Download Negative Regulation of Rho Signaling by Insulin and Its

Negative Regulation of Rho Signaling by Insulin and Its
Impact on Actin Cytoskeleton Organization in Vascular
Smooth Muscle Cells
Role of Nitric Oxide and Cyclic Guanosine Monophosphate
Signaling Pathways
Najma Begum,1,2 Oana A. Sandu,1 and Noreen Duddy1
Recent studies from our laboratory have shown that
insulin induces relaxation of vascular smooth muscle
cells (VSMCs) via stimulation of myosin phosphatase
and inhibition of Rho kinase activity. In this study, we
examined the mechanism whereby insulin inhibits Rho
signaling and its impact on actin cytoskeleton organization. Incubation of confluent serum-starved VSMCs with
thrombin or phenylephrine (PE) caused a rapid increase
in glutathione S-transferase-Rhotekin-Rho binding domain–associated RhoA, Rho kinase activation, and actin
cytoskeleton organization, which was blocked by preincubation with insulin. Preexposure to NG-monomethyl
L-arginine acetate (L-NMMA), a nitric oxide synthase
inhibitor, and Rp-8 CPT-cyclic guanosine monophosphate (RpcGMP), a cyclic guanosine monophosphate
(cGMP) antagonist, attenuated the inhibitory effect of
insulin on RhoA activation and restored thrombin-induced Rho kinase activation, and site-specific phosphorylation of the myosin-bound regulatory subunit
(MBSThr695) of myosin-bound phosphatase (MBP), and
caused actin fiber reorganization. In contrast, 8-bromocGMP, a cGMP agonist, mimicked the inhibitory effects
of insulin and abolished thrombin-mediated Rho activation. Insulin inactivation of RhoA was accompanied by
inhibition of isoprenylation via reductions in geranylgeranyl transferase-1 activity as well as increased
RhoA phosphorylation, which was reversed by pretreatment with RpcGMP and L-NMMA. We conclude that
insulin may inhibit Rho signaling by affecting posttranslational modification of RhoA via nitric oxide/cGMP
signaling pathway to cause MBP activation, actin cytoskeletal disorganization, and vasodilation. Diabetes
51:2256 –2263, 2002
From the 1Diabetes Research Laboratory, Winthrop University Hospital,
Mineola, New York; and the 2School of Medicine, State University of New York
at Stony Brook, Stony Brook, New York.
Address correspondence and reprint requests to Najma Begum, Diabetes
Research Laboratory, Winthrop University Hospital, 259 First St., Mineola, NY
11501. E-mail: [email protected].
Received for publication 2 November 2001 and accepted in revised form 3
April 2002.
ECL, enhanced chemiluminescence; FITC, fluorescin isothiocyanate;
GGTase, geranylgeranyl transferase; GST, glutathione S-transferase; HRP,
horseradish peroxidase; iNOS, inducible nitric oxide synthase; L-NMMA,
NG-monomethyl L-glargine acetate; MBP, myosin-bound phosphatase; MBS,
myosin-bound regulatory subunit; MLC, myosin light chain; NO, nitric oxide;
NOS, NO synthase; PI3-kinase, phosphatidylinositol-3 kinase; RBD, Rho
binding domain; ROK-␣, Rho-dependent kinase; VSMC, vascular smooth
muscle cell.
2256
T
he Rho family of small GTPases are well-recognized intracellular signaling proteins that act as
molecular switches to control actin cytoskeleton
organization in many cell types, including
smooth muscle (1– 4). In addition, RhoA-dependent signaling pathway controls important vascular smooth muscle
cell functions such as contraction, migration, and proliferation (5,6). In vascular smooth muscle cells (VSMCs),
the contracting effect of RhoA results from the activation
of Rho-dependent kinase (ROK-␣), which phosphorylates
the regulatory subunit of myosin light chain (MLC) phosphatase (myosin-bound regulatory subunit [MBS]) leading
to the inhibition of its function by reductions in the
phosphatase activity (7,8), thus allowing an increase in the
level of phosphorylated MLC and contraction at a constant
intracellular calcium level (7–10).
Activation of Rho GTPases can be regulated by several
mechanisms, including the activation of heterotrimeric
G-protein– coupled receptors (11). On agonist stimulation,
Rho is converted from the inactive guanosine diphosphate– bound form to the active GTP-bound form. Activation of RhoA by the agonists requires translocation of
inactive cytosolic RhoA to the membrane fractions. Thus,
appearance of RhoA in the membrane fraction is indicative
of Rho activation (12). Rho activation and/or membrane
localization is regulated by posttranslational modification
by geranylgeranylation and phosphorylation of Rho (13–
15). Phosphorylation by cAMP-dependent protein kinase
(protein kinase A) as well as by cyclic guanosine monophosphate (cGMP)-dependent protein kinase inhibits Rho
signaling by interfering with the membrane anchoring of
Rho (14 –16). Rho signaling pathway has been implicated
in the pathogenesis of hypertension as well as diabetes
(17–20). However, the agonists and the mechanism that
counterregulate Rho signaling in vivo have not been
clearly defined.
Recent studies from this laboratory have shown that
insulin rapidly stimulates myosin-bound phosphatase
(MBP) activity to cause MLC dephosphorylation and relaxation of VSMCs (20,21). MBP activation by insulin was
accompanied by a decrease in Rho kinase activity. These
DIABETES, VOL. 51, JULY 2002
N. BEGUM, O.A. SANDU, AND N. DUDDY
effects of insulin were severely impaired in VSMCs isolated
from diabetic Goto-Kakizaki rats.
Because Rho kinase is one of the downstream targets of
RhoA and activation of Rho kinase is dependent on RhoA
activation and translocation to membrane fraction, in this
study, we have examined the molecular mechanism by
which insulin may inhibit Rho signaling and its impact on
actin stress fiber organization.
RESEARCH DESIGN AND METHODS
Culture of VSMCs and treatment. VSMCs were isolated from rat thoracic
aortas and maintained in culture as described in our recent publications
(20 –24). All experiments were performed on highly confluent cells (7–9 days
in culture) at identical passages. Before each experiment, cells were serum
starved for 24 h in serum-free ␣-MEM containing 5.5 mmol/l glucose. The next
day, cells were exposed to insulin (0 –100 nmol/l) for 0 –30 min. In some
experiments, VSMCs were pretreated with various inhibitors for 30 min,
followed by exposure to insulin as detailed in the figure legends.
Metabolic labeling of VSMCs and measurement of RhoA phosphorylation by immunoprecipitation and Western blot analyses. Serum-starved
VSMCs labeled with [32P]orthophosphoric acid (0.3 mCi/ml ⫻ 4 h) were
exposed to various agonists as detailed in the figure legends. The cells were
lysed in a buffer containing 50 mmol/l HEPES, 2 mmol/l EDTA, 1% Triton
X-100, and a cocktail of phosphatase and protease inhibitors; pH 7.5 (20 –24).
Precleared lysates with equal amounts of proteins (200 ␮g) were immunoprecipitated with anti-RhoA antibody (10 ␮g) prebound to anti-mouse IgG
agarose (100 ␮l) for 3 h at 4°C followed by separation of the immunoprecipitates by SDS/PAGE and autoradiography. The level of RhoA phosphorylation
was measured by densitometric scanning of the autoradiograms. To overcome
variations in proteins due to immunoprecipitation, membranes were probed
with anti-RhoA antibody followed by incubation with horseradish peroxidase
(HRP)-conjugated secondary antibodies and detection by enhanced chemiluminescence (ECL). The extent of RhoA phosphorylation was quantitated by
dividing the intensity of radioactive signal with the protein signal.
Immunoprecipitation and in vitro assay of Rho kinase activity in the
immunecomplexes. Rho kinase was immunoprecipitated using anti–Rok-␣
antibody as detailed in our recent publications (20,21). Rho kinase activity in
the immunoprecipitates was assayed using the recombinant MBS as a
substrate (20).
Analyses of agonist-induced Rho translocation to the membrane fraction. Cytosolic and membrane fractions were prepared by differential centrifugation according to previously published protocols (20,21). Equal amounts of
membrane proteins were subjected to SDS/PAGE, transferred to polyvinylidine fluoride membrane, and probed with mouse anti-RhoA antibody followed
by incubation with HRP-labeled secondary antibody and subsequent detection
with ECL.
Analysis of RhoA activation by glutathione S-transferase-RhotekinRho binding domain pull down assay. [32P]-labeled VSMCs were exposed to
various agonists as detailed in Fig. 1. Cells were solubilized and equal amounts
of precleared lysate proteins (250 ␮g) were incubated with 25 ␮g (40 ␮l) of the
Rhotekin-Rho binding domain (RBD)-agarose slurry (UBI, Lake Placid, NY)
and incubated for 1 h at 4°C with shaking. The agarose beads were washed
with wash buffer according to the manufacturer’s protocol and resuspended in
20 ␮l Laemmli sample buffer followed by SDS/PAGE and Western blot analysis
with RhoA antibody.
Immunofluorescence studies on actin cytoskeleton organization. Serum-starved VSMCs were stimulated with thrombin (1 unit/ml) for 30 min or
treated first with insulin (100 nmol/l) for 10 min followed by thrombin. In the
inhibitor studies, cells were treated with the inhibitors for 60 min before
exposure to insulin and thrombin, as detailed in the figure legend. At the end
of the incubation, cells were rinsed with PBS, fixed for 10 min with 3.7%
formaldehyde in PBS, permeabilized in 0.1% Triton X-100 in PBS for 5 min, and
thrice rinsed in PBS. Cells were double stained with DNase-1 Texas Red
conjugate and fluorescin isothiocyanate (FITC)-conjugated smooth muscle
␣-actin antibody to colocalize monomeric globular actin (G-actin) and polymerized filamentous actin (F-actin), respectively (25). The chamber slides
were examined with an inverted fluorescence microscope (Eclipse TE-300;
Nikon). Images were collected with a Sony 3CCD DXC9000 video camera at
100⫻ magnification and digitally imported into Adobe photoshop 5.0 and
analyzed. For each area examined, images of FITC-actin and Texas RedDNase I fluorescence were collected. The time of measurements and image
capturing and the image intensity gain at both wavelengths were optimally
adjusted and kept constant between each treatment. The ratio of fluorescence
of FITC-actin and Texas Red-DNase I (F-actin–to–G-actin ratio) used to
DIABETES, VOL. 51, JULY 2002
FIG. 1. NO and cGMP inhibitors abolish the inhibitory effect of insulin
on RhoA translocation. VSMCs were pretreated with 1 mmol/l L-NMMA,
RpcGMP (100 ␮mol/l), and 8-bromo-cGMP (1 mmol/l) for 30 min
followed by insulin (100 nmol/l) for 10 min and then stimulated with
thrombin (1 unit/ml) for 5 min. Equal amounts of membrane proteins
were subjected to immunoblot analyses. A: Representative autoradiogram is shown. B: Quantitation of membrane-associated RhoA content
by densitometry. Results from different experiments are expressed
relative to control, which was assigned a value of 1. *P < 0.05 vs. basal;
**P < 0.05 vs. thrombin; ***P < 0.05 vs. insulin3thrombin.
quantify actin cytoskeleton organization was calculated for at least 20 cells in
each experimental condition and expressed as percentage of the ratio
obtained under control conditions. A decrease in the F-actin–to–G-actin ratio
was assumed to represent depolymerization of actin filaments.
Assay of geranylgeranyl transferase-1 activity. Geranylgeranyl transferase (GGTase)-1 activity was assayed in vitro using a modified method of
Moores et al. (26) with modifications (27). The enzymatic reaction was
initiated by adding 5-␮l aliquots of diluted VSMC extracts (0.5 mg/ml) to 45 ␮l
of assay buffer (50 mmol/l HEPES, 5 mmol/l MgCl2, 5 mmol/l DTT, 100 nmol/l
substrate Ras-CVLL, and 100 nmol/l [3H]geranylgeranyl pyrophosphate (15
mCi/mmol), pH 7.5. After 30 min incubation at 37°C, the reaction was stopped
with 1 ml ice-cold 1 mol/l HCl in ethanol, incubated on ice for 15 min, and then
the samples were filtered through Whatmann GF/C glass-fiber filters. Radioactivity bound to filters was measured by liquid scintillation spectrometry
(27).
Measurement of MBP activity. Myosin-enriched fractions of VSMCs were
prepared as described previously (21). MBP activity was assayed using
[32P]-labeled MLCs as a substrate (21). [32P]-labeled MLC was prepared
according to the published protocol (28) by incubating MLC (0.8 mg/ml) with
purified MLC kinase (50 ␮g/ml), 0.1 mg/ml calmodulin, and 50 ␮mol/l
[␥-32P]ATP.
Statistics. The results are presented as means ⫾ SE of three to six
independent experiments, each performed in triplicate. ANOVA with a Student-Newman-Keuls posthoc test in conjunction with a Tukey test was used to
compare mean basal values with those after various treatments. A P value
⬍0.05 was considered statistically significant.
RESULTS
Insulin inhibits thrombin/phenylephrine-mediated
RhoA translocation. Inhibitors of nitric oxide synthase
(NOS) and cGMP pathway abolish insulin-mediated RhoA
inactivation. As seen in Fig. 1, a considerable amount of
RhoA was present in the membrane fraction under basal
conditions. Thrombin (1 unit/ml) or phenylephrine (PE)
2257
INSULIN CAUSES ACTIN DISORGANIZATION VIA Rho INACTIVATION
(data not shown) treatment for 5 min caused a threefold
increase in membrane-associated RhoA content. Pretreatment with 100 nmol/l insulin for 10 min prevented thrombin-mediated increase in membrane RhoA content. Insulin
by itself caused a 40% decrease in membrane-associated
RhoA when compared with basal RhoA levels. The agonist-mediated increase in membrane-associated RhoA was
accompanied by a concomitant decrease in cytosolic
RhoA (data not shown). However, this observation was
not always consistent, probably because a large fraction of
the total Rho protein still remained cytosolic, as previously reported (29).
To evaluate the role of nitric oxide (NO)/cGMP pathway
in insulin inhibition of Rho translocation, VSMCs were
pretreated with 1 mmol/l NG-monomethyl L-arginine acetate (L-NMMA) (a NOS inhibitor) and 100 mmol/l Rp-8
CPT-cyclic guanosine monophosphate (RpcGMP) (a
cGMP antagonist) for 30 min followed by incubation with
100 nmol/l insulin for 10 min and then exposed to thrombin (1 unit/ml) for 5 min. VSMCs were examined for
translocation of RhoA and Rho kinase activity in the
anti-ROK-␣ immunoprecipitates. Both L-NMMA and
RpcGMP prevented the inhibitory effect of insulin on
thrombin-induced RhoA translocation (Fig. 1A, lanes 5
and 6 versus lane 4 and Fig. 1B). L-NMMA and RpcGMP
alone did not significantly alter RhoA content in the
membrane fraction (data not shown). Furthermore, exposure of VSMCs to 8-bromo-cGMP, a cGMP agonist, abolished thrombin-mediated RhoA translocation and
decreased membrane-associated RhoA content below
basal values (Fig. 1A, lane 7 versus lane 3 and Fig. 1B), as
seen with insulin treatment (Fig. 1A, lane 4). The concentrations of inhibitors used were determined in preliminary
experiments. L-NMMA at 1 mmol/l and Rp-CPT-cGMP at
100 ␮mol/l were most effective in inhibiting the effect of
insulin on Rho, Rho kinase, and MBP. Glut-1 and the ␣1
subunit of Na⫹/K⫹ ATPase, used as an internal control for
membrane proteins, revealed equal staining in all lanes
(data not shown).
NO and cGMP inhibitors prevent insulin-mediated
inactivation of Rho kinase activity. As detailed in our
recent publications (20,21), insulin treatment for 10 min
caused a 35% decrease in basal Rho kinase activity (Fig. 2).
Thrombin increased Rho kinase activity by 40% over the
basal values, an effect that was prevented by preincubation with insulin. Pretreatment with L-NMMA and RpcGMP
before insulin abolished insulin’s inhibitory effect and
restored thrombin’s effect on Rho kinase activity (Fig. 2).
These inhibitors alone had very little effect on basal Rho
kinase activity. Furthermore, treatment with 8-bromocGMP decreased thrombin-stimulated Rho kinase activity
to levels below basal values (Fig. 2).
NO/cGMP pathway mediates insulin-induced decrease in MBSThr695 phosphorylation. Recent studies
have shown that Rho kinase phosphorylates MBS on the
specific sites that influence MBP enzymatic activity
(28,30). Therefore, we examined the effect of insulin on
MBSThr695 phosphorylation. MBS is phosphorylated to a
considerable extent at Thr695 in the basal state. Exposure
to insulin for 10 min decreased basal MBSThr695 phosphorylation (Fig. 3A, lane 2 and Fig. 3B) and completely
abolished the thrombin-mediated increase in MBSThr695
2258
FIG. 2. NO and cGMP inhibitors abolish insulin-mediated inactivation
of Rho kinase activity. VSMCs were exposed to agents as detailed in
Fig. 1. Equal amounts of cell lysate proteins (100 ␮g) were immunoprecipitated with anti–Rok-␣ antibody. Rho kinase activity was assayed
using MBS as a substrate. Results are the means ⴞ SE of five separately
performed experiments. *P < 0.05 vs. basal; **P < 0.05 vs. thrombin;
***P < 0.05 vs. insulin3thrombin.
phosphorylation (Fig. 3A, lane 4 versus lane 3 and Fig.
3B). Pretreatment with L-NMMA and RpcGMP prevented
the insulin-mediated decrease in MBSThr695 phosphorylation and restored the effect of thrombin on MBS Thr695
phosphorylation (Fig. 3A, lanes 5 and 6 versus lane 4 and
Fig. 3B). In contrast, pretreatment with 8-bromo-cGMP, a
cyclic GMP agonist, prevented thrombin-mediated MBSThr695 phosphorylation in a manner comparable to insulin
(Fig. 3A, lane 7 and Fig. 3B). These inhibitors alone had
very little effect on basal MBSThr695 phosphorylation (Fig.
3A, lanes 8 and 9 and Fig. 3B). The alterations in MBSThr695
phosphorylation were accompanied by reciprocal changes
in MBP activation. Thus, the insulin-mediated reduction in
MBSThr695 phosphorylation was accompanied by a 65%
increase in MBP activity over the basal values. Moreover,
insulin prevented the inhibitory effects of thrombin on
MBP activation (Fig. 3C). Both L-NMMA and RpcGMP,
which increased MBSThr695 phosphorylation, prevented
the stimulatory effect of insulin on MBP activation (Fig.
3C).
Insulin inhibits Rho signaling by altering posttranslational modification of RhoA via the NO/cGMP signaling pathway. Recent studies indicate that
phosphorylation of RhoA by cAMP-dependent protein
kinase A as well as by cGMP-dependent protein kinases
impairs its biological activity (13,14), whereas geranylgeranylation of RhoA by GGTases is required for its activation
by agonists (5). To explore the possibility that insulin may
be affecting these two major processes of posttranslational
modification to cause Rho inactivation, we examined
GGTase-1 activity, the major isoform present in VSMCs,
and examined the effect of the NO/cGMP signaling pathway on GGTase-1 activity. As seen in Fig. 4, thrombin
treatment for 10 min resulted in a 1.8-fold increase in
GGTase-1. Preexposure to 10 nmol/l insulin for 10 min
abolished the thrombin-induced increase in GGTase-1
activity. Insulin alone did not affect basal levels of
GGTase-1 activity. Pretreatment with RpcGMP and
L-NMMA abolished the inhibitory effect of insulin on
GGTase-1 and restored the enzyme activity to values
DIABETES, VOL. 51, JULY 2002
N. BEGUM, O.A. SANDU, AND N. DUDDY
FIG. 4. Insulin pretreatment inhibits thrombin-mediated increase in
GGTase-1 activity. Inhibitors of NOS and cGMP block the insulin effect.
VSMCs were exposed to agents shown in Fig. 1. GGTase-1 activity was
measured using [3H]GGPP and Ras CVLL as a substrate. Results are the
means ⴞ SE of four independent experiments performed in duplicate.
*P < 0.05 vs. basal; **P < 0.05 vs. thrombin; ***P < 0.05 vs. insulin
pretreatment3thrombin.
An analyses of in vivo RhoA phosphorylation status
indicated that RhoA is considerably phosphorylated in the
unstimulated VSMCs (Fig. 6A, lane 1 and Fig. 6B). Insulin
caused a further twofold increase in RhoA phosphorylation when present alone or added before thrombin (Fig.
6A, lanes 2 and 4, and Fig. 6B). However, wortmaninn, a
phosphatidylinositol-3 kinase (PI3-kinase) inhibitor,
L-NMMA, and RpcGMP all decreased RhoA phosphorylation in untreated insulin-stimulated and insulin-exposed
thrombin-stimulated cells (Fig. 6A, lanes 5–11 and Fig.
6B), whereas 8-bromo-cGMP increased RhoA phosphorylation (Fig. 6A, lane 12 and Fig. 6B). ODQ, a guanyl cyclase
inhibitor, abolished RhoA phosphorylation when added
before insulin treatment (Fig. 6A, lane 13 and Fig. 6B).
RhoA antibody, pretreated with peptide antigen, did not
immunoprecipitate [32P]-labeled RhoA, suggesting that the
protein phosphorylated is RhoA and not some other
protein. Further confirmation of the phosphorylation of
FIG. 3. Inhibitors of NOS and cGMP signaling block insulin-mediated
decrease in MBSThr695 site-specific phosphorylation and MBP activation. VSMCs were exposed to inhibitors as detailed in Fig. 1. Equal
amounts of proteins from trichloracetic acid lysates were subjected to
SDS-PAGE followed by Western blot analysis with anti–phospho-threonine (695) MBS antibody. A: Representative autoradiogram is shown.
B: Quantitation by densitometric analyses. C: MBP activity was assayed in myosin-enriched fractions isolated from VSMCs exposed to
various inhibitors as detailed in Fig. 1. [32P]-labeled MLC was used as
a substrate. Results are the means ⴞ SE of three experiments performed in duplicate. *P < 0.05 vs. basal; **P < 0.05 vs. thrombin;
***P < 0.05 vs. insulin pretreatment B3thrombin.
comparable to thrombin-stimulated cells (Fig. 4). In contrast, 8-bromo-cGMP mimicked insulin’s effect by blocking
thrombin-induced GGTase-1 activity. Prolonged incubations for 24 h, with higher concentrations of insulin to
mimic hyperinsulinemia, have been reported to increase
GGTase-1 activity (27). Decreased GGTase-1 activity was
accompanied by a decrease in geranylgeranylated Rho A in
the membrane fraction in insulin and 8-bromo-cGMP–
treated cells (Fig. 5). This decrease was prevented by
RpcGMP and L-NMMA.
DIABETES, VOL. 51, JULY 2002
FIG. 5. Insulin inhibits geranylgeranylation of RhoA. VSMC lysates
were extracted with Triton X-114. Detergent soluble phase and aqueous phase were immunoprecipitated with anti-RhoA antibody followed
by SDS/PAGE, Western blot analyses with anti-RhoA antibody, and
quantitation by densitometry. Results are the means ⴞ SE of four
independent experiments performed in duplicate. *P < 0.05 vs. basal;
**P < 0.05 vs. thrombin; ***P < 0.05 vs. insulin pretreatment3
thrombin.
2259
INSULIN CAUSES ACTIN DISORGANIZATION VIA Rho INACTIVATION
FIG. 6. Inhibitors of PI3-kinase and NOS/cGMP signaling pathway decrease RhoA phosphorylation. [32P]-labeled VSMCs were pretreated with
wortmannin (100 nmol/l), L-NMMA, RpcGMP, 8-bromo-cGMP, and ODQ (100 ␮mol/l) for 30 min followed by treatment with and without insulin
(100 nmol/l) for 10 min and then exposed to thrombin (1 unit/ml) for 10 min. Equal amounts of precleared lysate proteins (200 ␮g) were
immunoprecipitated with anti-RhoA antibody followed by SDS/PAGE and autoradiography. A: Representative autoradiogram is shown. B: [32P]
signal from the autoradiograms was quantitated by densitometry and normalized for variations in immunoprecipitated RhoA by dividing the
intensity of [32P] signal with that of RhoA protein. Results are the means ⴞ SE of three separate experiments expressed relative to untreated
control, which was assigned a value of 1. C: [32P]-labeled VSMC lysates (250 ␮g protein) were incubated with 40 ␮l of GST-Rhotekin-RBD-agarose
for 1 h at 4°C. Beads were sedimented by brief centrifugation and washed with Rho wash buffer containing protease and phosphatase inhibitors.
Bound RhoA eluted with Laemmeli’s sample buffer followed by SDS/PAGE and Western blot analysis with anti-RhoA antibody (top panel).
Phosphorylation status of RBD-bound RhoA was examined by autoradiography (middle panel). To verify equal loading 50 ␮g of cell lysates were
subjected to Western blot analysis with RhoA antibody (bottom panel). A representative autoradiogram is shown. D: [32P] signal from the
autoradiograms was quantitated by densitometry and normalized for RhoA bound to Rhotekin by dividing the intensity of [32P] signal with that
of RhoA protein. Results are the means ⴞ SE of three separate experiments expressed relative to control, which was assigned a value of 1.
RhoA and its relationship to RhoA activation came from
studies in which active GTP-bound RhoA was precipitated
from [32P]-labeled VSMCs using glutathione S-transferase
(GST)-Rhotekin-RBD pull-down assay. Only active Rho
binds to Rhotekin-RBD. Bound RhoA was eluted and
subjected to SDS/PAGE and phosphorylation status was
2260
determined by autoradiography. The amount of RhoA
retained on GST-Rhotekin was determined by probing
membranes with Rho antibody. The phosphorylation status of active RhoA was quantitated by densitometric
analysis of phosphorylated RhoA bands and corrected for
Rho protein by dividing the [32P] signal with protein signal.
DIABETES, VOL. 51, JULY 2002
N. BEGUM, O.A. SANDU, AND N. DUDDY
As shown in Fig. 6C, upper panel, a small amount of RhoA
was associated with GST-Rhotekin-RBD under basal and
insulin-stimulated conditions. Thrombin induced a more
than threefold increase in active RhoA, which was 70% less
phosphorylated when compared with untreated control
(Fig. 6C middle panel, lane 3 and Fig. 6D). Both insulin and
8-bromo-cGMP prevented RhoA activation by thrombin
and increased Rho phosphorylation (Fig. 6C, top and
middle panels, lanes 4 and 12 versus lane 3 and Fig. 6D).
Pretreatment with wortmannin, L-NMMA, RpcGMP, and
ODQ all prevented the effect of insulin, restored Rho
activation by thrombin, and decreased Rho phosphorylation (Fig. 6C, top and middle panels, lanes 5–11 versus
lane 4 and Fig. 6D). The above effects were specific
because RhoA detection could be prevented by inclusion
of peptide antigen during incubation with the antibody
(see Fig. 6C, top and middle panels, lanes 13 and 14).
Insulin-mediated inactivation of Rho signaling is accompanied by inhibition of actin cytoskeleton organization. We also examined the ability of insulin to inhibit
actin stress fiber formation in low-passage VSMCs because
the higher passage was recently reported to increase actin
stress fiber formation (31). Figure 7 shows VSMCs double
stained with FITC-conjugated actin antibody and Texas
Red-labeled DNase I to colocalize F-actin and monomeric
G-actin, respectively. In untreated VSMCs, a few thin actin
filaments were observed (Fig. 7A). Insulin treatment
caused a decrease in actin filaments (Fig. 7B) and decreased the ratio of F-actin to G-actin by 30%. In contrast,
treatment with thrombin caused formation of dense and
organized network of thick and parallel actin stress fibers
(Fig. 7C), which was prevented by pretreatment with
insulin (Fig. 7D). Similar results were observed when PE
was used as an agonist (data not shown). More importantly, preincubation with RpcGMPS and L-NMMA for 1 h
decreased the insulin effect and restored thrombin-mediated actin fiber reorganization (Fig. 7E and F), but the
actin fibers were not as dense as compared with thrombin
alone. 8-Bromo-cGMP, a cGMP agonist, mimicked insulin’s
effect by decreasing thrombin-induced actin stress fiber
formation (Fig. 7G), as evidenced by increased TexasRed–labeled DNase I staining and a reduced ratio of
F-actin to G-actin.
FIG. 7. Insulin inhibits thrombin-induced actin cytoskeleton organization via the NO/cGMP signaling pathway. Serum-starved VSMCs grown
on eight-well chamber slides were exposed to various inhibitors as
detailed in Fig. 1. Cells were first stained with Texas-Red DNase I
conjugate for 20 min at RT, washed with PBS, and incubated with
monoclonal anti–FITC-conjugated smooth muscle ␣-actin antibody to
colocalize monomeric G-actin and polymerized F-actin, respectively.
Actin polymerization was visualized using a Nikon inverted fluorescent
microscope fitted with dual filters to simultaneously capture the
images of FITC-actin and Texas-Red-DNase I fluorescence at a 100ⴛ
magnification.
DISCUSSION
The results presented in this study indicate that insulin
negatively regulates Rho signaling by inhibiting thrombininduced RhoA activation via the NO/cGMP signaling pathway. More importantly, insulin, via NO/cGMP, affects
posttranslational modification of RhoA through a combination of an upregulation of Rho phosphorylation, resulting in inhibition of Rho activation and translocation, and a
decrease in RhoA geranylgeranylation due to inhibition of
GGTase-1 activity. Insulin inactivation of RhoA via NO/
cGMP inhibits Rho kinase activity, decreases MBSThr695
phosphorylation, and activates MBP, leading to inhibition
of actin cytoskeleton organization, which may contribute
to the well-known vasodilator actions of insulin.
Several lines of evidence presented in this study suggest
that inactivation of RhoA by insulin is mediated by multiple inputs from the NO/cGMP signaling pathway. First,
treatment with L-NMMA, a NOS inhibitor, and Rp-8-bromoDIABETES, VOL. 51, JULY 2002
cGMPS, a cGMP antagonist, prevented insulin inactivation
of RhoA as well as Rho kinase inactivation, whereas
8-bromo-cGMP, a cGMP agonist, mimicked the inhibitory
effect of insulin on thrombin-induced Rho activation/
translocation and Rho kinase activity. Second, both insulin
and 8-bromo-cGMP inhibit the thrombin-induced increase
in GGTase-1 activity, which is effectively blocked by
L-NMMA and Rp-8-bromo-cGMPS. Third, wortmannin, a
PI3-kinase inhibitor, L-NMMA, Rp-8-bromo-cGMPS, and
ODQ, a guanylcyclase inhibitor, all block RhoA phosphorylation and restore RhoA activation by thrombin. It should
be noted that although wortmannin prevented the insulinmediated increase in RhoA phosphorylation and MBP
activation, it failed to prevent insulin-mediated inactivation of Rho kinase (21). The reasons for this discrepancy
are not clear. It is plausible that Rho kinase inhibition by
2261
INSULIN CAUSES ACTIN DISORGANIZATION VIA Rho INACTIVATION
insulin may be mediated by other unidentified signaling
molecules in addition to inactivation of RhoA.
Earlier studies have shown that inhibition of protein
geranylgeranylation causes superinduction of inducible
NOS (iNOS) by interleukin-1␤ in VSMCs (32), suggesting
that these two pathways regulate the activation status of
each other via a complex cross talk. We have also reported
that insulin increases iNOS protein expression as well as
cGMP generation in VSMCs (21,23,24). Thus, insulininduced iNOS expression/cGMP generation attenuates
RhoA signaling by direct inactivation of Rho by altering
the phosphorylation as well as isoprenylation of RhoA. In
contrast, chronic activation of RhoA by vasoconstrictors,
as well as by disease states such as hypertension and
diabetes, may inhibit iNOS/cGMP generation by specifically blocking insulin signaling via the insulin receptor
substrate-1/PI3-kinase pathway (21,23,24). We and others
have recently reported that the PI3-kinase pathway mediates NOS activation in both VSMCs and endothelial cells.
In the endothelial cells, NOS is activated via phosphorylation by the PI3-kinase downstream target, Akt (33).
Several studies have reported that posttranslational
modification of RhoA by geranylgeranylation and phosphorylation plays a major role in RhoA activation (13–16).
Although prenylation of proteins is considered a stable
modification, we have observed that acute insulin treatment inhibits the thrombin-induced increase in GGTase-1
activity and is prevented by L-NMMA and RpcGMPS.
GGTase-1 is known to be phosphorylated. It is plausible
that thrombin may phosphorylate GGTase-1 to cause its
activation, and insulin may inhibit GGTase-1 activity by
decreasing the phosphorylation via an activated phosphatase, presumably MBP. Due to the nonavailability of
GGTase-1 antibody, we did not measure the phosphorylation status of GGTase-1. Further studies are needed to
identify the kinase and phosphatase that may mediate
thrombin-induced GGTase-1 phosphorylation and dephosphorylation. Alternatively, insulin may detach the geranylgeranylated moiety from RhoA by proteolysis.
In summary, we have demonstrated that insulin inactivates Rho and prevents Rho signaling by affecting posttranslational modification of RhoA via the NO/cGMP
pathway to cause Rho kinase inhibition, resulting in a
decrease in MBSThr695 phosphorylation, which leads to
MBP activation and actin cytoskeleton disorganization.
ACKNOWLEDGMENTS
This work was supported by the American Heart Association Established Investigator Award, an American Diabetes Association research grant, and medical education
funds from Winthrop University Hospital.
REFERENCES
1. Hall A: Regulation of Rho proteins. Science 279:509 –514, 1998
2. Seasholtz TM, Majumdar M, Brown JH: Rho as a mediator of G proteincoupled receptor signaling. Am Soc Pharm Exptl Ther 55:949 –956, 1999
3. Fukata Y, Kaibuchi K, Amano M, Kaibuchi K: Rho-Rho-kinase pathway in
smooth muscle contraction and cytoskeletal reorganization of non-muscle
cells. Trends Pharmacol Sci 222:32–39, 2001
4. Somlyo AP, Wu X, Walker LA, Somlyo AV: Pharmacomechanical coupling
and the role of calcium, G-proteins, kinases and phosphatases. Rev Physiol
Biochem.& Pharmacol 134:201–234, 1999
5. Gong MC, Iizuka K, Nixon G, Browne JP, Hall A, Eccleston JF, Sugai M,
Kobayashi S, Somlyo AV, Somlyo AP: Role of guanine nucleotide-binding
2262
proteins–ras-family or trimeric proteins or both–in Ca2⫹ sensitization of
smooth muscle Proc Natl Acad Sci U S A 93:1340 –1345, 1996
6. Loirand G, Cario-Toumaniantz C, Chardin P, Pacaud P. The Rho-related
protein Rnd1 inhibits Ca2⫹: sensitization of rat smooth muscle. J Physiol
(Lond) 516:825– 834, 1999
7. Kimura K, Ito M, Amano M, Ichihara K, Fukata Y, Nakafuku M, Yamamori
B, Feng J, Nakano T, Okawa K, Iwamatsu A, Kaibuchi K: Regulation of
myosin phosphatase by Rho and Rho-associated kinase (Rho kinase).
Science 273:245–248, 1996
8. Kureishi Y, Kobayashi S, Amano M, Kimura K, Kanaide H, Nakano T,
Kaibuchi K, Ito M: Rho-associated kinase dir Signal transduction and
regulation in smooth muscle directly induces smooth muscle contraction
through myosin light chain phosphorylation. J Biol Chem 272:12257–
12260, 1997
9. Somlyo AP, Somlyo AV: Signal transduction and regulation in smooth
muscle. Nature 372:231–236, 1994
10. Lee MR, Li L, Kitazawa T: cGMP causes Ca2⫹ desensitization in vascular
smooth muscle cells by activating the myosin light chain phosphatase
J Biol Chem 272:5063–5068, 1997
11. Fleming IN, Elliot CM, Exton JH: Differential translocation of Rho family
GTPases by lysophosphatidic acid, endothelin-1, and platelet-derived
growth factor. J Biol Chem 271:33067–33073, 1996
12. Gong MC, Fugihara H, Somlyo AV, Somlyo AP: Translocation of rhoA
associated with Ca2⫹ sensitization of smooth muscle. J Biol Chem
272:10704 –10709, 1997
13. Moomaw J, Casey P: Mammalian protein geranylgeranyltransferase: subunit composition and metal requirements. J Biol Chem 267:17438 –17443,
1992
14. James G, Goldstein J, Brown MJ: Polylysine and CVIM sequences of
K-RasB dictate specificity of prenylation and confer resistance to benzodiazepine peptidomimetic in vitro. Biol Chem 270:6221– 6226, 1995
15. Sauzeau V, Jeune HL, Cario-Toumaniantz, Smolenski A, Lohmann SM,
Chardin P, PacauP, Loirand G. Cyclic GMP-dependent: protein kinase
signaling pathway inhibits RhoA-induced Ca2⫹ sensitization of contraction in vascular smooth muscle. J Biol Chem 275:21722–21727, 2000
16. Dong JM, Leung T, Manser E, Lim L: A rapid and sensitive method for the
quantitation of microgram quantities of proteins utilizing the principle of
protein dye binding J Biol Chem 273:22554 –22562, 1998
17. Sowers JR, Epstein M: Diabetes mellitus and associated hypertension,
vascular disease, and nephropathy: an update. Hypertension 6:869 – 879,
1995
18. Hsueh WA, Law RE: Insulin signaling in the arterial wall. Am J Cardiol
84:21J–24J, 1999
19. Uehata M, Ishizaki T, Satoh H, Ono T, Kawahara T, Morishita T, Tamakawa
H, Yamagami K, Inui J, Maekawa M, Narumiya S: Calcium sensitization of
smooth muscle mediated by a Rho-associated protein kinase in hypertension. Nature 389:990 –994, 1997
20. Sandu OA, Ragolia, Begum N: Diabetes is accompanied by impaired
insulin-mediated myosin-bound phosphatase activation and vascular
smooh muscle cell relaxation due to defective Rho kinase and NOS
signaling. Diabetes 49:2178 –2189, 2000
21. Begum N, Duddy N, Sandu OA, Reinzie J, Ragolia L: Regulation of
myosin-bound protein phosphatase by insulin in vascular smooth muscle
cells: evaluation of the role of Rho kinase and phosphatidylinositol-3kinase dependent signaling pathways. Mol Endocrinol 14:1365–1376, 2000
22. Begum N, Ragolia L, Rienzie J, McCarthy M: Regulation of MKP-1 expression by insulin in vascular smooth muscle cells: evaluation of the role of
the nitric oxide signaling pathway and potential defects in hypertension.
J Biol Chem 273:25164 –25170, 1998
23. Begum N, Song Y, Rienzie J, Ragolia L: Vascular smooth muscle cell
growth and insulin regulation of mitogen-activated protein kinase in
hypertension. Am J Physiol 275:C42–C49, 1998
24. Begum N, Ragolia L: High glucose and chronic insulin inhibit MKP-1
expression in VSMCs by blocking iNOS induction via p38 MAPK activation
Am. J Physiol Cell Physiol 278:C81–C91, 2000
25. Knowles GC, McCulloh CAG: Simultaneous localization and quantification
of relative G and F actin content: optimization of fluorescence labeling
methods. J Histochem Cytochem 40:1605–1612, 1992
26. Moores SL, Schaber MD, Mosser SD, Rands E, O’Hara MB, Garsky VM,
Marshall MS, Pompliano DL, Gibbs JB: Sequence dependence of protein
isoprenylation. J Biol Chem 266:14603–14610, 1991
27. Golovchenko I, Goalstone ML, Watson P, Brownlee M, Draznin B: Hyperinsulinemia enhances transcriptional activity of nuclear factor-kappaB
induced by angiotensin II, hyperglycemia, and advanced glycosylation end
products in vascular smooth muscle cells. Circ Res 87:746 –752, 2000
28. Feng J, Ito M, Ichikawa K, Isaka N, Nishikawa M, Hartshorne DJ, Nakano
DIABETES, VOL. 51, JULY 2002
N. BEGUM, O.A. SANDU, AND N. DUDDY
T: Inhibitory phosphorylation site for Rho-associated kinase on smooth
muscle myosin phosphatase. J Biol Chem 274:37385–37390, 1999
29. Seasholtz TM, Majumdar M, Kaplan DD, Brown JH: Rho as a mediator of
G protein-coupled receptor signaling. Circ Res 84:1186 –1193, 1999
30. Ichikawa K, Ito M, Hartshorne DJ: Phosphorylation of the large regulatory
subunit of myosin phosphatase and inhibition of phosphatase activity
J Biol Chem. 271:4733– 4740, 1996
31. Sauzeu V, Jeune HL, Cario-Toumaniantz C, Vaillant N, Gadeau AP, Desgranges C, Scalbert E, Chardin P, Pacaud P, Loirand G: P2Y1, P2Y2, P2Y4,
and P2Y6 receptors are coupled to Rho and Rho kinase activation in
DIABETES, VOL. 51, JULY 2002
vascular myocytes. Am J Physiol Heart Circ Physiol 278:H1751–H1761,
2001
32. Finder JD, Litz JL, Blaskovich MA, McGuire TF, Qian Y, Hamilton AD,
Davies P, Sebti SM: Inhibition of protein geranylgeranylation causes a
superinduction of nitric-oxide synthase-2 by interleukin-1beta in vascular
smooth muscle cells. J Biol Chem 272:13484 –13488, 1997
33. Zeng G, Nystrom FH, Ravichandran LV, Cong LN, Kirby M, Mostowski H,
Quon MJ: Roles for insulin receptor, PI3-kinase, and Akt in insulinsignaling pathways related to production of NO in human vascular
endothelial cells. Circulation 101:1539 –1545, 2000
2263