568 JOURNAL OF COSMETIC SCIENCE
The coculture protocol was optimized before the testing of ingredients. More specifically,
the nutrient concentration of the medium was reduced, the pH was adjusted at 6.5 (lowest
pH to allow the growth of all bacteria) and the incubation temperature was set at 32°C
(data not shown). The selection of incubation time and atmosphere are presented in the
Results section.
INGREDIENTS ANALYSIS IN THE COCULTURE MODEL
Ingredients (glucose, inulin, glycine, proline, glycerol and a mix of ethyl- and methylparabens,
50:50) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Phenoxyethanol was
purchased from Honeywell Fluka (Morristown, New Jersey, USA), pentylene glycol
from Minasolve (Mont-Saint-Guibert, Belgium), butylene glycol from OXEA GmbH
(Oberhausen, Germany) and 1,3–propanediol from Knoell NL BV (Wageningen,
Netherland). SolagumTM AX is an ingredient from Seppic (La Garenne-Colombes, France)
obtained by a physical process between two polysaccharides: xanthan gum and acacia gum
(composite polymer, INCI name: Acacia Senegal Gum and Xanthan Gum).
The ingredients were dissolved in a minimal coculture medium and 180 µL were placed
in 96-well microplates at 1.1× of the test concentration. A positive control was included
in each test series (minimal medium without ingredient). Each sample was inoculated
with the consortium (20 µL). Microplates were incubated at 32°C. After incubation,
samples were diluted in sterile NaCl 0.9% (Merck, Darmstadt, Germany) and plated on
appropriate media (as described in the bacterial culture section) to determine the microbial
concentration of each species, in CFU/mL.
Figure 1. Process of isolation and banking of cutaneous wild type strains.

569 COSMETIC INGREDIENTS THAT RESPECT SKIN MICROBIOTA
The species proportion (%)in the consortium was calculated by dividing the species
concentration (in CFU/mL) with the concentration of total bacteria (in CFU/mL) and
multiplying by 100.
For each species, the logarithmic reduction (LR) was also calculated from the logarithmic
bacterial concentrations in the exposed consortium versus the non-exposed consortium
(control), according to the following calculation:
LR =log (concentration after contact with the ingredient (log CFU/mL)/concentration of
the control without ingredient
(log CFU/mL)
The LR ​​obtained for the five bacterial species were added, in order to obtain a score (Sc)
per ingredient:
Sc =LR
C.acnes +LR
S.epidermidis +LR
S.mitis +LR
M.luteus +LR
C.xerosis
STATISTICAL ANALYSIS
Each condition (control, ingredients) was analyzed in triplicates. Bacterial concentrations
were expressed as mean values and standard deviation. In Figure 2, percentages were
expressed as mean values and confidence intervals (CI 95%). The coefficient of variation
(CV) was calculated according to the following equation: CV =(σ/μ) × 100, where σ is
the standard deviation and μ the mean value.
RESULTS
VALIDATION OF THE IDENTIFICATION OF WILD TYPE STRAINS
To be as representative of the cheek microbiota as possible, five species were sampled from
healthy volunteers with sterile swabs. After strains isolation, the precise characterization
of each isolate was undertaken, by MALDI-TOF and/or 16S rRNA gene sequencing. The
MALDI-TOF technique provides a mass spectrum fingerprint of the biopolymers from an
isolated microbial colony. Each species has a unique fingerprint that can be compared with
those of the bioMérieux database for precise identification. Similarly, the full sequence of
the 16S rRNA encoding gene of each bacterial isolate was compared to sequences from
the 16S rDNA 500 microbial library to obtain identification at the species level. Results
confirmed the characterization of seven isolates of C acnes, seven isolates of S epidermidis,
seven isolates of S mitis, seven isolates of C xerosis, and seven isolates of M luteus. Isolated
strains were stored either alone or mixed in equal proportion (pools of species). The pools
were used for the development of the coculture model (Figure 1).
COCULTURE OPTIMIZATION
The coculture protocol was optimized before testing ingredients. More specifically, the
nutrient concentration in the growth medium was lowered and the incubation temperature