(Received for publication, October 3, 1996, and in revised form, January 23, 1997)
From the Department of Dermatology and the
Institute of Molecular Biology and Biochemistry, University
Medical Center Benjamin Franklin, The Free University of Berlin,
D-12200 Berlin, Germany
Treatment of the human keratinocyte cell line
HaCaT with 1,25-dihydroxyvitamin D3
(1,25-(OH)2D3) resulted in the hydrolysis of
sphingomyelin with peak elevations of ceramide levels after 2-3 h
(Geilen, C. C., Bektas, M., Wieder, Th., and Orfanos, C. E. (1996)
FEBS Lett. 378, 88-92). In the present paper, the
mechanism underlying the effect of 1,25-(OH)2D3
on sphingomyelin hydrolysis was investigated. Using the cell culture
supernatant of HaCaT cells treated with
1,25-(OH)2D3 for 2 h, it was possible to
induce sphingomyelin hydrolysis as early as 30-60 min after addition to resting cells. Several lines of experimental evidence indicated that
tumor necrosis factor
(TNF
) mediates sphingomyelin hydrolysis after 1,25-(OH)2D3 treatment: (i)
1,25-(OH)2D3 stimulated TNF
mRNA
expression after 1 h, (ii) newly synthesized TNF
occurred 2 h after 1,25-(OH)2D3 treatment, (iii)
indirect activation of sphingomyelin hydrolysis by the supernatant of
1,25-(OH)2D3-treated HaCaT cells was abolished
by preincubation of the supernatant with antibodies directed against
TNF
, and (iv) preincubation of HaCaT cells with neutralizing
antibodies directed against the 55-kDa receptor of TNF
blocked the
ability of 1,25-(OH)2D3 to induce sphingomyelin
hydrolysis in HaCaT cells. These data demonstrate that
1,25-(OH)2D3 activated sphingomyelin hydrolysis
by an autocrine mechanism via TNF
expression. Furthermore, this
indirect mode of action may serve as an explanation for the delayed
induction of sphingomyelin hydrolysis by vitamin D3.
1,25-Dihydroxyvitamin D3
(1,25-(OH)2D3)1 is
the first compound identified as an inducer of reversible sphingomyelin
(SM) hydrolysis in HL-60 human leukemia cells (1) leading to the
generation of ceramide. Using cell-permeable ceramide analogues, it was
possible to mimic the effects of 1,25-(OH)2D3
on cell proliferation and differentiation (2) and thereby confirm
ceramide to be an important mediator of
1,25-(OH)2D3-induced HL-60 cell differentiation
(3). In these studies, 1,25-(OH)2D3 caused SM
hydrolysis in a time- and concentration-dependent manner,
with a peak elevation in ceramide levels occurring 2 h after
treatment of cells with 100 nM
1,25-(OH)2D3. After 4 h, both
sphingolipids (SM and ceramide) returned to base-line levels by the
transfer of a choline head group from phosphatidylcholine to ceramide
(4). To date, several other agonists of the SM cycle have been
described, including tumor necrosis factor
(TNF
), interferon-
, dexamethasone, interleukin-1, nerve growth factor, complement, and brefeldin A (5-8). In human keratinocytes and in the
immortalized human keratinocyte cell line HaCaT, not only the well
known inducers 1,25-(OH)2D3 and TNF
but also
the vitamin D3 analogue calcipotriol activated SM
hydrolysis. In this study, 1,25-(OH)2D3 and
calcipotriol were shown to be late inducers of SM hydrolysis with peak
levels after 3 h, whereas TNF
was shown to be an early inducer
with peak levels after 30-60 min (9). However, the reason for these
different time courses occurring in human keratinocytes as well as in
HL-60 cells remained unclear, and the mechanisms underlying the
coupling of these various extracellular inducers to activation of
intracellular sphingomyelinase are largely unknown. In the case of
TNF
, it was shown that interaction of TNF
with its 55-kDa
receptor activates a phospholipase A2 generating arachidonic acid, which subsequently stimulates the cytosolic sphingomyelinase activity (10-12). As has been demonstrated recently, generation of ceramide then leads to activation of a protein kinase cascade, thereby implicating ceramide-activated protein kinase as a
link between the TNF receptor and Raf1 (13). On the other hand, it is
well established that 1,25-(OH)2D3 regulates
gene expression via binding to an intracellular vitamin D receptor that
belongs to the nuclear steroid hormone receptor superfamily (14, 15).
Human keratinocytes are known to possess the vitamin D receptor (16),
and 1,25-(OH)2D3 regulates the gene expression of several proteins in keratinocytes (17).
The aim of the present study was to investigate the mechanism
responsible for the delayed effect of
1,25-(OH)2D3 on SM hydrolysis, and strong
experimental evidence is provided that
1,25-(OH)2D3 induces SM breakdown indirectly
via expression and secretion of TNF in the human keratinocyte cell
line HaCaT.
Streptomyces sp. sphingomyelinase,
TNF, phenylmethylsulfonyl fluoride, and leupeptin were purchased
from Sigma (München, Germany). The TNF
ELISA
kit was purchased from R & D Systems (Minneapolis), and anti-TNF
receptor (human) antibodies were from Bender MedSystems (Vienna,
Austria). Solvents and reagents (reagent grade) were obtained from
Merck (Darmstadt, Germany) and Fluka (Neu-Ulm, Germany), and Triton
X-100 was purchased from Aldrich (Steinheim, Germany).
[methyl-3H]Choline chloride (2.8-3.1
TBq/mmol), [
-32P]dCTP (110 TBq/mmol), and
L-35S-labeled Pro-mixTM (in
vitro cell labeling mixture) (>37 TBq/mmol) were from
Amersham (Braunschweig, Germany). Anti-TNF
(human) antibodies were
purchased from PeproTech Inc. (London, United Kingdom).
1,25-(OH)2D3 was a gift from Dr. Lise Binderup
(Leo Pharmaceutical Products, Copenhagen, Denmark).
HaCaT cells (18) were grown in Roswell Park
Memorial Institute medium supplemented with 10% heat-inactivated fetal
calf serum, 0.35 g/liter glutamine, 100,000 IU/liter penicillin, and 0.1 g/liter streptomycin in plastic culture dishes (Nunc, Wiesbaden, Germany). Media and culture reagents were obtained from Life
Technologies, Inc. (Karlsruhe, Germany). Penicillin and streptomycin
were from Boehringer (Mannheim, Germany). For experimental purposes,
HaCaT cells were either maintained in keratinocyte basal medium (KBM) or in keratinocyte growth medium that was prepared from KBM by the
addition of 10 ng/ml epidermal growth factor, 5 µg/ml insulin, 0.5 µM hydrocortisone, 50 µg/ml bovine pituitary extract,
100 µg/ml penicillin/streptomycin, and 2.5 µg/ml Fungizone. KBM and supplements were purchased from Clonetics Corp. (San Diego, CA). 1,25-(OH)2D3 and TNF were diluted into KBM
from 10 µM stock ethanol solutions or from 600 nM stock solutions containing 0.1% bovine serum albumin,
respectively. NGF was diluted into KBM from 7.5 µM stock
solution containing 0.1% bovine serum albumin.
For choline labeling of HaCaT cells,
medium was removed, and pulse medium (keratinocyte growth medium
containing 3.7 × 104 Bq/ml
[methyl-3H]choline) was added. After
incubation for 72 h, cells were washed twice with
phosphate-buffered saline (PBS) and then treated with 100 nM 1,25-(OH)2D3 in KBM. Cells were
harvested in 400 µl of ice-cold PBS by the use of a cell lifter
(Costar Corp., Cambridge, MA). After freeze-drying of the cells, lipids
were extracted by a modified method of Bligh and Dyer (19) as described
(20). Total lipid extracts were dried under a stream of nitrogen and stored at 20 °C.
SM was quantified using
bacterial sphingomyelinase to release [3H]phosphocholine
as described recently (12). Briefly, the total lipid extracts were
resuspended in 100 µl of assay buffer (100 mM Tris-HCl,
pH 7.4, 6 mM MgCl2, and 0.1% Triton X-100).
The samples were sonicated for 5 min, and 1 unit/ml sphingomyelinase
from Streptomyces sp. was added. After incubation at
37 °C for 2 h, the reaction was stopped by the addition of 1 ml
of chloroform/methanol (2:1, v/v), and the liberated
[3H]phosphocholine was recovered with the aqueous phase
of a Folch extraction (22). Phase separation was completed by the
addition of 100 µl of water, and the aqueous phase was taken for
scintillation counting. The radioactivity of the aqueous phase normally
reached 10,000 cpm, and the radioactivity in control samples was set as 100%. Subsequently, SM in the samples of
1,25-(OH)2D3-, TNF-, and NGF-treated cells
was calculated as percent of control. The assay conditions described
above yielded maximal SM hydrolysis (>98%) without accompanying
phosphatidylcholine hydrolysis (<5%) (data not shown).
Confluent HaCaT cells were incubated with 100 nM 1,25-(OH)2D3 in KBM. After
different incubation times, cellular mRNA was extracted from cells
using a commercially available kit from Stratagene (Heidelberg,
Germany) according to the instructions of the manufacturer, and then
separated by electrophoresis in a 6% formaldehyde, 1.5% agarose gel
and blotted onto a nylon membrane (GeneScreen Plus, DuPont). The
filters were probed with the cDNA in question and reprobed using a
-actin cDNA probe. The Northern blot was hybridized (23) with
either of the following double-stranded cDNA probes: a 444-base
pair polymerase chain reaction (PCR) DNA fragment covering nucleotides
404-847 of human TNF
or an 838-base pair PCR DNA fragment covering
nucleotides 294-1131 of human
-actin. Both probes were generated
using PCR-Amplimers (Clontech, Palo Alto, CA) and a PCR amplification
kit (Perkin-Elmer, Überlingen, Germany) (24). The cDNA probes
were labeled with [
-32P]dCTP using the random priming
kit RPN 1601 (Amersham-Buchler, Braunschweig, Germany) (25).
Hybridizations were carried out overnight under stringent conditions at
60 °C in 1 M NaCl, 10% dextran sulfate, 1% SDS, 100 µg/ml salmon sperm DNA, and 2-4 × 105 cpm of the
respective probes/ml (26). The filters were then washed twice at room
temperature in 2 × SSC (0.3 M NaCl, 30 µM sodium citrate 2-hydrate, pH 7.0) plus 0.5% SDS
followed by washing twice in the same solution at 60 °C for 30 min.
Finally, the filters were washed twice in 0.1 × SSC for 5 min.
Autoradiography was carried out at
80 °C using Kodak XAR/5 film
with an intensifying screen.
Confluent HaCaT cells were washed twice with
PBS and treated with 100 nM
1,25-(OH)2D3 in KBM. After different incubation
times, an aliquot of the cell culture medium was removed from cells by centrifugation. TNF was detected in the supernatant by a
commercially available ELISA kit (QuantikineTM, R & D
Systems). 100 nM 1,25-(OH)2D3 in
KBM was used as a blank. Detection of standard TNF
was linear over
the range 2-25 pg/ml.
Confluent HaCaT cells were stimulated with 100 nM 1,25-(OH)2D3 in KBM for 2 h. The cell culture medium was removed, clarified from cells by centrifugation, and added to confluent HaCaT cells that have been previously labeled for 72 h with 3.7 × 104 Bq/ml [3H]choline. Cell culture medium from HaCaT cells stimulated with 1% ethanol was used as control. After different time periods, cells were washed with ice-cold PBS and harvested. The SM content of the samples was then determined as already described above.
Immunoprecipitation of TNFFor this,
confluent HaCaT cells were stimulated with 100 nM
1,25-(OH)2D3 in KBM for 2 h. The cell
culture medium was removed, briefly centrifuged, and treated with
rabbit anti-human TNF antibodies (PeproTech Inc.) or PBS as control
at 4 °C for 2 h under moderate shaking conditions. The immune
complex was precipitated with protein A-Sepharose for 1 h at
4 °C, and the protein A-Sepharose complex was sedimented by
centrifugation. Control supernatant was likewise treated with protein
A-Sepharose. The supernatants were incubated once more with the
respective antibodies or with PBS as control, and the procedure was
repeated. Subsequently, confluent HaCaT cells that had been previously
labeled for 72 h with 3.7 × 104 Bq/ml
[3H]choline were washed twice with PBS and incubated with
the immune-precipitated supernatant. After 10, 30, and 60 min cells
were washed with ice-cold PBS and harvested, and the SM content of the
samples was determined as already described above.
For this,
confluent HaCaT cells were preincubated for 2 h with
cysteine/methionine-free Dulbecco's minimal essential medium (Life
Technologies, Inc.). The cells were labeled with 10 µCi/ml L-35S-labeled Pro-mixTM (in
vitro cell labeling mixture) for 2 h and subsequently treated with 100 nM 1,25-(OH)2D3 or 1%
EtOH as a control. After different times, cell culture supernatant was
removed, and the cells were washed twice with PBS and harvested in 2 ml
of 50 mM Tris-HCl, pH 7.2, 150 mM NaCl, 1%
Triton X-100, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate,
1 mM phenylmethylsulfonyl fluoride, and 5 µg/ml
leupeptin. 20 µl of anti-TNF antibody (Sigma)
were then added to 2 ml of cell culture supernatant of 2 ml of cellular extract, and the samples were incubated overnight at 4 °C. To immunoprecipitate TNF
, 5 mg of protein A-Sepharose were used (27).
After 5 h at 4 °C under moderate shaking conditions, the protein A-Sepharose immune complex was sedimented by centrifugation at
4 °C and washed 3 times with ice-cold PBS. Finally, 90 µl of non-reducing SDS sample buffer were added, and the samples were boiled
for 5 min. The protein A-Sepharose was sedimented by centrifugation, and the supernatant was loaded on SDS-polyacrylamide gel
electrophoresis (15% acrylamide) and electrophoresed (28). The gels
were dried, and the bands corresponding to TNF
were visualized and
analyzed using a PhosphorImager (Molecular Dynamics GmbH, Krefeld,
Germany). Under the condition described above, TNF
migrated with an
apparent molecular mass of 17 kDa, which was confirmed with prestained SDS-polyacrylamide gel electrophoresis standards (Bio-Rad).
Confluent HaCaT cells that
had been previously labeled for 72 h with 3.7 × 104 Bq/ml [3H]choline were washed twice with
PBS and then pretreated for 2 h with purified neutralizing
antibodies against the human 55-kDa TNF receptor, non-immune serum,
or 0.1% bovine serum albumin in KBM as control. Subsequently,
1,25-(OH)2D3 in ethanol or ethanol vehicle
alone was added to the cell culture medium to achieve a concentration
of 100 nM and 1%, respectively. After 3 h of
incubation, cells were washed with ice-cold PBS and harvested. SM was
then quantified as described above.
Statistical comparisons were made in these studies with Student's t test.
Treatment of
HaCaT cells with 100 nM
1,25-(OH)2D3 or the
1,25-(OH)2D3 analogue calcipotriol resulted in
a time-dependent increase of sphingomyelin hydrolysis with
a maximum after 3 h that was accompanied by an elevation of
cellular ceramide levels (9). To establish the kinetics of the effect
of 1,25-(OH)2D3 and TNF on SM hydrolysis in
the cells utilized for the following studies, an initial time course
study was performed. Similar to previous studies,
1,25-(OH)2D3 induced a significant decrease of
SM levels after 3 h of treatment, whereas in the case of TNF
,
SM turnover of 12% was observed as early as 30 min following treatment
of cells (Table I). Maximal effects of SM hydrolysis
were observed 1 h after treatment with 30 nM TNF
.
These different time courses of 1,25-(OH)2D3
and TNF
are in accordance with results obtained in the leukemic cell
line HL-60 (1, 12, 29), which clearly demonstrated that
1,25-(OH)2D3 is a late and TNF
an early
inducer of SM hydrolysis. It is interesting that the early induction of SM hydrolysis by TNF
is even more pronounced in the human acute leukemic Jurkat T cell line and the human monocytic cell line U937,
which showed SM turnover with maximal effects after 2-3 min (30). In
additional experiments, NGF was tested for its ability to induce SM
hydrolysis. As shown in Table I, there was no effect of NGF on SM
turnover of HaCaT cells after 0.5, 1, and 3 h, and the slight
decrease of SM levels after 6 h was not significant.
|
Considering that
1,25-(OH)2D3 modulates gene transcription, it
seemed reasonable that the late activation of SM hydrolysis by
1,25-(OH)2D3 occurs indirectly via expression
and secretion of a mediator. To test this hypothesis, HaCaT cells were
treated with 1,25-(OH)2D3 for 2 h, then
the cell culture supernatant was removed and added to resting HaCaT
cells. As shown in Fig. 1, the supernatant of
1,25-(OH)2D3-treated HaCaT cells induced SM hydrolysis as early as 20 min after addition to the cells with peak
levels after 60 min. In an additional control experiment, HaCaT cells
were treated with 1,25-(OH)2D3-containing
medium that had been preincubated for 2 h in plastic culture
dishes, and no significant decrease of SM was observed after 20-60 min
of incubation (data not shown; see also Table I)
Besides 1,25-(OH)2D3, additional activators of
SM hydrolysis are described (e.g. TNF, interleukin-1,
interferon-
, NGF, or dexamethasone (6, 7)). In this context, it is
known that interleukin-1 is down-regulated and NGF is up-regulated by
1,25-(OH)2D3 (17). Thus NGF was a possible
candidate for the suggested mediator of indirect
1,25-(OH)2D3 action. As described above,
however, NGF did not influence SM hydrolysis in HaCaT cells (Table I) indicating that this growth factor does not mediate
1,25-(OH)2D3-induced effects in SM hydrolysis.
On the other hand, preliminary results showed that TNF
expression
may also be increased by
1,25-(OH)2D3.2 To
test if 1,25-(OH)2D3 indirectly stimulated SM
hydrolysis via TNF
, confluent HaCaT cells were treated with 100 nM 1,25-(OH)2D3 for 2 h. The
cell culture supernatant was then harvested, TNF
was
immunoprecipitated, and the supernatant was investigated for its
ability to induce SM hydrolysis in choline-prelabeled HaCaT cells. As
shown in Table II, precipitation of TNF
abolished the ability of the supernatant of
1,25-(OH)2D3-treated HaCaT cells to induce SM
hydrolysis after 30 and 60 min.
|
From the data obtained, we focused our
interest on TNF and the time courses of TNF
mRNA expression.
TNF
medium levels and TNF
biosynthesis were investigated in HaCaT
cells after stimulation with 1,25-(OH)2D3.
As shown in Fig. 2, 1,25-(OH)2D3
induced TNF mRNA expression after 1 h, and the signal
diminished after 2-3 h. These findings were confirmed by amplification
of the isolated mRNA using reverse transcription-PCR and detection
of TNF
cDNA by Southern blot (data not shown). In these
experiments, the slight increase of TNF
mRNA after 3 h of
incubation that is seen in Fig. 2 was not observed.
To investigate TNF biosynthesis in response to
1,25-(OH)2D3, HaCaT cells were labeled with
[35S]cysteine/methionine. After different incubation
times, TNF
was immunoprecipitated from the supernatant and cell
lysates. Whereas in the cell lysates an elevation of labeled TNF
was
detected after 1 h with a peak level after 2 h, labeled
TNF
occurred in the medium after 2 h (Fig. 3).
These data on biosynthesis were confirmed by determination of TNF
levels in the cell culture supernatant after treatment of HaCaT cells
with 100 nM 1,25-(OH)2D3 using the
ELISA technique as described under "Experimental Procedures," and
an elevated level of 5.5 pg/ml TNF
occurred after 2 h. In ethanol-treated control cells, the level of TNF
was below the detection limit of 2 pg/ml.
Neutralizing Antibodies against the 55-kDa TNF
The data on TNF production described above are in
accordance with immunohistological studies that have shown that
epidermal keratinocytes had intracellular plasma membrane and
cytoplasmic labeling for TNF
(31). To investigate the role of TNF
in mediating the 1,25-(OH)2D3 effect on SM
hydrolysis in more detail, the TNF
receptor was blocked by
preincubation with neutralizing antibodies against the human 55-kDa
TNF
receptor. This was done because it was shown in a previous study
that the 55-kDa TNF
receptor is crucial for the effect of TNF
on
SM hydrolysis (30). As shown in Table III, pretreatment
of HaCaT cells with these neutralizing antibodies abolished
1,25-(OH)2D3-induced SM hydrolysis after 3 h, indicating that the mechanism underlying the stimulation of SM
hydrolysis by 1,25-(OH)2D3 is mediated via
TNF
. Preincubation of control cells with bovine serum albumin or
non-immune serum did not influence SM hydrolysis. In this context, it
is interesting to note that the 55-kDa but not the 75-kDa TNF
receptor was shown to be expressed in human keratinocytes (32). Human
keratinocytes are therefore able to synthesize and simultaneously bind
TNF
, thereby representing an autocrine loop. A similar production
and auto-induction has been shown in keratinocytes for transforming growth factor
(21).
|
In conclusion, the present study has demonstrated for the first time by
several lines of experimental evidence that
1,25-(OH)2D3 indirectly induces the hydrolysis
of SM. Furthermore, TNF was shown to be the mediator of this
indirect activation. The time course of
1,25-(OH)2D3-induced TNF
mRNA
expression, TNF
biosynthesis, TNF
secretion, and TNF
-induced
SM hydrolysis fit well with the time course of
1,25-(OH)2D3-induced SM hydrolysis.
Dedicated to Professor Werner Reutter on the occasion of his 60th birthday.
We thank M. Hoffmann for excellent technical assistance and Dr. N. E. Fusenig, Deutsches Krebsforschungszentrum, Heidelberg, Germany, for the gift of HaCaT cells.