1. IntroductionIt is known that reactive species are naturally originated in biochemical processes such as energy production, inflammation, phagocytosis, and cell growth regulation, among others. External factors are also able to provide the production of species such as exposure to ultraviolet (UV) radiation, ozone, X-rays, and air pollutants [1,2,3,4]. In the skin, free radicals are constantly generated by cutaneous cells such as fibroblasts and keratinocytes. Nevertheless, healthy skin can eradicate these reactive species with nonenzymatic (glutathione, ubiquinol, vitamins C and E) and enzymatic (glutathione peroxidase, glutathione reductase catalase, superoxide dismutase, and thioredoxin reductase) antioxidants [5,6,7].When the number of free radicals produced in the tissues exceeds the antioxidant defenses, what is known as oxidative stress occurs. As a result, the peroxidation of lipid membranes, damage to DNA, protein, and enzymes appear, and these harms, in turn, worsen in cognitive dysfunction, degenerative cardiovascular diseases, and cancer, among other illnesses [8].Energetic photons of UV light can be transmitted through the layers of the skin and after that, can be absorbed by cellular chromophores. DNA bases are cellular chromophores; thus, they can directly absorb UV photons, initiating photo-induced reactions. Furthermore, damage in the cells may occur by photosensitization processes, where endogenous or exogenous sensitizers absorb UV photons. Depending on the type, the distribution and concentration of skin chromophores, epidermal thickness, and its functional condition, photobiochemical reactions occur, provoking changes in the cell and tissue biology due to the formation of reactive oxygen species (ROS). The interactions between ROS and cellular biomolecules provoke a final biological response, as illustrated in Figure 1 [5,6,9,10].Although the formation of these reactive species is inevitable, it is possible, with the administration of endogenous and exogenous antioxidants, to minimize their formation, thus avoiding injurious effects to the organisms. Molecules must perform one or more of the following actions to act as antioxidants: oxygen depletion; single oxygen extinction; chelation of metal ions that would otherwise catalyze reactions to form reactive oxygen species; elimination of reactive oxygen species or termination of the oxidation propagation chain reaction; and the prevention of oxidative damage. Hydroxyls attached to aromatic rings such as in polyphenols can react with free radicals, as shown in Figure 2 [8].Rosmarinic acid (RA) is a polyphenol whose molecular structure was first elucidated in 1958. Its molecule was first isolated from the leaves of Rosmarinus officinalis L. and later from plants belonging to the families Lamiaceae (e.g., lavender, sage and mint) and Boraginaceae (e.g., cordia and echium). The name “rosmarinic acid” was given precisely since it was first extracted from rosemary (R. officinalis L.). It is an ester derived from caffeic and 3,4-dihydroxyphenylacetic acids. The RA molecule has a carboxylic group, two aromatic rings A and A′ with ortho-catechol structures, the unsaturated C=C bond and the ester portion [11]. Different studies have shown that RA can act biologically in several ways, being an antiviral, antibiotic, anti-inflammatory, and antioxidant [12,13,14,15].The daily topical application of sunscreens can protect the skin from UV rays and thus delay or reduce the occurrence of skin harm such as the appearance of melanomas, spots, and wrinkles. Considering that antioxidants are capable of helping to minimize the occurrence of ROS that are related to the damages previously mentioned, the association of such molecules to the photoprotective formulations optimizes the beneficial effects of these products over the skin [9,14,15,16,17].

Considering the photoprotective and antioxidant potential of RA, in this research work, the safety and efficacy of a multifunction prototype sunscreen were investigated, aiming to evaluate the performance of this polyphenol with two known and widely used UV filters, bemotrizinol and octyl p-methoxycinnamate.

2. Materials and Methods 2.1. Materials

Rosmarinic acid (96%), agarose ethylenediaminetetraacetic acid disodium salt dihydrate, sodium chloride, Trizma, Triton X-100, dimethyl sulfoxide, phosphate buffer, methyl nicotinic, and sodium hydroxide were purchased from Sigma-Aldrich (St Louis, MO, USA). Octyl p-methoxycinnamate and bemotrizinol were purchased from Mapric (São Paulo, Brazil) and Brasquim (São Paulo, Brazil), respectively. The water was purified in a Milli-Q-plus System (Merck Milipore, USA). Long-lived, spontaneously immortalized human keratinocyte cells (HaCaT) were obtained from the Cell Bank of Rio de Janeiro. DMEM supplemented with 10% fetal bovine serum, trypsin, and SYBR Gold (1:10,000, Invitrogen—Cat S11494) were purchased from Thermo Fisher Scientific (Waltham, MA, USA).

2.2. Methods 2.2.1. Composition of the FormulationsFour formulations containing commercially available octyl p-methoxycinnamate (UVB filter), bemotrizinol (broad spectrum filter), plus 1.0% (w/w) of the bioactive compound rosmarinic acid were used in the study [17]. The active ingredients of the formulations are presented in Table 1. 2.2.2. In Vitro Assay In Vitro Evaluation of DNA Fragmentation of Human Cells Exposed to UV Radiation by the Comet Assay Method

Human Keratinocytes Cell Culture

Human keratinocyte (HaCaT) cells were seeded in 75 cm2 bottles, cultivated, and expanded in a humid atmosphere, in the presence of 5% CO2 at 37 °C using DMEM supplemented with 10% fetal bovine serum. Upon reaching 70% confluence, cells were trypsinized and seeded in 6-well plates for later treatments and the further evaluation of DNA fragmentation.

Incubation with the Formulations and Cells Irradiation

Cells were incubated with the samples at two non-cytotoxic concentrations, 0.0049 and 0.00245% (w/v), for 24 h for subsequent exposure to UV radiation. After the incubation time, cells were submitted to 3 J/cm2 of UVA/UVB in a sun simulation chamber (Suntest CPS+, Atlas, Linsengericht, Germany). Then, cells were kept in suitable culture conditions for an additional 24 h. Cell lysates were collected and proceeded with the comet assay.

Comet Assay

Cell lysates were centrifuged for 5 min at 1500 rpm. Afterward, the cell pellet was mixed with 1000 µL of 0.75% low melting point agarose. Immediately, 100 µL of this suspension was spread onto slides previously coated with 1% agarose. After solidification, the slides were immersed in a freshly prepared lysing solution [2.5 M NaCl, 100 mM ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA Na2S), and 10 mM Trizma Base, pH 10], with the addition of 1% Triton X-100 and 10% dimethyl sulfoxide, for at least 1 h. After lysis, slides were incubated for 5 min in cold PBS (free Ca2+ and Mg2+) and for 40 min in alkaline solution (pH >13). Cells on the slides were submitted to electrophoresis (300 mA, 25 V) for 30 min. Slides were neutralized by three washes with neutralization buffer (pH 7.5) and the DNA was precipitated. DNA was stained with SYBR Gold (1:10,000, Invitrogen—Cat S11494). A number of 150 cells was randomly selected on each slide and analyzed using an optical fluorescence microscope (Leica, DM 6000 B, Wetzlar, Germany) coupled with a camera of 2.8 MP (Leica, DFC7000 T, Wetzlar, Germany).

The comet tail length was classified into four classes (scores): 1 = low damage; 2 = medium damage; 3 = high damage; and 4 = almost all DNA in the tail. The final score of each experimental group was performed by multiplying the number that represented the class of the comet and the total number of comets found in this class. The sum of the results from the multiplications was carried out, corresponding to the final score [18]. 2.2.3. In Vivo Assays Subjects

The protocol was concluded with 12 healthy subjects with phototypes II to VI, after oral information and written consent. The procedures were in accordance with the ethical standards on human experimentation (Ethics Committee of Universidade Lusófona’s Research Center for Biosciences & Health Technologies) and the Declaration of Helsinki. Exclusion criteria were the presence of dermatitis or other skin or allergic disease and smoking. Subjects were instructed not to apply any topical products to the test sites during the study but were allowed to wash normally. The mean age was 32 ± 12 years old (20 to 44 years).

Skin BiocompatibilityAliquots of the formulations were applied to predefined sites on both forearms (two formulations per forearm of each subject, in a 9.0 cm2 area). The application site of the samples was randomized to minimize the occurrence of possible biases. Measurements were taken before and after 30 min of application with a corneometer (stratum corneum hydration), a Tewameter TM300 (skin barrier quantified through the transepidermal water loss, TEWL), a pH meter (pH value of the skin surface) (C&K Electronics GmbH, Germany), and a chromameter (red color, a *, according to CIE Lab system) (Minolta Camera Co., Osaka, Japan). Afterward, the subjects were instructed to apply the formulations twice a day for 7 days. On the eighth day, new measurements were performed. The non-invasive measurements were conducted according to the specialized literature and guidelines (n = 3) [19,20,21,22]. To reduce the interindividual variability, the results were treated as the ratio between the response and the basal/control value. Anti-Inflammatory ActivityAliquots of the samples were applied to the forearm of each volunteer in a 9.0 cm2 area. The study was conducted in two phases: a pre-treatment, which consisted in the application of the samples twice a day for 7 days. On the eighth day, the second phase began, where erythema formation was induced in each pre-treated area by applying filter paper (2.25 cm2) saturated with an aqueous solution of methyl nicotinate (0.5% w/w) for 60 s. Measurements of the skin microcirculation of individual test sites were recorded continuously for 15 min using laser doppler flowmetry equipment (PeriFlux System 5000, Perimed, Stockholm, Sweden). Therefore, the in vivo evaluation of the anti-inflammatory effect was based on the ability of the samples to decrease the extent of an erythema-induced response. The area under the curve of the perfusion profile obtained at each site was measured for all subjects, together with the slope of the tangent line in the hyperemia stage [23]. Results were analyzed as the ratio between the values obtained at each sample site and the control values for the volunteers, in the selected parameters. 2.3. Statistical Analysis

Experiments were conducted randomly, in triplicate, and with a significance level of 5% (p ≤ 0.05), conducted by Minitab software, version 18 (Minitab, Inc., State College, PA, USA).

4. DiscussionFormulations, at both concentrations, into the cell cultures, under the UV stress condition, protected the DNA fragmentation compared to UV control. Although we noticed the decrease in DNA fragmentation, we could not exclusively attribute this finding to the presence of RA in the samples, since F1 and F3 were absent in this phenolic compound. Possibly, the RA concentration at 1.0% was not completely enough to develop a distinctive performance among the samples, since all of them presented some level of DNA protection ranging from 11 to 27% including the blank sample (F1) [14,24,25]. Additionally, the test could have a limitation when complex samples are performed. Furthermore, our protocol similarly indicated that the blank sample and the sunscreen one (without RA) protected the cells from DNA fragmentation under UV stress, and specialized literature has reported that RA was able to reduce the extant of DNA injury by other mechanisms through doxorubicin or sepsis, for instance [26,27]. Although investigations point to RA as a molecule capable of protecting DNA fragmentation, studies that have investigated its influence in a multifunctional photoprotective formulation are restricted. For this reason, we used a complex multifunctional sample, compatible with the available commercial products, to study DNA fragmentation by the comet assay method. The effect of RA, as an isolated compound, in the protection of DNA damage against UV radiation is reported in the specialized literature. Pérez-Sánchez and coworkers evaluated the RA protective effects in human keratinocytes and reported an increased survival of the cells upon UVB radiation [28]. In similar research, Psotova and coworkers demonstrated that RA exhibited the ability to reduce the decrease in the cell viability of keratinocytes exposed to UVA radiation [29]. Although these results are promising, complex dermocosmetic formulations must be tested in scenarios more adjusted to consumer use.F1 and F3 provided an increase in skin hydration 30 min after their application and after daily application (p-value p-value 30,31]. We noticed that samples containing RA, independently of the presence of UV filters, were not able to maintain the cutaneous superficial hydration effect on the seventh day, however, all samples positively interacted with this skin attribute after 30 min. Tomazelli and coworkers and Ruscinc and coworkers observed compatible results from their multifunction bioactive sunscreens containing 0.1% rutin and 5.0% Vaccinium myrtillus L. extract, respectively, when in vivo skin hydration was not improved after the application of those products [32,33].Treatments with F1 and F2 resulted in an increase in the TEWL after 30 min of application (p-value 32,33]. Additionally, in the pH and a * parameters, no significant differences were found (p-value > 0.05) compared to the control, which indicated that, despite the increase in TEWL, the samples developed a safe profile for topical application as they did not cause redness nor changes on the skin pH value [23].It has been extensively reported that topical application of nicotinic acid and its esters causes local erythema [34,35] by vasodilation of peripheral blood capillaries in the dermal papillae of dermis through prostaglandin D2 and E2 releases. The prostaglandin D2 liberation also occurs by the incidence of UV radiation on the skin by the generation of ROS [36]. Considering this phenomenon, the association of antioxidants in photoprotective formulations can optimize skin protection. RA is known for its antioxidant potential by free radical scavenging action proven by several experimental protocols such as DPPH (2,2-diphenyl-1-picrylhydrazyl) free radical scavenging and TBARS (malonyl dialdehyde formation) [14,17,37]. Its molecule presents two catechol structures conjugated with a carboxylic acid group as an important structural element in the antioxidant activity of this compound [38]. Its activity as a photoprotective adjuvant ingredient was evaluated with successful results in the elevation of the sun protection factor (SPF) in vivo of a sunscreen system and RA also increased the tyrosinase activity and its expression level in B16 melanoma cells [15,39,40].Regarding the area under the curve, no significant differences were reached when comparing the sites pretreated with the samples and the control (p-value > 0.05). The slope of the tangent line in the hyperemia results followed a similar trend, and F2 slightly accelerated the formation of the erythema caused by ethyl nicotinate when compared with the control sample. Thus, RA formulations, divergent to that expected, were not able to decrease the onset and erythema intensity according to our protocol. Matwiejczuk and coworkers researched a RA protective effect against the influence of methylparaben and propylparaben on collagen in fibroblasts. Parabens can inhibit the biosynthesis of collagen and reduce the proliferation and viability of human skin fibroblasts [41]. In this investigation, RA provided protection against these changes, being the findings related to the RA antioxidant capacity [42]. Fernando and coworkers demonstrated the antioxidant capacity of RA by a reduction in UVB-induced intracellular ROS and weakening oxidative damage to protein and DNA [14]. Pattananandecha and coworkers studied the effects of an extract containing RA on UVA-irradiated human skin fibroblasts. The extract was able to inhibit ROS and matrix metalloproteinase-1 [43]. Several studies involving RA in modified systems has also been reported [44,45,46,47,48]. For instance, Perra and colleagues developed hyalurosomes loaded with an extract rich in RA and proved, in vitro, its protection against oxidative stress in skin fibroblasts [49]. 5. Conclusions

The samples in vitro protected the DNA fragmentation compared to the UV control and showed in vivo good skin compatibility. Formulations F1 and F3 were able to increase skin hydration, and, possibly, the RA interfered with this attribute. An increase in TEWL was observed for formulations F1, F2 and F4, which may be related to vehicle composition, containing or not the RA. Regarding the in vivo anti-inflammatory efficacy, no decreases were observed in the inflammatory reaction caused by the ethyl nicotinate with any of the evaluated formulations. As a perspective, we suggest trials with a greater number of subjects or protocol modifications such as the application of a monitored sample, a greater number of applications per day, or even the occlusion of the application site. Altering the qualitative and quantitative composition of the vehicle is also a pertinent perspective.