α-Thrombin mediated PI 3-Kinase activation through release of Gβγ dimers from Gαq and

Gαi2.

Reema Goel∗, Polly J. Phillips-Mason∗†, Alice Gardner∗±, Daniel M. Raben§ and Joseph J.

Baldassare∗¶

From the ∗Department of Pharmacological and Physiological Sciences, St Louis University

Medical School, St. Louis, Missouri 63104 and the §Department of Physiology, The Johns

Hopkins University School of Medicine, Baltimore, Maryland 21205.

¶To whom correspondence should be addressed: Department of Pharmacological and

Physiological Sciences, St Louis University Medical School, 1402 South Grand Blvd., St. Louis,

Missouri 63104; Tel.: 314-577-8543; Fax: 314-577-8233; E-mail: baldasjj@slu.edu

Running title: PI 3-Kinase pathway requires Gαq and Gαi2 associated Gβγ dimers.

1

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

SUMMARY

Chinese hamster embryonic fibroblasts (IIC9 cells) express the Gα subunits Gαs, Gαi2,

Gαi3, Gαo, Gαq/11 and Gα13. Consistent with reports in other cell types, α-thrombin stimulates a

subset of the expressed G proteins in IIC9 cells, namely Gi2, G13 and Gq as measured by an in

vitro membrane [35S]GTPγS binding assay. Using specific Gα peptides, which block coupling of

G-protein receptors to selective G proteins, as well as dominant negative xanthine nucleotide

binding Gα mutants, we show that activation of the Phosphatidylinositol 3-Kinase /Akt pathway

is dependent on Gq and Gi2. To examine the role of the two G proteins, we examined the events

upstream of PI 3-Kinase. The activation of PI 3-kinase/Akt pathway by α-thrombin in IIC9 cells

is blocked by the expression of dominant negative Ras and β-arrestin1 (Phillips-Mason, P. J.,

Raben, D. M., and Baldassare, J. J. (2000) JBC 275, 18046 and Goel, R., Phillips-Mason, P. J.,

Raben, D. M., and Baldassare, J. J. (2002) JBC 277, 18640), indicating a role for Ras and β-

arrestin1. Interestingly, inhibition of Gi2 and Gq activation blocks Ras activation and β-arrestin1

membrane translocation, respectively. Furthermore, expression of the Gβγ sequestrant, α-

transducin, inhibits both Ras activation and membrane translocation of β-arrestin1, suggesting

that Gβγ dimers from Gαi2 and Gαq activate different effectors to coordinately regulate the PI 3-

Kinase/Akt pathway.

2

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

INTRODUCTION

α-Thrombin is a potent mitogen for Chinese hamster embryonic fibroblasts (IIC9) (1).

The α-thrombin receptor is a member of the 7 membrane spanning family of G-protein coupled

receptors (GPCRs) known as the Proteinase Activated Receptors (PARs) (2). Upon activation,

GPCRs catalyze the exchange of GTP for GDP on the Gα subunit of specific G protein

trimers, promoting the dissociation of Gα-GTP from the Gβγ subunits, which remain tightly

associated (3). Both the Gα-GTP and the Gβγ dimers stimulate effectors including

phospholipase C (4), Ras (5) and PI 3-Kinases (6,7).

Twenty distinct Gα subunits have been cloned which can be divided into four families

based on sequence homology: Gs, Gi, Gq and G12 (8). Members of the Gi family can be

identified by their sensitivity to ADP-ribosylation by pertussis toxin (PTX), which uncouples

the G protein from the receptor (9). In addition to the twenty Gα subunits, five Gβ subunits

and twelve Gγ subunits have been identified (10). The downstream events associated with a

particular GPCR depend on the heterotrimeric G proteins that associate with the receptor.

Several G protein coupled receptors, including α-thrombin, are known to activate multiple G

proteins and these G proteins can activate different signaling pathways (11-20). IIC9 cells

express Gs, Gi2, Gi3, Go, Gq/11 and G13 (21). Pertussis toxin-sensitive G proteins are involved

in α-thrombin-induced Ras activation (22), and Gq family members mediate the activation of

PLC-β1 (4) and PKC (22).

We have previously found that in IIC9 cells α-thrombin induces PI 3-Kinase and Akt

activities that are dependent on Ras and β-arrestin1 (23,24). At present the Gα subunits that

mediate activation of PI 3-Kinase and Akt by α-thrombin are unknown. This paper identifies

the specific Gα subunits by using transient expression of specific Gα peptides, which block

coupling of GPCRs with selective G proteins, as well as dominant negative xanthine nucleotide

3

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

binding Gα mutants. We show that α-thrombin stimulates Gq, G13 and Gi2 in IIC9 cells.

Interestingly, the PI 3-Kinase and Akt stimulation are sensitive to uncoupling of both Gq and

Gi2. Efforts to delineate the individual role of Gq and Gi2 in this common pathway show that

Ras activation and the translocation of β-arrestin1 to membranes are mediated by Gβγ dimers

released from Gαi2 and Gαq respectively. These data indicate that α-thrombin induces PI 3-

Kinase and Akt stimulation via the Gβγ dimers from Gαq acting cooperatively with the Gβγ

dimers from Gαi2, and that PI 3-Kinase activation is a point of convergence in the action of the

two G proteins.

4

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

MATERIALS AND METHODS

Cell Culture and Reagents- IIC9 cells, a clonal population of Chinese hamster embryo

fibroblasts were maintained in Dulbeco’s modified Eagle’s Medium (DMEM) containing 4.5

g/L glucose and 2mM L-glutamine (BioWhittaker, Walkersville, MD) supplemented with 5%

(v/v) fetal calf serum. Subconfluent IIC9 cells (80%) were growth-arrested by washing once

with alpha-MEM (Life Technologies), containing 2mM L-glutamine (BioWhittaker) followed

by 48-hour incubation in the same medium. Human α-thrombin isolated from plasma (Sigma,

St. Louis, MO) was used at 1 NIH unit/ml in all experiments and PTX (BioMol, Plymouth

Meeting, PA) was used at 100 ng/ml.

Transient Transfection- The cDNA encoding pcDNA3.1 (Invitrogen), β-arrestin1 (a kind gift

from Marc Caron), α-transducin (25), MAS-GRK3-ct (a kind gift from Stephen R. Ikeda),

GoαC351G, Gi2αC352G, and Gi3αC351G (generated), carboxyl terminus Gα peptides (cue

Biotech, Chicago, IL) or xanthine nucleotide binding Gα mutants (Guthrie Research Institute,

Sayre, PA) were transfected into subconfluent (60-80%) IIC9 cells using LiptofectamineTM

(Gibco BRL, Gaithersburg, MD) as previously described (23). The following day, cells were

growth-arrested by washing once with alpha-MEM followed by 48-hour incubation in the same

medium prior to agonist stimulation. Transient transfection using LiptofectamineTM resulted in

80-90% expression efficiency as visualized by co-transfection of GFP.

Western Blot Analysis- Growth-arrested IIC9 cells were incubated in the absence or presence of

indicated mitogen for times indicated in the figure legends. At the indicated times, cell lysates

were prepared as previously described (23). Protein lysates (10-25 µg) were resolved by SDS-

PAGE and transferred to a polyvinylidene diflouride membrane (Millipore, Boston, MA).

Membranes were probed with polyclonal antibodies to Akt (Santa Cruz Biotechnology),

5

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

phospho-AktSer473 (Cell Signaling), phospho-ERK (Cell Signaling), ERK (Santa Cruz

Biotechnology), β-arrestin1 (Transduction Laboratories) and PI 3-Kinase p85 antibody (Upstate

Biotechnology, NY). Immunoreactive bands were visualized by enhanced chemiluminescence

(ECL) (Amersham Life Sciences, Arlington Heights, IL) as recommended by the manufacturer.

Preparation of IIC9 Membranes and β-Arrestin1 Translocation Assay- Plasma membranes from

IIC9 cells were prepared as previously described (23). The protein amounts of β-arrestin1 were

determined by Western Blot Analysis and quantified by using Molecular Dynamics

Phosphoimager.

Generation of PTX-Resistant Mutants- PTX-resistant Goα, Gi2α and Gi3α mutants were

generated using PCR to amplify these inserts from their current pcDNA vectors with the

addition of both a 5’ and 3’ restriction site and the C-terminal C→G mutation. For Goα a 5’

Hind III site and a 3’ Xba I site were introduced in addition to the C→G mutation using the

following primers (5’→3’): TAT AAA GCT TGG CCA CCA TGG GAT GTA CTC and TAT

ATC TAG ATC AGT ACA AGC CTC CGC CCC GGA GAT TGT T. For Gi2α a 5’ Hind III

site and a 3” Xho I site were introduced in addition to the C→G mutation using the following

primers (5’→3’): TAT AAA GCT TGG CAG GAT GGG CTG CA and TAT ACT CGA GTC

AGA AGA GGC CTC CGT CCT TCA GGT TGT TCT TG. For Gi3α a 5’ Hind III site and a 3”

Xho I site were introduced in addition to the C→G mutation using the following primers

(5’→3’): TAT AAA GCT TGG CCG CCG TCA TGG GCT GC and TAT ACT CGA GCC

TCT CAG TAA AGC CCA CCT TCC. Restriction sites are underlined. PCR was carried out in

the presence of 4 mM MgSO4 using the Vent enzyme (Promega) and the following cycling

conditions: 95°C, 5 minutes; 95°C, 1 minute, 55°C, 1 minute, 72°C, 1.5 minutes, 30 cycles. PCR

6

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

products were restricted with the appropriate enzymes and subcloned into pcDNA3 (Invitrogen).

Mutants were screened by restriction enzyme analysis and mutations confirmed by sequencing

using a Perkin Elmer automated sequencer.

ADP Ribosylation Assay- IIC9 cells were pretreated with pertussis toxin at 100 ng/ml. IIC9

membranes were prepared as described as mentioned above and resuspended in 30 µl of 50 mM

Tris-HCl, pH 7.5, 1 mM EDTA and 1 mM MgCl2 to each sample 50 µl 2x reaction buffer (2mM

ATP, 100 mM Tris-HCl, pH 7.5, 20 mM thymidine, 40 mM arginine, 0.4 mg/ml BSA, 200 mM

KPO4, pH 7.5, 10 mM ADP-ribose, 20 mM MgCl2, 2 mM EDTA and 200 µM GTP) and 4 µl

activated pertussis toxin (pertussis toxin was activated by incubating 20 µl pertussis toxin

[1mg/ml] with 42 mM DTT, 20 mM Tris-HCl, pH 7.5 and was incubated at 30 °C for 30

minutes ). Finally, 10 µl of 5 µM NAD containing 20 µCi [32P]NAD was added to each sample

and was incubated at 30 °C for 30 minutes. The reaction was stopped by adding 100 µl of NAD

wash solution (5 mM NAD, 50 mM Tris-HCl, pH 7.5, 2.5 mM EDTA) to each sample.

Membranes were pelleted by centrifugation at 14,000 x g at 4°C for 5 minutes, washed twice

with 200 µl of NAD wash solution, resuspended in Laemmli sample buffer and boiled for 5

minutes. Proteins were separated by SDS-PAGE gel containing 6M urea and quantified using a

Molecular Dynamics Phosphoimager.

Ras/Rho Activity Assays- Growth-arrested IIC9 cells or transfected cells were incubated in the

presence or absence of 1 unit/ml α-thrombin for 5 minutes. After 5 minutes, the cells were

washed two times with 4oC PBS and harvested by scraping into 500 µl lysis buffer (50 mM Tris-

HCl, pH 7.5, 20 mM MgCl2, 150 mM NaCl, 1% Triton X-100, 2 mM p-nitrophenylphosphate,

10 µg/ml pepstatin, 10 µg/ml aprotinin and 10 µg/ml leupeptin), and the cell extracts centrifuged

7

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

for 2 minutes at 4o C. The cell extract was added to glutathione beads complexed with GST-

RBD (fusion protein containing the Ras-binding domain of Raf-1 for Ras activity and fusion

protein containing the RhoA binding domain of Rhotekin for Rho activity) for 1 hour at 4o C.

Samples were washed three times with 500 µl of ice cold lysis buffer, analyzed on 15% SDS-

PAGE gel and transferred to PVDF. The membranes were probed with a 1/1000 dilution of a

pan Ras antibody (Santa Cruz) or RhoA monoclonal antibody (Santa Cruz). The amount of

activated Ras or Rho (complexed with GTP) was visualized by ECL detection.

Phosphatidylinositol 3-Kinase Assay- PI 3-Kinase activity determined as previously described

(23).

Assay of [35S]GTPγS Binding- Membranes (approximately 40µg protein) from serum arrested

IIC9 cells, were resuspended in 50µl of 50 mM Tris-HCl (pH 7.6), 2 mM EDTA, 100 mM

NaCl, 1 mM MgCl2, 1 µM GDP, 10 µg/ml pepstatin, 10 µg/ml aprotinin, 10 µg/ml leupeptin

and 50 nM [35S]GTPγS (2000 Ci/mmol), were incubated in the presence or absence of 1 unit/ml

α-thrombin at 37oC. After 5 minutes the reaction was terminated by the addition of 500µl of 50

mM Tris/HCl (pH 7.6), 150 mM NaCl, 20 mM MgCl2, 100 µM GDP, 100 µM GTP, 1%

Nonidet P-40, 10µg/ml pepstatin, 10 µg/ml aprotinin and 10 µg/ml leupeptin. The extracts were

incubated with 3µl of preimmune serum and 200µl of Pansorbin cells (Calbiochem) and

centrifuged after 30 minutes to remove nonspecifically bound proteins. Extracts were incubated

with antibody directed against a specific Gα subunit and incubated for 1 hour at 4oC. The

immune complex is then incubated with 50µl of a 50% protein G agarose suspension, the

complexes collected and washed three times in assay buffer plus 1% Nonidet P-40 and twice in

assay buffer without detergent and the presence of Gα subunits was analyzed by Western blot

8

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

analysis using Gα specific antibodies. [35S ]GTPγS binding in the immunoprecipitates were

quantified by scintillation counting.

Confocal microscopy- Growth-arrested or transfected IIC9 cells grown on chamber slides

(Nalgeen®Labware, Rochester, NY) were incubated in the absence or presence of 1 unit/ml α-

thrombin for indicated time points after pre-incubation in the absence or presence of 100 ng/ml

PTX for 6 hours. Subsequent to activation, the cells were fixed in a 3.7% Formalin (Sigma)

solution for 10 minutes at room temperature followed by 6 minute incubation in ice cold

methanol at -20°C. The cells were washed in PBS and then blocked in 1 ml of blocking buffer

(0.8 g Fatty Acid Free BSA (Sigma) in 100µl PBS) for 2 hours at room temperature. β-arrestin1

monoclonal antibody was added at a 1:75 dilution (antibody: blocking buffer) and incubated at

room temperature for 2 hours. The cells were washed three times with PBS. The secondary

antibody (Jackson Immunoresearch Inc) was added (1:5000 dilution in blocking buffer) for 45

minutes at room temperature. Again the cells were washed three times with PBS and then

mounted using gel mount (Biomedia Corp, Foster City, CA) and microscope cover slips (Fisher

Scientific, Pittsburgh, PA). Z series images were obtained using a Bio-Rad MRC 1024 confocal

microscope. The acquired images were assembled using Adobe Photoshop and MS PowerPoint.

Co-Immunoprecipitations- Growth-arrested or transfected IIC9 cells were incubated in the

presence or absence of 1 unit/ml α-thrombin. Cell lysates were prepared and protein

concentrations were determined as mentioned above. Protein lysates (100 µg) were incubated

with 5 µg β-arrestin1 antibody or PI 3-Kinase p85 antibody at 4 °C with gentle rocking for 2

hours. The immune complexes were then immunoprecipitated as described previously (26),

resolved by SDS-polyacrylamide gel electrophoresis and transferred to a PVDF. Membranes

9

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

were then probed with antibodies to β-arrestin1, PI 3-Kinase or G proteins (Calbiochem).

Immunoreactive bands were visualized by ECL detection.

PIP2 hydrolysis assay- PIP2 hydrolysis was determined as previously described (13).

10

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

RESULTS

Rapid PI 3-Kinase activity is pertussis toxin-sensitive and mediated by Gβγ dimers - Previous

data from our laboratory (23) show that in IIC9 cells, α-thrombin activates rapid β-arrestin1-,

and Ras-dependent PI 3-Kinase activity. Because changes in Akt phosphorylation and activity

are dependent on PI 3-kinase activity (23), we quantified differences in Akt phosphorylation as a

measure of changes in PI 3-kinase activity. Similar to our reported results (23,24), α-thrombin

induces a 10-fold increase in Akt activity as determined by Western blot analysis using

antibodies directed against phosphorylated Akt. To determine whether the increase in PI 3-

kinase is dependent on a member of pertussis toxin (PTX) sensitive G proteins, we treated IIC9

cells with 100ng/ml of pertussis toxin for 6 hours prior to stimulation. Pretreatment with

pertussis toxin completely blocks the α-thrombin-induced increase in Akt phosphorylation (Fig.

1A) and PI 3-Kinase activity (Fig. 1C), indicating that α-thrombin-induced PI 3-kinase/Akt

pathway is dependent on a member of the Gi/o subfamily of G proteins. While pertussis toxin

blocks α-thrombin-stimulated Akt phosphorylation, it is ineffectual on Akt phosphorylation by

PDGF (Fig. 1A), which acts through a receptor tyrosine kinase and is insensitive to PTX. To

determine whether this activation is mediated by the Gα subunit or the Gβγ dimers, we

examined the effect of Gβγ sequestrants on the α-thrombin-induced activation of PI 3-Kinase

and Akt. Transient expression of both Gαt (Fig. 1B) and a membrane anchored GRK3 carboxyl-

terminal polypeptide, MAS-GRK3ct (data not shown) block α-thrombin-stimulated Akt

phosphorylation by over 65% whereas; expression of the vector alone is ineffectual (Fig. 1B).

Similar results are seen for PI 3-Kinase activity (Fig. 1C). Consistent with the known

mechanism of signaling through tyrosine kinase receptors, Gαt does not affect EGF-induced PI

3-Kinase activity (Fig. 1C). To ensure that the Gβγ sequestrants do not affect a Gα subunit

11

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

mediated effect, we examined the result of expression of Gαt on PIP2 hydrolysis, which is

known to be mediated by Gαq (22). As previously found (13), addition of α-thrombin induces a

3- to 4-fold increase in inositol phosphate release (Fig. 3B). While expression of Gαt blocks Akt

phosphorylation (Fig. 1B), no change in the amount of thrombin-induced inositol phosphate

release is observed (Fig. 3B). Taken together, these data indicate that α-thrombin stimulates

rapid PI 3-kinase activity and Akt phosphorylation by the Gβγ dimers released from a PTX-

sensitive G protein.

α-Thrombin stimulates multiple heterotrimeric G proteins Gq, Gi2 and G13 - Several G protein

coupled receptors stimulate cellular responses by activating multiple G proteins that couple to

different signaling pathways. Because in many cell types, α-thrombin activates several G

proteins, including Gq/11, G13/12 and members of the pertussis toxin sensitive Gi subfamily,

we next investigated the G proteins activated by α-thrombin in IIC9 cells (Fig. 2). Previously we

found that IIC9 cells express Gαs, Gαi2, Gαi3, Gαo, Gαq/11 and Gα13 (21). We next examined

α-thrombin-induced binding of [35S]GTPγS to specific Gα subunits after immunoprecipitation

of the G proteins with Gα specific antibodies against Gαs, Gαi2, Gαi3, Gαo, Gαq, and Gα13

(27,28). Although significant binding of [35S]GTPγS occurs in unstimulated cells, α-thrombin

induces a 4-fold increase in [35S]GTPγS binding to Gαq, a 3-fold increase to Gα13 and an

approximate 5-fold increase to Gαi2 (Fig. 2A). α-Thrombin does not mediate significant binding

to Gαs, Gαi3 or Gαo (Fig. 2A). Because lysophosphatidic acid-mediated increases in Akt

phosphorylation are blocked by expression of Gα subunits C-terminal peptides to Gαo2 and not

Gαi2 (unpublished data), we reasoned that lysophosphatidic acid should couple to Go and not

Gi2. Consistent with the Akt phosphorylation data, lysophosphatidic acid stimulates a significant

increase in [35S]GTPγS binding to Gαo and not Gαi2, (Fig. 2B). Finally, treatment of isolated

12

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

IIC9 membranes with the active peptide of cholera toxin induces detectable binding [35S]GTPγS

to only Gs (Fig. 2C). The thrombin-induced increases in [35S]GTPγS binding to Gαi2, Gαq, and

Gα13 indicate that α-thrombin couples to, and therefore can signal through, these G proteins.

Because some of the G proteins may be expressed at levels that are difficult to detect, and we

have not identified ligands that stimulate all of the G-proteins expressed, it is not possible to

determine the G-proteins activated by α-thrombin with certainty. However, these studies suggest

that α-thrombin couples primarily to Gi2, Gq and G13. Furthermore, we find no detectable

increases in [35S]GTPγS binding to other members of the pertussis toxin class as observed in

other cell types (19).

Gαq and Gαi2 are critical for α-thrombin-stimulated Akt phosphorylation - Because pertussis

toxin blocks α-thrombin-induced increase in Akt phosphorylation (Fig. 1A), and α-thrombin

induces a significant increase in GTPγS binding to Gi2 (Fig. 2), we reasoned that Gi2 mediates

PI 3-Kinase stimulation. To further demonstrate the involvement of Gi2, we expressed peptides

containing the C-terminus of Gα subunits (Fig. 3A). Expression of the carboxyl terminus Gα

peptides of specific Gα subunits inhibits binding and activation of the targeted Gα subunit

(14,29), indicating that these peptides effectively inhibit specific GPCR-mediated responses.

Consistent with the data in Figures 1 and 2, expression of the C-terminal peptide of Gαi1/2

inhibits α-thrombin-induced Akt phosphorylation (Fig. 3A). Furthermore, expression of Gα13

(Fig. 3A), Gα01, Gα02 (data not shown) is ineffective. Surprisingly, rapid Akt phosphorylation

is inhibited by expression of the C-terminal peptide of Gαq (Fig. 3A), suggesting that α-

thrombin-induced PI 3-Kinase activation is dependent on Gαq. We obtained similar results for

PI 3-kinase activity (data not shown). The unexpected activity of the C-terminal peptide of Gαq

prompted us to characterize these peptides further. We reasoned that PIP2 hydrolysis and Rho

13

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

activity would be dependent on Gq and G13 (14), respectively. While expression of the Gαq C-

terminal peptide blocks PIP2 hydrolysis, expression of Gαi1/2 C-terminal peptide or PTX

treatment does not (Fig. 3B). Furthermore, only expression of the C-terminal peptide of Gα13

inhibits Rho activation, while Gαi1/2 and Gαq are ineffective (Fig. 3C). We also measured the

ability of the C-terminal peptides to inhibit sustained ERK phosphorylation, which is

dependent on PTX-sensitive G proteins (Gi2). We find that expression of the C-terminal

peptide of Gαi1/2 inhibits sustained ERK activation, while Gα13 and Gαq do not inhibit it (Fig.

3C). Taken, together these data indicate that the C-terminal peptides effectively inhibit targeted

Gα subunits and that Akt phosphorylation is dependent on both Gq and Gi2.

We were surprised that Akt phosphorylation is sensitive to uncoupling of Gαq

from α-thrombin. To ensure that these results are not selective for C-terminal peptides, we

decided to use another approach to inhibit G protein activation, utilizing xanthine nucleotide

binding Gα mutants (GαXs). Expression of GαX inhibits coupling of GPCR, including the

thrombin receptor, to specific G proteins indicating that these GαXs act as dominant negative

inhibitors (30). We transiently transfected IIC9 cells with Gαi2X, GαqX and Gα13X, and

quantified Akt phosphorylation (Fig. 3A). Similar to the results with the C-terminal peptides,

Gαi2X and GαqX block Akt phosphorylation, while Gα13X does not. Consistent with the ability

of these GαX constructs to inhibit the activation of specific G proteins by α-thrombin,

expression of GαqX blocks and Gαi2X does not inhibit PIP2 hydrolysis (data not shown).

Taken together, these data show that α-thrombin-induced PI 3-Kinase activity and Akt

phosphorylation is dependent on both Gq and Gi2.

Expression of Pertussis toxin-insensitive Gi2α rescues α-thrombin-stimulated Akt

phosphorylation in the presence of pertussis toxin – Because our data show that α-thrombin-

14

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

induced Akt phosphorylation is dependent on both Gq and Gi2, we reasoned that expression of

a pertussis toxin-resistant mutant of Gαi2 should rescue pertussis toxin inhibition of α-

thrombin-induced Akt phosphorylation. Pertussis toxin-resistant mutants were generated by

converting the C-terminal of Gα from cysteine residue to a glycine (GαoC351G, Gαi2C352G,

and Gαi3C351G). To verify that the C→G mutation renders the expressed Gα subunits

resistant to pertussis toxin treatment, IIC9 cells were transiently transfected with vector alone

or vector containing GαoC351G, Gαi2C352G, or Gαi3C351G, treated with 100ng/ml pertussis

toxin for 10 hours and then cell lysates were analyzed by SDS-PAGE in the presence of 6M

urea (Fig. 4A). Previous data in IIC9 cells show that pre-treatment with 100ng/ml pertussis

toxin for 10 hours results in the ADP-ribosylation of more than 95% of the pertussis toxin-

sensitive G proteins (22). For each construct, in the absence of pertussis toxin western blot

analysis with an antibody that recognizes all Gi family members shows a single band with

apparent molecular weight of 43kDa (Fig. 4A). Treatment with pertussis toxin results in a

single band shifted to higher molecular weight (Fig 4B). In the presence of pertussis toxin,

western blot analysis of lysate proteins from cells expressing each construct shows a band at

43kDa and a faster migrating band (Fig 4B). Presumably, the lower band is the unmodified Gα

subunit and is absent in lysates from cells transfected with empty vector. Furthermore, transient

expression of the three Gα C→G constructs results in similar protein levels (Fig. 4B) ensuring

that the ability to rescue is independent of expression levels. These data clearly demonstrate

that treatment with pertussis toxin for 10 hours results in ADP-ribosylation of the endogenous

Gi family members and the Gα C→G mutants are resistant to pertussis toxin. In agreement

with our data indicating a role for Gi2 in α-thrombin-stimulated Akt phosphorylation,

expression of Gαi2C352G but not GαoC351G or Gαi3C351G rescues α-thrombin-induced rapid

Akt phosphorylation in IIC9 cells treated with pertussis toxin (Fig. 4C).

15

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

α-Thrombin induced translocation of β-arrestin1 to the plasma membrane is mediated by Gβγ

released from Gαq - Because β-arrestin1 is required for thrombin-induced PI 3-Kinase and Akt

activities in IIC9 cells (23), we addressed the possibility that β-arrestin1 translocation is

dependent on a single G-protein. We have previously established that treatment of IIC9 cells

with α-thrombin results in the translocation of β-arrestin1 to the plasma membrane (23). We

have also shown that expression of a membrane anchored GRK3 carboxyl-terminal polypeptide

(MAS-GRK3ct) blocks the translocation, indicating the involvement of a Gβγ (23). We next

examined the effect of expression of C-terminal peptides on α-thrombin-induced translocation

of β-arrestin1 (Fig. 5) by both biochemical fractionation (Fig. 5A) and confocal microscopy

(Fig. 5B&C). We found it necessary to transiently express β-arrestin1 to examine the

translocation. α-Thrombin induces a significant increase in the translocation of β-arrestin1 to

the membrane fraction within 5 minutes (Fig. 5A) and in the translocation of a substantial

fraction of the cellular fluorescence to the plasma membrane within 10-30 minutes (Fig. 5B,

2&3). After 30 minutes, β-arrestin1 is found mainly in the cytoplasm (Fig. 5B, 4). While

expression of the C-terminal peptide of Gαq inhibits translocation (Fig. 5A & 5C, 4), treatment

with PTX or expression of the C-terminal peptide of Gαi1/2 does not block the amount of β-

arrestin1 translocation (Fig. 5A & 5C, 3&6). However, transient co-expression of Gαt blocks

the increase (Fig. 5A) and also results in an evenly distributed cytoplasmic fluorescence (Fig.

5C, 5). These data indicate that β-arrestin1 translocation to the plasma membrane in IIC9 cells

is mediated by the Gβγ released from Gαq and suggest that Gi2 affects PI 3-Kinase activation

by another mechanism.

α-Thrombin-mediated formation of complexes containing PI 3-Kinase and β-arrestin1 -

Because the ability of β-arrestin1 to associate with members of MAPK pathways is known to

16

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

affect activation of these pathways (31), we reasoned that β-arrestin1 serves as an adapter

molecule, associating with and recruiting PI 3-Kinase to the membrane. To determine whether

α-thrombin induces association of β-arrestin1 with PI 3-Kinase, we immunoprecipitated β-

arrestin1 pre and post α-thrombin stimulation and examined the immunoprecipites for the

presence of PI 3-Kinase (Fig. 6). In the absence of α-thrombin, low amounts of PI 3-Kinase are

found in the immunoprecipitates (Fig. 6). α-Thrombin induces an increase of PI 3-Kinase in β-

arrestin1 immunoprecipitates within 15 minutes that decreases to amounts similar to those

found in unstimulated cells after 30 minutes (data not shown). Similar results are found when

PI 3-Kinase immunoprecipitates are assayed for the presence of β-arrestin1 (data not shown).

Expression of Gαq C-terminal peptide inhibits complex formation while the Gαi1/2 C-terminal

peptide does not (Fig. 6). Taken together, these data indicate that α-thrombin stimulates

membrane translocation of β-arrestin1 and its association with PI 3-Kinase suggesting that, as

is found in MAPK pathway activation (32), β-arrestin1 functions as an adaptor molecule

recruiting PI 3-Kinase.

Gi2 mediates PI 3-Kinase activation by α-thrombin via Ras - While our results clearly

demonstrate a role for Gq in the activation of PI 3-kinase, the data do not provide a mechanistic

role for Gi2. Previously we found that Ras is required for α-thrombin-induced PI 3-Kinase

activation (24). Furthermore, Ras stimulation by α-thrombin is independent of β-arrestin1 (23),

suggesting that Ras activation is independent of Gq. These data suggest that Gi2 may affect PI

3-Kinase via Ras. To test this hypothesis, we next examined the role of G proteins in the

activation of α-thrombin-induced Ras activation using the RBD fragment of the Ras effector

Raf-1 (33). Treatment of IIC9 cells with α-thrombin results in an approximate 12-fold increase

of GTP-bound Ras (Fig. 7). The α-thrombin-induced increase is blocked by pretreatment of IIC9

17

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

cells with pertussis toxin or expression of the Gαi1/2 C-terminal peptide (Fig. 7A). In contrast to

the ability of the Gαi1/2 C-terminal peptide to block the Ras activation, expression of the Gαq C-

terminal peptide does not affect the increase (Fig. 7A). Also we find that expression of pertussis

toxin-insensitive Gαi2C352G but not GαoC351G nor Gαi3C351G rescues α-thrombin-induced

Ras activation in IIC9 cells treated with PTX (Fig. 7B), confirming the ability of the C-terminal

peptide of Gαi1/2 to block Ras activation.

Because GPCRs that are sensitive to pertussis toxin often stimulate Ras through

Gβγ subunits (5), we reasoned that expression of Gβγ sequestrants would block Ras activation

by α-thrombin. Expression of Gαt completely abolishes α-thrombin mediated GTP-binding to

Ras (Fig. 7A). Taken together, these data demonstrate the involvement of Gβγ dimers from Gαi2

in α-thrombin-stimulated Ras and thus a role in PI 3-Kinase activation.

18

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

DISCUSSION

α-Thrombin stimulates multiple effectors in fibroblasts (1,22,34), including PI-

PLC, ERK and PI 3-Kinase activities. In this paper, we identify the G proteins that mediate the

activation of PI 3-Kinase and Akt by α-thrombin. We find that treatment of IIC9 cells with PTX

markedly inhibits PI 3-Kinase and Akt activation in response to α-thrombin (Fig. 1). Because

PTX catalyzes the ADP-ribosylation and inactivation of members of the Gi/o family (9), these

data indicate the involvement of Gi/o. Consistent with these data, α-thrombin activates the PTX-

sensitive G-protein, Gi2 (Fig. 2). We then reasoned that Gi2 mediates PI 3-Kinase and Akt

activation in response to α-thrombin. Consistent with this notion, expression of either Gαi1/2 C-

terminal blocking peptide (29) or dominant negative Gαi2X mutant (30) blocks Akt

phosphorylation and PI 3-Kinase activity (Fig. 3). Surprisingly, expression of a Gαq C-terminal

peptide, which blocks the activation of Gq, or the dominant negative GαqX mutant, inhibits Akt

phosphorylation (Fig. 3A), indicating an important role for Gq in PI 3-Kinase activation by α-

thrombin.

Previously, we reported that β-arrestin1 is required for α-thrombin-induced PI 3-

Kinase and Akt activities in IIC9 cells (23,35). However, we had not identified the G proteins

that mediate activation of the PI 3-Kinase/Akt pathway. Expression of a dominant negative

GαqX mutant and a Gαq C-terminal peptide blocks β-arrestin1 translocation, while expression of

dominant negative Gαi2X mutant or the Gαi1/2 C-terminal peptide is ineffective (Fig. 5).

Furthermore, β-arrestin1 translocation is dependent on the Gβγ subunits of Gαq, indicating an

essential role for Gβγ in β-arrestin1 translocation.

In addition to Gq, there is also a role for Gi2 in α-thrombin-induced PI 3-Kinase

and Akt stimulation. Our results demonstrate that Ras is upstream of PI 3-Kinase activation and

19

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

the dependence of PI 3-Kinase on Ras is unrelated to the dependence of PI 3-Kinase on β-

arrestin1. First, inhibition of Gαi2 activation abrogates Ras activation (Fig. 7A). Second,

expression of only PTX-resistant Gαi2 rescues α-thrombin-induced Ras activation in PTX-

treated IIC9 cells (Fig. 7B). Furthermore, α-thrombin induces Ras via the Gβγ dimers released

from Gαi2 (Fig. 7A). Taken together, these results indicate that α-thrombin stimulates Ras via

the Gβγ from Gi2.

We were surprised to find that the translocation of β-arrestin1 and the activation of

Ras were mediated by the Gβγ subunits from Gq and Gi2, respectively. These results show that

the Gβγs associated with Gq and Gi2 can activate different effectors. There are several possible

explanations to account for the specificity of Gβγ signaling of Gq and Gi2. It is possible that Gαq

and Gαi2 associate with different subsets of Gβγs, and the distinct subsets activate distinct

effectors. Several studies lend support to the notion that select Gβγ subunits activate distinct

effectors (36). Co-transfection experiments in COS-7 cells found that expression of only specific

Gβ or Gγ subunits significantly activate PI-PLCβ2 (37). Whereas expression of Gβ1γ1, Gβ1γ5 or

Gβ2γ5 stimulates PI-PLCβ2 activity, expression of Gβ2γ1 has no effect. Similarly, expression of

Gβ5γ2 stimulates PI-PLBβ2 but not ERK, while expression of Gβ1γ2 activates both (38). Other

studies (39) found that expression of only specific Gβ subunits activate G protein receptor

kinases2 (GRK2). Of the Gβ subunits examined, Gβ1 and Gβ2 but not Gβ3 activates GRK2.

Recent data using immobilized Gβ1γ2 to screen phage-displayed random peptide libraries (40)

found a peptide, which contains a conserved sequence found in PI-PLCβ. Interestingly,

expression of this peptide blocks activation of PI-PLCβ, but not Gβγ dependent voltage-gated

channels or Gβγ-mediated inhibition of type I adenylate cyclase suggesting that effectors bind to

specific Gβγ peptide sequences (40).

20

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

For different G-proteins to activate distinct effectors via release of Gβγ suggests

that the Gαs are associated with distinct subsets of Gβγs. While there is strong evidence for the

activation of effectors by specific Gβγ subunits, there is little data on the identity of the Gβγ

associated with specific Gαs. Consistent with this requirement, analysis of the Gβ subunits

associated with Gαi2 and Gαq in IIC9 cells (Reema, G and Baldassare, JJ, unpublished results)

find different subsets of Gβs associated with Gαi2 and Gαq.

A second explanation for our results is that activation of Ras and translocation of

β-arrestin1 occurs in different membrane domains, for example in raft and non-raft membrane

fractions (41-44). These membrane domains are enriched in proteins important in intracellular

signaling including GPCRs, suggesting that these domains may play a role in signaling from

GPCRs to their effectors. Huang et al (45) have reported enrichment of Gi, Gs, Go and Gβγs

in detergent-resistant membrane domains, lipid rafts . Furthermore, H-Ras has been shown to

be localized to rafts, suggesting that Ras activation could occur in these domains (46).

At present our data cannot distinguish between these mechanisms. However, any

mechanism must take into account our results showing that the activation of PI 3-kinase by α-

thrombin (Fig. 8) is mediated via Gq and Gi2. Gq plays a crucial role in mediating α-thrombin-

induced PI 3-kinase activation through β-arrestin1 by stimulating β-arrestin1 association with

PI 3-kinase, suggesting that β-arrestin1 functions as an adaptor molecule recruiting PI 3-kinase.

PI 3-kinase activation also requires Ras activation, which is downstream of Gi2. Both effectors

are blocked by Gβγ sequestrants, indicating that they are Gβγ-regulated. Because Ras

activation and β-arrestin1 translocation are crucial for α-thrombin-mediated PI 3-kinase

activity and, therefore, likely occur in the same domain, we think that the association of

different Gβγs with specific Gαs is likely essential. However, it is also important to keep in

21

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

mind that these mechanisms are not mutually exclusive. Thus, both could play a role. The

possible importance of these mechanisms in Gβγ signaling remains to be elucidated.

22

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

References

1. Chambard, J., Paris, S., A'llemain, G., and Pouysségur, J. (1987) Nature 326, 800-803

2. Coughlin, S. R. (1999) Proc.Natl.Acad.Sci.U.S.A. 96, 11023-11027

3. Bourne, H. R. (1997) Curr.Opin.Cell Biol. 9, 134-142

4. Hawes, B. E., van Biesen, T., Koch, W. J., Luttrell, L. M., and Lefkowitz, R. J. (1995) Journal of Biological Chemistry 270, 17148-17153

5. Crespo, P., Xu, N. Z., Simonds, W. F., and Gutkind, J. S. (1995) Journal of Cellular Biochemistry 13

6. Brock, C., Schaefer, M., Reusch, H. P., Czupalla, C., Michalke, M., Spicher, K., Schultz, G., and Nurnberg, B. (2003) Journal of Cell Biology 160, 89-99

7. Kurosu, H., Maehama, T., Okada, T., Yamamoto, T., Hoshino, S., Fukui, Y., Ui, M., Hazeki, O., and Katada, T. (1997) Journal of Biological Chemistry 272, 24252-24256

8. Simon, M. I., Strathmann, M. P., and Gautam, N. (1991) Science 252, 802-808

9. Fields, T. A. and Casey, P. J. (1997) Biochemical Journal 321, 561-571

10. Dhanasekaran, N. and Prasad, M. V. (1998) Biol.Signals Recept. 7, 109-117

11. Aragay, A. M., Collins, L. R., Post, G. R., Watson, A. J., Feramisco, J. R., Brown, J. H., and Simon, M. I. (1995) Journal of Biological Chemistry 270, 20073-20077

12. Baffy, G., Yang, L., Raj, S., Manning, D. R., and Williamson, J. R. (1994) J.Biol.Chem. 269, 8483-8487

13. Gardner, A., Phillips-Mason, P. J., Raben, D. M., and Baldassare, J. J. (2002) Cellular Signalling 14, 499-507

14. Gilchrist, A., Vanhauwe, J. F., Li, A. L., Thomas, T. O., Voyno-Yasenetskaya, T., and Hamm, H. E. (2001) Journal of Biological Chemistry 276, 25672-25679

15. Kanthou, C., Kanse, S. M., Kakkar, V. V., and Benzakour, O. (1996) Cellular Signalling 8, 59-66

23

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

16. Ogino, Y., Tanaka, K., and Shimizu, N. (1996) European Journal of Pharmacology 316, 105-109

17. Post, G. R., Collins, L. R., Kennedy, E. D., Moskowitz, S. A., Aragay, A. M., Goldstein, D., and Brown, J. H. (1996) Mol.Biol.Cell 7, 1679-1690

18. Rahman, A., True, A. L., Anwar, K. N., Ye, R. D., Voyno-Yasenetskaya, T. A., and Malik, A. B. (2002) Circulation Research 91, 398-405

19. Vanhauwe, J. F., Thomas, T. O., Minshall, R. D., Tiruppathi, C., Li, A. L., Gilchrist, A., Yoon, E., Malik, A. B., and Hamm, H. E. (2002) Journal of Biological Chemistry 277, 34143-34149

20. Verrall, S., Ishii, M., Chen, M., Wang, L., Tram, T., and Coughlin, S. R. (1997) Journal of Biological Chemistry 272, 6898-6902

21. Cheng, J., Weber, J. D., Baldassare, J. J., and Raben, D. M. (1997) Journal of Biological Chemistry 272, 17312-17319

22. Gardner, A. J. (1998) Dissertation 1-2

23. Goel, R., Phillips-Mason, P. J., Raben, D. M., and Baldassare, J. J. (2002) Journal of Biological Chemistry 277, 18640-18648

24. Phillips-Mason, P. J., Raben, D. M., and Baldassare, J. J. (2000) Journal of Biological Chemistry 275, 18046-18053

25. Weber, J. D., Cheng, J., Raben, D. M., Gardner, A., and Baldassare, J. J. (1997) Journal of Biological Chemistry 272, 17320-17326

26. Lents, N. H., Keenan, S. M., Bellone, C., and Baldassare, J. J. (2002) Journal of Biological Chemistry 277, 47469-47475

27. Barr, A. J., Brass, L. F., and Manning, D. R. (1997) Journal of Biological Chemistry 272, 2223-2229

28. Windh, R. T. and Manning, D. R. (2002) G Protein Pathways Part B: G Proteins and Their Regulators 344, 3-14

29. Gilchrist, A., Bunemann, M., Li, A., Hosey, M. M., and Hamm, H. E. (1999) Journal of Biological Chemistry 274, 6610-6616

30. Yu, B., Gu, L. J., and Simon, M. I. (2000) Journal of Biological Chemistry 275, 71-76

31. Pierce, K. L., Luttrell, L. M., and Lefkowitz, R. J. (2001) Oncogene 20, 1532-1539

32. Miller, W. E. and Lefkowitz, R. J. (2001) Curr.Opin.Cell Biol. 13, 139-145

24

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

33. Taylor, S. J., Resnick, R. J., and Shalloway, D. (2001) Regulators and Effectors of Small Gtpases, Pt G 333, 333-342

34. Grand, R. J. A., Turnell, A. S., and Grabham, P. W. (1996) Biochemical Journal 313, 353-368

35. Goel, R. and Baldassare, J. J. (2002) Cell Signaling, Transcription, and Translation As Therapeutic Targets 973, 138-141

36. Albert, P. R. and Robillard, L. (2002) Cellular Signalling 14, 407-418

37. Wu, D., Katz, A., and Simon, M. I. (1993) Proc.Natl.Acad.Sci.U.S.A. 90, 5297-5301

38. Zhang, S., Coso, O. A., Lee, C., Gutkind, J. S., and Simonds, W. F. (1996) J.Biol.Chem. 271, 33575-33579

39. Daaka, Y., Pitcher, J. A., Richardson, M., Stoffel, R. H., Robishaw, J. D., and Lefkowitz, R. J. (1997) Proc.Natl.Acad.Sci.U.S.A. 94, 2180-2185

40. Scott, J. K., Huang, S. F., Gangadhar, B. P., Samoriski, G. M., Clapp, P., Gross, R. A., Taussig, R., and Smrcka, A. V. (2001) Embo Journal 20, 767-776

41. Brown, D. A. and London, E. (1998) Annu.Rev.Cell Dev.Biol. 14, 111-136

42. Dobrowsky, R. T. (2000) Cell Signal. 12, 81-90

43. Heximer, S. P., Srinivasa, S. P., Bernstein, L. S., Bernard, J. L., Linder, M. E., Hepler, J. R., and Blumer, K. J. (1999) J.Biol.Chem. 274, 34253-34259

44. Kurzchalia, T. V. and Parton, R. G. (1999) Curr.Opin.Cell Biol. 11, 424-431

45. Huang, C., Hepler, J. R., Chen, L. T., Gilman, A. G., Anderson, R. G., and Mumby, S. M. (1997) Mol.Biol.Cell 8, 2365-2378

46. Prior, I. A., Harding, A., Yan, J., Sluimer, J., Parton, R. G., and Hancock, J. F. (2001) Nat.Cell Biol. 3, 368-375

25

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

Footnotes

1The abbreviations used are: IIC9 cells, Chinese hamster embryonic fibroblasts; GPCR,

G-protein coupled receptor; Proteinase Activated Receptor, PAR; α-transducin, Gαt; PI3-

Kinase, phosphatidylinositol-3OH-Kinase; ERK, extracellular-related kinase; PAGE,

polyacrylamide gel electrophoresis; PLC, phospholipase C; PKC, protein kinase C; GTPγS,

guanosine-5’-O-(3-thio)triphosphate; Polyvinylidene difluoride membrane, PVDF.

This work is supported by GM59251(DMR)

†Present address is Department of Molecular Biology and Microbiology, Case Western

Reserve University, Cleveland, Ohio 44106.

±Present address is Department of Pharmaceutical Sciences, Massachusetts College of

Pharmacy & Health Sciences, Worcester, MA, 01608.

26

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

FIGURE LEGENDS

Figure 1: PI 3-Kinase pathway is pertussis toxin-sensitive and mediated by Gβγ dimers.

Cells were untreated or transfected with pcDNA (Invitrogen) or pcDNA containing Gαt. Growth

arrested IIC9 cells were untreated or incubated in the presence of 100ng/ml PTX 6 hours prior to

stimulation with the indicated mitogen for 5 minutes (1 unit/ml α-thrombin or 10 ng/ml PDGF

or 10 ng/ml EGF). A) & B) Lysate proteins (20 µg) were separated by SDS-PAGE,

immunoblotted with either anti-Akt or anti-phospho-Akt polyclonal antibody and the blots

quantified using a Molecular Dynamics densitometer. Relative activation of Akt (p-Akt /Akt) is

shown, with activity of serum arrested IIC9 as 1.0. Results for both A and B are representative

blots of four independent experiments. C) PI 3-Kinase complexes were immunoprecipitated

from lysates containing equal proteins using an anti-p85 polyclonal antibody and assayed for the

ability to phosphorylate PI in vitro. PI 3-Kinase activity was quantified as previously described

(1). The data shown are the mean ± S. D. for triplicates in one experiment and are representative

of four independent experiments. The Western blot is a representative blot.

Figure 2: α-Thrombin stimulates the binding of [35S]GTPγS to Gq, Gi2 and G13.

Membranes were prepared from serum arrested IIC9 cells and the G protein immune complexes

containing specific G proteins were immunoprecipitated as described in “Materials and

Methods” after treatment with A) 1 unit/ml thrombin or B) 20µM LPA or C) 1µg/assay of

cholera toxin. The presence of Gα subunits was analyzed by Western blot analysis using the

indicated Gα specific antibodies. In control samples, membranes proteins were

immunoprecipitated with anti-Gα subunit antibody (Calbiochem, CA). [35S]GTPγS binding in

the immunoprecipitates was quantified by scintillation counting. The data shown are the mean

± S. D. for triplicates in one experiment and are representative of four independent experiments.

27

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

Figure 3: Effect of C-terminal peptides of Gα subunits on α-thrombin-stimulated

effectors. IIC9 cells were untreated or transfected with the C-terminal peptides of Gα subunits

and then serum arrested. A) Inhibition of Gαq or Gαi2 activation blocks α-thrombin-

stimulated Akt phosphorylation. Cells were transfected with the GαXs mutants or the C-

terminal peptides as indicated in figure. The serum arrested cells were untreated or treated with

100ng/ml PTX for 6 hours prior to incubation in the absence or presence of 1 unit/ml α-

thrombin for 5 minutes. Cell lysates (20 µg) were separated by SDS-PAGE, and immunoblotted

with either anti-Akt or anti-phospho-Akt polyclonal antibody. Blots were quantified using a

Molecular Dynamics densitometer. The data shown are the mean ± S. D. for triplicates in one

experiment and are representative of three independent experiments. The Western blot is a

representative blot. B) Effect of C-terminal peptides on PIP2 hydrolysis. Serum arrested

cells were labeled with myo-[3H]inositol (1 µCi/ml) for 24 hours, and then incubated in serum

free medium supplemented with 20 mM LiCl in the absence or presence of 1 Unit/ml α-

thrombin for 20 minutes at 37oC, and IP3 amounts quantified. The data shown are the mean ±

S. D. for duplicates in one experiment and are representative of three independent experiments.

C) Effect of C-terminal peptides on Rho activity and sustained ERK phosphorylation.

Serum arrested cells were incubated in the absence or presence of 1 unit/ml α-thrombin for 5

minutes and 4 hours for Rho activity and sustained ERK phosphorylation, respectively. For

Rho activity, cells lysates were prepared, incubated with recombinant GST-RBD for 1 hour,

and the immucomplexes subjected to SDS-PAGE and Western blot analysis with anti-RhoA

antibody. For ERK activity the cell lysates were subjected to SDS-PAGE and Western blot

analysis with anti-phospho-ERK polyclonal antibody. Relative activation of RhoA (complexed

with GTP) and ERK are shown in the same graph. We calculated these activities in the

28

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

following manner. We subtracted the basal activities (for Rho and Phospho-ERK) from the α-

thrombin-induced activities in the untransfected cells and set these values equal to 100%. The

transfected samples are calculated by subtracting the basal from the experimental and dividing

these values by the untransfected values times 100. The data shown are representative of three

independent experiments.

Figure 4: C G Gi Mutants Are Not ADP-ribosylated: Cells were transiently transfected

with either vector (pcDNA), or pcDNA containing GoαC351G, Gi2αC352G, and Gi3αC351G.

The transfected cells were growth arrested, and cell lysates prepared from: A) cells treated with

100 ng/ml PTX for 6 hours prior to the preparation of lysates or B) untreated cells. Lysate

proteins (20 µg) were separated by SDS-PAGE containing 6M urea and immunoblotted with a

polyclonal antibody to an internal sequence recognized by Gα subunits. C) Expression of PTX-

insensitive Gαi2 rescues α-thrombin-stimulated Akt phosphorylation in IIC9 cells treated

with PTX. The transfected cells were incubated in the presence or absence of 100 ng/ml

pertussis toxin 6 hours prior to stimulation with 1 unit/ml α-thrombin for 5 minutes and cell

lysates prepared. Lysate proteins (20 µg) were immunoblotted with anti-phospho-Akt or anti-

Akt polyclonal antibody. Akt phosphorylation was quantified using a Molecular Dynamic

PhosphoimagerTM and reported as percent maximal stimulation. Data in A) and B) are

representative of three independent experiments. The data shown in C) are the mean ± S. D. for

triplicates in one experiment and are representative of four independent experiments. The

Western blot is a representative blot.

Figure 5: Expression of Gβγ sequestrant or C-terminal Gαq blocks α-thrombin induced

translocation of β-arrestin1. IIC9 cells were transfected with β-arrestin1 alone or with Gαt or

29

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

the C-terminal peptides of Gαq or Gαi1/2. The cells were then stimulated with 1 unit/ml α-

thrombin. A) Membranes were prepared after stimulation for 15 minutes. Membrane protein

lysates (50 µg) were separated by SDS-PAGE, immunoblotted, and the amount of β-arrestin1

quantified with a Molecular Dynamics densitometer. The data shown are the mean ± S. D. for

triplicates in one experiment and are representative of three independent experiments. The

Western blot is a representative blot. B) & C) IIC9 cells were grown on chamber slides and

transfected as described above. The cells were fixed, permeabilized and visualized by confocal

microscopy as described under “Material and Methods”. B) β-Arrestin1 was visualized after 0

minutes, 10 minutes, 30 minutes and 1 hour post stimulation with α-thrombin. C) The cells were

serum arrested, and were untreated or treated for 4 hour with 100 ng of PTX. After treatment the

cells were stimulated with 1 unit/ml α-thrombin for 30 min. Data in B and C are representative

of three independent experiments.

Figure 6: α-Thrombin-mediated formation of complexes containing PI 3-Kinase and β-

arrestin1. IIC9 cells were transiently co-transfected with 1µg/ml β-arrestin1 and C-terminal

peptides of Gαi1/2 or Gαq. The cells were serum arrested, and treated with or without 1 unit/ml of

α-thrombin for 15 minutes. Protein lysates (100 µg) were incubated with antibodies directed

against β-arrestin1 or PI 3-Kinase p85 for 2 hours. The immune complexes were

immunoprecipitated as described in “Materials and Methods”. Samples were subjected to SDS-

PAGE and probed with the indicated antibodies directed against β-arrestin1or PI 3-Kinase p85

and the amount of β-arrestin1 or PI 3-Kinase p85 quantified with a Molecular Dynamics

densitometer. This is a representative blot of five independent experiments

30

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

31

Figure 7: Gi2 mediates Ras activation. Cells were transfected with A) Gαt or the C-terminal

peptides Gαi1/2 or Gαq or B) pcDNA containing GαoC351G, Gαi2C352G, and Gαi3C351G. The

cells were growth arrested. Growth arrested cells were incubated in the presence or absence of

100ng/ml PTX 6 hours prior to stimulation and then incubated for 15 minutes in the absence or

presence of 1 unit/ml α-thrombin. Cells lysates were prepared and incubated with recombinant

GST-RBD for 1 hour. Samples were subjected to SDS-PAGE, probed with anti-pan-Ras

antibody and the amounts of Ras quantified. The data shown in A) are the mean ± S. D. for

triplicates in one experiment and are representative of three independent experiments. The

Western blot is a representative blot. Data in B) is representative of two independent

experiments.

Figure 8: Signal transduction pathway involved in α-thrombin-induced PI 3-Kinase.

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from

Figure 1

A

P-Akt

Akt

α-Thrombin PDGF

PTX - - + - +

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 1

0

20

40

60

80

100A

kt P

hosp

hory

latio

n (%

Max

imal

)

α-Thrombin - + + pcDNA - + -α-Transducin - - +

P-Akt

B

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 1

C

PI 3

-Kin

ase

Act

ivity

(Fol

d-ac

tivat

ion)

0

2

4

6

8

EGFα-Thrombin

Gαt PTX Gαt

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

0

1000

2000

3000

4000

5000

6000

7000

Gs Gi2 Gi3 Go Gq G13control

Figure 2AA

Imm

unop

rec i

pita

ted

[35S]

GT

P γS

( cpm

)

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 2

BB

0

2000

4000

6000

8000

Control Gi2 Gi3 Go

Imm

uno p

reci

pita

ted

[35S]

GTP

γ S (c

pm)

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 2

0

1000

2000

3000

4000

5000

6000

Control Gs Gi2 Gi3 Go Gq G13

CC

Imm

uno p

reci

pita

ted

[35S]

GTP

γ S (c

pm)

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

0

5

10

15

20

25

30

PTX - - + - - - - - -C terminal peptide Gαi/2 Gαq Gα13

GαXs Gα13 Gαi2 Gαq

Akt

Pho

spho

ryla

tion

(% M

ax)

α-Thrombinα-Thrombin

Figure 3A

P-Akt

Akt

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 3B

PTX - - - - - + C terminal peptide Gαq Gαi2 Transducin Gαt

[3 H]In

osito

l Pho

spha

tes

(cpm

/wel

l)

0

2000

4000

6000 α-Thrombin

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 3CC

C terminal peptide Gαi2 Gα13 Gαq

0

20

40

60

80

100

120

140R

hoRho

Act

ivat

ion

Act

ivat

ion

Phos

pho

Phos

pho --

ERK

Act

ivat

ion

ERK

Act

ivat

ion

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 4

Gα-ADP-riboseGα

PTX - + - + - + - +

Gαo Gαi2 Gαi3pcDNA C351G C352G C351GA.

Gαo Gαi2 Gαi3pcDNA C351G C352G C351G

Gα

B.

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 4

0

1

2

3

4

5

6

7

Akt

Pho

spho

ryla

tion

(% M

axim

al)

pcDNA Gαo Gαi2 Gαi3C351G C352G C351G

P-Akt

Akt

α-Thrombin

PTX

C

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 5

0

5

10

15

20

25

30α-Thrombin

β-ar

rest

in 1

tran

sloc

atio

n to

the

plas

ma

mem

bran

e(Fo

ld In

crea

se)

A

PTX - - - - - +C terminal peptide Gαq Gαi2

Transducin Gαt

β-Arrestin1

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 5

B1. Basal 2. Thrombin (10 mins)

3. Thrombin (30 mins) 4. Thrombin (1 hr)

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 5C

5. α-Thrombin + Gαt 6. α-Thrombin + C-terminal Gi1/2

1. Unstimulated

3. α-Thrombin + PTX

2. α-Thrombin

4. α-Thrombin + C-terminal Gq

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 6

IP : β-Arrestin1

α-Thrombin

IP : PI 3-kinase p85

β-Arrestin1

PI 3-kinase p85

β-Arrestin1

PI 3-kinase p85

C terminal peptide Gαq Gαi2

α-Thrombin

C terminal peptide Gαq Gαi2

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 7

A

0

10

20

30

Ras-GTP

PTX - - - - + -

C terminal peptide Gαq Gαi2

Transducin Gαt

Ras

α-Thrombin

Ras

-GTP

Pro

tein

/(tot

al R

as p

rote

in)

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 7

0

5

10

15

20

25

30

35

pcDNA Gαo Gαi2 Gαi3(C351G) (C352G) (C351G)

PTX - - + - + - + - +

α-Thrombin

Ras-GTP

Ras

B

Ras

-GTP

Pro

tein

/(tot

al R

as p

rote

in)

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

Figure 8

RasβγGi2α

p85 p110

Gqαβ γ

GRK3

β-arrestin1

α-Thrombin

PAR1

AKT

by guest on December 25, 2019 http://www.jbc.org/ Downloaded from

BaldassareReema Goel, Polly J. Phillips-Mason, Alice Gardner, Daniel M. Raben and Joseph J.

scripti2αscriptq and Gα dimers from Gγβ-Thrombin mediated PI 3-kinase activation through release of Gα

published online December 9, 2003J. Biol. Chem. 

  10.1074/jbc.M308753200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

by guest on Decem

ber 25, 2019http://w

ww

.jbc.org/D

ownloaded from