(Received for publication, October 22, 1996, and in revised form, February 10, 1997)
From CNRS, URA 583, Université Paris V, 78 350 Jouy-en-Josas, France
We have shown that estrogens and calcitriol, the
hormonally active form of vitamin D, increase the concentration of
intracellular calcium ([Ca2+]i) within
5 s by mobilizing calcium from the endoplasmic reticulum and the
formation of inositol 1,4,5-trisphosphate and diacylglycerol. Because
the activation of effectors as phospholipase C (PLC) coupled to
G-proteins is the early event in the signal transduction pathway
leading to the inositol 1,4,5-trisphosphate formation and to
[Ca2+]i increase, we described different PLC
isoforms (1,
2,
1, and
2, but not
4) in female rat
osteoblasts using Western immunoblotting. The data showed that
phospholipase C
was involved in the mobilization of
Ca2+ from the endoplasmic reticulum of Fura-2-loaded
confluent osteoblasts by calcitriol and 17
estradiol, and PLC
was ineffective. The data also showed that only a PLC
1 linked to a
Pertussis toxin-insensitive G-protein and a PLC
2 coupled to a
Pertussis toxin-sensitive G-protein are involved in the effects of
calcitriol and 17
estradiol on the mobilization of Ca2+
from intracellular Ca2+ stores. In conclusion, these
results may be an important step toward understanding membrane effects
of these steroids and may be an additional argument in favor of
membrane receptors to steroid hormones.
An increase in the turnover of inositol lipids in response to receptor is one of the most important molecular mechanisms used by cells for transmembrane signaling. The initial event is the hydrolysis of phosphatidylinositol 4,5-bisphosphate, a reaction catalyzed by a phosphoinositide-specific phospholipase C (PLC),1 which generates two intracellular second messengers, inositol 1,4,5-trisphosphate and 1,2-diacylglycerol (1-4). Inositol 1,4,5-trisphosphate binds to specific receptors on the endoplasmic reticulum (5) and mobilizes intracellular calcium, whereas diacylglycerol activates protein kinase C (6), which results in increased phosphorylation of cellular proteins.
Molecular cloning has revealed at least three major families of PLC,
,
, and
(7-9). Each of these families occurs in a number of
isoforms. The enzymes are classified on the basis of their size and
their immunological and structural similarities. The PLC isoforms have
two highly conserved domains, X and Y, which form the active site of
the protein. PLC
and PLC
proteins differ from PLC
in that
they have shorter C-terminal extensions past the end of the Y domain
(9). This diversity among the PLC isoforms also extends to distinct
mechanisms of regulation and function for the three PLC families. PLC
is regulated via the phosphorylation of tyrosine residues between
the X and Y domains by receptor tyrosine kinases (10-12). PLC
enzymes, of which there are four isoforms, PLC
1-4, are regulated
via heterotrimeric G-proteins in response to an agonist binding to a
receptor (13-15). The way in which PLC
is regulated is not yet
known, but enzyme activity is not affected by either the G-protein
subunits or by receptor tyrosine kinases (16).
The activation of molecules as PLC is an early event in the signal transduction pathways leading to a variety of cellular responses, including metabolism, proliferation, secretion, and motility. We have shown that estrogens (17) and calcitriol (18, 19), the hormonally active form of vitamin D, increase the concentration of intracellular calcium within 5 s by mobilizing calcium from the endoplasmic reticulum and the formation of inositol 1,4,5-trisphosphate and diacylglycerol. This process involves the activation of a phospholipase C linked to a Pertussis toxin-sensitive G-protein for estradiol (17) and an as yet uncharacterized G-protein for calcitriol (20).
However, no information is presently available on the PLC present in osteoblasts, the cells responsible for osteogenesis, or about the PLC isotypes involved in the membrane effects of calcitriol and estradiol.
We have therefore described PLC isoforms in the osteoblasts of female
rats. We have also identified the PLC isoenzymes involved in the rapid
actions of calcitriol and 17 estradiol on the mobilization of
calcium from the endoplasmic reticulum.
The ECL kit and Fura-2/AM were from Amersham
Corp. Polyclonal rabbit anti-PLC antibodies to PLC 1, PLC
2, PLC
3, PLC
4, PLC
1, and PLC
2 and antigens raised against PLC
1, PLC
2, PLC
3, PLC
1, and PLC
2 were from Santa Cruz
Biotechnology, Inc., and Tebu (Le Perray en Yvelines, France), and
peroxidase-conjugated goat anti-rabbit IgG was from Bio-Rad (Ivry sur
Seine, France). 1,25(OH)2D3 was from Hoffman-La
Roche (Bazel, Switzerland), and 17
estradiol was from Sigma.
-Minimum essential medium without phenol red and fetal calf serum
were from Eurobio (Paris, France).
Two-day-old female Wistar rats were from Charles River Breeding Laboratories (St Aubin les Elbeufs, France). Osteoblasts were isolated from parietal bones of the newborn rats by sequential enzymatic digestion (21). These cells had the following osteoblast characteristics: high alkaline phosphatase activity, high type I collagen synthesis, a cAMP and intracellular calcium response to parathyroid hormone, and an osteocalcin response to 1,25-dihydroxyvitamin D3.
Cells were grown on rectangular glass coverslips or in Petri dishes
(100 cm2) for 4 days in phenol red-free -minimum
essential medium supplemented with 10% heat-inactivated fetal calf
serum. Cells were then incubated for 72 h in phenol red-free
medium containing 1% heat-inactivated fetal calf serum and transferred
to serum-free medium 24 h before use.
Cells were washed three
times with ice-cold phosphate-buffered saline, pH 7.4, then scraped off
into ice-cold extraction buffer (20 mM Tris-HCl, pH 7.5, 0.5 mM EGTA, 2 mM EDTA, 0.6 mM
pepstatin, 0.5 mM benzamidine, 0.1 mM
leupeptin, 2 mM phenylmethylsulfonyl fluoride, 0.125 mM aprotinin, and 1 mM dithiothreitol). Cells were sonicated on ice (twice for 20 s at 40 KHz) and the
homogenate was centrifuged for 10 min at 600 × g to
remove nuclei. The remaining homogenate was centrifuged at 100,000 × g for 60 min, and the supernatant (cytosol fraction) was
saved. The pellet containing the plasma membranes was resuspended in
extraction buffer containing 0.3% Triton X-100 (w/w), left on ice for
60 min, and centrifuged again at 100,000 × g for 60 min. The resulting supernatant (solubilized membrane fraction) was
collected. All fractions (homogenate without nuclei, cytosol, and
membrane) were stored at 80 °C.
The brain was used as a positive control, because the cerebellum is
rich in PLC 3, PLC
4 and PLC
2, whereas the cerebral ventricules are rich in PLC
1, PLC
2 and PLC
1 (22). Extracts of these two tissues were treated in the same way as the osteoblast homogenate.
Protein was determined by the method of Bradford (23) with bovine serum albumin as standard. Alkaline phosphatase activity, as an enzyme marker of the plasma membrane, was assayed as described by Lieberherr et al. (24).
Protein Separation and ImmunoblottingProteins were
separated by SDS-polyacrylamide gel electrophoresis (7.5% resolving
gel) in 25 mM Tris-base, pH 8.3, 192 mM glycine, 0.1% SDS (25). They were then electrophoretically transferred to nitrocellulose membranes (Immobilon P) in the same buffer with 20%
ethanol for 2 h at 100 V (26). Nonspecific binding to
nitrocellulose was prevented by incubating the membranes in 50 mM Tris-buffered saline (TBS), pH 7.5, containing 150 mM NaCl, 5% skim milk powder, and 0.05% Tween 20 for
12 h at 4 °C. The membranes were given washes in TBS with 0.1%
Tween 20 and were incubated overnight at 4 °C with
isoenzyme-specific polyclonal rabbit antibodies (PLC 1, PLC
2,
PLC
3, PLC
4, PLC
1, and PLC
2). The concentrations of PLC
antibodies in TBS, 1.5% skim milk, 0.1% Tween 20 were as follows: 0.1 µg/ml for PLC
1, 0.5 µg/ml for PLC
2, 1 µg/ml for PLC
3,
0.5 µg/ml for PLC
4, 0.5 µg/ml for PLC
1, and 1 µg/ml for
PLC
2. Unbound antibodies were removed by four washes with TBS,
0.1% Tween 20; the antibodies bound to nitrocellulose were detected
using peroxidase-conjugated goat anti-rabbit IgG (1 mg/ml) (diluted
1/5000 in TBS, 1.5% skim milk, 0.1% Tween 20). The antigen was
detected by ECL. The molecular size standards used for calculating the
apparent molecular mass of the PLCs were: ovalbumin, 48 kDa; bovine
serum albumin, 87 kDa;
galactosidase, 120 kDa; and myosin, 199 kDa.
In some experiments, the specificity of the antibodies was verified by
incubating these antibodies at room temperature for 2 h with the
corresponding peptide (antibody:peptide ratio, 1:10 or 1:100 for Santa
Cruz Biotechnologies antibodies, according to the specifications of the
manufacturer) prior to use.
The film were scanned with Scanjet II CX/T (Hewlett-Packard) and labeled bands were quantified using Image Quant NT software (PhosphorImager SI, Molecular Dynamics).
Calcium Measurement and Experimental ProtocolThe cells were washed with Hanks' HEPES, pH 7.4 (137 mM NaCl, 5.6 mM KCl, 0.441 mM KH2PO4, 0.442 mM Na2HPO4, 0.885 mM MgSO4·7H2O, 27.7 mM glucose, 1.25 mM CaCl2, and 25 mM HEPES), and loaded with 1 µM Fura-2/AM for 30 min in the same buffer at room temperature. The glass coverslip carrying the cells was inserted into a cuvette containing 2.5 ml of Hanks' HEPES, pH 7.4. The cuvette was placed in a thermostatted (37 °C) Hitachi F-2000 spectrofluorometer. Drugs and reagents were added directly to the cuvette with continuous stirring.
The Fura-2 fluorescence response to the intracellular calcium concentration ([Ca2+]i) was calibrated from the ratio of the 340/380 nm fluorescence values after subtraction of the background fluorescence of the cells at 340 and 380 nm as described by Grynkiewicz et al. (27). The dissociation constant for the Fura-2-Ca2+ complex was taken as 224 nM. The values for Rmax and Rmin were calculated from measurements using 25 µM digitonin, 4 mM EGTA, and enough Tris base to raise the pH to 8.3 or higher. Each measurement on Fura-2-loaded cells was followed by a parallel experiment under the same conditions with non-Fura-2-loaded cells.
The direct effects of 100 pM calcitriol and estradiol on
[Ca2+]i were tested because this concentration of
steroids causes a maximal increase in [Ca2+]i in
confluent female osteoblasts (18, 17). Confluent female osteoblasts
were then permeabilized for 5 min with 50 µg/ml saponin in the
presence of anti-PLC antibody or nonimmune rabbit serum, used at 10 times the concentration used for Western blotting. Cells were washed
twice to remove saponin and incubated with the anti-PLC antibody or
nonimmune rabbit serum for 1 h at 37 °C. 1 µM
Fura-2/AM was added for the last 20 min of incubation. In some
experiments, anti-PLC 1 and anti-PLC
2 antibodies were set up in
competition with the antigens against which they were produced or with
the antigens corresponding to the other anti-PLC antibodies for 2 h at room temperature (antibody:peptide ratio, 1:10 or 1:100 for Santa
Cruz Biotechnologies antibodies, according to the specifications of the
manufacturer) prior to use.
The G-protein involved in the actions of calcitriol and estradiol was characterized by incubating cells with 100 ng/ml Pertussis toxin (PTX) for 16 h. Fura-2/AM loading and [Ca2+]i measurements were carried out with the toxin.
Statistical AnalysisThe data were analyzed by one-way analysis of variance. Treatment pairs were compared by Dunnett's method. A value of n represents n different cultures for a specific experiment.
SteroidsCalcitriol and 17 estradiol were dissolved in
ethanol; the final concentration of ethanol in the medium never
exceeded 0.01%. This concentration of ethanol was without effect on
intracellular calcium concentration (data not shown).
Table I shows the distribution of alkaline phosphatase activity in the subcellular fractions of osteoblasts and brain. Alkaline phosphatase was mostly in the plasma membrane fractions of osteoblasts and brain, because alkaline phosphatase is a membrane-bound enzyme (28).
|
The films were scanned
using a Scanjet II CX/T (Hewlett Packard) densitometer. The linear
range of protein concentrations on ECL Western blots was 5-50 µg of
proteins. All Western blots were done with 35 µg of proteins of each
subcellular fraction for each tissue or cell. Western blotting showed a
150-kDa immunoreactive band in soluble and membrane fractions of brain
and osteoblasts using the PLC 1 antibody (Fig. 1).
Most of the PLC
1 immunoreactivity was in the membrane fraction of
osteoblasts and in the soluble fraction of brain (Fig.
2). Immunoblots probed with the PLC
2 antibody showed
a 163-kDa immunoreactive band in osteoblasts and cerebellum (Fig. 1),
which was mainly in the osteoblast and cerebellum membrane fractions
(Fig. 2). There was a 153-kDa immunoreactive band in soluble and
particulate fractions of brain and osteoblasts using the PLC
3
antibody (Fig. 1), mainly in the cytosolic fractions in both brain and
osteoblasts (Fig. 2C). PLC
4 with an apparent molecular
mass of 160 kDa was mainly in the membrane fraction of brain, but no
immunoreactive band for PLC
4 was found in osteoblasts whatever the
protein (35-100 µg) concentration and the anti-PLC antibody (0.5-5
µg/ml) concentration (Fig. 1) (data shown for 35 µg in Fig. 2).
Immunoblots probed with the PLC
1 antibody showed a 156-kDa
immunoreactive band mostly in the cytosolic fractions of brain and
osteoblasts (Figs. 1 and 2). Immunoblots with the PLC
2 antibody
revealed a 145-kDa immunoreactive band in both tissues. Although the
signal intensities for this isoenzyme in the two subcellular fractions
of the brain were the same, the greatest signal was obtained from the
membrane fraction of osteoblasts (Figs. 1 and 2).
The competitive Western blot using polyclonal PLC 1, PLC
2, PLC
3, PLC
1, and PLC
2 antibodies and the antigens they were
raised against showed that the immunoreactivity was completely abolished only when the antigen was used at 100 times the concentration used for the corresponding antibody (data not shown). This also showed
that each of the antibodies was using its intended target isoform.
The basal intracellular calcium concentration ([Ca2+]i) in confluent female rat osteoblasts was 135 ± 5 nM (mean ± S.E.; n = 6). Pretreament of the cells with saponin for 5 min followed by incubation for 60 min with the anti-PLC antibody in the absence of saponin did not alter the basal [Ca2+]i. Nonimmune serum did not alter the basal [Ca2+]i or the [Ca2+]i response to the steroids.
Fig. 3 shows the transient increase
([Ca2+]i = 160 ± 5 nM,
mean ± S.E.; n = 6; p < 0.001)
in [Ca2+]i induced by 100 pM
calcitriol, which is the concentration of calcitriol that has the most
effect on this parameter (18). [Ca2+]i dropped
rapidly after 15 s but remained above the basal level (21 ± 2%, mean ± S.E.; n = 6; p < 0.001). The calcitriol-induced increase in
[Ca2+]i was partly inhibited by anti-PLC
1
antibody (Fig. 3A). The residual increase was due to a
Ca2+ influx from the extracellular medium because this
remaining increase was totally blocked by preincubating the cells for
30 s with 2 mM EGTA (18). On the other hand, the
antibodies to PLC
2, PLC
3, PLC
1, or PLC
2 did not block
the effect of calcitriol on [Ca2+]i (Fig. 3,
B and C).
Fig. 4 shows the transient increase
([Ca2+]i = 130 ± 4 nM,
mean ± S.E.; n = 6; p < 0.001)
in [Ca2+]i induced by 100 pM 17
estradiol, which is the concentration of estradiol that has most effect
on this parameter (17). [Ca2+]i dropped rapidly
after 15 s but remained above the basal level (19 ± 1%,
mean ± S.E.; n = 6; p < 0.001).
The estradiol-induced increase in [Ca2+]i was
partly inhibited by the anti-PLC
2 antibody (Fig. 4A).
The residual increase was due to a Ca2+ influx from the
extracellular medium because this remaining increase was totally
blocked by preincubating the cells for 30 s with 2 mM
EGTA (17). But anti-PLC
1, anti-PLC
3, anti-PLC
1, and anti-PLC
2 antibodies did not block the effect of estradiol on [Ca2+]i (Fig. 4, B and
C).
Polyclonal anti-PLC 1 and anti-PLC
2 antibodies were incubated
for 2 h with their corresponding antigens or with the antigens used for producing the anti-PLC
3, anti-PLC
1, and anti-PLC
2 antibodies (antibody:antigen ratio, 1:10 or 1:100) before use. The
inhibition of the calcitriol-induced increase in
[Ca2+]i due to the anti-PLC
1 antibody totally
disappeared only when the anti-PLC
1 antibody was co-incubated with
its antigen but not with the antigens corresponding to PLC
2, PLC
3, PLC
1, or PLC
2. Only the co-incubation with the PLC
2
antigen blocked the inhibition observed with the anti-PLC
2 antibody
on the estradiol-induced increase in [Ca2+]i.
This occurred when the antigen was used at 100 times the concentration
used for the corresponding antibody (data not shown).
Preincubation of the cells for 16 h
with Pertussis toxin did not alter the basal
[Ca2+]i or the [Ca2+]i
response to 1 pM-10 nM calcitriol (data not
shown). In contrast, PTX partly blocked the response to 100 pM estradiol (Fig. 5). The residual increase
disappeared when cells were incubated with 2 mM EGTA.
This is, to our knowledge, the first study showing a direct
involvement of phospholipase C in the membrane actions of
calcitriol or 17
estradiol in confluent female rat osteoblasts; PLC
is ineffective. The data also indicate that only PLC
1 linked to a Pertussis toxin-insensitive G-protein and a PLC
2 coupled to a
Pertussis toxin-sensitive G protein are involved in the effects of
calcitriol and 17
estradiol, respectively, on the mobilization of
Ca2+ from intracellular Ca2+ stores.
Female rat osteoblasts possess several isoforms of PLC: 1,
2,
3,
1, and
2, as shown by Western immunoblotting. Confluent female rat osteoblasts do not possess PLC
4, which is present in the
cerebellum (22) (see Fig. 3). This suggests that this isoenzyme may
have a tissue and cellular specificity or that the gene encoding PLC
4 is posttranscriptionally regulated. The gene may be expressed
during the logarithmic phase of proliferation and not when the cells
are confluent, corresponding to the differentiation of osteoblasts, or
the gene may not be expressed in cells during proliferation and
differentiation. These possibilities can be checked by in
situ hybridization and Northern blot analysis of osteoblasts at
different stages of maturation.
Most of the PLC 1 and PLC
2 in mature confluent osteoblasts is
linked to the plasma membrane, whereas most PLC
3 is in the cytosol.
PLC
1 and PLC
3 are predominantly in the cytosol of brain,
whereas PLC
2 is in the plasma membrane. The relationship between
membrane-bound and cytosolic PLC
is controversial (29-31). It is
not clear whether they are distinct enzymes or whether the same enzyme
is distributed between the two intracellular pools. It is generally
believed that only the membrane-bound enzyme is involved in
receptor-mediated phosphatidylinositol metabolism. Our finding that PLC
enzymes are distributed between soluble and membrane-bound
subcellular fractions in osteoblasts indicates that the membrane-bound
and cytosolic PLC
is the same enzyme. This is corroborated by our
immunocytochemical data showing that PLC
1 and
2 are mostly
membraneous, whereas PLC
3 is cytosolic in
osteoblasts.2 The differential distribution
of the enzymes may be a way of regulating enzyme activity that is
comparable to the regulation of protein kinase C (6, 32) rather than an
artifactual redistribution of the enzyme due to cell homogenization.
The presence of PLC
in osteoblasts is not surprising because these
osteogenic cells respond to growth factors that possess tyrosine kinase
domains (33).
We have previously shown that 17 estradiol and calcitriol rapidly
(within 5 s) increase the intracellular calcium concentration by
mobilizing Ca2+ from the endoplasmic reticulum (17, 18) and
by forming inositol 1,4,5-trisphosphate via a phospholipase C (17, 19,
20). Anti-PLC
1 and anti-PLC
2 antibodies inhibit the
steroid-induced increase in [Ca2+]i in much the
same way as do direct or indirect inhibitors of PLC (17, 19). Anti-PLC
antibodies, like PLC inhibitors (17, 19), block only the part of the
increase in [Ca2+]i that is due to the
mobilization of Ca2+ from the endoplasmic reticulum.
Moreover, the inhibition of the enzyme activity by anti-PLC antibodies
totally disappears in competitive experiments when polyclonal PLC
1
and PLC
2 antibodies and the antigens they were raised against are
used but not when the antigens corresponding to the other PLCs are
used. This type of enzyme inhibition by selective antibodies against
phosphoinositide-specific PLC has also been demonstrated in fresh
bovine erythrocytes (34). One explanation of the noninhibiting effects
of the other anti-PLC antibodies on the increase in
[Ca2+]i induced by the steroids may be that most
antibodies do not classically inhibit enzyme activity. On the other
hand, because the inhibition by the anti-PLC
1 and anti-PLC
2
antibodies can be reversed only by their antigens, the possible
blockade of the activity of the other isoenzymes by their respective
antibodies cannot be excluded. The inhibitory effect of polyclonal
antibodies against PLC
1 or PLC
2 may be best explained by one of
two hypotheses: (i) inhibition of the enzyme activity may result from a
stereochemical effect of the binding of the antibody to an epitope at
or near the active site of the enzyme; or (ii) alternatively, binding of the antibody to a site remote from the active site in
three-dimensional space may lead to a conformational change of the
enzyme, rendering it incapable of substrate binding or hydrolysis. It
is clear, however, that the anti-PLC
antibody binds to a site of
the enzyme that is critical to maintaining the correct geometry of the
active site. Further studies are needed to distinguish between these hypotheses, and it will be of interest to identify the site on the PLC
protein recognized by the inhibitory antibodies.
It is very likely that only PLC s are involved in the membrane
action of these steroids because only PLC
types may be regulated via heterotrimeric G-proteins in response to agonists binding to
receptors (1, 7, 8). This study shows that the PLC
1 involved in the
membrane action of calcitriol is linked to a Pertussis
toxin-insensitive G-protein, whereas the PLC
2 implicated in the
membrane effect of estradiol is coupled to a Pertussis toxin-sensitive
G-protein. The heterotrimeric G-proteins, which transduce a signal from
a hormone-bound receptor to a variety of downstream effectors, form a
large family of homologous proteins classified according to the amino
acid sequence of their
-subunits (35-37). G-proteins have been
divided into two types based on their sensitivity to the Bordella PTX
(7, 36). The PTX-sensitive G-proteins are inactivated by
ADP-ribosylation on the
-subunit and include the members of the
Gi and Go family. The PTX-insensitive G-proteins, which are resistant to ADP-ribosylation, belong to the
Gq class of G-proteins. The Gq class of
G-proteins, including G
q, G
11,
G
14, G
15, and G
16, can
activate all four PLC
isoforms (38-41), but receptor activation of
PLC
via G-proteins occurs by both Pertussis toxin-sensitive and
Pertussis toxin-insensitive signaling pathway. It is now clear that the
-subunits of the Gq family mediate the toxin-insensitive
pathway, but the nature of the G-proteins mediating the toxin-sensitive
pathway is not clear. There is no direct evidence that phospholipase C
can be activated by the
-subunits of Gi or
Go (42), but the recent discovery that G-protein
-subunits can activate PLC suggests an alternative mechanism
(41-44). Because calcitriol uses a Pertussis toxin-insensitive pathway
coupled to PLC
1 and estradiol uses a Pertussis toxin-sensitive
pathway linked to a PLC
2, the next step will be to describe the
G-proteins and the kind of subunits involved in the membrane action of
these steroids. Neither PLC
3 nor PLC
4 is involved in the signal
transduction. Similarly, PLCs
1 and
2 take no part in the
membrane effects of the two steroids as expected because PLC
enzymes are substrates for growth factor receptor protein-tyrosine
kinases (45).
Finally, these results, showing that both calcitriol and estradiol use
PLC to increase the intracellular calcium concentration via
inositol 1,4,5-trisphosphate formation (17, 19), may be an important
step toward understanding the membrane effects of these steroids. These
PLC
enzymes are also linked to two classes of G-proteins, one
insensitive to Pertussis toxin, as for calcitriol, the other linked to
a Pertussis toxin-sensitive G-protein, as for estradiol. These findings
may be an additional argument in favor of membrane receptors for
steroid hormones (46-52).
We gratefully acknowledge Dr. Owen Parkes for careful review of the English in the manuscript.