(Received for publication, November 22, 1996, and in revised form, December 30, 1996)
From the The pivotal role of phosphatidylinositol 3-kinase
(PI 3-kinase) in signal transduction has been well established in
recent years. Receptor-regulated forms of PI 3-kinase are thought to phosphorylate phosphatidylinositol 4,5-bisphosphate
(PtdIns(4,5)P2) at the 3-position of the inositol ring to
give the putative lipid second messenger, phosphatidylinositol
3,4,5-trisphosphate (PtdIns(3,4,5)P3). Cellular levels of
PtdIns(3,4,5)P3 are currently measured by time-consuming procedures involving radiolabeling with high levels of
32PO4, extraction, and multiple chromatography
steps. To avoid these lengthy and hazardous procedures, many
laboratories prefer to assay PI 3-kinase activity in cell extracts
and/or appropriate immunoprecipitates. Such approaches are not readily
applied to measurements of PtdIns(3,4,5)P3 in extracts of
animal tissues. Moreover, they can be misleading since the association
of PI 3-kinases in molecular complexes is not necessarily correlated
with the enzyme's activity state. Direct measurements of
PtdIns(3,4,5)P3 would also be desirable since its
concentration may be subject to additional control mechanisms such as
activation or inhibition of the phosphatases responsible for
PtdIns(3,4,5)P3 metabolism. We now report a simple,
reproducible isotope dilution assay which detects
PtdIns(3,4,5)P3 at subpicomole sensitivity, suitable for measurements of both basal and stimulated levels of
PtdIns(3,4,5)P3 obtained from samples containing
approximately 1 mg of cellular protein. Total lipid extracts,
containing PtdIns(3,4,5)P3, are first subjected to alkaline
hydrolysis which results in the release of the polar head group
Ins(1,3,4,5)P4. The latter is measured by its ability to
displace [32P]Ins(1,3,4,5)P4 from a highly
specific binding protein present in cerebellar membrane preparations.
We show that this assay solely detects PtdIns(3,4,5)P3 and
does not suffer from interference by other compounds generated after
alkaline hydrolysis of total cellular lipids. Measurements on a wide
range of cells, including rat-1 fibroblasts, 1321N1 astrocytoma cells,
HEK 293 cells, and rat adipocytes, show wortmannin-sensitive increased
levels of PtdIns(3,4,5)P3 upon stimulation with appropriate
agonists. The enhanced utility of this procedure is further
demonstrated by measurements of PtdIns(3,4,5)P3 levels in
tissue derived from whole animals. Specifically, we show that
stimulation with insulin increases PtdIns(3,4,5)P3 levels
in rat skeletal muscle in vivo with a time course which
parallels the activation of protein kinase B in the same samples.
PI 3-kinases1 represent a family of
enzymes which phosphorylate phosphoinositides on the 3-position of the
inositol ring (1). The role of PI 3-kinases in mitogenic signaling,
membrane ruffling, trafficking, cell motility, inflammatory and immune
cell responses, activation of neutrophils, and the metabolic effects of
insulin (2-8) has been well established using biochemical,
pharmacological, and genetic approaches. At least two major classes of
PI 3-kinases have been identified which differ in their substrate
specificity. Agonist-stimulated forms of PI 3-kinase can utilize
PtdIns, PtdIns 4-phosphate, and PtdIns(4,5)P2 as substrates
in vitro (9, 10), but several lines of evidence suggest that
PtdIns(4,5)P2 is the preferred substrate in intact cells
(11, 12). The resulting PtdIns(3,4,5)P3 has been shown to
increase rapidly and transiently in many cells in response to a wide
variety of stimuli and since there are no known phospholipases that can
metabolize this lipid (13), it has been suggested that
PtdIns(3,4,5)P3 itself is a second messenger. At present
there are reports that PtdIns(3,4,5)P3 can activate
atypical forms of protein kinase C (14) and can bind to protein kinase
B (also named Akt or Rac) (15-17) and the recently cloned centaurin
(18), encoding for a novel gene, which is highly expressed in brain and
which shows homology to yeast and mammalian Arf GTPase-activating
proteins.
Measuring PtdIns(3,4,5)P3 levels has so far been done by
labeling cells with [3H]inositol or
[32P]orthophosphate. Cell labeling with inositol is often
not very efficient, and labeling to isotopic equilibrium can take days. Labeling cells with [32P]orthophosphate is inconvenient
because of the precautions needed for the large amounts of
radioactivity required. Both labeling procedures suffer from the
drawback of elaborate, expensive and time-consuming HPLC analysis of
deacylated lipid extracts which are required to resolve the glycerol
derivative, GroPtdIns(3,4,5)P3, from
32P-labeled contaminants. We therefore developed a
straightforward, highly specific and sensitive method that allows the
detection of picomole amounts of PtdIns(3,4,5)P3. The assay
is based on cleavage of the polar head group of
PtdIns(3,4,5)P3 to yield Ins(1,3,4,5)P4, the
mass of which can be measured using an isotope dilution assay with a
sensitivity limit of approximately 0.3 pmol. In general, lipid extracts
from samples containing 1 mg of cellular protein are sufficient for
determination of the level of PtdIns(3,4,5)P3. The
preparation of a highly specific Ins(1,3,4,5)P4-binding
protein (obtained from sheep cerebellum) and high specific activity
[3-32P]Ins(1,3,4,5)P4 (prepared from
Ins(1,4,5)P3 and [ [ Expression and purification of
recombinant rat brain Ins(1,4,5)P3 3-kinase was performed
according to Ref. 20 with minor modifications. A single
Escherichia coli colony, containing the Bluescript plasmid
with the cloned DNA insert (C5), was used to inoculate LB medium
supplemented with 50 µg/ml ampicillin at 37 °C to an
A600 of 1.5. The culture was then diluted to an
A600 of 0.5 with fresh prewarmed medium
(30 °C), and expression was induced by the addition of
isopropyl- [3-32P]Ins(1,3,4,5)P4
was prepared from Ins(1,4,5)P3 and
[ 1321N1 Astrocytoma
cells were grown to confluence on 6-well plates as described previously
(22). Prior to treatment, cells were washed twice with modified
Krebs-Henseleit buffer and incubated in this buffer for 30 min at
37 °C (22). Cells were treated for 7.5 min with dimethyl sulfoxide
carrier or wortmannin (100 nM) and then stimulated for 10 min with insulin (10 µg/ml). The medium was aspirated and the cells
were quenched with 1 ml of 10% trichloroacetic acid. After 15 min on
ice, dishes were scraped and washed once with 0.5 ml of 10%
trichloroacetic acid. The cell lysates were centrifuged for 5 min at
13,000 × g, and the resulting pellet was washed twice
with 0.5 ml of 5% trichloroacetic acid, 1 mM EDTA. Lipids
were extracted in 0.75 ml of CHCl3/MeOH/HCl (40:80:1, by
volume) for 20 min, and phases were then split by the addition of 0.25 ml of CHCl3 and 0.45 ml of 0.1 M HCl. The lower
phase, obtained after a 1-min centrifugation at 13,000 × g, was transferred to a screw cap tube, and the upper phase
was re-extracted once with 0.45 ml of the synthetic lower phase. The lower phases were pooled and dried down. Alkaline hydrolysis of dried
lipids was carried out by vortexing in 50 µl of 1.0 M KOH and boiling for 30 min. Following neutralization with 50 µl of 1.0 M acetic acid, the fatty acids were removed by 2 extractions with 0.5 ml of water-saturated butan-1-ol/petroleum ether
(40-60 °C)/ethyl acetate (20:4:1, by volume). Finally, the
resulting water-soluble fractions were dried down and resuspended in
100 µl of 0.2 M acetic acid, resulting in a 0.5 M potassium acetate solution with a final pH of 5.0, which
could be used directly in the isotope dilution assay. The volumes could
be adjusted to take into account the amount of cellular lipid present
in each sample. As a general rule, samples which originally contained 1 mg of cell protein (about 0.3 mg of total lipid) could be efficiently
hydrolyzed using the volumes of reagents noted above (see "Results
and Discussion").
The
Ins(1,3,4,5)P4-binding protein was obtained from sheep
cerebellum and prepared as described in Ref. 23. Briefly, cerebella were homogenized in ice cold buffer (20 mM
NaHCO3, pH 8.0, 1 mM dithiothreitol, 2 mM EDTA) and centrifuged for 10 min at 5000 × g. The pellet was re-extracted once, and the pooled
supernatants were centrifuged for 20 min at 38,000 × g. The pellet was washed twice and resuspended in
homogenization buffer at a final protein concentration of 10-20
mg/ml.
Ins(1,3,4,5)P4 concentration was determined as in Ref. 23
which is an adaptation of Ref. 24. Assays of 320 µl comprised 80 µl
of assay buffer (0.1 M NaAc, 0.1 M
KH2PO4, pH 5.0, 4 mM EDTA, 80 µl
of 3 × 105 dpm of
[32P]Ins(1,3,4,5)P4), 80 µl of sample, and
80 µl of binding protein. Both standard Ins(1,3,4,5)P4
and samples were in 0.5 M KOH/acetic acid (pH 5.0) and were
assayed directly. Samples were diluted in 0.5 M KOH/acetic
acid (pH 5.0) to allow measurements in the most sensitive range of the
displacement curve. After addition of binding protein, samples were
incubated on ice for 30 min and subsequently subjected to rapid
filtration using GF/C filters. Filters were washed twice with 5 ml of
ice cold buffer (25 mM NaAc, 25 mM
KH2PO4, pH 5.0, 1 mM EDTA, and 5 mM NaHCO3). Radioactivity was determined after
the filters were extracted for 12 h in 4 ml of scintillant.
Male Wistar rats
(200 g) were starved overnight and anesthetized with sodium
pentobarbital, and one hind limb muscle was isolated prior to
treatment. Rats were injected via the hind limb saphenous vein with 0.5 ml of 150 mM NaCl, containing propranolol (3.0 mg/kg) or
propranolol (3.0 mg/kg) plus insulin (1.0 unit/kg). After various times, one hind limb skeletal muscle was freeze-clamped (with aluminum
tongs cooled in liquid nitrogen), excised, and then powdered under
liquid nitrogen using a precooled pestle and mortar. One gram of
powdered muscle was used for the PtdIns(3,4,5)P3 mass determination, and three rats were used for each time point. To each
gram of powdered muscle, 10 ml of 10% trichloroacetic acid was added,
and after 10 min the samples were centrifuged at 3000 rpm for 10 min.
The pellets were washed and extracted as described above, but using 10 volumes of extraction solution. The extracted lipids were hydrolyzed in
0.8 ml of 1 M KOH, to account for the greater mass of
starting material. Efficient hydrolysis was checked by HPLC analysis of
spiked [32P]PtdIns(3,4,5)P3. For measurements
of PKB activity, 1 g of powdered muscle was homogenized in 3 ml of
ice-cold homogenization buffer (4 mM EDTA (pH 8.0), 50 mM NaF, 1.0 mM orthovanadate (pH 10.0), 1 µM microcystin-LR, 0.1% The assay procedure is based upon
alkaline cleavage of a total cell phospholipid extract which generates
Ins(1,3,4,5)P4 from PtdIns(3,4,5)P3. The
water-soluble head group can then be measured using a highly specific
radioligand displacement assay. This approach, which avoids tedious
chromatographic procedures, is based on a previously reported assay for
PtdIns(4,5)P2, which utilized alkaline hydrolysis coupled
to the radioligand displacement-based measurement of
Ins(1,4,5)P3 (26).
In order to assess the utility of such an approach for analysis of
PtdIns(3,4,5)P3, it was first necessary to determine the recovery of Ins(1,3,4,5)P4 through the lipid extraction and
hydrolysis procedures. The recovery of PtdIns(3,4,5)P3 was
analyzed by spiking tracer
[32P]PtdIns(3,4,5)P3 into total cell lipid
extracts or lipids from Folch fraction I from bovine brain.
A single back-extraction of the upper phase of such extracts with
synthetic lower phase (see "Experimental Procedures") was sufficient to recover 98% of the spiked radioactivity (not shown). The
recovery of the radiolabeled Ins(1,3,4,5)P4 was then
followed by anion exchange HPLC analysis of the products. For these
experiments, it was important to ensure an excess of KOH for hydrolysis
of the ester bonds in the sample. Cell samples containing 1 mg of protein contain approximately 0.3 mg of lipid. Assuming an average molecular weight of 1000 and 3 susceptible bonds per mol of lipid, 50 µl of 1 M KOH gives a 50-fold excess of hydroxyl ions.
When this approach was applied to astrocytoma cell lipid
extracts, HPLC analysis revealed 4 major peaks of radioactivity which
accounted for >95% of the radioactivity applied to the column and
which corresponded with the mobilities of Ins(3,4,5)P3
(10.9 ± 0.59%), GroPtdIns(3,4,5)P3 (3.9 ± 0.66%), Ins(1,3,4,5)P4 (62 ± 0.41%), and
Ins(2,3,4,5)P4 (24 ± 1.2%). These are the expected
products of alkaline hydrolysis of PtdIns(3,4,5)P3 based on
the original studies of PtdIns(4,5)P2 hydrolysis by
Brockerhoff and Ballou (27). As expected, the recovery of
Ins(1,3,4,5)P4 through this procedure depended on the
relative amounts of lipid and KOH. Thus, hydrolysis of 0.3 mg of
astrocytoma cell lipids with 0.1 M and 0.5 M KOH yielded 17.0% and 52.4% Ins(1,3,4,5)P4,
respectively. All subsequent experiments used 50 µl of 1 M KOH to hydrolyze samples containing not more than 0.6 mg
of lipid; the volume of 1 M KOH was increased
correspondingly for samples containing greater amounts of lipid.
Previous reports established the
presence of highly specific Ins(1,3,4,5)P4 binding sites in
membranes prepared from mammalian cerebella and platelets (23, 24, 28),
which formed the basis of a radioreceptor assay for
Ins(1,3,4,5)P4 (24). This assay requires a radioligand with
a higher specific radioactivity than the commercially available
[3H]Ins(1,3,4,5)P4, and hence it was
necessary to synthesize [32P]Ins(1,3,4,5)P4
as detailed under "Experimental Procedures." Sheep cerebellum
proved in our hands to be a convenient source of membranes with a
sufficiently high density of high affinity Ins(1,3,4,5)P4
binding sites, although pig and rat cerebellum could also be used (23,
24). At least 5 different preparations of
Ins(1,3,4,5)P4-binding protein, isolated from sheep
cerebellum, showed consistent characteristics with regard to the
Kd for Ins(1,3,4,5)P4 (3.97 ± 1.23; n = 5) and Ins(1,4,5)P3, maximal binding, nonspecific binding, and the presence of a low affinity binding site.
Fig. 1 shows a typical calibration curve in which
[32P]Ins(1,3,4,5)P4 binding to cerebellar
membranes from sheep was displaced by increasing concentrations of
unlabeled ligand. The binding assays contained 125 mM
potassium acetate (pH 5.0) to mimic the conditions for samples which
had undergone alkaline hydrolysis and were analyzed using a
computer-assisted curve-fitting program. Ins(1,3,4,5)P4
bound with a Kd of 4.1 ± 0.96 nM
and nonspecific binding, defined in the presence of 2.5 µM unlabeled ligand, was approximately 20% of total
binding. Scatchard analysis indicated the presence of a second, low
affinity site which never amounted to more than 10% of total specific
binding. The sensitivity range for accurate measurements of
Ins(1,3,4,5)P4 was approximately 0.3 to 5 pmol (25-75%
displacement of specific binding). Since the assay volume was 0.32 ml,
this could be improved slightly by reducing the assay volume. These
results are essentially similar to those reported in Ref. 23.
When the above assay was applied to synthetic
PtdIns(3,4,5)P3 that had been subjected to alkaline
hydrolysis, an apparent Kd of 8.1 ± 2.9 nM was obtained, consistent with the expected yield of
Ins(1,3,4,5)P4 of 62%. This suggests that none of the
side-products of PtdIns(3,4,5)P3 hydrolysis have a
significant impact on displacement of the radioligand in the assay.
This was established directly in the case of
GroPtdIns(3,4,5)P3, which gave a Kd of
7.6 ± 1.72 nM, but amounts to less than 5% of the
products of alkaline hydrolysis (see above).
Another issue to be dealt with was the affinity of the receptor for
Ins(1,4,5)P3 since this is the polar head group derived from alkaline hydrolysis of PtdIns(4,5)P2, which in
unstimulated cells is usually present at a 1000-fold excess over
PtdIns(3,4,5)P3 (29). As shown in Fig. 1,
Ins(1,4,5)P3 did not displace any [32P]Ins(1,3,4,5)P4 at concentrations less
than 1 µM (confirming previous reports (23, 24, 28)) and
therefore is unlikely to contribute to the measured values obtained by
the radioligand displacement assay.
A calibration
curve, as shown in Fig. 1, was constructed for each experiment, and the
assay values obtained for each sample were corrected for recovery of
PtdIns(3,4,5)P3 during extraction (98%) and yield of
Ins(1,3,4,5)P4 through alkaline hydrolysis (62%). To
establish that displacement in the assay was due to authentic,
cell-derived PtdIns(3,4,5)P3, we made use of a highly specific
PtdIns(3,4,5)P3/Ins(1,3,4,5)P4-5-phosphatase
that was partially purified from bovine
brain.2 A large scale lipid extract from
astrocytoma cells was hydrolyzed as described above. The
resulting sample was dissolved in water (giving a final solution of 0.5 M potassium acetate, pH 7.5) and split into 2 portions,
which were incubated with the 5-phosphatase under conditions in which
hydrolysis of the available Ins(1,3,4,5)P4 was between 50 and 90%. One sample was spiked with
[32P]Ins(1,3,4,5)P4, and the products were
analyzed by anion exchange to determine the degradation of authentic
Ins(1,3,4,5)P4. The other sample was analyzed by the
radioligand displacement assay. The 2 assays gave similar results with
73% hydrolysis of the internal radiolabeled standard versus
85% loss of radioligand displacing activity (no significant loss was
detected by either approach using boiled enzyme as control). These
results confirm that the radioligand displacement assay detects only
material present in alkali-hydrolyzed total cell lipid extracts that is
degraded by a specific 5-phosphatase preparation at a similar rate to
authentic Ins(1,3,4,5)P4, strongly suggesting that the
assay is specifically measuring the latter compound.
We next applied the assay to measurements of
PtdIns(3,4,5)P3 in control and insulin-stimulated 1321N1
astrocytoma cells in parallel with analysis of cells which
had been labeled to isotopic equilibrium with
[3H]inositol as described previously (22). The latter
procedure allowed an independent assessment of the mass of
PtdIns(3,4,5)P3 which can be determined from the
radioactivity recovered in this lipid and measurement of the specific
radioactivity of the inositol lipid pool. The results from these
experiments are shown in Table I. The basal levels of
PtdIns(3,4,5)P3 (2.16 ± 1.0 pmol/well) increased
approximately 5-fold upon stimulation with insulin (11.3 ± 1.85 pmol/well). These values are remarkably close to those determined from
the specific radioactivity measurements.
PI 3-kinase stimulation in 1321N1 astrocytoma cells
Department of Biochemistry,
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
FOOTNOTES
Acknowledgments
REFERENCES
-32P]ATP using a
recombinant Ins(1,4,5)P3 3-kinase from rat brain) are
described. We show that measurements of PtdIns(3,4,5)P3
mass determined by isotope dilution match well with measurements from isotopic equilibrium cell labeling experiments. Furthermore, the assay
can be applied in situations where radioactive labeling is impractical.
Insulin administration to anesthetized rats resulted in a rapid
increase of PtdIns(3,4,5)P3 levels, which was paralleled by
an increase in PKB activity, in extracts from freeze-clamped hind limb
skeletal muscle.
Materials
-32P]ATP (3000 Ci/mmol) and
[3H]Ins(1,4,5)P3 (20-60 Ci/mmol) were from
Amersham. Ins(1,4,5)P3 and Ins(1,3,4,5)P4 were
from CellSignals. Folch fraction I from bovine brain and calmodulin immobilized on agarose beads were from Sigma. Calmodulin was from Calbiochem. [32P]PtdIns(3,4,5)P3 was prepared
from PtdIns(4,5)P2 and [
-32P]ATP using
immunoprecipitated PI 3-kinase from U937 cells. Anti-PKB antibodies
were raised in sheep. GroPtdIns(3,4,5)P3 was prepared from
PtdIns(3,4,5)P3 by deacylation for 20 min with methylamine (25%)/methanol/butanol (42.8:45.7:11.5 (v/v/v)) at 53 °C (19). A
parallel reaction with [32P]PtdIns(3,4,5)P3
internal standard showed a 95% conversion to GroPtdIns(3,4,5)P3.
-D-thiogalactopyranoside (1 mM) after which the cells were grown for an additional 2 h at
30 °C. Bacteria were harvested (1200 × g, 15 min)
and resuspended in cold lysis buffer (50 mM Tris-HCl, pH
8.0, 1 mM EDTA, 0.4 mM phenylmethylsulfonyl fluoride, 0.4 mM benzamidine, 5 µM leupeptin,
5 µM pepstatin, and calpain inhibitors I and II at 5 µg/ml). After sonication in ice, Triton X-100 was added to a final
concentration of 1% (v/v) and the lysate was shaken for 1 h at
4 °C. After a 30-min centrifugation at 15,000 × g,
the supernatant was applied to a CaM-affinity column which was eluted
as described in Ref. 20. The active fraction was concentrated using an
Amicon Centriprep 30 filter.
-32P]ATP using recombinant Ins(1,4,5)P3
3-kinase partially purified as described above. A reaction mixture of
200 µl, containing 100 µM Ins(1,4,5)P3, 20 mM MgCl2, 50 mM Tris-HCl (pH 7.5),
1.018 mM CaCl2, 1 mM EGTA, 10 µM CaM, 1 mg/ml bovine serum albumin, 1 mCi of
[
-32P]ATP (3000 Ci/mmol), and 20 µl of enzyme, was
incubated for 1 h at 37 °C, and the reaction was terminated by
the addition of 0.8 ml of 10 mM EDTA followed by 2 min of
boiling. The sample was applied to a HPLC Partisphere-SAX column,
eluted with a nonlinear gradient made of water and 1.0 M
NH4H2PO4 (pH 3.8). Fractions (2 ml)
containing [3-32P]Ins(1,3,4,5)P4 were
identified by Cerenkov counting, pooled, and dialyzed 3 times against
1000 ml of water for 45 min (the desalting of inositol polyphosphates
by dialysis is described in more detail in Ref. 21). This effectively
removed the inorganic phosphate which would otherwise interfere in the
Ins(1,3,4,5)P4 binding assay. The recovery of
[32P]Ins(1,3,4,5)P4 following dialysis was
approximately 50%. Starting with 100 µM
Ins(1,4,5)P3 and 1 mCi of [
-32P]ATP
(~3000 Ci/mmol), 0.5 mCi of
[32P]Ins(1,3,4,5)P4 with a specific
radioactivity of ~3000 Ci/mmol was routinely produced.
-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, and 1 mM
benzamidine) using a Polytron homogenizer at 4 °C. The homogenate
was spun at 13,000 × g for 10 min in a refrigerated centrifuge. Immunoprecipitation of PKB from 500 µg of muscle extract and subsequent measurement of enzymatic activity was done as in Ref.
25.
Principles of the Assay
Fig. 1.
Displacement of
[32P]Ins(1,3,4,5)P4 by authentic
Ins(1,3,4,5)P4 and Ins(1,4,5)P3 using
Ins(1,3,4,5)P4-binding protein isolated from sheep
cerebellum. The displacement of
[32P]Ins(1,3,4,5)P4 (3 × 105 dpm) by Ins(1,3,4,5)P4 () and
Ins(1,4,5)P3 (
) at indicated concentrations was measured
in a final assay volume of 0.32 ml after a 30-min incubation on ice.
Each data point represents a determination in triplicate ± S.D.
obtained from seven independent experiments.
[View Larger Version of this Image (19K GIF file)]
1321N1 astrocytoma
cells
PtdIns(3,4,5)P3 (HPLC
analysis)
PtdIns(3,4,5)P3 (specific
activity)
PtdIns(3,4,5)P3 (mass assay)
dpm/well
pmol/well
pmol/well
Control
2467 ± 154
0.82
± 0.051
2.16 ± 1.00
Insulin
23862 ± 775
7.95
± 0.258
11.32 ± 1.85
Wortmannin
ND
ND
1.41
± 0.13
The use of wortmannin, a potent inhibitor of PI 3-kinases, gave further confirmation of the assay's specificity. In 1321N1 astrocytoma cells preincubated with wortmannin at a concentration of 100 nM, a dose which is considered to be selective for PI 3-kinases, the insulin-stimulated increase in PtdIns(3,4,5)P3 was completely abolished. The radioligand displacement assay was also used for measurements on Rat-1 fibroblasts, rat adipocytes, and HEK 293 cells, which all showed wortmannin-sensitive increases in PtdIns(3,4,5)P3 upon stimulation with platelet-derived growth factor or insulin (not shown).
Regulation of PtdIns(3,4,5)P3 Levels and PKB Activity in Skeletal Muscle after in Vivo Stimulation of Rats with InsulinPKB is the cellular homologue of the v-Akt protein encoded in the genome of the Akt-8 retrovirus, isolated from a rodent T-cell lymphoma (30). Some isoforms of PKB are significantly overexpressed in several types of cancer (17, 31, 32). PKB is activated rapidly in response to a wide range of stimuli which also activate PI 3-kinase. A direct role for PI 3-kinase in the activation of PKB has been suggested by a combination of pharmacological and molecular genetic approaches (33-35). Moreover, PtdIns(3,4,5)P3 can bind directly to PKB although this in itself is not sufficient for activation (36) which requires phosphorylation of specific serine and threonine residues by an unidentified protein kinase.3
Using the new assay it is possible to monitor
PtdIns(3,4,5)P3 levels in tissues and/or cells which are
not suitable for labeling to high specific radioactivity. Thus we now
report the first measurements of this lipid second messenger in an
animal tissue, rat hind limb skeletal muscle, and compare the effects
of intravenous injection of insulin on PtdIns(3,4,5)P3
levels and PKB activity in the same tissue samples. The results are
shown in Fig. 2. PtdIns(3,4,5)P3 levels
increased approximately 3-fold, 5 min after insulin administration, remained elevated at 10 min, and returned to basal levels within 15 min. PKB activity was elevated up to 4-fold and followed a similar time
course which persisted at 15 min, but also returned to basal levels
within 30 min. The larger and more prolonged PKB response is compatible
with the hypothesis that PI 3-kinase lies upstream of PKB in a pathway
which serves to amplify this initial signal.
In summary, the present report establishes the feasibility of measuring PtdIns(3,4,5)P3 in cell and tissue extracts by a method which avoids time-consuming chromatographic procedures and hazardous amounts of radioactivity. The specificity and precision of the assay was demonstrated enzymologically, using an enzyme which specifically degrades PtdIns(3,4,5)P3 and Ins(1,3,4,5)P4; pharmacologically, by showing that the signal detected by the assay was enhanced by agonists which stimulate PI 3-kinase in a wortmannin-sensitive manner; and by comparison with PtdIns(3,4,5)P3 levels determined by an independent method. In addition to the ease of analysis, an important advance is the ability to monitor PtdIns(3,4,5)P3 in animal tissues which are not suited to metabolic labeling procedures. As PtdIns(3,4,5)P3 has been implicated as a critical signal in both normal and pathological growth control, an important application of the assay will be the measurement of this lipid in clinical samples such as tumor biopsies.
The E. coli clone expressing the recombinant Ins(1,4,5)P3 3-kinase was a gift from C. Erneux (Free University of Brussels). PtdIns(3,4,5)P3 was provided by R. Gigg, and [3-32P]PtdIns(3,4,5)P3 was prepared by D. M. Hickinson.