2.1.

Animals

Male Swiss mice, weighing 25 to 30 g, were kept in an environment under controlled temperature ($22°\mathrm{C}\pm 2°\mathrm{C}$) and a $12\mathrm{h}/12\mathrm{h}$ light/dark cycle with free access to a standard pellet diet and water. The experimental protocol was approved by the Ethics Committee for Animal Experimentation of the State University of Maringá (ECAE/UEM 045/2012).

2.2.

Fish Oil Preparation

The FO was commercially obtained from Naturalis^®. According to the manufacturer, each 500-mg capsule contains 250 mg DHA and 50 mg EPA. A previous study⁷ quantified the fatty acids that were present in the capsule, validating the aforementioned information from the manufacturer (data not shown).

The contents of the FO capsule were diluted in olive oil (OO) or in a solution of 70% acetone to obtain the fish oil preparations (${\mathrm{FOP}}_{\mathrm{OO}}$ and ${\mathrm{FOP}}_{\mathrm{AC}}$) in amounts of (i) 0.25 mg DHA and 0.05 mg EPA, (ii) 0.5 mg DHA and 0.1 mg EPA, (iii) 1 mg DHA and 0.2 mg EPA, (iv) 2 mg DHA and 0.4 mg EPA, and (v) 4 mg DHA and 0.8 mg EPA. These preparations were topically applied to the ears of different groups of animals. For simplicity, we refer to the preparations below according only to the amount of DHA in each preparation.

2.3.

Croton Oil-Induced Ear Edema

Ear edema was induced on the ventral side of the left ear by applying $20\mu \mathrm{l}$ of CO ($200\mu \mathrm{g}$) diluted in 70% acetone (vehicle), adapted from the technique of Schiantarelli et al.²² The right ear received only topical application of $20\mu \mathrm{l}$ of vehicle (70% acetone) as a noninflamed control or negative control. Immediately after CO application, the animals received a topical application of $20\mu \mathrm{l}$ of 70% acetone (inflamed control or positive control), ${\mathrm{FOP}}_{\mathrm{OO}}$ (0.25, 0.5, 1, 2, or $4\mathrm{mg}\mathrm{DHA}/\text{ear}$) or ${\mathrm{FOP}}_{\mathrm{AC}}$ (0.25, 0.5, 1, 2, or $4\mathrm{mg}\mathrm{DHA}/\text{ear}$). The other groups of animals received a topical application of $20\mu \mathrm{l}$ of OO, dexamethasone (Dexa; $0.1\mathrm{mg}/\text{ear}$, used as anti-inflammatory reference drug), or indomethacin (Indo; $1\mathrm{mg}/\mathrm{ear}$, used as anti-inflammatory reference drug) immediately after CO application. After 6 h, the animals were anesthetized and euthanized. The ears were sectioned into 6.0-mm diameter discs and weighed (in milligrams) using an analytical balance. Ear edema was evaluated as an increase in ear weight ($n=7\text{animals}/\text{group}$). The percent inhibition of edema was determined according to the following equation:

2.4.

Myeloperoxidase Activity

Myeloperoxidase (MPO) activity was assessed in the supernatants of homogenates of ear sections ($n=7/\text{group}$) in the control groups and mice that received the aforementioned treatments (CO-induced ear edema) according to the technique of Bradley et al.²³ The ear tissue was placed in a solution that contained 0.5% hexadecyltrimethyl ammonium bromide in potassium phosphate buffer (50 mM) and processed in a homogenizer (60 s at 0°C). The homogenate was centrifuged at 2500 rotations per minute (rpm) for 5 min. Afterward, $10\mu \mathrm{l}$ of the supernatant (in triplicate) was added to a 96-well microplate, followed by the addition of $200\mu \mathrm{l}$ of a mixture that contained $O$-dianisidine dihydrochloride ($0.167\mathrm{g}/\mathrm{ml}$), hydrogen peroxide (0.0005%), and potassium phosphate buffer (50 mM). The reaction was stopped by the addition of 1.46 M sodium acetate solution. Enzyme activity was determined by the end-point technique by measuring absorbance at a wavelength of 460 nm and expressed in optical density (OD). The percent inhibition of MPO activity was determined according to the following equation:

2.5.

Determination of Cytokine Levels

Cytokine levels were determined in the supernatants of homogenates from ear sections in the control groups and mice that were treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ ($4\mathrm{mg}\mathrm{DHA}/\text{ear}$) or OO alone 6 h after topical CO application ($n=7/\text{group}$). The ear tissue was placed in phosphate-buffered saline and processed in a homogenizer (60 s at 0°C). The homogenate was centrifuged at 2500 rpm for 10 min, and the supernatant was separated and stored at $-70°\mathrm{C}$ until the assay was performed. The analyses were performed using a magnetic bead reader (Luminex 100TM/200TM) and Cytokine Mouse Magnetic 20-Plex Kit (Invitrogen, Waltham, Massachusetts, catalog no. LMC006M, lot no. 1000793) according to the manufacturer’s recommendations.

2.6.

Determination of Percutaneous Penetration and Topical Inflammatory Response by Photoacoustic Spectroscopy

The measurements were performed on sections of right ear tissue (negative control) and 6 h after applying CO in the positive control group, ${\mathrm{FOP}}_{\mathrm{OO}}$ ($4\mathrm{mg}\mathrm{DHA}/\text{ear}$) and OO alone, using the experimental setup shown in Fig. 1 ($n=5/\text{group}$). The PAS homemade experimental setup is composed of a 1000-W Xenon arc lamp (Oriel, model 68820) used as a light source, which is diffracted when passing through the monochromator (Oriel, model 77250) with 3.16-mm input and output slits. The monochromator was equipped with a diffraction grade (Oriel, model 77296) for UV–Vis spectral range between 200 up to 800 nm. The nominal power of the light source was 800 W. The mechanical chopper (Stanford Research Systems, model SR 540) was tuned at 15 Hz modulating the light that impinges the sample. Higher order diffractions were eliminated by bandpass filters. Thus, the monochromatic beam was focused in the sample placed inside the photoacoustic cell, which was sealed with a transparent quartz window (8-mm diameter and 2-mm thickness). A Brüel and Kjaer model 2669 capacitive microphone was coupled to collect the photoacoustic signal that was generated from the pressure variation resulting from the periodic heating of the sample. The used lock-in amplifier was a EG&G Instruments, model 5110. The depth of the skin sample contributing to the photoacoustic signal was estimated using the thermal diffusion length ($\mu s$): $\mu s=(d/\pi f)1/2$, in which $d$ is the sample thermal diffusivity and $f$ is the light modulation frequency. Taking the thermal diffusivity of the skin (ventral and dorsal) as $d=4.0\times {10}^{-4}{\mathrm{cm}}^2{\mathrm{s}}^{-1}$ (Ref. 24) and $f=15\mathrm{Hz}$, the estimated $\mu s$ value of our measurements was $30\mu \mathrm{m}$. The final photoacoustic signal is proportional to the sample absorption coefficient, then the photoacoustic spectra can be interpreted by means of absorption bands.¹⁷ To correct the source emission intensity in each wavelength, all spectra were normalized with respect to a carbon black sample signal.

Fig. 1

Experimental setup of the used PAS.

The spectra of each ear were obtained positioning the sample into the photoacoustic cell, illuminating the face to be measured. Afterwards, the ear was turned upside down to illuminate the opposite side, as shown in Figs. 2(a) and 2(b). Considering the chosen light modulation frequency at 15 Hz ($\mu s$ of about $30\mu \mathrm{m}$) and that the total sample thicknesses were always around $445\mu \mathrm{m}$, in all cases the measurements were performed near the surface of the samples exposed to the incident light. Then, detecting the presence of ${\mathrm{FOP}}_{\mathrm{OO}}$ band on the dorsal side (opposite side of ${\mathrm{FOP}}_{\mathrm{OO}}$ application) indicates that the topically applied formulation permeated through the ear tissue. In addition, the quantification of the bands associated with the inflammatory response in the inflamed and treated ears may provide information regarding the inflammatory processes and the treatment effect.

Fig. 2

Schematic of spectral readings for illumination of the (a) ventral surface and (b) dorsal surface of ear tissue in mice (modified from Ref. 15).

2.7.

Statistical Analysis

Edema values, MPO activity, and cytokine levels are presented as $\text{mean}\pm \text{standard error}$ of the mean (SEM). The data were analyzed using analysis of variance (ANOVA) followed by Tukey’s test. Values of $p<0.05$ were considered statistically significant.

3.1.

Effect of Fish Oil Preparations on Croton Oil-Induced Ear Edema and Myeloperoxidase Activity

The application of CO on the left ear (inflamed control or positive control) caused an evident inflammatory response at 6 h compared with the right ear (noninflamed control or negative control). Topical application of ${\mathrm{FOP}}_{\mathrm{OO}}$ at 0.25, 0.5, 1, 2, and $4\mathrm{mg}\mathrm{DHA}/\text{ear}$ and OO alone significantly reduced the formation of ear edema [93.3%, 81.4%, 85.5%, 72%, 83.7%, and 85.5%, respectively; Fig. 3(A)]. Because these reductions of ear edema were similar, independent of treatment, we conducted a further experiment using 70% acetone as the diluent to discard the possible action of OO on the FO response.

Fig. 3

Effect of FOP on ear edema induced by CO ($200\mu \mathrm{g}$) in mice. (A) ${\mathrm{FOP}}_{\mathrm{OO}}$ at the amounts indicated and OO alone were topically applied immediately after the application of CO on the left ear. (B) ${\mathrm{FOP}}_{\mathrm{AC}}$ at the amounts indicated and OO diluted in 70% acetone were topically applied immediately after the application of CO on the left ear. Dexa and Indo were used as anti-inflammatory reference drugs and diluted in 70% acetone and topically applied at doses of 0.1 and $1\mathrm{mg}/\text{ear}$, respectively. V = right ears that received only application of the vehicle (70% acetone) as a negative control. The data are expressed as the $\text{mean}\pm \mathrm{SEM}$ ear weight for each group ($n=7$) 6 h after the application of CO. ^a$p<0.001$, compared with V; ^b$p<0.001$, compared with positive control group (CO); ^c$p<0.05$, compared with positive control group (CO; ANOVA followed by Tukey’s test).

${\mathrm{FOP}}_{\mathrm{AC}}$ at 0.5, 1, 2, and 4 mg DHA/ear and OO diluted in 70% acetone significantly reduced the intensity of ear edema (31.4%, 51.3%, 72.4%, 74.4%, and 59.9%, respectively). These reductions were less than those of Dexa ($0.1\mathrm{mg}/\text{ear}$) and Indo [$1\mathrm{mg}/\mathrm{ear}$; 98.5% and 98.3%, respectively; Fig. 3(B)].

MPO activity increased after CO application. The topical application of ${\mathrm{FOP}}_{\mathrm{OO}}$ at 0.25, 0.5, 1, 2, and $4\mathrm{mg}\mathrm{DHA}/\text{ear}$ and OO significantly inhibited MPO activity [98.3%, 89%, 99%, 99%, 97.8%, and 93.6%, respectively; Fig. 4(A)].

Fig. 4

Effect of FOP on MPO activity in supernatants of homogenates of ear tissue in mice. (A) ${\mathrm{FOP}}_{\mathrm{OO}}$ at the amounts indicated and OO alone were topically applied immediately after the application of CO on the left ear. (B) ${\mathrm{FOP}}_{\mathrm{AC}}$ at the amounts indicated and OO diluted in 70% acetone were topically applied immediately after the application of CO on the left ear. Dexa and Indo were used as anti-inflammatory reference drugs and diluted in 70% acetone and topically applied at doses of 0.1 and $1\mathrm{mg}/\text{ear}$, respectively. V = right ears that received only application of the vehicle (70% acetone) as a negative control. The data are expressed as the $\text{mean}\pm \mathrm{SEM}$ OD for each group ($n=7$) 6 h after the application of CO. ^a$p<0.001$, compared with positive control group (CO; ANOVA followed by Tukey’s test).

${\mathrm{FOP}}_{\mathrm{AC}}$ at 0.25, 0.5, 1, 2, and $4\mathrm{mg}\mathrm{DHA}/\text{ear}$ and OO diluted in 70% acetone significantly reduced MPO activity (76.1%, 86.6%, 91.6%, 96.9%, 92.3%, and 80.1%, respectively). These reductions of MPO activity were greater than the effect of Indo ($1\mathrm{mg}/\text{ear}$; 48%), which did not significantly inhibit MPO activity. Dexa ($0.1\mathrm{mg}/\text{ear}$) also inhibited MPO activity [80.6%; Fig. 4(B)].

3.2.

Effect of Fish Oil Preparation on Cytokine and Chemokine Levels in Supernatants of Homogenates from Ear Sections

The levels of the proinflammatory cytokines tumor necrosis factor (TNF), interleukin-$1\beta$ ($\mathrm{IL}\text{-}1\beta$), and IL-6 as well as chemokines keratinocyte-derived chemokine (KC/CXCL-1) and monocyte chemoattractant protein-1 (MCP-1/CCL2) increased in the supernatants of homogenates from the ear sections 6 h after the induction of the inflammatory response by topical application of CO. Topical treatment with ${\mathrm{FOP}}_{\mathrm{OO}}$ at $4\mathrm{mg}\mathrm{DHA}/\text{ear}$ reduced TNF, $\mathrm{IL}\text{-}1\beta$, IL-6, KC, and MCP-1 levels in the supernatants of homogenates from the ear sections [Fig. 5(A)–5(E), respectively]. Topical application of OO reduced $\mathrm{IL}\text{-}1\beta$, IL-6, and MCP-1 levels [Fig. 5(B), 5(C), and 5(E), respectively] but did not reduce TNF or KC levels [Fig. 5(A) and 5(D), respectively]. Topical treatment with ${\mathrm{FOP}}_{\mathrm{OO}}$ at $4\mathrm{mg}\mathrm{DHA}/\text{ear}$ and OO alone did not alter the levels of the anti-inflammatory cytokines IL-10 and IL-4 (data not shown).

Fig. 5

Effect of FOP on the levels of the cytokines (A) TNF, (B) $\mathrm{IL}\text{-}1\beta$, and (C) IL-6 and chemokines, (D) KC, and (E) MCP-1 in supernatants of homogenates of ear tissue in mice. ${\mathrm{FOP}}_{\mathrm{OO}}$ ($4\mathrm{mg}\mathrm{DHA}/\text{ear}$) and OO alone were topically applied immediately after the application of CO on the left ear. Cytokine and chemokine levels were calculated by subtracting the value obtained for the right ear (V, negative control) from the value obtained for the left ear in the different groups. The data are expressed as the $\text{mean}\pm \mathrm{SEM}$ for each group ($n=7$) 6 h after the application of CO. ^a$p<0.05$, compared with positive control group (CO; ANOVA followed by Tukey’s test).

3.3.

Determination of Percutaneous Penetration of Fish Oil Preparation and Its Effect on the Topical Inflammatory Response by Photoacoustic Spectroscopy

The photoacoustic spectra for CO, FO, OO, and ${\mathrm{FOP}}_{\mathrm{OO}}$ at $4\mathrm{mg}\mathrm{DHA}/\text{ear}$ (Fig. 6) indicated that these substances featured absorbing chromophores in the region from 200 to 350 nm. The CO spectrum was broadband, without the presence of well-defined bands compared with the FO, OO, and ${\mathrm{FOP}}_{\mathrm{OO}}$ spectra. The FO spectrum had two absorption bands with peaks around 246 and 280 nm. The OO spectrum had a band centered around 234 nm. The ${\mathrm{FOP}}_{\mathrm{OO}}$ spectrum had two absorption bands with peaks around 246 and 273 nm. The detection of these characteristic bands along the thickness of the ear tissue indicated the presence of ${\mathrm{FOP}}_{\mathrm{OO}}$ and OO.

Fig. 6

Photoacoustic spectra of CO, FO, OO, and ${\mathrm{FOP}}_{\mathrm{OO}}$ ($4\mathrm{mg}\mathrm{DHA}/20\mu \mathrm{l}$).

The readings from the ventral and dorsal sides of the right (negative control) and the left (positive control) ear tissue 6 h after the application of CO showed the presence of two bands in the left ear tissue spectrum centered around 225 and 270 nm (Fig. 7, arrows). These bands were absent in the right ear tissue spectrum. These results show that PAS detected the inflammatory response in ear tissue 6 h after the CO-induced inflammatory response.

Fig. 7

Photoacoustic spectra for the ventral and dorsal sides of right ear tissue (negative control, skin ventral, skin dorsal) and left ear tissue (positive control, skin + CO ventral, skin + CO dorsal) 6 h after the application of CO on the ventral side of the left ear ($n=5$).

Fig. 8(a) shows the spectra that were obtained from readings on the ventral side of the right and left ears tissue (negative and positive controls, respectively) and ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ ($\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$) and OO (skin + CO + OO). The spectra for ear tissue that was treated with OO have spectral characteristics that were similar to the positive control ear tissue spectrum, with bands at 225 and 270 nm. The ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ presented a band at 270 nm, but at a lower intensity compared with this same band from the positive control and no band at 225 nm. This strong reduction of the 225 nm band appears to be a fingerprint showing that the ${\mathrm{FOP}}_{\mathrm{OO}}$ presented an anti-inflammatory effect on the ventral side of the ear.

Fig. 8

(a) Photoacoustic spectra obtained from readings of the ventral side of right ear tissue (negative control, skin ventral), left ear tissue (positive control, skin + CO ventral), ear tissue treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ ($4\mathrm{mg}\mathrm{DHA}/\text{ear}$; $\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$ ventral) and ear tissue treated with olive oil (skin + CO + OO ventral), 6 h after topical application of CO on the ventral side of the left ear ($n=5$). (b) Gaussian adjustments of spectra for $\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$ ventral. The dashed lines show the bands that make up the spectrum, and the solid line indicates the sum of all bands.

Because of the overlapping of the 210 to 250 nm bands from ${\mathrm{FOP}}_{\mathrm{OO}}$ and OO with that characteristic of inflamed tissue, we could not directly observe the presence of substances that were topically applied on the ears. Therefore, fittings were made using Gaussian functions [Fig. 8(b)] to decompose the photoacoustic spectra into constituent bands, thus allowing the identification of characteristic bands for ${\mathrm{FOP}}_{\mathrm{OO}}$ and OO in the ear tissue. This procedure was performed for all of the photoacoustic spectra for the treated ear tissue and pure substances. In Fig. 8(b), the dashed lines show the bands in the photoacoustic spectrum for the ventral side of the ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$. The solid line is the Gaussian functions sum, which represents the best fitting of the sample spectrum.

With Gaussian fitting, the separation of the spectrum of the positive control ear tissue (skin + CO) showed peaks centered at 200, 225, 270, and 280 nm. The separation of the spectrum of the ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ ($\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$) had bands centered at 200, 246, 270, 273, and 280 nm. The separation of the spectrum of ear tissue that was treated with OO (skin + CO + OO) exhibited Gaussian peaks centered at 200, 225, 234, 270, and 280 nm. Thus, the bands of the substances that were topically applied that differed from the characteristic bands of inflamed tissue (225 and 270 nm) were centered at 246 and 273 nm for ${\mathrm{FOP}}_{\mathrm{OO}}$ and 234 nm for OO, indicating the presence of such substances on the ventral surface of the treated ears. These same bands were obtained with the Gaussian fittings of the pure substances.

The spectra that were obtained from reading the dorsal side of the right and left ears tissue (negative and positive controls, respectively) and ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ ($\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$) and OO (skin + CO + OO) are shown in Fig. 9. Figure 9(a) shows the presence of the optical absorption bands that characterize inflamed tissue in the photoacoustic spectra of the ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ and OO, but at a lower intensity compared with the positive control ear, indicating an inhibitory effect. Similar to the ventral side, measurements of the dorsal side were also performed using Gaussian fittings [Fig. 9(b)]. In the photoacoustic spectra of the samples that were treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ and OO, decomposed by Gaussian functions, bands were centered at 246 and 273 nm for ${\mathrm{FOP}}_{\mathrm{OO}}$ and 234 nm for OO, indicating the presence of these substances on the dorsal side of the ear tissue. This shows that both ${\mathrm{FOP}}_{\mathrm{OO}}$ and OO permeated the ear tissue to the dorsal side. The topical application of both substances occurred on the ventral side.

Fig. 9

(a) Photoacoustic spectra obtained from readings of the dorsal side of right ear tissue (negative control, skin dorsal), left ear tissue (positive control, skin + CO dorsal), ear tissue treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ ($4\mathrm{mg}\mathrm{DHA}/\text{ear}$; $\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$ dorsal) and ear tissue treated with olive oil (skin + CO + OO dorsal), and 6 h after topical application of CO on the ventral side of the left ear ($n=5$). (b) Gaussian adjustments of spectra for $\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$ dorsal. The dashed lines show the bands that make up the spectrum, and the solid line indicates the sum of all bands.

Fig. 10 shows the average areas of the optical absorption bands that characterize inflamed tissue that was induced by CO, centered at 225 and 270 nm, obtained from readings from the ventral and dorsal sides of inflamed control ear tissue (skin + CO) and ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ ($\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$) and OO (skin + CO + OO). The areas of each photoacoustic spectra were obtained using Gaussian fittings. On the ventral side [Fig. 10(a)], there was a reduction of the areas of the bands that characterize inflamed tissue for ear tissue that was treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ compared with the inflamed control ear. Ear tissue that was treated with OO presented areas of the bands that characterize inflamed tissue that were similar to the control inflamed ear. Figure 10(b) shows a reduction of the areas of the optical absorption bands that characterize inflamed tissue on the dorsal side for both ${\mathrm{FOP}}_{\mathrm{OO}}$ and OO treatment compared with the control inflamed ear. Note that in the treated ear with ${\mathrm{FOP}}_{\mathrm{OO}}$, the areas of the bands related to the inflamed tissue were smaller than those treated with OO.

Fig. 10

Average areas of optical absorption bands that characterize inflamed tissue (225 and 270 nm) in photoacoustic spectra after the CO-induced inflammatory response in positive control ear tissue (skin + CO), ear tissue treated with ${\mathrm{FOP}}_{\mathrm{OO}}$ [$4\mathrm{mg}\mathrm{DHA}/\text{ear}$ ($\text{skin}+\mathrm{CO}+{\mathrm{FOP}}_{\mathrm{OO}}$)], and ear tissue treated with olive oil (skin + CO + OO) ($n=5$). (a) Ventral side. (b) Dorsal side.