| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Articles |
Surgical Service, Veterans Administration Connecticut Healthcare System, and the Department of Surgery and Cell Biology, Yale University School of Medicine, New Haven, Connecticut 06250-8062
Address all correspondence and requests for reprints to: Lisa M. Matovcik, Ph.D., Department of Surgery, P.O. Box 208062, New Haven, Connecticut 06520-8062. E-mail: matovcik{at}aol.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
S) increased PTH
secretion from SLO-permeabilized cells in a dose-dependent manner from
0.1–100 µM. In the absence of GTPS there was no
relationship between the ambient Ca2+ concentration and the
rate of PTH secretion. However, in the presence of GTPS,
intracellular Ca2+ inhibited PTH secretion with an
EC50 of approximately 0.1 µM, corresponding
to physiological intracellular Ca2+ levels. Thus, the
addition of GTPS to SLO-permeabilized parathyroid cells
reconstituted the inverse relationship between extracellular
Ca2+ and PTH secretion that is observed in
vivo and in intact cells. The data indicate that this effect is
mediated at least in part by heterotrimeric guanosine triphosphatases.
In addition, calcium/calmodulin-dependent protein kinase II appears to
mediate low Ca2+-dependent PTH secretion from these cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In intact cells, individual steps of the pathways that lead to
exocytosis can be difficult to resolve. In permeabilized cell systems,
the barrier imposed by the plasma membrane is eliminated, and the
intracellular environment can be manipulated, facilitating the analysis
of cellular signal transduction mechanisms. The bacterial toxin
streptolysin-O (SLO) binds to cholesterol and causes the formation of
stable pores selectively in the plasma membrane (6). SLO has been
widely used to permeabilize cells, manipulate the composition of the
cytosol, and study the trafficking of intracellular organelles. Most
methods of cell permeabilization, such as electroporation and
-toxin, permit the exchange only of ions and small molecules.
However, the larger pores formed by SLO permit the entry of probes of
cellular function, such as peptides and antibodies. The
cytoarchitecture of SLO-permeabilized cells is largely preserved, and
intracellular organelles such as secretory granules are too large to
leak out of the cell (7). Thus, SLO-permeabilized cells provide a
useful model system to examine the regulation of hormone secretion.
Previous studies of permeabilized parathyroid cells have shown
that although some features of the intact cell remain, others, most
notably the ability to suppress PTH secretion in response to
physiological levels of Ca2+, were often lost. Studies
using electropermeabilized parathyroid cells to examine
Ca2+-dependent PTH secretion have yielded conflicting
results; Ca2+ had no effect (8), generated a biphasic
response (9), or was stimulatory at high Ca2+
(10-5-10-3 M) (10). Likewise,
guanine nucleotides and their analogs increased PTH secretion from
electropermeabilized parathyroid cells in some studies (10, 11) and had
no effect in another (9). In the present study we report that bovine
parathyroid cells permeabilized with SLO in the presence of guanosine
5'-O-(thio)triphosphate (GTPS) both retain the ability to
undergo inverse, Ca2+-dependent PTH secretion and allow the
entry of macromolecular probes of cellular function. This experimental
model system facilitates the direct examination of the factors that
regulate PTH secretion.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation of isolated parathyroid cells and measurement of PTH
secretion
Dispersed cells were isolated from calf parathyroids as
previously described (12, 13, 14). Briefly, parathyroid glands were trimmed
of connective tissue, minced to a fine slurry with scissors, placed
into a polypropylene tube, and digested by shaking with 0.024%
collagenase D and 0.004% deoxyribonuclease I in
Ca2+/Mg2+-free MEM (Eagle’s) supplemented with
0.5 mM CaCl2, 0.5 mM
MgSO4, 0.1 mM L-methionine, and 20
mM HEPES, pH 7.47. The resulting bovine parathyroid cells
were separated from the rest of the tissue by mechanical disruption and
filtration, then from debris, nonviable cells, and red cells on a 35%
(vol/vol) self-forming Percoll gradient. Secretion was monitored by
measuring intact PTH released into the medium by commercial double
antibody RIA in which the sample containing PTH is incubated
simultaneously with an antibody to the midregion/C-terminal fragment of
PTH (amino acids 39–84) immobilized on a bead and another radiolabeled
antibody to the N-terminal region 1–34. The assay is specific for the
biologically active and intact molecule (Allegro Intact PTH, Nichols
Diagnostics, San Juan Capistrano, CA).
SLO permeabilization
SLO was dissolved in water, stored in aliquots at -70 C, and
thawed no more than once. Dispersed bovine calf parathyroid cells
(1.5 x 105 in a volume of 30 µl) were added to 270
µl permeabilization buffer, consisting of 139 mM
potassium glutamate, 5.0 mM HEPES (pH 7.0), 2.0
mM ATP, 5.0 mM EGTA, 4.0 mM
MgSO4, and 0.4 IU/ml SLO. Sufficient CaCl2 was
added to result in the desired free Ca2+ concentration at
pH 7.00 (15).
Preparation of parotid cytosol
Parotid glands were removed from Sprague-Dawley rats and trimmed
of blood vessels and connective tissue. The parotids were homogenized
using six strokes by hand with a ground glass Dounce homogenizer at 5%
(wt/vol) in 0.3 M sucrose, 10 mM Tris (pH 7.6),
2.0 mM EDTA, 1.0 mM dithiothreitol, 5 µg/ml
leupeptin, aprotinin, and antipain, and 2.5 µg/ml pepstatin and
chymostatin. This homogenate was centrifuged at 600 x
g for 10 min at 4 C, and the resulting postnuclear
supernatant was centrifuged at 180,000 x g for 10 min
at 4 C. The supernatant is parotid cytosol.
Transmission electron microscopy
Bovine parathyroid glands were cut into approximately 5-mm
cubes; isolated dispersed parathyroid cells were pelleted at 600
x g. Both were fixed overnight in 3% glutaraldehyde and
0.1 M sodium cacodylate, pH 7.4, then postfixed for 1
h with 1% osmium tetroxide and 0.1 M sodium cacodylate, pH
7.4. After Epon embedding, 80-nm sections were cut and stained with
10% uranyl acetate in 25% ethanol and lead citrate (80 mM
lead nitrate, 120 mM sodium citrate in 0.15 N
NaOH), then examined with a Philips 300 electron microscope (Philips
Electronics, Rahway, NJ).
Other methods
The total Ca2+ concentration in the secretion medium
was measured photometrically using arsenazo III as a calcium-binding
agent (16) by the clinical chemistry laboratory, Yale University School
of Medicine. Lactate dehydrogenase (LDH) activity was measured
according to manufacturer’s instructions, using a quantitative
colorimetric assay (Sigma procedure 500), in which pyruvic acid is
converted to lactic acid in the presence of NADH. In the presence of
2,4-dinitrophenylhydrazine, residual pyruvate produces a colored
phenylhydrazone that is inversely proportional to LDH activity. ATP
levels were determined by the quantitative bioluminescent measurement
of light emitted by ATP-dependent luciferase catalysis of the oxidation
of D-luciferin (Sigma FL-AA). A Beckman LS6000IC
scintillation counter equipped with a single photon monitor (Beckman,
Fullerton, CA) was used to measure light emission.
Statistical significance was calculated using the paired t
test (Figs. 4
and 5B) or ANOVA (Figs. 5A, 7A, and 8). A four-parameter
model of the sigmoidal relationship between PTH secretion and
Ca2+ concentration in intact (Fig. 1) and permeabilized
(Fig. 7B) cells was performed using the Systat program, based on the
model described by E. M. Brown (3). In this model, A is the maximal
value of PTH secretion, B is the slope of the curve at the midpoint, C
is the midpoint, and D is the minimum value of PTH secretion. The
values were calculated based on the equation (17): Y = {(A
- D)/[1 + (X/C)B]} + D.
|
|
|
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
100 mg
each) and yielded 1.2 x 107 cells. To test the
sensitivity of these isolated dispersed cells to
Ca2+e, they were incubated in medium containing
0.3–3.9 mM Ca2+, and the amount of intact PTH
(iPTH) secreted into the medium was measured. As shown in Fig. 1
, PTH release has a sigmoidal relationship to
Ca2+e. The resulting curve can be defined by
four parameters (3), A is the percent maximum PTH release, B is the
slope of the curve at the set-point, C is the set-point
(Ca2+ concentration at which half-maximal inhibition of
secretion is observed), and D is the percent minimum PTH release. In
this preparation of intact dispersed cells, A = 93.03, B =
14.51, C = 1.44, and D = 37.55, similar to the findings of
Brown (3), who observed values of A = 100, B = 11.9, C =
1.27, and D = 32 in dispersed bovine calf parathyroid cells, and
with those of Habener and Potts (18) and Shoback et al.
(19), who observed set-points of 1.4 and 1.3 mM
Ca2+, respectively, in bovine calf parathyroid tissue
slices and bovine calf dispersed parathyroid cells.
Parathyroid cells treated with SLO are permeable
Uptake of the dye trypan blue (mol wt, 961) was measured to assess
the permeability of parathyroid cells treated with SLO to small
molecules. After permeabilization with 0.4 IU/ml SLO for 10 min at 37
C, greater than 90% of the cells were stained with trypan blue. Uptake
was observed in 2–8% of intact unpermeabilized cells, demonstrating
that 92–98% of the cells were viable.
The permeability of SLO-permeabilized cells to large molecules was assessed by the degree of leakage of the cytosolic 140-kDa protein LDH. After permeabilization with 0.4 IU/ml SLO for 10 min at 37 C, cells were separated from their medium by centrifugation at 100 x g for 10 min. The amount of LDH released into the medium was compared to that present in the cells. Before incubation in SLO-containing permeabilization buffer, no LDH was released into the medium. After 5 min, more than 50%, and after 10 min, more than 70% of the total cellular LDH were released into the medium. No LDH was detectable in the medium after a 10-min incubation in the absence of SLO. Thus, bovine parathyroid cells are rendered permeable to both small and large molecules by SLO, as assessed by the criteria of trypan blue uptake and LDH release.
The ultrastructural appearance of permeabilized parathyroid cells
was examined by transmission electron microscopy and compared to that
of the intact gland (Fig. 2). In the intact gland (Fig. 2A), the cisternae of the endoplasmic reticulum (er) were arranged in
parallel, and the ribosomes were attached. The mitochondria (m)
appeared cylindrical, and the secretory granules (sg) and plasma
membrane (arrowheads) were intact. In the dispersed intact
cells (Fig. 2, B and C), the endoplasmic reticulum was disrupted, but
the ribosomes remained attached for the most part, the mitochondria
were somewhat swollen, and the cristae were disrupted; the secretory
granules and plasma membrane remained intact. In the permeabilized
cells (Fig. 2, D and E), the secretory granules and some Golgi stacks
(g) remained intact. There was considerable vacuole formation in both
the dispersed and the permeabilized cells; vacuoles were larger and
more abundant in the SLO-treated cells, and mitochondrial disruption
was more pronounced in some instances. The plasma membrane remained
grossly intact in the permeabilized cells; SLO-induced pores have been
described to be about 100 Å in diameter and are not visible in Fig. 2E.
|
), is identical to that used to permeabilize
pancreatic acinar cells and examine regulated amylase secretion (15, 20). Identical rates of PTH secretion were observed from cells
permeabilized for 10 min at 37 C or for 10 min on ice followed by
incubation in the potassium glutamate permeabilization buffer at 37 C
for 10 min in the absence of SLO. A higher concentration of SLO was
required to attain maximum release at 0 C (2.0 IU/ml) than at 37 C (0.4
IU/ml).
|
). Three independent experiments
were performed in which more than 93% of maximum PTH release was
reached by 20 min. No further secretory activity was observed after
this time. No significant loss of ATP from the permeabilized cells was
detected after 60 min of secretion, thus, the availability of ATP is
not the limiting factor in sustaining secretion. Based on the above
findings, a 10-min incubation was chosen for subsequent
experiments. The secretory rate of eucalcemic permeabilized cells was faster than but comparable to that of intact parathyroid cells. SLO-permeabilized cells secreted 63 ± 31 (n = 51) pg iPTH/105 cells in 10 min at 37 C when incubated in 10-7 M Ca2+, an approximation of the [Ca2+i]. When intact cells were incubated at their set-point of 1.4 mM Ca2+ for 1 h, 61 ± 26 pg iPTH/105 cells (n = 8) were secreted; intact cells secreted PTH in a linear manner during the first hour of incubation.
Energy dependence of PTH release in SLO-permeabilized
parathyroid cells
To determine whether PTH secretion from permeabilized
parathyroid cells is energy dependent, cells were metabolically
depleted by incubation for 30 min with 10 µM antimycin A
and 6 mM 2-deoxyglucose. PTH secretion from permeabilized
depleted cells was less than 10% of that from undepleted cells. After
recovery in the absence of antimycin A and 2-deoxyglucose, secretion
from permeabilized cells returned to more than 70% of that from
undepleted cells.
Parotid cytosol stimulates PTH secretion from
SLO-permeabilized parathyroid cells
To determine whether there are cytosolic cofactors necessary
for the regulated secretion of PTH, rat parotid cytosol was added upon
permeabilization (Fig. 4). Parotid cytosol resulted in a
2-fold increase in PTH secretion from the SLO-permeabilized parathyroid
cells; in the presence of cytosol, secretion was 205 ± 19% of
that in the absence of cytosol (P = 0.01). However, the
addition of cytosol did not replace a soluble factor(s) necessary to
confer Ca2+ sensitivity to the permeabilized cells; changes
in the ambient Ca2+ concentration over the range of
10-9-10-5 M had no effect on
cytosol-induced secretion (data not shown). Boiling the parotid cytosol
for 3 min eliminated its ability to stimulate secretion; however,
heating it at 56 C for 30 min had no effect on secretion. Parathyroid
cytosol was not used in these studies because it contains a small
amount of PTH that is detected in the radioimmunometric secretion
assay; this PTH is most likely liberated from secretory granules or the
biosynthetic pathway upon preparation of the cytosol. No PTH
immunoreactivity was detected in parotid cytosol by radioimmunometric
assay.
SLO-permeabilized parathyroid cells do not secrete in response to
cAMP
Although agents that increase intracellular cAMP are secretagogues
in intact cells (21), the addition of cAMP to permeabilized parathyroid
cells did not result in increased PTH secretion. At physiological
intracellular Ca2+ (10-7 M), there
was no correlation between the cAMP concentration
(10-8-10-3 M) and PTH secretion
(r = 0.08; Fig. 5). Similarly, cAMP had no effect
on PTH secretion from permeabilized cells at either 10-8
or 10-3 M Ca2+ (data not shown).
The phosphodiesterase inhibitor 3-isobutyl-1-methylxanthene (1.0
mM) had no effect when added in combination with cAMP. The
addition of cAMP-dependent protein kinase catalytic subunit to
SLO-permeabilized cells increased secretion by about 30%, but this
effect was small compared to the severalfold increase in secretion
induced by agents that increase cAMP in intact cells.
Guanine nucleotides stimulate PTH secretion
Poorly hydrolyzable GTP analogs, such as GTPS, stimulate
regulated exocytosis from a number of different permeabilized cell
types, including the parathyroid (22, 23). The addition of either
GTPS or GppNHp to the permeabilization buffer resulted in a
dose-dependent increase in PTH secretion from parathyroid cells (Fig. 6). The maximally effective concentration of GTPS was
100 µM. The maximally effective concentration of GppNHp
was 10 µM. Addition of GTP in the same manner (100
µM) had no effect on PTH secretion, suggesting that a
stable GTP analog is required. Likewise, permeabilization in the
presence of GDPßS (100 µM) had no effect on PTH
secretion at 10-9, 10-7, or 10-5
M Ca2+ (data not shown). Simultaneous addition
of 100 µM GDPßS had no effect on the secretion induced
by 10 µM GTPS. This finding is consistent with the
previous observation by Oetting at al. that GDPßS had no effect on
secretion from electropermeabilized parathyroid cells when added alone,
but differs from this previous work in which it did inhibit the
secretion induced by GppNHp (10, 22). Addition of 100 µM
GTP to the permeabilization buffer did not alter the free calcium
concentration (24).
|
S reconstitutes calcium-dependent suppression of PTH release
from permeabilized parathyroid cellsS there was no relationship between the
Ca2+ concentration (10-8-10-3
M) and PTH secretion in SLO-permeabilized parathyroid cells
(r = 0.20; Fig. 7A). In the presence of
10-4 M GTPS, however, there was an inverse
relationship between PTH release and the ambient Ca2+
concentration (Fig. 7B). The resulting sigmoidal curve could be
accurately described using the same four-parameter model (3) that was
applied to the intact cells (Fig. 1), where A, the percent maximal
secretion, = 97.88; B, the slope of the curve at the midpoint, = 13.29;
C, the set-point, = -10-7.12; and D, the percent minimal
secretion, = 53.86. The set-point of SLO-permeabilized cells
corresponded to physiological intracellular Ca2+ levels
(10-7 M). The slopes of the curves
generated by the intact and the permeabilized cell preparations were
similar. GTPS conferred Ca2+ dependence on PTH secretion
from SLO-permeabilized parathyroid cells and restored the inverse
relationship between [Ca2+]e and PTH
secretion that was observed in vivo and in intact cells.
The addition of parotid cytosol did not induce a further increase in
the maximum amount of secretion observed in the presence of GTPS.
PTH secretion at 10-7 M Ca2+ in
the presence of 100 µM GTPS and 30 µg cytosol was
99 ± 4% of that in the absence of cytosol (n = 3,
experiments performed in duplicate). The parotid cytosol did not
reconstitute the inverse dependence of PTH secretion on
Ca2+ in the absence of GTPS. An unanticipated finding
occurred; GppNHp (100 µM) did not confer Ca2+
dependence to PTH secretion in this cell model. There was no
correlation between secretion and Ca2+ concentration
(10-10-10-4 M; n = 4; data
not shown).
Aluminum fluoride increases PTH secretion from permeabilized
cells
To examine the role of heterotrimeric GTP-binding proteins in this
system, parathyroid cells permeabilized with SLO were treated with
AlF4- to selectively activate heterotrimeric and not small
G proteins. In six of seven experiments performed, AlF4-
increased PTH secretion (P < 0.01; n = 6; Fig. 8); in two of these experiments the increase was large,
and in four the increase was modest. AlF4- did not further
increase PTH secretion induced by GTPS (not shown).
Calcium/calmodulin-dependent protein kinase II (CaM kinase II)
mediates PTH secretion from permeabilized cells
CaM kinase II, abundant in the parathyroid gland (25, 26), is
activated by increased Ca2+ and has been implicated in
regulating secretion from nerve terminals (27). To examine the role of
CaM kinase II in PTH secretion, parathyroid cells permeabilized with
SLO were treated with the CaM kinase II inhibitor CKIP, a synthetic
peptide that corresponds to amino acids 281–302 of the -subunit in
which alanine is substituted for threonine 286 and inhibits CaM kinase
II with a Ki of approximately 5 µM (27). In
the presence of 100 µM GTPS, CKIP inhibited PTH
secretion in a dose-dependent manner at low ambient Ca2+
(10-9 M; Fig. 9A); 30–50%
inhibition was observed at 30 µM CKIP. CKIP had no effect
on secretion at high ambient Ca2+ (10-5
M). In a preliminary experiment shown in Fig. 9B, KN62, a
membrane-impermeant CaM kinase II inhibitor that blocks calmodulin
binding, inhibited PTH secretion at 10-9 M
Ca2+ to approximately the same extent as CKIP.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-toxin
or SLO (28). Ions, nucleotides, and other molecules smaller than
1000 daltons can pass through pores formed by
electropermeabilization or -toxin, whereas larger molecules (up to
140 kDa) can permeate pores formed by SLO. SLO binds to cholesterol and introduces stable pores selectively into the plasma membrane (7). Regulated secretion is maintained in many types of secretory cells permeabilized by SLO (6). For example, in the pancreas, investigations of individual steps in the pathways that lead to exocytosis show that Ca2+-dependent amylase secretion occurs with an EC50 similar to that of intact cells and is modulated by protein kinase C and a G protein(s) (20). In the SLO-permeabilized mast cell, GTP-dependent hexosaminidase secretion is enhanced by Ca2+ (29). Thus, SLO-permeabilized cells can provide a useful model system to examine the regulation of exocytosis.
Several studies have been performed to examine
Ca2+-dependent PTH secretion from electropermeabilized
parathyroid cells. An early study reported that PTH secretion was not
responsive to Ca2+ concentration, but could be induced in
response to protein kinase A activation by cAMP and to protein kinase C
activation by phorbol ester (8). Subsequent studies reported that
electropermeabilized parathyroid cells increased PTH secretion in
direct (not inverse) proportion to Ca2+, but only at very
high levels (1 µM to 1 mM); no effect was
observed at 1 nM to 1 µM Ca2+
(10). Pocotte et al. observed a biphasic response to
Ca2+, with peak secretion occurring near the physiological
concentration (10-7 M) (9, 30). Our results
showed no dependence of PTH secretion on Ca2+ concentration
in SLO-permeabilized cells unless a stable guanine nucleotide was
added, and then secretion changed in inverse relation to
Ca2+. These findings suggest that GTP-mediated pathways
play a role in modulating PTH secretion in response to the
Ca2+i concentration.
There are limitations to the ability of a permeabilized cell model to examine the role of second messenger systems that rely on the structural integrity of intracellular organelles and their organization. The minimum level of PTH secretion is higher than that in the intact cells, reflecting a loss of the ability of the cell to suppress secretion in response to high Ca2+ levels. A similar loss has been observed when dispersed cells are compared to the intact animal; cells in vitro have a higher level of Ca2+-insuppressible secretion than parathyroid glands in vivo (3). The SLO-permeabilized cells described above did not exhibit cAMP-dependent PTH secretion, confirming findings from the majority of studies in electropermeabilized parathyroid cells (10, 11, 22). With the exception of one study (8), permeabilized parathyroid cells have not been found to retain the capacity of the intact cell to secrete PTH in response to cAMP. This loss of sensitivity to cAMP, and to the cAMP-dependent protein kinase catalytic subunit, suggests that soluble downstream effectors are lost or that a scaffolding complex with the cAMP-dependent protein kinase anchoring protein(s) necessary for the correct localization and function of the enzyme was not intact (31).
The first evidence that G proteins are involved in vesicular transport
was the observation that addition of GTPS to permeabilized mast
cells was sufficient to induce their degranulation (32). Subsequently,
both large heterotrimeric G proteins and small G proteins of the
rab family have been implicated in nearly every step of
transport along the exocytic pathway in a variety of cell types, and
the roles of specific G proteins are now being identified (23). GTP
analogs stimulate exocytosis in many kinds of secretory cells, such as
parathyroid cells, pancreatic exocrine cells, mast cells, and the
insulinoma cell line RINm5F. In adrenal chromaffin cells, GTP analogs
potentiate Ca2+-induced secretion via a protein kinase C
pathway; when PKC is inhibited, the GTP analogs directly inhibit
exocytosis (33). Recently, a complex of the small G protein
rac complexed to the guanine nucleotide exchange inhibitor
rho (GDI) has been identified as a factor that modulates
secretion from SLO-permeabilized mast cells (34).
It was recognized a decade ago that G proteins are necessary for
Ca2+-dependent suppression of PTH secretion. The
hydrolysis-resistant guanine nucleotide analog GppNHp stimulated PTH
secretion in electropermeabilized cells via a mechanism independent of
phosphoinositide hydrolysis or cAMP accumulation (11). Inhibition of
Gi with pertussis toxin abolished the inhibitory effects
of agents that inhibit cAMP production in the intact cell and blocked
the suppression of PTH secretion by high Ca2+ (35).
Recently, several G proteins have been described in the bovine
parathyroid; messenger RNA for heterotrimeric Gi
subunits belonging to each of the four major functional classes have
been identified (36), and the small G protein rabphilin 3 (Rab 3) has
been identified by two-dimensional immunoblot (37). Our preliminary
work suggests that the addition of a synthetic peptide corresponding to
the effector domain of Rab 3A also increases secretion (Kinder, B. K.,
and L. M. Matovcik, unpublished data). Thus, both large and small G
proteins are likely to be involved in PTH secretion, either at
different steps in the secretory pathway or by release from separate
pools of secretory granules. The mechanism by which either large or
small G proteins mediates the suppression of PTH secretion in response
to an increase in Ca2+e remains largely
undefined, however. An approach that has been used to implicate G
proteins in cellular processes in other cell types is the addition of
peptides or antibodies with known specificity to SLO-permeabilized
cells. The development of a parathyroid cell model that undergoes
regulated PTH secretion facilitates this approach.
Just as past studies of electropermeabilized parathyroid cells have
differed in their findings with respect to the relationship between
calcium and PTH secretion, they also differ in the observed effects of
GTP analogs. In one study, GTPS had no effect on PTH secretion from
electropermeabilized adult bovine parathyroid cells (9). In another
study, GppNHp acted as a potent secretagogue for calf
electropermeabilized parathyroid cells, increasing PTH secretion 7-fold
and, notably, altering the Ca2+ responsiveness of the cells
so that they secrete less PTH as Ca2+ rises from
10-7 to 10-4 M (22). The site of
action of the guanine nucleotide was later determined to be distal to
cAMP accumulation and phosphoinositide metabolism (11). In the present
study we unexpectedly found that although GppNHp was somewhat more
potent in inducing PTH secretion than GTPS, it did not reconstitute
the inverse Ca2+ dependence of the intact cell. Structural
differences between the imido- and the thio-phosphorylated GTP analogs
may confer different affinities for selective G proteins and result in
different effects. Differential effects of GTP analogs on secretion
have been observed; for example, in SLO-permeabilized adrenal
chromaffin cells, basal levels of catecholamine exocytosis are elevated
by GppNHp, but not GTPS (38). Structural differences between the
imido- and the thio-phosphorylated GTP analogs may confer different
affinities for selective G proteins and result in different
effects.
Another unexpected observation in this study was that
AlF4-, an orthophosphate analog that binds to and
activates heterotrimeric, but not monomeric, G proteins (38), activated
PTH secretion from SLO-permeabilized parathyroid cells. In intact
cells, fluoride stimulates the accumulation of inositol trisphosphate,
increases Ca2+i, and inhibits PTH secretion
(40, 41). There are several possible reasons for this apparent
discrepancy. AlF4- diffusing into intact cells may first
come into contact with and activate G proteins at the plasma membrane
that are linked to the Ca2+ receptor, increasing
intracellular Ca2+. There is evidence that some of the
-subunits of heterotrimeric G proteins partially leak out of the
SLO-permeabilized cell (39); thus, the predominant action of
AlF4- may be activation of a more distal G protein that
couples Ca2+i to secretion. Another possibility
is that AlF4- is acting not on GTP but, rather, on
ATP-dependent pathways and binding to cytoskeletal proteins,
i.e. actin, myosin, or tubulin, and is disrupting secretory
granule interactions with the cytoskeleton (42). Finally,
AlF4- may be inhibiting protein phosphatases. The fluoride
ion, via its property as a phosphate analog, is a nonspecific inhibitor
of most classes of protein phosphatases. Phosphatase inhibition has
been reported to increase secretion from some cell types and decrease
secretion from others.
Another potential mediator of calcium signaling in the parathyroid is CaM kinase II, a ubiquitous multifunctional effector of Ca2+-dependent processes (43). At least 11 mammalian isoforms have been identified at the RNA level; they are found in different ratios as a holoenzyme complex in a tissue-specific manner (43). A 550-kDa holoenzyme complex composed of 50-kDa monomers is both abundant and active in parathyroid cytosol (25, 26). Many proteins, including but not limited to ones involved in cytoskeletal function, intermediary metabolism, and ion transport have been identified as in vitro or in vivo substrates for CaM kinase II (43). Phosphorylation of synapsin I releases an inhibitory constraint on neurotransmitter vesicles at the active zone of the nerve terminals, demonstrating a direct involvement of CaM kinase II in the secretion of neurotransmitters (27).
CaM kinase II is a sensor of complex patterns of changes in intracellular Ca2+. Under conditions of low Ca2+, its known isoforms have no activity in the basal state, but can be activated by calmodulin. Under conditions of high Ca2+ the enzyme is autophosphorylated and can remain autonomously active, i.e. Ca2+ independent, after the Ca2+ signal has ended (43, 44). As inhibition of the kinase at low Ca2+ blocked PTH secretion, the parathyroid form of the enzyme may be very sensitive to Ca2+, remaining active at low Ca2+. At high Ca2+, inhibiting the kinase had no effect on secretion; therefore, other Ca2+-dependent processes, i.e. hyperautophosphorylation or activation of calcineurin, may eclipse its effects (45). Alternatively, parathyroid CaM kinase II may be an uncharacterized isoform that is active at low Ca2+i and suppressed at high Ca2+i.
There are many possible substrates for CaM kinase II in the parathyroid that could be involved in the regulation of PTH secretion or PTH degradation. The cytoplasmic C-terminus of the bovine calcium receptor has consensus sequences for CaM kinase II phosphorylation (R/K x X S/T) at T877, S901, S903, and T1000 (1). It has recently been reported that rabphilin 3A is an efficient substrate for CaM kinase II (46). Rab 3A, a Ca2+-binding protein found on the synaptic vesicle and in neuroendocrine cells, binds to Rab 3A in a GTP-dependent manner. Rabphilin stimulates the GDP/GTP exchange of Rab 3A and inhibits the Rab 3A guanosine triphosphatase-activating protein-stimulated guanosine triphosphatase activity, with the net result of maintaining Rab 3A in the GTP-bound form (47). Thus, calcium-dependent phosphorylation of rabphilin could regulate (either positively or negatively) a GTP-dependent exocytic event via a parathyroid Rab 3-like protein. There are many other ways that Ca2+i might modulate GTP-dependent pathways involved in the exocytosis of PTH; the SLO-permeabilized parathyroid cell model should prove useful in defining them.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received October 3, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-toxin and streptolysin-O: an approach to analyze
intracellular processes. Methods Cell Biol 31:63–90[Medline]
and -adrenergic agonists regulate
parathyroid cell function via the inhibitory guanine nucleotide
regulatory protein. Endocrinology 118:2115–2119[Abstract]
-subunits in
bovine parathyroid. Endocrinology 136:4390–4396[Abstract]
-thio]triphosphate and guanosine
5'-[ß-imido]triphosphate. Biochem J 114:321–326
This article has been cited by other articles:
 |
|
 |
|
  R. D. Burgoyne and A. Morgan Secretory Granule Exocytosis Physiol Rev, April 1, 2003; 83(2): 581 - 632. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
