(Received for publication, December 18, 1996, and in revised form, February 14, 1997)
From the Neuroscience Laboratory, Institute of Biochemistry, Faculty of Medicine, University of Cologne, Joseph-Stelzmann-Strasse 52, D-50931 Cologne, Germany
Tumor necrosis factor (TNF-
) is one of the
most potent inducer of the nuclear transcription factor
B (NF-
B).
Activation of NF-
B is initiated by phosphorylation of the inhibitory
subunit of the I
B-
-NF-
B complex. This leads to the
dissociation of the complex and degradation of I
B-
. NF-
B is
translocated into the nucleus. The sphingomyelin pathway is thought to
mediate the TNF-
-induced activation of NF-
B by its second
messenger ceramide. We have used the recently established acid
sphingomyelinase-deficient mouse line
(asmase
/
mice) to evaluate the role of acid
sphingomyelinase in the TNF-
-induced signal transduction pathway.
Here we present experimental evidence that acid sphingomyelinase is not
involved in the TNF-
-induced activation of NF-
B. TNF-
treatment induced the dissociation and degradation of I
B-
and the
nuclear translocation of NF-
B in embryonic fibroblasts derived from
asmase
/
and wild type mice
indiscriminately.
A ubiquitous, evolutionarily conserved signaling pathway with sphingomyelin as precursor lipid for the second messenger ceramide has been postulated. It is comparable with the other well known signal transduction systems (for reviews, see Refs. 1-3).
The second messenger ceramide is generated by hydrolysis of sphingomyelin by sphingomyelinases. At least two sphingomyelinases are reported to participate in signal transduction, acid (ASM)1 and neutral sphingomyelinase (NSM). Acid sphingomyelinase is a lysosomal hydrolase with a pH optimum of 4.5-5.5 required for membrane turnover (4). The neutral, Mg2+-dependent isoform (pH optimum 7.4) is localized on the outer leaflet of the plasma membrane (5). While the ASM is well characterized at the molecular level (6, 7) and an ASM null-allelic mouse model is available (8, 9), the NSM has only partially been purified (10, 11).
Sphingomyelinase activation has been linked to several cell surface
receptors (e.g. the 55-kDa TNF- receptor (TR55) or the 80-kDa interleukin 1 receptor (2)). Activation of ASM has been described for TNF-
, Fas, and CD28 (12-14), and of NSM for TNF-
, vitamin D3, interleukin 1-1
, and ionizing radiation
(12, 15-17).
As direct targets for ceramide action ceramide-activated protein kinase
(18), ceramide-activated protein phosphatase (19), protein kinase C-
(20, 21) and the putative guanine nucleotide exchange factor Vav (22)
have been identified.
Ceramide is thought to mediate several physiological effects in
different cell types, including the activation of the transcription factor NF-B (23), induction of apoptosis (24, 25), and mitogenic
signaling (26).
NF-B, first described as a B-cell-specific factor responsible for
expression of the immunoglobulin-
gene (27), participates in the
regulation of several genes, most of them being involved in the early
events of immune, acute phase, or inflammatory responses (for review,
see Ref. 28). NF-
B is also responsible for the transcriptional
activation of many viral genes, e.g. of the human immunodeficiency virus-1 (29). Inactive NF-
B is a heterodimer consisting of a 50-kDa and a 65-kDa subunit retained in the cytoplasm by association with a 36-kDa inhibitory subunit I
B-
(30-32). This complex can be activated by several effector molecules like TNF-
or interleukin 1 (33), starting with serine phosphorylation and
subsequent ubiquitination and proteolytic degradation of I
B-
(34-37). The free NF-
B dimer translocates into the nucleus and activates the transcription of several genes (28).
Recently it has been reported that different receptor domains of the
TR55 activate the different sphingomyelinases (12). Truncated TR55
mutants, which lack the so-called death domain, an 80-amino acid
residue C-terminal sequence, neither showed NF-B activation nor an
increase of ASM activity, while the NSM activation was not affected
when expressed in 70Z/3 cells, suggesting that activation of NF-
B is
mediated exclusively by the acid sphingomyelinase (12, 23). Following
these results activation of NF-
B by TNF-
should be precluded in
ASM-deficient cells.
In the present study we used primary embryonic fibroblasts (EMFIs) from
the recently established ASM-deficient mouse line (8)
(asmase/
EMFIs) to elucidate the role of the
acid sphingomyelinase in the TNF-
-induced activation of NF-
B.
In contrast to previous reports (12, 23), we found no moderation of the
NF-B activation pathway in the response to TNF-
in cells derived
from asmase
/
mice in comparison with wild
type mice.
Neither changes in ASM or NSM activitiy nor alterations in
sphingomyelin or ceramide concentration have been determined after TNF- treatment.
EMFIs (fourth to sixth passage)
derived from wild type (C57/Bl6) and asmase/
(C57/Bl6 × 129/01a asmase
/
) e 14 embryos were grown in Dulbecco`s modified Eagle's medium (Seromed)
supplemented with 10% fetal calf serum (Life Technologies, Inc.), 2 mM glutamine, 100 units/ml penicillin, and 100 µg/ml streptomycin in a humidified incubator containing 5%
CO2.
Rabbit polyclonal anti-NF-B p65, rabbit polyclonal anti-NF-
B p50,
and rabbit polyclonal anti-I
B-
antibodies were purchased from
Santa Cruz Biotechnology Inc.
Alkaline phosphatase-labeled goat anti-rabbit IgG second antibody,
phenylmethylsulfonyl fluoride, benzamidine, pepstatin A, and DTT were
purchased from Sigma and [-32P]ATP and
[
-32P]dATP from Amersham Corp.
Recombinant human TNF- (specific activity, 6.6 × 106 units/mg) was kindly provided by BASF/Knoll,
Germany.
EMFIs were grown in serum-free medium for 4 h prior
to stimulation, and 100 ng/ml medium TNF- was added for the
indicated times. Stimulation was stopped by removing the medium and
washing the cell layer with cold PBS.
Proteins were fractionated as described previously (33) with some modifications. Briefly, cells were scraped with a rubber policeman into fresh, cold PBS and washed twice. After low-speed centrifugation (200 × g) the cell pellet was resuspended in lysis buffer (10 mM Tris/HCl, pH 8.0, 60 mM KCl, 1 mM EDTA, 0.5% Nonidet P-40, 1 mM DTT, 1 mM phenylmethylsulfonyl fluoride, and 1 mM benzamidine) and placed for 10 min on ice. Nuclei were separated from the cytoplasm by centrifugation at 2500 rpm in a microcentrifuge (Eppendorf) at 4 °C for 5 min, washed briefly with lysis buffer without Nonidet P-40, resuspended in nuclear extraction buffer (20 mM Tris/HCl, pH 8.0, 20 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, and 25% glycerol), and lysed by adding NaCl to a final concentration of 400 mM. The suspension was kept for 10 min on ice, vortexed, and centrifugated for 10 min (14.000 rpm). The supernatant contained the nuclear protein extract.
The supernatant of the first centrifugation step, which contained the cytoplasmic fraction, was centrifugated at 14,000 rpm and the resulting supernatant used for Western blot analysis.
Western Blot Analysis50 µg of protein of cytoplasmic
extract was subjected to SDS-polyacrylamide gel electrophoresis on 15%
polyacrylamide gels. Proteins were transferred onto a nitrocellulose
membrane (Schleicher & Schuell) for 3 h with 1 mA/cm2.
The membrane was blocked for 2 h with 3% BSA in Tris-buffered saline and incubated overnight at 4 °C with 300 ng of anti-IB-
polyclonal antibody/ml of Tris-buffered saline, 1% BSA. After incubation with the second antibody for 1 h at room temperature in
Tris-buffered saline, 1% BSA, the bands were stained using 5-bromo-4-chloro-3-indolyl phosphate (Sigma) and nitro blue tetrazolium (Boehringer Mannheim).
5-10 µg of nuclear
proteins (determined by the BCA assay (Pierce)) were incubated with 3 µg of poly(dI-dC) (Boehringer Mannheim), 0.4 pmol of double-stranded
32P-labeled oligonucleotide containing the NF-B binding
site of the human immunodeficiency virus long terminal repeat
(5
-ATCAGGGACTTTCGCTGGGGACTTTCCG-3
) in a total volume of 20 µl in a
buffer containing 20 mM HEPES, pH 7.9, 2 mM
MgCl2, 40 mM KCl, 0.1 mM EDTA, 0.5 mM DTT, and 10% glycerol. After 30 min at room temperature
samples were separated on native 4% polyacrylamide gels using
0.25 × TBE (0.0225 M Tris borate, 0.0005 M EDTA, pH 8.0) as running buffer. Competition studies were
carried out with two unrelated double-stranded oligonucleotides A
(5
-ATCAAGATATGAAAGAGTCTGAACATAGCACCA-3
) and B
(5
-GTCTTATTCACTGCTAGCAACCATGGGCTGAAG-3
). For supershift analysis the
NF-
B-double-stranded oligodeoxynucleotide complexes were further
exposed to a polyclonal antibody recognizing the p50 subunit of
NF-
B. Gels were dried, and radioactive signals were analyzed with a
PhosphorImager (Molecular Dynamics).
EMFIs from wild type and
asmase/
mice were grown in serum-free medium
for 4 h and stimulated for 20 min with 100 ng/ml TNF-
. After
discarding the medium, cells were fixed with a mixture of acetone/methanol (1:1) and incubated with anti-NF-
B p65 antibody (300 ng/ml in PBS, 3% BSA) for 1 h. Cells were washed three times with PBS, 0.5% Triton X-100, incubated with a Cy3-coupled anti-rabbit IgG antibody (Jackson ImmunoResearch) for 1 h and after further washings analyzed under a fluorescence microscope (Zeiss).
40 µg of total RNA prepared by the guanidine thiocyanate-phenol method (38) was separated on a 2% formaldehyde-agarose gel and blotted onto a GeneScreen Plus nylon-membrane (DuPont) following the manufacturer's protocol.
The blot was probed with a 125-bp fragment derived from the 5 end of
the I
B-
cDNA sequence and a 348-bp fragment derived from the
-actin cDNA.
For first strand cDNA synthesis RNA was transcribed in a reverse transcriptase reaction. The reaction mixture contained in a total volume of 10 µl: 0.5 mM of each deoxynucleotide (Boehringer Mannheim), 5 µM random hexanucleotide primers (Boehringer Mannheim), 10 mM DTT (Life Technologies, Inc.), 20 units of RNase inhibitor (Promega), 100 units of Superscript II (Life Technologies, Inc.), and 100 ng of total RNA in a reverse transcriptase buffer (Life Technologies, Inc.). Samples were incubated for 1 h at 37 °C.
One-half of the RT reaction mixture was used for the PCR reaction with
the primers mIB-s (5
-GCCCCGCACAGCCATGTTTCAG-3
) and mI
B-as
(5
-CATGGAGTCCAGGCCGCTGTCGTG-3
) for I
B-
. The other half was used
for the
-actin control with the oligonucleotides m
-actin-s
(5
-TGGAATCCTGTGGCATCCATGAA-3
) and m
-actin-as
(5
-TAAAACGCAGCTCAGTAACAGTC-3
).
Fragments were amplified in a thermal cycler (Perkin-Elmer) with the following program: 5 min 94 °C (1 ×); 45 s 94 °C, 60 s 60 °C, 90 s 72 °C (25 ×); 10 min 72 °C (1 ×).
PCR products were separated on a 2% TBE-agarose gel and analyzed under UV light.
Sphingomyelinase AssaysEMFIs were incubated with 100 ng/ml
TNF-, and ASM activities were measured following a modified method
described previously (39). Briefly, 4 × 106 cells
were grown in serum-free medium for 4 h and incubated with TNF-
for 2 and 4 min, respectively (12). The medium was discarded, cells
were scraped into cold PBS, washed once, resuspended in 200 µl of
0.2% Triton X-100, and placed on ice for 10 min. After centrifugation
at 800 × g for 5 min, 10 µg of protein of the
supernatant was diluted to a final volume of 90 µl with buffer (250 mM sodium acetate, pH 4.5, 1 mM EDTA). 10 µl
of substrate (1 nmol of
[N-14CH3]sphingomyelin/µl of
buffer (specific activity, 8000 dpm/nmol) (40)) was added and the
mixture incubated for 60 min at 37 °C.
For detection of NSM activity (12, 41), cells were stimulated with
TNF- for 1 and 2 min, transferred into cold PBS, washed, resuspended
in 200 µl of buffer (20 mM HEPES, pH 7.4, 2 mM EDTA, 10 mM MgCl2, 5 mM DTT, 1 mM phenylmethylsulfonyl fluoride, 10 µM pepstatin A, 750 µM ATP, 0.2% Triton
X-100), and placed for 10 min on ice. Conditions of incubation and
enzyme assay were as described for ASM except that 10 µg of protein
was resolved in 20 mM HEPES, pH 7.4, 1 mM
MgCl2, and incubation time was extended to 120 min.
[14C]Phosphorylcholine was extracted by adding 200 µl
of H2O and 800 µl of chloroform/methanol (2:1) mixture,
and the radioactivity of the aqueous phase was determined by
scintillation counting.
Fibroblasts were
metabolically labeled with [1-14C]palmitic acid (specific
activity, 0.81 × 107 dpm/mg, 20 µg/ml culture
medium) for 36 h. Before TNF- stimulation the medium was
removed, and cells were grown in serum-free medium for 4 h.
Stimulation of cells was stopped by removing the medium and immediate washing with cold PBS. Cells were scraped from the culture dishes, centrifugated, and the cell pellet suspended in chloroform/methanol (2:1) for lipid extraction. Phases were separated by centrifugation for 10 min at 4000 × g. The organic phase was dried under nitrogen, and lipids were dissolved in chloroform/methanol (2:1). Radioactive lipids were measured in a liquid scintillation counter (Wallac 1409), and aliquots were separated on TLC plates (Merck) in chloroform/methanol/water (65:25:4). Radioactivity was detected and quantified with a PhosphorImager (Molecular Dynamics) using internal standards.
We studied the role of sphingomyelin and ASM within the postulated
signal transduction chain initiated by TNF- and leading to the
NF-
B activation in EMFIs derived from wild type and
asmase
/
mice. The genotype of
asmase
/
EMFIs was confirmed by Southern blot
hybridization analysis of the EcoRI-restricted DNA using the
previously described probe (8). In addition the ASM activity of wild
type and the null mutant fibroblasts was assayed using
[N-14CH3]sphingomyelin as
substrate. The asmase
/
fibroblasts showed no
ASM activity.
Southern blot analysis and enzyme assay therefore clearly proved the complete ASM deficiency of the mutant EMFIs.
Degradation and Resynthesis of ITNF- rapidly activates NF-
B (28) and leads to
degradation of I
B-
(33, 37), which is rapidly resynthesized. We
studied first the kinetics of I
B-
degradation of wild type and
asmase
/
EMFIs after TNF-
stimulation over
the time period indicated in Fig. 1. Western blot
hybridization analysis showed the disappearance of the I
B-
signal
within 10-20 min identically in cells with both genotypes. Resynthesis
restored the I
B-
level within 60 min to the level of the
unstimulated cells. I
B-
was recognized by the polyclonal antibody
as a 36-kDa protein.
We also monitored the NF-B-induced resynthesis of I
B-
following its degradation (28, 42, 43) by measuring the increase of
I
B-
transcription in Northern blot hybridization analysis (Fig.
2a). I
B-
mRNA was hardly measured
in unstimulated wild type and asmase
/
cells.
However within 60 min after TNF-
stimulation a strong I
B-
mRNA signal appeared in wild type and in the mutant fibroblasts. Relating the I
B-
mRNA signals with the respective
constitutively expressed
-actin signal and quantification of the
signals by phosphorimaging clearly indicated that
asmase
/
and wild type EMFIs showed an
identical expression rate.
IB-
is constitutively expressed at a very low level. The low
concentration of mRNA was detected by amplification of a specific 125-bp fragment by RT-PCR. Fig. 2b shows the RT-PCR pattern
of EMFIs of the wild type and asmase
/
genotype, quiescent and after TNF-
stimulation for 30 and 60 min,
respectively. Identical patterns of the expected 125-bp PCR fragment
were observed in wild type and asmase
/
EMFIs, the intensity of the 348-bp
-actin PCR fragment in all lanes
proved the Northern blot hybridization analysis.
TNF- stimulation induces
the dissociation of the I
B-NF-
B complex liberating NF-
B for
nuclear translocation and DNA binding (28). We followed the kinetics of
NF-
B nuclear translocation and DNA binding in the electrophoretic
mobility shift assay. Double-stranded, 32P-labeled
oligodeoxynucleotides resembling the NF-
B binding site of the human
immunodeficiency virus long terminal repeat were incubated with nuclear
protein extracts of wild type and asmase
/
EMFIs stimulated with TNF-
for the time periods indicated in Fig.
3a.
Within 5 min of TNF- stimulation the strong nuclear import reached a
plateau, which was stable during prolonged TNF-
stimulation. The
intensities of NF-
B oligonucleotide complexes formed in nuclear extracts of wild type and asmase
/
EMFIs were
identical.
The specificity of the signals was proven in the control experiments
(Fig. 3b). For an in vivo proof the nuclear
import of NF-B after TNF-
stimulation in wild type and
asmase
/
EMFIs was followed by
immunofluorescence microscopy. After TNF-
stimulation cells were
permeabilized and incubated with the anti-NF-
B p65 antibody and
subsequently with a Cy3-conjugated anti-rabbit IgG second antibody.
The cellular distribution of the NF-B-
signal in untreated and
TNF-
-stimulated wild type and asmase
/
EMFIs is compared in Fig. 4. The nuclear staining of
both TNF-
-treated cells (Fig. 4, b and d) was
equal, whereas the untreated cells (Fig. 4, a and
c) showed no nuclear immunofluorescence.
The immunocytochemical data fully support the results of our
biochemical analysis. After TNF- stimulation, wild type as well as
mutant EMFIs indiscriminately translocated NF-
B within a short time
into the nucleus for DNA binding.
Previous reports indicate that
TNF- stimulation of U937 cells increases NSM activity 2-3-fold
within 90 s and ASM activity within 4 min (12). A rapid 70%
increase of ceramide concentration within 2-3 min and a decrease of
sphingomyelin concentration reaching 20% of control values of
untreated cells after 5 min has been reported for Jurkat cells
(23).
When mouse embryonic fibroblasts with the wild type and ASM deficiency
genotype were assayed after 0, 2, and 4 min of TNF- stimulation, no
significant increase of ASM activity of wild type was observed. The NSM
assay (41) at 0, 1, and 2 min after TNF-
stimulation indicated
neither a measurable NSM activity nor an increase of the enzyme
activity in wild type and asmase
/
EMFIs.
Neutral sphingomyelinase is present only at a low level in nonneuronal
tissue (8, 44).
These results are supported by the measurement of the intracellular
ceramide concentrations in wild type and
asmase/
EMFIs after TNF-
stimulation. We
prelabeled EMFIs with [1-14C]palmitate for 36 h.
Cells were thoroughly washed and stimulated with TNF-
for 1, 2, 5, and 20 min. Total lipids of EMFIs were isolated and separated by thin
layer chromatography. The intensity of the radioactive ceramide and
sphingomyelin bands were calculated and correlated with the individual
phospholipid classes using the Image Quant software of the
PhosphorImager (Fig. 5). Neither EMFIs from wild type
nor ASM-deficient mice showed significant alterations of the
sphingomyelin and ceramide concentrations after TNF-
stimulation.
NF-
Jurkat T-cells, which overexpress recombinant ASM,
constitutively show a dose-dependent NF-B-specific
transcription of a reporter gene without an external stimulus (12).
We generated a stable, murine ASM-expressing human embryonic kidney
(HEK 293) cell line with an ASM activity 20-fold above basis level.
They were stimulated with TNF- for the periods indicated in Fig.
6 and compared with wild type HEK cells.
The time course of the NF-B nuclear translocation followed by the
electrophoretic mobility shift assay was indistinguishable in wild type
and ASM-overexpressing HEK cells and showed the same kinetics as in the
EMFIs of wild type and asmase
/
mice.
The asmase/
mouse line offers a unique
opportunity for studies on the function of ASM and NSM in the recently
postulated signal transduction pathways triggered by TNF-
and
interleukin 1 (2) or other effector molecules, e.g. CD28,
Fas, or vitamin D3 (1). TNF-
is supposed to activate
both acid and neutral sphingomyelinase. Initiation of NF-
B nuclear
translocation is supposed to be transmitted exclusively by acid
(lysosomal) sphingomyelinase generating the proposed second messenger
ceramide. Diacylglycerol, which is released by a
phosphatidylcholine-specific phospholipase C, has been postulated as a
link between TNF-
receptor 55 and lysosomal sphingomyelinase (23).
However, the existence of the enzyme remains uncertain.
Taking these reports for granted, NF-B activation should be
precluded in ASM-deficient cells.
The results described here indicate that NF-B is activated in
embryonic fibroblasts from wild type and
asmase
/
mice when stimulated with TNF-
.
The I
B-
-NF-
B complex is dissociated as demonstrated by the
rapid transient proteolysis and resynthesis of I
B-
. Free
cytosolic NF-
B is translocated into the nucleus shortly after
TNF-
stimulation in EMFIs of both genotypes, which was proven by the
electrophoretic mobility shift assay with nuclear protein extracts and
by immunocytochemistry.
Our results clearly exclude acid sphingomyelinase from the signaling
pathway leading to the TNF--induced activation of NF-
B. Additionally no increase in ASM enzyme activity after short TNF-
incubation in wild type EMFIs was observed.
Similar observations were made in studies with ASM-deficient Niemann-Pick disease A human fibroblasts (45, 46).
NSM activity of wild type and asmase/
EMFIs
is apparently also not stimulated by TNF-
. The enzyme activity in
extraneural cells so far analyzed is very low, and high activities were
only measured in the central nervous system (8). Embryonic fibroblasts seem to be almost devoid of NSM activity, as shown by the enzyme assay,
which is in agreement with a previous report (47). If NSM of EMFIs is
participating in the ceramide-activated pathway, then the activity is
below the detection level of the assay used. Further studies will
address the analysis of the function of NSM.
It should be mentioned that most data for the function of ASM and NSM in signal transduction have been derived mainly from myeloid-lymphoid cell lines.
Recent reports showed that NF-B activation protects mouse embryonic
fibroblasts from induction of apoptosis, suggesting a dissection of the
TNF-
-induced signaling pathways (48-51). In our analysis we showed
that neither acid nor neutral sphingomyelinase were activated, and no
ceramide was generated in the TNF-
-induced activation of NF-
B.
Therefore we conclude that sphingomyelinases are not of general
importance in TNF-
responses at all.
We thank D. Newrzella for providing the HEK cell line.