Identification of Bakuchiol Targeting Proteins in Human Skin Cells

Yucheng Li

Department of Biochemistry & Molecular Biology, University of Miami, Coral Gables, FL 33146, U.S.A.

Keywords: Bakuchiol, Retinol, Antiaging, Natural Product, Target Identification.

Abstract: Purpose: Bakuchiol is a natural product that is widely used for skin antiaging. Previous studies show that

bakuchiol and retinol have similar biological activities despite having no structural resemblance, but the

mechanism of action of bakuchiol remains largely unknown. We proposed a regulatory pathway of bakuchiol

that is based on the downstream activities of retinol. In this work, we will test a small portion of the proposed

pathway—that is, bakuchiol can bind to target proteins different than that of retinol (RARs). Methods: We

will design experiments to identify the molecular targets of bakuchiol. We will perform affinity

chromatography to isolate bakuchiol targeting proteins, followed by mass spectroscopy to sequence the target

proteins. The sequences will be identified using a database. Possible Results: At the end, there are three

possible results: (1) Bakuchiol does not bind the target proteins of retinol (RARs) and instead binds other

proteins; (2) Bakuchiol only binds RARs. (3) Bakuchiol binds RARs and other proteins. Conclusion: The

results of our study will pave the way for understanding the mechanism of action of bakuchiol.

1 INTRODUCTION

Bakuchiol is a natural product mainly obtained from

the seeds of the plant Psoralea corylifolia. Previous

studies have shown that bakuchiol can serve as a

functional analog to retinol even though bakuchiol

does not look like a retinoid (Figure 1). For example,

retinol has long been used as a therapeutic agent to

treat photo-aged human skin. Similarly, Chaudhuri

and Bojanowski showed through a pilot clinical study

that the clinical appearance of photo-aged human skin

is improved by bakuchiol, similar to retinol

(Chaudhuri, Bojanowski 2014). The functional

similarity of bakuchiol and retinol is also reflected in,

for example, their antioxidant and antiinflammation

effects (Chaudhuri, Sivamani, Jagdeo, Elsner,

Maibach 2015). To investigate the similarity of

bakuchiol and retinol on a molecular biology level,

the authors applied a comparative gene expression

profiling with both bakuchiol and retinol by using the

technique of DNA microarray (Chaudhuri,

Bojanowski 2014). They revealed that the volcano

plots of the two substances are similar in shape

(Figure 2), indicating their similar regulation on gene

expression. This leaves a question on why bakuchiol

and retinol, having no structural resemblance with

each other, regulates gene expression in a similar

pattern. One possibility is that bakuchiol binds to the

same targets as that of retinol (i.e., retinoic acid

receptors, or RARs). Another possibility is that

bakuchiol binds to different targets than retinol,

which somehow improves the availability of

endogenous retinol. The latter possibility is supported

by Chaudhuri and Bojanowski’s observation that

bakuchiol upregulates proteins (CRBP II, CRBP IV,

CRABP I, LRAT) that can help increase the cellular

storage of retinol to an extent greatly higher than that

of retinol (Chaudhuri, Bojanowski 2014). This

indicates a possible regulation pathway that

bakuchiol achieves its functional similarity to retinol

by binding to protein targets different than that of

retinol; the targets directly or indirectly activate

transcription factors that mediate the expression of

the retinol-storage-related proteins, which increases

the availability of endogenous retinol. In other words,

bakuchiol increases available retinol in cells, and it is

the increased retinol that ultimately acts to exhibit the

anti-aging activity of bakuchiol in human skin cells.

Figure 1: Structure of bakuchiol (left) and retinol (right).

Li, Y.

Identiﬁcation of Bakuchiol Targeting Proteins in Human Skin Cells.

DOI: 10.5220/0011296600003443

In Proceedings of the 4th International Conference on Biomedical Engineering and Bioinformatics (ICBEB 2022), pages 803-809

ISBN: 978-989-758-595-1

803

Figure 2. Volcano plots of DNA microarray data obtained

from retinol- (A) and bakuchiol- (B) treated skin

substitutes. The overall shapes of the two volcano plots are

similar, indicating that retinol and bakuchiol have similar

regulation on gene expression.

However, more evidence is still needed to support

this hypothesized pathway. It is also not clear what

are the targets of bakuchiol and how exactly they

regulate the expression of the retinol-storage-related

proteins. In this study, I will test a small portion of

the hypothesized pathway—bakuchiol can bind to

different targets than retinol. To test this hypothesis,

I will design an experiment to identify the bakuchiol

targeting proteins (BTPs) in cells.

I will perform bakuchiol affinity chromatography

on human dermal fibroblast lysates, followed by mass

spectrometry to sequence the BTPs. The use of

affinity chromatography assumes that modifications

at the 4-hydroxyl group of bakuchiol do not

significantly alter its anti-aging biological activity in

human skin cells. To test the assumption, I will

perform a structure-function study by substituting the

4-hydroxyl group of bakuchiol with methoxy group

and then comparing the biological activity of

bakuchiol and 4-methoxy-bakuchiol in stimulating

the expression of collagens by ELISA.

The result will provide hints for future researchers

to investigate the mechanism of action of bakuchiol.

2 MATERIALS AND METHODS

2.1 Structure-Function Study

2.1.1 Materials

Bakuchiol (INCI name), also known as Sytenol A

(trade name), will be purchased from Sytheon.

Sytheon derives bakuchiol from the plant Psoralea

corylifolia, which contains edible seeds that serve as

the source of bakuchiol. The plant itself is psoralene-

depleted bakuchiol with a purity of about 95%. The

Williamson-Ether method will be used to generate

methoxy-bakuchiol.

2.1.2 Synthesis of Methoxy-Bakuchiol

The Williamson-Ether method used to generate

methoxy-bakuchiol is as follows: 0.9 mmol/L

bakuchiol in acetone, treated with 4.8 mmol/L methyl

iodide, and 3.6 mmol/L potassium carbonate. The

solution will be stirred and refluxed for 24 hr;

afterward, there will be further room temperature

stirring for an additional 72 hr. The reaction will be

followed with thin-layer chromatography (25 ml of

methyl chloride and 3 ml of methanol). The acetone

and excess methyl iodide will be removed with a

stream of N2 gas. The remaining solid will be

resuspended in approximately 50 ml of ether and

vacuum filtered to remove the potassium carbonate.

The ether will be removed with N2 gas, leaving an

oily, yellow residue behind. This residue will be

resuspended in 50 ml of methanol and triturated with

water, then refrigerated (48C) overnight to produce

pale beige crystals of methoxy-bakuchiol. The

structural confirmation of methoxy-bakuchiol will be

determined through infrared spectrum (Nicolet IR

spectrometer, in KBr pellet) and a proton nuclear

magnetic resonance spectrum (Varian UnitccccPlus,

400 MHz, in DMSO). A sample of methoxy-

bakuchiol will be submitted for elemental analysis

(Desert Analytical, Tucson, AZ). The purity of the

sample will be assessed in two ways: 1) by measuring

its melting point, Rf on thin layer chromatography; 2)

by using high performance liquid chromatography

(Beckman Gold) that incorporates an isocratic 0.1%

trifluoro acetic acid:acetonitrile (50/50) solvent

system and a C18 reverse phase column at 0.8

ml/min.

2.1.3 Collagen ELISA

Bakuchiol and methoxy-bakuchiol will be assayed at

10 ug/mL on normal human fibroblasts grown in

DMEM with 5% calf serum (Hyclone, Salt Lake City,

UT, U.S.A.). Neonatal human dermal fibroblasts will

be used in analysis of Type I and IV collagen

quantification (low passage; American Type Culture

Collection, Manassas, VA, U.S.A. cat. no. PCS-201-

010, lot no. 58243223). Human epidermal fibroblasts

from a 68-year-old female donor will be used in

analysis of Type III collagen quantification (p. 5,

Zen-bio, cat. no. KR-F). Type I collagen

quantification requires cells to be subjected to test

ICBEB 2022 - The International Conference on Biomedical Engineering and Bioinformatics

804

materials for 3 days, whereas Type III and IV

collagen quantification requires cells to be subjected

to test materials for 7 days. Afterward, per standard

ELISA protocol, the sandwich ELISA, which uses

affinity-purified antibodies, will be used to assay the

collected cell culture conditioned media for type I,

type III or type IV collagen, which is then followed

by streptavidin-avidin-HRP conjugate and ABTS

(Dobak, Grzybowski, Liu et al 1994, Zhao, Alexeev,

Chang, et al. 2005). The collagen content is

proportional to a colorimetric signal. A BioRad

microplate spectrophotometer 3550-UV at 405 nm

with background subtraction at 660 nm will be used

to measure this colorimetric signal. Further analysis

of this colorimetric signal will be performed with

Microplate Manager v.2 software for Macintosh

(BioRad, Hercules, CA, U.S.A.).

2.2 Affinity Chromatography and

Mass Spectroscopy

2.2.1 Materials

Epoxy-activated agarose resin (12 atom linker, 33

µmol of epoxy group/ml of packed gel) will be

purchased from Sigma Chemical (St. Louis, MO).

Bakuchiol stock at a concentration of 12.5 mM will

be made in DMSO and stored at −20°C. Other

biochemical reagents will be procured from a variety

of chemical suppliers. The National Cell Culture

Center in Minneapolis, MN will source the large

amounts of cultured fibroblasts.

2.2.2 Preparation of Immobilized Bakuchiol

Affinity Column (BAC)

1 g of epoxy-activated agarose will be held in ice-cold

water for 5 min and thoroughly washed to get rid of

any additives or impurities. 23 mg of Bakuchiol

dissolved in 2.5 ml of 0.1 M NaOH will be added and

incubated with 1 ml of resuspended epoxy-activated

agarose for a night at room temperature to ensure that

the resin binds to the bakuchiol. 6 ml of 1 M sodium

acetate buffer (pH 5.0) containing 1 mM

dithiothreitol (DTT) will be added to the mixture to

neutralize unreacted epoxy groups and eliminate

further bakuchiol oxidation, thus halting the reaction.

Immobilized bakuchiol resin will be washed

successively with 0.1 M sodium acetate, pH 5.0,

containing 1 mM DTT and 70%, 30%, 10%, and 0%

ethanol, respectively. Mock-treated beads (which

utilize identical procedure except that there is no

added bakuchiol) or beads immobilized with a

tyrosine ligand will make up the controls.

2.2.3 Fractionation of Cytoplasmic Extracts

on BAC

Cultured mammalian cells will be lysed by 3 freeze–

thaw cycles with buffer containing 10 mM Hepes, pH

7.5, 90 mM KCl, 1.5 mM Mg (OAc)2, 1 mM DTT,

0.5% NP-40, 5% glycerol, 0.5 mM

phenylmethylsulfonyl fluoride (PMSF), and 10 µl/ml

of the protease inhibitor cocktail sourced from Sigma

Chemical. 10-min centrifugation using a refrigerated

microcentrifuge will yield the cell-free extracts.

Figure 3. Purification strategy for the isolation of BTPs

from BAC. In step 1, a cell lysate is passed over BAC or a

mock-coupled or tyrosine-linked control column in

parallel, and washed exhaustively with lysis buffer to

remove nonspecific proteins. In step 2, the column is eluted

stepwise eluted with 0.35 and 1 M NaCl. In step 3, column-

bound proteins are eluted using 1 mM ATP. The last step

involves elution with 1–2 mM bakuchiol dissolved in 2%

DMSO.

To characterize BAC (Figure 3), 200 µl fibroblast

extract containing 0.6–1.0 mg protein will be

combined with 50 µl control (mock-treated or

tyrosine-linked) or bakuchiol-immobilized agarose

beads in a 1.5-ml Eppendorf tube. The tube will be

stored overnight at 4 °C using a modest tumbling

process. The protein extract, which resembles a gel

slurry, will be loaded onto a minicolumn (from Pierce

Chemical) and rinsed with 10–20 ml lysis buffer to

eliminate any proteins that did not successfully bind.

Elution of the column will be next occur 5–7 times,

each time with 0.5 ml lysis buffer containing 0.35 M

NaCl, and will be followed by the same number of

rinsing using 1 M NaCl supplemented buffer. Next,

the column will be equilibrated with the lysis buffer

and eluted with 1 mM ATP. For the final step, the

column will once more be re-equilibrated with the

lysis buffer and eluted with 1–2 mM bakuchiol

dissolved in 2% DMSO.

Identiﬁcation of Bakuchiol Targeting Proteins in Human Skin Cells

805

2.2.4 Using MALDI-TOF MS to Identify

BTPs from Fibroblasts

We will use fibroblasts to prepare cytoplasmic

extracts which will be fractionated as described

(Figure 3). The middle silver-stained protein bands

will be excised, reduced, carbamidomethylated, and

cleaved with trypsin. The subsequent peptide mixture

will be desalted and concentrated using Zip-Tip C18

micro-columns (Millipore), and applied to the

MALDI target using solution phase nitrocellulose

method, as defined by Landry et at. (Landry,

Lombardo, Smith 2000). We will use a Voyager-DE

PRO MALDI-TOF mass spectrometer (Applied

Biosystems, Foster City, CA) set in reflector mode

with standard settings to calculate peptide masses.

Trypsin autodigestion products will be used for

internal mass calibration. MALDI-TOF MS

generates peptide mass fingerprinting which will be

compared to the theoretical tryptic digests of proteins

stored in NCBI nonredundant protein database using

the ProFound software

(http://prowl.rockefeller.edu/cgi-bin/ProFound).

This technique is used to identify proteins. The

identity of the BTPs will be obtained from the

database search result.

2.2.5 Statistical Analysis

The statistical significance of all numerical data will

be analyzed using the student’s T-Test on GraphPad

Prism® at (p <0.05).

3 POSSIBLE RESULTS

3.1 Confirmation of the Applicability

of Affinity Chromatography

Table 1 lists all the possible stimulatory effects of

bakuchiol and 4-methoxy bakuchiol on collagen I, II,

and IV.

Table 1. Possible stimulatory effects of bakuchiol and 4-

methoxy bakuchiol on collagen I, II, and IV. Bakuchiol and

4-methoy bakuchiol may have the same or different

stimulatory effects on each of collagen I, II, and IV. The

same stimulatory effect is defined as the two substances

decrease or increase the concentration of a collagen by the

same amount (p<0.05). In this table, the same stimulatory

effect is represented by “+”, and different stimulatory effect

is represented by “-“.

Collagen I Collagen III Collagen IV

Possible result 1 + + +

Possible result 2 + + -

Possible result 3 + - +

Possible result 4 + - -

Possible result 5 - + +

Possible result 6 - + -

Possible result 7 - - +

Possible result 8 - - -

3.2 Identification of Molecular Target

by Affinity Chromatography

Possible result 1: There is no silver-stained protein

shown in the gel.

Possible result 2: There is at least one silver-

stained proteins shown in the gel.

3.3 Identification of Molecular Target

by Mass Spectroscopy

Possible result 1: Bakuchiol does not bind RARs.

Possible result 2: Bakuchiol only binds RARs.

Possible result 3: Bakuchiol binds RARs and

other proteins.

4 DISCUSSION

4.1 Confirmation of the Applicability

of Affinity Chromatography

To secure and detect BTPs, we will immobilize

bakuchiol onto epoxy-activated agarose beads.

Residual bakuchiol found on column will be

eliminated with thorough rinsing. Chemical coupling

of bakuchiol occurs at the 4-hydroxyl group, which

yields a bakuchiol-immobilized affinity column

(BAC). However, it is not guaranteed that the

coupling between bakuchiol and epoxy-activated

agarose beads does not significantly affect the

antiaging biological activity of bakuchiol. We

therefore will perform a structure-function study to

examine if modifications at the 4-hydroxyl group of

ICBEB 2022 - The International Conference on Biomedical Engineering and Bioinformatics

806

bakuchiol alters the antiaging biological activity of

bakuchiol.

We synthesized 4-methoyl bakuchiol and will

compare its stimulatory effect with bakuchiol on the

expression of collagens in skin model.

The main components of the skin extracellular

matrix (type I and type III collagens) and basement

membrane (type IV collagen) include collagens

produced by dermal fibroblasts. Dermal fibroblasts

are reduced in amount and quality as found in aged

and photodamaged skin in addition to having less

new collagen pool. Consequently, we choose to

measure select collagens by ELISA method to

confirm the assumption that modifications at the 4-

hydroxyl group of bakuchiol does not alter the

antiaging biological activity of bakuchiol.

It is known from Chaudhuri and Bojanowski’s

work that bakuchiol can stimulate the amount of

collagen I, III, and IV in human dermal fibroblasts in

vitro (Chaudhuri, Bojanowski 2014). We therefore

choose collagen I, III, and IV to test our assumption.

Possible results are summarized in Table 1.

Possible result 1 the most strictly indicates that

modifications on the 4-hydroxyl group of bakuchiol

does not significantly alter its biological effect. As a

result, bakuchiol and 4-methoxyl bakuchiol has the

same stimulatory effect.

Possible results 2, 3, and 5 indicate that

modifications on the 4-hydroxly group of bakuchiol

can significantly alter its biological effect. As a result,

in the case of a substitution of 4-hydroxyl group with

4-methoxy group, the stimulation of one of the three

collagens has changed compared to bakuchiol.

Possible results 4, 6, and 7 indicate that

modifications on the 4-hydroxly group of bakuchiol

can alter its biological effect to a greater extent than

that of possible results 2, 3, and 5. As a result, in the

case of a substitution of 4-hydroxyl group with 4-

methoxy group, the stimulation of two of the three

collagens has changed compared to bakuchiol.

Possible result 8 indicates that the 4-hydroxyl

group of bakuchiol is necessary for it to exhibit its

biological effect, and that modifications on the 4-

hydroxyl group of bakuchiol can the most greatly

alter its biological effect. As a result, in the case of a

substitution of 4-hydroxyl group with 4-methoxy

group, the stimulation of all of the three collagens has

changed compared to bakuchiol.

The alteration of the biological effect of bakuchiol

as a result of modifications on the 4-hydryoxyl group

is most likely because the bakuchiol derivatives have

different binding affinities with biomolecules in the

cells. This can result in an alteration of the molecular

regulation, leading to an alteration of biological

effects.

If possible results 2-8 occur, the subsequent

affinity chromatography will not be applicable,

because the immobilization of bakuchiol at the 4-

hydroxyl group will affect the binding of bakuchiol

to putative BTPs. Other experimental method should

be designed to identify the molecular target of

bakuchiol.

If possible result 1 occurs, we are confident that

the subsequent affinity chromatography will be

applicable. The reason is that, as possible result 1

indicates, the immobilization of bakuchiol at the 4-

hydryoxyl group will not affect the binding of

bakuchiol to putative BTPs, and that the only position

of bakuchiol for epoxy-activated agarose beads to

bind is the 4-hydroxyl group. We can therefore claim

that column-bound bakuchiol should maintain similar

binding characteristics as bakuchiol in solution, and

thus affinity chromatography should be an applicable

method for the identification of the molecular target

of bakuchiol in photo-aged human skin cells.

4.2 Identification of Molecular Target

by Affinity Chromatography

Fibroblast lysates will be incubated with BAC and

eluted, as indicated in Materials and methods. We

will use SDS–PAGE to analyze eluted samples.

Furthermore, we will use silver staining to visualize

eluted samples. There may or may not be proteins that

show specific retention on BAC.

Possible result 1: There is no silver-stained

proteins.

The absence of silver-stained proteins suggests

that bakuchiol does not bind to any molecular target

in Fibroblasts. This is a very unlikely result given that

it have been proved that bakuchiol does have

biological affinity in photoaged human skin cells. If

this result occurs, there might be human errors during

the experiment, and further action should be done to

examine, e.g., if the cell lysates have active proteins.

Possible result 2: There is at least one silver-

stained proteins.

The presence of at least one silver-stained protein

suggests that bakuchiol has molecular target(s) in

fibroblasts, which bind to BAC and retain in the gel.

It is likely that some of the targets are involved in the

mechanism of action of the anti-aging activity of

bakuchiol. Next, we will perform mass spectroscopy

to reveal the identity of them.

Identiﬁcation of Bakuchiol Targeting Proteins in Human Skin Cells

807

4.3 Identification of Molecular Target

by Mass Spectroscopy

Silver-stained proteins will be excised from the gel

and subjected to in-gel trypsin digestion to generate

peptide fragments that will be further determined by

MALDITOF MS. This method will produce

recognizable patterns that will be used in database

searches to match predicted tryptic peptide masses of

proteins with known identity, thus leading to

determining what the silver-stained proteins are.

We hypothesized that bakuchiol binds to different

molecular targets than that of retinol (RARs), but

there are other possibilities.

Possible result 1: Bakuchiol does not bind RARs

and instead binds other proteins.

This result is consistent with our hypothesis that

bakuchiol can bind to different target proteins other

than that of retinol. By binding to different target

proteins, it is plausible that bakuchiol exhibits its

biological activity by facilitating the intracellular

storage of retinol, and it is retinol that ultimately

function to achieve the biological activity of

bakuchiol. This explains why bakuchiol has similar

but not exactly the same gene regulation pattern as

retinol, and why bakuchiol has antiaging activity as

retinol does.

The hypothesized pathway proposed in the

Introduction part of this paper suggests that there are

transcription factors that mediate the enhanced retinol

storage. Further studies should examine whether or

not the identified targets are the above-mentioned

transcription factors, or the identified targets can

activate those transcription factors. It also needs to be

clarified whether or not the above-mentioned

transcription factors mediate the expression of

retinol-storage-related proteins (e.g. CRBP II, CRBP

IV, CRABP I, LRAT).

In addition, the hypothesized pathway may not be

sufficient to describe the biological activity of

bakuchiol. Maybe some bakuchiol targeting proteins

facilitates cellular uptake and/or retinol activation

and/or intracellular transportation of retinol. These

possibilities also need to be tested in future studies.

Possible result 2: Bakuchiol only binds RARs.

This result is contradictive with our hypothesis. If

bakuchiol only binds RARs, theoretically bakuchiol

should have the same biological activity as retinol,

which contradicts with the different gene expression

pattern (Figure 2). Therefore, this is not a likely

result.

Possible result 3: Bakuchiol binds RARs and

other proteins.

This result is consistent with our hypothesis that

bakuchiol can bind to different target proteins other

than that of retinol. Therefore, everything discussed

in possible result 1 is also applicable in possible result

Furthermore, since bakuchiol can directly bind to

RARs, it is likely that bakuchiol can directly activate

the retinol-mediated gene regulation pathway. We

recommend conducting further studies to evaluate

this possibility.

5 CONCLUSION

Generally, this study tries to identify the molecular

target of bakuchiol in human skin cells. However, this

study assumes that the chemical coupling between

bakuchiol and epoxy-activated agarose beads does

not significantly affect the antiaging activity of

bakuchiol. If this assumption turns out to be false, the

subsequent affinity-chromatography is not

applicable, and other experimental approaches should

be designed to identify the molecular targets of

bakuchiol. If the assumption is true, most likely some

protein targets can be isolated and then identified by

mass spectroscopy. After the identification of target

proteins, we will be able to test the hypothesis that

bakuchiol can bind to proteins different than that of

retinol (i.e., RARs). After all, this hypothesis is not

likely to be false because Chaudhuri and Bojanowski

have revealed different gene expression patterns

regulated by bakuchiol and retinol (Chaudhuri,

Bojanowski 2014).

Further studies should be done to test the

hypothesized regulatory pathway of bakuchiol

proposed in the Introduction part of this paper.

Molecular details involved in this pathway needs to

be clarified. Further studies should also examine

whether or not the proposed pathway is sufficient to

explain the biological activity of bakuchiol. Overall,

after a series of studies, the mechanism of action of

bakuchiol can be elucidated.

REFERENCES

Chaudhuri, R. K., & Bojanowski, K. (2014). Bakuchiol: a

retinol‐like functional compound revealed by gene

expression profiling and clinically proven to have anti‐

aging effects. International journal of cosmetic science,

36(3), 221-230.

Chaudhuri, R. K., Sivamani, R., Jagdeo, J. R., Elsner, P., &

Maibach, H. I. (2015). Bakuchiol: a retinol-like

functional compound, modulating multiple retinol and

ICBEB 2022 - The International Conference on Biomedical Engineering and Bioinformatics

808

non-retinol targets. Cosmeceuticals and active

cosmetics, 1-18.

Dobak, J., Grzybowski, J., Liu, F.T. et al. 1,25-

Dihydroxyvitamin D3 increases collagen production in

dermal fibroblasts. J. Dermatol. Sci. 8, 18–24 (1994).

Landry F, Lombardo CR, Smith JW. A method for

application of samples to matrix-assisted laser

desorption ionization time-of-flight targets that

enhances peptide detection. Anal. Biochem. 2000;

279:1–8.

Zhao, H., Alexeev, A., Chang, E. et al. Lycium barbarum

glycoconjugates: effect on human skin and cultured

dermal fibroblasts. Phytomedicine 12, 131–137 (2005).

Identiﬁcation of Bakuchiol Targeting Proteins in Human Skin Cells

809