588 JOURNAL OF COSMETIC SCIENCE
hours the Cutibacterium species were seeded at 106–108 CFU/mL in an insert above the
keratinocyte’s cultures grown in KSFM medium for 48 hours. To evaluate the impact
of the different C. acnes strains on the cell viability, a MTT assay was performed. The
MTT assay is a rapid colorimetric assay based on the cleavage of the tetrazolium ring of
MTT (3-(4,5-dimethylthazolk-2-yl)-2,5-diphenyl tetrazolium bromide) by dehydrogenases
in active mitochondria of living cells as an estimate of viable cell number. Sodium dodecyl
sulfate at 0.025% was used as cytotoxic positive control.
To study direct interaction, normal human keratinocytes were seeded in 96-well culture
plates at 50,000 cells/cm2 in KSFM (Thermo Fisher Scientific). C. acnes strains were heat-
killed at 95°C for 1 hour with a heated dry bath and were then seeded at 106–108 CFU/
mL in direct interaction with keratinocytes for 24 hours. Heat-killed C. acnes strains were
used instead of living C. acnes strains to avoid their interaction in MTT viability test run
in parallel to normalize the IL-8 secretion from keratinocytes. To evaluate the impact of
the different C. acnes strains on the secretion of IL-8 by keratinocytes, the media was then
collected and assessed using an IL-8 immunoassay, a technique relying on the ability of an
antibody to recognize and bind a specific macromolecule (IL-8). An inflammatory cocktail
(12.5 ng/ml of tumor necrosis factor alpha (TNFα), 12.5 ng/mL of interferon gamma
(IFNγ), and 0.5 µg/mL of Poly I:C) was used as positive control. Statistics: n =3, Student
t-test or Anova versus Untreated, *p 0.05, ***p 0.001, NS: not significant.
RESULTS
THE MICROBIAL SEQUENCING ANALYSIS BETWEEN NORMAL AND SENSITIVE SKIN
The analysis was completed for 73 volunteers (40 NS subjects and 33 SS subjects). The
diversity was analyzed using the Shannon index and highlighted no significant differences
between the two cohorts (Figure 1).
The prevalence of the species was analyzed by determining the absence or presence
of different bacterial genera on NS or SS. Figure 2 represents the 20 genera having a
Figure 1. The alpha diversity analysis between normal skin and sensitive skin by using the Shannon index.

589 Specificities of Microbiota From Sensitive Skin
prevalence of greater than 30% for either cohort. Considering genera having a prevalence
higher than 50% for either cohort (NS or SS), we observed that the genus Cutibacterium was
found in all the volunteers regardless of cohort. The genus Staphylococcus was also found
in all NS volunteer, but the prevalence slightly decreased by 3% in SS. The prevalence
was decreased by 23% in the Streptococcus genus, by 19% in the Anaerococcus, and by 9%
in the Paracoccus and Bradyrhizobium. On the contrary, we observed an increase in SS of
the prevalence in Corynebacterium genus by 7%, Acinetobacter by 47%, Kocuria by 45%, and
Enhydrobacter by 65%.
Next, we compared the two cohorts to examine the presence of common genera and to
determine variations in abundance (Figure 3). Cutibacterium and Staphylococcus are the two
most abundant genera, but the proportions were reversed between the two cohorts. NS
showed 47% Cutibacterium abundance and 39% Staphylococcus, compared to the SS with
60% Cutibacterium, and only 22 %Staphylococcus. Ruminococcus, Anaerococcus, and Romboutsia
were present in the top 10 of most abundant genera in NS, but were absent from this
ranking in SS. On the contrary, Bacillus, Acinetobacter, Actinomyces, Kocuria, and Micrococcus
were more abundant in SS when compared to NS.
When looking at the species level Cutibacterium genus, the more prevalent and abundant
genus in both cohorts, and found increased in SS by 1.28-fold (from 47% to 60%, Figure 3),
we observed that, in NS the most prevalent and abundant members of the Cutibacterium
were 4 species: Cutibacterium acnes, Cutibacterium granulosum, Cutibacterium namnetense and
Cutibacterium avidum with a respective relative abundances of 46%, 2%, 0.4%, and 0.1%
(data not shown). In SS subjects, only 3 species of Cutibacterium were recovered, with
Cutibacterium avidum no longer being detected. In addition, Cutibacterium acnes was increased
Figure 2. Sequencing analysis between normal and sensitive skin. Illustration of selected 20 genera having
more than 30% prevalence in either cohort (percentage of negative or positive variation in SS versus NS are
given for species having a prevalence greater than 50% for either cohort).