Considering the overall variations of the microbiota that colonized the volunteers' facial skin before the sunscreen application, both the right and left sides revealed significant similarities regarding bacterial families/genera/species (Fig. 2).
The main entity found, in terms of overall abundance, was Cutibacterium (Propionibacterium) acnes (p < 0.05), although it was not the most found species in two volunteers (V3 and V5). However, this finding is not surprising since this species is the most reported in metagenomics studies [19, 20]. Another Propionibacterium, C. granulosum, was also found in all volunteers and sampling times; however, in lower relative abundance (< 0.85%) and without inter-sample variations (p ≥ 0.7851).
The average abundance of C. acnes was compiled, and the results showed no differences between the right and left sides of this species before the experiment (p ≥ 0.8438). This shows a certain balance in the abundance of C. acnes, with no predominance on one face over the other (Fig. 3).
C. acnes is an aerotolerant anaerobic bacterium found both on the skin surface and inside the pilosebaceous follicles [21]. Therefore, it is possible to assume that during the collections, the slight pressure exerted by the swabs against the skin exposed the pilosebaceous follicles (pores), where C. acnes is usually found in greater abundance [22]. Such facts may collaborate in understanding the higher proportional collection of this bacterium for 66.6% (4/6) of the volunteers.
The volunteers reported performing facial hygiene equally (two to three times a day) and sleeping on either side of their body (data not shown). This may have contributed to the homogeneity in C. acnes density on both sides of the face.
After exposure to the base gel (Rt1) and the photoprotective formulation (Lt1), no differences in species abundance were detected either (p ≥ 0.8438) (Fig. 3). Therefore, it is possible to assume that the presence of the sun protection factors ethyl methoxycinnamate, ethyl salicylate, methyl anthranilate, and octocrylene added to the base gel do not determine an increase or reduction of C. acnes in the follicles/skin. It is also interesting to consider the fact that the photoprotective formulation also did not promote variations in abundance at the collection times for base-gel (Rt0 and Rt1) nor the photoprotective formulation (Lt0 and Lt1) (p ≥ 0.8438). This indicates that the formulation, as presented, does not seem to determine changes in the proportions of this Propionibacteriaceae and is not considered a factor in dysbiosis for this species.
This finding is unprecedented and gains importance if we consider that, contrary to the old (and deposed) dominant idea that the exacerbated proliferation of C. acnes would be the main trigger for acne, this species may have a probiotic [23] or antioxidant [24] function, contributing to skin homeostasis [5]. In addition, it is believed that the reduction in skin microbial diversity associated with the constant activation of TLR-2-activated innate immunity may be associated with acne chronification [22].
The second largest group in terms of overall abundance is composed of Staphylococcus spp. All volunteers were colonized before the experimental phase, ranging from 3.1% (V5) to 69.2% (V3). The main species found was S. epidermidis, and its abundance was shown to be invariant between the different experimental groups (p ≥ 0.6878) (Fig. 4).
It is known that, like C. acnes, S. epidermidis is an important commensal of the skin [25–27].
Recently, it has been postulated that acne may arise from a proportional imbalance of C. acnes and S. epidermidis in the microbiota colonizing the skin [28]. According to this hypothesis, an event causing dysbiosis with reduced S. epidermidis subpopulation could precipitate an over-colonization by virulent variants of C. acnes in follicles, with multipoint activation and exacerbated innate immunity and consequent acne inflammation.
Since no significant variations in the abundance of S. epidermidis were observed after using the photoprotective formulation, this event indicated that it should not induce acne due to an imbalance of these two bacterial species.
Staphylococcus aureus is a species of clinical importance [29], whose role in acne is debatable [30], but which induces α-toxin-dependent TLR-2 response [31, 32]. This species was also constant in all volunteers and at all time points (Fig. 5); however, with relatively low frequency (< 1.3% in overall abundance) compared to S. epidermidis.
Regardless of these Firmicutes' low relative abundance and virulence profile, no differences were found between samples collected from the same volunteer at different times and treatments (p ≥ 0.584). This shows that the photoprotective formulation could not cause any imbalance in terms of abundance for this species.
Other Staphylococcus species such as S. equorum, S. sciuri, and S. succinus occurred at extremely low frequencies (≥ 0.12%) within the genus and were not detected in all volunteers. The same occurred with non-fermenting Gram-negatives such as Pseudomonas spp. (P. citronellolis, P. fragi, P. nitroreducens, P. stutzeri, and P. viridiflava) and Acinetobacter sp. with spot detections and at low frequencies (≥ 0.035%). Other bacterial entities for which there is no evidence or indication of an association with acne were not considered [33]. For these reasons, these species, genera, or families were not included in this discussion.
Metagenomics is a valuable tool for studying the taxonomic structures of microbial populations that coexist in each ecological niche without the need to culture these microbial entities. For some, metagenomics analysis allows one to know more than "who is there," it also allows one to infer "what they are doing there" [34]. Because of the practicality in simultaneously detecting multiple species, including those with anaerobic and fastidious growth, and the high-throughput capability [35], we chose to employ this technology to conduct this study.
Male volunteers were recruited since acne vulgaris is less common in men than in women [36]. Among several factors, acne results from elevations in androgen levels [37]; since progesterone blocks androgen receptors, inhibiting their acnegenic action, and considering that progesterone reductions occur in the late stage of the menstrual cycle, this dermatosis tends to be exacerbated in the premenstrual period [38]. As this was a complex variable to control, recruitment of female volunteers was avoided. However, this subpopulation warrants future study, as women use sunscreens more frequently than men [39].
The recruited volunteers had no complaints nor clinical signs of acne during the recruitment period, nor did they develop dermatosis during the experimental course, indicating that there was no clinical disturbance in the regions of application of the base gel, a base gel containing filters, nor of the liquid soap formulation, which was shown to control skin oiliness in a balanced manner, both in the washout period and during the experiment, avoiding obliteration of follicles and maintaining non-comedogenic conditions [40].
As limitations of this study, we can point out that a larger sample size would undoubtedly lead to more data and that after a larger-scale analysis, it would be possible to conclude with a greater degree of certainty whether the formulation of sunscreen may or may not cause dysbiosis in facial skin. In addition, a study with a longer intervention time could generate data different from those presented here, also considering the possibility of divergent results when evaluating sunscreens with a sun protection factor (SPF) value higher than those applied in this study.
Based on the results obtained and considering study limitations, it was possible to infer that there is no evidence that the SPF 28.8 photoprotective formulation used in this study causes microbial dysbiosis involving the species of microorganisms related to the occurrence of acne vulgaris.