Characteristic differences in barrier and hygroscopic properties between normal and cosmetic dry skin. I. Enhanced barrier analysis with sequential tape-stripping

Characteristic differences in barrier and hygroscopic properties

between normal and cosmetic dry skin. I. Enhanced barrier

analysis with sequential tape-stripping

N. Lu, P. Chandar, D. Tempesta, C. Vincent, J. Bajor and H. McGuiness

Unilever Research and Development Trumbull, 40 Merritt Boulevard, Trumbull, CT 06611, U.S.A.

Received 9 October 2013, Accepted 2 January 2014

Keywords: barrier function, chemical analysis, claim substantiation in vivo/in vitro, dry skin, moisturization

Synopsis

OBJECTIVE: Cosmetic dry skin often has a lower hydration level

but a similar apparent barrier function, as measured by transepi-

dermal water loss (TEWL), than that of the normal skin. To

investigate the intrinsic difference in barrier property and moisture-

holding ability between the cosmetic dry and normal skin, we

developed a new clinical and data analysis procedure based on

sequential tape-stripping with TEWL measurement, coupled with

chemical analysis for protein and natural moisturizing factors

(NMF) in the stratum corneum.

METHODS: A clinical study consisting of 64 healthy Caucasian

female subjects with normal and cosmetic dry skin was conducted

according to our clinical and data collection protocols. After the

baseline visual dryness assessment, 20 tape-strips were placed and

removed on each test site using D-Squame tapes. TEWL was mea-

sured at baseline and after the 5th, 10th, 15th and 20th tape-

strips. All tapes were analysed for protein mass via chemical

extraction and the Pierce BCA protein assay, as well as using an

infrared densitometry device SquameScan 850A. The stratum cor-

neum thickness and barrier quality (water transport resistance per

thickness of the stratum corneum) were decoupled from the appar-

ent barrier function using the TEWL and protein data.

RESULTS: A linear relationship between 1/TEWL and cumulative

protein removal was observed for both normal and cosmetic dry

skin. However, the slope of the linear relation was significantly

steeper for normal skin, and significantly more protein was

removed from cosmetic dry skin. The results showed that on aver-

age, the barrier quality of the stratum corneum of the normal skin

is about 40% higher than that of the dry skin, whereas the stratum

corneum of the dry skin is about 30% thicker than that of the nor-

mal skin. In addition, the amount of SC removal in sequential

tape-stripping is generally non-uniform.

CONCLUSION: Our results demonstrated that there are character-

istic differences in the barrier property between normal and cos-

metic dry skin. In comparison to the normal skin, the stratum

corneum of the cosmetic dry skin is considerably thicker, however,

with a lower barrier quality. The results also showed that the

amount of the SC removal in sequential tape-stripping is generally

non-uniform. Therefore, the number of tape strips is not a good

indicator for the tape-stripping depth.

R�esum�eOBJECTIF: La peau s�eche (selon les crit�ere cosm�etiques) a souvent

un niveau d’hydratation plus faible, mais une fonction barri�ereapparente similaire, telle que mesur�ee par la perte insensible en eau

(PIE), �a celle de la peau normale. Pour �etudier la diff�erence int-

rins�eque entre la propri�et�e de barri�ere et la capacit�e de r�etention

d’eau entre la peau s�eche et la peau normale, nous avons d�evelopp�eune nouvelle proc�edure clinique et d’analyse des donn�ees sur la base

s�equentielle d’arrachement d’adh�esifs (‘tape-stripping) avec mesure

de la PIE, coupl�ee �a l’analyse chimique des prot�eine et des facteurs

d’hydratation naturels (NMF) dans le stratum corneum.

M�ETHODES: L’ �epaisseur de la couche corn�ee et de la qualit�e de

la barri�ere (r�esistance au transport de l’eau par l’�epaisseur de la

couche corn�ee) sont d�ecoupl�es de la fonction de barri�ere apparente.

Une �etude clinique comprenant 64 sujets f�eminins sains de type

caucasien avec la peau normale et la peau s�eche a �et�e effectu�ee

selon nos protocoles cliniques et de collecte de donn�ees. Apr�es l’�evaluation de base de la s�echeresse visuelle, 20 bandes D-Squame

ont �et�e plac�es et retir�es sur chaque site d’essai. Deux mesures

s�equentielles de PIE ont �et�e prises pour assurer un �etat stable a �et�e

prise apr�es la bande -froid. La quantit�e de prot�eine sur des bandes

D- Squame a �et�e analys�ee en utilisant un appareil de densitom�etrieinfrarouge SquameScan 850A. Toutes les bandes ont �egalement �et�e

analys�es pour la teneur en prot�eine par analyse chimique.

R�ESULTATS: Une relation lin�eaire entre 1/TEWL et l’enl�evement

cumulatif de prot�eine a �et�e observ�ee pour la plupart des sujets indi-

viduels dans cette �etude, ind�ependamment de leurs qualit�es de

s�echeresse visuels. En accord avec la conclusion par Kali et al. nos

r�esultats indiquent que malgr�e le caract�ere mol�eculaire h�et�erog�enedu SC, la r�esistance de transport de l’eau �a travers le SC semble etre

homog�ene. En outre, la meme conclusion vaut non seulement pour

les peaux normales, mais aussi pour les peaux s�eches. Les r�esultats

montrent que, en moyenne, la qualit�e de la fonction barri�ere de la

couche corn�ee de la peau normale est d’environ 40% sup�erieure �a

celle de la peau s�eche, tandis que la couche corn�ee de la peau s�eche

est d’environ 30% plus �epaisse que celle de la peau normale.

CONCLUSION: Les r�esultats pr�esent�es ici et dans les �etudespr�ec�edentes ont montr�e clairement que la quantit�e du SC arrach�ee

Correspondence: Nandou Lu, Unilever Research and Development,

Trumbull, 40 Merritt Boulevard, Trumbull, CT 06611, U.S.A. Tel.:

1 203 381 5442; fax: 1-203-381-5476; e-mail: Nandou.Lu@Unilever.

com

Part of this work was presented at International Investigative Dermatol-

ogy Conference, 8–11 May 2013, Edinburgh, Scotland.

© 2014 Society of Cosmetic Scientists and the Soci�et�e Franc�aise de Cosm�etologie 167

International Journal of Cosmetic Science, 2014, 36, 167–174 doi: 10.1111/ics.12112

s�equentiellement est de type non - uniforme. L’analyse s’appuyant

sur des chiffres du tape-stripping peut conduire �a des erreurs

importantes et meme �a des conclusions abusives et devrait donc

etre �evit�ee.

Introduction

Cosmetic dry skin (xerosis cutis) is one of the main problems con-

fronting dermatologists and skin researchers. Besides the lack of

proper hydration, it has long been speculated that cosmetic dry

skin may be associated with impaired barrier function caused by

imbalanced epidermal proliferation and differentiation, inadequate

lipid synthesis and defective processing of filaggrin and natural

moisturizing factors (NMF).

It is well known that stratum corneum (SC), the outmost layer

of the skin, constitutes the main permeability barrier of the skin to

water loss and the transport of other endogenous and exogenous

chemicals. In clinical studies, the barrier function typically is

assessed by measuring the transepidermal water loss (TEWL) [1–4].Diseased skin conditions, such as atopic dermatitis, psoriasis and

contact dermatitis, are associated with inferior barrier function [5]

as characterized by abnormally high levels of TEWL. However, the

relation between TEWL and skin dryness condition remains incon-

clusive [6, 7]. The baseline TEWL of cosmetic dry skin typically

shows very small or even no difference from that of normal skin.

In addition, studies of the long-term effects of moisturizer on the

barrier function of normal and dry skin as measured by baseline

TEWL have led to inconsistent conclusions [8–12]. Such observa-

tions have raised the question of whether dry skin has intrinsically

different barrier properties from normal skin besides its superficial

(visual) problem.

In this study, our primary objective was to investigate whether

there are characteristic differences in the SC barrier property

between normal and dry skin. It is clear that baseline TEWL mea-

surement alone is insufficient for distinguishing the barrier differ-

ence between normal and dry skin, because TEWL only represents

an apparent measure of skin’s overall resistance to the passive water

diffusion through the skin. TEWL is a function of multiple factors.

In addition to the skin barrier function, both internal (physiologi-

cal, such as blood flow and skin temperature) and external (tem-

perature, humidity) conditions can affect TEWL significantly.

Clinical studies typically are conducted in a controlled environ-

ment, on the same designated body sites, and with subjects remain-

ing at rest condition during the measurement [2, 13], thus

minimizing the difference in both internal and external conditions.

Nevertheless, baseline TEWL alone is not sufficient to quantify the

intrinsic barrier function of the skin, which is determined by the

molecular composition and organization of intracellular lipids, nat-

ural moisturizing factors, desquamation enzymes and biological

and physiological processes in viable skin layers [14–22].However, it is possible to extract useful information about SC

biophysical properties that phenomenologically determine the skin

barrier function by performing TEWL measurement along with

sequential tape-stripping. Kalia et al. [23] first developed a proce-

dure that allows the decoupling of water diffusivity and SC thick-

ness from sequential tape-stripping and TEWL measurement. This

method has been applied to studies of the barrier function and SC

thickness of normal skin [24, 25], but it has not been applied to

investigate the difference in barrier property between normal and

cosmetic dry skins. The sequential tape-stripping and TEWL analy-

sis requires knowledge of the amount of SC removal by the tapes.

In the original method, tapes were weighed both prior to and after

the tape-stripping in order to determine the amount of SC removed.

This process is time-consuming and thus is not practical for clinical

studies with a large number of subjects. In this clinical study, we

instead detected the amount of protein removed by tape-stripping

and used it to provide an estimate for the amount of SC removal.

Accordingly, the original barrier analysis method of Kalia et al.

was modified to allow the analysis for relative barrier property dif-

ferences between the normal and dry skin.

Materials and methods

Clinical design and execution

The study consisted of a 5–7-day conditioning phase followed by a

1-day baseline and challenge assessment phase. Informed consent

was obtained from all subjects, and the study was approved by an

Institutional Review Board. The clinical study was conducted over

2 weeks from early to middle December.

The clinical subjects consisted of 64 Caucasian females, aged

35–50 years, with either visually normal or dry skin at the mid-

dle of the outer legs. These subjects were free from any skin dis-

ease, had no history of hormone replacement therapy and were

not atopic. No moisturizer was used on the test site 7 days prior

to the study. All subjects refrained from washing/cleaning the test

site 24 h before the study. The subjects acclimated for at least

20 min in an environmentally controlled room (17–23°C, relativehumidity 25–45%) before any visual and instrumental assess-

ments were obtained. The visual dryness and erythema of the test

sites were assessed using a 5-point scale (0 to 4) in 0.5 incre-

ments where 0 indicates none and four indicates severe dryness

[26]. Table 1 summarizes the number and age of subjects of each

dryness grade.

After the baseline visual dryness assessment, 20 tape-strips were

placed and removed on each test site using D-Squame tapes

(CuDerm, Dallas, TX, U.S.A.) with a diameter of 22 mm (area

3.8 cm2). Each D-Squame tape was placed on the skin under

225 g cm�2 of pressure for 5 s using the D500 D-Squame pressure

applicator (CuDerm, Dallas, TX, U.S.A.). The D-Squame tapes were

peeled from different directions (every 90°) in rotation until all 20

samples had been obtained. TEWL values were measured at baseline

and after the 5th, 10th, 15th and 20th sequential tape strips using a

cyberDERM RG1 Evaporimeter (Broomall, PA, U.S.A.) with TEWL

probes manufactured by Cortex Technology (Hadsund, Denmark).

Two sequential TEWL measurements were made to ensure a stable

condition was reached following tape-stripping, and the average was

used in further analysis. Safety checks for erythema were conducted

prior to every tape-strip and after the final tape strip.

Table 1 Summary of the number, age and visual dryness grade of subjects

Visual dryness

grade

Number of

subjects

Age range

(years)

Median age

(years)

0 12 35–45 39.5

0.5 19 37–50 41

2 8 35–50 42

2.5 15 35–50 44.5

3 10 38–50 45.5

168 © 2014 Society of Cosmetic Scientists and the Soci�et�e Franc�aise de Cosm�etologie

International Journal of Cosmetic Science, 36, 167–174

Intrinsic barrier property of normal and dry skin N. Lu et al.

Protein analysis

The amount of protein on D-Squame tapes was analysed using an

infrared densitometry device SquameScan 850A (Heiland Elec-

tronic, Wetzlar, Germany) [27]. All tapes also were analysed for

protein content using chemical analysis, as described below.

Free amino acids (FAA) and proteins on the tapes were obtained

using a series of extractions as described by Dreher et al. [28, 29].

During the first extraction, 1.8 mL of 6 mM HClO4 was added to

the tapes in the initial 96-Deepwell plate (DWP) (Greiner from

VWR, West Chester, PA, U.S.A.) and incubated 48 h at room tem-

perature to extract FAA. After recovering the extraction into a sec-

ond Deepwell plate for FAA analysis, the original plate was then

rinsed with water to neutralize the pH from any residual HClO4

that may have been present. Protein was extracted by adding

1.8 mL of 0.1 M NaOH and then incubating for 2 h at 60°C. TheTecan Genesis liquid handling system was used in the addition of

extraction solvents, washing of Deepwell plate as well as the trans-

fer of material from one Deepwell plate to another. The NaOH

extract was analysed using the Pierce BCA assay kit (Rockford, IL,

U.S.A.) for proteins. Samples were quantified by reading the absor-

bance at 562 nm wavelength.

Sequential tape-stripping and TEWL analysis

At steady state, TEWL can be expressed using Fick’s first law of dif-

fusion:

TEWL ¼ KDCDh

; ð1Þ

where K is the partition coefficient of water between SC and the

underneath viable epidermis, ΔC is the difference in water con-

centration between the skin surface and the bottom of the SC, D

is the apparent diffusivity of water in the SC and h is the SC

thickness.

Kalia et al. [23] first demonstrated that during progressive

removal of the SC by repeated tape-stripping, Equation (1) can be

rewritten as

TEWL�1x ¼ ðh� xÞ

KDC�Dx; ð2Þ

where TEWLx is the TEWL value when thickness x of SC has

been removed, and �Dx is the average apparent diffusivity of water

through the remaining (h�x) of the SC. Assuming TEWLx mea-

surement was performed after the skin reaches a stable condition

following tape-stripping, KΔC can reasonably be assumed to be

constant [23]. Kalia et al. further found that �Dx is independent of

h � x (i.e. uniform across the SC depth) which was also observed

in other studies [24, 25]. In such a case, Equation (2) represents

a straight line in a TEWL�1x vs. x plot. The SC thickness h and

the average apparent diffusivity �D can be obtained from a linear

fitting or graphic analysis of the TEWL�1x vs. x plot (the intercept

at the x-axis gives h, and the �1/slope is proportional to �D).

The thickness x of SC removal is approximately proportional to

the SC protein removed by tape-stripping as the protein in the cor-

neocytes is the major component of the SC [30]. (For an excellent

summary of methods for determining the amount of SC removal,

including methods based on protein, see [31] and references

therein.) Thus, in terms of protein removal, Equation (2) can be

written as

TEWL�1p ¼ ðP� pÞ

aKDCD; ð3Þ

where p is the cumulative protein removal by sequential tape-strip-

ping, and P is the total protein mass for the entire SC under the

tape-stripping area. a is the proportional constant, p = ax. A linear

fitting or graphic analysis for TEWL�1p , and p relation leads to esti-

mates for the total protein P and the group parameter aKDCD.In modelling diffusion transport, Equation (1) is also commonly

expressed in terms of (total) water transport resistance R:

TEWL = KΔC/R, where R = h/�D, or in terms of resistivity~R : TEWL ¼ KDC=ðh~RÞ [32]. The resistivity ~R is the reciprocal of

the diffusivity �D, reflecting water transport resistance per unit SC

thickness (i.e. R/h). Relating it to SC barrier function, here, we call

the resistivity ~R as barrier quality. The decoupling of the SC thick-

ness and barrier quality from the apparent TEWL allows better

characterization of SC barrier function that by TEWL alone. For

example, under a given environmental condition, a lower baseline

TEWL may be the result of higher barrier quality, or thicker SC or

both.

Assuming aKΔC is the same for different subjects (or subject

groups) under the same measurement condition (room temperature

and humidity), the relative barrier quality among the subjects (or

subject groups) can be obtained by comparing the slopes of the

TEWL�1p and p relation for each subject (or subject group). The rel-

ative SC thickness can be determined from the values of cumulative

protein removal p that can be quantified either by parameter fitting

or the intercept at x-axis.

The analysis based on Equations (2) or (3) has some drawbacks,

including the overestimate of the SC thickness (or total SC protein)

as 1/TEWL does not reach 0 (intercepting at x-axis) when all the

SC is removed from the skin. This can be improved via, for exam-

ple, using a modified model equation and a nonlinear fitting proce-

dure of Russell et al., [33] or including the mass transport

resistance in the air phase above the skin [34] to account for the

fact that water loss rate from skin approaches a finite value equal

to the free-water evaporation when all the SC is removed. The

main goal of this study was to determine the relative difference in

SC thickness and barrier quality between normal and cosmetic dry

skin, for which the simple analysis presented above is sufficient.

Results and discussion

The 64 subjects were placed into groups with visual dryness grades

of 0, 0.5, 2, 2.5 and 3, with each group having 12, 19, 8, 15 and

10 subjects, respectively. Table 1 summarizes the number and age

of subjects of each dryness grade. Thirty-one subjects had normal

skin (visual dryness grade of 0 or 0.5), and 33 subjects had moder-

ately dry skin (visual dryness grades of 2–3). For all the subjects,

the baseline TEWL on the test site was less than 10.0 g m�2 h�1.

Protein removal and visual dryness

The protein mass (in lg) removed by each D-Squame tape was

individually analysed using the BCA protein assay, and cumulative

protein mass removed during the sequential tape-stripping was cal-

culated for each subject. The average cumulative amount of pro-

tein removal (under the tape-stripping area of 3.8 cm2) as a

function of the tape-strip number for each visual dryness group,

and for normal (visual dryness grade 0–0.5) and dry (2.0–3.0) sub-jects is presented in Figs 1 and 2, respectively. From both figures, it

is evident that on average, more protein (hence more SC materials)




was removed by equal numbers of tape strips from skin with higher

visual dryness grades. Also, note that the slopes of the curves grad-

ually decrease as the number of tape-strips increases, reflecting the

fact that on average, less SC is removed by tape-stripping from the

deeper region of the SC. Such non-uniform SC removal in sequen-

tial tape-stripping can be clearly seen in Fig. 3, where average pro-

tein levels in each sequential tape for different visual dryness

groups are plotted against the number of tape-strips.

Using the number of tape strips as a representation of the SC

removal (and SC depth) was a common practice in sequential

TEWL and tape-stripping analyses. Such an analysis implies an

assumption that each tape removes the same amount of SC. Our

results (Fig. 3) clearly showed that SC removal in sequential tape-

stripping is typically non-uniform, especially for the first several

tapes and for subjects with dry skin. The non-uniform SC removal

in sequential tape-stripping also has been reported in several pre-

vious studies [23, 35–37]. Therefore, the practice of using the

number of tape-strips for SC removal is not optimal and can even

lead to misleading conclusions. Here, we emphasize the importance

of using a proper quantity (e.g. protein level or SC mass) in the

analysis of sequential tape-stripping data.

Comparison of SquameScan and BCA protein assay results

Several studies have demonstrated linear correlations between

SquameScan pseudo-absorption at 850 nm and protein level deter-

mined using BCA assay for human forearm [27, 35, 38] and por-

cine ear skin [39]. The reported protein mass removed by each

tape strip was typically below 50 lg cm�2 (as detected by BCA

protein assay). In our study, the protein mass removed by tape-

stripping, particularly for the first two tapes, was generally higher

than those reported in previous studies. This difference is likely due

to the difference in anatomical site – our tape-stripping samples

were collected at the middle of the outer legs. Fig. 4 presents the

comparison between SquameScan pseudo-protein absorption at

850 nm and the protein mass per area detected using the BCA pro-

tein assay. Our results suggest that 50–60 lg cm�2 or 25–30%absorption at 850 nm roughly represents the upper limit for a lin-

ear correlation between the results of these two methods. As the

amount of protein removal per tape goes higher, the sensitivity of

the pseudo-protein absorption to the protein mass is reduced. To

better illustrate the results of individual subjects, in Fig. 4(b), a

comparison for the SquameScan absorption and BCA protein assay

was made for a subset of 14 subjects (6 with normal skin and 8

with dry skin) where each subject is specified by a unique maker

and colour combination. As one can see, there are subject-to sub-

ject-variations that can be large in some cases. Concerned that a

large number of tape-stripping samples may fall outside of the opti-

mal working range of SquameScan, only the protein levels detected

by BCA assay were used in our further analyses.

Figure 1 Average cumulative amount of protein removed by D-Squame

tapes during sequential tape-stripping for subject groups with visual dryness

grades of 0, 0.5, 2.0, 2.5 and 3.0. The protein mass corresponds to the area

of tape-stripping (3.8 cm2). Data shown as mean � standard error.

Figure 2 Average cumulative amount of protein removed during sequential

tape-stripping for subject groups with normal and dry skin. Normal skin has

a visual dryness grade of 0 or 0.5; dry skin has a visual dryness grade of

2.0–3.0. The protein mass corresponds to the area of tape-stripping

(3.8 cm2). Data shown as mean � standard error.

Figure 3 Average amount of protein removed in each tape-strip for subject

groups with different visual dryness grades. The protein mass corresponds to

the area of tape-stripping (3.8 cm2). Data shown as mean � standard error.




TEWL and barrier property analysis

Results of 1/TEWL vs. cumulative protein removal (p) during the

sequential tape-stripping are presented in Figs. 5a and 5b for all

individual subjects with normal and dry skin, respectively. A linear

relationship between 1/TEWL and cumulative protein removal was

observed for most of the individual subjects in this study, regardless

of their visual dryness grades. In agreement with the conclusion by

Kalia et al. [23], our results indicate that in spite of the heteroge-

neous molecular character of the SC, the water transport resistance

across the SC appears to be homogeneous. Furthermore, the same

conclusion holds not only for normal skin, but also for dry skin. It

is interesting to note that in general, we do not observe different

slopes in 1/TEWL – cumulative protein plots for tape-stripping near

the skin surface and deeper in the SC, for both normal and dry

skin. An initial plateau (a slope close to 0) was thought to exist

near the skin surface for dry skin due to loose outer SC layers that

have an insignificant contribution to the barrier function. No such

a plateau was observed in this study.

The group average of 1/TEWL – cumulative protein relationships

is presented in Fig. 6 for normal and dry skin groups. A linear rela-

tionship between the inverse of TEWL and cumulative protein

removal again was observed for each subject group. Note that the

slopes in Fig. 6 correspond to �~R/aKΔC; thus, a steeper (more neg-

ative) slope indicates a higher SC barrier quality.

To further study the differences in the SC barrier quality and its

thickness, we performed the least-square linear fitting for the

sequential 1/TEWL and cumulative protein results of each individ-

ual subject and computed the corresponding slope and x-intercept.

Results of the average (negative) slope and the total SC protein

mass are presented in Figs 7 and 8, respectively, for subject groups

with different visual dryness conditions. As mentioned earlier, the

absolute value of the slope is proportional to the barrier quality ~R,

while the x-intercept of the linear extrapolation provides an esti-

mate for the total protein mass under the tape-stripping area which

is proportional to the SC thickness. From Fig 7, it is obvious that

the absolute value of the slope decreases as the visual dryness

grade increases and that normal skin has a higher absolute value

of the slope than the dry skin. Meanwhile, the total SC protein

mass increases as the visual dryness grade becomes higher (Fig 8).

The difference between normal and dry skin was statistically signifi-

cant for both the slope (P < 0.0001) and the total SC protein mass

(P < 0.0001) based on unpaired two-tail t-tests. The differences in

the slope also were significant (P < 0.05) between groups with

visual dryness grades differing by 1.0 or higher. Such results

clearly indicate that on average, the barrier quality of the normal

a b

Figure 4 Comparison between SquameScan pseudo-protein absorption (%) at 850 nm and protein mass detected using the BCA assay (in lg cm�2) for each

tape. Plot a: results for all 64 subjects; plot b: results for a subset of 14 subjects (6 with normal skin and 8 with dry skin) each identifiable by different colour

and marker.

a b

Figure 5 Plots of 1/TEWL vs. cumulative protein (p) removed during sequential tape-stripping for all subjects with normal (a) and dry skin (b). The cumula-

tive protein mass corresponds to the area of tape-stripping (3.8 cm2).




skin is significantly higher than that of the dry skin, whereas dry

skin has a significantly thicker SC. Quantitatively, the average

slopes for the 1/TEWL vs. p plot were �5.8 9 10�5 and

�3.4 9 10�5 for normal and dry skin, respectively; and the

x-intercepts were 2858 and 3985 for normal and dry skin, respec-

tively. Such results indicated that the barrier quality of normal skin

is about 40% higher than that of the dry skin, whereas the SC of

the dry skin is about 30% thicker than that of the normal skin.

Concluding remarks

The apparent barrier function of cosmetic dry skin, as measured by

baseline TEWL, often shows no significant difference from that of

normal skin. In this study, we developed a clinical method that

couples TEWL measurement along with sequential tape-stripping

and protein analysis, and revealed that there is a significant differ-

ence in barrier property between the cosmetic dry skin and normal

skin. The stratum corneum of the cosmetic dry skin was consider-

ably thicker than that of the normal skin; however, the barrier

quality of the dry skin was intrinsically lower. Our findings support

the earlier speculation that dry skin may be associated with intrin-

sically inferior barrier function. The conclusion that dry skin has a

thicker SC also supports the hypothesis that dry skin may result

from the imbalance between the proliferation and differentiation

processes – a hyperproliferation condition that can lead to thicker

but premature SC formation which results in poor barrier quality

[40, 41].

The decoupling of the SC thickness and barrier quality from

TEWL measurement during sequential tape-stripping is the key to

our analysis. Measuring TEWL during sequential tape-stripping is a

minimally invasive method that is frequently performed in clinical

studies. However, a good estimate of the amount of SC removal by

tape-stripping is practically difficult in clinical studies with a large

number of subjects. As a compromise, the number of tape-strips

often was used as a marker for tape-stripping depth. The results

reported here and in previous studies [23, 35–37] clearly showed

that the amount of SC removal in sequential tape-stripping is typi-

cally non-uniform. Analysis relying on tape strip numbers can lead

to large errors and even improper conclusions [42, 43], and thus

should be avoided.

We overcame the obstacle of estimating actual SC removal by

tape-stripping by using the BCA assay for protein level on the

tapes, which was performed after the completion of the clinical

study. This method does not increase the workload during the clin-

ical study, making it feasible for studies with a large number of

subjects. The same tapes collected during the clinical also were

Figure 6 Group average of 1/TEWL vs. cumulative protein mass on tapes

for normal (visual dryness grades 0 and 0.5, line with open circle) and dry

(visual dryness grades 2.0, 2.5 and 3.0 line with open square) subjects. The

cumulative protein mass corresponds to the area of tape-stripping (3.8 cm2).

The error bar represents standard error.

Figure 7 The group average of slope in Equation (3) (in absolute value, i.e.

ðaKDCDÞ�1 which is proportional to barrier quality) for subject groups of dif-

ferent visual dryness grade, and of normal and dry skin. The error bar repre-

sents standard error. The difference in the average slopes was statistically

significant (P < 0.05) based on an unpaired two-tail t-test for any pair of

groups with visual dryness grades differing by 1 unit. The difference between

normal and dry skin groups is also significant (P < 0.0001).

Figure 8 The average estimated total protein mass for the SC (under the

area of tape-stripping, 3.8 cm2) for subject groups of different visual dryness

grade, and of the normal and dry skin. The total SC protein mass was taken

as the intercept at the x-axis by linear extrapolation for 1/TEWL vs. cumula-

tive protein plot. The error bar represents the standard error. The difference

in total SC protein mass was statistically significant between normal and dry

skin groups (P < 0.0001) based on an unpaired two-tail t-test, and also

between different visual dryness groups (P < 0.05) except the group pairs

with visual dryness grades of (0, 0.5), (0.5, 2.0), (2.0, 3.0) and (2.5, 3.0).




analysed for the contents of free amino acids that are components

of natural moisturizing factors, thus providing further insights into

the difference between normal and cosmetic dry skin. This part of

the analysis will be presented in the second paper of this series.

Acknowledgements

This work is fully funded by Unilever Research and Development.

The authors thank Dr. Stephen Madison for insightful discussion.

References

1. Distante, F. and Berardesca, E. Transepider-

mal water loss. In: Bioengineering of the Skin:

Methods and Instrumentation (Berardesca, E.,

Elsner, P., Wilhelm, K.P. and Maibach, H.I.

eds.), pp. 1–4. CRC Press, Boca Raton (1995).

2. Pinnagoda, J., Tupker, R.A., Agner, T. and

Serup, J. Guidelines for transepidermal

water loss (TEWL) measurement. Contact

Dermatitis 22, 164–178 (1990).

3. Lotte, C., Rougier, A., Wilson, D.R. and Mai-

bach, H.I. In vivo relationship between

transepidermal water loss and percutaneous

penetration of some organic compounds in

man: effect of anatomic site. Arch. Dermatol.

Res. 279, 351–356 (1987).

4. Nangia, A., Berner, B. and Maibach, H.I.

Transepidermal water loss measurements

for assessing skin barrier function during

in vitro percutaneous absorption experi-

ments. In: Percutaneous Absorption: Drugs,

Cosmetics, Mechanisms, Methodology, 3rd

edn. (Bronaugh, R.L. and Maibach, H.I.

eds.), Marcel Dekker, New York (1999).

5. Denda, M. Methodology to improve epider-

mal barrier homeostasis: how to accelerate

the barrier recovery? Int. J. Cosmet. Sci. 31,

1–8 (2009).

6. Fischer, T., Wigger-Alberti, W. and Elsner,

P. Assessment of ‘dry skin: current bioengi-

neering methods and test designs. Skin Phar-

macol. App.l Skin Physiol. 14, 183–195

(2001).

7. Yosipovitch, G. Dry skin and impairment of

barrier function associated with itch - new

insights. Int. J. Cosmet. Sci. 26, 1–7 (2004).

8. Buraczewska, I., Berne, B., Lindberg, M.,

Torma, H. and Loden, M. Changes in skin

barrier function following long-term treat-

ment with moisturizers, a randomized con-

trolled trial. Br. J. Dermatol. 156, 492–498

(2007).

9. Ganemo, A., Virtanen, M. and Vahlquist, A.

Improved topical treatment of lamellar ich-

thyosis: a double-blind study of four differ-

ent cream formulations. Br. J. Dermatol.

141, 1027–1032 (1999).

10. Held, E., Sveinsdottir, S. and Agner, T. Effect

of long-terms use of moisturizer on skin

hydration, barrier function and susceptibility

to irritants. Acta Derm. Venereol. 79, 49–51

(1999).

11. Loden, M. Barrier recovery and influence of

irritant stimuli in skin treated with a

moisturizing cream. Contact Dermatitis 36,

256–260 (1997).

12. Loden, M., Andersson, A.C. and Lindberg, M.

Improvement in skin barrier function in

patients with atopic dermatitis after treat-

ment with a moisturizing cream (Canoderm).

Br. J. Dermatol. 140, 264–270 (1999).

13. Rogiers, V. and EEMCO group. EEMCO guid-

ance for the assessment of transepidermal

water loss in cosmetic sciences. Skin Pharma-

col. App.l Skin Physiol. 14, 117–128 (2001).

14. Harding, C.R. and Scott, I.R. Stratum corne-

um moisturizing factors. In: Skin Moisturiza-

tion (Leyden, J.J., Rawlings, A.V., eds.), pp.

61–80. Marcel Dekker, New York (2002).

15. Tabachnick, J. and LaBadie, J.H. Studies on

the biochemistry of epidermis. IV. The free

amino acids, ammonia, urea, and pyrroli-

done carboxylic acid content of conventional

and germ-free albino guina pig epidermia. J.

Invest. Dermatol. 54, 24–31 (1970).

16. Egelrud, T. Purification and preliminary

characterization of stratum corneum chymo-

tryptic enzyme-A proteinase that may be

involved in desquamation. J. Invest. Derma-

tol. 101, 200 (1993).

17. Elias, P.M. and Friend, D.S. The permeability

barrier in mammalian epidermis. J. Cell Biol.

65, 180–191 (1975).

18. Elias, P.M. Epidermal lipids, barrier function

and desquamation. J. Invest. Dermatol. 80,

44–49 (1983).

19. Elias, P.M. The importance of epidermal lip-

ids for the stratum corneum barrier. In: Top-

ical Drug Delivery Formulations (Osbourne, D.

and Amann, A., eds.), Marcel Dekker, New

York (1990).

20. Lampe, M.A., Burlingame, A.L., Whitney, J.,

Williams, M.L., Brown, B.E., Roitman, E.

and Elias, P.M. Human stratum corneum

lipids: characterisation and regional varia-

tions. J. Lipids Res. 24, 120–130 (1983).

21. Norlen, L., Nicander, I., Lundh, R., Ollmar,

S. and Forslind, B. Inter- and intra-individ-

ual differences in human stratum corneum

lipid content related to physical parameters

of skin barrier function in vivo. J. Invest. Der-

matol. 112, 72–77 (1998).

22. Rawlings, A.V. and Harding, C.R. Moistur-

ization and skin barrier function. Dermatol.

Therapy. 17, 43–48 (2004).

23. Kalia, Y.N., Pirot, F. and Guy, R.H. Homoge-

neous transport in a heterogeneous mem-

brane: water diffusion across human

stratum corneum in vivo. Biophys. J . 71,

2692–2700 (1996).

24. Pirot, F., Berardesca, E., Kalia, Y.N., Singh,

M., Maibach, H.I. and Guy, R.H. Stratum

corneum thickness and apparent water dif-

fusivity: facile and noninvasive quantifica-

tion in vivo. Pharm. Res. 15, 492–494

(1998).

25. Schwindt, D.A., Wilhelm, K.P. and Maibach,

H.I. Water diffusion characteristics of human

stratum at different anatomical sites in vivo. J.

Invest. Dermatol. 111, 385–389 (1998).

26. Sharko, P.T., Murahata, R.I., Leyden, J.J.

and Grove, G.L. Arm wash with instrumen-

tal evaluations - a sensitive technique for

differentiating the irritation potential of per-

sonal washing products. J. Dermal. Clin.

Eval. Soc. 2, 19–27 (1991).

27. Voegeli, R., Heiland, J., Doppler, S., Rawlings,

A.V. and Schreier, T. Efficient and simple

quantification of stratum corneum proteins

on tape strippings by infrared densitometry.

Skin Res. Tech. 13, 242–251 (2007).

28. Dreher, F., Modjtahedi, B.S., Modjtahedi,

S.P. and Maibach, H.I. Quantification of

stratum corneum removal by adhesive tape

stripping by total protein assay in 96-well

microplates. Skin Res. Tech. 11, 97–101

(2005).

29. Dreher, F., Arens, A., Hostynek, J.J.,

Mudumba, S., Ademola, J. and Maibach,

H.I. Colorimetric method for quantifying

human stratum corneum removed by adhe-

sive-tape-stripping. Acta Derma. Venereol 78,

186–189 (1998).

30. Boncheva, M., de Sterke, J., Caspers, P.J.

and Puppels, G.J. Depth profiling of stratum

corneum hydration in vivo: a comparison

between conductance and confocal Raman

spectroscopic measurements. Exp. Dermatol.

18, 870–876 (2009).

31. Lademann, J., Jacobi, U., Surber, C., Weig-

mann, H.J. and Fluhr, J.W. The tape strip-

ping procedure - evaluation of some critical

parameters. Eur. J. Pharm. Biopharm. 72,

317–323 (2009).

32. Cussler, E.L. Diffusion Mass Transfer in Fluid

Systems, 2nd edn. Cambridge University

Press, Cambridge, U.K. (1997).

33. Russell, L.M., Wiedersberg, S. and Delgado-

Charro, M.B. The determination of stratum

corneum thickness – an alternative




approach. Eur. J. Pharm. Biopharm. 69,

861–870 (2008).

34. Imhof, R.E., De Jesus, M.E., Xiao, P., Ciortea,

L.I. and Berg, E.P. Closed-chamber transepi-

dermal water loss measurement: microcli-

mate, calibration and performance. Int. J.

Cosmet. Sci. 31, 97–118 (2009).

35. Mohammed, D., Yang, Q., Guy, R.H., Matts,

P.J., Hadgraft, J. and Lane, M.E. Comparison

of gravimetric and spectroscopic approaches

to quantify stratum corneum removed by

tape-stripping. Eur. J. Pharm. Biopharm. 82,

171–174 (2012).

36. Klang, V., Schwarz, J.C., Lenobei, B., Nadj,

M., Aubock, J., Wolzt, M. and Valenta, C.

In vitro vs. in vivo tape stripping: validation

of the porcine ear model and penetration

assessment of novel sucrose stearate emul-

sions. Eur. J. Pharm. Biopharm. 80, 604–

614 (2012).

37. Mohammed, D., Crowther, J.M., Matts, P.J.,

Hadgraft, J. and Lane, M.E. Influence of nia-

cinamide containing formulations on the

molecular and biophysical properties of the

stratum corneum. Int. J. Pharm. 441, 192–

201 (2013).

38. Hahn, T., Hansen, S., Neumann, D., Kostka,

K.H., Lehr, C.M., Muys, L. and Schaefer,

U.F. Infrared densitometry: a fast and non-

destructive method for exact stratum corne-

um depth calculation for in vitro tape-strip-

ping. Skin Pharm. Physiol. 23, 183–192

(2010).

39. Klang, V., Schwarz, J.C., Hartl, A. and Valenta,

C. Facilitating in vitro tape stripping: applica-

tion of infrared densitometry for quantification

of porcine stratum corneum proteins. Skin

Pharm.Physiol.24,256–268(2011).

40. Jackson, S.M., Williams, M.L., Feingold, K.R.

and Elias, P.M. Pathobiology of the stratum

corneum. West. J. Med. 158, 279–285

(1993).

41. Engelke, M., Jensen, J.-M., Ekanayake-Mudi-

yanselage, S. and Proksch, E. Effects of

xerosis and ageing on epidermal prolifera-

tion and differentiation. Br. J. Dermatol.

137, 219–225 (1997).

42. Weigmann, H.J., Lademann, J., Meffert, H.,

Schaefer, H. and Sterry, W. Determination

of the horny layer profile by tape stripping

in combination with optical spectroscopy in

the visible range as a prerequisite to quan-

tify percutaneous absorption. Skin Pharma-

col. Appl. Skin Physiol. 12, 34–45 (1999).

43. Marttin, E., Neelissen-Subnel, M.T.A., De

Haan, F.H.N. and Bodde, H.E. A critical

comparison of methods to quantify stratum

corneum removed by tape stripping. Skin

Pharmacol. Appl. Skin Physiol. 9, 69–77

(1996).




Documents

Characteristic differences in barrier and hygroscopic properties between normal and cosmetic dry skin. I. Enhanced barrier analysis with sequential tape-stripping