Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Arch Dermatol Res (1999) 291 : 47–53 © Springer-Verlag 1999

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 Key words Keratinocyte fatty acid uptake · Essential
acid (EFA), is essential for cutaneous barrier function. fatty acid metabolism · Keratinocyte culture
Cultured human keratinocytes, routinely used for
studies of lipid metabolism, are grown in a keratino-
cyte serum-free medium (KSFM), under conditions Introduction
that reveal EFA-deficient cells. Here, fatty acid (FA)
uptake was analysed in human adult keratinocytes Mammalian cells synthesize saturated fatty acids (FA),
grown either under EFA-deficient conditions [KSFM mainly C18 : 0 via the action of cytosolic acetyl-CoA car-
supplemented with 10% FCS (A) or 1% UltroserG boxylase and FA synthases. Monounsaturated FA (i.e.
(B)] or EFA-supplemented conditions [KSFM supple- C16 : 1 and C18 : 1) are formed by desaturation of the cor-
mented with a devised FA cocktail (C) or evening responding saturated FA by the enzyme ∆9 desaturase.
primrose oil (D)]. The FA composition of the total cel- Mammals cannot desaturate beyond the ∆9 position, thus
lular lipid and major lipid fractions was analysed by C18 : 2, ∆9,12 (linoleic acid) and 20 : 4 ∆5,8,11,14 (arachidonic
gas chromatography. Cells grown with supplements A acid) cannot be synthesized by mammalian tissues such as
or B balanced their EFA-deficient state primarily with the epidermis, and must be supplied by the diet. There-
oleic acid. Cells grown with supplements C or D nor- fore, they are essential [1].
malized to the epidermal FA composition in vivo with Essential fatty acids (EFA) are integral to normal cell
raised linoleic and lower oleic acid contents. When behaviour. In whole epidermis linoleic acid is central to
cells were grown longer than 48 h with supplements C permeability barrier function [2, 3]. Alterations in cell
or D decreased cell growth was observed. FA uptake membrane EFA composition are associated with alter-
was curvilinear with preference for linoleic over oleic ations in the activity of a number of membrane-bound
acid under all culture conditions. The uptake of enzyme systems [4–7]. EFA deficiency is accompanied
linoleic acid by cells cultured with supplement B was by proliferative epidermal changes which to some extent
twice the uptake of those cultured with supplement A, mimic atopic dermatitis. In patients with atopic dermati-
while the uptake of oleic acid was similar under both tis the level of ceramide 1, a carrier of linoleic acid, is
culture conditions. Oleic acid uptake of cells cultured decreased and levels of esterified oleic acid are in-
with supplement C or D was lower. These results show creased [8]. The importance of systemic EFA supple-
that the uptake of linoleic, but not that of oleic acid, is mentation as a treatment for atopic dermatitis has been
influenced by the extracellular FA composition, and discussed for more than half a century [9, 10]. In this
that EFA-supplemented keratinocytes compared to context, the oil from the seeds of evening primrose has
EFA-deficient cells might serve as an in vitro model for attracted special attention for the treatment of atopic
the study of EFA metabolism. 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-
N. Y. Schürer · K. Vogelsang · V. Schliep · T. Ruzicka take and incorporation into proliferating basal keratino-
Department of Dermatology, University of Düsseldorf, Germany cytes is not known.
F. Rippke Human adult keratinocytes cultured in serum-free me-
Beiersdorf AG Hamburg, Germany dium reveal a more rapid uptake of linoleic acid than of to
oleic acid (C18 : 1) during the initial influx [11]. Under
N. Y. Schürer (쾷)
Judenbühlweg 28, 97082 Würzburg, Germany these culture conditions, the preference of a keratinocyte
Tel. +49-931-88-36-24; Fax +49-931-88-36-30 FA transporter for linoleic acid may function to ensure
48

capture of sufficient linoleic acid. However, serum-free Table 2 Composition of total LCFA of evening primrose oil as
media are virtually EFA-free and keratinocytes cannot determined by GC/MS
synthesize EFAs. The rapid division of the cells in culture Compound % of total content
results in an altered lipid composition and in the predom-
inance of extremely EFA-deficient keratinocytes, which Palmitic acid (16 : 0) 7.65
could serve as a model to study EFA deficiency [12–15]. Stearic acid (18 : 0) 0.99
The aim of the present study was to characterize the Oleic acid (18 : 1) 9.81
uptake of linoleic acid in comparison with that of oleic Linoleic acid (18 : 2) 70.78
acid in human adult keratinocytes grown under FA-defi- γ-Linolenic acid (18 : 3) 8.96
cient or FA-supplemented culture conditions. Further- Tocopherolacetate 1.81
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- lowed by 2 days of cultivation in KSFM with medium supple-
tions. 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
Materials and methods the exclusion of trypan blue, and > 96% of cells were found to be
viable.
Cell culture and media formulae

Normal human adult keratinocytes (NHAK) were obtained from Harvesting and homogenizing of cells
reduction mammoplasties, and cultures were initiated according to
the Gibco (Berlin) protocol. Cells were seeded at a density of 4 × NHAK were rinsed three times with cold PBS and scraped from
105 cells per T75 flask. The growth medium for NHAK was kera- the dish into microfuge tubes (Falcon, Heidelberg) using a rubber
tinocyte serum-free medium (KSFM) supplemented with bovine policeman. The total volume of each sample was 1 ml. Cells were
pituitary extract (50 µg/ml) and epidermal growth factor (5 ng/ml) homogenized by sonication on ice with three or four 20 s bursts at
(all from Gibco, Berlin, Germany). a relative output of 35% in a Sonic Dismembranator (Branson
FA-enriched media (Table 1, C and D) contained the long- Sonic Power Company, Geneva), and 50-µl aliquots were taken
chain FAs (LCFAs) palmitic, oleic, linoleic and γ-linolenic acids, for protein determination (BioRad Protein Assay, BioRad, Mu-
FA-free bovine serum albumin (BSA) at a BSA : FA molar ratio of nich). The remainder were used for lipid extraction.
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 Lipid extraction, fractionation and determination
supplement C. The amount of evening primrose oil (medium sup-
plement D) added to the medium was adjusted to the linoleic acid Total lipid was extracted from the cell homogenate according to
concentration in medium supplement C (70 µM), equivalent to the the method of Bligh and Dyer with modification, using a 0.25 M
approximate physiological linoleic acid concentration of human KCl solution to ensure the complete extraction of all polar lipids
serum. The calculated final concentrations of other LCFA natu- [17]. Phases containing organic lipid were dried under a stream of
rally occurring in evening primrose oil were based on analysis by nitrogen and the residues were dissolved in 800 µl chloro-
gas chromatography (Table 2), and were in accordance with the form/methanol (2 : 1 v/v/v). The extracted lipids were separated by
findings of Horrobin et al. [16]. The final concentrations of satu- thin layer chromatography (TLC; Silica gel 60, Merck, Darmstadt)
rated (C16 : 0 and C18 : 0) and monounsaturated FA (C16 : 1) were using a one-dimensional system. The separation of 100 µg total
higher in supplement C than in supplement D medium. Further- lipid was performed using three consecutive developing systems:
more, arachidonic acid (C20 : 4) does not naturally occur in (1) chloroform/methanol/water 95 : 14 : 1 (v/v/v) up to 10 cm, (2)
evening primrose oil. n-hexane/diethyl ether/glacial acetic acid 80 : 20 : 7 (v/v/v) up to
For experimental purposes, third-passage NHAK were plated 15 cm, (3) petroleum ether up to 18 cm. Lipid fractions were iden-
in 60-mm culture dishes (Falcon, Heidelberg) and maintained in an tified by cochromatography with authentic standards [18]. Lipid
atmosphere here containing 5% CO2 at 37 °C until near conflu- quantitation was performed using a Desaga scanner providing peak
ence. NHAK were incubated for 6 days in growth medium fol- integration and calculation of percent composition and absolute
amount in relation to authentic standards [18].

Table 1 Concentrations (µM of long-chain fatty acids (LCFA) in


Fatty acid analysis
human serum and final concentrations (µM) in KSFM with various
supplements: A 10% fetal calf serum, B 1% UltroserG, C FA cock- FA methyl esters (FAME) were prepared by a modification of the
tail containing LCFA according to the approximate physiological method of Morrison and Smith [19]. Briefly, 5–8 mg total cellular
human serum concentration, D evening primrose oil (Epogam) lipid extract in 500 µl chloroform was applied on an Isolute spe
column according to the IST protocol (IST, Mid Glamorgan, UK).
LCFA Human serum KSFM supplement
Neutral lipids were eluted using chloroform/isopropanol 2 : 1 (v/v),
(normal range)
free FA were separated from the remainder with glacial acetic
A B C D
acid/diethyl ether 2 : 98 (v/v), and finally the phospholipid fraction
was eluted with methanol. Triglyceride separation from the neutral
16 : 0 84–196 23 0.9 100.0 5.5
lipid fraction was achieved using diethyl ether/dichloromethane/
18 : 0 42– 98 1.8 0.7 – 1.7 n-hexane 1 : 10 : 90 (v/v/v) as solvent. This procedure resulted in
18 : 1 84–196 1.8 1.1 100.0 6.9 the recovery of 92% of the applied total lipid. However, the free
18 : 2 42– 98 0.4 0.2 70.0 70.0 FA and triglyceride fractions were found to be contaminated with
18 : 3 – – – 8.6 alkanes. Therefore, the free FAs were separated from the alkanes
by one-dimensional TLC as described previously and the triglyc-
20 : 4 9– 21 0.6 – 30.0 –
erides were separated from the alkanes using the following devel-
49
oping systems (all to the top of the plate): (1) petroleum ether/di- by trypan blue exclusion. However, even in the presence
ethyl ether/glacial acetic acid 78 : 28 : 1 (v/v/v), (2) toluene, and (3) of BSA, stearic acid was not dissolved to the desired con-
n-hexane. Lipid fractions were visualized with 1% aqueous
8-anilino-1-naphthalene sulphonic acid and were then scraped off centration of 50–100 µM.
the plates, extracted and evaporated to dryness under nitrogen. The LCFA content of evening primrose oil (supple-
For transmethylation of FAs the free FA fraction was incubated ment D) differed from that of the devised FA cocktail
for 30 min, and the phospholipid, cholesterol ester and triglyceride (Table 1). Medium supplement C was composed of free
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-
LCFA at a defined concentration, while evening primrose
dorf, Germany) was performed using 50 mg of the crude evening oil is a naturally occurring EFA supplement. Based on the
primrose oil dissolved in 2 ml 2% H2SO4 in methanol at 50 °C analysis by GC both esterified and free LCFA accounted
overnight. FAMEs were extracted three times in 2 ml n-hexane for the final LCFA concentration (90 µM total LCFA at a
and analysed on a Chrompack gas chromatograph (GC; Model CP BSA:FA molar ratio of 1 : 3).
9000, Chrompack, Frankfurt) using a capillary column (Chrom-
pack silica gel, length 50 m, diameter 0.25 mm). The column tem- Table 3 summarizes quantitative data on the cellular
perature was programmed to increase from 160 °C to 230 °C at a lipid fractions. The triglyceride and sphingolipid fractions
rate of 1° C/min with the prepressure being 115 kPA at a split ratio increased in cells cultivated in EFA-supplemented media
of 1 : 10. FAMEs were identified by comparison of retention times (supplements C and D) compared with cells grown in
with those of known standards and the relative percentages were
calculated by integration of the area under each peak using a Shi- EFA-deficient media (supplements A and B). However,
madzu integrator (Model C-R5A, Shimadzu, Duisburg). no statistically significant differences were detected when
the data were analysed as micrograms of lipid per mil-
ligram of protein.
Assay of fatty acid uptake
As shown in Table 4, the content of cellular protein
The radiolabelled ligands, [1-14C]linoleic acid (specific activity 59 was comparable in cells grown for up to 48 h with sup-
mCi/mmol; Amersham, Braunschweig, Germany) and [9,10(n)- plements A, B, C and D. If cells were maintained in the
3H]oleic acid (specific activity 10 Ci/mmol; Amersham), were dis-
media supplemented with C and D for more than 48 h, i.e.
solved in 50 µl 0.1 N NaOH at 37 °C, to which quantities of the up to 96 h, the protein content remained unchanged, while
corresponding unlabelled FA were added in order to achieve the
desired final concentration. BSA dissolved in PBS was added to the protein content of cells cultured in KSFM supple-
the FA solution to obtain the desired FA : BSA molar ratio of 1 : 1. mented with A and B still continued to increase. Visual
The pH was adjusted to 7.4 and the FA/BSA solution diluted to a examination of proliferating cells grown for 48 h in the
final concentration of 173 µM in PBS [11]. After bringing both extreme EFA-deficient state (medium supplements A and
cells and solutions to the appropriate temperature (37 °C), the cul-
tures were incubated with 500 µl of the FA solution by gentle agi- B) showed normal, rounded morphology while those cells
tation in a 37 °C water bath. At various times 3 ml ice-cold 0.5% grown in KSFM supplemented with C presented a flatter,
BSA in PBS was added to stop cellular influx and to remove sur- granular cell morphology. Medium supplemented with D
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) Table 3 Quantitation of cellular lipid fractions. Cells were cul-
and the radioactivity was determined using a Packard 1600 TR tured with supplements A, B, C or D for 48 h and harvested. Lipids
scintillation counter (Packard, Frankfurt) at a counting efficiency (µg lipid / mg protein) were fractioned by TLC and quantitated.
of 92% for [14C] and 61% for [3H]. Additional 50-µl aliquots were Values are the means ± SD of three seperate experiments (A–D as
processed for determination of protein content (BioRad, Mün- Table 1)
chen).
Lipid fraction Medium supplement
Statistical analysis A B C D
Each experimental condition was performed in triplicate dishes Sterol ester 1.4 ± 1.1 0.9 ± 1.0 2.0 ± 0.8 2.4 ± 0.9
and all experiments were replicated at least three times. The values
Triglycerides 22.5 ± 7.4 19.4 ± 6.4 37.9 ± 13.4 33.8 ± 12.5
shown in the figures and tables are means ± SD of three different
experiments. The significance of differences was determined using Free fatty acids 2.6 ± 1.3 1.7 ± 1.5 2.5 ± 0.6 2.4 ± 0.5
a two-tailed Student’s t-test. 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
Results 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 tween NHAK grown in KSFM supplemented with B and
in KSFM with supplement B, i.e. extremely FA-deficient NHAK grown in KSFM supplemented with A, we se-
conditions, and was linear for both linoleic and oleic acid lected the 30-s time period. Linoleic acid uptake by cells
during the initial 30-s incubation period and gradually de- grown with supplement B (extremely EFA-deficient) was
creased thereafter. Therefore, cellular influx rates were twice that by cells grown with supplement A (compared
determined from the slope of this initial rate of uptake. to B less EFA-deficient, Table 1; 0.67 ± 0.17 vs 0.36 ±
The uptake of linoleic acid during the initial 30-s influx 0.04 nmol/mg protein; P < 0.0001), while the initial up-
period significantly exceeded that of oleic acid (0.67 ± take rate of oleic acid was comparable (0.36 ± 0.12 vs
0.17 vs 0.36 ± 0.12 nmol/mg protein, P < 0.001) at the 0.29 ± 0.09 nmol/mg protein; NS).
same concentration. To compare initial influx rates be-
52

The linoleic acid uptake rates of cells grown with sup- takes place in cultured keratinocytes [25]. This enzy-
plements C and D were similar and low (0.12 ± 0.12 vs matic activity is not specifically reflected under the cul-
0.09 ± 0.02 nmol/mg protein, NS) over the initial phase ture conditions used in our studies. Under the conditions
(Fig. 1 A). The uptake of [14C]-linoleic acid by cells grown employed and over the time-period examined the cellu-
under EFA-supplemented culture conditions (supplement lar lipid composition of NHAK reflected the FA compo-
C was lower than that by cells grown with supplement A sition of the medium.
(0.12 ± 0.12 vs 0.36 ± 0.12 nmol/mg protein; P < 0.001) Palmitic acid is essential for normal cellular FA com-
and even lower than the uptake by cells grown with sup- position [23], linoleic acid is essential for barrier lipid
plement B (0.12 ± 0.12 vs 0.67 ± 0.17 nmol/mg protein; synthesis [2, 3] and γ-linolenic acid plays an important
P < 0.0001). However, the differences in oleic acid uptake role in cellular membrane viscosity [26]. Only supple-
were not significant (Fig. 1 B): supplement C vs A, 0.21 ± ment D contained γ-linolenic acid (C18 : 3) in substantial
0.06 vs 0.29 ± 0.09 nmol/mg protein (NS) and supplement amounts (8.9% of total evening primrose oil LCFA). C18 : 3
B vs A, 0.21 ± 0.06 vs 0.36 ± 0.12 nmol/mg protein (NS). 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-
Discussion rally occurring in evening primrose oil (Epogam) ac-
counted for the observed in vitro effect of reduced cell
Epidermal cell cultures are commonly grown in an FCS- growth accompanied by a normal appearance deserves
containing or a specific KSFM [21]. Since the FA profile further attention.
of cultured cells reflects that of the growth medium, kera- Keratinocytes metabolize FAs that can be synthesized
tinocytes become EFA-deficient under these culture con- by the cell [27, 28], or those FAs present in the medium
ditions [12–14, 22, 23]. However, the addition of EFA (in vitro condition) or in the serum (in vivo condition).
alone to the medium fails to normalize the FA profile of Despite its relative autonomy from the circulation in vivo,
EFA-deficient cells. Addition of C16 : 0 to the medium is the epidermis incorporates some circulating lipids, such
also required for normalization of the cellular FA compo- as certain sterols, EFA and polyunsaturated FA. Prior to
sition [24]. For FA supplementation in vitro a devised FA EFA uptake, FAs (bound and/or unbound) must traverse
cocktail was employed in the present study and in the the endothelial cells and basal membrane, and dissociate
studies of Boyce and Williams [15]. While Boyce and from the carrier protein albumin. In vivo, serum FAs are
Williams supplemented the lipid-enriched medium with a found mainly in the form of lipoproteins. Furthermore, al-
total of 72 µM free FA (BSA : FA 1 : 3), in these studies ei- bumin is known to have high- (≈ 3) and low-affinity bind-
ther a total FA supplement of 300 µM (supplement C) or ing sites (≈ 15) for FA [20]. An equilibrium between lipo-
90 µM (supplement D) was employed. In the presence of protein FA, FA bound to albumin and free FA influences
BSA at a BSA : FA molar ratio of 1 : 3, only a small frac- cellular FA uptake. These variables are diffcult to include
tion of free FA (≈ 1%) dissolves in the medium and is as- in in vitro studies, where FAs equilibrate only between the
sumed to be in equilibrium with the BSA-bound fraction free and BSA-bound fractions. Taking these methodolog-
[20]. Binding to albumin minimizes the toxic effects of ical difficulties into consideration, the results obtained
high FA concentrations, such as 300 µM. Using medium from in vitro uptake studies cannot be directly linked to
supplements C and D the free FA concentrations were the in vivo situation.
calculated to be 3 µM and 0.9 µM, respectively. The total In these studies the uptake of FA bound to albumin
free FA concentration for cells grown in KSFM supple- (FA:BSA at a ratio of 1 : 1, 173 µM) was studied by kera-
mented with 1% UltroserG (supplement B) was 2.9 µM. tinocytes cultured with medium supplements A, B, C or
Neither FCS nor FA-free BSA was added to this medium. D. From the results presented in Fig. 1 A, B one can spec-
The total FA concentration in the medium with supple- ulate that uptake of linoleic and oleic acids is a two-step
ment A was 27.6 µM. The addition of 10% FCS (contain- process: (1) a rapid FA uptake phase during the first 30 s
ing about 75 µM serum albumin) reduced the concentra- mediated by an FA carrier system and/or due to rapid dis-
tion of free FA to 0.27 µM. Briefly, even though supple- solution of the FA into the cell membrane and (2) a slower
ments C and D were FA-enriched in respect of the total uptake process possibly due to passive diffusion of FA
FA concentration, the presence of BSA reduced the into the cell. However, the observation that keratinocytes
amount of free FA significantly. Toxic effects were negli- take up linoleic acid more effectively than oleic acid dur-
gible under these culture conditions as also indicated by ing the first 30 s of incubation depending on the FA con-
trypan blue exclusion. centration of the medium suggests that cultured keratino-
Under lipid-enriched culture conditions precursor cytes have an uptake mechanism with an apparent acyl
barrier lipids and lamellar bodies are expressed [15]. specificity for linoleic acid. A rapid influx of linoleic acid
Here, cultured NHAK grown under EFA-supplemented from the medium has also been found by Marcelo and
conditions normalized to the epidermal FA composition Dunham in experiments in which cells were incubated for
in vivo with raised EFA (C18 : 2) and lower monounsat- 1–24 h with radioactively labelled FA [28]. Once inside
urated FA (C18 : 1 and C16 : 1) as compared to those the cell C18 : 2 remained within the EFA pool. Here, we
cells grown under EFA-deficient conditions. Contrary to demonstrated the velocity of linoleic acid uptake over the
the in vivo situation, conversion of C18 : 2 into C20 : 4 first 2-min uptake period.
53

The uptake of linoleic acid (C18 : 2) is influenced by 12. Isseroff RR, Ziboh VA, Chapkin RS, Martinez DT (1985) Al-
the cellular EFA content. The uptake of C18 : 2 by cells terations in fatty acid composition of murine keratinocytes with
in vitro cultivation. J Invest Dermatol 85 : 131–134
cultured in 1% UltroserG for 48 h was significantly 13. Marcelo CL, Duell EA, Rhodes LM, Dunham WR (1992) In
greater than by cells cultured in KSFM supplemented vitro model of essential fatty acid deficiency. J Invest Derma-
with 10% FCS, while the uptake of oleic acid was similar. tol 99 : 703–708
C18 : 2 uptake by cells cultured under EFA-supplemented 14. Kennedy AH, Golden GM, Gay CL, Guy RH, Francoeur ML,
Mak VHW (1996) Stratum corneum lipids of human epidermal
conditions was relatively low. However, since linoleic keratinocyte air-liquid cultures: implications of barrier func-
acid was readily incorporated from the culture medium tion. Pharm Res 13 : 1162–1167
into the phospholipid and triglyceride fraction of NHAK 15. Boyce ST, Williams ML (1993) Lipid supplemented medium
(Table 6), these results suggest that supplementation of induces lamellar bodies and precursors of barrier lipids in cul-
the culture medium with EFAs results in cultured kera- tured analogues of human skin. J Invest Dermatol 101 : 180–
184
tinocytes more able to reproduce the physiology of kera- 16. Horrobin DF, Ellis KM, Morse-Fisher N, Manku NS (1991)
tinocytes in vivo. This finding is of particular relevance to The effects of evening primrose oil, safflower oil and paraffin
in vitro studies using keratinocytes unless the effects of on plasma fatty acid levels in humans: choice of an appropriate
EFA deciency are being studied. 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–
References 917
18. Schürer NY, Köhne A, Schliep V, Barlag K, Goerz G (1993)
1. Vance DE, Vance JE (1995) Biochemistry of lipids and mem- Lipid composition and synthesis of HacaT cells, an immortal-
branes. The Benjamin/Cummings Publishing Company, Menlo ized human keratinocyte line, in comparison to normal human
Park adult keratinocytes. Exp Dermatol 2 : 179–185
2. Elias PM, Brown BE, Ziboh VA (1980) The permeability bar- 19. Morrison WR, Smith LM (1964) Preparation of fatty acid
rier in essential fatty acid deficiency: evidence for a direct role methyl esters and dimethyl-acetals from lipids with boron fluo-
for linoleic acid in barrier function. J Invest Dermatol 74 : 230– ride-methanol. J Lipid Res 5 : 600–608
233 20. Potter BJ, Sorrentino D, Berk PD (1989) Mechanisms of cellu-
3. Prottey C (1976) Essential fatty acids and the skin. Br J Der- lar uptake of free fatty acids. Annu Rev Nutr 9 : 253–270
matol 94 : 579–587 21. Ponec M (1991) Lipid metabolism in cultured keratinocytes.
4. Feingold KR, Brown BE, Lear SR, Moser AH, Elias PM Adv Lipid Res 24 : 83–118
(1986) Effect of essential fatty acid deficiency on cutaneous 22. Madison K, Wertz P, Strauss J, Downing D (1986) Lipid com-
sterol synthesis. J Invest Dermatol 87 : 588–591 position of established cell lines. J Invest Dermatol 87 : 253–
5. Grubauer G, Feingold KR, Elias PM (1987) Relationship of 259
epidermal lipogenesis to cutaneous barrier function. J Lipid 23. Wright S, Nawsaria H, Leigh IM (1991) Essential fatty acid
Res 28 : 746–752 composition of cultured keratinocytes. Pharmacol Skin 4 : 196–
6. Proksch E, Elias PM, Feingold KR (1990) Regulation of 3-hy- 198
droxy-3-methylglutaryl-coenzyme A reductase activity in murine 24. Marcelo CL, Rhodes LM, Dunham WR (1994) Normalization
epidermis: modulation of enzyme content and activation state of essential-fatty-acid-deficient keratinocytes requires palmitic
by barrier requirement. J Clin Invest 85 : 874–882 acid. J Invest Dermatol 103 : 564–568
7. Ziboh VA (1996) The significance of polyunsaturated fatty 25. Isseroff RR, Ziboh VA, Chapkin RS, Martinez DT (1987) Con-
acids in cutaneous biology. Lipids 31 : S249-S253 version of linoleic acid into arachidonic acid by cultured
8. Yamamoto A, Seriwaza S, Ito M, Sato Y (1991) Stratum murine and human keratinocytes. J Lipid Res 28 : 1342–1349
corneum lipid abnormalities in atopic dermatitis. Arch Derma- 26. Dunham WR, Klein SB, Rhodes LM, Marcelo CL (1996) Oleic
tol Res 283 : 219–223 acid and linoleic acid are the major determinants of changes in
9. Brown WR, Hansen AE (1937) Arachidonic and linoleic acid keratinocyte plasma membrane viscosity. J Invest Dermatol
of the serum in normal and eczematous human species. Proc 107 : 332–335
Soc Exp Biol Med 36 : 113–117 27. Cullis PR, Hope MJ (1985) Physical properties and functional
10. Horrobin DF (1989) Essential fatty acids in clinical dermatol- roles of lipids in membranes. In: Vance DE, Vance JE (eds)
ogy. J Am Acad Dermatol 20 : 1045–1053 Biochemistry of lipids and membranes. The Benjamin/Cum-
11. Schürer NY, Stremmel W, Grundmann J-U, Schliep V, Klein- mings Publishing Company, Menlo Park, pp 25–30
ert H, Bass NM, Williams ML (1994) Evidence for a novel ke- 28. Marcelo CL, Dunham WR (1993) Fatty acid metabolism stud-
ratinocyte fatty acid uptake mechanism with preference for ies of human epidermal cell cultures. J Lipid Res 34 : 2077–
linoleic acid: comparison of oleic and linoleic acid by cultured 2090
human keratinocytes, fibroblasts and a human hepatoma cell
line. Biochim Biophys Acta 1211 : 51–60

You might also like