566 JOURNAL OF COSMETIC SCIENCE
to their habitat. Indeed, most of them are able to thrive in poorly nutritious conditions,
essentially dry, characterized by an acidic pH (around five) and exposed to varying climatic
conditions (ultraviolet radiations or UV, temperature, etc.).3 Despite a great variability in
the composition of the skin microbiota from one individual to another, common features
are described. On the face, the most abundant genera are Cutibacterium sp. (formerly
known as Propionibacterium sp.), coagulase-negative Staphylococci (CoNS), Corynebacterium sp.,4
Micrococcus sp. and Streptococcus sp.1
Most microorganisms are not harmful and some are known to benefit their host.5 For
instance, they can inhibit pathogens through nutrient competition or the secretion of
harmful molecules like bacteriocins.5 They can also play a role in the immune system
training, or contribute to the maintenance of the barrier function.6 Hence, the microbiota,
in symbiosis with its host, plays an important role for the skin. Nonetheless, the skin
is constantly exposed to harsh environmental conditions or to pollutants or xenobiotic
substances that can alter the homeostasis of this ecosystem.7,8 It has been reported that
the regular use of hygienic products can alter the skin microbiota.9–11 Most cosmetics also
contain preservatives to prevent the proliferation of potential pathogens.12
To ensure that cosmetic products are not harmful to the skin microbiota, researchers have
designed a variety of assays, as described hereafter. Some commercial solutions (“Microbiome
Friendly” certifications by My Microbiome or Byome Labs “Microbiome-Friendly+” seal
from Labskin, etc.)—more or less complex—are already proposed by contractors. The most
common assays consist in putting into contact a cosmetic ingredient or a formulation with
targeted commensal species of the skin (e.g., Staphylococcus epidermidis, Cutibacterium acnes,
Corynebacterium spp., etc.).14 These methods are easy and rapid to implement.13 However,
they are often achieved by working with microorganisms from collections, such as the
American Type Culture Collection (ATCC) or the Biological Resource Center of Institut
Pasteur (CRBIP) and grown in optimal laboratory conditions that are not relevant to the
conditions found on the skin. Moreover, they do not take into consideration the fact that
the skin microbiota is a consortium of several species interacting with each other.
Some models gain in complexity by depositing tested products on 3D (three-dimensional)
skin models colonized by microorganisms.13,15 Here, not only can the microorganisms
be analyzed following the adjunction of products, but also the impact on the skin
tissues. Another advantage is that several species of microorganisms can be inoculated
simultaneously on a reconstructed human epidermis (RHE), hence mimicking the skin
microbiota and giving more relevance to the test. However, mixing more than three
different species in 3D models is a technical challenge.
Finally, the “friendliness” of a product is sometimes evaluated through clinical assays. With
the use of next generation sequencing techniques, such trials offer accurate data about the
formulas’ effect on the microbiome.9,6 Nevertheless, the screening of multiple ingredients
can hardly be achieved with a single clinical trial, and this method seems more appropriate
for finished products.
The present work aims at designing an in vitro model to test the ability of cosmetic ingredients
to preserve the microbiota, and takes into account the complexity and specificity of the skin
ecosystem. Wild isolates from the main species of the human face microbiota (Cutibacterium
acnes, Staphyloccocus epidermidis, Corynebacterium xerosis, Micrococcus luteus, and Streptococcus
mitis) were collected from healthy individuals. Given the environmental conditions found
on the skin, the aim was to develop a coculture protocol (pH near 5.5, temperature near

567 COSMETIC INGREDIENTS THAT RESPECT SKIN MICROBIOTA
32°C, defined atmosphere, and low nutrient concentration) to obtain a consortium with
relevant bacteria proportions, near that of the skin microbiota: 30–40% of C acnes, 15–20%
of S epidermidis, 15–25% of S mitis, 1–10% of xerosis and 2–10% of M luteus.16,9
MATERIALS AND METHODS
COLLECTION AND IDENTIFICATION OF STRAINS AND BACTERIAL CULTURE
All growth media were purchased from Oxoïd, Thermo Fischer Scientific, Waltham,
Massachusetts, USA, except Brucella broth (Condalab, Madrid, Spain).
Wild-type isolates from five species of bacteria were collected from the cheek of 10 healthy
volunteers by rubbing sterile swabs (Thermo Fischer Scientific) for five minutes on the skin.
Swabs were placed in Tryptone Soy Broth (TSB) and incubated aerobically at 37°C for the
enrichment of aerobic strains. Another swab was placed in Brucella broth and incubated
anaerobically at 37°C for anaerobic strains.
After enrichment, selective media were used to grow the strains of interest: Hoyle agar was
used to select potassium tellurite-reducing bacteria like Corynebacteria, Chapman agar and
Baird Parker for halophilic bacteria such as Micrococci and Staphyloccoci, and Columbia agar
with 5% sheep blood for the recovery of fastidious microorganisms like Cutibacteria and
Streptococci.
BACTERIAL IDENTIFICATION AND BANKING
Isolates with the appropriate phenotypic characteristics on selective media were identified
using microscopic observations (Gram staining) and biochemical tests (catalase, oxidase,
RapID CB PLUS identification system, RapID Staph PLUS identification system, Remel
Thermo Scientific, Lenexa, Kansas, USA). Finally, matrix-assisted laser desorption
ionization–time of flight (MALDI-TOF) mass spectrometry (VITEK MS method,
bioMérieux, Marcy-l’Étoile, France) and/or 16S rRNA gene sequencing (MicroSEQ ID
technology combined with 16S rDNA 500 microbial library v2019, Remel Thermo
Scientific) were used to confirm colonies’ identification. Seven isolated colonies of each
species were mixed in equal proportions to obtain pools of species and stored at –80°C
until further use (Figure 1).
CONSORTIUM PREPARATION
Each species was initially grown individually in the appropriate liquid medium (TSB for
S epidermidis, S mitis and M luteus Brain Heart Infusion broth (BHI) for C xerosis, and
Brucella broth for C acnes), and in atmospheres corresponding to the respiratory type of
the species. Then, the five species were calibrated (microscopic counting with a Malassez
cell followed by colony counting on agar plates) and mixed together to the desired initial
concentrations, in the coculture medium (pancreatic digest of casein, 1.7 g/L papaic digest
of soybean meal, 0.3 g/L glucose monohydrate, 0.25 g/L NaCl, 0.5 g/L, and K
2 HPO
4 ,
0.25 g/L (Oxoïd, Thermo Fisher Scientific), and 1.5% Tween 80 (Sigma-Aldrich, St. Louis,
Missouri, USA, pH 6.5).