion cells, a thermally controlled circulating water bath and a magnetic stirrer. Abdominal rat skin samples (excised from rats aged ∼8 weeks) [26] were obtained for this study. The shaved skins were cut into squares and the subcutaneous fat was carefully removed. The prepared skin was pre‐equilibrated in PBS buffer at 4°C a day before the experiments. The skins were placed with the SC facing the donor cell. The effective diffusion area of the cell was 1.07 cm². The receiver compartments were filled with 2.5 ml while the donor compartments with 1.25 ml of fresh buffer (pH 7.4) (Fig. 1). The diffusion cells were equilibrated at 32 ± 2°C, corresponding to the temperature of the human skin surface using a re‐circulating water bath and the fluid in the donor and receptor compartments was stirred continuously at 300 rpm [27]. 1.25 ml samples were then pipetted into the donor compartment. The diffusion cells were covered with an aluminium foil to prevent light exposure.

Fig. 1.

Side‐by‐side diffusion cell

A sample of 1.0 ml was taken out at 0, 0.5, 1, 2, 4, 8, 14, 20, 26, 32, 38, and 44 h, respectively. Each time a sample aliquot was drawn from the receiver compartment, it was compensated by the addition of an equivalent volume of PBS buffer. The concentration of ferulic acid in receptor fluid was analysed using HPLC. Before analysis, the collected samples were extracted with ethanol (99%) in a ratio of 1:1 for 20 min in an ultrasonic bath (as described by Lai et al. [28]) in order to extract the oil. The extracted oil was filtered using 0.2 µm pore sized filter before it could be analysed.

Following the last receptor collection, the skin samples were removed from the diffusion cell. The ferulic acid content remaining in the skin (epidermis and dermis) was determined by washing the skin surface with ethanol and water followed by drying it with cotton wool. The skin was placed in a glass bottle containing 2 ml of ethanol and vortexed for 10 min. The bottle was placed in the sonication bath for 20 min for further extraction of VCO‐SLP. Finally, the solution was filtered and injected into HPLC.

Cumulative active ingredient penetration, Q_t /S (ug/cm²) of the VCO‐SLP was calculated using (2) [23, 24, 25] for 44 h per unit of skin surface area, S (1.07 cm²)

$$Q_t=V_r{C}_t+\sum _{i=0}^n{V}_s{C}_i$$ (2)

Here, C _t is the drug concentration [ug/ml] of the receiving solution at each sampling time, C _i the drug concentration [ug/ml] of the i th sample, and V _r and V _s are the volumes (ml) of the receiver solution and the sample, respectively.

Data was expressed as the cumulative drug permeation per unit of skin surface area, Q _t /S. The steady state fluxes, J _ss (ug/cm² h), were calculated using the linear regression interpolation of the experimental data at a steady state using Fick's Law [24] (3).

$$J_{ss}=\mathrm{\Delta }{Q}_t/(\mathrm{\Delta }t\times S)$$ (3)

The permeability coefficient (K _p) of active ingredients into the skin was calculated using (4)

$$K_{\mathrm{p}}=J_{\mathrm{ss}}/C_d$$ (4)

where; K _p = Permeability coefficient, cm/h, J _ss = Flux, (ug/cm² h), C _d = active ingredients concentration at the beginning in donor compartment, ug/cm³.

3 Results and discussion

3.1 Effect of size reduction processing parameters: intensity of ultrasonication process

VCO‐SLP were synthesised at various power amplitude of ultrasonication, that is, 60, 70, 80 and 90% while other parameters; 60 s homogenising time, 2.5% of total emulsifier, 5% of VCO, 10% of stearic acid and 82.5% of distilled water were kept constant. Particle size distribution (volume weighted mean), zeta potential and polydispersity index (PDI) analysis are shown in Table 1.

Table 1.

Ultrasonication intensity and particle size distribution of VCO‐SLP at 180 s (n = 3)

Sample	Power of ultrasonication, %	Vol. weighted mean [4, 3], µm	Zeta potential, mV	PDI
A	60	0.608 ± 0.002	−47.5 ± 0.42	0.276
B	70	3.071 ± 1.453	−44.3 ± 0.28	0.282
C	80	39.255 ± 1.654	−43.2 ± 0.28	0.303
D	90	44.265 ± 1.870	−45.5 ± 0.50	0.304

The results showed that particle size had a direct relationship with ultrasonicating power because the more the power, the larger the particle size. Table 1 shows that the particle size ranged inside the nanometer scale at 60% ultrasonicating power, but it gradually progressed to the micrometer (at 70% power) and multimicrometer size (at 80–90% power) with the increment in ultrasonicating power. Ultrasonication at 90% power amplitude resulted in the largest size distribution of the particles compared with the other intensities. The increase in power resulted in elevated temperature of the sonicator's probe which in turn increased the kinetic energy of the particles. The higher kinetic energy led to coalescence of the particles. The same phenomenon has been also described in a number of other articles [29, 30, 31].

Zeta potential values of ± 30 mV and greater indicate a stable formulation [32]. The zeta potential measurement returned that the VCO‐SLP were a stable formulation as the values of the samples were distributed between −43.2 to −47.5 mV. Ultrasonicating at 60% power amplitude produced the most stable particles, that is, lowest PDI and mean particle size. The PDI is used as a measure of the homogeneity distribution where PDI < 0.1 indicates a monodisperse population, while PDI > 0.3 indicates a higher heterogeneity [28, 33]. Table 1 shows that the samples had slightly non‐uniform distribution because of the fact that the particles coalesced due to high power amplitude [34].

3.2 Morphology of VCO‐SLP

The morphology of the VCO‐SLP was investigated using a TEM. Fig. 2 shows that the size of VCO‐SLP is between 300–600 nm at 60% ultrasonicating power. The particle appeared as a dark solid sphere (Fig. 2 : inset). The electron microscopy imaging confirmed the particle size already obtained by the volume weighted mean as shown in Table 1.

Fig. 2.

TEM image of the surface morphology of VCO‐SLP. The dark sphere represents the VCO encapsulated SLP. Inset shows a magnified version of VCO loaded SLP

3.3 Entrapment efficiency of VCO‐SLP

The effect of the proportional amount of VCO on the EE was examined by varying the ultrasonication power at 60, 70, 80 and 90%. This EE was determined after the separation of the free active ingredients using the mini column Sephadex G50. The results are expressed as percentile of active ingredients (Table 2). In this part of the experiment, ferulic acid was used as a marker compound for the VCO‐SLP. Table 2 summarises the EE of the VCO‐SLP prepared at different ultrasonicating intensities (A: 60%, B: 70%, C: 80% and D: 90%). The results of the EE of VCO‐SLP ranged from 98%–99% and there was no significant difference between them. This confirmed that our formulation was capable of entrapping VCO inside the SLP.

Table 2.

Results of EE% of the VCO‐SLP

Sample	Vol. weighted mean [4, 3], µm	EE, %
A	0.608 ± 0.002	99.97 ± 0.014
B	3.071 ± 1.453	99.95 ± 0.007
C	39.255 ± 1.654	99.47 ± 0.071
D	44.265 ± 1.870	98.93 ± 0.212

*A: 60%, B: 70%, C: 80% and D: 90% power amplitude

However, the EE decreased slightly when the particle size enlarged. Similar results were found in the work done by Yadav et al. [35] where the EE of DOX‐loaded HA‐PEG‐PLGA decreased from 94.36% to 88.74% when there was an increase in the particle size. In a similar study Yassin et al. [36] found that the smaller the particle size, the better the EE.

3.4 Penetration study of VCO‐SLP through rat skin

To study the effects of the VCO‐SLP application on skin, a penetration study was performed using rat skin as a barrier. The VCO‐SLP penetrating the rat skin was carried out using horizontal diffusion cells as shown in Table 3. VCO‐SLP samples of different sizes were selected for the penetration analysis (Sample A, 0.608 µm; B, 3.071 µm; C, 39.255 µm). These samples were used to evaluate the rate of penetration in a skin with 0.77 mm thickness. Fig. 3 shows the penetration profile of VCO‐SLP at different particle sizes through the rat skin. The initial concentration of ferulic acid (used to calculate the rate of penetration) for samples A, B and C were 0.182 mg/l, 0.186 mg/l and 0.182 mg/l, respectively.

Table 3.

Comparison of transport parameters for penetration and the amount of ferulic acid extracted (n = 2)

Sample with different particle sizes	Cumulative permeation Qt /S (µg/cm2)	J ss, µg/cm2 h	K p, × 10−1 cm/h	Amount of FA in the skin, µg
A (0.608 µm)	3.83 ± 0.01	4.96 × 10−1 ± 0.0006	1.09 ± 0.002	0.345
B (3.071 µm)	3.48 ± 0.20	4.89 × 10−1 ± 0.002	1.05 ± 0.007	0.341
C (39.255 µm)	3.59 ± 0.02	4.72 × 10−1 ± 0.001	1.04 ± 0.019	0.340

*FA = Ferulic acid

Fig. 3.

Plot of cumulative ferulic acid mass as a function of time

Fig. 3 illustrates the trend for flux at the steady state (J _ss) and permeation coefficient, (K _p) through rat skin, calculated and shown in Table 3. J _ss was calculated from the slopes of the linear portion. From the graph, the linear lines started from 0 hour to the 2nd hour. The particles of VCO‐SLP diffused faster at smaller particle size (Sample A) compared with the larger size (Samples B and C). This could be due to the occlusivity on the outer layer of SC and the rapid adsorption and permeation of the particles into the skin surface. It resulted in a higher thermodynamic activity gradient of the marker compound at the interface which is the driving force for permeation. This result supports the findings of a number of previously reported studies, that is [37, 38, 39, 40]. A maximum cumulative permeation was observed in sample A (Table 3). It had been proved that the VCO‐SLP with the lowest particle size resulted in the best relative permeation. The increase in particle size of VCO‐SLP decreases the permeation through rat skin (Fig. 3). The high permeation of sample A is related to the inherent characteristics of SLP, such as compatibility of the SLP lipid matrix with skin lipids, ‘occlusive effect’, and the small size of the particles that would enhance the ability of VCO‐SLP to penetrate the skin [11, 41].

Literature suggests that SLP with larger particles tend to aggregate at SC surface [42, 43]. The penetration appeared to accommodate the smaller particles, which would diffuse first, followed by the larger ones, and some of the larger particles would only be entrapped on the skin surface. Thus, it can be interpreted that the permeation of the drug across the skin is affected by the particle size. In this study, the particle size around 600 nm is the ideal size to be an occlusive agent. This extends what Du Plessis, Badran and Neubert [44, 45, 46] had already reported that an intermediate particle size of 300 nm resulted in both the highest reservoir in the deeper SCs as well as having the highest drug concentration in the reservoir.

Table 3 shows that the cumulative mass of ferulic acid in the receptor after 44 h for the lowest VCO‐SLP particle size was 0.345 µg (A), as compared with 0.341 µg (B) and 0.340 µg (C). The VCO‐SLP were retained in the receptor chamber although the sizes were quite large. Our results proved that sample A harbours more amount of ferulic acid in the skin than sample B and C. This is due to the smallest particle size (0.608 µm). This verified that VCO‐SLP sample with 0.608 µm size that was retained in the skin, is most suitable to be used as an occlusive ingredient.

4 Conclusion

VCO‐SLP prove to be one of the promising cosmeceutical carrying scaffolds for improved dermal delivery. The EE can be vastly improved especially when the size is kept at the nanometer scale.