Make-up brushes: hygienic or not?: COSSMA

The cosmetics industry has had to deal with the issues of hygiene, infections and microbial dangers for some time. In the recent past, make-up brushes in particular have come to the fore as potential germ carriers. Givaudan took a closer look at this prejudice.

The use of make-up brushes was recorded for the first time in ancient Egypt. Centuries later, whether for comfort, camouflage or seduction, almost every culture embraced cosmetic rituals involving the use of these tools. For foundation, 63% of women declare using a make-up brush for the application of dry products and 36% for liquid products¹. Despite this popularity, more and more magazines highlight how dangerous make-up brushes could be. During the last past years, they have been described as “places full of germs”² and “able to transmit potentially lethal infections”³. Scientifically, only a few studies have been conducted regarding the microbial contamination of make-up tools. They revealed varying levels of pathogenic micro-organisms in application tools⁴ and showed that beauty blender is the application tool carrying the highest bacterial load5. The following study aims to describe more precisely the microbial communities living on this make-up tool using next generation sequencing technologies for the first time. This study also aims to show whether or not this make-up tool can shape the skin microbiota in a dangerous way.

Experimental design

Six Caucasian women between 26 and 39 were recruited, a written informed consent was completed, and no personal identifiable data were collected. Volunteers were in good general health, were not pregnant and did not present any major skin pathology. Questions were asked about their skin types and cosmetic habits.

Skin microbiota was swabbed twice on the volunteers. Firstly, the swabbing of nude skins was done before make-up routine, corresponding to more than nine hours after the demake- up of the day before. The second swabbing was done five hours after the morning routine, including face cleansing and the application of skin care products and make-up. Each volunteer was asked to perform her normal cosmetic routine, using her own cosmetic products and brushes. All samplings were done on the entire face, corresponding to the area where foundation was applied with the brush. Finally, microbiota was collected from the foundation brushes of each volunteer, inside and outside the brush.

The detailed protocol for sampling, sequencing and data analysis is described by Jarrin C. et al^.6 excepted for ITS1 DNA study, where amplification has been performed as described by Park T. et al.⁷

fig 1: Microbial composition of nude skins and make-up brushes. figures: Givaudan

Experimental analysis

The microbial composition of the make-up brushes is derived from the microbial composition of the skin. The microbial composition of the make-up brushes were found to be proclose to that of the skin itself, with four major bacterial communities detected: Staphylococcus, Cutibacterium, Corynebacterium and Streptococcus; and one major fungal genus: Malassezia (fig 1). Interestingly, bacterial composition allows clustering samples into two groups according their origin: nude skin or make-up brushes, while fungal compositions don’t (fig 2). Also, the microbial composition profile not always allow linking a make-up brush to its volunteer. Futhermore, we do not notice any correlation between the microbiota diversity of brushes and the age of the brush, frequency of washing, type of foundation applied, frequency of use, bristle properties, storage conditions or skin types (Tab 1).

fig 2: Nude skins (before), make-up brushes (brushes) and skin five hours after beauty routine (after) bacterial (2.a) and fungal (2.b) compositions for each volunteer (where Sample_1 corresponds to all the samples collected from Volunteer 1) represented by PCoA on weightedUNIFRAC distance.

for each volunteer (where Sample_1 corresponds to all the samples collected from Volunteer 1) represented by PCoA on weighted — PCoA: principal component analysis, a technique for feature extraction —the “least important” variablescan be dropped while still retaining the most valuable parts of all of the variables. figures: Givaudan

Bacteria in higher proportions on make-up brushes are not able to proliferate on the skin. We report significant higher proportions (p-value<0.05, defining the statistical relevance of the result) of five microbial genera, only bacterial, on brushes in comparison to nude skin: Xanthomonas (0.01% on nude skin to 1.11% on brushes), Brachybacterium (0.01% to 1.69%), Chryseobacterium (0.01 to 0.47%), Actinomyces (0.03% to 0.54%) and Acinetobacter (0.12% to 1.05%). Xanthomonas is a genus common in the environment8 and all others are usually found on the skin. Some of them include species identified as skin opportunistic pathogen⁹ and one species of Acinetobacter (Acinetobacter ursingii) has been previously identified in make-up brushes5. Nevertheless, five hours after make up, in comparison to nude skins, none of these bacterial genera are found in higher proportions on the skin. This result indicates that, even if these bacteria are able to grow on brushes, they are not capable of proliferating on the skin and unbalance the microbiota and the genus level. Cosmetic routine shapes the skin microbiota. Once again, we notice that samples can be clustered according their origin: nude skin or skin after cosmetic routine. In addition, a similar shift of bacterial compositions of volunteers 1, 2, 3 and 4 after beauty routine has been observed (fig 2.a). This cluster of volunteers has a similar nude skin bacterial profile dominated by Cutbacterium and Staphylococcus genera, and the shift observed reflects the important and significant reduction of the proportion of the Staphylococcus genus (p-value<0.05). The decrease of Staphylococcus proportion after cosmetic application has already been reported10. Oppositely, microbiota of volunteers 5 and 6 (respectively dominated by Staphylococcus and Corynebacterium and by Staphylococcus genera and Streptococcus) do not react in the same way, with a lower impact on Staphylococcus proportions. In addition, an important increase of Malassezia proportions is noticed on volunteers 5 and 6 after their cosmetic routine (from 34.3% to 96.2% for volunteer 5 and from 9.6% to 39% for volunteer 6) while the mycobiota of other volunteers is not impacted (data not shown) (fig 2.b). Interestingly, there are no significant differences in the bacterial/fungal ratio between volunteers 1, 2, 3 and 4 and volunteers 5 and 6. This result suggests that some communities are more stable than others depending on the individual, and that the effect of a cosmetic routine highly depends on the individual

Conclusion

This study provides a wider view of bacterial and fungal compositions of make-up brushes and reports the colonization of this beauty accessory by microorganisms from the skin or from the environment. Interestingly, even if some bacteria are found in higher proportions in make-up brushes in comparison to the nude skin, none of them can proliferate on the skin after make-up. According to this study, modifications of the skin microbiota at the genus level after cosmetic routine cannot be imputed to make-up brushes. This study also highlights that the modifications of the skin microbiota after cosmetic routine is highly correlated to the personal microbiota profile. This observation gives scientific support to a consumer trend: 96% of consumers declare that personalisation is important for beauty products and 73% would like to try a cosmetic product with a microflora concept¹¹.

References:

1 Lightspeed/Mintel, September 2019

2 https://www.grazia.fr

3 https://eu.jsonline.com

4 Dadashi L, Dehghanzadeh R. Investigating incidence of bacterial and fungal contamination in shared cosmetic kits available in the women beauty salons. Health Promot Perspect. 2016;6(3):159-163. doi:10.15171/ hpp.2016.25

5 Bashir A, Lambert P. Microbiological study of used cosmetic products: highlighting possible impact on consumer health. J Appl Microbiol. 2020;128(2):598-605. doi:10.1111/jam.14479

6 Jarrin et al. Sensitive skin: insight into microbiota composition and comparison with microbi-ota of normal skin. IFSCC Mag. 2020; Number 1, Volume 23

7 Park et al. Collapse of human scalp microbiome network in dandruff and seborrhoeic derma-titis. Exp Dermatol. 2017;26(9):835-838. doi:10.1111/exd.13293

8 Ryan R. et al. Pathogenomics of Xanthomonas: understanding bacterium–plant interac-tions. Nat Rev Microbiol 2011; 9, 344–355

9 Zheng S. et al. A Microbiome-Based Index for Assessing Skin Health and Treatment Effects for Atopic Dermatitis in Children. mSystems 2019; 4 (4) e00293-19; DOI:10.1128/mSystems.00293-19

10 Lee HJ et al. Effects of cosmetics on the skin microbiome of facial cheeks with different hy-dration levels. Microbiologyopen. 2018;7(2):e00557

11 Givaudan Survey, 2019