![]()
|
|
||||||||
(Received for publication, June 14, 1995) From the
Cellular responses initiated by tumor necrosis factor (TNF) are
mediated by two different cell surface receptors with respective
molecular masses of 55 kDa (p55) and 75 kDa (p75). p55 is functional in
almost every cell type and can independently transmit most biological
activities of TNF. In contrast, TNF signaling via p75 seems so far
largely restricted to cells of lymphoid origin, where it can induce
proliferation, cytokine production, and/or apoptosis. The mechanisms
that regulate TNF receptor activity are largely unknown. Here we report
that the p75 of unstimulated p75-responsive PC60 T cells is
phosphorylated on serine by a kinase activity present in p75 immune
complexes. Several lines of evidence indicate that the latter kinase is
casein kinase-1 (CK-1). Previous results have shown that the p75 TNF
receptor is constitutively phosphorylated in vivo. Our data
show that the latter in vivo phosphorylation is also at least
partially due to CK-1. Pretreatment of cells with TNF had no detectable
effect on p75 phosphorylation in vitro or in vivo.
However, a specific CK-1 inhibitor potentiated TNF-induced apoptosis
mediated by p75, suggesting an inhibitory role for phosphorylation by
CK-1. Although in vivo p75 phosphorylation could be seen in
both p75-unresponsive and p75-responsive cell lines, in vitro p75 phosphorylation in p75 coimmunoprecipitates could not be
observed in cell lines that were biologically unresponsive to p75
stimulation. The latter observation further indicates a regulatory role
for p75 phosphorylation in p75-mediated signaling. Taken together, our
data demonstrate that the p75 TNF receptor is phosphorylated and
associated with CK-1, which negatively regulates p75-mediated TNF
signaling.
Tumor necrosis factor (TNF) ( The cytoplasmic domains of p55
and p75 lack sequence homology, suggesting that they generate distinct
activation signals. To date, it is not clear how TNF receptor
aggregation leads to signal transduction. Oligomerization of
cytoplasmic domains might unveil discrete peptide motifs, activate
receptor-associated enzymes, or create a composite, tertiary binding
site to attract signal transduction molecules. The fact that neither
TNF receptor contains intrinsic protein kinase activity further
suggests that associated proteins, rather than the intracellular domain
of either receptor, act as essential elements in signal transduction.
Recently some evidence has been presented for the presence of proteins
that are specifically interacting with p55 or p75 (Darnay et
al., 1994a, 1994b; Rothe et al., 1994; VanArsdale and
Ware, 1994; Song et al., 1995). Some of these proteins were
shown to phosphorylate one or both TNF receptors. The biological
significance as well as the identity of the latter kinase(s) remain
unknown. Interestingly, it was demonstrated that the unspecific protein
kinase inhibitor staurosporine (STS) could inhibit p55 phosphorylation
in U937 cells (VanArsdale and Ware, 1994). The finding that STS can
sensitize tumor cells to TNF-induced cell death (Beyaert et
al., 1993) suggests that phosphorylation of p55 might negatively
regulate its activity. In contrast, the functional role of p75
phosphorylation as well as the identity of p75-associated kinase(s)
still must be investigated. In an attempt to gain more insight into
this matter, we analyzed p75 immune complexes from a rodent hybridoma
PC60 T cell line, engineered to overexpress the human p55 and p75 (in
this paper referred to as PC60p55p75). We show that p75 becomes
phosphorylated in vitro and in vivo by a kinase
activity present in p75 immune complexes. We identified this kinase as
CK-1, and we present evidence that phosphorylation by CK-1 negatively
regulates p75-mediated signaling to TNF-induced apoptosis.
Phosphorylation of the
CK-1-specific peptide (1 mg/ml) was analyzed as described above, except
that phosphorylation reactions were stopped by spotting an aliquot of
the supernatant of the incubation mixture onto square papers of Whatman
P-81 and subsequent immersion and washing in 75 mM phosphoric
acid (Agostinis et al., 1987). Controls with either substrate
or immunoprecipitates alone were performed at the same time and
subtracted from the peptide phosphorylation data.
Figure 1:
In vitro phosphorylation of a
75-kDa protein in utr1 immunoprecipitates of PC60 cells. A,
immunoprecipitates (protein G-Sepharose 4 Fast Flow was used as
carrier) of control PC60 transfectants (PC60 control) or PC60 cells
expressing either full-length (PC60p55p75) or intracellularly truncated
(PC60p55p75
As
immunoprecipitates of control transfectants did not contain the 75-kDa
phosphoprotein, it is very likely that the latter is the p75 TNF
receptor itself. This conclusion was substantiated by the following
observations. First, p75 immunoprecipitated from PC60p55p75 cells
metabolically labeled with
Figure 2:
The 75-kDa phosphoprotein generated in
vitro in p75 immunoprecipitates comigrates with p75 labeled in
vivo with
To
investigate the effect of TNF stimulation in vivo on p75
kinase activity in vitro, PC60p55p75 cells expressing
full-length human p75 were treated with human TNF, after which p75
immune complexes were prepared and assayed for p75 kinase activity.
Immunoprecipitates were made with the utr4 monoclonal antibody, which
binds human p75, even when the latter is occupied by TNF (as assessed
by flow fluorimetric analysis) (data not shown; see also Brockhaus et al.(1990)). TNF doses, varying between 0.1 and 500 ng/ml,
were given for different time periods (30 s up to 4 h). In some
experiments, cells starved by a 24-h incubation in the absence of serum
were used. Under none of these conditions a TNF-stimulated enhancement
of p75 kinase activity could be demonstrated, as judged from the
absence of an increase of
Figure 3:
Effect of protein kinase inhibitors on p75
kinase activity in vitro. p75 immunoprecipitates (Trisacryl
GF-2000 protein A was used as carrier) of PC60p55p75 cells were
pretreated for 10 min with or without the indicated drugs, after which
an in vitro kinase assay was performed. Immunoprecipitates
were then resolved on SDS-polyacrylamide gels as described under
``Materials and Methods.'' The position of p75 is indicated
by an arrowhead. A, effect of STS; B, effect
of CKI-7 and heparin.
Figure 4:
Effect of CKI-7 on p75 kinase activity in vivo. 2
Figure 5:
Absence of in vitro p75
phosphorylation in p75 immunoprecipitates of U937 and HL60 cell lines. A, immunoprecipitates of the indicated cell lines were made
without (lanes1, 3, and 5) or with (lanes2, 4, and 6) utr1, subjected
to in vitro phosphorylation in the presence of labeled ATP and
analyzed as described under ``Materials and Methods.''
Trisacryl GF-2000 protein A was used for immunoprecipitation. The arrowhead points to the position of p75 from PC60p55p75 cells. B, immunoblotting of p75 from PC60p55p75 and U937 cell lines.
Lysates of PC60p55p75 (lane1) or U937 (lane2) were immunoprecipitated with utr1 and immunoblotted as
described under ``Materials and Methods.'' C,
lysates of U937 (lane1) or PC60p55p75 (lane2) labeled with
To date, receptor-associated mechanisms involved in TNF
signal transduction, as well as the mechanisms determining whether the
signal is initiated by p55 or p75, are largely unknown. Covalent
modification of proteins by phosphorylation-dephosphorylation plays a
central role in cellular regulation, including signal transduction. The
cytoplasmic domains of p55 and p75 TNF receptors, like that of other
members of the cytokine receptor superfamily, lack any protein kinase
motifs. However, there is increasing evidence that signal transduction
by TNF receptors involves tyrosine and serine/threonine phosphorylation
of proteins. The p75 receptor has been shown to be constitutively
phosphorylated on serine residues in vivo (Pennica et
al., 1992). Recent results have provided evidence for physical
association of a protein kinase with the cytoplasmic domain of p75
(Darnay et al., 1994a). However, the identity of the kinase as
well as the TNF response in which it is involved was not stated.
Therefore, we have performed biochemical analyses of p75
imunoprecipitates in vitro and in vivo, with the aim
to characterize the p75 kinase and its biological role. In this study,
we made use of rodent PC60 T cells transfected with human p55 and p75
receptors. Specific triggering of the transfected p75 receptor (either
by species-specific antisera or receptor-specific mutants of human TNF)
has been shown to induce granulocyte/macrophage colony-stimulating
factor production in these cells, especially in the presence of
interleukin-1 (Vandenabeele et al., 1992). Recently, we found
that selective p75 triggering was also capable of inducing apoptosis in
PC60 cells, but, surprisingly, only in cells that were also transfected
with human p55 (Vandenabeele et al., 1995). Here we report
that p75 receptors from these p75-responsive PC60 cells became
phosphorylated in vitro by a kinase activity that was present
in p75 immunoprecipitates. Serine was the only detectable phosphoamino
acid identified in p75. At present, we have no indication that p75
kinase activity is affected by TNF treatment of cells. This finding is
reminiscent of the absence of a detectable effect of TNF on in vivo p75 phosphorylation (Pennica et al., 1992). Several lines of evidence point to
CK-1 as the p75 kinase identified in the present study. First, CKI-7,
an inhibitor with relative specificity for CK-1 (Chijiwa et
al., 1989), completely inhibited p75 kinase activity that could be
coimmunoprecipitated with p75. Second, casein and a CK-1-specific
peptide substrate are both phosphorylated by the p75 kinase in
vitro. Third, a partially purified CK-1 phosphorylates p75 in
vitro. The p75 kinase could be further distinguished from CK-2 by
its inability to use GTP as phosphate donor (Hathaway and Traugh, 1979)
and its insensitivity to heparin (Hathaway et al., 1980). In
addition, the involvement of a whole range of other kinases could be
excluded by the absence of an effect of the broad spectrum protein
kinase inhibitor STS. Similar to the in vitro phosphorylation,
CKI-7 also inhibited p75 phosphorylation in vivo. Although we
cannot exclude that CKI-7 only reaches suboptimal concentrations in
cultured cells, the residual in vivo phosphorylation of p75
that is still observed in cells treated with CKI-7 might be due to
other kinases for which multiple potential phosphorylation sites can be
found in the intracellular domain of p75. CK-1 prefers substrate sites
with acidic residues N-terminal to the modified residue, of which eight
can be found in the intracellular domain of human p75. In addition, the
phosphorylated sequence -S(P)-X-X-S- (where
S(P) indicates a phosphorylated serine) is also a substrate for CK-1,
suggesting CK-1 can act in a process termed hierarchical protein
phosphorylation (Kennelly and Krebs, 1991). If one takes the latter in
consideration, a total of 22 serine residues are potential
phosphorylation sites for CK-1. Recently it was found that a C-terminal
region of 78 amino acids within the cytoplasmic domain of p75,
mediating the interaction with TRAF proteins, was indispensable for
signal transduction toward cell proliferation and NF- Recently, a
p75-associated kinase activity that phosphorylated p75 in U937 cells
has been described (Darnay et al., 1994a). In these studies
p75 was phosphorylated on serine and threonine residues, and a 59-kDa
p75-binding protein also served as a substrate. These properties
distinguish it from the p75 kinase activity described in the present
study. It has also been reported that cell extracts of TNF- or
IL-1-treated cells contain an activated kinase that specifically
phosphorylates Several cDNA clones for CK-1 s have been
isolated and sequenced (reviewed by Issinger(1993)). They have been
found in the cytosol, associated with membranes, and in the nucleus. A
large number of proteins become phosphorylated by CK-1. Some of these
substrates undergo defined functional changes when phosphorylated by
CK-1. For example, the regulatory M-subunit (inhibitor-2) of protein
phosphatase-1 appears to be phosphorylated by CK-1 and this inhibits
the phosphatase activity (Agostinis et al., 1987). CK-1 has
also been shown to phosphorylate the insulin receptor and the
progesterone receptor (Rapuano and Rosen, 1991; Chauchereau et
al., 1992). The latter could also be copurified with CK-1.
Interestingly, the p75 kinase as well as CK-1 could also phosphorylate
the cytoplasmic domain of the p55 receptor, No previous studies have linked CK-1 or p75 phosphorylation with
biological activities of TNF. Our observation that inhibition of CK-1
increases p75-mediated apoptosis indicates that CK-1 negatively
regulates TNF-induced signaling. The level at which this regulation
takes place is still unclear. Based on our observations of a
correlation between p75 phosphorylation in vitro and p75
signaling capacity in vivo (in vitro phosphorylation
and signaling in PC60p55p75 cells, not in U397 or HL60), it is tempting
to speculate that the phosphorylation level of p75 regulates its
signal-transducing capacity. p75 is constitutively phosphorylated in vivo (Pennica et al., 1992; Fig. 3A) and the turnover of this phosphorylation seems
to be high. For example, in PC60p55p75 cells, Taken together,
our data demonstrate that p75 is phosphorylated by a coprecipitating
kinase, which has been identified as CK-1. In addition, we have
obtained evidence that CK-1 negatively regulates TNF-induced signaling
to apoptosis. These findings provide a framework for future studies.
Ongoing studies with PC60 cells expressing intracellular p75 deletion
mutants of varying length have already indicated that phosphorylation
takes place in a C-terminal region of 78 amino acids within the
cytoplasmic domain of p75 that has been shown to be required for signal
transduction and association of TRAF proteins (Rothe et al.,
1994).
Volume 270,
Number 40,
Issue of October 06, pp. 23293-23299, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)is a pluripotent
cytokine inducing an extraordinary wide variety of biological
responses, mainly related to immune regulation and inflammation
(reviewed by Vassalli(1993) and Fiers(1995)). Several signal
transducing mechanisms have been shown to be involved in the action of
TNF, dependent on the cell type and the cell response studied (reviewed
by Beyaert and Fiers(1994)). Cellular signals occurring within minutes
after addition of TNF include generation of cAMP in some fibroblast
lines (Zhang et al., 1988), activation of
phosphatidylcholine-specific phospholipase C and sphingomyelin
degradation in myeloid cell lines (Kim et al., 1991;
Schütze et al., 1992; Dressler et
al., 1992), activation of the transcription factor NF-
B
(Lowenthal et al., 1989), and phosphorylation of several
cellular proteins, including heat shock protein 27, epidermal growth
factor receptor, and eukaryotic initiation factor 4E (Hepburn et
al., 1988; Saklatvala et al., 1991; Marino et
al., 1991). Enhanced phosphorylation of cellular proteins could be
due to the documented activation of a wide variety of kinases (Zhang et al., 1988; Guy et al., 1991; Kim et al.,
1991; Mathias et al., 1991; Van Lint et al., 1992;
Guesdon et al., 1993) and/or to inhibition of phosphatases
(Guy et al., 1993; Tan, 1993). TNF-induced signaling is
initiated by oligomerization of TNF receptors upon binding of the
trimeric TNF ligand (Engelmann et al., 1990; Van Ostade et
al., 1991). Two distinct, high affinity TNF receptors of 55 kDa
(p55) and 75 kDa (p75) have been identified and cloned (reviewed by
Loetscher et al.(1991) and Tartaglia and Goeddel(1992)).
Although both TNF receptor types are expressed by most cell types, p55
is responsible for the majority of TNF activities (reviewed by
Tartaglia and Goeddel(1992)). The contribution of p75 can be explained
in part by the so-called ligand passing model, in which p75 presents
TNF to neighboring p55 molecules, the latter being signal transducing
(Tartaglia and Goeddel, 1992). Direct p75 signaling involves mainly
effects on lymphoid cells, such as proliferation of the CT6 T cell line
and induction of granulocyte/macrophage colony-stimulating factor
secretion by a T cell hybridoma (Gehr et al., 1992;
Vandenabeele et al., 1992; Tartaglia et al., 1993).
In addition, an independent signaling role for p75 has been
demonstrated in TNF-mediated cytotoxicity in some specific cell lines
(Heller et al., 1992; Grell et al., 1993;
Vandenabeele et al., 1995).
Reagents
All reagents were from Sigma unless
otherwise stated. STS was from Boehringer (Mannheim, Federal Republic
of Germany). CKI-7 was from Seikagaku Corp. (Tokyo, Japan). The
synthetic peptide DDDEESITRR, referred to as CK-1-specific peptide in
this paper, was obtained from Cambridge Research Biochemicals
(Cambridge, United Kingdom). CK-1 was purified from pig spleen
essentially as described (Agostinis et al., 1989); 1 unit of
CK-1 incorporates 1 nmol of phosphate/min in casein (2 mg/ml) at 37
°C. Recombinant human TNF and the muteins R32WS86T and D143F,
specific for human p55 and human p75, respectively, have been described
(Van Ostade et al., 1992, 1994). Monoclonal antibodies to
human p55 and p75 were from the htr and utr series, all of which are
directed to the extracellular part of the p55 or p75 TNF receptor,
respectively (Brockhaus et al., 1990; Espevik et al.,
1990).Cell Lines
The rat/mouse T cell hybridoma PC60
(Conzelmann et al., 1982) and derivatives (Vandenabeele et
al., 1992 and 1995), and the human U937 histiocytoma and HL60
promyelocytic leukemia cell lines were cultured in RPMI 1640 medium
(Life Technologies, Inc./BioCult, Paisley, United Kingdom) supplemented
with 10% (v/v) fetal bovine serum, 1 mM sodium pyruvate, 50
µM 2-mercaptoethanol, and antibiotics. PC60p55p75 cells
are derived from transfection of rodent PC60 cells with the human p55
and p75 genes as reported and have been described elsewhere
(Vandenabeele et al., 1992 and 1995). Human p55 expression in
these cells was found to be lower than 50 molecules/cell. Human p75 was
expressed at 5000 receptors/cell, numbers that are in the same range as
endogenous p75 levels found in human U937 and HL60 cells. We reported
previously that PC60p55p75 cells coexpressing the two types of human
TNF receptors undergo TNF-mediated apoptosis, in contrast to cells
expressing only one kind of TNF receptor (Vandenabeele et al.,
1995). We further showed that, although the presence of both receptor
types was required, triggering of only one type was sufficient to
induce intermediate levels of apoptosis. Cells expressing a truncated
p75 lacking its intracellular domain were obtained after transfection
with a p75 expression vector in which the codons 293 and 294 (encoding
Cys and Leu, respectively) of the human p75 were mutated to stop codons
(TGA and TAG, respectively) according to Morinaga et al. (1984).Cell Stimulation, Cell Lysis, and Immunoprecipitation of
p75
Immunoprecipitation of p75 from cells labeled with P
in vivo was performed essentially
as described (Beyaert et al., 1990). Isolation of p75 immune
complexes for in vitro phosphorylation reactions was as
follows. After harvesting of the cells at room temperature, conditioned
medium was collected for subsequent cellular stimulations. Prewarmed
(37 °C) conditioned medium (8 ml) with or without human TNF was
then added to a 2-ml suspension of 10
cells in 200-ml
tubes. After various incubation periods at 37 °C, an excess of
ice-cold phosphate buffer containing 50 µM Na
VO
was added, followed by pelleting of
the cells at 350 g for 10 min in a cooled (0 °C)
centrifuge. The number of cells used per immunoprecipitate varied from
2
10
for PC60p55p75 up to 10
in case of
U937 and HL60 cell lines. After one additional wash in ice-cold
phosphate buffer, the cell pellet was resuspended at 2-5
10
cells/ml in lysis buffer and rotated for 15 min at 4
°C. The lysis buffer consisted of 1% (w/v) digitonin, 10 mM triethanolamine-HCl, pH 7.8, 139 mM NaCl, 10 mM NaF, 1 mM Na
VO
, 400 nM microcystin-LR, 180 µg/ml phenylmethylsulfonyl fluoride, 0.27
trypsin inhibitory units/ml aprotinin, and 10 µg/ml leupeptin.
Following 10 min of centrifugation at 31,000 g, the
supernatant was rotated for 2 h at 4 °C in the presence of
antibodies directed against human p75 (5 µg/10
cells).
The lysates were then mixed with 100 µl of a 50% (v/v) slurry of
either protein G-Sepharose 4 Fast Flow (Pharmacia LKB Biotechnology,
Uppsala, Sweden), TSK-protein A (Affiland, Liège,
Belgium), or Trisacryl GF-2000 protein A (Pierce). After a 90-min
rotation at 4 °C, immune complexes were spun down and washed three
times in 4 ml of lysis buffer.In Vitro Kinase Assay
Washed immunoprecipitates
were resuspended in 60 µl of 2 phosphorylation buffer (40
mM Tris-HCl, pH 7.4, 20 mM MgCl
, 200
µM Na
VO
, 40 mM
-glycerophosphate, 200 µM ATP, and 80 µCi/ml (27
nM) of [-
P]ATP (3000 Ci/mmol;
Amersham International, Amersham, United Kingdom)). Reaction mixtures
(in some cases containing 1 mg/ml whole casein) were incubated for 10
min at 30 °C and stopped by the addition of Laemmli sample buffer
containing 2-mercaptoethanol (Laemmli, 1970). After heating, the
mixtures were separated on a 10% SDS-polyacrylamide gel. For inhibition
studies with protein kinase inhibitors, immunoprecipitates were
pretreated for 10 min at 4 °C in phosphorylation buffer without
cold ATP and containing 100 µg/ml ovalbumin as a stabilizing
protein. Inhibitors remained present during the subsequent
phosphorylation reaction, which was also performed in the absence of
cold ATP. Quantification of phosphorylated bands was done by
phosphorimaging analysis or by excising and counting the corresponding
bands in a scintillation counter.
Re-immunoprecipitation of Phosphoproteins Generated by in
Vitro Phosphorylation
In some experiments, p75 was
immunoprecipitated using utr4 antibody covalently coupled to Hydrazide
AvidGel (BioProbe International, Tustin, CA). After washing and
phosphorylation as described above, the phosphorylation reaction was
now stopped by addition of 1 ml of lysis buffer supplemented with 1%
SDS. After a 30-min rotation at room temperature, the supernatant was
harvested and diluted with 12 ml of lysis buffer supplemented with 1.5%
(w/v) Nonidet P-40 and 1 ml of lysate of untransfected PC60 cells,
prepared at 5 10
cells/ml. Following a 45-min
rotation at 4 °C, this supernatant mixture was again subjected to
immunoprecipitation now using protein G-Sepharose and utr1 monoclonal
antibodies.Immunoblotting of p75
Following
immunoprecipitation of p75, the washed precipitate was taken up in
Laemmli sample buffer without 2-mercaptoethanol and loaded on a 10%
SDS-polyacrylamide gel. Electrophoresed proteins were electroblotted to
nitrocellulose membranes in 40 mM glycine, 50 mM Tris, 20% (v/v) methanol. After a 30-min incubation in blocking
buffer (1% (w/v) bovine serum albumin, 5 mM EDTA, 140 mM NaCl, 50 mM Tris-HCl, pH 7.5, 0.02% NaN
), the
membranes were incubated for 4 h at room temperature in blocking buffer
containing 1 µg/ml biotinylated utr1. Following three 10-min washes
in wash buffer (10 mM Tris-HCl, pH 8.0, 150 mM NaCl,
0.05% (w/v) Tween 20, 0.02% NaN
), the membranes were
incubated for 90 min with 0.5 µCi of I-streptavidin
(20-40 mCi/mg; Amersham International)/ml of wash buffer, washed
and autoradiographed. For competition with TNF, recombinant human TNF
was added at 50-100 µg/ml during blocking and development
steps.
Phosphoamino Acid Analysis
The P-labeled p75 bands in SDS-polyacrylamide gels were
localized by exposure of the gels to film, excised from the gel, and
analyzed for phosphoamino acid content by two-dimensional
electrophoresis as described (Hunter and Sefton, 1980).
Partial Proteolytic Peptide Mapping
Peptide
mapping was performed by limited proteolysis in SDS using Staphylococcus V8 protease, followed by analysis via gel
electrophoresis as described (Van Roy et al., 1981).Apoptosis Assay
Cells were treated with ligands
for 24 h, after which apoptosis was assayed by adding 30 µM propidium iodide to the cells. Propidium iodide exclusion
(PC60p55p75 cells) or hypoploidy in propidium iodide fluorescence
histograms was measured by quantitative flow cytometry on an EPICS,
Luton, United Kingdom) as described previously (Vandenabeele et
al., 1995). Similar results, but requiring longer incubation
times, were obtained when cytotoxicity was assayed colorimetrically by
staining with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide as described (Tada et al., 1986).
Phosphorylation of the p75 TNF Receptor by a Kinase
Present in p75 Immunoprecipitates
Using protein G-Sepharose and
utr1, a species-specific antibody directed to the extracellular domain
of human p75, immunoprecipitates were prepared from unstimulated rodent
PC60 cells engineered to express human p55 and p75 (referred to as
PC60p55p75). Following short term incubation of these precipitates with
[-
P]ATP, a major phosphorylated protein
with a M
of approximately 75,000 was revealed on
SDS-polyacrylamide gels (Fig. 1A). This 75-kDa
phosphoprotein was absent in immunoprecipitates of control PC60
transfectants not expressing human p75, or expressing a human p75
lacking its cytoplasmic domain (p75
CD). Expression levels of
wild-type and truncated p75 were similar in both transfectants, as
judged by immunoblotting and flow cytometry (data not shown). The
75-kDa phosphoprotein could also be generated in immune complexes using
the monoclonal antibodies utr2 and utr4, all of which are reactive with
the extracellular part of human p75 (data not shown). Phosphoamino acid
analysis of the 75-kDa phosphoprotein demonstrated that phosphorylation
was exclusively on serine (Fig. 1B). Phosphate
distribution found in p75 immunoprecipitated from cells that have been
metabolically labeled with P
is also mainly on
serine, with a very low level of incorporation (3%) in threonine
residues (Pennica et al., 1992). (
)
CD) human p75 were made with (+) or
without(-) utr1 and subjected to in vitro phosphorylation as described under ``Materials and
Methods.'' Reaction mixtures were then analyzed on a 10%
SDS-polyacrylamide gel followed by autoradiography. The arrowhead points to the position of the 75-kDa phosphoprotein. B,
phosphoamino acid analysis of the in vitro phosphorylated p75
protein band shown in A. Radioactivity on the thin layer
plates was detected by phosphorimaging. The migration positions of
phosphoserine, phosphothreonine, and phosphotyrosine are
indicated.
P
in vivo comigrated and showed a largely overlapping proteolytic
fingerprint with the 75-kDa protein phosphorylated in vitro (Fig. 2A and data not shown). In addition, the
75-kDa phosphoprotein generated in vitro was found to be
reactive with antibodies to p75. This was demonstrated as follows.
Immunoprecipitates were made with antibodies to p75 covalently coupled
to an agarose matrix. Following its in vitro phosphorylation
and SDS-driven, mild release from the immobilized antibodies in the
immunoprecipitate, the 75-kDa protein could be re-immunoprecipitated
with p75 TNF receptor-specific antibodies (Fig. 2B).
From these data we conclude that the 75-kDa phosphoprotein generated in
p75 immune complexes is the p75 TNF receptor itself, and the associated
kinase activity will be further referred to as p75 kinase.
P
and is reactive with
antibodies to p75. A, immunoprecipitates of PC60p55p75 cells
made without(-) or with (+) utr1 were subjected to in
vitro phosphorylation, followed by analysis on a 10%
SDS-polyacrylamide gel and autoradiography. A similar analysis was
performed on lysates of PC60p55p75 cells in vivo labeled with P
and immunoprecipitated with (+) or
without(-) utr1. The arrowhead points to the position of
the 75-kDa phosphoprotein. B, immunoprecipitates of PC60p55p75
cells were made with AvidGel, which was either untreated (lane1) or coupled to utr4 (lanes2-4)
and subjected to in vitro phosphorylation. In lanes1 and 2, the phosphorylation reaction was
stopped by adding Laemmli loading buffer and directly analyzed on an
SDS-polyacrylamide gel. In lanes3 and 4,
the phosphorylation reaction was stopped by the addition of
SDS-containing lysis buffer, and after a 30-min rotation at room
temperature, the supernatant was diluted in Nonidet P-40 and
protein-containing lysis buffer as detailed under ``Materials and
Methods.'' Following a 45-min rotation at room temperature, this
supernatant mixture was again subjected to immunoprecipitation now
using protein G-Sepharose without (lane3) or with (lane4) utr1. The arrowhead points to the
position of p75.
P incorporated in p75 on
SDS-polyacrylamide gels (analyzed by phosphorimaging; data not shown).
This finding is reminiscent of the absence of an effect of TNF on in vivo phosphorylation of p75 (Pennica et al.,
1992).
![]()
Characterization of p75 Kinase Activity
To
characterize the p75 kinase, we examined the effect of various drugs on in vitro p75 phosphorylation. We first tested STS, a very
broad spectrum protein kinase inhibitor blocking, among others, most
members of the protein kinase C family, cAMP-dependent kinase,
cGMP-dependent kinase, Ca/calmodulin-dependent
protein kinase, and myosin light chain kinase (Beyaert et al.,
1993). Under none of the conditions tested was STS found to affect p75
kinase activity (Fig. 3A), which means that a
contribution of the above-mentioned kinases in the in vitro p75 phosphorylation is unlikely. The observation that, in contrast
to p75 phosphorylation, STS did inhibit the phosphorylation of an
unknown, faster migrating doublet, which was unspecifically bound by
the Trisacryl GF-2000 protein A matrix, shows that STS is active under
the conditions tested. In addition, the protein kinase C inhibitors H-7
and calphostin C, as well as the cAMP-dependent kinase inhibitor H-8
had no effect on p75 phosphorylation (data not shown). Although STS
inhibits a broad spectrum of kinases, it has no effect on casein
kinases. We therefore tested the contribution of the latter by testing
p75 kinase activity in the presence of CKI-7 and heparin, which inhibit
CK-1 and CK-2, respectively (Hathaway et al., 1980; Chijiwa et al., 1989). CKI-7 clearly inhibited p75 phosphorylation,
while heparin had no effect (Fig. 3B). The inhibition
of p75 phosphorylation was quantitated by phosphorimager analysis and
found to be 100%, 80% and 45%, respectively at a CKI-7 concentration of
500 µM, 100 µM and 20 µM.
Similarly, in vivo phosphorylation of p75 from
P
-labeled PC60p55p75 cells was also diminished
by CKI-7 pretreatment (75%, 45% and 24% inhibition, at CKI-7
concentrations of 500, 100, and 20 µM, respectively; Fig. 4). In contrast, STS pretreatment had no effect on in
vivo p75 phosphorylation (data not shown). The above observations,
together with our finding that the p75 kinase can not use GTP as
phosphate donor (further excluding CK-2; data not shown), indicate that
p75 is phosphorylated by CK-1, or a CK-1-like enzyme, which is also
present in p75 immunoprecipitates. The observation that p75 is a good
substrate for a partially purified CK-1 also points in the same
direction (data not shown). To further demonstrate that the p75 kinase
has CK-1 like properties, we tested whether whole casein and the
CK-1-specific peptide DDDEESITRR (Agostinis et al., 1989), are
substrates for the p75 kinase identified in p75 immunoprecipitates of
PC60p55p75 cells. A partially purified preparation of CK-1 that clearly
phosphorylated casein and the specific peptide substrate was taken as a
positive control (Table 1). As expected, both substrates were
also phosphorylated by the p75 kinase. However, casein, but not the
CK-1-specific peptide, was also highly phosphorylated when incubated
with control precipitates that were prepared without an antibody
recognizing p75. The latter observation suggests that in addition to
CK-1 which is specifically retained by p75, p75 immunoprecipitate still
contains some aspecifically sticking kinase(s) that also
phosphorylate(s) casein, thereby masking the effect of the p75 kinase
on the latter substrate. Attempts to identify and eliminate this
aspecifically binding kinase have so far been unsuccessful.
10
PC60p55p75 cells were
metabolically labeled with 100 µCi of P
for 2.5 h. Treatment with CKI-7 started 30 min before the
labeling. At the end of the incubation period, cells were lysed and p75
was immunoprecipitated with utr1 as described under ``Materials
and Methods.'' Immunoprecipitates were then resolved on an 8%
SDS-polyacrylamide gel. The position of p75 is indicated by an arrowhead.
Inhibition of CK-1 Increases p75-mediated
Apoptosis
Selective triggering of either the human p55 or p75
with receptor-specific muteins of human TNF has been shown to induce
apoptosis in PC60 cells expressing human p55 and p75 (Vandenabeele et al., 1995). To investigate a possible correlation between
phosphorylation by CK-1 and TNF responsiveness, we analyzed the effect
of the specific CK-1 inhibitor CKI-7 on p55-mediated and p75-mediated
apoptosis in PC60p55p75. Triggering was either with human TNF, which
binds both TNF receptors, or by R32WS86T and D143F, which are specific
for p55 and p75, respectively. TNF-induced apoptosis was maximal when
both receptors were stimulated (Table 2). Interestingly, CKI-7
significantly potentiated both p55-mediated and p75-mediated apoptosis,
suggesting a negative regulation of the TNF response by CK-1. The above
increase in apoptosis could not be explained by enhanced TNF binding,
since the latter was not affected by CKI-7 treatment of the cells (data
not shown).
p75 Phosphorylation in Vitro Correlates with Cellular p75
Responsiveness
To further analyze the role of CK-1-mediated
phosphorylation in p75-mediated signaling, we examined p75 kinase
activity in p75 immunoprecipitates prepared from human U937 and HL60
myeloid cell lines which, in contrast to PC60p55p75, are not responsive
to selective p75 triggering (analyzed for apoptosis, cytokine
production, NF-
B activation, and hsp27 phosphorylation; data not
shown). As shown in Fig. 5A (lanes2 and 4), no in vitro phosphorylated 75-kDa
protein was observed in p75 immunoprecipitates of U937 or HL60 cells,
even when 20 times more cells were used (data not shown). The absence
of in vitro phosphorylation of p75 was not due to differences
in molecular forms of p75 in PC60p55p75 versus U937 or HL60
cells as similar p75 bands were revealed on immunoblots (Fig. 5B; data for HL60 are not shown). Moreover, p75
phosphorylation in PC60p55p75 immune complexes was found to be
unaltered after mixing with U937-derived p75 immunoprecipitates,
excluding the presence in the latter of a p75-directed protease or
phosphatase activity (data not shown). Also in the latter mix, p75
immunoprecipitates from U937 cells were not phosphorylated by the p75
kinase present in p75 immunoprecipitates from PC60p55p75 cells. In
contrast to in vitro p75 phosphorylation, there was no obvious
difference in phosphorylation of p75 immunoprecipitated from PC60p55p75
and U937 cells labeled with P
in vivo (Fig. 5C). The above results might be explained by
assuming that, in contrast to p75-unresponsive U937 and HL60 cells,
p75-responsive PC60p55p75 cells regulate p75 phosphorylation to a level
at which some of the 43 possible phosphorylation sites are not
phosphorylated, as reflected by its ability to be further
phosphorylated in vitro, and resulting in a p75 that is able
to transduce signals.
P
in vivo were immunoprecipitated with utr1 and analyzed by
SDS-polyacrylamide gel electrophoresis and
autoradiography.
However, the fact that the intracellular domain of p75 contains a
very high number of potential phosphoamino acids (32 serines and 11
threonines, but no tyrosine residues, out of 174 amino acids), renders
this a very complex substrate to study possible minor TNF-induced
alterations in phosphorylation.
B activation
in a T cell line (Rothe et al., 1994). The latter stretch of
amino acids contains 13 serine residues as potential acceptor sites for
CK-1-mediated phosphorylation. Preliminary results suggest that p75 is
mainly phosphorylated in this region. (
)
-casein on serine and threonine residues, with
little activity on
-casein (Guesdon et al., 1993). The
latter observation, together with the sensitivity of
-casein
kinase to STS, makes it unlikely that the latter kinase is involved in
p75 phosphorylation.
providing a
mechanism for functional cross-talk between p55 and p75 TNF receptors. P
was found to be incorporated in p75 even after short (e.g. 15 min) in vivo labeling with P
.
In addition, mild permeabilization of these cells (with 0.01%
digitonin) and incubation in the presence of
[-
P]ATP resulted even after 5 s in
P labeling of p75. (
)Possibly, the p75 of
p75-responsive cells is associated with (a) constitutively active
phosphatase(s), the action of which results in the formation of an
effective substrate for in vitro phosphorylation. A role for
phosphatase activity in TNF signaling has already been suggested by the
observation that phosphatase inhibitors can inhibit TNF cytotoxicity in
certain cell lines (Totpal et al., 1992).
Mutational analysis of the many intracellular serine
residues in this region may provide more insight regarding the
importance of p75 kinase activity in p75-mediated TNF signaling.
)
)
)
)
We thank Drs. M. Brockhaus and W. Buurman for
providing antibodies to human TNF receptors. The cDNA genes coding for
human p55 and p75 TNF receptors were generously provided by Drs. W.
Lesslauer and H. Loetscher. We acknowledge S. Vermeulen for technical
assistance in phosphoamino acid analysis.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
![]()
CiteULike
Complore
Connotea
Del.icio.us
Digg
Reddit
Technorati What's this?
This article has been cited by other articles:
![]() |
S. Xu, C. C. L. Wong, E. H. Y. Tong, S. S. M. Chung, J. R. Yates III, Y. Yin, and B. C. B. Ko Phosphorylation by Casein Kinase 1 Regulates Tonicity-induced Osmotic Response Element-binding Protein/Tonicity Enhancer-binding Protein Nucleocytoplasmic Trafficking J. Biol. Chem., June 20, 2008; 283(25): 17624 - 17634. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. H. Bustos, A. Ferrarese, A. Venerando, O. Marin, J. E. Allende, and L. A. Pinna The first armadillo repeat is involved in the recognition and regulation of beta-catenin phosphorylation by protein kinase CK1 PNAS, December 26, 2006; 103(52): 19725 - 19730. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Murata, M. Tsuboi, K. Hikita, and N. Kaneda Protective Effects of Neurotrophic Factors on Tumor Necrosis Factor-related Apoptosis-inducing Ligand (TRAIL)-mediated Apoptosis of Murine Adrenal Chromaffin Cell Line tsAM5D J. Biol. Chem., August 11, 2006; 281(32): 22503 - 22516. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Izeradjene, L. Douglas, A. B. Delaney, and J. A. Houghton Casein Kinase I Attenuates Tumor Necrosis Factor-Related Apoptosis-Inducing Ligand-Induced Apoptosis by Regulating the Recruitment of Fas-Associated Death Domain and Procaspase-8 to the Death-Inducing Signaling Complex Cancer Res., November 1, 2004; 64(21): 8036 - 8044. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Okamura, N. Iwata, A. Nagata, A. Tamekane, M. Shimoyama, H. Gomyo, K. Yakushijin, N. Urahama, M. Hamaguchi, C. Fukui, et al. Involvement of casein kinase I{epsilon} in cytokine-induced granulocytic differentiation Blood, April 15, 2004; 103(8): 2997 - 3004. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Yoshii, T. Fukumori, Y. Honjo, H. Inohara, H.-R. C. Kim, and A. Raz Galectin-3 Phosphorylation Is Required for Its Anti-apoptotic Function and Cell Cycle Arrest J. Biol. Chem., February 22, 2002; 277(9): 6852 - 6857. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. V. Faundez and R. B. Kelly The AP-3 Complex Required for Endosomal Synaptic Vesicle Biogenesis is Associated with a Casein Kinase Ialpha -Like Isoform Mol. Biol. Cell, August 1, 2000; 11(8): 2591 - 2604. [Abstract] [Full Text] |
||||
![]() |
A. A. Van Linden, V. Cottin, C. Leu, and D. W. H. Riches Phosphorylation of the Membrane Proximal Region of Tumor Necrosis Factor Receptor CD120a (p55) at ERK Consensus Sites J. Biol. Chem., March 15, 2000; 275(10): 6996 - 7003. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.-t. Uh, A. Van Linden, and D. W. H. Riches Phosphorylation of 130- and 95-kDa Substrates Associated with Tumor Necrosis Factor-alpha Receptor CD120a (p55) J. Biol. Chem., January 14, 2000; 275(2): 793 - 800. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Cottin, A. Van Linden, and D. W. H. Riches Phosphorylation of Tumor Necrosis Factor Receptor CD120a (p55) by p42mapk/erk2 Induces Changes in Its Subcellular Localization J. Biol. Chem., November 12, 1999; 274(46): 32975 - 32987. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. B. Bioukar, N. C. Marricco, D. Zuo, and L. Larose Serine Phosphorylation of the Ligand-activated beta -Platelet-derived Growth Factor Receptor by Casein Kinase I-gamma 2 Inhibits the Receptor's Autophosphorylating Activity J. Biol. Chem., July 23, 1999; 274(30): 21457 - 21463. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Guesdon, C. G. Knight, L. M. Rawlinson, and J. Saklatvala Dual Specificity of the Interleukin 1- and Tumor Necrosis Factor-activated beta Casein Kinase J. Biol. Chem., November 28, 1997; 272(48): 30017 - 30024. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Dubois, C. Rommel, S. Howell, U. Steinhussen, Y. Soneji, N. Morrice, K. Moelling, and A. Aitken 14-3-3 Is Phosphorylated by Casein Kinase I on Residue 233. PHOSPHORYLATION AT THIS SITE IN VIVO REGULATES Raf/14-3-3 INTERACTION J. Biol. Chem., November 14, 1997; 272(46): 28882 - 28888. [Abstract] [Full Text] [PDF] |
||||
![]() |
T.-T. Yami |