heparin responses in vascular smooth muscle cells involve cgmp-dependent protein kinase (pkg)

Heparin Responses in VascularSmooth Muscle Cells InvolvecGMP-Dependent Protein Kinase(PKG)ALBERT C. GILOTTI,1 WUTIGRI NIMLAMOOL,1 RAYMOND PUGH,2 JOSHUA B. SLEE,1

TRISTA C. BARTHOL,1 ELIZABETH A. MILLER,1 AND LINDA J. LOWE-KRENTZ1*1Department of Biological Sciences, Lehigh University, Bethlehem, Pennsylvania2Department of Chemistry, Lehigh University, Bethlehem, Pennsylvania

Published data provide strong evidence that heparin treatment of proliferating vascular smooth muscle cells results in decreased signalingthrough the ERK pathway and decreases in cell proliferation. In addition, these changes have been shown to bemimicked by antibodies thatblock heparin binding to the cell surface. Here, we provide evidence that the activity of protein kinase G is required for these heparineffects. Specifically, a chemical inhibitor of protein kinase G, Rp-8-pCPT-cGMS, eliminates heparin and anti-heparin receptor antibodyeffects on bromodeoxyuridine incorporation into growth factor-stimulated cells. In addition, protein kinase G inhibitors decrease heparineffects on ERK activity, phosphorylation of the transcription factor Elk-1, and heparin-induced MKP-1 synthesis. Although transient, thelevels of cGMP increase in heparin treated cells. Finally, knock down of protein kinase G also significantly decreases heparin effects ingrowth factor-activated vascular smooth muscle cells. Together, these data indicate that heparin effects on vascular smooth muscle cellproliferation depend, at least in part, on signaling through protein kinase G.

J. Cell. Physiol. 229: 2142–2152, 2014. © 2014 Wiley Periodicals, Inc.

Following injury, migration of vascular smooth muscle cells(VSMCs) from the tunica intima into the vessel lumen andsubsequent hyperplasia, are key events in the development ofatherosclerosis. Molecules that can decrease VSMCproliferation have been examined for possible treatments toslow disease advancement. The discovery that heparinsuppresses VSMC growth was reported more than 30 yearsago (Clowes and Karnovsky, 1977); yet the mechanism bywhich heparin treatment of VSMCs inhibits their proliferationremains unclear. Heparin blocks PKC-dependent c-fos induc-tion and activation of ERK, a MAPK activated in response tonumerous treatments of sub-cultured VSMCs (Castellotet al., 1989; Ottlinger et al., 1993). In addition, heparintreatment results in decreases in cyclin-dependent kinase 2activity by increasing levels of p27kip1 (Fasciano et al., 2005).However, sequestration of growth factors is not likely toexplain all of the effects of heparin on VSMCs (Reilly et al., 1989;Pukac et al., 1997; Savage et al., 2001; Blaukovitch et al., 2010).

VSMCs specifically bind and endocytose heparin (Castellotet al., 1985). This specific binding activity, in combination withheparin's effects on cell signaling pathways, supports a modelwhereby heparin binds to cell surface proteins and initiates itsown signaling pathways. To identify putative heparin receptorproteins, Patton et al. (1995) produced monoclonal antibodiesthat specifically inhibit heparin binding to cells in vitro. Theseantibodies identified a single heparin-binding cell surfaceprotein and act as agonists of heparin, mimicking both heparin'seffects on cell signaling and its anti-proliferative effects incultured VSMCs (Savage et al., 2001; Blaukovitch et al., 2010).

MAPK activity is regulated by the reversible phosphorylationof specific tyrosine and threonine residues, and active ERKaccumulation in the nucleus is critical in cell cycle progressionthrough G1 (Brunet et al., 1999) where the sustained ERKactivity results in Elk-1 phosphorylation (Shin et al., 2003).Dual-specificity protein phosphatases play important roles inERK inactivation, and MAPK Phosphatase-1 (MKP-1) localizesto the nucleus (Rohan et al., 1993) where it is able to regulateERK signaling in VSMCs. Loss of active ERK in the nucleusresults in decreased Elk-1 activity (Shin et al., 2003). Our

laboratory has demonstrated that heparin and the anti-heparinreceptor antibodies increaseMKP-1 levels in VSMCs, playing animportant role in heparin-induced ERK activity decreases(Blaukovitch et al., 2010). However, the signaling intermediatesbetween heparin's interaction with its receptor and inductionof MKP-1 expression remain unknown. One possibility is

Abbreviations: VSMC, vascular smooth muscle cell; MKP-1, MAPKphosphatase-1; IGF, insulin-related growth factor; iNOS, inducibleNOS; PKG, cGMP-dependent protein kinase; ANF or ANP, atrialnatriuretic factor or peptide; PMA, phorbal myristic acid; pElk,phospho Elk-1; IBMX, 3-isobutyl-1-methylxanthine; DAB, 3,30-diaminobenzidine; TBST, TBS with 0.1% Tween-20; TRITC,tetramethyl rhodamine isothiocyanate; PKA, cAMP-dependentprotein kinase.

Current address of Raymond Pugh is Penn State University, 403Wartik Laboratory, University Park, PA 16802.

Current address of Albert C. Gilotti is Toxological Sciences, MerckResearch Laboratories, Building K-15, Block F-2516, 2000Galloping Hill Road, Kenilworth, NJ 07033.

Current address of Joshua B. Slee is Department of Pediatrics, TheChildren’s Hospital of Philadelphia, Philadelphia, PA 19104; TheUniversity of Pennsylvania, Perelman School of Medicine,Philadelphia, PA 19104.

Contract grant sponsor: Public Health Service;Contract grant number: HL54269.

*Correspondence to: Linda J. Lowe-Krentz, Department ofBiological Sciences, Lehigh University, 111 Research Drive,Bethlehem, PA 18015.E-mail: [email protected]

Manuscript Received: 5 July 2013Manuscript Accepted: 20 May 2014

Accepted manuscript online in Wiley Online Library(wileyonlinelibrary.com): 9 June 2014.DOI: 10.1002/jcp.24677

ORIGINAL RESEARCH ARTICLE 2142J o u r n a l o fJ o u r n a l o f

CellularPhysiologyCellularPhysiology

© 2 0 1 4 W I L E Y P E R I O D I C A L S , I N C .

suggested by studies of insulin signaling (Begum et al., 1998;Jacob et al., 2002). These studies report the expression ofMKP-1 in VSMCs in response to insulin and insulin-relatedgrowth factor (IGF) where insulin and IGF induce theexpression of inducible NOS (iNOS), eventually increasing thelevels of cGMP in response to the NO activation of solubleguanylyl cyclase. The increase in cGMP levels was shown to besufficient to induce MKP-1 expression and attenuate ERKactivity. Similarly, leptin treatment induces decreased VSMCproliferation, and this depends on iNOS induction (Rodríguezet al., 2010). As well as in heparin treated cells, p27kip is inducedin response to cGMP rises and cGMP-dependent proteinkinase (PKG) activity in VSMCs (Sato et al., 2000).

Another agent that elevates cGMP in VSMCs is atrialnatriuretic factor or peptide (ANF or ANP). Upon ligandbinding, the ANP receptor undergoes a conformational changethat activates an intracellular guanylate cyclase domain therebyelevating levels of cGMP. Although better known for their rolein vasorelaxation, both ANP and cGMP have been shown todecrease the proliferation of VSMCs (Baldini et al., 2002;Tantini et al., 2005). These effects on VSMCs imply that cGMPmediates a coordinated control of acute cellular function.Elevations in cGMP (by cGMP analogues, activators ofguanylate cyclase, or ANP) induce MKP-1 expression insmooth muscle, mesangial, and endothelial cells by way of PKG(Sugimoto et al., 1996; Baldini et al., 2002; Jacob et al., 2002;Furst et al., 2005). This increase in MKP-1 expression results ina decrease of ERK activation, and thus provides a mechanismfor the anti-proliferative activity of ANP in VSMC andmesangialcells (Sugimoto et al., 1996; Baldini et al., 2002; Tantiniet al., 2005). As with heparin, ANP treatment induces increasesin p27kip1 levels (Tete et al., 2001).

Heparin and cGMP affect VSMCs similarly. First, both inhibitgrowth of VSMCs late in the G1 phase of the cell cycle. Second,the proximity of the endothelium to VSMCs in vivo provides asource for both endogenous heparin and cGMP-elevating agentssuch as NO. Endogenous heparin from endothelial cells couldmaintain quiescence in VSMCs (Castellot et al., 1981). Third, inreducing VSMC growth, both cGMP and heparin cause aninactivationofERKdue, at least inpart, to the inductionofMKP-1(Baldini et al., 2002; Blaukovitch et al., 2010). Because of thesimilarities in theway that heparin,ANP, andNO-induced cGMPincreases affect VSMCs, we hypothesize that heparin's cellulareffects are mediated through the second messenger cGMPtarget, PKG.Consistentwith this idea is evidence that reductionsin cGMP signaling occur with neointimal proliferation andvascular dysfunction in late-stage atherosclerosis (Melicharet al., 2004). Also consistent with this hypothesis is the fact thatexpression of constitutively active PKG inhibits VSMC prolifer-ation in response to high glucose (Wang and Li, 2009). In thepresent report, we present evidence that PKG activity is indeedrequired for heparin-induced decreases in VSMC ERK activity,Elk-1 phosphorylation, and VSMC proliferation.

Materials and MethodsMaterials

Cell culture chemicals, DMEM and MEM, 2.5% trypsin/EDTA,gelatin, heparin, penicillin/streptomycin, phorbal myristic acid(PMA), PDGF, and glutamate were obtained from Sigma ChemicalCo. (St. Louis, MO). Pre-tested FBS was obtained from Invitrogen(Gaithersburg, MD), Atlanta Biologicals (Atlanta, GA), or Biowest(St. Louis, MO). Anti-active ERK (rabbit, against phosphorylatedERK, but called active ERK in the text to distinguish it from themouse antibody) and anti-phospho Elk-1 (pElk) antibodies werefrom Cell Signaling (Beverly, MA). Anti-MKP-1 (V-15), anti-phospho ERK (pERK, mouse, employed when both pERK and pElkwere detected using double-immunofluorescence), and anti-PKG(a mixture of antibodies against PKG Ia and Ib was used) were

from Santa Cruz Biotechnology (La Jolla, CA). siRNA (speciesspecific) for PKG was also from Santa Cruz. Anti-smooth musclemyosin, and Extra-avidin-alkaline phosphataseTM were obtainedfrom Sigma Chemical Co. Biotin-labeled and fluorescent-taggedsecondary antibodies (in donkey or bovine, with minimal cross-reactivity) were from Jackson ImmunoResearch Laboratories, Inc.(West Grove, PA). 8-Br-cAMP, 8-Br-cGMP, the PKG inhibitorKT5823, 8-pCPT-cAMS and Rp-8-pCPT-cGMS, and Mowiol werefrom Calbiochem (EMD, San Diego, CA). cGMP ELISA kits werefrom R&D Systems, Inc. (Minneapolis, MN) or Cayman Chemical(Ann Arbor, MI).

Cell culture

A7r5 rat smooth muscle cells were obtained from ATCC(Rockville, MD). Porcine aortic smoothmuscle cells were obtainedfrom Clonetics, a division of BioWhitaker (Walkersville, MD) orCell Applications, Inc. (San Diego, CA). Commercially availableVSMCs were grown as recommended by the supplier and porcinecells were exchanged into MEM media over time beforeexperiments. For some experiments, VSMCs were isolated fromporcine aortas, characterized as smooth muscle, and cultured asdescribed previously (Savage et al., 2001; Blaukovitch et al., 2010).

BrdU incorporation assays

The effects of heparin and other reagents on cell growth wereexamined by monitoring BrdU incorporation as describedpreviously (Savage et al., 2001). Cells were routinely activated with1.5mg/ml PDGF after 48 h culture inmedia without serum. Prior toactivation, indicated cells were treated for 10min with 200mg/mlheparin, 100 nM ANP, 100mM 8-Br-cAMP, 100mM 3-isobutyl-1-methylxanthine (IBMX), or 100mM 8-Br-cGMP. cAMP-dependentprotein kinase (PKA) or PKG inhibitors, Rp-8-pCPT-cAMS and Rp-8-pCPT-cGMS, were added at a concentration of 2mM for 10minbefore the addition of heparin or analogous treatments. After 24–48 h, the cells were processed, and cell staining was visualized withthe addition of 0.66mg/ml 3,30-diaminobenzidine (DAB) in TBSwith 0.1% Tween-20 (TBST) with 0.8ml/ml hydrogen peroxide(Savage et al., 2001). Data collection and statistical analysis of theseexperiments were also carried out as described (Savageet al., 2001). Briefly, treatments were run in duplicate or triplicatewithin an experiment and tested in at least three independentexperiments. Data collection was carried out by individuals blindto experimental treatments. For BrdU incorporation in cellstreated with siRNA for knock down of PKG, the cell staining wasvisualized by Alexa 488 labeled secondary antibodies and intensityanalysis using Image J as described for immunofluorescence below.The relative effects of each treatment from each independentexperiment were analyzed by unpaired t-tests as appropriate.

Western blotting

For immunoblotting, cells starved as above were incubated with200mg/ml heparin, 100 nM ANP, 100mM 8-Br-cAMP, 100mM 8-Br-cGMP, or anti-heparin receptor antibodies (at 0.8mg/ml) for10–20min as noted, and then activated with PMA (at 50 ng/ml),10% serum, or 1.5mg/ml PDGF as noted. The PKG inhibitor, Rp-8-pCPT-cGMS, was added at a concentration of 2mM for 10minbefore the addition of heparin or analogous treatments. Cells wereharvested and ERK activity or MKP-1 levels were determined asdescribed previously (Savage et al., 2001; Blaukovitch et al., 2010).

Measurement of intracellular cGMP levels by ELISA

Porcine VSMCs or A7r5s were treated with heparin (200mg/ml),anti-receptor antibodies (0.8mg/ml), or ANP (100 nM) for theindicated time periods. Samples were extracted with ice-coldethanol, and concentrated to volumes required according to the

JOURNAL OF CELLULAR PHYSIOLOGY

H E P A R I N R E S P O N S E S I N V O L V E P K G 2143

kits. The concentration of GMP present in each sample wasmeasured according to the protocol of the kit used.

Indirect immunofluorescence analysis of Elkphosphorylation and ERK phosphorylation

A7r5 or porcine VSMC were cultured on gelatin-coated glasscover slips individual wells of a 6-well plate and starved as above.Starved cells were activatedwith 1.5mg/ml PDGF at 37°C for 10 or20min and cells were fixed with 4% paraformaldehyde for 15min,washed three times in PBS and permeabilized with 0.3% Triton X-100 in PBS for 5min. Some cell samples were treated with 200mg/ml heparin 10min prior to PDGF activation. Rp-8-pCPT-cGMS(2mM final) or KT5823 (0.2mM final) was added 10min prior toheparin in cell samples so labeled. Fixed and permeabilized cellswere treated with primary antibodies against pElk and pERK (1:100in 1% BSA/PBS) overnight at 4°C. In some cases, cells were treatedwith heparin as above without PDGF and stained for MKP-1. Afterthree washes with PBS, tetramethyl rhodamine isothiocyanate(TRITC)-labeled anti-rabbit and FITC or Alexa 488-labeled anti-mouse secondary antibodies (1:200) were added and incubated for

1 h at room temperature in the dark. Cover slips were washedthree times in PBS, mounted with mowoil or Vectashield (VectorLaboratories, Burlingame, CA), observed using a Nikon eclipse TE2000-U fluorescence microscope (Nikon, Tokyo, Japan), andmicrographs were captured with a SPOT RT KE camera.Fluorescent intensity in cell nuclei was determined using Image J.

PKG knockdown

A7r5 cells were transfected using 20mg/ml of siRNA in the Bio-Rad Gene Pulser X-Cell System (Hercules, CA). Briefly, 100mmconfluent plates of cells were trypsinized, rinsed with PBS,suspended in HEPES-buffered saline, electroporated with the pre-set HeLa protocol (modified to 170V), seeded onto glass coverslips in 6-well plates as described above, and freshmedia was addedafter approximately 24 h, and monitored for PKG knockdownafter various times by immunofluorescent staining for PKG. Toevaluate effects of knockdown on heparin induced signaling, cellsexposed to siRNA for approximately 72 h total exposure wereemployed. PKG knock-down cells were treated with heparin at200mg/ml heparin for 20min, followed by PDGF for 10, 15, or

Fig. 1. cGMPmimics heparin and PKG inhibition blocks heparin effects on VSMC DNA synthesis. A7r5 VSMC (A,C,D) and porcine VSMC (B)were grown to approximately 60% confluence and synchronized by starvation. Cells were incubated with 20mM BrdU for at least 4 h andinhibitors were added (200mg/ml heparin [n¼ 5 in A, 6 in B], 100mM 8-Br-cGMP [n¼ 4 in A, 3 in B], 100 nM ANP [n¼ 3 in A, 3 in B]. After10min, 1.5mg/ml PDGF was added. After 48h incubation, cells were fixed and the percentage of the total cells demonstrating BrdUincorporation was measured histochemically. The values reported reflect the average (�SE) relative BrdU incorporation from separateexperiments compared to non-treated controls (0%) and PDGF treatments (100%). All three treatments in A and B blocked PDGF-stimulatedDNA synthesis (P< 0.0001). For part C, indicated samples were pretreated with 2mMRp-8-pCPT-cGMS for 20min prior to the PDGF addition(n¼ 3). Inhibition of PKG with Rp-8-pCPT-cGMS prevented the effects of anti-proliferative agents (heparin, 8-Br-cGMP and ANP) on PDGF-stimulated DNA synthesis (P< 0.0001, 0.0002, and 0.0028, respectively). For part D, anti-heparin receptor antibodies (18H6 or 18E9, 0.8mg/ml) were added to some cells, as noted. Rp-8-pCPT-cGMS was added, as described for part C. The inhibitor blocked the anti-heparin receptorantibody effects on BrdU incorporation (P< 0.0001). Comparisons with significant differences are noted by *.


2144 G I L O T T I E T A L .

20min. Controls were left untreated, or treated with PDGF alone.Identical cells were electroporated with control siRNA andadditional control cells were not electroporated, but were treatedidentically. PDGF was not included in cells where MKP-1 synthesiswas monitored. Each cell sample was monitored to measure PKGknockdown by staining for PKG, and cells were also stained forpERK, pElk, or MKP-1 as above to determine heparin effects. PKGknockdown was also carried out on cells that were then analyzedfor BrdU incorporation as above. As in the pERK, pElk, and MKP-1experiments, cells were not starved before treatments tomaximize PKG knockdown. To determine statistics withknockdown experiments, cells where PKG was knocked downwere evaluated. Cells in knockdown samples were not counted ifbright PKG staining remained in the treated samples.

ResultsHeparin,ANP, and 8-Br-cGMPall blockPDGF-stimulatedDNA synthesis

To test the hypothesis that heparin and ANP function similarlyin VSMCs, the effects of heparin, ANP, and 8-Br-cGMP onPDGF-stimulated BrdU incorporation were examined as ameasure of DNA replication. Cells were serum-starved for48 h and then grown in the presence of 20mM BrdU for 4 h.Heparin, ANP, and 8-Br-cGMP were added for 10min prior tostimulation of the cells with PDGF. Figures 1A and Bdemonstrates that heparin, ANP, and 8-Br-cGMP added 10minprior to PDGF all block the incorporation of BrdU duringPDGF-stimulated DNA synthesis in both rat and porcineVSMCs. In rat A7r5 cells, all three treatments blocked 90% ormore of the PDGF-stimulated BrdU incorporation (Fig. 1A). Inporcine VSMCs, these treatments blocked approximately 75%of the BrdU incorporation (Fig. 1B).

To further examine anti-proliferative effects in this pathway,the ability of a PKG inhibitor to attenuate heparin's effects wastested. Cells were treated as above, but the indicated cellswere treated with the PKG inhibitor, Rp-8-pCPT-cGMS, for20min prior to treatment with PDGF. Addition of Rp-8-pCPT-cGMS almost completely reversed the effects of heparinand ANP (Fig. 1C). The inhibitor also attenuated the ability of8-Br-cGMP to block PDGF-stimulated BrdU incorporation, butBrdU incorporation remained lower than in the other inhibitortreatments. This may be the result of the direct competitionfor PKG between the PKG inhibitor and 8-Br-cGMP.

The effects of anti-heparin receptor antibodies on VSMCsare PKG-dependent

Heparin as well as anti-receptor antibodies, 18H6 and 18E9,blocked PDGF-stimulated BrdU incorporation. Inhibition ofPKG with Rp-8-pCPT-cGMS reversed the anti-proliferativeeffects of heparin and the anti-receptor antibodies (Fig. 1D).Therefore, as in studies of the effects of these anti-heparinreceptor antibodies on cell proliferation and MAPK activity(Savage et al., 2001; Blaukovitch et al., 2010), the sensitivity ofthese anti-heparin receptor antibody effects to the PKGblocker matches the sensitivity of heparin effects. Interestingly,Rp-8-pCPT-cGMS also increased the level of BrdU incorpo-ration in cells not treated with PDGF (Fig. 1D). The reason forthis remains unclear.

PKA is not involved in heparin effects

To rule out the possibility that cAMP cross-talk wasresponsible for the anti-proliferative effects of heparin, theability of the PKG inhibitor to block cAMP-dependentinhibition of BrdU incorporation was tested. Cells weretreated as in Figure 1; 10min prior to PDGF stimulation, cellswere treated with 8-Br-cGMP, 8-Br-cAMP, or IBMX. IBMX is a

non-specific phosphodiesterase inhibitor, and treatment withthis agent should result in elevations of cAMP and cGMPsimultaneously. Indicated cells were treated with the PKGinhibitor, Rp-8-pCPT-cGMS for 20min prior to treatment withPDGF. Each treatment (8-Br-cGMP, 8-Br-cAMP, and IBMX)significantly reduced PDGF-stimulated BrdU incorporation inthe A7r5 cells (Fig. 2A). 8-Br-cGMP inhibited almost 95% ofthe BrdU incorporated into cells treated with PDGF alone.8-Br-cAMP and IBMX inhibited PDGF-stimulated BrdUincorporation by 70% and 55%, respectively. When the PKGinhibitor was added to each of these treatments, Rp-8-pCPT-cGMS only reversed the inhibition of BrdU incorporation by8-Br-cGMP. Both 8-Br-cAMP and IBMX still inhibitedPDGF-stimulated BrdU incorporation in the presence of thePKG inhibitor in the A7r5 cells. These results demonstrate thatthere are also PKG-independent signaling pathways capable ofblocking PDGF stimulation of VSMCs. To further support thePKG-dependence of heparin's effects on PDGF-induced BrdU

Fig. 2. PKG, and not PKA, mediates heparin inhibition of DNAsynthesis. A7r5 (A, n¼ 3) and porcine (B, n¼ 3) VSMC were treatedas in Figure 1. In these assays, some cells were treated with 100mM8-Br-cAMP, 100mM IBMX, and/or Rp-8-pCPT-cAMS at 2mM.Heparin, 8-Br-cGMP, 8-Br-cAMP, and IBMX blocked PDGF-stimulated BrdU incorporation (P< 0.0001). Inhibition of PKA withRp-8-pCPT-cAMS blocked the effects of 8-Br-cAMP (P< 0.0001).However, only Rp-8-pCPT-cGMS completely blocked the effects ofheparin and 8-Br-cGMP (P< 0.0001). Interestingly, Rp-8-pCPT-cAMS did have a partial effect on 8-Br-cGMP in the porcine cells (-*).Both inhibitors decreased the effects of IBMX (P< 0.0001).



incorporation into DNA, we tested the ability of a PKAinhibitor to block heparin's effects in porcine VSMCs (Fig. 2B).Heparin's inhibitory effects were only blocked by Rp-8-pCPT-cGMS, but not by Rp-8-pCPT-cAMS. Interestingly, the effectsof IBMX were almost totally decreased by the inhibitors ofboth PKG and PKA. In each case, the effect of the 8-Br-form ofthe cyclic nucleotide was only completely blocked by the Rp-8-form of that nucleotide and only the PKG inhibitor blocked theheparin effect. These results strongly support the role of PKG,not PKA, in mediating the anti-proliferative effects of bothcGMP and heparin in VSMCs.

Decreasing PKG levels results in decreased heparineffects on BrdU incorporation

To further confirm the results obtained using kinase inhibitors,A7r5 cells were treated with siRNA for PKG. BrdUincorporation was monitored by immunofluorescencedetection of BrdU to allow simultaneous detection of PKGlevels. Consistent knock down of PKG required maintainingcells in standard culture medium. Therefore, cells were notstarved for these studies. Under these conditions, BrdUincorporation was observed in control and PDGF treated cellsthat had been subjected to knock down of PKG with siRNA ortreated with control siRNA. In control siRNA treated cells,PDGF significantly increased BrdU incorporation. It wasinteresting that cells treated with siRNA for PKG exhibited amuch higher than control level of BrdU incorporation, andPDGF induced limited additional incorporation (see Fig. 3).This result is consistent with the data using inhibitor alone in

Figure 1D that resulted in BrdU incorporation without PDGF.In cells treated with control siRNA, heparin was able tosignificantly decrease BrdU incorporation. However, heparinhad little effect on BrdU incorporation if the cells had reducedPKG levels after treatment with PKG siRNA (Fig. 3).

Deactivation of ERK by heparin is PKG-dependent

Treatment of VSMCs with heparin leads to an accelerateddeactivation of PMA-stimulated ERK activation (Ottlingeret al., 1993; Pukac et al., 1997; Savage et al., 2001) which isdependent upon heparin induction of MKP-1 expression that inturn deactivates ERK (Blaukovitch et al., 2010). To establish therole of PKG in deactivating ERK, porcine VSMCs and A7r5 cellswere treated with the PKG inhibitor Rp-8-pCPT-cGMS for10min before the addition of heparin or 8-Br-cGMP. Again,heparin and 8-Br-cGMP decreased PMA-stimulated ERK2activation (Fig. 4). Inhibition of PKG with Rp-8-pCPT-cGMSpartly blocked the effects of 8-Br-cGMP and heparin on ERKphosphorylation. Similarly, PKG inhibition caused serumactivated A7r5 cells to exhibit little heparin-induced change inERK phosphorylation (data not shown). Variability in PMAinduced activation between experiments, relatively smallheparin-induced decreases, and complexity of the serumsystem led us to develop assays to monitor heparin signalingunder conditions where we could examine responses inindividual cells.

To further examine heparin effects on ERK activity, A7r5VSMC were cultured on cover slips to facilitate examinationsignaling events in individual cells. Because heparin effects on

Fig. 4. Inhibiting PKG negates heparin effects on ERK activity.A7r5 (A, black) and porcine VSMC (A, gray) were synchronized bystarvation and stimulated for 30min with 20mM PMA. Indicatedcells were pretreated for 10min (A) with heparin at 200mg/ml or100mM 8-Br-cGMP. Indicated cells were incubated with 2mM Rp-8-pCPT-cGMS for 10min prior to heparin addition. Treated cells wereanalyzed by Western blotting for active ERK as described in theMaterials and Methods Section. Each mean (�SE) represents theactivation of ERK2 (n¼ 3) compared to untreated controls (0%activation) and stimulated controls (100% activation). Heparin and8-Br-cGMP both decreased PMA stimulated ERK activity(P< 0.005). The effects of the PKG inhibitor in PMA-activated cellsreached significance for the 8-Br-cGMP treated cells (P< 0.01). Rp-8-pCPT-cGMS effects on heparin treated cells (**) in these threerepeats were not quite at the significance level (P< 0.058).

Fig. 3. Knockdown of PKG decreases the ability of heparin to blockDNA synthesis. A7r5 cells were treated with siRNA designed toknock down PKG or non-specific siRNA as described in the Materialsand Methods Section. After 48h, BrdU was added to all cells. Cellswere left untreated or treated with PDGF in the presence orabsence of 200mg/ml heparin. Cells were stained with antibodiesagainst PKG (not shown) or antibodies against BrdU followed byappropriate secondary antibodies. Fluorescent intensity as apercentage of the highest intensity sample from the experiment wascalculated for each cell. Data from control siRNA-treated cells(black bars) are shown from one experiment run in duplicate (morethan 100 cells/treatment). For control siRNA-treated cells (blackbars), PDGF treated cells had significantly increased fluorescentintensity compared to controls (*P< 0.05), and these cells alsotreated with heparin had significantly decreased fluorescentintensity compared to cells treated with PDGF alone (*P< 0.05).Data from two experiments (more than 150 cells/treatment) areshown for PKG siRNA treated cells (gray bars). There were nosignificant differences in fluorescent intensity between treatmentsfor knockdown cells.


2146 G I L O T T I E T A L .

ERK activity depend at least partly on nucleardephosphorylation (Blaukovitch et al., 2010), we examined anuclear transcription factor target of ERK in VSMC, Elk-1(Yordy and Muise-Helmericks, 2000; Shin et al., 2003) forheparin sensitivity. The ability of heparin to cause decreased

ERK activity and Elk phosphorylation was observed byimmunofluorescence (Fig. 5, e.g., compare images D with F andL with N in part A), where we found consistent heparin-induced decreases in ERK and Elk-1 PDGF-induced phos-phorylation. Note that all heparin induced decreases in Elk-1

Fig. 5. PKG inhibitors interfere with heparin effects on ERK activation and Elk phosphorylation. A7r5 cells were cultured on cover slips asdescribed in the Materials and Methods Section, starved and treated with PDGF in the presence or absence of 200mg/ml heparin. Cells werefixed, permeabilized, stained with primary antibodies against pERK (mouse) and pElk (rabbit) followed by FITC-labeled anti-mouse secondaryantibodies and TRITC-labeled anti-rabbit secondary antibodies, and visualized as described in the Materials and Methods Section. Some cellswere treated with Rp-8-cCPT-cGMPS (part A) or KT5823 (part B) 10min prior to heparin treatment. Pictures are representative of at leastthree experiments. Cells were analyzed using image J to determine intensity of pERK and pElk staining. Three experiments with 50–80 cellsper experiment in each treatment were analyzed. For statistical analysis black bars represent untreated cells, medium intensity barsrepresent PDGF activated cells, diagonal striped bars represent heparin treated cells stimulated with PDGF, and the lightest bars representcells treated with inhibitor prior to heparin and PDGF treatments. In all cases except for pERK staining at 20min in part C PDGF stimulationresulted in significant (P< 0.05) increases in staining (*). Heparin-induced decreases in staining were significant (P< 0.05) in all cases (**). PartC data indicate pERK staining with Rp-8-cCPT-cGMPS which significantly reversed heparin effects (P< 0.05) at both 10 and 20min (y). Part Ddata indicate pElk staining with Rp-8-cCPT-cGMPS which significantly reversed heparin effects (P< 0.05) at 20min (y). Part E data indicatepERK staining with KT5823 which significantly reversed heparin effects (P< 0.05) at 20min (y). Part F data indicate pElk staining with KT5823which significantly reversed heparin effects (P< 0.05) at 10min (y).



were significant in parts C through F. We then employed thisfluorescence assay for active ERK and Elk-1 phosphorylation toexamine the requirement of PKG activity in the heparinresponses (Fig. 5). PDGF treatment of starved A7r5 VSMCsinduced significant increases in staining with antibodies to pERKat 10 and 20min (e.g., compare part A pictures A to C and I toK). At both times (10 and 20min PDGF treatment) heparinpre-treatment of cells decreased ERK activation and Elkphosphorylation. In the presence of Rp-8-pCPT-cGMPS (Fig. 5,part A) or KT5823, a second inhibitor of PKG, (Fig. 5, part B),heparin treatment had little effect on PDGF induced ERKactivation and Elk-1 phosphorylation. Figure 5 parts C–F showanalysis of cells from three separate experiments (at least50 cells/experiment) for each condition in parts A and B. Ineach condition, at least one of the time points indicatedsignificant differences between the heparin treated cells withinhibitor and those without. The remaining time points in eachcondition were consistent, but did not reach significance. Toconfirm that PKG inhibition decreased active ERK and Elk-1phosphorylation in primary VSMC, similar experiments werecarried out with porcine VSMC (Fig. 6, parts A and B). In thesecells, PKG inhibitors Rp-8-pCPT-cGMPS and KT5823 (notshown) also decreased heparin effects on ERK activation andElk-1 phosphorylationwith the result that phosphorylation wassimilar to that in cells not treated with heparin.

Because there is a short time frame required to add theinhibitors, heparin, and during which PDGF activation can beobserved in ERK and Elk assays, we confirmed the importanceof PKG activity in heparin effects on ERK phosphorylation incells where PKG levels were significantly decreased throughthe use of siRNA. To ensure that siRNA was effectively gettinginto the cells, FITC-labeled control siRNA was used in oneexperiment. Essentially all cells took up the labeled controlsiRNA (data not shown). PKG knockdown efficiency variedbetween experiments from occasionally greater than 90%

knockdown to about 50% decrease in staining in A7r5 cellstreated with siRNA compared to a scrambled control asdescribed in the Materials and Methods Section. Cells allowedto recover in starvation media typically exhibited insignificantknockdown of PKG. Therefore, cells were treatedwith controlsiRNA or PKG siRNA and were cultured in regular growthmedia for 48 or 72 h and then treated with heparin for 20minfollowed by PDGF for 15min. Controls with PDGF or notreatment were compared. The presence of PKG, pERK, andpElk-1 was determined by immunofluorescence staining(Fig. 7A,B). Despite incomplete knockdown, it is clear thatheparin induced decreases in pERK and pElk-1 staining werevery limited in knockdown cells compared to cells with controlsiRNA. Western Blots confirm that heparin treatment doesnot have much effect on ERK activation in cells where PKGlevels have been significantly decreased (data not shown). Themaximum differences between untreated, PDGF-treated,and heparin/PDGF-treated cells are smaller because there is nostarvation to decrease the basal ERK activity. These resultsare similar to those with BrdU incorporation in that wholecells treated with PKG siRNA retain more active ERK thancells treated with control siRNA. Repeats from immunofluo-rescence stained knockdown experiments were analyzed fornuclear staining and relative intensity comparisons areillustrated in (Figs. 7C and D). Phosphorylation of ERK and Elkwas significantly increased by PDGF in both cell types, butheparin significantly decreased nuclear ERK and Elk phos-phorylation only in cells treated with control siRNA.

Heparin-stimulated increases in MKP-1 expression arePKG-dependent

Because heparin-induced expression of MKP-1 in VSMCs wasfound to be important for heparin-induced decreases in ERKactivity (Blaukovitch et al., 2010) and treatments that increase

Fig. 6. Inhibitors of PKG interfere with heparin responses in primary VSMC. Part A: Porcine VSMC were treated identically to A7r5 cellswith PDGF stimulation for 10min. Heparin and inhibitor treatments and staining were identical to that in Figure 5. Part B: Porcine VSMCwere treated identically and active ERK analyzed by Western blotting. Lanes 1—untreated control, 2—PDGF 10min, 3—heparin only, 4—heparin, and PDGF, 5—heparin and Rp8-cCPT-cGMPS, 6—heparin, Rp8-cCPT-cGMPS and PDGF, 7—Rp8-cCPT-cGMPS and PDGF.


2148 G I L O T T I E T A L .

cGMP levels result in MKP-1 increases (Furst et al., 2005), therole of cGMP/PKG in the process of heparin induced MKP-1expression was evaluated. Figure 8A demonstrates that theinduction of MKP-1 expression by heparin was dependent onPKG; serum-starved rat A7r5 cells were treated with heparinwith or without the PKG inhibitor Rp-8-pCPT-cGMS (added10min before the heparin). Inhibiting PKG significantlydecreases the heparin-induced MKP-1 expression. This resultwas confirmed using PKG siRNA induced decreases in PKGlevels. To maximize PKG decreases, cells were not starvedafter knockdown, and heparin-induced MKP-1 expression wascompared between cells where PKG was knocked down andcontrol cells where non-specific siRNA was employed(Fig. 8C).

Heparin increases cGMP concentration in VSMCs

The ability of heparin to increase cGMP levels was tested inVSMCs using cGMP ELISA assays. In order to obtainmeasurable levels of cGMP, it was necessary to use 150mmculture dishes of cells for each data point. Figure 9A illustratesthe time course of heparin-stimulated cGMP increases instarved A7r5 cells. cGMP concentrations in the cell lysateswere elevated approximately four times above basal levels after5, 10, and 20min of stimulation with 200mg/ml heparin andreturned to near basal levels after 40min. Logarithmicallygrowing A7r5 VSMCs (Fig. 9B) exhibited higher levels of cGMPin response to heparin, and a 10min treatment with eitherheparin or an anti-receptor antibody (18H6) increased the

Fig. 7. Knockdown of PKG decreases heparin effects on ERK activity and pElk. A7r5 cells were electroporated with siRNA designed to knockdown PKG in rat cells or scrambled RNA, and the cells were allowed to proliferate in growth media for 72h with feeding at 24h. At 72h, cellswere untreated, treated with PDGF for 15min or heparin for 20min before PDGF was added for 15min. Part A illustrates the pElk levels(pictures A–F). Staining for PKG is illustrated in the same cells in pictures G–L. Scrambled siRNA (A–C,G–I) as compared to PKG siRNA (D–F,J–L) is shown for cells not stimulated (A,D,G,J), PDGF treated cells (B,E,H,K) and heparin plus PDGF (C,F,I,L). Part B illustrates anexperiment where pERK was monitored (A0–F0) PKG staining for these cell samples is shown in pictures G0–L0. The treatment pattern isidentical to that for part A. These experiments are representative of two similar experiments each. Part C provides quantitative analysis ofElk activation in PKG and scrambled siRNA. More than 100 cells from two separate experiments were analyzed and fluorescent intensity ofpElk staining in the nucleus was determined. In both control (black bars) and PKG (gray) siRNA, pElk staining was significantly increased(*P< 0.05). Heparin significantly (*P< 0.05) decreased pElk staining in control (black), but actually further increased pElk in PKG siRNA-treated cells. Part D illustrates ERK activation as described for Elk in part C. Nuclear pERK was determined for more than 100 cells from twoexperiments. As with Elk phosphorylation, PDGF increased ERK phosphorylation significantly (*P< 0.05) for both control siRNA (black bars)and PKG siRNA (gray bars) treated cells. In addition, heparin decreased PDGF-induced activation of ERK significantly (P< 0.05,*) in controlsiRNA-treated cells, but not in PKG siRNA-treated cells.



cGMP concentration. In this experiment, the increase in cGMPin response to heparin was approximately a 15-fold induction(compared to untreated control levels). 18H6 induced a moremodest increase in cGMP levels and a 10min treatment withANP did not show a statistically significant increase in cGMP atthe 10min time point used in this experiment (Fig. 9B). Thistime point was used to maximize the cGMP production inresponse to heparin, andmay not represent the optimal time atwhich to measure ANP induced cGMP increases. Starvedporcine VSMC did not yield measurable increases in cGMP.However, proliferating porcine VSMC treated with heparin orANP exhibited modestly elevated cGMP (not shown).

Discussion

The proliferation of VSMCs in healthy vessels is carefullyregulated, and heparin's ability to block cell signaling events inVSMCs has been well documented (e.g., Ottlinger et al., 1993;Pukac et al., 1997). Antibodies to a heparin receptor proteinwere shown to mimic heparin's effects in these cells, therebyassociating heparin’s anti-proliferative effects with thisreceptor protein (Savage et al., 2001). An intracellular signaltransduction cascade starting at the receptor presumably isinvolved in heparin-induced decreases in PDGF-stimulatedBrdU incorporation and deactivation of ERK MAPK activity(Savage et al., 2001), expression of MKP-1 (Blaukovitch et al.,2010), and decreased Elk-1 phosphorylation (this study) inVSMCs. In fact, the changes in Elk-1 phosphorylation areconsistent with reports that Elk-1 phosphorylation results inchanges in gene expression. Obvious mechanisms for heparin-induced PKG-mediated modulation of VSMC proliferationinvolve phosphorylation of transcription factors (reviewed inPilz and Broderick, 2005) and reported for Elk-1 (Choi et al.,2010). The results reported here provide evidence for the

downstream events of heparin receptor activation involvingactivation of PKG. Heparin has previously been demonstratedto have an anti-proliferative effect on VSMCs in rats and in anendothelial/VSMC co-culture system, but while NOproduction in the heparin-treated rats played a role in theresponse, it did not appear to be involved in heparin-induceddecreases in VSMC proliferation (Horstman et al., 2002). Toour knowledge, the present study is the first to report that theeffects of heparin on VSMC DNA synthesis in culture, MKP-1synthesis, Elk-1 phosphorylation changes, and ERK activity aresensitive to PKG inhibitors.

NO is a cell permeable activator of soluble guanylylcyclase. Treatment of cells with synthetic NO donors oractivators of NO synthase causes a rapid increase inintracellular concentrations of cGMP, while particulateguanylyl cyclases often result in limited cGMP increases (Suet al., 2005). In addition, production of cGMP at themembrane does not produce cGMP increases throughout thecell (Nausch et al., 2008), consistent with the idea that limitedlocalized signaling might lead to physiological results withoutsignificant increases in total cGMP levels. ANP treatment ofVSMCs does limit their proliferation (Baldini et al., 2002), andthe limited increase in cGMP induced by heparin (Fig. 9)seems to be an important mechanism by which heparininteraction with the receptor induces changes in VSMCsignaling and proliferation.

There are PKG-independent effects of cGMP in VSMCs, butPKG is involved in the maintenance of the contractilephenotype characteristic of healthy VSMCs in vitro (Lincolnet al., 2006). Cells of this contractile phenotype are not highlyproliferative. Therefore, many agents that can prevent aproliferative response to mitogenic signals might require PKGto be effective. In fact, a recent report directly links cGMPincreasing agents to decreases in cell proliferation that result in

Fig. 8. Heparin-induced increases in MKP-1 require PKG activity. A: A7r5 cells were synchronized and left untreated or treated with 200mg/ml heparin (dark bars). Identically treated cells were pre-treated with 2mM Rp-8-pCPT-cGMS for 20min prior to heparin (light bars) and cellswere analyzed as described in the Materials and Methods Section (n¼ 7). MKP-1 expression was significantly different (*) between heparin andinhibited cells at 20, 30, and 40min (P< 0.05 in each case). B: OneWestern blot experiment included in the data is illustrated. The blot was cutand aligned to show the same heparin points without inhibitor (above) and with inhibitor (below). No inhibitor without heparin was includedin this blot. C: A7r5 cells were treated with siRNA to knock down PKG and treated with heparin as in A (light bars). For comparison, identicalcells were treated with scrambled siRNA and then with heparin (dark bars). Cells were stained for PKG and MKP-1. Cells were analyzed usingimage J to determine intensity of MKP-1 staining. In knock down samples, cells were not analyzed if they stained strongly for PKG. Twoexperiments with more than 100 cells each were analyzed. In scrambled siRNA treated cells there were significant (*) increases in MKP-1expression at 30 and 40min (P< 0.005). There were no increases in cells where PKG siRNA was used.


2150 G I L O T T I E T A L .

therapeutic responses (Schwappacher et al., 2013). In addition,PKG levels typically decrease as VSMCs are sub-cultured(Cornwell and Lincoln, 1989; Lin et al., 2004) a result thatcoincides with reports of a decrease in heparin sensitivity withsubculture (e.g., Ottlinger et al., 1993; Pukac et al., 1997). In thisregard, it is interesting that our PKG knockdown cells appearto maximally incorporate BrdU into DNA without furtherPDGF activation. Our data indicating the involvement of PKGin heparin-induced changes in signaling and proliferation areconsistent with these results. Interestingly, it has been shownthat activation of PKG results in its down-regulation, indicatinga mechanism by which the response to cGMP-inducing signalscan also be limited (Dey et al., 2009) and both NO and PDGFinduced PKG decreases appear to require protein tyrosinephosphatase 1B (Zhuang et al., 2011). Although studies withgenetically modified mice do not confirm decreased VSMCproliferation by PKG signaling pathways (Feil et al., 2003), areview of this system indicates complicated changes in generegulation in these mice, making conclusions regarding cGMP/PKG signaling drawn from studies with these mice less clear(Hofmann et al., 2006).

ANP, 8-Br-cGMP, heparin, and anti-heparin receptorantibodies reduced PDGF-stimulated BrdU incorporation intoVSMCs. Treatment of the cells with PKG inhibitor released theinhibitory effects of these agents. 8-Br-cAMP and IBMX alsoattenuated BrdU incorporation under these conditions, buttheir effects were insensitive to the PKG inhibitor, Rp-8-pCPT-

cGMS. cAMP has anti-proliferative effects on VSMCs (e.g., Wuet al., 1993); however, our results indicate that while bothcGMP- and cAMP-dependent mechanisms induce anti-proliferative effects on VSMCs, the heparin-induced responseis not operating through a cAMP-dependent mechanism.Heparin treatment caused an elevation in cGMPconcentrations further supporting a role for PKG in heparin'santi-proliferative effects in VSMCs.

One possible mechanism whereby heparin could affectVSMCs while inducing modest elevations of cGMP levels isthrough localization in caveolae. Peterson et al. (1999)reported that heparin affects the expression of caveolin andthat caveolin expression influences VSMC growth, but whetherthese changesmeasured at 24 h explain ERK activity differencesobserved in less than an hour is unclear. Such a mechanismcould also explain other heparin effects identified as involvingcaveolae (Liu et al., 2007). If the heparin receptor is localized incaveolae, binding could result in localized signaling throughPKG by activating a NOS protein in VSMC (Cheah et al., 2002)or caveolae might modulate PKG activity adjacent to thecaveolae.

Different PKG locations result in different cellular responses(Su et al., 2005; Nausch et al., 2008; Schlossmann andDesch, 2011). For example, PKG modulates calcium levelsthrough several mechanisms including: decreasing the fre-quency of agonist-induced Ca2þ oscillations (Perez-Zoghbiet al., 2010), decreasing channel activity of TRPC1/TRPC3 andTRPC6 channels (Takahashi et al., 2008; Chen et al., 2009), andmodulation of Ca2þ-dependent big potassium channels (Fellnerand Arendshorst, 2010).

Whatever the mechanism and down-stream signaling fromPKG, our data suggest that heparin treatment causes theelevation of cGMP levels and indicate a role for PKG inheparin-induced decreased VSMC proliferation. Theexistence of a heparin receptor on the cell surface of VSMCsand vascular endothelial cells has been reported by ourlaboratory (Patton et al., 1995; Savage et al., 2001; Blaukovitchet al., 2010). Here, we provide evidence for a mechanismwhereby heparin interacts with VSMCs and either directly orindirectly activates a guanylyl cyclase that causes a rapidelevation of cGMP which, in turn, activates PKG. This idea issupported by (1) the rapid elevation of intracellular cGMP inheparin-treated cells, (2) the sensitivity of the heparin effectsto PKG inhibitors, and (3) the sensitivity to partial knockdown of PKG protein. Identification of the heparin receptorrecognized by the anti-receptor antibodies produced by thislab will provide important information regarding the link(s)between the heparin receptor and cGMP/PKG.

Acknowledgments

The authors acknowledge preliminary PKG knock-downstudies by Kristen Cornell and the technical help of Sara LynnN. Farwell, Matthew M. Miller, and Traci Rickert.

Literature Cited

Baldini P, De Vito P, Fraziano M, Mattioli P, Luly P, Di Nardo P. 2002. Atrial natriuretic factorinhibits mitogen-induced growth in aortic smooth muscle cells. J Cell Physiol 193:103–109.

BegumN, Song Y, Rienzie J, Ragolia L. 1998. Vascular smooth muscle cell growth and insulinregulation of mitogen-activated protein kinase in hypertension. Am J Physiol 275:C42–C49.

Blaukovitch CI, Pugh R, Gilotti AC, Kanyi D, Lowe-Krentz LJ. 2010. Heparin treatment ofvascular smooth muscle cells results in the synthesis of the dual-specificity phosphataseMKP-1. J Cell Biochem 110:382–391.

Brunet S, Roux D, Lenomand P, Dowd S, Keyse S, Pouyss�egur J. 1999. Nuclear translocationof p42/p44 mitogen-activated protein kinase is required for growth factor-induced geneexpression and cell cycle entry. EMBO J 18:664–674.

Castellot JJ, Jr., Addonizio ML, Rosenberg R, Karnovsky MJ. 1981. Cultured endothelial cellsproduce a heparinlike inhibitor of smooth muscle cell growth. J Cell Biol 90:372–379.

Castellot JJJ,Wong K, Herman B, Hoover RL, Albertini DF,Wright TC, Caleb BL, KarnovskyMJ. 1985. Binding and internalization of heparin by vascular smooth muscle cells. J CellPhysiol 124:13–20.

Fig. 9. Heparin treatment induces increases in cGMPconcentration. Part A: A7r5 cells were cultured in 150mm dishes,starved as described in the Materials and Methods Section andtreated with 200mg/ml heparin for indicated times. The cGMPconcentrations were determined as described in the assay protocoland are reported as fold increases. The means (�SE) representthree independent experiments. Heparin increased cGMP after 5,10, and 20min (P< 0.0051, 0.0011, and 0.0222, respectively). Part B:A7r5 cells were cultured in 150mm dishes and treated with heparin(200mg/ml), 18H6 antibody (0.8mg/ml) or ANP (100 nM) for 10min.The data represent triplicate repeats. Values are reported as foldincreases. Heparin treatment resulted in a statistically significantincrease in cGMP (P< 0.0032).



Castellot JJJ, Pukac LA, Caleb BL, Wright TC, Karnovsky MJ. 1989. Heparin selectivelyinhibits a protein kinase C-dependent mechanism of cell cycle progression in calf aorticsmooth muscle cells. J Cell Biol 109:3147–3155.

Cheah LS, Gwee M, Das R, Ballard H, Yang YF, Daniel EE, Kwan CY. 2002. Evidence for theexistence of a constitutive nitric oxide synthase in vascular smooth muscle. Clin ExpPharmacol Physiol 29:725–727.

Chen J, Crossland RF, Noorani MM, Marrelli SP. 2009. Inhibition of TRPC1/TRPC3 by PKGcontributes to NO-mediated vasorelaxation. Am J Physiol Heart Circ Physiol 297:H417–H424.

Choi C, Sellak H, Brown FM, Lincoln TM. 2010. CGMP-dependent protein kinase and theregulation of vascular smooth muscle cell gene expression: possible involvement of Elk-1sumoylation. Am J Physiol Heart Circ Physiol 299:H1660–H1670.

Clowes AW, Karnovsky MJ. 1977. Suppression by heparin of smooth muscle cellproliferation in injured arteries. Nature 265:625–626.

Cornwell T, Lincoln T. 1989. Regulation of intracellular Ca2þ levels by atriopeptin and8-bromo-cyclic GMP is mediated by cGMP-dependent protein kinase. J Biol Chem264:1146–1155.

Dey N, Busch J, Frnacis S, Corbin J, Lincoln T. 2009. Cyclic GMP specifically suppresses type-1a cGMP-dependent protein kinase expression by ubiquitination. Cell Signal 21:859–866.

Fasciano S, Patel R, Handy I, Patel C. 2005. Regulation of vascular smooth muscleproliferation by heparin. Inhibition of cyclin-dependent kinase 2 activity by p27kip1. J BiolChem 280:15682–15689.

Feil R, Lohmann S, de Jonge H, Walter U, Hofmann F. 2003. Cyclic GMP-dependent proteinkinases and the cardiovascular system. Insights from genetically modified mice. Circ Res93:907–916.

Fellner SK, Arendshorst WJ. 2010. Complex interactions of NO/cGMP/PKG systems onCa2þ signaling in afferent arteriolar vascular smooth muscle. Am J Physiol Heart CircPhysiol 298:H144–H151.

Furst R, Brueckl C, Kuebler W, Zahler S, Krotz F, Gorlach A, Vollmar A, Kiemer A. 2005.Atrial natriuretic peptide induces mitogen-activated protein kinase phosphatase-1 inhuman endothelial cells via Rac1 and NAD(P)H oxidase/Nox2-activation. Circ Res 96:43–53.

Hofmann F, Feil R, Kleppisch T, Schlossmann J. 2006. Function of cGMP-dependent proteinkinases as revealed by gene deletion. Physiol Rev 86:1–23.

Horstman D, Fischer L, Kouretas P, Hannan R, Rich G. 2002. Role of nitric oxide in heparin-induced atenuation of hypoxic pulmonary vascular remodeling. J Appl Physiol 92:2012–2018.

Jacob A, Molkentin JD, Smolenski A, Lohmann SM, Begum N. 2002. Insulin inhibits PDGF-directed VSMCmigration via NO/cGMP increase of MKP-1 and its inactivation of MAPKs.Am J Physiol Cell Physiol 283:C704–C713.

Lin G, Chow S, Lin J, Wang G, Lue T, Lin C-S. 2004. Effect of cell passage and density onprotein kinase G expression and activation in vascular smoothmuscle cells. J Cell Biochem92:104–112.

Lincoln T,Wu X, Sellak H, Dey N, Choi CS. 2006. Regulation of vascular smooth muscle cellphenotype by cyclic GMP and cyclic GMP-dependent protein kinase. Front Biosci 11:356–367.

Liu YT, Song L, Templeton DM. 2007. Heparin suppresses lipid raft-mediated signaling andligand-independent EGF receptor activation. J Cell Physiol 211:205–212.

Melichar VO, Behr-Roussel D, Zabel U, Uttenthal LO, Rodrigo J, Rupin A, Verbeuren TJ,Kumar HSA, Schmidt HH. 2004. Reduced cGMP signaling associated with neointimalproliferation and vascular dysfunction in late-stage atherosclerosis. Proc Natl Acad SciUSA 101:16671–16676.

Nausch LW, Ledoux J, Bonev AD, Nelson MT, DostmannWR. 2008. Differential patterningof cGMP in vascular smooth muscle cells revealed by single GFP-linked biosensors. ProcNatl Acad Sci USA 105:365–370.

Ottlinger ME, Pukac LA, Karnovsky MJ. 1993. Heparin inhibits mitogen-activated proteinkinase activation in intact rat vascular smoothmuscle cells. J Biol Chem 268:19173–19176.

Patton WAI, Granzow CA, Getts LA, Thomas SC, Zotter LM, Gunzel KA, Lowe-Krentz LJ.1995. Identification of a heparin-binding protein using monoclonal antibodies that blockheparin binding to porcine aortic endothelial cells. Biochem J 311:461–469.

Perez-Zoghbi J, Bai Y, Sanderson M. 2010. Nitric oxide induces airway smooth muscle cellrelaxation by decreasing the frequency of agonist-induced Ca2þ oscillations. J Gen Physiol135:247–259.

Peterson T, Kleppe L, Caplice N, Pan S, Mueske C, Simari R. 1999. The regulation of caveolinexpression and localization by serum and heparin in vascular smooth muscle cells.Biochem Biophys Res Commun 265:722–727.

Pilz R, Broderick K. 2005. Role of cyclic GMP in gene regulation. Front Biosci 10:1239–1268.

Pukac LA, Carter JE, Ottlinger ME, Karnovsky MJ. 1997. Mechanisms of inhibition by heparinof PDGF stimulated MAP kinase activation in vascular smooth muscle cells. J Cell Physiol172:69–78.

Reilly C, Kindy M, Brown K, Rosenberg R, Sonenshein G. 1989. Heparin prevents vascularsmooth muscle cell progression through the G1 phase of the cell cycle. J Biol Chem264:6990–6995.

Rodríguez A, G�omez-Ambrosi J, Catal�an V, Fortuno A, Frühbeck G. 2010. Leptin inhibits theproliferation of vascular smooth muscle cells induced by angiotensin II through nitricoxide-dependent mechanisms. Mediators Inflamm. DOI: 10.1155/2010/105489.

Rohan PJ, Davis P, Moskaluk CA, Kearns M, Krutzsch H, Siebenlist U, Kelly K. 1993. PAC-1:A mitogen-induced nuclear protein-tyrosine phosphatase. Science 259:1763–1769.

Sato J, Nair K, Hiddinga J, Eberhardt N, Fitzpatrick L, Katusic Z, O'Brien T. 2000. ENOS genetransfer to vascular smooth muscle cells inhibits cell proliferation via upregulation of p27and p21 and not apoptosis. Cardiovasc Res 47:697–706.

Savage JM, Gilotti AC, Granzow CA, Molina F, Lowe-Krentz LJ. 2001. Antibodies against aheparin receptor slow cell proliferation and decreaseMAPK activation in vascular smoothmuscle cells. J Cell Physiol 187:283–293.

Schlossmann J, Desch M. 2011. IRAG and novel PKG targeting in the cardiovascular system.Am J Physiol Heart Circ Physiol 301:H672–H682.

Schwappacher R, Kilic A, Kojonazarov B, LangM, Diep T, Zhuang S, Gawlowski T, SchermulyRT, Pfeifer A, Boss GR, Pilz RB. 2013. A molecular mechanism for therapeutic effects ofcGMP-elevating agents in pulmonary arterial hypertension. J Biol Chem 288:16557–16566.

Shin HS, Lee HJ, Nishida M, Lee MS, Tamura R, Yamashita S, Matsuzawa Y, Lee IK, Koh GY.2003. Betacellulin and amphiregulin induce upregulation of cyclin D1 and DNA synthesisactivity through differential signaling pathways in vascular smooth muscle cells. Circ Res93:302–310.

Su J, Scholz P, Weiss H. 2005. Differential effects of cGMP produced by soluble andparticulate guanylyl cyclase on mouse ventricular myocytes. Exp Biol Med 230:242–250.

Sugimoto T, Haneda M, Togawa M, Isono M, Shikano T, Araki S, Nakagawa T, Kashiwagi A,Guan KL, Kikkawa R. 1996. Atrial natriuretic peptide induces the expression of MKP-1, amitogen-activated protein kinase phosphatase, in glomerular mesangial cells. J Biol Chem271:544–547.

Takahashi S, Lin H, Geshi N, Mori Y, Kawarabayashi Y, Takami N, Mori M, Honda A, Inoue R.2008. Nitric oxide-cGMP-protein kinase G pathway negatively regulates vasculartransient receptor potential channel TRPC6. J Physiol 586:4209–4223.

Tantini B, Manes A, Fiumana E, Pignatti C, Guarnieri C, Zannoli R, Branzi A, Galie N. 2005.Antiproliferative effect of sildenafil on human pulmonary artery smoothmuscle cells. BasicRes Cardiol 100:131–138.

Tete H, Schroeder R, Stahl R, Wolf G. 2001. Atrial natriuretic peptide attenuates ANG II-induced hypertrophy of renal tubular cells. Am J Physiol Renal Physiol 281:F81–F90.

Wang S, Li Y. 2009. Expression of constitutively active cGMP-dependent protein kinaseinhibits glucose-induced vascular smooth muscle cell proliferation. Am J Physiol HeartCirc Physiol 297:H2075–H2083.

Wu JJ, Dent P, Jelinek T, Wolfman A, Weber M, Sturgill T. 1993. Inhibition of the EGF-activated MAP kinase signaling pathway by adenosine 30 ,50-monophosphate. Science262:1065–1069.

Yordy JS, Muise-Helmericks RC. 2000. Signal transduction and the Ets family of transcriptionfactors. Oncogene 19:6503–6513.

Zhuang D, Balani P, Pu Q, Thakran S, Hassid A. 2011. Suppression of PKG by PDGF or nitricoxide in differentiated aortic smooth muscle cells: Obligatory role of protein tyrosinephosphatase 1B. Am J Physiol Heart Circ Physiol 300:H57–H63.


2152 G I L O T T I E T A L .

heparin responses in vascular smooth muscle cells involve cgmp-dependent protein kinase (pkg)

Documents