Summary:
The present study was carried out towards an EU funded project, named
“MICROSMETICS”. The aim of MICROSMETICS project was to discover and carry to
the stage of development innovative products in the area of cosmeceuticals
originating from global biodiversity using emerging and state of the art
technologies in the field of biotechnology, natural products chemistry and
applied microbiology.
More specifically, MICROSMETICS scientific concept involved the discovery of
novel natural products originating from global microbial biodiversity. Already
existing culture collections of fungi and actinomycetes were exploited
incorporating modern high throughput platforms (in silico & in vitro) for the
rational and targeted selection of the most promising strains. Advanced
analytical approaches and techniques were applied for the efficient,
accelerated and advantageous isolation and identification of natural
constituents as well as the quality assessment of the lead products. A broad
spectrum of bioassays was incorporated for the evaluation of anti-ageing, more
specifically anti-oxidant, skin-protecting, and skin-whitening activity of all
derived products. Attention was given to the selection and optimization of
fermentation technologies used at pilot scale for the production of new active
raw materials to be used in the cosmetic industry.
In order to achieve those goals, as a first step a Rational Drug Design
approach has been developed. Starting from the CosIng Repository of the
European Commission, an accurate functional prediction model was created for
all known cosmetic functions. Additionally it was constructed the homology
models of specific cosmetic target receptors (tyrosinase, elastase,
hyaluronidase and collagenase) for which the appropriate in vitro tests have
already been developed. More than 40.000 known microbial metabolites were
processed through a consensus scoring prediction protocol using: a) functional
prediction model b) virtual screening procedure for the above 4 selected
receptors c) similarity search based on all known molecules from literature
that bind to the specific receptors and d) toxicological profile filtering.
Combining all the results with functional prediction model and toxicological
profile filtering, 110 microorgnanisms were selected that can produce those
metabolites or analogues.
Among them 54 fungi and 56 actinomycetes from the global biodiversity
(including strains from Antarctica, Alaska, Spain, New Caledonia, Hawaii, South
Africa, Comoros Island etc.) were cultivated under “nutritional arrays”
(different culture conditions) in order each single strain to produce ten
different extracts and thus exploit all the potential chemodiversity that
microorganisms can produce. In total 1082 sample extracts (614 actinomycetous
extracts & 425 fungal extracts) have been generated, then were filtered and
prepared in 96 barcoded well format and were forwared to high throughput
screening.
Initially for the safety of those extracts cytotoxicity was evaluated on A2058
and HepG2 cell lines by the MTT method. The results of the above bioassays
permitted the elimination of all toxic and non active extracts.
A broad spectrum of bioassays have being incorporated for the evaluation of
anti-ageing, more specifically anti-oxidant, skin-protecting, and
skin-whitening activity of all derived products. For the evaluation of the 1082
extracts concerning the antioxidant activity the DPPH and ABTS assays were
used. Skin-protecting was evaluated by measuring spectrophotometrically the
inhibitory properties of samples against enzymes which are related to the
elasticity and moisture of the skin (elastase, collagenase, and protease). The
skin whitening activities were determined by the tyrosinase assay, using L-DOPA
as substrate.
Among the 1082 extracts the top 104 extracts demonstrated significant activity
in at least one target and were forwarded to metabolomics analysis. Among the
104 extracts, 75 belonged to actinomycetes and 29 to fungi. Only the 11% and 6%
of all the actinomycetes’ and fungi’s extracts respectively were non toxic and
active. Data analysis revealed the likely effect in the activity that the
culture medias may have.
In order to select the 20 most active extracts, a strategy combining
UHPLC/Orbitrap-HRMS, in positive and negative modes, with multivariate
statistical methods was applied. In parallel, biological assays were repeated
to exclude the false positive samples and select the most active ones. All
derived chromatograms have been analyzed and a positive correlation between the
profiles of the extracts with the bioassays’ results was observed. Thus,
through analysis the 20 most promising extracts (19 strains under 14 conditions
have been forwarded for large-scale cultivation and for confirmation of the
activity. In total 8 fungal and 12 actinomycetous extracts were selected for
further investigation.
All strains were cultivated with the optimum media for producing the active
metabolites in a 1 L scale fermentation and 66 extracts were generated, using
various extraction protocols in order to optimize the extraction method. Among
them some confirmed the activity while in others the large scale cultivation
failed to produce the same activity and metabolites and thus their cultivation
is currently re-evaluated. The chemical profile of those extracts was examined
by using TLC, UPLC and LC-HRMS.
The EtOAc extract of Cercospora sp. (strain CF-223709) with the most
siginificant antioxidant activity (70.76% ± 1.28 free radicals scavenging, DPPH
assay) and the EtOAc_LL extract of Streptomyces hygroscopicus angustmyceticus
(strain CA-129531) with the most significant skin-whitening activity (74.35% ±
0.48 inhibition of the tyrosinase enzyme), were chosen in order to isolate
their bioactive metabolites and elucidate their structures.
The profiling and fractionation of those extracts was performed using both
classical chromatographic methods (semi-prep HPLC), as well as more innovative
techniques, like UPLC, SFC-MS and FCPC. All the fractions were evaluated for
their antioxidant and skin-whitening activity and the most active ones were
chosen for further fractionation or to direct elucidation of the structure of
the main metabolite they contained. The full set of spectroscopic data (LC-HRMS
and 1D & 2D NMR) was recorded for all isolated compounds. During the process of
the bio-guided isolation we managed to isolate and identify:
• three bioactive metabolites that cause the antioxidant activity of the
strain CF-223709, which include the well-known active compounds fulvic acid
(IC50=21.58 μg/ml, DPPH assay) and vinaxanthone (IC50=35.08 μg/ml, DPPH assay),
as well as one new natural product with the chemical formula C13H12O6
(IC50=18.99 μg/ml, DPPH assay).
• two bioactive metabolites that contribute to the skin-whitening
activity of the strain CA-129531: trichostatin A and trichostatic acid.
It is anticipated that following the same extraction and fractionation protocol
to other active strains we can identify more potent candidates for industrial
development. This can serve as a proof of concept that microbial ingredients
can have successful applications in cosmeceutical industry.
Keywords:
Actinomycete, Fungi, Natural products, Skin-whitening activity, Antioxidant activity
