Interactions between growth factors, skincells and the extra-cellular matrix (ECM) areessential for tissue regeneration in woundhealing1 as well as intrinsic ageing of theskin. Wound healing is a complex processcomprising different phases such asinflammation, proliferation and remodelling,all of which require growth factors toregulate the fine balance between thesynthesis of extracellular matrix and itsdegradation by proteases.2 Intrinsic skinageing is accompanied by an accumulationof reactive oxygen species (ROS) due to animpaired mitochondrial oxidativemetabolism. Enhanced ROS formation willaffect cell signalling pathways which lead toan increased break down of extracellularmatrix components.3 The thinning andfragility of elderly skin is the result of animbalance of degradation and regenerationof skin tissue. More recently, studies byQuan and colleagues find an additionalreason for the loss of matrix structure in adecreased synthesis and release of growthfactors.4 Specifically this work suggests thata decrease in the growth factors CTGF andtransforming growth factor-β (TGF-β) inaged skin and fibroblasts leads to adecrease in the expression of extracellularstructural proteins. CTGF is primarilyinduced by TGF-β and appears to functionas a down-stream mediator in theactivation of extracellular matrix synthesis.Therefore one might ask: could ageing skinbe rejuvenated by supplementing growthfactors? The use of natural human growthfactors in cosmetic products is prohibited.Biotechnologically-produced growth factorscould theoretically be used for cosmeticproducts, but their efficacy is very limitedfor the following reasons: most growthfactors are proteins of high molecularweight and complex conformation, whichreach a limited penetration depth into theskin. Moreover, growth factors were shownto be unstable in cosmetic formulations,they lose their functionality and are quicklydegraded by proteases following superficialapplication.

Here we describe an approach tostimulate the growth factor-dependentsynthesis of matrix proteins in the dermis

by improving keratinocyte-fibroblastcommunication. In the present study weinvestigated the effect of specific plant bulbextracts on human keratinocyte cellcultures for the purpose of enhancing thesynthesis and release of growth factors,which stimulate fibroblast cells tosynthesise extracellular matrix proteins. In aplant the bulb serves as a food reserve inorder to support the newly growing plantwith nutrients. We identified a bulb extractfrom Crocus chrysanthus with the ability toactivate intercellular communication.

Materials and methodsPreparation of extracts from plant bulbsBulb extracts were prepared from speciesof the Amaryllidaceae, Asparagacea,Iridaceae, Liliaceae and Myrsinaceaefamilies. A mixture containing 12% (w/w)plant bulbs and 15% (w/w) ethanol in water was incubated at 50˚C for 4 hours in a DIG-MAZ 50 system (SamtechExtraktionstechnik GmbH). The extractionprocess was controlled by chromatographicfingerprint analysis (Waters, ACQUITY UPLC system, C18 column).

Screening assay Primary human keratinocytes from a 50-year-old donor were cultured in anappropriate liquid medium for 24 hours.The medium was then replaced with theassay medium (Epilife medium) containingthe different plant bulb extracts at non-toxicconcentrations as was previouslydetermined in a cytotoxicity assay (MTT).The keratinocytes were then incubated for 72 hours. Afterwards, the keratinocytecell cultures were centrifuged and the cellpellets as well as the culture supernatants(keratinocyte-conditioned media) collected.The keratinocyte cells were used to analyse the expression of 30 genes coding for different growth factors andcytokines typically expressed in this celltype. The gene-expression analysis wasdone by quantitative PCR (LightCycler

September 2012 PERSONAL CARE 51

Jasmin Lozza, Daniel Schmid, Esther Belser, Fred Zülli – Mibelle Group Biochemistry, Switzerland

SKIN CARE

Crocus bulb extract promptsepidermis/dermis crosstalk

ABSTRACT

The epidermis communicates with thesubjacent dermis via signalling molecules.This crosstalk is particularly important for the regulation of complex events such as wound healing, but also forcontinuous tissue repair and regeneration.The messenger molecules involved aremainly cytokines and growth factors. An imbalance in the synthesis and releaseof these signals as it is observed in elderlyskin may lead to a reduced biosynthesisof important extracellular-matrix (ECM)proteins such as collagen and elastin.Here we present a strategy to activate

matrix protein production by stimulatinggrowth factor synthesis and release bykeratinocytes, which can easily bereached by superficially applied cosmeticcompounds. In vitro assays demonstratethat a crocus bulb extract can activateexpression of elastin, lysyl oxidase-like 2enzyme (LOXL2) and the connectivetissue growth factor (CTGF). Non-invasiveskin structure analysis by two-photonmicroscopy performed after a two-weektreatment with a cream containing thecrocus bulb extract demonstrates a clearaugmentation of collagen and elastin.

Figure 1: Effect of an extract from Crocuschrysanthus bulbs on specific growth factor

gene expression in keratinocytes.

Gen

e ex

pres

sion

rel

ativ

e to

con

trol

(%

)

400

350

300

250

200

150

100

50

0After 72 hours

■ Fibroblast growth factor 5■ Fibroblast growth factor 7■ Insulin-like growth factor

binding protein 6■ CTGF

pp51-54 PCEuroSept12 21/8/12 12:26 Page 51

system; Roche Molecular System Inc.). The keratinocyte-conditioned media

were used to treat normal human dermalfibroblasts (pool of donors over 50 yearsold). The fibroblasts cells were cultured for24 hours in a standard culture mediumand then for 72 hours in an assay medium(Epilife medium). The medium was thenremoved and replaced by either, a) thekeratinocyte-conditioned medium, b) theassay medium with TGF-β (10 ng/mL), c) the assay medium containing the plantbulb extracts or d) the assay medium alone serving as control. After incubationfor 24 or 72 hours, the fibroblasts wereisolated and subjected to gene expressionanalysis of 61 genes known to play a keyrole in fibroblast physiology and cutaneousageing. These experiments were performedin triplicate for verification.

Clinical study using two-photon microscopy0.4% crocus bulb extract [DermCom/INCIname: Crocus Chrysanthus Bulb Extract(and) Acacia Senegal Gum (and)Aqua/Water] was formulated into a creamand tested in a study over a period of 4 weeks on a 53-year-old Caucasianwoman. First the pre-application baselineimages were characterised. Then a creamwas applied twice-a-day to one innerforearm. The other forearm was treatedwith the corresponding placebo cream.Two-photon microscopic images of the skinwere taken after two and after four weeksof treatment. Stacks of images at depthsincreasing up to 200 μm below the skinsurface were acquired using the OlympusFluoview 1000 MP two-photon microscopysetup customised for human skin imagingin vivo. Two-photon microscopic imageswere obtained simultaneously in twomodes: auto-fluorescence (AF) and

second-harmonic generation (SHG). Theimages reveal changes in collagen andelastin levels, which were quantified andstatistically compared to the baseline andto the placebo.

Clinical anti-ageing studyA cream containing 2% crocus bulb extractwas tested in a clinical study with 20healthy volunteers, average age 47. Thecream was applied twice daily for 4 weeks

to the inner side of the forearm and on thecrow’s feet area. The other forearm andcontralateral side of the face were treatedwith the placebo cream. Eight to twelvehours after the last product application ondays 14 and 28, skin firmness andelasticity on the forearms and wrinkle deptharound the eyes were measured by meansof the Cutometer MPA 580, Courage &Khazaka GmbH and PRIMOS,GFMesstechnik GmbH.

SKIN CARE

52 PERSONAL CARE September 2012

Figure 4: Two-photon microscopic images taken at different skin depths. Skin area was

treated with 0.4% DermCom in a cream. Auto-fluorescence signal shown in green, second-

harmonic generation signal in red.

Figure 2 (left): Effect on Elastin, LOXL2 and CTGF gene expression in fibroblasts after incubation for 24 and 72 hours with the keratinocyte-

conditioned medium. The keratinocytes were treated with the crocus bulb extract. The values are compared to gene expression values of fibroblasts

after incubation with the conditioned medium of untreated keratinocytes (100%). Figure 3: Effect of TGF-β on gene expression in fibroblast cells.

Gen

e ex

pres

sion

rel

ativ

e to

con

trol

(%

)

1400

1200

1000

800

600

400

200

0

After 24 hours After 72 hours

■ Elastin■ Lysyl oxidase-like 2■ CTGF

Gen

e ex

pres

sion

rel

ativ

e to

con

trol

(%

)

3000

2500

2000

1500

1000

500

0

After 24 hours After 72 hours

■ Elastin■ Lysyl oxidase-like 2■ CTGF

Dermal-epidermal junction Dermis

0.4

% D

erm

Com

Pla

cebo

Elastin

Collagen

pp51-54 PCEuroSept12 21/8/12 12:26 Page 52

Results and discussionBulb extracts from species of theAmaryllidaceae, Asparagaceae, Iridaceae,Liliaceae and Myrsinaceae families weretested in a screening assay for growthfactor signalling in order to identify one thatbest triggers intercellular communication.Treatment of a keratinocyte cell culture with a bulb extract obtained from Crocus

chrysanthus (Iridaceae) was shown to up-regulate the expression of genes coding for growth factors, which are known toplay an important role in the epidermis/dermis crosstalk (Fig. 1). Members of thefibroblast growth factor family are involvedin a variety of biological processes includingcell growth, morphogenesis and tissuerepair. Insulin-like growth factor bindingproteins modulate the activity of growthfactors.

In another set of experiments, humanfibroblast cells were incubated with theconditioned medium derived fromkeratinocyte cultures treated with theabove plant bulb extract. Figure 2 showsthat this keratinocyte-conditioned mediumgreatly enhances the expression of genesinvolved in the synthesis of matrix proteinsin comparison to the control cultureincubated with the conditioned mediumfrom untreated keratinocytes. Results alsoreveal that elastin expression increases12-fold and LOXL2 increases 5-fold after24 hours of incubation.

LOXL2 plays a critical role in theformation of the extracellular matrix bycross-linking elastin, which is essential forthe stabilisation of this fibrous protein. It has been shown that the concentrationof lysyl oxidase-like enzymes is reduced in aged skin.5 Also, CTGF expression wasfound to be increased after 72 hours ofincubation. The role of CTGF in adulttissue has previously been linked to

SKIN CARE

September 2012 PERSONAL CARE 53

Figure 5: Increase of collagen content in the upper dermis after

two weeks of treatment with a cream containing 0.4% DermCom.

Figure 6: Increase of elastin content in the upper dermis after two

weeks of treatment with a cream containing 0.4% DermCom.

Start After 2 weeks Start After 2 weeks

Col

lage

n am

ount

(ar

bitr

ary

unit

s)

7000

6000

5000

4000

3000

2000

1000

0

Placebo 0.4% DermComStart After 2 weeks Start After 2 weeks

Elas

tin

amou

nt (

arbi

trar

y un

its)

900

800

700

600

500

400

300

200

100

0

Placebo 0.4% DermCom

pp51-54 PCEuroSept12 21/8/12 12:26 Page 53

fibrosis, a group of diseases with excessiveformation of fibrous connective tissue.However, Quan and co-workers showed thatthe down-regulation of CTGF likely mediatesreduced collagen and elastin expression inaged human skin.4 In our experiment, TGF-β was used as a positive control sinceit mediates the synthesis of extracellularmatrix proteins. The comparison of Figure 2and 3 illustrates that fibroblast cells treatedwith TGF-β exhibit a similar though strongerexpression pattern compared to the cellsincubated with the conditioned medium. A direct incubation of fibroblast cells withthe crocus bulb extract resulted in no effecton the gene expression pattern. Thisindicates that the observed stimulatingeffect on fibroblasts depends on signallingmolecules secreted by keratinocytes.

A clinical study using a two-photonmicroscopy was performed for the purposeof analysing the effect of a topically appliedcream containing 0.4% crocus bulb extracton skin structure. In the two-photonmicroscopic image shown in Figure 4 theAF signal in green is generated for exampleby elastin fibres and compounds such asNADH and FAD. The SHG signal in redhighlights collagen fibres. The results of thisstudy demonstrate that the verum iseffective as it significantly increases theamount of collagen in the papillary dermisand in the upper part of the reticular dermis(Fig. 4 lower panel) after only two weeks oftreatment compared to the placebo treatedskin (upper panel). Statistical analysisreveals an increase in collagen content by115% relative to initial conditions (Fig. 5).Less pronounced but also enhanced is theincrease in elastin content by 25%compared to initial conditions (Fig. 6). Inboth cases, the difference between theverum and placebo treated forearms washighly significant. The increase in collagenand elastin observed after two weekscontinues until the end of the experimentat week four. However, the differencebetween placebo and verum-treated skin isless pronounced after four weeks.

Furthermore, the two-photonmicrographs reveal an improvement inwrinkle width and depth. This effect isclearly seen in the upper panel of Figure 7,where wrinkles appear as black, crack-likeopenings that are wide at a skin depth of12 microns and become narrower at 24and 48 microns in depth. By contrast,images of the verum-treated skin (lowerpanel) show a complete absence of suchmicro-cracks.

These results are further supported byour second in vivo study on 20 volunteers,where a cream containing 2% crocus bulbextract was applied to the crow’s feet area.The wrinkle depth was reduced significantlyafter two weeks treatment, on average by

9% relative to initial conditions. After fourweeks of treatment on inner forearms, skin firmness was improved by 13% bothcompared to initial conditions anduntreated skin (Fig. 8).

ConclusionsAged skin and fibroblasts exhibit adecrease in the synthesis of growth factors.Wrinkles, which originate in the dermis arethe main manifestation of skin ageing andindicate a loss in elasticity and tensilestrength. The dermis is not easily reachedby topically applied cosmetic actives.DermCom derived from Crocus chrysanthus

bulbs, overcomes this problem bystimulating the keratinocytes in the toplayer of the skin to synthesise and releasegrowth factors. These act as messengermolecules to induce the production ofextracellular matrix proteins in the dermis,thus improving skin firmness. PPCC

References1 Schultz GS, Wysocki A. Interactions between

extracellular matrix and growth factors in woundhealing. Wound Repair Regen 2009; 17 (2):153-62.

2 Ghahary A, Ghaffari A. Role of keratinocyte-fibroblast cross-talk in development ofhypertrophic scar. Wound Repair Regen 2007;15 (Suppl 1): S46-53.

3 Mehta RC, Fitzpatrick RE. Endogenous growthfactors as cosmeceuticals. Dermatol Ther 2007;20 (5): 350-9.

4 Quan T, Shao Y, He T, Voorhees JJ, Fisher GJ.Reduced expression of connective tissue growthfactor (CTGF/CCN2) mediates collagen loss inchronologically aged human skin. J Invest

Dermatol 2010; 130 (2): 415-24.5 Pascual G, Mendieta C, Mecham RP,

Sommer P, Bellón JM, Buján J. Down-regulationof lysyl oxydase-like in aging and venousinsufficiency. Histol Histopathol 2008; 23 (2):179-86.

SKIN CARE

54 PERSONAL CARE September 2012

Figure 7: Two-photon microscopic images of the epidermis after two weeks of treatment

with a cream containing 0.4% DermCom.

Figure 8: Increase in skin firmness.

Mea

n in

crea

se r

elat

ive

to in

itia

lco

ndit

ions

, an

d to

unt

reat

ed (

%)

14

12

10

8

6

4

2

0After 14 days After 28 days

■ Placebo■ 2% DermCom

*p<0.05 versus untreated**p<0.05 versus untreated

and placebo

Granular layer Spinous layer Dermal-epidermal

12 μμm 24 μμm 48μμm

Veru

mP

lace

bo

pp51-54 PCEuroSept12 21/8/12 12:26 Page 54