|
|
|
|||||||
|
|||||||||
(Received for publication, March 9, 1995; and in revised form, May 23, 1995 ) From the
A cytoskeletal fraction of porcine tracheal smooth muscle (PTSM)
was found to contain >90% of total cellular aldolase (fructose
1,6-bisphosphate aldolase, EC 4.1.2.13) activity. PTSM aldolase was
purified by DEAE and inositol 1,4,5-trisphosphate
(Ins(1,4,5)P
Many reports have demonstrated the importance of inositol
1,4,5-trisphosphate (Ins(1,4,5)P In a previous
study(11) , we obtained evidence that there may be a bound or
sequestered Ins(1,4,5)P Ins(1,4,5)P In the present study we developed and studied a
cytoskeleton fraction in PTSM. Data were obtained which indicate that
aldolase is almost entirely associated with the cytoskeleton in this
muscle. We purified aldolase recovered from this fraction and showed
that it is aldolase C and that it binds and releases Ins(1,4,5)P
The equilibrium dialysis assay conditions
were the same as in the PEG assay except that aldolase and
[
Figure 1:
SDS electrophoresis of K-released and
ghost proteins. Aliquots (20 µg of protein) of either K-released or
ghost proteins were separated by SDS-PAGE and the gels stained with
Coomassie Brilliant Blue. R
Figure 2:
Western blots. A, study of
K-released proteins using to antibodies to vinculin,
Since it is possible
that high tonicity could break bonds between aldolase and plasma
membrane components, we solubilized ghosts using the detergents Triton
X-100 (0.5%) or octyl glucoside (30 mM). This treatment did
not release aldolase activity into the supernatant nor change release
of aldolase evoked by a subsequent increase in KCl. We concluded that
particulate aldolase was released from the cytoskeleton during 0.6 M KCl wash of ghosts and not from the plasma membrane. These
data were similar to those reported in other tissues (27, 35) . Assay of aldolase activity in K-released
proteins was entirely dependent on the presence of added GAPDH.
However, with ghosts aldolase activity in the entire absence of added
GAPDH was as large as 60% of that in the presence of added GAPDH. This
indicates that endogenous GAPDH and aldolase are associated with the
intact ghosts.
Figure 3:
Chromatography of aldolase on
Ins(1,4,5)P
Figure 4:
SDS-gel electrophoresis of purified PTSM
aldolase and Western blot against anti aldolase C antibody.
Densitometric tracings (aligned at the front) of silver-stained
aldolase (solid line) (1 µg) following SDS-PAGE and a
Western blot run against anti-aldolase C (dotted line; blot is
shown Fig. 2B) as described under ``Experimental
Procedures.'' The upper x axis was determined from the
molecular weights of standard proteins.
An estimate of the
molecular weight of native aldolase was obtained by Sephacryl S-300 HR
chromatography of K-released proteins in standard buffer containing 0.6 M KCl. Aldolase activity and Ins(1,4,5)P
The total aldolase content in the tissue was computed by measurement
of aldolase activity in the entire crude fraction and the specific
activity of purified smooth muscle aldolase. This assumes aldolase
specific activity is constant throughout the tissue and is the same as
the specific activity of purified aldolase. This calculation is given
under ``Appendix.'' Total aldolase content was computed to be
0.52 µg/mg cellular protein. The total content of the aldolase
tetramer (aldolase
Fig. 5shows the
effects of altering pH on purified aldolase activity. These data were
obtained using [Fru(1,6)P
Figure 5:
Effect of pH on aldolase activity and
Ins(1,4,5)P
Figure 6:
Binding
of Ins(1,4,5)P
We were not able to saturate Ins(1,4,5)P
where K We also used a Scatchard analysis
(Klotz plot multiplied by
[aldolase Thus, considering the above, we
conclude that the K
Figure 7:
Inhibition of Ins(1,4,5)P
Fig. 7also shows results of
experiments which aimed at determining if altering
[Ca
Figure 8:
Displacement of bound
[
Major findings in this study relate to smooth muscle aldolase
and Ins(1,4,5)P The finding that virtually all
cellular aldolase was found in the K-released fraction had practical
importance in that the enzyme was already 17-fold purified in this
fraction, when related to total cellular proteins (Table 2). The
apparent molecular mass of native aldolase was 138 kDa while
SDS-denatured enzyme was 35 kDa indicating that there are four aldolase
monomers which associate as a tetramer. The K Immunoblot data indicate the presence
of an aldolase C isoform, and we could not detect reaction with
aldolase A and B antibodies. The presence of aldolase C was supported
by determining the ratio of PTSM aldolase activities driven by
Fru(1,6)P Strong evidence was obtained that
Ins(1,4,5)P The sharp loss of
Ins(1,4,5)P The maximal
molar binding of 1/tetramer bound in PTSM contrasts with molar binding
of Ins(1,4,5)P The question of
whether or not Ins(1,4,5)P It was shown here that Ins(1,3,4)P
Total cellular aldolase
activity/mg of total cellular protein (Ald
mg aldolase/mg cellular total protein was computed from the
specific activity of purified aldolase (SA
Volume 270,
Number 35,
Issue of September 01, pp. 20459-20465, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
APPENDIX
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
) affinity chromatography and found to react
with an antibody directed against human aldolase C, but not
anti-aldolase A and B. The molecular mass of native aldolase was about
138 kDa (on Sephacryl S-300); SDS-denatured enzyme was 35 kDa
(comigrated with rabbit skeletal muscle aldolase). Total cellular
aldolase tetramer (aldolase
) content was 34.5 pmol/100 nmol
lipid P
. Ins(1,4,5)P) binding activity coeluted
with aldolase during Sephacryl 300, DEAE, and Ins(1,4,5)P affinity chromatography. Ins(1,4,5)P bound to
purified aldolase (at 0 °C) in a dose-dependent manner over the
range [Ins(1,4,5)P] 20 nM to 20
µM, with maximal binding of 1 mol of
Ins(1,4,5)P/mol aldolase and a K of 12-14 µM.
Fru(1,6)P
and Fru(2,6)P displaced bound
Ins(1,4,5)P) with a 50% inhibition at 30 and 170
µM, respectively. Ins(1,3,4)P (20
µM) and glyceraldehyde 3-phosphate (2 mM) were
also potent inhibitors of Ins(1,4,5)P binding, but not
inositol 4-phosphate or inositol 1,4-bisphosphate (20 µM each). Aldolase-bound Ins(1,4,5)P may play a role in
phospholipase C-independent increases in free
[Ins(1,4,5)P].
)(
)-evoked
Ca release from endoplasmic reticulum (ER) in various
non-muscle cells(1, 2) . Although a role of
Ins(1,4,5)P
and sarcoplasmic reticulum (SR) Ca release is established for smooth
muscle(3, 4, 5) , it is not proven that
Ins(1,4,5)P
can be formed rapidly enough from
phosphatidylinositol 4,5-bisphosphate (PIP) during receptor
activation to be involved in initial increases in free
[Ca] which drive force development. This
requires that increases in [Ins(1,4,5)P
] to levels which release SR Ca
occur within a
few 100 ms of receptor activation of the muscle(5) . We did not
detect any rapid or sustained increases in total Ins(1,4,5)P
content in porcine tracheal smooth muscle (PTSM) during
muscarinic receptor-evoked force development. However, phospholipase C
was activated as determined by measurements of flux in inositol
phospholipids and increases in Ins(1,4,5)P by-products(6) . In other smooth muscles, transient
increases in Ins(1,4,5)P have been measured for a few
seconds following receptor
activation(7, 8, 9) . In a vascular smooth
muscle, large stretch-activated increases in Ins(1,4,5)P content occur within 200-300 ms after stretch to L
(10) . store in PTSM. The
Ins(1,4,5)P content in resting, unstimulated muscle
averaged 2.7 µM (assuming uniform mixing in cellular
water), a value much higher than that required to release ER and SR
Ca(12, 13) . (High values of resting
Ins(1,4,5)P
contents have been reported for other tissues,
as well(14, 15, 16) .) Ins(1,4,5)P content decreased during atropine-induced relaxation of
carbachol-contracted PTSM to levels about 60% of the unstimulated
level. This suggests that a portion of the sequestered store is
released during the carbachol-stimulated contraction and is not
refilled during an atropine-induced relaxation. can bind to membranes via Ins(1,4,5)P-receptors (17) and to cytosolic proteins (18, 19, 20) ; however, it is unknown if this
binding could explain the high unstimulated Ins(1,4,5)P content found in PTSM. Ins(1,4,5)P can bind to
aldolase (fructose 1,6-bisphosphate aldolase, EC 4.1.2.13) A (skeletal
muscle) and B (liver); the substrate for the enzyme, fructose
1,6-bisphosphate (Fru(1,6)P), can reduce the affinity of
aldolase A and B for Ins(1,4,5)P(18) . Hirata and
Kanematsu (19, 20, 21) have determined that
in rat brain, Ins(1,4,5)P binds to cytosolic phospholipase
C and an unidentified 130-kDa protein. The total bound
Ins(1,4,5)P
store in brain has not yet been determined, and
it is unknown if, in brain cells, metabolic events release bound
Ins(1,4,5)P which could then exert effects on cellular
function. under conditions which may occur physiologically. Total cellular
aldolase content was quantitated and shown to be very large suggesting
there is a potential site for binding large amounts of
Ins(1,4,5)P.
Tissue Preparation
Porcine tracheal smooth
muscle strips were prepared as described previously(22) .
Unstimulated strips were incubated in Krebs solution at 37 °C for 1
h and then freeze-clamped with liquid N-cooled tongs. The
frozen tissue was ground to a fine powder in a liquid
N-cooled mortar and pestle and stored at -70 °C
until used.Fractionation of Tissue
The method described below
was adapted from that described by Jones et al.(23) ,
Caroni and Carafoli (24) and Popescu and Ignat(25) ;
this preparation was originally used to prepare muscle sarcolemma.Preparation of Ghosts and Cytosol
Frozen, powdered
tissue was thawed and dispersed by slow addition to 6 ml of
homogenization buffer: 20 mM HEPES-Na, pH 7.3, 250 mM sucrose, 2 mM EGTA, 0.5 mM MgCl (0.45 mM free), 1 mM phenylmethylsulfonyl
fluoride, 1 mM benzamidine, 1 µM calpain
inhibitor peptide, and 5 µg/ml each of aproptinin, leupeptin, and
pepstatin A. All steps in this preparation were performed at 0 °C.
The suspension was homogenized in a 7-ml Tenbroeck homogenizer
(Wheaton, Wheaton, NJ) and centrifuged for 5 min at 200 g. The supernatant was collected, the pellet resuspended in
homogenization buffer and the above procedure repeated five more times.
The pooled supernatant from multiple washes was centrifuged two times
(1,000
g for 20 min; 100,000
g for 1
h). The 100,000
g supernatant is the
``cytosolic'' fraction. The pellet obtained from the 1,000
g centrifugation was added to the 200
g pellet (resulting in a 10% increase in total aldolase activity).
This fraction has been termed the ``ghost'' fraction, so
named because of its microscopic appearance(25) .
Preparation of K-released Proteins
Ghost
suspensions were homogenized in 6 ml of release buffer: 20 mM
HEPES-Na, pH 7.3, and 0.6 M KCl, 2 mM EGTA, and
centrifuged for 20 min at 1,000 g. The pellet was
rehomogenized and centrifuged. The resulting pellet is the sarcolemmal
fraction studied previously (24, 25) but not used
extensively in the present study. Pooled supernatants were centrifuged
at 100,000
g for 1 h to removal all membrane particles
and then dialyzed against 50 mM Tris-Cl, pH 7.5, 1 mM
dithiothreitol, and 1 mM HEDTA (standard buffer). Following
dialysis and reduction of the ionic strength, the solution became
cloudy and was clarified by centrifugation (15 min at 15,000
g) which removed proteins consisting mainly of myosin and
actin. The supernatant is the ``K-released'' fraction used in
this study.
Purification of Aldolase
K-released proteins were
applied to a DE32 (microgranular DEAE-cellulose (Whatman, Clifton, NJ),
5 ml bed volume, and eluted with standard buffer. Aldolase-containing
fractions of the DEAE wash-through were applied to a 1-ml column of no.
204 resin (26) (2-O-[4-(5-aminoethyl-2-hydroxyphenylazo)
benzoyl]-1,4,5-tri-O-phosphono-myo-inositol-Sepharose
4B) (Ins(1,4,5)P affinity column). This column was washed
with 24 ml of 10 mM HEPES-Na, pH 7.3, 50 mM NaCl, 1
mM EDTA, 1 mM dithiothreitol, and 2 mM
NaN (buffer A) and then with buffer A containing the
indicated [NaCl]: 26 ml of 0.15 M, 30 ml gradient of
0.15-0.65 M and 8 ml of 2 M.Aldolase Activity Assay
Aldolase was assayed by
coupling the reaction to the oxidation of glyceraldehyde 3-phosphate
(GAP) to 1,3-diphosphoglyceric acid by glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) and following the reduction of NAD (2 mol/mol
Fru(1,6)P hydrolyzed) at 340 nm. The assay was modified by
using arsenate (27) instead of phosphate as the product,
1-phospho-3-arsenoglyceric acid, spontaneously hydrolyzes. Thus, there
is no need for additional enzymes to remove products. Early assays were
performed at pH 7.6; the reaction mixture contained 200 mM triethanolamine, pH 7.6, 10 mM EDTA, 2 mM
NaAsO, 1 mM Fru(1,6)P, and
1 unit of rabbit skeletal muscle GAPDH in 1 ml (addition of triose
phosphate isomerase did not increase the rate and so was not included).
Later experiments contained 100 mM HEPES-Tris, pH 7.0, 1
mM EDTA, 100 mM KCl, 2 mM
NaAsO, 1 unit of GAPDH, and 2.5-80
µM Fru(1,6)P. When aldolase was assayed at
different pH values, 0.1 M HEPES-Tris (i.e. [HEPES] + [Tris] = 0.1 M)
was used instead of triethanolamine. Mixtures were incubated at 37
°C, and base-line NADH formation was determined. The reaction was
started by addition of enzyme. 1 unit of aldolase was taken to equal to
1 µmol of Fru(1,6)P hydrolyzed/min at 37 °C.Aldolase Isoforms Present in PTSM
This was
determined using antibodies to aldolase A, B, and C, and by determining
preferential substrates (Fru(1,6)Pversus Fru(1)P)
for PTSM aldolase(28) .Ins(1,4,5)P
We used two
different assays: protein precipitation with polyethylene glycol (PEG)
and equilibrium dialysis. The PEG assay was adapted from the method of
Kanematsu et al.(19) and utilized 50 mM HEPES-Tris, pH 7.3, 1 mM EDTA, 100 mM KCl, 0.6
mg/ml Binding Assay-globulins, and 3 nM [
H]-Ins(1,4,5)P in a volume of
50 µl. Assays were performed at 0 °C. The sample was added and
the mixture incubated for 20 min. 50 µl of 30% PEG (average M
8,000) was added, vortexed, and left on ice for
an additional 20 min. The mixture was centrifuged two times (first
horizontally at 15,000 g for 15 min with about 95% of
the supernatant removed and second in an angled rotor at 15,000
g for 5 min with the remaining supernatant removed). The
pellet was dissolved in 0.2 ml of 0.1 N NaOH, transferred to a
counting vial, and the tube rinsed with 0.4 ml of H
O which
was added to the vial. Vials were counted in a scintillation counter.
An aliquot of the mixture was taken prior to the addition of sample to
determine the [H]Ins(1,4,5)P radioactivity in the assay. PEG did not completely precipitate
aldolase, as determined by measuring aldolase activity in the pellet
(obtained by dissolving the pellet in 20 mM HEPES-Na, pH 7.3);
this factor (0.753 ± 0.009) was used to correct for the amount
of aldolase in the PEG precipitation assay. Specific binding was
determined by the addition of unlabeled Ins(1,4,5)P.
Complete displacement of radioactivity was achieved with 200
µM unlabeled Ins(1,4,5)P and was the same
value as -globulin alone (with or without 200 µM
Ins(1,4,5)P
). When binding was assayed at different pH
values, the same HEPES-Tris buffers were used as described in the
aldolase activity assay.H]Ins(1,4,5)P were added on opposite
sides; aliquots of each side were counted after 3 days of equilibration
at 4 °C.Binding of Other Inositol Phosphates and Aldolase
Substrates and Products of Aldolase
Binding of these compounds
was assessed indirectly by determining if they displaced bound
Ins(1,4,5)P.Calculation of [Ca
The computer program MAXC, v6.50 (29) was used.] in
Assayed Fractions
SDS-Gel Electrophoresis
Proteins were dialyzed
against water, SDS and dithiothreitol added to 1.1% and 20 mM,
respectively, placed in a boiling water bath for 5 min, and 10%
glycerol added. Electrophoresis was carried out in 0.4 M Tris-Cl, pH 8.8, in an 8% acrylamide running gel with a 4%
acrylamide stacking gel at 45 milliamps/gel (30) . Gels were
fixed and stained with 0.05% Coomassie Brilliant Blue R in
methanol/water/acetic acid (20:73:7, by volume). When gels were silver
stained(31) , they were first destained in
methanol/water/acetic acid and then rinsed with water to remove all
methanol and acetic acid. Densitometric traces (Shimadzu CS-930) of the
gels were performed at 555 nm for Coomassie Brilliant Blue R and at 474
nm for silver.Western Blots
Proteins were transferred to
nitrocellulose as described previously(32) . Densitometric
traces (Shimadzu CS-930) of the blots were performed at 552 nm.Protein and Total Phospholipid
Measurements
Protein was measured using the Micro BCA Protein
Reagent Kit (Pierce) (prior precipitation with 7% trichloroacetic acid
was used to remove interfering substances). We measured total
phospholipid P by the method of Ames(33) .5`-Nucleotidase Assay
5`-Nucleotidase activity was
measured by incubating aliquots of tissue extracts with 5 mM AMP, 5 mM MgCl, and 50 mM Tris-Cl,
pH 7.4, for 30 min at 37 °C. The reaction was stopped by the
addition of 10% trichloroacetic acid, the samples centrifuged for 5 min
at 1,000 g, and the supernatants analyzed for released
phosphate. Phosphate detected with 0.41%
(NH
)
Mo
O
4O,
0.375 M H
SO, and 1.1% FeSO at 660 nm.Curve Fitting
Constants and standard deviations
were obtained by using non-linear curve fitting with the computer
program SYSTAT (ver. 3.0). Sigmoidal curves were fitted to the general
equation y = (a-c)/(1 + (x/k)) + c(34) .
Materials
[H]Ins(1,4,5)P was purchased from DuPont NEN. Non-radioactive Ins(1,4,5)P was from American Radiolabeled Chemicals, Inc., St. Louis, MO.
HEPES, polyethylene glycol, rabbit skeletal muscle aldolase A,
-globulins, EGTA, EDTA, HEDTA, rabbit skeletal muscle GAPDH, DL-GAP, NAD, arsenic acid (disodium salt),
Fru(1,6)P
, Fru(2,6)P, calpain inhibitor
peptide, aprotinin, leupeptin, pepstatin A, and antibodies to vinculin,
talin, and 
-actinin were from Sigma. Octyl glucoside (OG) was from
Calbiochem, La Jolla, CA. Triton X-100 was from Rohm and Haas,
Philadelphia, PA. Antibodies to human aldolase A-C were kindly provided
by Dr. Kanefusa Kato, Aichi Prefectural Colony, Japan. All other
chemicals were reagent grade.
Characteristics of Ghosts and K-released
Proteins
Gels prepared using SDS-solubilized ghosts (Fig. 1) showed that the major solubilized proteins are actin
and myosin. Ghost suspensions contained about 50% of total cellular
phospholipid and 80% of 5`-nucleotidase activity. About 21% of ghost
proteins could be released from ghosts with 0.6 M KCl. After
removal of particulate material by a 100,000 g centrifugation for 1 h, 16% of cellular proteins remained in the
supernatant (the K-released preparation used in this study). The
K-released fraction contained 11 and 17% of total cellular myosin and
actin, respectively, as assessed by areas of peaks in gels. K-released
proteins reacted strongly with antibodies to vinculin, talin, and
-actinin (Fig. 2). No phospholipid or 5`-nucleotidase
activity was detected in the K-released fraction. Table 1summarizes data relevant to various components and
properties of ghosts and K-released proteins. We have concluded from
the above data that ghosts consist of extracellular
matrix-sarcolemma-cytoskeleton complexes and that KCl-released protein
fractions contain cytoskeleton proteins and no plasma membrane. The
absence of lipid in this preparation suggests the absence of intact
organelles, i.e. intact ER, SR, and nucleus. Small amounts of
myosin are still present in this preparation. Although this could
reflect myosin in the cytoskeleton, i.e. myosin motors, it is
possible that some contractile proteins are still attached to the
cytoskeleton.
for actin was
confirmed, using antibodies to actin, to be 0.6 corresponding to a
molecular mass of 40 kDa. The peak at R
0.06 is myosin.
-actinin, and
talin. B, identity of aldolase isoforms. Lanes 1-3 contained K-released proteins (15 µg), and lane 4 contained 0.75 µg of purified PTSM aldolase. Antibodies to
aldolase A-C were applied as shown in the
figure.
Aldolase Activity in Various Fractions
93% of
cellular aldolase activity was recovered in ghost suspensions from
either unstimulated or 55 µM carbachol-stimulated (20 s or
5 min) tissue and could be entirely released by treatment with high
tonicity (0.6 M KCl) into the K-released fraction. The
remainder of aldolase activity was recovered mainly in the cytoplasm.
No activity was detectable in the sarcolemma.Purification of Aldolase and
Ins(1,4,5)P
K-released proteins
were passed through a DEAE column. Aldolase activity, and coincident
Ins(1,4,5)P-binding Proteins binding, did not stick to the DEAE column, but
passed through the column in the washes. These fractions were then
passed through an Ins(1,4,5)P affinity column. This
revealed (Fig. 3) one major peak of aldolase activity (>80%
of total aldolase activity) which was eluted between 0.2-0.4 M NaCl. Ins(1,4,5)P, up to 10 µM, did
not elute any aldolase or Ins(1,4,5)P-binding protein.
GAPDH activity (an enzyme which can contaminate aldolase preparations
as observed in other tissues) was not detected. SDS-gel electrophoresis
of aldolase recovered from the affinity column contained only one major
peak, at 35-40 kDa (Fig. 4). SDS-gel electrophoresis of
rabbit skeletal muscle aldolase A showed a peak at an identical
molecular weight as PTSM aldolase (not shown).
affinity column. DEAE column fractions
containing aldolase were applied and eluted from the Ins(1,4,5)P affinity column as described under ``Experimental
Procedures.'' The column fractions were monitored for aldolase
activity (open squares) and adsorption at 280 nm (solid
line); estimated [NaCl] is also indicated (interrupted dashed line).
binding
were coincident at a molecular mass of 138 kDa.PTSM Total Aldolase Content
Table 2shows
that aldolase was purified from K-released proteins with only an
11-fold purification. Note that the use of K-released proteins already
resulted in a 17-fold purification of aldolase, given as activity/total
cellular protein. Therefore, the total purification to apparent
homogeneity was about 190-fold. Thus, aldolase contributes about 9% of
the total K-released proteins and about 0.5% of total cellular protein.
) given per 100 nmol phospholipid P was 34.5 pmol/100 nmol lipid P.Aldolase Isoforms
Aldolase in the K-released
fraction and purified aldolase reacted strongly with anti-aldolase C
antibody, but not with anti-aldolase A or B antibodies (Fig. 2).
PTSM aldolase activity was 29 ± 3 times greater with
Fru(1,6)P as substrate than when Fru(1)P was used. This
finding supports the conclusion that aldolase C is the major aldolase
isoform in PTSM(28) . In rabbit tissue, aldolase A (skeletal
muscle) and B (liver) ratios of activity with
Fru(1,6)P/Fru(1)P were reported to be 55, and 1.2,
respectively(28) .Effects of Fru(1,6)P
The
apparent K,
Ins(1,4,5)P, and pH on Aldolase Activity for Fru(1,6)P
(range 2.5-75 µM; 37 °C,
[Ca] = 0, pH 7.0) was 18.5 ±
2.0 µM. The addition of 10 µM Ins(1,4,5)P
had no effect on purified aldolase
activity either at [Fru(1,6)P] of 2.5 or 25
µM (however, a higher [Ins(1,4,5)P]
(20 µM) caused a small (30.1 ± 4.7%), significant
inhibition (p < 0.002)).] 1 mM and
standard conditions quoted above (i.e. [Ca] = 0). Maximal activity was
recorded at pH 8.0. The low Fru(1,6)P
hydrolytic activity
observed at pH 6.0 was not due to a reduced activity of the added GAPDH
used in the coupled enzymatic assay. This was verified by increasing
the amount of GAPDH by 10-fold.
binding. Aldolase activity (closed
circles, solid line (Guassian fit)) and Ins(1,4,5)P binding (open squares, long dashed line (fit),
interrupted dotted line (± 1 S.D.)) were measured at different
pH values as described under ``Experimental Procedures.''
Ins(1,4,5)P binding versus pH was transposed to versus [OH] and the data fitted to
the curve: y = (a-c)/(1+(x/k)
) + c where y =
[
H]Ins(1,4,5)P bound (counts/min), x = [OH] (M), a = maximum binding = 131.2 ± 5.4 counts/min, c = minimum binding = 24.9 ± 4.8
counts/min, k = [OH
] midway
between the maximal and minimal binding transitions = 2.57
± 0.09
10
M (which occurred
at pH = 7.41 ± 0.02), and b = the
steepness factor or the number of H
-binding sites
undergoing transition = 14.6 ±
9.0.
Ins(1,4,5)P
Data were obtained over the range
[Ins(1,4,5)P Binding to Purified
Aldolase] 3 nM to 20 µM (Fig. 6). These data show concentration-dependent binding
of Ins(1,4,5)P to aldolase and that there was no difference
with the two different methods of determining Ins(1,4,5)P binding.
to aldolase. Data from eight separate
determinations (total n = 68) were normalized to
Ins(1,4,5)P bound/aldolase (mol/mol). Data were
obtained by PEG precipitation assay () and equilibrium dialysis
(+). The graph is presented in log-log form for easy visualization
of all data; the inset shows data in linear form. The solid line is the fit to the K
equation described in the text.
binding to aldolase at 20 µM free
Ins(1,4,5)P. At this concentration binding reached about
0.6 mol of Ins(1,4,5)P bound/mol of aldolase.
The K for binding and the total capacity
of aldolase to bind Ins(1,4,5)P
were computed using the
dissociation equation for the reaction:
= dissociation
constant, [A]
= the free
concentration of aldolase
, [I] = the free concentration of Ins(1,4,5)P
,
[AI] = the concentration of
aldolase-Ins(1,4,5)P
complex, n = the
number of Ins(1,4,5)P bound/aldolase,
[A] = the total concentration of
aldolase
and thus,
n[AI]/[A]
= the total amount of Ins(1,4,5)P
bound/total aldolase (mol/mol). Using
[A] = [A]
+ [AI
], the equation was
transposed to
n[AI
]/[A]
=
n[I]
/(K+[I]
).
The values obtained from the mathematical fitting (Fig. 6) of
n[AI]/[A]
versus [I]
were 13.6
± 0.8 µM for K
and
0.99 ± 0.01 for n.
]; not shown) to compute K
and n. Using this method, the
computed n was unreasonably small (0.023), which was
apparently due to the large deviation of the fit from the points at
values <6 µM
for
[Ins(1,4,5)P
] (=
[Ins(1,4,5)P
] > 160 nM). However,
using n = 0.99, as obtained above, Kcomputed by Scatchard analysis was 11.9
± 0.3 µM.
for Ins(1,4,5)P
binding to aldolase was in the range of 11.9-13.6
µM (at 0 °C, pH 7.3, 100 mM KCl, 0
Ca).
Effects of [Fru(1,6)P
Fru(1,6)P] and
[Fru(2,6)P] on Ins(1,4,5)P Binding to
Purified Aldolase had a profound effect on
Ins(1,4,5)P binding (Fig. 7). Ins(1,4,5)P binding was significantly inhibited at a
[Fru(1,6)P] of 8 µM (t test, p < 0.03) and progressively decreased as
Fru(1,6)P was increased to 1 mM. Inhibition of
binding was not dependent upon the [Ins(1,4,5)P]
tested. The [Fru(1,6)P] for half-maximal
inhibition of maximal Ins(1,4,5)P binding was calculated at
be 30 ± 5 µM. Fru(2,6)P was less
effective in inhibiting Ins(1,4,5)P binding. The
[Fru(2,6)P] for half-maximal inhibition was 170
± 28 µM.
binding to aldolase by Fru(1,6)P and Fru(2, 6)P. Purified
PTSM aldolase was incubated with either 10 or 100 nM Ins(1,4,5)P and Fru(1,6)P or
Fru(2,6)P at indicated concentrations
([Ca] = 0 except where indicated):
Fru(1,6)P
(open squares, n =
7-11/point except 2 mM, n = 3);
Fru(1,6)P, [Ca] = 1
µM (
, n = 3/point); Fru(2,
6)P
, (open triangles, n =
5-6/point)). All points >8 µM Fru(1,6)P were significantly different (p > 0.05) than 0
µM Fru(1,6)P. The data
([Ca] = 0; Fru(1,6)P
, n = 56, solid line; Fru(2, 6)P, n = 16, long dashed line) were fitted to the
equation: y = 100/(1 + (x/k)), where y = %
Ins(1,4,5)P
bound (100% at [Fru(X,
6)P] = 0 and 0% at
[Fru(X,6)P]) = , x = [Fru(X,6)P
], k = [Fru(X,6)P] for
displacement of half of the maximum amount of bound Ins(1,4,5)P (Fru(1,6)P, 0.030 ± 0.005 mM;
Fru(2,6)P, 0.170 ± 0.028 mM), and b = the steepness factor (Fru(1,6)P, 0.89
± 0.09; Fru(2,6)P, 0.93 ±
0.13).
] changed the affinity for binding of
Ins(1,4,5)P
to aldolase. Ca (1 µM) did not cause a significant change in
Ins(1,4,5)P
binding at different
[Fru(1,6)P] compared to values obtained at 0
[Ca].
Effects of Ins(1,3,4)P
[,
Ins(1,4)P, Ins(4)P, and GAP on Ins(1,4,5)P Binding to
AldolaseH]Ins(1,4,5)P binding was determined at 10 nM. Ins(1,3,4)P (20 µM) displaced about 70% bound
[H]Ins(1,4,5)P (p <
0.04). Fru(1,6)P and GAP (2 mM) displaced all
bound Ins(1,4,5)P. Ins(1,4)P and Ins(4)P (20
µM) had no effect on bound
[H]Ins(1,4,5)P. Data are shown in Fig. 8.
H]Ins(1,4,5)P by other compounds.
[H]Ins(1,4,5)P was bound to purified
PTSM aldolase by incubation with 10 nM
[H]Ins(1,4,5)P. 20 µM Ins(1,3,4)P, Ins(1,4)P or Ins(4)P, or 2
mM GAP was added to the reaction mixture. Displacement of H was determined by measuring bound counts using the PEG
assay. The graph also shows displacement (data similar to that given
above) caused by addition of 20 µM cold Ins(1,4,5)P and 2 mM Fru(1,6)P. In these experiments
bound [H]Ins(1,4,5)P was determined
by subtracting total H from
[H]-globulin blank. Parentheses indicate number of data points. * indicates statistical
significant displacement of bound
[
H]Ins(1,4,5)P at p <
0.001.
Effects of pH on Ins(1,4,5)P
Ins(1,4,5)P Binding to
Aldolase binding to aldolase was
essentially independent of pH between values of 6.0 and 7.3 (Fig. 5). Between 7.3 and 7.5 there was a sharp drop in binding
of about 80%, which then remained constant to pH 8.5. The halfway point
of the transition was pH 7.41 ± 0.02.
binding to smooth muscle aldolase. The
total aldolase content of PTSM was surprisingly large, i.e. about 0.5% of total cellular protein. 93% of aldolase was found in
the K-released fraction which contains cytoskeletal proteins. On the
basis of studies on other tissues, it is likely that aldolase is bound
to cytoskeletal actin(36) . Since some contractile proteins
were present in ghost and K-released fractions, it is possible that
some or all cellular aldolase was bound to contractile proteins, as
suggested by previous workers who used skeletal or smooth
muscle(27, 37) . Our finding that aldolase was
resistant to detergent solubilization of ghosts is consistent with
aldolase binding to the cytoskeleton. for the substrate Fru(1,6)P
, 18.5
µM, is similar to that reported for pig skeletal muscle,
29 µM(38) , and higher than the rabbit aldolases:
muscle, 6 µM; liver, 1 µM; and brain, 2
µM(39) . compared to Fru(1)P, an early method of
identifying aldolase isoforms. The presence of aldolase C in uterine
and gastric smooth muscle has been reported previously (40) but
in combination with aldolase A. binding to purified aldolase is specific.
Binding was displaced by cold Ins(1,4,5)P,
Fru(1,6)P, Ins(1,3,4)P, and GAP, but not by
Ins(1,4)P or Ins(4)P. The finding that Fru(2,6)P displaced Ins(1,4,5)P only at very high
concentrations argues for the specificity of Fru(1,6)P displacement of Ins(1,4,5)P. binding, at pH > 7.3 (Fig. 5) is of
interest in respect to the binding site of the highly negatively
charged Ins(1,4,5)P to aldolase. Protein pH effects between
pH 6 and 8 may be associated with titration of positively charged
histidyl residues (41, 42, and analysis of our pH data (using the
``steepness factor'' described under ``Results'')
suggests 15 histidyl residues were titrated/aldolase over
this pH range. Aldolase isoforms are rich in histidyl residues (i.e. A = 44, B = 36, and C = 28 per
tetramer(43) . Thus, Ins(1,4,5)P may be bound to
aldolase C histidyl residues. These may also be the same
phosphate-binding residues which were reported to be ``capable of
immobilizing ten negative charges''(44) . to skeletal muscle aldolase A of
4/tetramer(18) . The K for
binding to PTSM aldolase, which was estimated to be in the range
11.9-13.6 µM, was higher than that reported for
skeletal muscle aldolase, 0.58 µM, and liver, 0.83
µM(18) . These differences may be explained, in
part, by different properties of aldolase C (PTSM), aldolase A (rabbit
skeletal muscle), and aldolase B (rabbit liver).
binding to aldolase C can
entirely explain sequestered or bound Ins(1,4,5)P in
unstimulated smooth muscle (11) has not been fully answered by
our study. The large content of aldolase in PTSM supports an additional
non-glycolytic role of this protein and provides a large, potential
sink for Ins(1,4,5)P binding. In our previous
study(11) , 13.3 pmol/100 nmol lipid P of
Ins(1,4,5)P was sequestered or bound in unstimulated PTSM.
As indicated above, total aldolase content determined in
the present study was 34 pmol/100 nmol lipid P. Thus, the
potential Ins(1,4,5)P binding to aldolase is more than
2-fold greater than the measured sequestered or bound Ins(1,4,5)P in unstimulated muscle. The high K for binding found in our study argues that only a fraction
of total aldolase binds Ins(1,4,5)P
under physiological
conditions. The finding that easily measurable Ins(1,4,5)P binding to purified PTSM smooth muscle aldolase occurred even at
[Ins(1,4,5)P] as low as 3 nM suggests a
physiological importance, if binding sites are in close proximity to
Ins(1,4,5)P-sensitive SR. Uncertainties in projecting our
current binding data to amplification of free
[Ins(1,4,5)P] in intact cells are related to: 1)
effects of temperature on binding (our binding studies were performed
at 0 °C), 2) whether aldolase bound to the cytoskeleton has a
different affinity for Ins(1,4,5)P than does the soluble
form of the enzyme, 3) whether changes in Fru(1,6)P and
associated metabolites concentration occur, and 4) other unknown
factors. releases
Ins(1,4,5)P from the aldolaseIns(1,4,5)P complex (this is apparently the first described function for
Ins(1,3,4)P). Further studies are required to determine if
Ins(1,3,4)P formed from phospholipase C-derived
Ins(1,4,5)P functions to release Ins(1,4,5)P from aldolase-binding sites. Fru(1,6)P and GAP also
release aldolase-bound Ins(1,4,5)P. Further work is planned
in our laboratory to determine if the large, rapid increases in
Fru(1,6)P content which occur in other tissues during
stimulation(45, 46, 47) , also occur in PTSM.
A steep pH-sensitive Ins(1,4,5)P binding plot was shown in
this study. It still needs to be determined if pH increases which occur
in stimulated smooth muscle (48, 49, 50) could trigger Ins(1,4,5)P release from the Ins(1,4,5)Paldolase complex.
Calculation of Total Aldolase Stores
Units of
aldolase activity and total cellular proteins were converted/100 nmol
of lipid P giving values of 0.0165 ± 0.0005
units/100 nmol lipid P and 0.91 ± 0.13 mg/100 nmol
lipid P, respectively. units/mg cellular protein) was computed as:
) units/mg
purified aldolase) and Ald
(units/mg cellular
protein):
), inositol 1,4,5-trisphosphate; PIP,
phosphatidylinositol 4,5-bisphosphate; PI, phosphatidylinositol; PIP,
phosphatidylinositol 4-phosphate; PTSM, porcine tracheal smooth muscle;
ER, endoplasmic reticulum; SR, sarcoplasmic reticulum;
Fru(1,6)P, fructose 1,6-bisphosphate;
Fru(2,6)P, fructose 2,6-bisphosphate; HEDTA, N-hydroxyethlethylenediamine triacetic acid; GAP, D-glyceraldehyde 3-phosphate; GAPDH, D-glyceraldehyde
3-phosphate dehydrogenase; PEG, polyethylene glycol; I(4)P, inositol
4-phosphate; Ins(1,4)P, inositol 1,4-bisphosphate;
Ins(1,3,4)P. inositol 1,3,4-trisphosphate;
aldolase, aldolase tetramer.
We acknowledge, with kind appreciation, Dr. Kanefusa
Kato, Department of Biochemistry, Institute for Developmental Research,
Aichi Prefectural Colony, Kamiya, Kasugai, Aichi 480-30, Japan
for providing the aldolase anti-bodies.
©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:
|
|
 |
|
  T. Saeki, J.-B. Shen, and A. J. Pappano Inositol-1,4,5-trisphosphate increases contractions but not L-type calcium current in guinea pig ventricular myocytes Cardiovasc Res, March 1, 1999; 41(3): 620 - 628. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Wilson and P. W. Majerus Isolation of Inositol 1,3,4-Trisphosphate 5/6-Kinase, cDNA Cloning, and Expression of the Recombinant Enzyme J. Biol. Chem., May 17, 1996; 271(20): 11904 - 11910. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |