(Received for publication, March 25, 1996, and in revised form, October 4, 1996)
From the Departments of Biochemistry and
§ Microbiology and Immunology, Emory University School of
Medicine, Atlanta, Georgia 30322-3050
Sphingosine, sphinganine, and other long-chain
(sphingoid) bases are highly bioactive intermediates of sphingolipid
metabolism that have diverse effects when added to cells, including the
inhibition of protein kinase C (PKC) as evaluated by both enzymatic
activity and [3H]phorbol dibutyrate
([3H]PDBu) binding. Nonetheless, changes in endogenous
sphingoid bases have not been proven to affect PKC or other signal
transduction pathways. We have discovered recently that changing
J774A.1 cells to new medium results in up to 10-fold increases in
sphingoid bases (Smith, E. R., and Merrill, A. H., Jr. (1995)
J. Biol. Chem. 270, 18749-18758); therefore, this
system was used to elevate sphingosine and sphinganine and determine if
PKC was affected. Incubation of J774A.1 cells in new medium for 30 min
increased the levels of these endogenous sphingoid bases to
approximately 0.5 nmol/mg of protein and decreased
[3H]PDBu binding by 40-60%. Addition of
NH4Cl, which suppresses the change in sphingosine, restored
[3H]PDBu binding. Elevation of endogenous sphinganine by
a second method (addition of fumonisin B1, an inhibitor of
ceramide synthase) also reduced [3H]PDBu binding;
therefore, elevations in sphingosine and sphinganine can both affect
PKC. The elevation in sphingoid bases was also associated with an
increase in the amount of PKC- (the major PKC isozyme in J774A.1
cells) in the cytosol, as determined by activity assays and immunoblot
analyses. Changing the culture medium affected other PKC isozymes,
increased cellular levels of diacylglycerol, dihydroceramide, and
ceramide, and altered the expression of two genes (the expression of JE
was increased, and the induction of MnSOD by TNF-
was potentiated).
Thus, changing the culture medium has numerous effects on J774A.1
cells, including the modulation of PKC by endogenous sphingoid
bases.
Sphingosine and other long-chain (sphingoid) bases, the structural backbones of sphingolipids, have been found to affect diverse cellular systems when added to in vitro assays, cells, and even applied to skin (1). These affected systems include, but are not limited to, protein kinase C (PKC)1 (2), Na+,K+-ATPase (3), phosphatidic acid phosphatase (4-7), phospholipases (including phospholipase D) (8, 9), retinoblastoma protein phosphorylation (10, 11), and sphingosine-activated protein kinase(s) (12). The inhibition of PKC has been studied most thoroughly in vitro using mixed micellar assays of the purified enzyme (2), as well as by evaluation of cellular functions dependent on this enzyme in platelets (2), neutrophils (13), HL-60 cells (14), and many other systems (1). Sphingosine inhibits PKC by acting as a competitive inhibitor of activation by diacylglycerol (DAG), phorbol dibutyrate (PDBu), and (for some isozymes) calcium (2) and also blocks activation by unsaturated fatty acids and other lipids (15). The exact mechanism by which sphingoid bases inhibit PKC remains unknown; however, since PKC binds to membranes through interactions with DAG and negatively charged phosphatidylserine (PS), sphingosine may be localized in regions of acidic lipids and block enzyme binding or activity (2, 14, 16, 17).
Cells normally contain low levels of free long-chain bases (1-10
nmol/g of tissue (wet weight) or 10-100 pmol/106 cells)
(18-21), and the levels have been found to change in response to
various treatments (22, 23). Nonetheless, there has been no direct link
between changes in endogenous sphingoid bases and inhibition of PKC. We
have found that the change from conditioned to new culture medium
stimulates relatively rapid increases in free sphingoid bases in J774
macrophages (24, 25). Similar observations have been made in other
cultured cell systems, including Swiss 3T3 fibroblasts (26), NIH-3T3
fibroblasts, A431 cells, and NG108-15 cells (27). The levels of free
sphingoid bases that are achieved within 30-60 min of incubation of
J774 cells in new medium (0.5 nmol/mg of protein) are comparable
with the levels of exogenous long-chain bases that inhibit PKC in human neutrophils (22). Therefore, this system serves as a model to determine
if endogenous sphingoid bases can affect PKC, as well as to explore
some of the implications of changing culture medium on cellular signal
transduction pathways.
Using this model, we have found that changes in endogenous sphingoid
bases affect [3H]PDBu binding, PKC activity, and the
subcellular localization of one PKC isozyme in J774 cells. In contrast,
modulation of endogenous ceramide after the medium change did not
affect the expression of two genes that are known to be induced by
TNF-: JE (28) and MnSOD (29). The change in culture medium did,
however, alter JE expression and potentiate TNF-
induction of
MnSOD.
[n-20-3H]Phorbol
12,13-dibutyrate (44.8 Ci/mmol), [-32P]ATP (>3000
Ci/mmol), [
-32P]dATP (>3000 Ci/mmol), and the ECL kit
were purchased from Amersham Corp. Escherichia coli DAG
kinase was purchased from Lipidex, Inc. (Westfield, NJ).
1,2-Dioleoyl-sn-glycerol and cardiolipin came from Avanti
Polar Lipids (Alabaster, AL). Monoclonal antibodies to PKC isozymes
(
,
I,
II,
,
,
, and
) came from Calbiochem. Fumonisin B1 (FB1) was purchased either from
Sigma or from the Division of Food Sciences and Technology, Council for
Scientific and Industrial Research (Pretoria, South Africa). Fetal
bovine serum was purchased either from Life Technologies, Inc. or from Biocell (Rancho Dominguez, CA). QuikHyb and the random-primed DNA
labeling kit were purchased from Stratagene (La Jolla, CA) and
Boehringer Mannheim, respectively. Zeta-Probe GT membrane (nylon
membrane for nucleic acid transfer) was purchased from Bio-Rad. Cetus
Corp. generously supplied human recombinant TNF-
. C2-Ceramide was prepared according to the method of Gaver
and Sweeley (30). All other reagents were from Sigma or were of the
highest available research quality.
J774A.1 cells (American Type Culture Center number TIB 67), a murine macrophage-like transformed cell line, were routinely grown in suspension at 37 °C in a spinner flask (Corning Glass) in DMEM supplemented with 10% FBS, penicillin-G (100 units/ml), streptomycin sulfate (100 µg/ml), and sodium biocarbonate (3.7 g/liter). Cells were passaged every 2-3 days by 1:4 dilution with fresh culture medium to yield a cell density of approximately 2 × 105 cells/ml. Unless indicated differently, cells were removed from the spinner flask, pelleted by gentle centrifugation, resuspended in new culture medium, and plated and allowed to adhere to 60-mm tissue culture dishes (Corning) at 5-7.5 × 105 cells in 2 ml of medium (for PKC studies) or plated at 1-2.5 × 106 cells/100-mm dish in 6 ml of medium (for RNA studies). Dishes were incubated 3 days at 37 °C, 5% CO2 before beginning the experiments.
C3HA cells, a mouse embryo fibroblast line (31), were grown as
described previously (32). These cells are an immortalized, contact-inhibited cell line and were grown on 100-mm dishes in 7 ml of
DMEM supplemented with 10% fetal calf serum (Intergen, Inc.), 2 mM glutamine, 100 units/ml penicillin, and 100 µg/ml streptomycin. For isolation of RNA, conditioned medium was removed from
dishes of confluent C3HA cells, the cells were washed quickly with
ice-cold sterile PBS, and TNF- was added in 7 ml of fresh culture
medium for 6 h to yield a final concentration of 400 units/ml.
For these experiments, J774 cells were removed from suspension culture and plated in 12-well dishes (Corning) at a density of 106 cells/well in 0.5 ml of conditioned medium (>36 h). The dishes were incubated for 24 h at 37 °C, 5% CO2. Except for conditioned medium treatments, the cells were washed twice with warm Kreb's Ringer phosphate-buffered saline, KRPG (0.5 mM MgCl2·6H2O, 0.12 M NaCl, 0.7 mM Na2HPO4, 1.5 mM NaH2PO4, 15 mM NaHCO3, 11 mM glucose), and new medium (0.5 ml) with or without NH4Cl was added and the cells incubated for 30 min. [3H]PDBu was added to the medium to make a final concentration of 20 nM. Dishes were returned to the incubator for 30 min, after which the medium was aspirated, and each well washed quickly three times with ice-cold KRPG. Cells were treated with 0.75 ml of 2% Triton X-100 for 30-60 min at 37 °C, and the solubilized mixture was transferred to scintillation vials, 4 ml of scintillation mixture added, and radioactivity determined by counting for 5 min. Specific binding was determined by the difference between total binding and binding in the presence of excess (>1 µM) cold PDBu. (The actual amount of cold PDBu varied in different experiments but generally was between 1 and 10 µM.)
For determination of binding in FB1-treated cells, the cells were plated in 12-well dishes at a density of 2 × 105 cells/well in 1 ml of fresh DMEM containing 10% FBS. Sterile FB1 (in water) was added to experimental dishes for a final concentration of 10 µM. The medium was changed every 24 h for 3 days and replaced with the same medium and concentration of FB1. To begin the experiments, conditioned medium was removed and replaced with new DMEM supplemented with 10% FBS, and the cells were incubated for various times before measuring [3H]PDBu binding as described (above). Sphingoid base mass was determined in identically treated dishes, as described completely in Merrill et al. (19).
Determination of DAG MassDAG mass was measured according to the method of Priess et al. (33), in which DAG from a crude lipid extract of cells was solubilized in defined mixed micellar conditions and quantitatively converted to [32P]phosphatidic acid by E. coli DAG kinase.
Cell Lysis and Subcellular FractionationFor identification of isozymic species and PKC activity assays, 2.5 × 105 cells were plated in 2.5 ml of DMEM containing 10% FBS in 60-mm dishes and incubated at 37 °C, 5% CO2 for 3 days. Following the specified treatments, cells were washed quickly with 3 ml of ice-cold PBS (0.12 M NaCl, 2.7 mM KCl, 1.2 mM K2HPO4, 1 M NaH2PO4), scraped into ice-cold buffer A (20 mM Tris, pH 7.5, 1 mM phenylmethylsulfonyl fluoride; 2 mM EDTA, 0.5 mM EGTA, 5 mM dithiothreitol, 50 µg/ml leupeptin, 20 µg/ml aprotinin), treated with 4 mM diisopropyl fluorophosphate for 20 min on ice and subcellular fractions obtained as described (33, 34). (We found that diisopropyl fluorophosphate treatment was necessary to obtain consistent, reliable results.) For in vitro activity assays, membrane and cytosolic fractions were supplemented with glycerol to a final concentration of 15% (w/v) to prevent loss of PKC activity.
Protein Kinase C AssayFor determination of activity, PKC was partially purified over DEAE-Sephacel columns (34, 35). Membrane and cytosolic fractions were passed over individual 0.5-ml DEAE-Sephacel columns equilibrated with buffer B (20 mM Tris, pH 7.5, 1 mM phenylmethylsulfonyl fluoride, 0.5 mM EDTA, 0.5 mM EGTA, 0.5 mM dithiothreitol, 50 µg/ml leupeptin, 20 µg/ml aprotinin). The columns were washed with approximately 5 column volumes of buffer B, and the bound enzyme was eluted with 1-1.5 ml of buffer B containing 0.25 M NaCl and immediately assayed for activity.
PKC activity was determined by measuring the incorporation of 32P into histone III-S protein (34). Calcium-independent activity was measured in the presence of DAG (0.04 µg/µl), PS (0.04 µg/µl), and 5 mM EGTA, whereas calcium-dependent activity was measured in the presence of 0.4 mM CaCl2, PS, DAG, but without EGTA. Values have been normalized for protein, which was determined by the Bio-Rad Bradford method (36).
Immunoblot AnalysesImmediately following subcellular fractionation, the isolated cytosolic and membrane extracts were boiled for 10-15 min in SDS sample buffer (37). Approximately 50-100 µg per sample was analyzed by SDS-polyacrylamide gel electrophoresis electrophoresis on 10% gels. Proteins were electroblotted at 30 V overnight onto nitrocellulose membranes (Schleicher & Schuell), which were blocked for 30 min with 5% powdered milk, incubated with primary rabbit immunoglobulin G antibodies against PKC isozymes (1:1000 dilution in 1% powdered milk in TBS (20 mM Tris, pH 8.0, 150 mM NaCl)) for 2 h at room temperature, washed three times with 0.5% Triton X-100 in TBS, and probed for 1 h with a donkey anti-rabbit IgG conjugated to horseradish peroxidase (1:10,000 dilution). Blots were probed with ECL detection reagents according to the manufacturer's directions and exposed to film for 1-60 min.
RNA IsolationRNA was isolated from cells according to the
acid guanidinium thiocyanate-phenol-chloroform extraction protocol
(38). Following the indicated treatments and incubations, the medium
was aspirated from dishes, the cells were washed twice with ice-cold
PBS (0.12 M NaCl, 2.7 mM KCl, 1.2 mM K2HPO4, 1 M
NaH2PO4), and 1 ml of denaturation solution was
added to each dish. The final RNA extract was suspended in 100 µl
0.1% diethyl pyrocarbonate-treated water and stored at 75 °C.
Northern blot hybridizations were performed as described by Thomas (39), with the modifications specified in Boss et al. (40). Five micrograms of total RNA were separated by electrophoresis on 1.5% agarose (in 1.5 M formaldehyde) gels (50-60 V for 2 h). The amount and purity of RNA loaded was verified by ethidium bromide staining of the gels. RNA was transferred overnight to nylon membranes in 20 × SSC (3 M NaCl, 0.3 M sodium citrate, pH 7.0) (41) and cross-linked to the filters using 1200-µJ UV light (Stratalinker). Filters were prehybridized for 2-3 h and hybridized overnight (12-18 h) at 42 °C in deionized formamide:2 × Northern solution (0.1 M sodium phosphate, pH 7.0, 10 × SSC, 2% SDS, 10 × Denhardt's solution, 20% dextran sulfate) (1:1, v/v), which contained 100 µg/ml boiled sonicated calf thymus DNA. Alternatively, some filters were hybridized in 10-15 ml of QuikHyb solution/filter according to the manufacturer's instructions (Stratagene, Inc.), with identical results.
Labeling of DNA ProbesJE and MnSOD hybridization probes
were generated by the random primer labeling method (42, 43) or
according to the prescribed protocol in the random primer labeling kit
(Boehringer Mannheim). The amount of RNA loaded was normalized to
-actin, which was monitored by hybridization to a mouse
-actin
oligonucleotide labeled at the 5
-end with [
-32P]ATP
(41). Autoradiography was carried out for 4-24 h at
75 °C using
intensifying screens. Autoradiographs were analyzed by laser densitometry (Personal Densitometer, Molecular Dynamics) using the
ImageQuant software system.
To examine whether endogenous long-chain bases can affect PKC,
several indices of the state of this enzyme were examined in J774 cells
under conditions that increase free sphingosine and sphinganine (24,
25) (e.g. changing the medium) versus conditions where this increase is suppressed (conditioned medium or new medium containing NH4Cl (Fig. 1).
[3H]PDBu Binding
After 30 min in fresh culture medium, [3H]PDBu binding decreased 40-60%, from 0.189 ± 0.028 to 0.083 ± 0.021 pmol [3H]PDBu bound per 106 cells compared with cells that were returned to conditioned medium (cf. 0 mM NH4Cl versus "conditioned medium" in Fig. 1). Since exogenous sphingosine and other long-chain bases cause dose-dependent inhibition of [3H]PDBu binding by PKC in mixed micellar assays (2) and intact cells (1, 2), this reduction in [3H]PDBu binding suggests that endogenous long-chain bases can affect PKC, the major intracellular receptor for phorbol esters in most cells (44).
[3H]PDBu binding was also assessed in cells incubated in new culture medium with increasing concentrations of NH4Cl, which blocks the generation of total sphingoid bases (Fig. 1), although this is due to a reduction in sphingosine and not sphinganine (25). Ammonium chloride restored [3H]PDBu binding in a concentration-dependent manner (Fig. 1), and by 4 mM NH4Cl, the binding had returned to the level obtained in conditioned medium. At this concentration of NH4Cl, sphinganine levels were reduced by 20% and sphingosine by over 50%. At a higher NH4Cl concentration (8 mM), [3H]PDBu binding exceeded that found in conditioned medium.
Similar studies were conducted using chloroquine (25 µM) instead of NH4Cl. This lysoosmotrophic agent suppressed the elevation in sphingosine to the same extent as 4 mM NH4Cl and restored [3H]PDBu binding to the level in conditioned medium (data not shown). Chloroquine was not studied further, however, since it is a hydrophobic amine that, like sphingoid bases, might affect PKC activity.
These results establish that changing the medium to elevate endogenous long-chain bases results in a reduction of [3H]PDBu binding by PKC and that changing the medium with suppression of the generation of sphingosine restores [3H]PDBu binding. To elevate free sphingoid bases independently of changing the medium, the cells were treated with fumonisin B1 (FB1), a fungal toxin that inhibits ceramide synthase (45). After 3 days, sphingoid base levels were elevated 2-3-fold (Table I), with the sphinganine mass increasing 600% over control at time 0 and 1800% after 60 min in new medium. Phorbol ester binding was 30% lower in the FB1-treated cells at time 0 (i.e. 0.68 ± 0.07 versus 0.95 ± 0.21 pmol/mg of protein), and after 60 min in new medium, binding was approximately 50% lower than control levels (p < 0.05). These results establish that the elevation of sphinganine is also associated with reduced [3H]PDBu binding.
|
To
examine another aspect of PKC behavior, and to take into account the
possibility that only some isozymes might be affected (46-48),
immunoblot analyses were conducted using cytosolic and membrane
fractions prepared from cells after incubation for 30 min in new or
conditioned medium, with or without 1.6 µM PMA (Figs. 2 and 3). This concentration of PMA was
chosen, since it fully activates superoxide production in J774
macrophages (49) and would thus be expected to elicit significant
changes in PKC.
Immunoblot analysis of cytosolic and membrane fractions indicated that
J774A.1 cells contain PKC-, PKC-
I, PKC-
II, and PKC-
. PKC-
, a member of the nPKC class that does not require calcium for
activation (46, 47), was found to be the major isozyme, based on
Western blot analysis and activity assays (discussed below). This
finding is consistent with its prominence in hematopoietic cells (34,
50). PKC-
was located primarily in the membrane fraction in cells
incubated in either conditioned or new medium (Fig. 2A);
however, the change to new culture medium reproducibly increased
PKC-
in the cytosolic fractions. Incubation in conditioned medium,
or with PMA in either new or conditioned medium, did not evoke a
similar redistribution. PKC-
appeared to have a slightly reduced
mobility on 10% SDS-polyacrylamide gels after PMA treatment, which
indicates phosphorylation (34, 48, 51), but this was not explored
further in this study. Other bands that did not correspond to the
molecular weight of PKC were also detected; it is unclear whether these
bands represent proteolytic fragments of the enzyme or nonspecific
binding of the antibody.
The appearance of PKC- in the cytosol indicated a displacement of
PKC from the membrane (presumably due to increases in free long-chain
bases), as has been seen in studies in which exogenous sphingosine was
used (2, 14, 16). PKC-
localization, therefore, was determined in
cells incubated with sphingosine in conditioned medium. The
concentrations of sphingosine were similar to the increases in
sphingoid bases determined after cells are incubated in new medium, and
which have also been found to affect PKC in other cells, including
Chinese hamster ovary cells (0.75-4 µM] (52),
neutrophils (1-3 µM) (13, 16), and HL-60 cells (
4 µM) (53). As shown in Fig. 2B, 0.5 µM sphingosine caused little or no change in PKC-
distribution, but 5.0 µM increased the amount of PKC-
in the cytosolic fractions compared with conditioned medium alone (and
to approximately the same level found in new medium). Addition of 10 mM NH4Cl in new medium prevented PKC-
redistribution to the cytosolic fraction. Thus, endogenous (and exogenous) sphingoid bases increase the amount of PKC-
in the cytosol, and the factors that suppress the endogenous sphingoid bases
reversed this change.
The classical PKC (cPKC) isozymes, which require calcium in addition to
PS and DAG for activation (42), were detected in both cytosol and
membrane fractions of cells at time 0, although the amount of cytosolic
isozyme exceeded membrane-bound in all cases (Fig. 3). Incubation of
cells in new medium eliminated PKC- and substantially reduced the
amount of PKC-
I and
II in the cytosol; all three were eliminated
in the membrane fractions. When incubated, instead, in conditioned
medium, these losses of PKC-
,
I, and
II did not occur.
Addition of PMA with new medium further reduced the amount of cPKCs
present in the cytosolic fraction, but PMA did not alter the amount or
distribution of cPKCs in conditioned medium (Fig. 3, lanes 4 and 5 under "Cytosol" and
"Membrane" fractions). As shown in Fig.
4, NH4Cl did not prevent the loss of PKC-
in new medium alone. Similar results were obtained for PKC-
I and PKC-
II (not shown).
PKC Activity in Isolated Cytosolic and Membrane Fractions
To
determine whether the changes in PKC distribution and isozyme
corresponded to altered enzyme activity, PKC activity was measured in
membrane and cytosolic fractions obtained from cells treated as
described for Figs. 2, 3, 4. To determine total PKC activity, the
fractions were assayed in the presence of DAG and PS with 400 µM calcium, whereas calcium-independent activity was determined with 5 mM EGTA and no calcium (Fig.
5A). Calcium-dependent activity
was calculated as the difference between total and calcium-independent activity (Fig. 5B). With each treatment, the activity in
membrane fractions was 4-5-fold higher than cytosolic activity, which
agreed with the predominant localization of PKC- (the most abundant isozyme) in membrane fractions. Calcium-independent and
-dependent activity increased in the membrane fraction from
cells incubated in new medium for 30 min. In agreement with the
appearance of PKC-
in the cytosol only after the change to new
medium, incubation in new medium increased the activity in the cytosol;
calcium-independent activity entirely accounted for this increase (Fig.
5A). Cytosolic calcium-dependent activity was
undetectable under the conditions examined.
To determine whether the inability of PMA to activate PKC in conditioned medium (Figs. 2, 3, 4) extended to a biological response, two other PKC-related activities were examined in J774 cells. The rate of superoxide generation of cells in new medium containing 5 µM PDBu was found to be 1.21 nmol/min/107 cells, whereas the rate was 0.17 nmol/min/107 cells in conditioned medium (or 7-fold lower than in new medium). Additionally, [14C]glucose oxidation was approximately 2-fold higher in new medium. PMA stimulated a 4.4-fold increase in [14C]CO2 release from cells in new medium, but only a 2.6-fold increase from cells in conditioned medium. Thus, conditioned medium may contain various factors that blunt PKC-activation (including, but not limited to, ammonia).
Changes in DAG in New MediumSince there is a seemingly
paradoxical increase in both cytosolic and membrane PKC when cells were
incubated in new culture medium (Fig. 5), it appeared that other
modulators of PKC might contribute to some of these changes. For
example, changes in DAG might account for the the reduction in
[3H]PDBu binding, since phorbol esters are analogs of DAG
(46). As shown in Fig. 6A, DAG mass did rise
60% (to 0.388 ± 0.046 nmol/mg of cell protein) within 30 min
following the change from conditioned to new medium and remained
elevated (0.346 ± 0.035 pmol/mg of protein) after 60 min.
However, unlike sphingosine, the increase in DAG was not blocked, and
sometimes rose, upon adding NH4Cl (Fig. 6B).
Additionally, 25 µM chloroquine had no effect on DAG mass. Thus, changes in DAG mass are not likely to be responsible for
the alterations in [3H]PDBu binding that occurred with
the change to new medium.
Effect of New Medium on Gene Expression
Since changing the
culture medium has profound effects on numerous bioactive lipids
(sphingosine, DAG, and ceramide (this study and Ref. 25), we explored
whether this might have an impact on gene expression. For this
analysis, we selected two genes that are TNF-inducible, JE and MnSOD
(29, 40, 58, 59), since sphingolipids have been implicated as mediators
for this cytokine (54-57). However, incubation of J774A.1 cells with
bacterial sphingomyelinase (1-10 units/ml) did not induce JE at 1 or
4 h nor alter the JE RNA levels in TNF- treated cells; and
addition of C2-ceramide (0.10-10 µM) was
also without effect in either new medium alone or in conditioned medium
(data not shown). Thus, ceramide per se does not appear to
affect JE or MnSOD expression in this system.
Nonetheless, changing J774 cells to new medium influenced the
expression of these genes. Following the incubation of cells in new
medium, JE expression was elevated 2-4-fold at 30 min and 1 h and
thereafter returned to conditioned medium levels by 2-4 h (data not
shown); these changes were not due to differences in RNA loading as
determined by -actin levels. The increase in JE RNA did not appear
to be affected by TNF; MnSOD expression was induced by TNF-
(by
8 h, MnSOD expression was increased approximately 6-8-fold by
TNF-
). (In comparison, C3HA fibroblasts showed a similar increase in
MnSOD by 6 h, and an 80-fold elevation in JE expression after
6 h, which depended strongly on TNF-
for induction.)
The changes in MnSOD and JE RNA expression were not seen if J774 cells
were kept in conditioned medium, even if TNF- was added (Fig.
7). In cells incubated in conditioned medium for 1 h, JE did not increase over time 0 conditioned medium levels, and
although elevated in new medium, JE remained unresponsive to further
induction by TNF-
in new medium (Fig. 7). Conditioned medium
inhibited the induction of MnSOD by TNF that occurred by 4 h in
new medium. Invariably, no change was seen in MnSOD at 1 h. Even
though we do not know if these changes are mediated by sphingosine or
other bioactive lipids, the observations that new culture medium can
have such large effects on gene expression should be borne in mind in
studies of these, and possibly other, cells.
This study examined several indices of PKC function to understand whether the changes in endogenous sphingoid bases in J774A.1 cells upon adding new medium (which induces sphingolipid turnover) or FB1 (which inhibits ceramide synthase) affect cellular signaling pathways. Increases in endogenous long-chain bases by changing the culture medium or addition of FB1 correlated with inhibition of PKC, as reflected by decreased [3H]PDBu binding and increases in PKC in the cytosolic fraction. The effects of sphingosine and sphinganine could be evaluated separately, since NH4Cl alters intracellular sphingosine levels and FB1 affects primarily sphinganine mass. Both agents increase [3H]PDBu binding and each sphingoid base can apparently alter PDBu binding and PKC. These results agree with in vitro studies, which demonstrated that sphingosine and sphinganine are equally effective in inhibiting PKC in mixed micellar assays and in cells (16). The effects of FB1 are particularly interesting, because Huang et al. (60) have reported that FB1 causes dose-dependent repression of PKC in CV-1 cells. Although that study did not determine whether FB1 caused sphingoid base accumulation, this has been seen in every other cell system examined to date (61); therefore, this result is consistent with our finding that increases in endogenous sphingosine and sphinganine can affect PKC.
As we are aware, this is the first analysis of the PKC isozymes in J774
cells, with the finding that PKC- is the major species plus smaller
amounts of PKC-
and PKC-
I and
II; although, a previous study
detected a polypeptide that reacted with antibodies directed against
the C-terminal region of the regulatory domain of PKC-
(62). PKC-
is located primarily in the membrane fraction of J774 cells, as well as
in many other cell types (34, 63, 64). Thus, translocation of PKC-
to the membrane would not seem to be a determinant of the activation of
this PKC isozyme. One would predict, nonetheless, that the shift in the
distribution of PKC-
from the membrane to the cytosol by endogenous
sphingoid bases would in some way affect its activity and/or the types
of protein substrates that are phosphorylated.
The change to new medium alters cellular levels of several bioactive lipids: DAG (this study), sphingoid bases, ceramide, and phospholipids (25), all of which may influence PKC or other signal transduction pathways directly or indirectly. The complexity of these changes make this difficult to study in detail; however we have found that changing the cells to new medium affected the induction of JE and MnSOD. These effects of new and conditioned media should be borne in mind in signaling studies using J774 cells and other cell lines since similar "bursts" of sphingolipid metabolism have been seen in many systems (26, 27).
It is not known if there is any physiological significance to these changes in endogenous sphingoid bases and their effects on PKC. It has been proposed that sphingoid bases might establish a "set point" for PKC activation by positive effector stimuli (65). Even if this is not the case, the elevation of sphingoid bases (and aberrant regulation of PKC) in diseases caused by FB1 (61) and in some sphingolipidoses (13, 66) may be significant contributors to the pathophysiology associated with these diseases.
We gratefully acknowledge and appreciate the technical assistance and advice of Drs. Edward Bowman and David Uhlinger for PKC isolation, Dr. Shiv-Raj Tyagi for DAG measurements, Dr. Xiang-Xi Xu for immunoblot analysis, and Drs. James Riley and Joseph Schroeder for RNA isolation and Northern blot analysis.