Arch Dermatol Res (2000) 292 : 164–172
O R I G I N A L PA P E R
Martina Seifert · Wolfram Sterry · Elke Effenberger ·
Anett Rexin · Markus Friedrich ·
Antje Haeussler-Quade · Hans-Dieter Volk ·
Khusru Asadullah
The antipsoriatic activity of IL-10 is rather caused
by effects on peripheral blood cells than by a direct effect
on human keratinocytes
Received: 19 July 1999 / Revised: 6 December 1999 / Accepted: 10 December 1999
Abstract IL-10 is a promising candidate for the treat-
ment of cutaneous disorders. Antipsoriatic efficacy of
systemic IL-10 treatment has been already demon-
strated. This includes histomorphological changes in
the epidermis, suggesting effects on keratinocytes.
However, less is known about direct effects of IL-10 on
this cell population, although effects are likely since
IL-10 receptor expression on keratinocytes has been
demonstrated recently. Therefore we analysed the ef-
fects of IL-10 on keratinocytes in vitro, using concen-
trations of human recombinant IL-10 corresponding
to those detectable in plasma during therapy. Prolifer-
ation, cytokine formation (IL-6, IL-8, IL-1ra), and ex-
pression of surface molecules (MHC class I and II, cos-
timulatory molecules CD80 and CD86, CD29, CD54,
CD95) were measured in stimulated and unstimulated
cells. Although stimulation influenced the expression
levels of certain surface markers, no or only slight ef-
fects of IL-10 were found. In contrast considerable in-
hibitory effects of IL-10 on surface molecule expres-
sion and cytokine secretion by peripheral blood hu-
man monocytes were observed. Our results suggest
that the antipsoriatic activity of IL-10 is rather caused
by modulatory effects on circulating immune cells,
which subsequently might infiltrate the skin, than by
direct effects on human keratinocytes. Considering the
remarkable antipsoriatic activity of IL-10 and the ob-
servation that IL-10 seem to act on peripheral blood
mononuclear cells but not on keratinocytes provide
M. Seifert (쾷) · E. Effenberger · H. D. Volk
Department of Medical Immunology, Medical School Charité,
Humboldt University Berlin, Schumannstr. 20/21,
10098 Berlin, Germany
e-mail: martina.seifert@charite.de,
Tel.: +49-30-28026546, Fax: +49-30-28026166
K. Asadullah · A. Haeussler-Quade · M. Friedrich · A. Rexin ·
W. Sterry
Department of Dermatology, Medical School Charité,
Humboldt University Berlin
© Springer-Verlag 2000
further evidence that circulating immune cells play a
key role in the pathology of psoriasis. Finally, our re-
sults argue against the value of IL-10 therapy in der-
matoses strictly limited to keratinocyte involvement.
Key words IL-10 · Keratinocytes · Monocytes ·
Psoriasis
Abbreviations APC antigen-presenting cell ·
CD cluster of differentiation · ELISA enzyme-linked
immunosorbent assay · HLA human leucocyte antigen ·
IFN interferon · IL interleukin · KC keratinocyte(s) ·
LPS lipopolysaccharide · MFI mean fluorescence
intensity · MHC major histocompatibility complex ·
PBMC peripheral blood mononuclear cells · TNF tumor
necrosis factor
Introduction
Abnormal KC proliferation, differentiation and function is
a common phenomenon in different cutaneous disorders,
most prominently in psoriasis, a chronic relapsing disor-
der characterized by inflammation and increased epider-
mal proliferation with a prevalence of 2–3% in the general
population (Christophers and Sterry 1993). Recent data
have shown that peripheral blood mononuclear cells
(PBMC) are of major pathophysiological importance in
the development of psoriasis (Gilhar et al. 1997; Wrone-
Smith and Nickoloff 1996).
Very recently the therapeutic effect in psoriatic patients
of systemic IL-10 administration has been demonstrated.
Significant antipsoriatic activity has been found in phase
2 pilot trials with subcutaneous injections of 4 and 8 µg/kg
per day into nonlesional skin as well as 20 µg/kg three
times per week (Asadullah et al. 1998; Asadullah et al.
1999a; Reich et al. 1998). Although its mode of action is
not yet fully understood, we favour the concept that IL-10
exerts its antipsoriatic activity by effects on different cell
populations (Asadullah et al. 1998). This in particular in-
cludes T cells and antigen-presenting cells (APCs) as well

as their interactions. Psoriasis is considered to be a T-cell-
dependent (auto)immune disease (Valdimarsson et al.
1986), probably initiated by presentation of so-far-un-
known ‘psoriasis-related antigens’ by specialized cuta-
neous APCs (Mitra et al. 1995; Weinstein 1996). IL-10 is
able to suppress the antigen-presenting activity of mono-
cytes/macrophages and dendritic cells (De Waal Malefyt
et al. 1991; Fiorentino et al. 1991; Mitra et al. 1995). Pre-
sentation of antigens and/or superantigens activates T-ef-
fector cells (Norris et al. 1997; Weinstein 1996). In partic-
ular, type 1 cytokine-producing T cells seem to play an
important role in the pathogenesis of psoriasis (Gilhar et
al. 1997; Norris et al. 1997; Schlaak et al. 1994; Uyemura
et al. 1993). In fact we have observed immunosuppression
(decrease in monocytic HLA-DR expression, IL-12 plasma
levels and responsiveness to recall antigens) as well as a
shift towards a type 2 cytokine pattern (increasing propor-
tion of IL-4, IL-5, and IL-10 producing T cells, increase
in IgE serum levels) in our patients treated with IL-10,
which supports this concept (Asadullah et al. 1998; Asa-
dullah et al. 1999a, 1999b).
Besides the increasing evidence for the crucial role of
PBMC in psoriasis (Krueger et al. 1990; Nickoloff 1991;
Schlaak et al. 1994; Uyemura et al. 1993; Valdimarsson et
al. 1986), keratinocytes (KC) have also been suggested to
have a pathophysiological impact in psoriasis. They show
an abnormal behaviour in this disease, including en-
hanced proliferation, expression of surface molecules in-
dicating activation (such as MHC class II and ICAM-1),
and proinflammatory cytokine formation (Baker et al.
1988; Bruch-Gerharz et al. 1996; Krueger et al. 1990;
Mussi et al. 1994; Ruissen et al. 1996). Interestingly, in
our histological investigations of the use of IL-10 in the
treatment of psoriasis, we have observed a decrease in
epidermal thickness along with a decrease in parakerato-
sis, clearly demonstrating effects of IL-10 treatment on
KC (Asadullah et al. 1998; Asadullah et al. 1999a).
Although these might be indirect effects, mediated by ef-
fects of IL-10 on the inflammatory cells mentioned above,
direct effects of IL-10 on KC may also contribute to
the clinical response observed (Asadullah et al. 1999a,
1999b).
There is so far only very limited information about the
effect of IL-10 on KC. Recently it has been reported that
human KC express the IL-10 receptor and upregulation is
amplified by vitamin D3 and calcipotriol, a potent antip-
soriatic compound (Michel et al. 1997a, 1997b). This
strongly suggests that KC may be a target for IL-10. In
fact there is initial evidence that IL-10 inhibits TNF-α and
IL-6 synthesis (Becherel et al. 1995) and proliferation
(Michel et al. 1997a) by KC.
The aim of the present study was to investigate the ef-
fects of human recombinant IL-10 on human KC to deter-
mine whether direct effects of IL-10 on KC may con-
tribute to its antipsoriatic activity and to develop a further
basis for IL-10 administration in other dermatoses. There-
fore, the effects of IL-10 on proliferation, cytokine forma-
tion, and expression of surface molecules with a major
impact on immune regulation were analysed. Since under
165
pathopysiological conditions KC are usually stimulated the
effect of IL-10 in corresponding in vitro models was of
particular interest.
Materials and methods
Cell cultures
Human epidermal cell suspensions were obtained from normal
donors undergoing foreskin surgery. KC were propagated in a 3:1
DMEM/F12 mixture (Seromed, Biochrom, Germany) with full
supplements (0.1 ng/ml epidermal growth factor, 0.5 µg/ml hydro-
cortisone, 5 µg/ml insulin, choleratoxin; all from Sigma, Deisen-
hofen, Germany), 1% v/v penicillin-streptomycin (Life Technolo-
gies, Eggenstein, Germany) or KGM (Biowhittaker Europe, Bel-
gium). KC were passaged by dissociating the monolayer with
trypsin/EDTA (0.025% and 0.01% w/v, respectively; PAA Labo-
ratories, Cölbe, Germany). For the experiments, cells were derived
from the third to fifth passages grown as a monolayer to subcon-
fluency.
Additionally cells of the human KC cell line HaCaT (kindly
provided by N.E. Fusenig) (Boukamp et al. 1988) were cultured in
DMEM with 5% (v/v) FCS and 1% (v/v) antibiotic solution at
37°C in an atmosphere containing 5% CO2. HaCaT cells have
some similarities to psoriatic keratinocytes, such as keratin 17 ex-
pression (Bonnekoh et al. 1995).
Stimulation assays
Cell cultures of freshly isolated human KC (obtained from
foreskin) and HaCaT cells were used. IL-10 (SCH 52000; kindly
provided by Schering-Plough Research Institute/Essex Pharma)
was added to the cultures at the following concentrations: 0, 1, 10,
100 ng/ml. These concentrations correspond well with plasma con-
centrations found in volunteers after the subcutaneous administra-
tion of IL-10. In other cultures, cells were additionally stimulated
with 10 ng/ml IFN-γ (Genzyme Corporation, Cambridge, Mass.)
and/or 100 ng/ml endotoxin (LPS E. coli 0127:B8; Sigma, Deisen-
hofen, Germany). The cells were cultured for 6–48 h, as indicated.
Experiments were performed in triplicate with KC from different
donors. The specificity of the effects observed was confirmed by
the addition of neutralizing monoclonal anti-IL-10 antibody (Sabat
et al. 1996). Control experiments included the simultaneous mea-
surement of the effects of IL-10 on human monocytes. Therefore,
PBMC from three different donors were prepared according to
standard procedures and cultured in VLE-RPMI 1640/10% FCS
(Biochrom, Berlin, and Life Technologies, Eggenstein, Germany,
respectively).
Proliferation
Normal human KC and HaCaT cells were harvested and seeded in
96-well microtitre plates (104 cells/well) with 0, 1, 10 or 100 ng/ml
human recombinant IL-10 in quintuplicate assays. The rate of
DNA synthesis was determined after 24 and 48 h from the incor-
poration of 3[H]-thymidine (1 µCi/well). Cells were harvested and
counted (Inotech, Dunn Labortechnik, Asbach, Germany). Addi-
tionally, cell numbers were determined in triplicate by counting
with a standard haemocytometer chamber.
In order to demonstrate the validity of the test system, control
experiments using 1, 10 or 100 ng/ml IFN-γ were performed, since
IFN-γ is known to be a strong inhibitor of cell proliferation.
Cytokine formation
Supernatants of 24 h cultures of human KC or HaCaT cells were
harvested after appropriate stimulation with IFN-γ or LPS and for
166
a further 24 h with IL-10 and stored at –70 °C until cytokine analy-
sis by ELISA. Determination of cytokines was performed with the
following commercial ELISA kits: IL-6 (Quantikine; R&D Sys-
tems, Wiesbaden, Germany), IL-8 (Immunlite; DPC Biermann,
Germany) and IL-1ra (R&D Systems).

Expression of surface antigens

The expression of surface antigens on KC and monocytes was de-

termined by FACS analysis (Asadullah et al. 1998) after 24 h of
culture. Data are expressed as mean fluorescence intensity (MFI),
indicating the density of antigen expression per cell. Flow cyto-
metric analysis was performed using FACScan equipment and Ly-
sis II software (all from Becton Dickinson, Heidelberg, Germany).
Monoclonal antibodies labelled directly with FITC or PE were
used against the following antigens (clones in parentheses): HLA-
DR (L243) (Becton Dickinson), CD80 (BB1), CD 86 (2331FUN-1)
(all from Pharmingen, Hamburg, Germany), CD95 (UB2), CD29
(K20), CD54 (84H10) (all from Coulter-Immunotech, Hamburg,
Germany). Additionally isotype controls (IgG1 and IgG2a) were
used (Coulter-Immunotech, Hamburg, Germany).

Statistical analysis

Statistical analysis was performed using the Wilcoxon test, and

P < 0.05 was considered to be significant. Data are presented as
medians with confidence intervals (25th, 50th and 75th percentile
values).

Results

KC proliferation is not influenced

by IL-10 administration

The analysis of fresh human KC obtained from different

healthy donors and of the transformed KC of the cell line
HaCaT in vitro for DNA-synthesis by 3[H]-thymidine in-
corporation showed a normal doubling rate of 24 h for un-
treated cells. Administration of increasing amounts of re-
combinant IL-10, corresponding to the IL-10 plasma lev-
els occurring during IL-10 treatment, to the cultures
(Huhn et al. 1997) did not alter the cell proliferation after
24 h and 48 h of culture (Fig. 1a). Even the administration
of very high, almost nonphysiological, IL-10-concentra-
tions (up to 800 ng/ml) in two experiments did not affect
the proliferation (104.9% for 800 ng/ml IL-10 versus
Fig. 1a, b Effects of human IL-10 and human IFN-γ on the prolif-
100% for the medium control after 24 h of culture). In eration of primary human KC (a) and HaCaT cells (b). Cells were
contrast to IL-10 the addition of recombinant human IFN-γ cultured for 24 h and 48 h in medium (KGM, DMEM), or with IL-
at two concentrations to the cell cultures significantly re- 10 (1, 10 and 100 ng/ml) and IFN-γ (1 and 10 ng/ml). a 3[H]-thymi-
duced the 3[H]-thymidine incorporation as has been re- dine incorporation is expressed as percentage of proliferation com-
pared with KC cultured with medium only (n = 5; data are medi-
ported previously (Baker et al. 1988; Matsue et al. 1993; ans with confidence intervals shown as 25th, 50th and 75th per-
Marionnet et al. 1997). So after 48 h of culture cell repli- centile values; the error bars denote the 5th and 95th percentile
cation could hardly be detected, demonstrating the valid- values; the square symbol denotes the mean; *P < 0.05 versus me-
ity of the test system. Similar results were obtained with dium control). b Cell proliferation was determined by cell count-
the HaCaT cells (data not shown). The experimental 3H- ing with a haemocytometer chamber and is shown for one experi-
ment as the mean of triplicate determinations
thymidine incorporation data for the primary KC and for
the HaCaT cells were confirmed by the cell counting ex-
periments, shown here for a single experiment with Ha- IL-10 inhibits cytokine formation in monocytes
CaT cells (Fig. 1b). IL-10 had no effect on the numbers of but not in KC
KC or HaCaT cells.
To investigate the capacity of human KC to produce
different cytokines after triggering with inflammatory
 167

Fig. 2 Increase in IL-8 secretion by primary KC with prolonged

culture. IL-10 treatment of cells induced no significant changes in
the secretion level after 6, 12 and 24 h in culture. A representative
experiment with primary human KC is shown

agents, we compared the cytokine secretion pattern of un-

stimulated and IFN-γ-stimulated cells after 6, 12 and 24 h
of culture in low-calcium KGM and in high-calcium
DMEM/F12 medium. Culture supernatants were analysed
for the presence of the proinflammatory cytokines IL-6
and IL-8 and the antiinflammatory cytokine IL-1ra.
Analysis at different time-points showed an increased cy-
tokine secretion with increasing incubation time. So the
highest levels for all cytokines were detected after 24 h of
incubation. Figure 2 shows the kinetics of IL-8 secretion
by primary human KC from a single donor as an example.
Consequently, 24 h preincubation cultures were used for
the subsequent experiments analysing the effect of IL-10
on stimulated and unstimulated cytokine secretion by KC.
So, KC monolayers grown up to 90–100% confluence
were first stimulated with IFN-γ or LPS to simulate in-
flammation in vitro for 24 h. After that time cells were
washed and incubated for a further 24 h with appropriate
concentrations of IL-10, and then the supernatants were
analysed for the cytokines mentioned above to determine
whether a reversal of inflammation had taken place. Si-
multaneously the coadministration of IL-10 and stimuli
was performed. Fig. 3a–c IL-10 treatment of primary human KC had no influence
Remarkably, no effects of IL-10 on IL-6, IL-8 or IL- on the secretion of pro- and antiinflammatory cytokines. Secretion
1ra protein secretion by KC were observed (Figs. 2 and 3). of IL-6 (a), IL-8 (b) and IL-1ra (c) was analysed after 24 h of cul-
This was the case when using unstimulated KC as well as ture. No significant changes in cytokine release occurred in un-
stimulated cells or in cells stimulated with IFN-γ (10 ng/ml), LPS
KC prestimulated with different agents. So IL-10 could (100 ng/ml) or IFN-γ + LPS. The percentage of cytokine release
not reverse the secretion of the proinflammatory cy- compared to the IL-10-untreated level is shown. The means ± SEM
tokines IL-6 and IL-8 nor did it have an influence on IL- were calculated from three independent experiments using cells
1ra release after stimulation with IFN-γ or LPS. Figure 3 from different donors
shows the experimental results with cells from three dif-
ferent donors as mean secretion values of the appropriate
168

cytokine in comparison with the cultures without IL-10

administration. The same reaction pattern reflected the
analysis of the gene expression level: enhanced mRNA
expression after stimulation with LPS and/or IFN-γ but no
significant modulation by IL-10 administration (data not
shown).
In contrast to the lack of effect of IL-10 on KC, a dose-
dependent inhibition of proinflammatory cytokine secre-
tion by monocytes was clearly shown, confirming recent
in vitro results and in accordance with our results from ex
vivo investigations during IL-10 therapy (Asadullah et al.
1998). Figure 4 shows the effect of IL-10 on LPS-induced
IL-8 release by PBMC as an example. This downregula-
tion of IL-8 secretion could be reversed by the application
of a neutralizing anti-IL-10 antibody, indicating the speci-
ficity of the reaction observed (Fig. 4).

IL-10 modulates surface antigen expression

on monocytes but not on KC

We also determined the effects of human recombinant IL-

Fig. 4 Inhibition of IL-8 secretion of human PBMC by IL-10 treatment.
Strong inhibition of IL-8 secretion was detected for LPS-stimulated cul-
tures after subsequent 24 h culture with IL-10. A neutralizing anti-IL-10
antibody inhibited this effect. The means ± SEM were calculated from
three independent experiments using cells from different donors
10 on the expression of different surface antigens on hu-
man KC in comparison with human isolated PBMC as a
known reference system. The cells were incubated for 24 h
with IFN-γ (10 ng/ml), endotoxin (100 ng/ml), a combina-

Fig. 5a–d IL-10 treatment de-

creased HLA-DR expression
by human monocytes but not
by KC. Human KC (a) and
PBMC (b) were stimulated for
24 h by IFN-γ or LPS. Cells
were subsequently treated with
IL-10 (1, 10 or 100 ng/ml) for
a further 24 h. Surface expres-
sion was measured by flow cy-
tometry as described above and
the MFI was determined. The
means ± SEM were calculated
from three different experi-
ments. FACS histograms of
MHC class II expression by
human keratinocytes (c) and
monocytes (d) stimulated with
IFN-γ and further incubated
without IL-10 (thin line) or
with 10 ng/ml IL-10 (bold line)
are shown. IL-10 only down-
regulated MHC class II expres-
sion by human monocytes. The
dashed line shows the isotype
staining control
 169

CD95 was determined by FACS analysis. Values are mean fluorescence intensity ± SEM
ng/ml shown here). Surface expression of MHC class I, CD80, CD86, CD54, CD29 and
tion of both reagents and with medium alone. After preac-

60 ± 14

43 ± 14
74 ± 14

89 ± 4
77 ± 10

52 ± 2
54 ± 9
51 ± 7
tivation the supernatant was removed and the cells were

KC
incubated for a further 24 h with 0–100 ng/ml recombinant
human IL-10 so that any antiinflammatory effect of IL-10

14 ± 0.6
could be detected.

14 ± 6

17 ± 8
13 ± 2

26 ± 6
26 ± 12

44 ± 23
32 ± 16
PBMC
CD95
The FACS analysis of the cells showed the expected
results for monocytes. We observed a dose-dependent in-
hibition of MHC class II (HLA-DR) expression by stimu-

1059 ± 109
992 ± 123

1330 ± 128

1108 ± 44
1254 ± 217

1090 ± 63
1289 ± 9
1235 ± 110
lated and unstimulated cells (Fig. 5b). Coadministration of
a neutralizing anti-IL-10-specific murine monoclonal an-
tibody blocked this effect. In contrast, IL-10 did not affect
KC

HLA-DR expression by the KC preactivated by IFN-γ or

endotoxin (Fig. 5a). Although the MFI of HLA-DR ex-
55
11

56
38
2
17
35
28
pression by human KC was increased up to 25 times fol-
PBMC
CD29

190 ±
117 ±

458 ±
460 ±
164 ±
197 ±
178 ±
196 ±
lowing IFN-γ stimulation compared with the value for the
expression by unstimulated cells, IL-10 did not inhibit
this increase at all. HLA-DR expression by LPS-preacti-
vated KC was also not influenced by IL-10 (Fig. 5a).
26 ± 3
27 ± 5

638 ± 142
662 ± 120
29 ± 3
28 ± 4
560 ± 184
560 ± 168
Similarly, MHC class I expression by KC and mono-
cytes was upregulated by IFN-γ stimulation, but only
KC

monocytes showed a slight decrease in the MFI following

IL-10 stimulation (Table 1).
271 ± 0.2
155 ± 44
70 ± 38

396 ± 42
387 ± 63
243 ± 13

470 ± 139
479 ± 73

Analysis of the costimulatory molecules (CD80 and

PBMC

CD86) showed inhibition of CD86 expression only on hu-

CD54

man monocytes, whereas neither unstimulated nor stimu-

lated KC expressed this molecule (Table 1). This down-
regulating capacity of IL-10 on monocytes was inhibited
18 ± 6
20 ± 4

21 ± 8
24 ± 11
19 ± 6
18 ± 6
19 ± 8
19 ± 9

by coincubation with a neutralizing anti-IL-10 mono-

KC

clonal antibody. No influence of IL-10 on CD80 expres-

sion by monocytes or KC was found (Table 1). Although
Table 1 Comparison of surface marker expression by PBMC and by human KC. KC and
PBMC were cultured with medium or different stimuli (IFN-γ, LPS, combination of both)
for 24 h and were subsequently treated with three concentrations of human IL-10 (only 10

29 ± 6
18 ± 6

53 ± 21
62 ± 16
64 ± 20
56 ± 18
295 ± 24
264 ± 7

not influenced by IL-10, the increase in CD80 expression

PBMC
CD80

by monocytes but not by KC after stimulation with a com-

bination of IFN-γ and endotoxin was striking (Table 1).
The adhesion molecule ICAM-1 (CD54) was ex-
pressed by KC with a 23 time higher MFI after stimula-
14.9 ± 7
14.6 ± 7

17.3 ± 7
18.3 ± 6
14.0 ± 6
16.2 ± 8
17.8 ± 5
18.8 ± 6

tion with IFN-γ or endotoxin and IFN-γ compared with

KC

untreated cells, but this was also not influenced by IL-10.

The expression of CD54 by monocytes was increased by
all the stimulatory molecules, but also failed to react to
79.5 ± 23
141.8 ± 39

284.5 ± 117
106.9 ± 49
170.6 ± 113
106.5 ± 68
267.0 ± 152
195.2 ± 115

human IL-10 administration in this regard. The Fas-recep-

PBMC

tor (CD95) was induced on human KC and monocytes by

CD86

IFN-γ, but no influence of IL-10 on the expression was

detected in either cell type (Table 1). Stimulation of KC
by TNF-α induced strong upregulation of CD54, as well
345 ± 119
399 ± 64

1536 ± 122
1360 ± 146
364 ± 82
327 ± 92
953 ± 98
987 ± 167

as of other surface molecules such as MHC class I and

Fas, but no effects of IL-10 on the expression were found
KC

(data not shown).

Taken together, whereas IL-10 reverses the inflamma-
MHC class I

1041 ± 36
1111 ± 197

1932 ± 426
1728 ± 156
1019 ± 19
927 ± 5
2401 ± 311
1846 ± 177

tion-induced changes of the expression of certain surface

antigens (MHC class I and II, CD86) by human mono-
PBMC

cytes it does not affect the expression by KC at all.

IFNγ + LPS + IL-10

Discussion
IFN-γ (10 ng/ml)
IL-10 (10 ng/ml)

LPS (100 ng/ml)

IFN-γ + IL-10

IFN-γ + LPS
LPS + IL-10

Recently it has been reported that human KC express the

IL-10 receptor, strongly suggesting that these cells, which
Medium

show an abnormal behaviour in several dermatoses, may

be a target for IL-10 (Michel et al. 1997a, 1997b; Asadul-
170
lah et al. 1999a, 1999b). Moreover, in histological inves-
tigations during systemic IL-10 treatment as a new thera-
peutic approach in psoriasis, we and others have observed
a decrease in epidermal thickness along with a decrease in
parakeratosis, clearly demonstrating effects of IL-10 on
KC in vivo (Asadullah et al. 1998; Asadullah et al. 1999a;
Reich et al. 1998). Since these observations may also re-
flect indirect effects of IL-10, mediated via modulatory
effects of the IL-10 on immune cells, direct effects of IL-
10 on human KC were investigated in this study. We in-
vestigated the effects of IL-10 on KC proliferation, cyto-
kine formation and expression of surface molecules with
an impact on immunoregulation.
No effects of IL-10 on unstimulated or stimulated pri-
mary human KC or cultured HaCaT cells were found. De-
spite the well-known limitations of cell lines, in general
the lack of effects of IL-10 on HaCaT cells is of impor-
tance, since this cell line has some similarities concerning
hyperproliferation and the keratin expression pattern with
psoriatic KC (Bonnekoh et al. 1995). The lack of effects
of IL-10 on prestimulated KC is of particular interest,
since stimulation of KC is a common phenomenon in sev-
eral dermatoses. One potent ‘in vivo’ stimulator is IFN-γ,
which is considerably overexpressed in psoriasis (Baker
et al. 1988; Krueger et al. 1990; Nickoloff 1991; Schlaak
et al. 1994). Furthermore, bacteria (partly containing en-
dotoxin) might also contribute to the inflammatory re-
sponse of KC in certain dermatoses. Our in vitro prestim-
ulation of KC and monocytes with IFN-γ and endotoxin
led to enhanced expression levels of several immunologi-
cally important surface antigens (MHC molecules) and
enhanced cytokine secretion. This is in agreement with re-
cent observations demonstrating secretion of IL-6 and IL-
8 by KC after stimulation with IFN-γ (Fujisawa et al.
1997). Remarkably, IL-10 was able to inhibit these effects
on human monocytes, but interestingly not on cultured
human KC or HaCaT cells. It has been shown recently
that IL-10 is able to inhibit baseline and cytokine-stimu-
lated CD54 expression only by human Langerhans cells,
but not by dermal vascular endothelial cells, fibroblasts or
keratinocytes (Chatelain et al. 1998). This is in agreement
with our own observations under different stimulatory
conditions (data not shown in detail).
We can fairly exclude the possibility that the failure of
IL-10 to induce considerable changes in KC behaviour re-
sulted from technical problems. The validity of the KC
proliferation test system used was proven by simultane-
ously performed experiments showing dramatic inhibitory
effects on 3H-thymidine incorporation and cell number by
human recombinant IFN-γ treatment. This corresponds
well with previous data from other groups (Baker et al.
1988; Matsue et al. 1993; Marionnet et al. 1997). Simi-
larly, deactivating effects of IL-10 on human PBMC with
regard to their cytokine production and surface antigen
expression were found in parallel experiments using
IL-10 in the very same concentration range used by us and
others before (Asadullah et al. 1998; Mitra et al. 1995;
Sironi et al. 1993; Taga and Tosato 1992). This observa-
tion demonstrates the biological activity of the IL-10 we
used. Moreover, exactly the same human recombinant
IL-10 (identical charge) was used for the treatment of pso-
riatic patients and showed remarkable clinical effects
(Asadullah et al. 1998; Asadullah et al. 1999a).
Our results contradict previous reports of in vitro ef-
fects of IL-10 on KC. Recently it has been shown that IL-
10 inhibits the synthesis of TNF-α and IL-6 (Becherel et
al. 1995) as well as proliferation (Michel et al. 1997a) of
KC. This discrepancy may have resulted from differences
in culture conditions, the IL-10 proteins used, or the ex-
perimental procedures. Other groups have used higher
doses of IL-10 (up to 400 ng/ml) for a preincubation of
KC before stimulation. Although these results might have
some impact of the understanding of the biological effects
of IL-10, such experimental designs do not simulate
pathophysiological conditions. Consequently, a preincu-
bation of KC with IL-10 should not be appropriate as a
model for the effect of IL-10 therapy. Interestingly, other
groups have reported a possible influence of the timing of
the exposure of retinal pigment epithelial cells to IL-10,
which reacted to pre-/costimulation only by HLA-DR
downregulation, whereas coincubation alone had no effect
(Boorstein et al. 1997), which might explain our dis-
crepant results. Moreover, low concentrations of contami-
nating LPS in IL-10 preparations can induce LPS hypore-
sponsiveness (Randow et al. 1995; Ziegler-Heitbrock
1992) which may be of relevance to the effects seen by
others in preincubation experiments.
In our culture systems we used primary KC of a con-
venient passage number (three to five) and pharmacologi-
cally relevant concentrations of a highly purified IL-10
preparation which has shown clear effects in the treatment
of psoriasis patients in vivo. IL-10 concentrations in vitro
between 0.1 and 100 ng/ml were selected according to the
IL-10 plasma levels found 6 and 24 h after subcutaneous
administration of IL-10 at 0.3–0.5 pg/ml (Döcke WD,
manuscript in preparation). This corresponds well with
the peak values of IL-10 detected after subcutaneous ad-
ministration of IL-10 at 1–50 µg/kg to healthy volunteers
(Huhn et al. 1997; Radwanski et al. 1998). Moreover,
prestimulation of KC with the proinflammatory cytokine
IFN-γ or endotoxin mimics the inflammatory processes in
lesional psoriatic skin. The coadministration of IL-10 af-
ter a preactivation of skin cells should be a more suitable
model of the in vivo situation in psoriasis. However, even
parallel application of IL-10 with IFN-γ or LPS did not
show significant effects on the properties of KC in our ex-
periments.
Based on our results the reasons for the lack of signif-
icant effects of IL-10 on KC remain unclear because the
expression of the IL-10 receptor has clearly been shown
in the past in vitro and in situ (Michel et al. 1997a,
1997b). Moreover, we have recently observed a biological
effect of IL-10 on human KC, suggesting appropriate in-
tracellular signaling pathways. We have demonstrated that
cytokine-induced Fas-Ligand expression is downregu-
lated by IL-10 (Arnold et al. 1999).
Taken together, our results suggest that the antipsori-
atic activity of IL-10 is rather caused by immunomodula-

tory effects on peripheral blood cells than by direct effects
on human KC. This conclusion is in agreement with phar-
macokinetic analyses in IL-10-treated volunteers suggest-
ing that recombinant human IL-10 after a single dose does
not distribute extensively outside the vascular compart-
ment (Huhn et al. 1996). However, we would not exclude
the possibility that administration of IL-10 to psoriatic pa-
tients could reach all immune organs as well as lesional
skin regions and act on tissue-homing APCs and T cells.
Our observations further support the concept that pso-
riasis is mainly an immune system-mediated disease
(Gilhar et al. 1997; Nickoloff 1991; Nickoloff et al. 1994;
Wrone-Smith and Nickoloff 1996) initiated by pathogenic
blood-derived immune cells leading to activation and disso-
nant growth of epithelial cells (Wrone-Smith and Nicko-
loff 1996). Finally, our results argue against strategies
aiming at the use of IL-10 therapy for dermatoses where
the defect is fully restricted to KC alone.
Acknowledgements This work was partly supported by DFG (Ste
366/7) and (Vo SFB 421/B2) and Essex Pharma, Germany/Schering
Plough Research Institute, USA. We thank Prof. Schönberger for
providing skin probes.
References
Arnold R, Seifert M, Asadullah K, Volk HD (1999) Crosstalk be-
tween keratinocytes and lymphocytes via Fas/Fas ligand inter-
action: modulation by cytokines. J Immunol 162: 7140–7147
Asadullah K, Sterry W, Stephanek K, Jasulaitis D, Leupold M,
Audring H, Volk HD, Döcke W (1998) IL-10 is a key cytokine
in psoriasis. J Clin Invest 101: 783–794
Asadullah K, Döcke WD, Ebeling M, Friedrich M, Belbe G, Au-
dring H, Volk HD, Sterry W (1999a) IL-10 treatment of psori-
asis – clinical results of a phase II trial. Arch Dermatol 135:
187–192
Asadullah K, Döcke WD, Sabat R, Ebeling M, Volk HD, Sterry W
(1999b) Interleukin-10 in der Dermatologie. Hautarzt 50: 12–
19
Baker BS, Powles AV, Valdimarsson H, Fry L (1988) An altered
response by psoriatic keratinocytes to gamma interferon. Scand
J Immunol 28: 735–740
Becherel PA, LeGoff L, Ktorza S, Ouaaz F, Mencia Huerta JM,
Dugas B, Debre P, Mossalayi MD, Arock M (1995) Inter-
leukin-10 inhibits IgE-mediated nitric oxide synthase induction
and cytokine synthesis in normal human keratinocytes. Eur J
Immunol 25: 2992–2995
Bonnekoh B, Huerkamp C, Wevers A, Geisel J, Sebok B, Bange
FC, Greenhalgh DA, Bottger EC, Krieg T, Mahrle G (1995)
Up-regulation of keratin 17 expression in human HaCaT kera-
tinocytes by interferon-gamma. J Invest Dermatol 104: 58–61
Boorstein SM, Elner SG, Bian ZM, Strieter RM, Kunkel SL, Elner
VM (1997) Selective inhibition of HLA-DR expression in IFN-
gamma-stimulated human retinal pigment epithelial cells. Curr
Eye Res 16: 547–555
Boukamp P, Petrussevska RT, Breitkreuz D, Hornung J, Markham
A, Fusening NE (1988) Normal keratinization in a sponta-
neously immortalized aneuploid human keratinocyte line.
J Cell Biol 106: 761–771
Bruch-Gerharz D, Fehsel K, Suschek C, Michel G, Ruzicka T,
Kolb-Bachofen V (1996) A proinflammatory activity of inter-
leukin 8 in human skin: expression of the inducible nitric oxide
synthase in psoriatic lesions and cultured keratinocytes. J Exp
Med 184: 2007–2012
171
Chatelain R, Wollenberg A, Martin C, Panhans-Gross A, Bieber T,
Degitz K, Heckmann M (1998) IL-10 inhibits ICAM-1 expres-
sion on human Langerhans cells but not on human keratino-
cytes, dermal endothelial cells or fibroblasts. Arch Dermatol
Res 209: 477–482
Christophers E, Sterry W (1993) Psoriasis. In: Fitzpatric TB, Eisen
AZ, Wolff K, Freedberg IM, Austen KF (eds) Dermatology in
general medicine. McGraw-Hill, New York, pp 489–515
De Waal Malefyt R, Abrams J, Bennett B, Figdor CG, Vries JE de
(1991) Interleukin 10 (IL-10) inhibits cytokine synthesis by hu-
man monocytes: an autoregulatory role of IL-10 produced by
monocytes. J Exp Med 174: 1209–1220
Fiorentino DF, Zlotnik A, Mosmann TR, Howard M, O’Garra A
(1991) IL-10 inhibits cytokine production by activated macro-
phages. J Immunol 147: 3815–3822
Fujisawa H, Wang B, Sauder DN, Kondo S (1997) Effects of in-
terferons on the production of interleukin-6 and interleukin-8
in human keratinocytes. J Interferon Cytokine Res 17: 347–353
Gilhar A, David M, Ullmann Y, Berkutski T, Kalish R (1997) T-
lymphocyte dependence of psoriatic pathology in human psori-
atic skin grafted to SCID mice. J Invest Dermatol 109: 283–
288
Huhn RD, Radwanski E, O’Connell SM, Sturgill MG, Clarke L,
Cody RP, Affrime MB, Cutler DL (1996) Pharmacokinetics
and immunomodulatory properties of intravenously adminis-
tered recombinant human interleukin-10 in healthy volunteers.
Blood 87: 699–705
Huhn RD, Radwanski E, Gallo J, Affrime MB, Sabo R, Gonyo G,
Monge A, Cutler DL (1997) Pharmacodynamics of subcuta-
neous recombinant human interleukin-10 in healthy volunteers.
Clin Pharmacol Ther 62: 171–180
Krueger JG, Krane JF, Carter DM, Gottlieb AB (1990) Role of
growth factors, cytokines, and their receptors in the pathogene-
sis of psoriasis. J Invest Dermatol 94: 135–140
Marionnet AV, Lizard G, Chardonnet Y, Scmitt D (1997) Com-
parative evaluation of the antiproliferative effect of cyclosporin
A and gamma-interferon on normal and HPV-transformed ke-
ratinocytes by cell counting, MTT assay and tritiated thymidine
incorporation. Cell Biol Toxicol 13: 115–123
Matsue H, Bergstresser PR, Takashima A (1993) Reciprocal cy-
tokine-mediated cellular interactions in mouse epidermis: pro-
motion of gamma delta T-cell growth by IL-7 and TNF-α and
inhibition of keratinocyte growth by gamma IFN. J Invest Der-
matol 101: 543–548
Michel G, Mirmohammadsadegh A, Olasz E, Jarzebska-Deussen
B, Müschen A, Kemény L, Abts HF, Ruzicka T (1997a)
Demonstration and functional analysis of IL-10 receptors in
human epidermal cells: decreased expression in psoriatic skin,
down-modulation by IL-8, and up-regulation by an antipsori-
atic glucocorticosteroid in normal cultured keratinocytes. J Im-
munol 159: 6291–6297
Michel G, Gailis A, Jarzebska-Deussen B, Muschen A, Mirmo-
hammadsadegh A, Ruzicka T (1997b) 1,25-(OH)2-vitamin D3
and calcipotriol induce IL-10 receptor gene expression in hu-
man epidermal cells. Inflamm Res 46: 32–34
Mitra RS, Judge TA, Nestle FO, Turka LA, Nickoloff BJ (1995)
Psoriatic skin-derived dendritic cell function is inhibited by ex-
ogenous IL-10. Differential modulation of B7-1 (CD80) and
B7-2 (CD86) expression. J Immunol 154: 2668–2677
Mussi A, Bonifati C, Carducci M, Viola M, Tomaselli R, Sacerdoti
G, Fazio M, Ameglio F (1994) IL-10 levels are decreased in
psoriatic lesional skin as compared to the psoriatic lesion-free
and normal skin suction blister fluid. J Biol Regul Homeost
Agents 8: 117–120
Nickoloff BJ (1991) The cytokine network in psoriasis. Arch Der-
matol 127: 871–884
Nickoloff BJ, Fivenson DP, Kunkel SL, Strieter RM, Turka LA
(1994) Keratinocyte interleukin-10 expression is upregulated in
tape stripped skin, poison ivy dermatitis and Sézary syndrome,
but not in psoriatic plaque. Clin Immunol Immunopathol 73:
83–86
172
Norris DA, Traves JB, Leung DYM (1997) Lymphocyte activation
in the pathogenesis of psoriasis. J Invest Dermatol 109: 1–4
Radwanski E, Chakraborty A, Van Wart S, Huhn RD, Cutler DL,
Affrime MB, Jusko WJ (1998) Pharmacokinetics and leuko-
cyte response of recombinant interleukin-10. Pharm Res 15:
1895–1901
Randow F, Syrbe U, Meisel C, Krausch D, Zuckermann H, Platzer
C, Volk HD (1995) Mechanisms of endotoxin desensitization:
involvement of interleukin 10 and transforming growth factor
β. J Exp Med 181: 1887–1892
Reich K, Brück M, Gräfe A, Vente C, Neumann C, Garbe C
(1998) Treatment of psoriasis with interleukin-10. J Invest Der-
matol 6: 1235–1236
Ruissen F van, Jongh GJ de, Erp PE van, Boezman JB, Schalkwijk
J (1996) Cell kinetic characterization of cultured human kera-
tinocytes from normal and psoriatic individuals. J Cell Physiol
168: 684–694
Sabat R, Seifert M, Volk HD, Glaser RW (1996) Neutralizing
murine monoclonal anti-interleukin 10 antibodies enhance
binding of antibodies against different epitopes. Mol Immunol
33: 1103–1111
Schlaak JF, Buslau M, Jochum W, Hermann E, Girndt M, Gallati
H, Meyer zum Büschenfelde KH, Fleischer B (1994) T cells in-
volved in psoriasis vulgaris belong to the Th 1 subset. J Invest
Dermatol 102: 145–149
Sironi M, Munoz C, Pollicino T, Siboni A, Sciacca FL, Bernasconi
S, Vecchi A, Colotta F, Mantovani A (1993) Divergent effects
of interleukin-10 on cytokine production by mononuclear
phagocytes and endothelial cells. Eur J Immunol 23: 2692–
2695
Taga K, Tosato G (1992) IL-10 inhibits human T cell proliferation
and IL-2 production. J Immunol 149: 1143–1148
Uyemura K, Yamamura M, Fivenson DF, Modlin RL, Nickoloff
BJ (1993) The cytokine network in lesional and lesion-free
psoriatic skin is characterized by a T-helper type 1 cell-medi-
ated response. J Invest Dermatol 101: 701–705
Valdimarsson H, Baker BS, Johnsdottir I, Fry L (1986) Psoriasis:
a disease of abnormal keratinocyte proliferation induced by T
lymphocytes. Immunol Today 7: 256–259
Weinstein GD (1996) Can immunomodulatory molecules work
topically for psoriasis? J Invest Dermatol 106: 589
Wrone-Smith T, Nickoloff BJ (1996) Dermal injection of im-
munocytes induces psoriasis. J Clin Invest 98: 1878–1887
Ziegler-Heitbrock HWL (1992) In vitro desensitization to lipo-
polysaccharide suppresses tumor necrosis factor, interleukin-1
and interleukin-6 gene expression in similar fashion. Immunol-
ogy 75: 264–268