Arch Dermatol Res (1999) 291 : 47–53
O R I G I N A L PA P E R
Nanna Y. Schürer · Frank Rippke · Kathrin Vogelsang ·
Viola Schliep · Thomas Ruzicka
Fatty acid uptake by cultured human keratinocytes grown
in medium deficient in or supplemented with essential fatty acids
Received: 10 February 1997 / Received in revised form: 10 September 1998 / Accepted: 17 September 1998
Abstract Epidermal linoleic acid, i.e. essential fatty
acid (EFA), is essential for cutaneous barrier function.
Cultured human keratinocytes, routinely used for
studies of lipid metabolism, are grown in a keratino-
cyte serum-free medium (KSFM), under conditions
that reveal EFA-deficient cells. Here, fatty acid (FA)
uptake was analysed in human adult keratinocytes
grown either under EFA-deficient conditions [KSFM
supplemented with 10% FCS (A) or 1% UltroserG
(B)] or EFA-supplemented conditions [KSFM supple-
mented with a devised FA cocktail (C) or evening
primrose oil (D)]. The FA composition of the total cel-
lular lipid and major lipid fractions was analysed by
gas chromatography. Cells grown with supplements A
or B balanced their EFA-deficient state primarily with
oleic acid. Cells grown with supplements C or D nor-
malized to the epidermal FA composition in vivo with
raised linoleic and lower oleic acid contents. When
cells were grown longer than 48 h with supplements C
or D decreased cell growth was observed. FA uptake
was curvilinear with preference for linoleic over oleic
acid under all culture conditions. The uptake of
linoleic acid by cells cultured with supplement B was
twice the uptake of those cultured with supplement A,
while the uptake of oleic acid was similar under both
culture conditions. Oleic acid uptake of cells cultured
with supplement C or D was lower. These results show
that the uptake of linoleic, but not that of oleic acid, is
influenced by the extracellular FA composition, and
that EFA-supplemented keratinocytes compared to
EFA-deficient cells might serve as an in vitro model for
the study of EFA metabolism.
N. Y. Schürer · K. Vogelsang · V. Schliep · T. Ruzicka
Department of Dermatology, University of Düsseldorf, Germany
F. Rippke
Beiersdorf AG Hamburg, Germany
N. Y. Schürer (쾷)
Judenbühlweg 28, 97082 Würzburg, Germany
Tel. +49-931-88-36-24; Fax +49-931-88-36-30
© Springer-Verlag 1999
Key words Keratinocyte fatty acid uptake · Essential
fatty acid metabolism · Keratinocyte culture
Introduction
Mammalian cells synthesize saturated fatty acids (FA),
mainly C18 : 0 via the action of cytosolic acetyl-CoA car-
boxylase and FA synthases. Monounsaturated FA (i.e.
C16 : 1 and C18 : 1) are formed by desaturation of the cor-
responding saturated FA by the enzyme ∆9 desaturase.
Mammals cannot desaturate beyond the ∆9 position, thus
C18 : 2, ∆9,12 (linoleic acid) and 20 : 4 ∆5,8,11,14 (arachidonic
acid) cannot be synthesized by mammalian tissues such as
the epidermis, and must be supplied by the diet. There-
fore, they are essential [1].
Essential fatty acids (EFA) are integral to normal cell
behaviour. In whole epidermis linoleic acid is central to
permeability barrier function [2, 3]. Alterations in cell
membrane EFA composition are associated with alter-
ations in the activity of a number of membrane-bound
enzyme systems [4–7]. EFA deficiency is accompanied
by proliferative epidermal changes which to some extent
mimic atopic dermatitis. In patients with atopic dermati-
tis the level of ceramide 1, a carrier of linoleic acid, is
decreased and levels of esterified oleic acid are in-
creased [8]. The importance of systemic EFA supple-
mentation as a treatment for atopic dermatitis has been
discussed for more than half a century [9, 10]. In this
context, the oil from the seeds of evening primrose has
attracted special attention for the treatment of atopic
dermatitis owing to its unusually high content of linoleic
and γ-linolenic acid (C18 : 3), i.e. 70% and 10%, respec-
tively. However, the exact pathway of cellular EFA up-
take and incorporation into proliferating basal keratino-
cytes is not known.
Human adult keratinocytes cultured in serum-free me-
dium reveal a more rapid uptake of linoleic acid than of to
oleic acid (C18 : 1) during the initial influx [11]. Under
these culture conditions, the preference of a keratinocyte
FA transporter for linoleic acid may function to ensure
48
capture of sufficient linoleic acid. However, serum-free
media are virtually EFA-free and keratinocytes cannot
synthesize EFAs. The rapid division of the cells in culture
results in an altered lipid composition and in the predom-
inance of extremely EFA-deficient keratinocytes, which
could serve as a model to study EFA deficiency [12–15].
The aim of the present study was to characterize the
uptake of linoleic acid in comparison with that of oleic
acid in human adult keratinocytes grown under FA-defi-
cient or FA-supplemented culture conditions. Further-
more, the FA contents of total cellular lipid and major cel-
lular lipid fractions (i.e. phospholipids, triglycerides and
free fatty acids) were analysed under these culture condi-
tions.
Materials and methods
Cell culture and media formulae
Normal human adult keratinocytes (NHAK) were obtained from
reduction mammoplasties, and cultures were initiated according to
the Gibco (Berlin) protocol. Cells were seeded at a density of 4 ×
105 cells per T75 flask. The growth medium for NHAK was kera-
tinocyte serum-free medium (KSFM) supplemented with bovine
pituitary extract (50 µg/ml) and epidermal growth factor (5 ng/ml)
(all from Gibco, Berlin, Germany).
FA-enriched media (Table 1, C and D) contained the long-
chain FAs (LCFAs) palmitic, oleic, linoleic and γ-linolenic acids,
FA-free bovine serum albumin (BSA) at a BSA : FA molar ratio of
1 : 3 to allow dissolution of LCFA at the concentrations used and
the antioxidant α-tocopherol acetate (1 mM). The cellular lipid
transport factor L-carnitine (100 µM) was added to the medium
supplement C. The amount of evening primrose oil (medium sup-
plement D) added to the medium was adjusted to the linoleic acid
concentration in medium supplement C (70 µM), equivalent to the
approximate physiological linoleic acid concentration of human
serum. The calculated final concentrations of other LCFA natu-
rally occurring in evening primrose oil were based on analysis by
gas chromatography (Table 2), and were in accordance with the
findings of Horrobin et al. [16]. The final concentrations of satu-
rated (C16 : 0 and C18 : 0) and monounsaturated FA (C16 : 1) were
higher in supplement C than in supplement D medium. Further-
more, arachidonic acid (C20 : 4) does not naturally occur in
evening primrose oil.
For experimental purposes, third-passage NHAK were plated
in 60-mm culture dishes (Falcon, Heidelberg) and maintained in an
atmosphere here containing 5% CO2 at 37 °C until near conflu-
ence. NHAK were incubated for 6 days in growth medium fol-
Table 1 Concentrations (µM of long-chain fatty acids (LCFA) in
human serum and final concentrations (µM) in KSFM with various
supplements: A 10% fetal calf serum, B 1% UltroserG, C FA cock-
tail containing LCFA according to the approximate physiological
human serum concentration, D evening primrose oil (Epogam)
LCFA Human serum KSFM supplement
(normal range)
A B C D
16 : 0 84–196 23 0.9 100.0 5.5
18 : 0 42– 98 1.8 0.7 – 1.7
18 : 1 84–196 1.8 1.1 100.0 6.9
18 : 2 42– 98 0.4 0.2 70.0 70.0
18 : 3 – – – 8.6
20 : 4 9– 21 0.6 – 30.0 –
Table 2 Composition of total LCFA of evening primrose oil as
determined by GC/MS
Compound % of total content
Palmitic acid (16 : 0) 7.65
Stearic acid (18 : 0) 0.99
Oleic acid (18 : 1) 9.81
Linoleic acid (18 : 2) 70.78
γ-Linolenic acid (18 : 3) 8.96
Tocopherolacetate 1.81
lowed by 2 days of cultivation in KSFM with medium supple-
ments A–D (Table 1). Media were changed every 48 h. Prior to
cultivation in the defined media A–D, the growth medium was re-
moved and cultures were washed twice with calcium- and magne-
sium-free PBS (pH 7.4). Culture viability was routinely tested by
the exclusion of trypan blue, and > 96% of cells were found to be
viable.
Harvesting and homogenizing of cells
NHAK were rinsed three times with cold PBS and scraped from
the dish into microfuge tubes (Falcon, Heidelberg) using a rubber
policeman. The total volume of each sample was 1 ml. Cells were
homogenized by sonication on ice with three or four 20 s bursts at
a relative output of 35% in a Sonic Dismembranator (Branson
Sonic Power Company, Geneva), and 50-µl aliquots were taken
for protein determination (BioRad Protein Assay, BioRad, Mu-
nich). The remainder were used for lipid extraction.
Lipid extraction, fractionation and determination
Total lipid was extracted from the cell homogenate according to
the method of Bligh and Dyer with modification, using a 0.25 M
KCl solution to ensure the complete extraction of all polar lipids
[17]. Phases containing organic lipid were dried under a stream of
nitrogen and the residues were dissolved in 800 µl chloro-
form/methanol (2 : 1 v/v/v). The extracted lipids were separated by
thin layer chromatography (TLC; Silica gel 60, Merck, Darmstadt)
using a one-dimensional system. The separation of 100 µg total
lipid was performed using three consecutive developing systems:
(1) chloroform/methanol/water 95 : 14 : 1 (v/v/v) up to 10 cm, (2)
n-hexane/diethyl ether/glacial acetic acid 80 : 20 : 7 (v/v/v) up to
15 cm, (3) petroleum ether up to 18 cm. Lipid fractions were iden-
tified by cochromatography with authentic standards [18]. Lipid
quantitation was performed using a Desaga scanner providing peak
integration and calculation of percent composition and absolute
amount in relation to authentic standards [18].
Fatty acid analysis
FA methyl esters (FAME) were prepared by a modification of the
method of Morrison and Smith [19]. Briefly, 5–8 mg total cellular
lipid extract in 500 µl chloroform was applied on an Isolute spe
column according to the IST protocol (IST, Mid Glamorgan, UK).
Neutral lipids were eluted using chloroform/isopropanol 2 : 1 (v/v),
free FA were separated from the remainder with glacial acetic
acid/diethyl ether 2 : 98 (v/v), and finally the phospholipid fraction
was eluted with methanol. Triglyceride separation from the neutral
lipid fraction was achieved using diethyl ether/dichloromethane/
n-hexane 1 : 10 : 90 (v/v/v) as solvent. This procedure resulted in
the recovery of 92% of the applied total lipid. However, the free
FA and triglyceride fractions were found to be contaminated with
alkanes. Therefore, the free FAs were separated from the alkanes
by one-dimensional TLC as described previously and the triglyc-
erides were separated from the alkanes using the following devel-

oping systems (all to the top of the plate): (1) petroleum ether/di-
ethyl ether/glacial acetic acid 78 : 28 : 1 (v/v/v), (2) toluene, and (3)
n-hexane. Lipid fractions were visualized with 1% aqueous
8-anilino-1-naphthalene sulphonic acid and were then scraped off
the plates, extracted and evaporated to dryness under nitrogen.
For transmethylation of FAs the free FA fraction was incubated
for 30 min, and the phospholipid, cholesterol ester and triglyceride
fractions for 90 min at 100 °C in 1 ml 14% BF3 in methanol. Trans-
methylation of total FA of evening primrose oil (Epogam, Beiers-
dorf, Germany) was performed using 50 mg of the crude evening
primrose oil dissolved in 2 ml 2% H2SO4 in methanol at 50 °C
overnight. FAMEs were extracted three times in 2 ml n-hexane
and analysed on a Chrompack gas chromatograph (GC; Model CP
9000, Chrompack, Frankfurt) using a capillary column (Chrom-
pack silica gel, length 50 m, diameter 0.25 mm). The column tem-
perature was programmed to increase from 160 °C to 230 °C at a
rate of 1° C/min with the prepressure being 115 kPA at a split ratio
of 1 : 10. FAMEs were identified by comparison of retention times
with those of known standards and the relative percentages were
calculated by integration of the area under each peak using a Shi-
madzu integrator (Model C-R5A, Shimadzu, Duisburg).
Assay of fatty acid uptake
The radiolabelled ligands, [1-14C]linoleic acid (specific activity 59
mCi/mmol; Amersham, Braunschweig, Germany) and [9,10(n)-
3H]oleic acid (specific activity 10 Ci/mmol; Amersham), were dis-
solved in 50 µl 0.1 N NaOH at 37 °C, to which quantities of the
corresponding unlabelled FA were added in order to achieve the
desired final concentration. BSA dissolved in PBS was added to
the FA solution to obtain the desired FA : BSA molar ratio of 1 : 1.
The pH was adjusted to 7.4 and the FA/BSA solution diluted to a
final concentration of 173 µM in PBS [11]. After bringing both
cells and solutions to the appropriate temperature (37 °C), the cul-
tures were incubated with 500 µl of the FA solution by gentle agi-
tation in a 37 °C water bath. At various times 3 ml ice-cold 0.5%
BSA in PBS was added to stop cellular influx and to remove sur-
face-bound FA. Cells were dissolved by the addition of 1 N NaOH
(1 ml) for 12 h at 4 °C. Aliquots (300 µl) were subsequently added
to 10 ml of a scintillation cocktail (Zinsser Analytics, Frankfurt)
and the radioactivity was determined using a Packard 1600 TR
scintillation counter (Packard, Frankfurt) at a counting efficiency
of 92% for [14C] and 61% for [3H]. Additional 50-µl aliquots were
processed for determination of protein content (BioRad, Mün-
chen).
Statistical analysis
Each experimental condition was performed in triplicate dishes
and all experiments were replicated at least three times. The values
shown in the figures and tables are means ± SD of three different
experiments. The significance of differences was determined using
a two-tailed Student’s t-test.
Results
49
by trypan blue exclusion. However, even in the presence
of BSA, stearic acid was not dissolved to the desired con-
centration of 50–100 µM.
The LCFA content of evening primrose oil (supple-
ment D) differed from that of the devised FA cocktail
(Table 1). Medium supplement C was composed of free
LCFA at a defined concentration, while evening primrose
oil is a naturally occurring EFA supplement. Based on the
analysis by GC both esterified and free LCFA accounted
for the final LCFA concentration (90 µM total LCFA at a
BSA:FA molar ratio of 1 : 3).
Table 3 summarizes quantitative data on the cellular
lipid fractions. The triglyceride and sphingolipid fractions
increased in cells cultivated in EFA-supplemented media
(supplements C and D) compared with cells grown in
EFA-deficient media (supplements A and B). However,
no statistically significant differences were detected when
the data were analysed as micrograms of lipid per mil-
ligram of protein.
As shown in Table 4, the content of cellular protein
was comparable in cells grown for up to 48 h with sup-
plements A, B, C and D. If cells were maintained in the
media supplemented with C and D for more than 48 h, i.e.
up to 96 h, the protein content remained unchanged, while
the protein content of cells cultured in KSFM supple-
mented with A and B still continued to increase. Visual
examination of proliferating cells grown for 48 h in the
extreme EFA-deficient state (medium supplements A and
B) showed normal, rounded morphology while those cells
grown in KSFM supplemented with C presented a flatter,
granular cell morphology. Medium supplemented with D
Table 3 Quantitation of cellular lipid fractions. Cells were cul-
tured with supplements A, B, C or D for 48 h and harvested. Lipids
(µg lipid / mg protein) were fractioned by TLC and quantitated.
Values are the means ± SD of three seperate experiments (A–D as
Table 1)
Lipid fraction Medium supplement
A B C D
Sterol ester 1.4 ± 1.1 0.9 ± 1.0 2.0 ± 0.8 2.4 ± 0.9
Triglycerides 22.5 ± 7.4 19.4 ± 6.4 37.9 ± 13.4 33.8 ± 12.5
Free fatty acids 2.6 ± 1.3 1.7 ± 1.5 2.5 ± 0.6 2.4 ± 0.5
Free sterols 12.9 ± 4.4 11.8 ± 3.4 10.8 ± 1.1 12.3 ± 1.6
Sphingolipids 5.1 ± 2.4 4.5 ± 2.8 9.9 ± 3.2 8.6 ± 3.9
Phospholipids 63.8 ± 10.3 60.7 ± 13.1 59.4 ± 8.9 58.2 ± 7.2
Total lipid 106.6 ± 13.7 96.5 ± 14.5 120.2 ± 25.3 116.3 ± 20.4

FA composition
Table 4 Protein fractions. Cells were cultured for 5 days in
growth medium followed by a 48- or 96-h cultivation in supple-
Medium supplement C was composed of LCFA naturally mented media A–D. Values are the mean (± SD) weight (mg) from
occurring in human serum. The addition of LCFA at con- three 60-mm culture dishes (A–D) as Table 1)
centrations of 30–100 µM each (300 µM total, supplement Duration of Medium supplement
C, Table 1) to KSFM required the presence of BSA. At a cultivation
BSA : FA molar ratio of 1 : 3 most FA is tightly bound to (h) A B C D
albumin and only a very small fraction of free FA (≈ 1%,
i.e. 3 µM free FA) is actually dissolved in aqueous solu- 48 0.93 ± 0.7 0.97 ± 0.6 0.89 ± 0.9 0.91 ± 0.8
tion [20]. Cytotoxic effects were not observed, as shown 96 109.9 ± 1.3 112.3 ± 1.6 0.93 ± 0.6 0.96 ± 0.9
50
Table 5 Fatty acid content. Cells were cultured in KSFM supple- total lipid), while those of C18 : 1 were relatively low
mented with A, B, C or D for 48 h and harvested. The values are (39.9 ± 3.2% and 34.6 ± 2.9% of total lipid). NHAK cul-
the fatty acid contents expressed as percent of total lipid (A–D as
Table 1) tured in KSFM supplemented with C showed an increase
in C20 : 4 (3.4 ± 0.3% of total lipid) reflecting the high
Fatty Medium supplement medium concentration (bound and unbound, 30 µM).
acid However, evening primrose oil contains C18 : 3, which is
A B C D
reflected by the relative whole cell FA composition (3.1 ±
14 : 0 2.1 ± 0.6 3.7 ± 0.5 2.0 ± 0.8 2.4 ± 0.6 0.5% of total lipid). The low concentration of cellular
16 : 0 26.3 ± 1.2 29.9 ± 2.1 25.9 ± 3.4 25.8 ± 3.5 LCFA (> C20) may indicate that a high proportion of pro-
16 : 1 7.6 ± 0.3 4.8 ± 0.3 2.6 ± 0.6 3.4 ± 0.5 liferating cells were used in these studies.
18 : 0 10.9 ± 0.4 14.0 ± 1.2 10.8 ± 1.1 9.3 ± 0.6 In cells grown under EFA-deficient conditions (supple-
18 : 1 45.1 ± 2.4 42.6 ± 1.9 39.9 ± 3.2 34.6 ± 2.9 ments A and B) the relative amounts of C18 : 2 in phos-
18 : 2 3.8 ± 0.3 2.7 ± 0.6 13.4 ± 1.9 16.2 ± 2.2 pholipids, triglycerides and free FA were less than in cells
18 : 3 0.2 ± 0.0 0.6 ± 0.0 0.5 ± 0.0 3.1 ± 0.5 grown under the EFA-supplemented conditions (supple-
20 : 4 1.3 ± 0.1 1.1 ± 0.3 3.4 ± 0.3 0.9 ± 0.4 ments C and D; Table 6). Relatively high amounts of C18
22 : 0 1.1 ± 0.2 0.3 ± 0.0 0.8 ± 0.1 0.4 ± 0.0 : 1 were recovered from the phospholipids and triglyc-
24 : 0 0.4 ± 0.1 0.0 ± 0.0 0.4 ± 0.0 1.1 ± 0.5 erides when cells were grown with supplement A or B.
26 : 0 0.4 ± 0.0 0.1 ± 0.0 0.0 ± 0.0 0.9 ± 0.3 When NHAK were cultured with supplement C, C20 : 4
28 : 0 0.5 ± 0.0 0.2 ± 0.0 0.2 ± 0.0 1.3 ± 0.1 was found to account for 2.2% of the FA recovered from
n 6 4 4 4 cellular phospholipids and 5.9% from triglycerides. Low
amounts of C16 : 1 were recovered from the phospho-
lipids, triglycerides and free FA when NHAK were cul-
tured under EFA-supplemented conditions. Briefly, the to-
yielded cultures with a more normal, rounded morphol-
tal cellular FA composition described above reflected the
ogy.
FA composition of cellular phospholipids and triglyc-
The percent composition of major FA extracted from
erides, and to some extent that of cellular free FA.
the whole cell lipid is presented in Table 5. The FA values
reflect the FA composition of the medium supplements
A–D. Cells grown under EFA-deficient conditions (sup- FA uptake by NHAK grown in FA-depleted
plements A and B) showed low levels of C18 : 2 (3.8 ± and -supplemented media
0.3% and 2.7 ± 0.6% of total lipid) and consequently of
the metabolites C18 : 3 (0.2 ± 0.0% and 0.6 ± 0.0% of to- These experiments compared the uptake of [3H]-oleic and
tal lipid) and C20 : 4 (1.3 ± 0.1% and 1.1 ± 0.3% of total [14C]-linoleic acid by NHAK cultivated in KSFM supple-
lipid), while levels of the unsaturated nonessential FA mented with A, B, C or D for 48 h. (Table 1; Fig. 1 A, B).
C18 : 1 were relatively high (45.1 ± 2.4% and 42.6 ± 1.9% For these uptake studies a total FA concentration of 173 µM
of total lipid). The FA profile of the whole cell lipid ex- (FA : BSA molar ratio 1 : 1) was chosen to allow compari-
tract reflected also the FA composition of the EFA-sup- son with previous studies of FA uptake [11]. FA uptake
plemented media (supplements C and D). Levels of C18 : 2 was curvilinear with a preference for linoleic over oleic
were relatively high (13.4 ± 1.9% and 16.2 ± 2.2% of acid under all culture conditions.

Table 6 FA composition (%) of confluent cells grown in KSFM supplemented with 10% FCS (A), 1% UltroserG (B), LCFA (C) or
evening primrose oil (D). The FA concentration was aligned with the approximate physiological linoleic acid concentration of human
serum (ND not detectable)
Fatty acid Phospholipids Triglycerides Free fatty acids

A B C D A B C D A B C D

14 : 0 2.2 3.3 2.9 3.4 4.4 5.9 2.9 2.4 1.3 1.0 1.4 1.4
16 : 0 30.4 29.2 39.6 30.7 21.6 27.0 20.4 25.5 29.5 28.5 25.3 29.9
16 : 1 3.9 4.3 2.1 2.5 5.6 5.1 1.9 2.3 5.0 6.8 3.1 2.8
18 : 0 18.7 15.1 12.5 16.7 19.8 21.4 8.5 13.9 25.9 26.1 25.9 24.6
18 : 1 38.9 42.6 25.4 23.3 43.1 36.3 31.6 28.2 34.4 33.8 33.8 34.4
18 : 2 4.2 3.5 12.2 23.5 4.2 3.1 26.9 19.7 0.7 0.6 6.4 3.1
18 : 3 ND ND ND 0.4 0.5 0.8 0.6 3.1 0.4 0.4 0.6 0.5
20 : 4 0.7 1.2 2.2 ND 0.4 0.3 5.9 0.3 2.5 2.1 2.1 2.6
22 : 0 0.7 0.6 2.6 1.3 ND ND ND ND 0.3 0.4 0.7 0.3
24 : 0 0.0 0.0 0.1 0.0 0.0 0.0 0.2 1.3 0.0 0.0 0.2 0.4
26 : 0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0
28 : 0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 1.1 0.0 0.0 0.0 0.0
 51
Fig. 1 Time course of linoleic A
acid uptake by cultured kera-
tinocytes. Cells were incubated
in KSFM supplemented with
1% UltroserG (closed trian-
gles), 10% FCS (open trian-
gles), FA cocktail containing
LCFA according to the approx-
imate physiological human
serum concentration (closed
circles), or evening primrose
oil (open circles). Uptake of
labeled linoleic acid (173 µM
FA/BSA, 1 : 1) was determined
at the times indicated. Values
are the means ± SD of nine ex-
periments. B Time course of
oleic acid uptake by cultured
keratinocytes (see legend
above for A)

The rate of FA uptake was greatest in cells cultivated
in KSFM with supplement B, i.e. extremely FA-deficient
conditions, and was linear for both linoleic and oleic acid
during the initial 30-s incubation period and gradually de-
creased thereafter. Therefore, cellular influx rates were
determined from the slope of this initial rate of uptake.
The uptake of linoleic acid during the initial 30-s influx
period significantly exceeded that of oleic acid (0.67 ±
0.17 vs 0.36 ± 0.12 nmol/mg protein, P < 0.001) at the
same concentration. To compare initial influx rates be-
tween NHAK grown in KSFM supplemented with B and
NHAK grown in KSFM supplemented with A, we se-
lected the 30-s time period. Linoleic acid uptake by cells
grown with supplement B (extremely EFA-deficient) was
twice that by cells grown with supplement A (compared
to B less EFA-deficient, Table 1; 0.67 ± 0.17 vs 0.36 ±
0.04 nmol/mg protein; P < 0.0001), while the initial up-
take rate of oleic acid was comparable (0.36 ± 0.12 vs
0.29 ± 0.09 nmol/mg protein; NS).
52
The linoleic acid uptake rates of cells grown with sup-
plements C and D were similar and low (0.12 ± 0.12 vs
0.09 ± 0.02 nmol/mg protein, NS) over the initial phase
(Fig. 1 A). The uptake of [14C]-linoleic acid by cells grown
under EFA-supplemented culture conditions (supplement
C was lower than that by cells grown with supplement A
(0.12 ± 0.12 vs 0.36 ± 0.12 nmol/mg protein; P < 0.001)
and even lower than the uptake by cells grown with sup-
plement B (0.12 ± 0.12 vs 0.67 ± 0.17 nmol/mg protein;
P < 0.0001). However, the differences in oleic acid uptake
were not significant (Fig. 1 B): supplement C vs A, 0.21 ±
0.06 vs 0.29 ± 0.09 nmol/mg protein (NS) and supplement
B vs A, 0.21 ± 0.06 vs 0.36 ± 0.12 nmol/mg protein (NS).
Discussion
Epidermal cell cultures are commonly grown in an FCS-
containing or a specific KSFM [21]. Since the FA profile
of cultured cells reflects that of the growth medium, kera-
tinocytes become EFA-deficient under these culture con-
ditions [12–14, 22, 23]. However, the addition of EFA
alone to the medium fails to normalize the FA profile of
EFA-deficient cells. Addition of C16 : 0 to the medium is
also required for normalization of the cellular FA compo-
sition [24]. For FA supplementation in vitro a devised FA
cocktail was employed in the present study and in the
studies of Boyce and Williams [15]. While Boyce and
Williams supplemented the lipid-enriched medium with a
total of 72 µM free FA (BSA : FA 1 : 3), in these studies ei-
ther a total FA supplement of 300 µM (supplement C) or
90 µM (supplement D) was employed. In the presence of
BSA at a BSA : FA molar ratio of 1 : 3, only a small frac-
tion of free FA (≈ 1%) dissolves in the medium and is as-
sumed to be in equilibrium with the BSA-bound fraction
[20]. Binding to albumin minimizes the toxic effects of
high FA concentrations, such as 300 µM. Using medium
supplements C and D the free FA concentrations were
calculated to be 3 µM and 0.9 µM, respectively. The total
free FA concentration for cells grown in KSFM supple-
mented with 1% UltroserG (supplement B) was 2.9 µM.
Neither FCS nor FA-free BSA was added to this medium.
The total FA concentration in the medium with supple-
ment A was 27.6 µM. The addition of 10% FCS (contain-
ing about 75 µM serum albumin) reduced the concentra-
tion of free FA to 0.27 µM. Briefly, even though supple-
ments C and D were FA-enriched in respect of the total
FA concentration, the presence of BSA reduced the
amount of free FA significantly. Toxic effects were negli-
gible under these culture conditions as also indicated by
trypan blue exclusion.
Under lipid-enriched culture conditions precursor
barrier lipids and lamellar bodies are expressed [15].
Here, cultured NHAK grown under EFA-supplemented
conditions normalized to the epidermal FA composition
in vivo with raised EFA (C18 : 2) and lower monounsat-
urated FA (C18 : 1 and C16 : 1) as compared to those
cells grown under EFA-deficient conditions. Contrary to
the in vivo situation, conversion of C18 : 2 into C20 : 4
takes place in cultured keratinocytes [25]. This enzy-
matic activity is not specifically reflected under the cul-
ture conditions used in our studies. Under the conditions
employed and over the time-period examined the cellu-
lar lipid composition of NHAK reflected the FA compo-
sition of the medium.
Palmitic acid is essential for normal cellular FA com-
position [23], linoleic acid is essential for barrier lipid
synthesis [2, 3] and γ-linolenic acid plays an important
role in cellular membrane viscosity [26]. Only supple-
ment D contained γ-linolenic acid (C18 : 3) in substantial
amounts (8.9% of total evening primrose oil LCFA). C18 : 3
is not a major LCFA constituent of human serum FAs and
therefore was not included in medium supplement C.
Whether C18 : 3 alone or the combination of LCFA natu-
rally occurring in evening primrose oil (Epogam) ac-
counted for the observed in vitro effect of reduced cell
growth accompanied by a normal appearance deserves
further attention.
Keratinocytes metabolize FAs that can be synthesized
by the cell [27, 28], or those FAs present in the medium
(in vitro condition) or in the serum (in vivo condition).
Despite its relative autonomy from the circulation in vivo,
the epidermis incorporates some circulating lipids, such
as certain sterols, EFA and polyunsaturated FA. Prior to
EFA uptake, FAs (bound and/or unbound) must traverse
the endothelial cells and basal membrane, and dissociate
from the carrier protein albumin. In vivo, serum FAs are
found mainly in the form of lipoproteins. Furthermore, al-
bumin is known to have high- (≈ 3) and low-affinity bind-
ing sites (≈ 15) for FA [20]. An equilibrium between lipo-
protein FA, FA bound to albumin and free FA influences
cellular FA uptake. These variables are diffcult to include
in in vitro studies, where FAs equilibrate only between the
free and BSA-bound fractions. Taking these methodolog-
ical difficulties into consideration, the results obtained
from in vitro uptake studies cannot be directly linked to
the in vivo situation.
In these studies the uptake of FA bound to albumin
(FA:BSA at a ratio of 1 : 1, 173 µM) was studied by kera-
tinocytes cultured with medium supplements A, B, C or
D. From the results presented in Fig. 1 A, B one can spec-
ulate that uptake of linoleic and oleic acids is a two-step
process: (1) a rapid FA uptake phase during the first 30 s
mediated by an FA carrier system and/or due to rapid dis-
solution of the FA into the cell membrane and (2) a slower
uptake process possibly due to passive diffusion of FA
into the cell. However, the observation that keratinocytes
take up linoleic acid more effectively than oleic acid dur-
ing the first 30 s of incubation depending on the FA con-
centration of the medium suggests that cultured keratino-
cytes have an uptake mechanism with an apparent acyl
specificity for linoleic acid. A rapid influx of linoleic acid
from the medium has also been found by Marcelo and
Dunham in experiments in which cells were incubated for
1–24 h with radioactively labelled FA [28]. Once inside
the cell C18 : 2 remained within the EFA pool. Here, we
demonstrated the velocity of linoleic acid uptake over the
first 2-min uptake period.

The uptake of linoleic acid (C18 : 2) is influenced by
the cellular EFA content. The uptake of C18 : 2 by cells
cultured in 1% UltroserG for 48 h was significantly
greater than by cells cultured in KSFM supplemented
with 10% FCS, while the uptake of oleic acid was similar.
C18 : 2 uptake by cells cultured under EFA-supplemented
conditions was relatively low. However, since linoleic
acid was readily incorporated from the culture medium
into the phospholipid and triglyceride fraction of NHAK
(Table 6), these results suggest that supplementation of
the culture medium with EFAs results in cultured kera-
tinocytes more able to reproduce the physiology of kera-
tinocytes in vivo. This finding is of particular relevance to
in vitro studies using keratinocytes unless the effects of
EFA deciency are being studied.
References
1. Vance DE, Vance JE (1995) Biochemistry of lipids and mem-
branes. The Benjamin/Cummings Publishing Company, Menlo
Park
2. Elias PM, Brown BE, Ziboh VA (1980) The permeability bar-
rier in essential fatty acid deficiency: evidence for a direct role
for linoleic acid in barrier function. J Invest Dermatol 74 : 230–
233
3. Prottey C (1976) Essential fatty acids and the skin. Br J Der-
matol 94 : 579–587
4. Feingold KR, Brown BE, Lear SR, Moser AH, Elias PM
(1986) Effect of essential fatty acid deficiency on cutaneous
sterol synthesis. J Invest Dermatol 87 : 588–591
5. Grubauer G, Feingold KR, Elias PM (1987) Relationship of
epidermal lipogenesis to cutaneous barrier function. J Lipid
Res 28 : 746–752
6. Proksch E, Elias PM, Feingold KR (1990) Regulation of 3-hy-
droxy-3-methylglutaryl-coenzyme A reductase activity in murine
epidermis: modulation of enzyme content and activation state
by barrier requirement. J Clin Invest 85 : 874–882
7. Ziboh VA (1996) The significance of polyunsaturated fatty
acids in cutaneous biology. Lipids 31 : S249-S253
8. Yamamoto A, Seriwaza S, Ito M, Sato Y (1991) Stratum
corneum lipid abnormalities in atopic dermatitis. Arch Derma-
tol Res 283 : 219–223
9. Brown WR, Hansen AE (1937) Arachidonic and linoleic acid
of the serum in normal and eczematous human species. Proc
Soc Exp Biol Med 36 : 113–117
10. Horrobin DF (1989) Essential fatty acids in clinical dermatol-
ogy. J Am Acad Dermatol 20 : 1045–1053
11. Schürer NY, Stremmel W, Grundmann J-U, Schliep V, Klein-
ert H, Bass NM, Williams ML (1994) Evidence for a novel ke-
ratinocyte fatty acid uptake mechanism with preference for
linoleic acid: comparison of oleic and linoleic acid by cultured
human keratinocytes, fibroblasts and a human hepatoma cell
line. Biochim Biophys Acta 1211 : 51–60
53
12. Isseroff RR, Ziboh VA, Chapkin RS, Martinez DT (1985) Al-
terations in fatty acid composition of murine keratinocytes with
in vitro cultivation. J Invest Dermatol 85 : 131–134
13. Marcelo CL, Duell EA, Rhodes LM, Dunham WR (1992) In
vitro model of essential fatty acid deficiency. J Invest Derma-
tol 99 : 703–708
14. Kennedy AH, Golden GM, Gay CL, Guy RH, Francoeur ML,
Mak VHW (1996) Stratum corneum lipids of human epidermal
keratinocyte air-liquid cultures: implications of barrier func-
tion. Pharm Res 13 : 1162–1167
15. Boyce ST, Williams ML (1993) Lipid supplemented medium
induces lamellar bodies and precursors of barrier lipids in cul-
tured analogues of human skin. J Invest Dermatol 101 : 180–
184
16. Horrobin DF, Ellis KM, Morse-Fisher N, Manku NS (1991)
The effects of evening primrose oil, safflower oil and paraffin
on plasma fatty acid levels in humans: choice of an appropriate
placebo for clinical studies on primrose oil. Prostaglandins
Leukot Essent Fatty Acids 42 : 245–249
17. Bligh EG, Dyer WJ (1959) A rapid method of total lipid ex-
traction and purification. Canad J Biochem Physiol 37 : 911–
917
18. Schürer NY, Köhne A, Schliep V, Barlag K, Goerz G (1993)
Lipid composition and synthesis of HacaT cells, an immortal-
ized human keratinocyte line, in comparison to normal human
adult keratinocytes. Exp Dermatol 2 : 179–185
19. Morrison WR, Smith LM (1964) Preparation of fatty acid
methyl esters and dimethyl-acetals from lipids with boron fluo-
ride-methanol. J Lipid Res 5 : 600–608
20. Potter BJ, Sorrentino D, Berk PD (1989) Mechanisms of cellu-
lar uptake of free fatty acids. Annu Rev Nutr 9 : 253–270
21. Ponec M (1991) Lipid metabolism in cultured keratinocytes.
Adv Lipid Res 24 : 83–118
22. Madison K, Wertz P, Strauss J, Downing D (1986) Lipid com-
position of established cell lines. J Invest Dermatol 87 : 253–
259
23. Wright S, Nawsaria H, Leigh IM (1991) Essential fatty acid
composition of cultured keratinocytes. Pharmacol Skin 4 : 196–
198
24. Marcelo CL, Rhodes LM, Dunham WR (1994) Normalization
of essential-fatty-acid-deficient keratinocytes requires palmitic
acid. J Invest Dermatol 103 : 564–568
25. Isseroff RR, Ziboh VA, Chapkin RS, Martinez DT (1987) Con-
version of linoleic acid into arachidonic acid by cultured
murine and human keratinocytes. J Lipid Res 28 : 1342–1349
26. Dunham WR, Klein SB, Rhodes LM, Marcelo CL (1996) Oleic
acid and linoleic acid are the major determinants of changes in
keratinocyte plasma membrane viscosity. J Invest Dermatol
107 : 332–335
27. Cullis PR, Hope MJ (1985) Physical properties and functional
roles of lipids in membranes. In: Vance DE, Vance JE (eds)
Biochemistry of lipids and membranes. The Benjamin/Cum-
mings Publishing Company, Menlo Park, pp 25–30
28. Marcelo CL, Dunham WR (1993) Fatty acid metabolism stud-
ies of human epidermal cell cultures. J Lipid Res 34 : 2077–
2090