cellular biology - circulation...

Cellular Biology

FSP-1 Silencing in Bone Marrow Cells SuppressesNeointima Formation in Vein GraftJizhong Cheng, Yun Wang, Anlin Liang, Lixin Jia, Jie Du

Rationale: Fibroblast-specific protein 1 (FSP-1) plays multiple roles in promoting cell proliferation and motility.Increased FSP-1 expression in smooth muscle cells (SMCs) has been associated with their enhanced proliferation.

Objective: To study how FSP-1 contributes to neointima formation of vein grafts.Methods: Arteriovenous grafts were created in wild-type or FSP-1–GFP mice (green fluorescent protein

expression regulated by FSP-1 promoter). The effects of FSP-1 on bone marrow (BM) cell migration and on SMCproliferation were studied in vivo and in vitro.

Results: On creation of a vein graft, there was rapid deposition of platelets on the denuded surface leading tosecretion of the chemokine stromal cell–derived factor-1� (SDF-1�). This was followed by recruitment ofBM-derived cells expressing the SDF-1� receptor CXCR4; homing of FSP-1–positive cells was found to bedependent on platelet-derived SDF-1�. FSP-1 was expressed in 8% of the BM cells, and 20% of these expressCD45; 85% of FSP-1–positive cells express CD11b. We found that the FSP-1–positive cells migrated into the veingraft in a Rac-1–dependent fashion. FSP-1 expression was also found to stimulate proliferation of SMCs througha MEK5-ERK5 signaling pathway that can be suppressed by a dominant-negative Rac1. Consequently, knockingdown FSP-1 expression in BM cells prevented neointimal formation.

Conclusions: BM-derived FSP-1� cells enhance neointima formation through an increase in transendothelialinvasion with stimulation of SMC proliferation. The Rac1 and ERK5 signaling cascade mediate FSP-1–inducedresponses in SMCs and BM cells. This novel pathophysiology suggests a new therapeutic target, FSP-1, forpreventing the development of neointima in vein grafts. (Circ Res. 2012;110:230-240.)

Key Words: fibroblast-specific protein 1 � neointima formation � platelets � bone marrow � smooth muscle cells

Arteriovenous (AV) fistula or vein grafts used in bypasssurgery are commonly created to relieve angina or

function as an access for hemodialysis. Failure of these graftsis characterized by hyperplasia of venous smooth musclecells (VSMCs), neointima formation, and atherosclerosis.1–4

Several processes including endothelial dysfunction, neointi-ma cell hyperplasia, and atherosclerosis are present in failedvein grafts.5 The pathogenesis of this lesion is incompletelyunderstood. Identifying the pathogenesis of intimal hyperpla-sia in vein grafts could lead to strategies for facilitating theregression of lesions or reducing the likelihood of developingnew lesions.

Neointima cells are derived from local VSMCs, adventitialfibroblasts, stem cells, or bone marrow (BM)-derived progen-itors.6,7 In AV grafts, BM-derived, c-Kit–positive cells canassume the phenotype of either VSMC or endothelial cells, asdetermined by the expression of calponin or endothelial nitricoxide synthase, respectively.7 BM-derived cells can also

assume other roles including conversion into inflammatory orprogenitor cells that participate in the formation of neointimain AV grafts. A signal affecting the recruitment of BM cellsduring the development of the neointima could be SDF-1�

and its receptor, CXCR4.Fibroblast-specific protein (FSP-1, S100A4) is a member

of the S100 family of calcium-binding proteins. It wasisolated from mouse mammary adenocarcinoma cells, and itsmRNA is detectable in cells of the BM, spleen, thymus, andlymphocytes.8 FSP-1 functions to regulate cellular motilitythrough a direct interaction with myosin-IIA9; FSP-1 also isassociated with cancer cell metastasis,10 including invasivegrowth of mouse endothelial cells11 or neurons.12 How FSP-1might affect the formation of neointima in vein grafts is notdefined.

We show that BM-derived FSP-1� cells are major compo-nents of remodeled AV grafts. BM-derived, FSP-1� cells areattracted to the graft through an SDF-1�–dependent mecha-

Original received April 2, 2011; revision received November 10, 2011; accepted November 16, 2011. In October 2011, the average time fromsubmission to first decision for all original research papers submitted to Circulation Research was 15 days.

From the Nephrology Division, Baylor College of Medicine, Houston, TX (J.C., Y.W., A.L.); and Beijing Anzhen Hospital Affiliated to the CapitalMedical University, The Key Laboratory of Remodeling-Related Cardiovascular Diseases, Capital Medical University, Ministry of Education, Beijing,China (L.J., J.D.).

Correspondence to Jie Du, MD, Beijing Anzhen Hospital Affiliated to the Capital Medical University (E-mail [email protected]); or Jizhong Cheng, MD,Department of Medicine, Baylor College of Medicine, Houston, TX 77030 (E-mail [email protected]).

© 2012 American Heart Association, Inc.

Circulation Research is available at http://circres.ahajournals.org DOI: 10.1161/CIRCRESAHA.111.246025

230

by guest on June 30, 2018http://circres.ahajournals.org/

Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from

http://circres.ahajournals.org/







nism and induce transendothelial invasion of inflammatorycells into the neointima. We also find that secreted FSP-1influences the proliferation of VSMCs in AV grafts. Theseresults provide new insights into the mechanisms by whichFSP-1 BM-derived cells influence the formation ofneointima.

MethodsAn expanded Methods section is available in the Data Supplement athttp://circres.ahajournals.org.

Mice and Vein Graft ProcedureAll animal protocols were approved by IACUC. Wild-type (WT)C57/B6 mice aged 3 months were purchased from Jackson Labora-tory (Bar Harbor, ME). FSP-1–GFP mice expressing the greenfluorescent protein (GFP) under the control of the FSP-1 promoterwere a generous gift from E.G. Neilson, Vanderbilt UniversitySchool of Medicine. The vein graft procedure was performed as wepreviously described.13 See detailed information in the OnlineMethods section.

Reagents and VirusPenicillin, streptomycin, DMEM, and FBS were obtained fromInvitrogen Life Technologies (Carlsbad, CA). The protein assay kitwas purchased from Bio-Rad (Hercules, CA). The FSP-1 antibodywas obtained from DAKO (Carpinteria, CA); PCNA, F4/80,pMEK5, and GFP antibodies were from Abcam (Cambridge, MA);the anti-MEK5 antibody was from Millipore, Inc (Billerica, MA); theSMA-� antibody was from Sigma, Inc (Sigma-Aldrich); the anti-ERK5 and Rac-1 antibodies were from Cell Signaling (Danvers,MA); pERK5 antibody was from Affinity Bioreagent (Golden, CO);and fluorescence-700 or -800 secondary antibodies were obtainedfrom Invitrogen (Carlsbad, CA). Anti-CXCR4 and anti–SDF-1�antibodies were purchased from eBioscience (San Diego, CA).Recombinant SDF-1� was obtained from R & D (Minneapolis, MN).FSP-1 shRNA lentiviruses (total 5 clones) in TRC2-pLKO-purovectors were purchased from Sigma Aldrich (Louis, MO). The PAK1PBD agarose beads (to detect Rac1 activation) were from CellBiolabs, Inc (San Diego, CA). The FSP-1 expression vector waskindly provided by Dr Anne R. Bresnick (Albert Einstein College ofMedicine, Bronx, NY), and the FSP-1 recombinant protein waspurified as described14 with endotoxin removed using the EndotoxinFree kit from Qiagene (Valencia, CA). The FSP-1 expressionrecombinant adenovirus was kindly provided by Dr T.C. He (Uni-versity of Chicago); and the ERK5 siRNA was purchased fromDharmacon RNAi Technologies, (Lafayette, CO). Dominant Rac1expression plasmid (Rac N17) was obtained from Addgene (Cam-bridge, MA).15 The MEK5 dominant-negative vector was used asdescribed.16

ImmunohistochemistryFor histological analysis, vein grafts were perfused through the leftventricle as described.11 Grafts were obtained by cutting the trans-planted segments from the native vessels at the cuff end. Sampleswere fixed with 4% phosphate-buffered formaldehyde at 4°C for24hours and processed as described.13 See detailed information in theOnline Methods section.

Real-Time RT-PCRTotal RNA from freshly removed vena cavas from mice or from veingrafts was isolated using the RNeasy kit (Qiagen, Valencia, CA).Real-time RT-PCR was performed using the Opticon real-timeRT-PCR machine (MJ Research, Waltham, MA). The specificityof real-time RT-PCR was confirmed through agarose gel electro-phoresis and melting-curve analysis. GAPDH was used as aninternal standard. Primers used were mouse FSP-1: forward5�-TTCCAGAAGGTGATGAG-3 � ; reverse 5 � -TCATGGCAATGCAGGACAGGAAGA-3�; and GAPDH forward 5�-

AGTGGGAGTTGC TGTT GAAATC-3�, reverse 5�-TGCTG-AGTATGTCGTGGAGTCTA-3�.

Cell Culture and TransfectionVSMCs were isolated from the inferior vena cavas of WT mice andcultured as monolayers in DMEM supplemented with 10% heat-in-activated fetal bovine serum (Hyclone, Logan, UT), 100 �g/mLpenicillin, and100 �g/mL streptomycin. The identity of VSMCs wasdetermined by immunostaining with SMA-� actin antibody. TheVSMC transfection (for knockdown ERK5) was performed by usingprogram U23 with Nucleofector (Lonza, Basel, Switzerland). Cellproliferation assay was performed with XTT kit (Sigma-Aldrich),following the manufacture’s instructions.

Western Blot Analysis and ImmunoprecipitationCells were lysed in RIPA buffer, and �30 �g of proteins wereseparated by SDS-PAGE. After transferring to PVDF membranes,antibodies were added. To deplete FSP-1 from conditioned media, itwas treated with anti–FSP-1 antibodies (2 �g/mL) for 2 hours beforeprotein A/G agarose beads (Santa Cruz, CA) were added; the mixturewas incubated overnight at 4°C. After centrifugation, FSP-1–de-pleted supernatants were collected.

BM Isolation and TransplantationBM cells were harvested by flushing femurs and tibias of donormice. The transplant was created by injecting 5�106 BM cellsinto the lateral tail vein of lethally irradiated (1100 rads) mouserecipients.

Lentivirus Transduction to Knockdown FSP-1 inBM CellsThe short hairpin RNA of FSP-1 in lentivirus particles or controllentivirus particles expressing GFP only (Sigma Aldrich Inc) wereincubated with BM cells from male WT mice. Lentivirus transduc-tions of BM cells were performed over 12 hours in serum-free,StemSpan TM SFEM media (Stem Cell Technologies), in thepresence of cytokines (20 ng/mL IL-3, 100 ng/mL IL-6, and 100ng/mL SCF, R & D Biosynthesis). Transduction efficiencies weredetermined by an analysis of GFP expression produced by controllentivirus particles. Red blood cells were lysed with an ammoniumchloride lysis buffer (150 mmol/L NH4Cl, 10 mmol/L KHCO3,0.1 mmol/L EDTA, pH 7.4).

Coculture and ConditionedMedia-Induced ResponsesFSP-1–GFP� BM cells were isolated by cell sorting and about2�105 purified cells were cultured in the upper chamber of 24-wellBoyden transwell plates (BD Biosciences, San Jose, CA) in DMEMsupplemented with 10% FBS. After 48 hours, 5�104 primary venousSMCs were placed in the cover slip in the lower transwell andexamined with/without added BM cells. After 24 hours, BrdU wasadded for 45 minutes before fixing VSMCs. BrdU immunostainingwas determined according to the manufacturer’s protocol (Roche

Non-standard Abbreviations and Acronyms

AV arteriovenous

BM bone marrow

FSP-1 fibroblast-specific protein 1

GFP green fluorescent protein

MMP matrix metalloproteinase

SDF-1� stromal cell–derived factor 1 alpha

VSMCs venous smooth muscle cells

Cheng et al FSP-1 and Neointima Formation 231


Dow

nloaded from

http://circres.ahajournals.org


Applied Science, Indianapolis, IN). At least 3 independent experi-ments were performed for each experimental condition.

Transendothelial Migration and Invasion AssayMonolayers of 2�105 HUVEC were seeded in 24-well BoydenTranswell inserts for 24 hours and formed a confluent monolayer at37°C in EGM2 media plus SingleQuots (Cambrex Bio Science,Walkersville, Inc). Inserts were placed over 24-well plates coatedwith a thin layer of 1% agarose in 1� PBS to facilitate removal oftransmigrated cells for further analysis. Freshly isolated BM cells(2�106) from WT or FSP-1–GFP� mice were added to the topchamber and allowed to transmigrate through HUVEC at 37°C. After3 or 24 hours, the top chamber was removed and the cell number inthe bottom chamber was determined using a Z1 Beckman Coultercounter (Beckman Coulter, Fullerton, CA) GFP� cells were countedusing fluorescent microscope. For analysis of invasion, collagen gel(100 �L) (Millipore Corp, Billerica, MA) was added to the transwellinsert and kept at 37°C to solidify the gel. BM cells (2�106) fromWT or FSP-1–GFP� mice were added to the gels; after culturing for7 days, cells on the membrane were fixed and counterstained withDAPI. The total number of cells and the number of GFP� cells werecounted using a fluorescent microscope. Each experiment wasrepeated at least 3 times to confirm our results. To determine the roleof Rac-1 on BM cell transmigration, BM cells (5 million) wereelectroporated with or without dominant-negative Rac N17 expres-sion plasmid (5 �g) by Neon system (Invitrogen); the workingcondition was set as voltage at 1650 V, 10 ms for 2 pulses.

Statistical AnalysisAll data are presented as mean�SEM. Comparison between groupswas made using 1-way ANOVA; P�0.05 was considered statisti-cally significant.

ResultsFSP-1 Expression in AV GraftsVein grafts were created in FSP-1–GFP� mice that expressedGFP under the control of the FSP-1 promoter.17 Histologi-cally, remodeled vein grafts in FSP-1–GFP� mice weresimilar to those in WT mice. Specifically, in vena cavas fromFSP-1–GFP mice, there were no FSP-1� cells, but 1 monthafter creation of the vein grafts, approximately one-third of

the cells in vein grafts were FSP-1 positive (Figure 1A). MostFSP-1� cells were located in the adventitia, but they alsowere present in the neointima and endothelial layers (Figure1A). Results from Western blots and real-time PCR con-firmed the increase in FSP-1 expression in vein grafts after 1month (42�7.2-fold compared with FSP-1 in isolated venacavas; Figure 1B and 1C). GFP staining in vein grafts placedin FSP-1–GFP� mice colocalized with FSP-1 (Figure 1D); noGFP staining was found in vein grafts placed in WT mice(data not shown).

Accumulation of BM-Derived FSP-1� Cells inVein GraftsTo explore the origin of FSP-1� cells present in AV grafts,we created AV grafts using donor vena cavas from FSP-1–GFP� or WT mice and placing them in WT or FSP-1–GFP�

mice, respectively. GFP� cells were only found in AV graftscreated in FSP-1–GFP� mice using veins from WT mice(Figure 2A). GFP� cells were absent in grafts created in WTmice using vena cavas taken from FSP-1–GFP� mice (Figure2B). Thus, FSP-1� cells did not arise from donor vena cavas.GFP� cells were also found in AV grafts created in lethallyirradiated WT mice that had BM transplants from FSP-1–GFP� mice (Figure 2D); a finding not present in AV grafts inWT mice transplanted with WT BM cells (Figure 2C). ThusFSP-1� cells found in AV grafts originate from BM cells.

BM-Derived FSP-1 Cells Express Markers ofInflammatory Cells in Vein GraftsUsing flow cytometry, we found no detectable FSP-1 expres-sion in cells of either arteries or vena cavas freshly isolatedfrom WT mice (Online Figure I, A). However, �8% of BMcells were FSP-1 positive and 85.0�6.2% of the FSP-1–positive cells also stained positively for CD11b and20.3�1.7% were CD45 positive (Figure 3A and 3B). Nota-bly, GFP� BM cells from FSP-1–GFP mice also stained

Figure 1. FSP-1� cells are present in AVgrafts. A through C, Increased FSP-1expression in AV grafts was detected byimmunostaining (A), by Western blot (B),and by real-time RT-PCR (C). D, Colocal-ization of GFP� and FSP-1� cells wasdetected in AV grafts created in FSP-1–GFP� mice. The FSP-1 signal is in red,GFP in green, and nuclei are stained withDAPI in blue.

232 Circulation Research January 20, 2012


Dow

nloaded from


positively for collagen I and fibronectin but not for collagenIII or SMA-� (Online Figure I, B and C). In vein grafts,FSP-1� cells costained positively for CD45 and CD11b(Figure 3, C and D). Most FSP-1� cells did not costainpositively with F4/80 (Online Figure I, D) in vein grafts after1 month.

Stromal Cell–Derived Factor 1� From PlateletsAttracts FSP-1� Cells Into Vein GraftsIn donor vena cavas placed in aorta, endothelial cells werelost over the initial 24 hours after creating the AV graft.These cells were not replaced by 72 hours (Figure 4A).Instead, the exposed subendothelium was covered with acti-vated platelets that stained positively for the platelet marker,CD41 (Online Figure II). At 5 days after creating the AVgraft, there were FSP-1� cells surrounded by platelets (Figure4B). Notably, SDF-1� is secreted by activated platelets andfunctions as a chemoattractant for BM-derived progenitorcells.18 Therefore, we evaluated SDF-1� expression. It wasdetected within days of creating the vein graft and disap-peared after 1 month (Figure 4C). CD41� platelets wereaggregated on the vein graft lacking endothelial cells andstained positively for SDF-1� (Figure 4D). Notably, somecells in the grafted vein coexpressed FSP-1 and CXCR4(Figure 4E), suggesting that SDF-1� that is secreted byplatelets attracts FSP-1� cells into AV grafts and that thesecells can respond to SDF-1� through its receptor, CXCR4.

Transendothelial Migration and InvasiveProperties of BM FSP-1� cellsIn evaluating the function of FSP-1 cells from the BM, wefound that transmigration of BM cells across an endothelial

cell layer was significantly blocked when FSP-1 was knockeddown in BM cells (Figure 5A). Likewise, pretreatment of BMcells with anti-CXCR4 antibodies led to a marked decrease inthe transmigration potential of BM cells (Figure 5A). Thepercentage of GFP� cells (ie, from FSP-1–GFP transgenicmice) that transmigrated increased significantly (47�6.3%)when compared with only 8�1.3% of GFP� cells in BM ofFSP-1–GFP� mice (Figure 5B). Addition of SDF-1� in-creased the transmigration of GFP� BM cells; this responsewas dramatically suppressed in BM cells that had beentreated with FSP-1 shRNA or incubated with CXCR4 anti-bodies (Figure 5B). Treatment of BM cells to knockdownFSP-1 or addition of anti-CXCR4 antibodies to the BM cellsalso reduced the invasion of BM cells into 3D matrix gels,even though SDF-1� was present (Figure 5C). Furthermore,we found that increased invasion of the GFP�-BM cells into3D collagen gels was much higher compared with invasion ofthe GFP�-BM cells (Figure 5D).To determine the underlyingmechanism for FSP-1–mediated migration, we evaluated therole of small G-protein, Rac1, because its activity is relatedwith cell migration19,20; we found that overexpression ofFSP-1 stimulated PAK-bound Rac1 activity (Figure 5E) andenhanced Rac1 membrane translocation (Figure 5F) in BMcells. Finally, FSP-1–stimulated BM-cell transendothelialmigration was markedly blocked by dominant-negative Rac1expression (Figure 5G).

Expression of FSP-1 Induces VSMC ProliferationFSP-1� cells in an AV graft were present in neointima areasthat also exhibited PCNA� proliferating cells (Figure 6A). Toexplore whether FSP-1� cells could be involved in stimulat-

Figure 2. Bone marrow–derived-FSP-1�

cells accumulate in AV grafts. A,FSP-1� cells come from the hostbecause GFP� cells were present in thevein grafts created in FSP-1–GFP� micewith vena cavas from WT mice. GFP andSMA-� expression were detected in thegrafted vena cavas. Arrow indicates posi-tive staining of GFP in endothelial cells.B, Expression of SMA-� but not GFP wasdetected in vein grafts created in WTmice using vena cavas from FSP-1–GFP�

mice. C and D, Bone marrow–derivedFSP-1� cells accumulated in vein graftsonly in mice with BM reconstituted fromFSP-1–GFP� mice.



Dow

nloaded from


ing VSMCs hyperplasia, we isolated BM-derived FSP-1�

cells by sorting GFP� cells from BM of FSP-1–GFP� mice.Coculture of these GFP� BM cells with VSMCs isolatedfrom vena cavas of WT mice led to proliferation of theVSMCs (Figure 6C). The link FSP-1 with proliferation wasstrengthened by our finding that lentivirus-mediated knock-down of FSP-1 in BM cells (Figure 6B) abolished its effect onVSMC proliferation (Figure 6C). Moreover, we found thatconditioned media from BM cells isolated from FSP-1–GFP�

mice stimulated the growth of VSMCs, whereas the condi-tioned media from BM cells expressing shRNA to FSP-1failed to stimulate VSMC proliferation (Figure 6D). Finally,depletion of FSP-1 from the conditioned media with a FSP-1antibody significantly attenuated VSMC proliferation in-duced by conditioned media of FSP-1–GFP� BM cells(Figure 6E).

FSP-1 Stimulates VSMC Proliferation ThroughActivating Rac1 and ERK5We found that VSMCs exposed to the recombinant FSP-1protein exhibited increased PCNA expression and BrdUincorporation in VSMCs (Figure 7A and 7B). To determine

FSP-1–stimulated signaling responsible for this event, wefound that FSP-1 significantly stimulated ERK5 phosphory-lation; the earliest stimulation was observed at 5 minutes andpeaked at 30 minutes (Figure 7C). FSP-1 also dose-dependently stimulated ERK5 phosphorylation (Online Fig-ure III). In contrast, Treatment with FSP-1 has no significanteffect on pAKT and pJNK levels (Figure 7C). Moreover, wefound that inhibition of ERK5 by transfecting VSMCs withERK5 siRNA or dominant-negative MEK5 suppressed FSP-1–induced expression of cell proliferation marker PCNA andcyclin D1 (Figure 7D and 7E). Similar responses were foundin cell growth curve analysis (Figure 7F). Therefore, ourresults showed that both Rac1 and ERK5 were activated byFSP-1 and were critical for FSP-1–mediated cell migrationand proliferation. To elucidate the sequence of activation ofthese molecules by FSP-1, we measured ERK5 and Rac1activation when one of molecules was inhibited. The expres-sion of dominant-negative Rac1(N17) inhibited the FSP-1–induced phosphorylation of MEK5 and ERK5 (Figure 7G)and proliferation (Figure 7H) in VSMCs. Although theexpression of dominant-negative MEK5 blocked phosphory-lation of ERK5, it had no effect on FSP-1–stimulated Rac1

Figure 3. Characterization of FSP-1� cells. A, Flow cytometic analysis of FSP-1 expression in mouse bone marrow; bone marrowfrom WT or FSP-1–GFP� mice were collected to identify GFP� cells. B, BM cells were isolated from FSP-1–GFP� mice and incubatedwith antibodies to cell markers leukocyte CD45 and mononuclear cells CD11b before flow cytometic analysis. C and D, Characteriza-tion of FSP-1� cells in vein grafts. Costaining of vein grafts for FSP-1 and CD45 (C), CD11b (D). FSP-1 staining in these panels is inred. DAPI in blue and arrows indicate colocalization between FSP-1 and other cell markers. Representative data from at least 3 AVgrafts are shown.



Dow

nloaded from


activation (Figure 7I) and could not block FSP-1–inducedBM cell transendothelial migration (data not shown). Theseresults suggest that Rac1 acts upstream of MEK5/ERK5 inresponse to FSP-1.

BM Transplantation With FSP-1 Knockdown CellsSuppresses Neointima Formation and VSMCProliferation in AV GraftsTo determine the role of FSP-1 in vivo, WT mice that hadreceived a BM transplant with BM cells in which FSP-1 hadbeen knocked down, there was downregulation of FSP-1expression (data not shown). In vein grafts that were createdin WT mice that had been transplanted with knockdownFSP-1–BM, the function of the AV graft (measured as thelumen-to-neointima ratio) was significantly improved (Figure8A through 8C). AV grafts in these mice also had fewerFSP-1� cells as well as reduced SMA-� expression (Figure8D and 8E). PCNA� proliferative cells were decreased in AVgrafts in mice transplanted with FSP-1 knockdown BM cells(Figure 8F and 8G).

DiscussionFSP-1 (also known as S100A4, mts1, p9Ka, and metastasin)can mediate cellular proliferation, including the proliferationof VSMCs.21,22 FSP-1 was first cloned from serum-stimulatedmurine fibroblasts,23 and its overexpression was associatedwith neointima developing in the pulmonary artery; it alsowas found to be expressed in plexogenic arteriopathy ofhumans.24 These associations with neointima are similar tothose we have uncovered (Figure 1).

FSP-1 has been implicated in controlling cell motility,possibly through the formation of complexes with a numberof intracellular proteins (eg, the heavy chain of nonmusclemyosin II,9,25 liprin �1,26 and p5327). Moreover, cancer cells

expressing FSP-1 can exhibit aggressive migration propertieswith upregulation of matrix metalloproteinase 9,28 raising thepossibility that FSP-1 could influence the migration of cellsto form a neointima. For example, FSP-1 transgenic mice willform neointima in pulmonary arteries.21,29 Our data indicatethat FSP-1 is expressed in BM-derived inflammatory cells(Figures 2 and 3), suggesting that FSP-1 may influence themotility of inflammatory cells infiltrating AV grafts. Inter-estingly, only a small fraction (8�1.3%) of BM cells areFSP-1� in mice (Figure 3A), but these cells have an increasedpotential to migrate across an endothelial cells layer andinvade in collagen matrix (Figure 5C and 5D). Because thecreation of an AV graft results in rapid loss of endothelialcells and since the neointima contains inflammatory cells, ourresults suggest that the FSP-1� cells regulate these processesthat occur early in AV graft dysfunction.

We found that knockdown of FSP-1 in BM cells markedlyinhibited transendothelial migration and the invasive activi-ties of the cells. Because monocytes, leukocytes, and macro-phages coexpress FSP-1, it is not surprising that knockdownof FSP-1 can suppress the inflammatory responses present inan AV graft. The increased motility occurring in FSP-1expressing cells could be mediated in 2 fashions: First, FSP-1can suppress the expression of a tissue inhibitor of metallo-proteinase while increasing matrix metalloproteinases(MMP) expression.30,31 These responses would degrade theextracellular matrix improving cell migration and invasiveability. Second, FSP-1 can directly bind with cytoskeletalproteins to stimulate cell motility. For example, FSP-1 canbind non–smooth muscle myosin-IIA with alternation of cellmotility.9,32 Rac1 is a small G-protein that regulates cytoskel-eton organization and cell migration. We found that Rac1activation mediated FSP-1–induced BM-cell transmigration(Figure 5E through 5G). This is consistent with previous

Figure 4. Platelet secretion of SDF-1�leads to FSP-1� cells homing to AVgrafts. A, At 24 hours after creating theAV graft, the endothelium was denudedas detected by the staining of PECAM(red). Nuclei were stained by DAPI. B,FSP-1� and CD41 cells were found onthe lumen surface at 5 days after AVgrafts. C, AV grafts collected at differenttimes were stained for SDF-1� and DAPI.D, Double staining with the plateletmarker CD41, and SDF-1� indicates thatSDF-1� released from CD41� cells. E,Double staining of FSP-1 and the SDF-1�receptor CXCR4 suggests that SDF-1�recruits FSP-1–positive cells. FSP-1�

cells homing to AV grafts was associatedwith platelet-produced SDF-1�.



Dow

nloaded from


report that FSP-1 stimulates VSMC migration through Rhoand Rac activation.33

The hallmark of AV graft failure is neointima formationcharacterized by VSMC proliferation. Our results demon-strate that BM-derived FSP-1� cells accumulate in the AVgraft, suggesting that they participate in the formation of aneointima. More concrete evidence of their role is thattransplantation of the BM cells treated to knockdown FSP-1expression significantly suppresses neointima formation (Fig-ure 8). Thus, the actions of FSP-1� cells are important inmediating the dysfunction of AV grafts. FSP-1–mediatedresponses could involve at least 2 pathways: (1) an increase intransendothelial migration and collagen gel invasion ofFSP-1� inflammatory cells and (2) local paracrine stimula-tion of VSMC proliferation in AV grafts. In fact, FSP-1 isknown to be secreted,34,35 and our results suggest it acts in aparacrine fashion to increase proliferation of VSMC. Specif-ically, we demonstrated that coculture of FSP-1� BM cellsstimulates the growth of VSMC (Figure 6). We also foundthat VSMC proliferation was blocked by inhibition of MEK5-ERK5 signaling pathway, suggesting that FSP-1 stimulates

proliferation through ERK5 activation (Figure 7). ERK5 hasbeen linked with cell proliferation and differentiation.16 Wefound that FSP-1 stimulated ERK5 phosphorylation in a time-and dose-dependent manner. Furthermore, we found thatRac1 can regulate FSP-1–induced MEK5/ERK5 phosphory-lation and VSMC proliferation (Figure 7G and 7H). Similarresponses have been reported that Rac1 involved in regulat-ing MAPK activation through p21-activated kinases36 orJNK/AP1 pathway.37 In contrast, FSP-1–stimulated Rac1activation could not be inhibited by negative dominantMEK5, suggesting that Rac1 acts as an upstream of ERK5.

FSP-1 signaling could be stimulated by activating thereceptor for advanced glycation end products (RAGE). In thisformulation, FSP-1 proteins (ie, S100 proteins), released frominflammatory cells, could activate RAGE and induce proin-flammatory adhesion molecules as well as cytokines, tissue-destroying elastases, and MMPs.35 Serotonin has been shownto stimulate the release of FSP-1 from SMCs,38 and FSP-1can bind with RAGE and induce migration of SMCs throughan intracellular chloride channel 4.33 Oligomerization of S100proteins appears to be essential for RAGE-mediated stimula-

Figure 5. SDF-1� stimulates the motilityof FSP-1� BM cells. A, Using an in vitrotransmigration system, BM cells wereexamined for their ability to cross an en-dothelial cell layer. The transmigration wasblocked by knocking down FSP-1 in BMcells or by adding neutralizing antibodiesagainst the SDF-1� receptor CXCR4.Wild-type BM cells were added to the topchamber and SDF-1� (5 ng/mL) to thelower chamber. BM cells that had trans-migrated to the endothelial cell layer after3 or 24 hours were counted. B, FSP-1�

BM cells exhibited increased transmigra-tion after SDF-1� exposure. Similar exper-iments as described in A were performed,except that BM cells were isolated fromFSP-1–GFP� mice and placed in the topchamber. The percentage of transmigrat-ing cells that were GFP� cells was calcu-lated. The inset shows the transmigratedcells. The bar graphs represent resultsfrom 3 repeated experiments. C, Collagengel (100 �L) was solidified in the top Boy-den chamber, and wild-type BM cells orBM cells treated with FSP-1 shRNA wereadded to the top of the gel. In otherexperiments, BM cells incubated with orwithout anti-CXCR4 antibodies wereadded to the top chamber. SDF-1� wasadded to the lower chamber, and the cel-lular invasion was assessed after 7 days.D, Bar graph shows the percentage ofGFP� BM cells (ie, from FSP-1–GFP�

mice) invading into the collagen gel.Graph represents results from 3 separateexperiments. E and F, Overexpression ofFSP-1–induced Rac1 activation. BM cellswere infected with recombinant adenovi-rus AdFSP-1; the Rac1 activation was an-alyzed by Western blot after the Rac1-GTP was pulled down with PAK1 PBDagarose beads (E). The Rac1 localizationwas detected by immunostaining andimaging (F). G, Rac1 mediated FSP-1–

stimulated transmigration of BM cells. The transmigration induced by FSP-1 overexpression was suppressed by expression ofdominant-negative Rac1 (N17) in BM cells.



Dow

nloaded from


tion of MMP in chondrocyes.39 RAGE also mediatesS100A4-stimulated MMP13 release.30

We found that platelets (stained positively for CD41, theplatelet marker) were deposited in the denuded AV graft

(Figure 4). Platelets also expressed SDF-1�, which is re-leased from apoptotic VSMCs and platelets.18 Because adhe-sion of platelet to the exposed subendothelium is the firstresponse to vascular injury,40 we explored the involvement of

Figure 6. FSP-1 induces VSMC proliferation. A, At 1 month after creating the AV graft, serial slides of the vein grafts were examined.FSP-1–positive cells were costained with SMA-� antibodies (left panel); proliferation was assessed by PCNA staining (right panel). B,BM cells infected with FSP-1 shRNA lentivirus for 24 hours had reduced the expression of FSP-1, even though BM FSP-1� cells hadbeen exposed to PDGF-BB for 24 hours. C, VSMCs were cocultured with BM cells that had been treated to knockdown FSP-1. In 3different experiments, there was a decrease in BrdU-positive cells counted in 5 different areas. D, Conditioned-media (CM) from BMFSP-1� cells induced VSMC proliferation as measured by BrdU incorporation. This response was blocked by adding CM from BM cellsthat had been treated to knockdown FSP-1. E, CM from FSP-1� cells was incubated with control or anti–FSP-1 antibodies and subse-quently immunoprecipitated with agarose-A/G to remove FSP-1. VSMCs treated with FSP-1-conditioned media were subjected to BrdUanalysis.

Figure 7. FSP-1 stimulates VSMC prolifer-ation and migration through Rac1-dependent pathway. A and B, After 48hours of exposure to recombinant FSP-1,VSMC proliferation increased as detected byPCNA expression (A) and BrdU incorporation(B). C, FSP-1 stimulated ERK5 phosphoryla-tion. VSMCs were treated with FSP-1 (100ng/mL) after serum starvation for 24 hours;cell lysates were collected, and the phosphor-ylation of ERK5, ERK1/2, p38, JNK, and AKTwere detected by Western blots. GAPDH wasused as a loading control. D, ERK5 siRNAknocked down ERK5 expression in VSMCs.E and F, FSP-1–induced cell proliferation wasinhibited by blocking MEK5-ERK5 signalingpathway. VSMCs were transfected with ERK5siRNA or MEK5 dominant-negative plasmids;the FSP-1–induced proliferative markers weredetected by Western blot (E); and the XTTassay were used to measure the cell prolifer-ation (F). The effects of FSP-1 on MEK5/ERK5 phosphorylation (G) and proliferation (H)were suppressed by dominant-negative Rac1expression plasmids in VSMCs. I, Theexpression of dominant-negative MEK5 hadno effect on FSP-1–induced Rac1 activation.



Dow

nloaded from


the SDF-1�/CXCR4 axis in the recruitment of BM-derivedFSP-1� cells into the AV graft. SDF-1� expression wasincreased in AV grafts. In denuded veins of the AV graft, wefound cells expressing both FSP-1 and the SDF-1� receptor,CXCR4 (Figure 4). Because SDF-1� expression can mobilizeand recruit peripheral blood progenitor cells into injuredvessels,41 our results suggest that the recruitment of FSP-1�

cells occurs when platelets are deposited on the denuded venacava of the AV graft.

It is widely accepted that neointima formation involvesmigration and proliferation of adjacent cells. These also is,however, accumulating evidence that cells from the BM giverise to endothelial cells and VSMCs, which contribute tovascular healing, remodeling, and neointima formation.42

Because BM cells can be directly involved in vascularremodeling, strategies to prevent neointima formation havebeen based on genetically engineering BM cells. For exam-ple, knockout of ROCK-143 or ASC44 (apoptosis-associatedspeck-like protein containing a caspase recruitment domain)in BM cells was found to decrease neointima formation aftercarotid artery ligation. Moreover, we found that knockdownFSP-1 in BM cells will decrease the accumulation of inflam-matory cells and the formation of neointima in AV grafts(Figure 8).

We conclude that after creation of an AV graft, theendothelial layer cell layer is rapidly lost, and platelets coverthe subendothelium. SDF-1� is secreted by these platelets torecruit BM-derived FSP-1� cells into the vein graft. Cellsexpressing FSP-1 exhibit a higher Rac1 activation and anincreased capacity for migration and invasion. This results inthe development of inflammation from penetration and infil-tration of FSP-1� cells into the vein grafts. Secreted FSP-1

also exhibits a paracrine role, which stimulates MEK5-ERK5signaling pathway through Rac1 activation to promote pro-liferation of surrounding VSMCs, leading to neointimaformation.

AcknowledgmentsWe acknowledge Dr William E. Mitch for inspirational support.

Sources of FundingThis study was supported by grants from the Chinese Ministry ofScience and Technology (2012CB945100 and 2009CB522205),grants from the National Science Foundation of China (30888004,31090363), an American Heart Association Grant 10SDG2780009(to J.C.), and a generous grant from Dr and Mrs Harold Selzman.

DisclosuresNone.

References1. Dietrich H, Hu Y, Zou Y, Huemer U, Metzler B, Li C, Mayr M, Xu Q.

Rapid development of vein graft atheroma in apoE-deficient mice. Am JPathol. 2000;157:659–669.

2. Kobayashi K, Yokote K, Fujimoto M, Yamashita K, Sakamoto A,Kitahara M, Kawamura H, Maezawa Y, Asaumi S, Tokuhisa T, Mori S,Saito Y. Targeted disruption of TGF-�-Smad3 signaling leads toenhanced neointimal hyperplasia with diminished matrix deposition inresponse to vascular injury. Circ Res. 2005;96:904–912.

3. Wernig F, Mayr M, Xu Q. Mechanical stretch-induced apoptosis insmooth muscle cells is mediated by beta1-integrin signaling pathways.Hypertension. 2003;41:903–911.

4. McKee JA, Banik SS, Boyer MJ, Hamad NM, Lawson JH, Niklason LE,Counter CM. Human arteries engineered in vitro. EMBO Rep. 2003;4:633–638.

5. Motwani JG, Topol EJ. Aortocoronary saphenous vein graft disease:pathogenesis, predisposition, and prevention. Circulation. 1998;97:916–931.

Figure 8. Knockdown FSP-1 in bonemarrow cells suppresses neointimaformation in AV grafts. A, AV grafts werecreated in WT mice that had been trans-planted with BM from WT mice or BMcells treated to knockdown FSP-1. Hema-toxylin and eosin staining was performed.B and C, the areas of the neointima andlumen from grafts in each group are pre-sented. D and E, Vein graft segmentswere inmunostained for FSP-1 andSMA-�. F and G, PCNA in vein graft seg-ments was detected by immunofluores-cent staining. Results were comparedwith those present in vein grafts from WTmice that had FSP-1 knockdown in BMtransplants.



Dow

nloaded from


6. Werner N, Nickenig G. Influence of cardiovascular risk factors on endo-thelial progenitor cells: limitations for therapy? Arterioscler Thromb VascBiol. 2006;26:257–266.

7. Caplice NM, Wang S, Tracz M, Croatt AJ, Grande JP, Katusic ZS, NathKA. Neoangiogenesis and the presence of progenitor cells in the venouslimb of an arteriovenous fistula in the rat. Am J Physiol Renal Physiol.2007;293:F470–F475.

8. Ebralidze A, Tulchinsky E, Grigorian M, Afanasyeva A, Senin V,Revazova E, Lukanidin E. Isolation and characterization of a gene spe-cifically expressed in different metastatic cells and whose deduced geneproduct has a high degree of homology to a Ca2�-binding protein family.Genes Dev. 1989;3:1086–1093.

9. Li ZH, Bresnick AR. The S100A4 metastasis factor regulates cellularmotility via a direct interaction with myosin-IIA. Cancer Res. 2006;66:5173–5180.

10. Chen M, Sinha M, Luxon BA, Bresnick AR, O’Connor KL. Integrinalpha6beta4 controls the expression of genes associated with cell motility,invasion, and metastasis, including S100A4/metastasin. J Biol Chem.2009;284:1484–1494.

11. Schmidt-Hansen B, Ornas D, Grigorian M, Klingelhofer J, Tulchinsky E,Lukanidin E, Ambartsumian N. Extracellular S100A4 (mts1) stimulatesinvasive growth of mouse endothelial cells and modulates MMP-13matrix metalloproteinase activity. Oncogene. 2004;23:5487–5495.

12. Kiryushko D, Novitskaya V, Soroka V, Klingelhofer J, Lukanidin E,Berezin V, Bock E. Molecular mechanisms of Ca(2�) signaling inneurons induced by the S100A4 protein. Mol Cell Biol. 2006;26:3625–3638.

13. Cheng J, Du J. Mechanical stretch simulates proliferation of venoussmooth muscle cells through activation of the insulin-like growth factor-1receptor. Arterioscler Thromb Vasc Biol. 2007;27:1744–1751.

14. Li ZH, Spektor A, Varlamova O, Bresnick AR. Mts1 regulates theassembly of nonmuscle myosin-IIA. Biochemistry. 2003;42:14258–14266.

15. Nobes CD, Hall A. Rho GTPases control polarity, protrusion, andadhesion during cell movement. J Cell Biol. 1999;144:1235–1244.

16. Guo Y-S, Cheng J-Z, Jin G-F, Gutkind JS, Hellmich MR, Townsend CM.Gastrin stimulates cyclooxygenase-2 expression in intestinal epithelialcells through multiple signaling pathways. J Biol Chem. 2002;277:48755–48763.

17. Iwano M, Plieth D, Danoff TM, Xue C, Okada H, Neilson EG. Evidencethat fibroblasts derive from epithelium during tissue fibrosis. J ClinInvest. 2002;110:341–350.

18. Massberg S, Konrad I, Schurzinger K, Lorenz M, Schneider S, Zohln-hoefer D, Hoppe K, Schiemann M, Kennerknecht E, Sauer S, Schulz C,Kerstan S, Rudelius M, Seidl S, Sorge F, Langer H, Peluso M, Goyal P,Vestweber D, Emambokus NR, Busch DH, Frampton J, Gawaz M.Platelets secrete stromal cell-derived factor 1alpha and recruit bonemarrow-derived progenitor cells to arterial thrombi in vivo. J Exp Med.2006;203:1221–1233.

19. Azab AK, Azab F, Blotta S, Pitsillides CM, Thompson B, Runnels JM,Roccaro AM, Ngo HT, Melhem MR, Sacco A, Jia X, Anderson KC, LinCP, Rollins BJ, Ghobrial IM. RhoA and Rac1 GTPases play major anddifferential roles in stromal cell derived factor-1� induced cell adhesionand chemotaxis in multiple myeloma. Blood. 2009;114:619–629.

20. Li C, Hu Y, Sturm G, Wick G, Xu Q. Ras/Rac-dependent activation ofp38 mitogen-activated protein kinases in smooth muscle cells stimulatedby cyclic strain stress. Arterioscler Thromb Vasc Biol. 2000;20:E1–E9.

21. Brisset AC, Hao H, Camenzind E, Bacchetta M, Geinoz A, Sanchez JC,Chaponnier C, Gabbiani G, Bochaton-Piallat ML. Intimal smooth musclecells of porcine and human coronary artery express S100A4, a marker ofthe rhomboid phenotype in vitro. Circ Res. 2007;100:1055–1062.

22. Sherbet GV. Metastasis promoter S100A4 is a potentially valuablemolecular target for cancer therapy. Cancer Lett. 2009;280:15–30.

23. Jackson-Grusby LL, Swiergiel J, Linzer DI. A growth-related mRNA incultured mouse cells encodes a placental calcium binding protein. NucleicAcids Res. 987;15:6677–6690.

24. Greenway S, van Suylen RJ, Du Marchie Sarvaas G, Kwan E, Ambart-sumian N, Lukanidin E, Rabinovitch M. S100A4/Mts1 produces murinepulmonary artery changes resembling plexogenic arteriopathy and isincreased in human plexogenic arteriopathy. Am J Pathol. 2004;164:253–262.

25. Kriajevska MV, Cardenas MN, Grigorian MS, Ambartsumian NS,Georgiev GP, Lukanidin EM. Non-muscle myosin heavy chain as apossible target for protein encoded by metastasis-related mts-1 gene.J Biol Chem. 1994;269:19679–19682.

26. Kriajevska M, Fischer-Larsen M, Moertz E, Vorm O, Tulchinsky E,Grigorian M, Ambartsumian N, Lukanidin E. Liprin beta 1, a member ofthe family of LAR transmembrane tyrosine phosphatase-interactingproteins, is a new target for the metastasis-associated protein S100A4(Mts1). J Biol Chem. 2002;277:5229–5235.

27. van Dieck J, Fernandez-Fernandez MR, Veprintsev DB, Fersht AR.Modulation of the oligomerization state of p53 by differential binding ofproteins of the S100 family to p53 monomers and tetramers. J Biol Chem.2009;284:13804–13811.

28. Saleem M, Kweon MH, Johnson JJ, Adhami VM, Elcheva I, Khan N, BinHafeez B, Bhat KM, Sarfaraz S, Reagan-Shaw S, Spiegelman VS,Setaluri V, Mukhtar H. S100A4 accelerates tumorigenesis and invasion ofhuman prostate cancer through the transcriptional regulation of matrixmetalloproteinase 9. Proc Natl Acad Sci U S A. 2006;103:14825–14830.

29. Spiekerkoetter E, Alvira CM, Kim YM, Bruneau A, Pricola KL, Wang L,Ambartsumian N, Rabinovitch M. Reactivation of gammaHV68 inducesneointimal lesions in pulmonary arteries of S100A4/Mts1-overexpressingmice in association with degradation of elastin. Am J Physiol Lung CellMol Physiol. 2008;294:L276–L289.

30. Yammani RR, Carlson CS, Bresnick AR, Loeser RF. Increase in produc-tion of matrix metalloproteinase 13 by human articular chondrocytes dueto stimulation with S100A4: role of the receptor for advanced glycationend products. Arthritis Rheum. 2006;54:2901–2911.

31. Bjornland K, Winberg JO, Odegaard OT, Hovig E, Loennechen T, AasenAO, Fodstad O, Maelandsmo GM. S100A4 involvement in metastasis:deregulation of matrix metalloproteinases and tissue inhibitors of matrixmetalloproteinases in osteosarcoma cells transfected with an anti-S100A4ribozyme. Cancer Res. 1999;59:4702–4708.

32. Kallis P, Unsworth-White J, Munsch C, Gallivan S, Smith EE, Parker DJ,Pepper JR, Treasure T. Disability and distress following cardiac surgeryin patients over 70 years of age. Eur J Cardiothorac Surg. 1993;7:306–312.

33. Spiekerkoetter E, Guignabert C, de Jesus Perez V, Alastalo T-P, PowersJM, Wang L, Lawrie A, Ambartsumian N, Schmidt A-M, Berryman M,Ashley RH, Rabinovitch M. S100A4 and bone morphogenetic protein-2codependently induce vascular smooth muscle cell migration viaphospho-extracellular signal-regulated kinase and chloride intracellularchannel 4. Circ Res. 2009;105:639–647.

34. Ambartsumian N, Klingelhofer J, Grigorian M, Christensen C, KriajevskaM, Tulchinsky E, Georgiev G, Berezin V, Bock E, Rygaard J, Cao R, CaoY, Lukanidin E. The metastasis-associated Mts1(S100A4) protein couldact as an angiogenic factor. Oncogene. 2001;20:4685–4695.

35. Basta G, Lazzerini G, Massaro M, Simoncini T, Tanganelli P, Fu C,Kislinger T, Stern DM, Schmidt AM, De Caterina R. Advanced glycationend products activate endothelium through signal-transduction receptorRAGE: a mechanism for amplification of inflammatory responses. Cir-culation. 2002;105:816–822.

36. Deleris P, Trost M, Topisirovic I, Tanguay P-L, Borden KLB, Thibault P,Meloche S. Activation loop phosphorylation of ERK3/ERK4 by group Ip21-activated kinases (PAKs) defines a novel PAK-ERK3/4-MAPK-ac-tivated protein kinase 5 signaling pathway. J Biol Chem. 2011;286:6470–6478.

37. Bianchi R, Kastrisianaki E, Giambanco I, Donato R. S100B proteinstimulates microglia migration via RAGE-dependent up-regulation ofchemokine expression and release. J Biol Chem. 2011;286:7214–7226.

38. Lawrie A, Spiekerkoetter E, Martinez EC, Ambartsumian N, ShewardWJ, MacLean MR, Harmar AJ, Schmidt A-M, Lukanidin E, RabinovitchM. Interdependent serotonin transporter and receptor pathways regulateS100A4/Mts1, a gene associated with pulmonary vascular disease. CircRes. 2005;97:227–235.

39. Donato R. RAGE: a single receptor for several ligands and differentcellular responses: the case of certain S100 proteins. Curr Mol Med.2007;7:711–724.

40. Massberg S, Schurzinger K, Lorenz M, Konrad I, Schulz C, Plesnila N,Kennerknecht E, Rudelius M, Sauer S, Braun S, Kremmer E, EmambokusNR, Frampton J, Gawaz M. Platelet adhesion via glycoprotein IIb integrinis critical for atheroprogression and focal cerebral ischemia: an in vivostudy in mice lacking glycoprotein IIb. Circulation. 2005;112:1180–1188.

41. Tanaka K, Sata M, Hirata Y, Nagai R. Diverse contribution of bonemarrow cells to neointimal hyperplasia after mechanical vascular injuries.Circ Res. 2003;93:783–790.

42. Tanaka K, Sata M. Contribution of circulating vascular progenitors inlesion formation and vascular healing: lessons from animal models. CurrOpin Lipidol. 2008;19:498–504.



Dow

nloaded from


43. Noma K, Rikitake Y, Oyama N, Yan G, Alcaide P, Liu PY, Wang H,Ahl D, Sawada N, Okamoto R, Hiroi Y, Shimizu K, Luscinskas FW,Sun J, Liao JK. ROCK1 mediates leukocyte recruitment and neointimaformation following vascular injury. J Clin Invest. 2008;118:1632–1644.

44. Yajima N, Takahashi M, Morimoto H, Shiba Y, Takahashi Y, MasumotoJ, Ise H, Sagara J, Nakayama J, Taniguchi S, Ikeda U. Critical role ofbone marrow apoptosis-associated speck-like protein, an inflammasomeadaptor molecule, in neointimal formation after vascular injury in mice.Circulation. 2008;117:3079–3087.

Novelty and Significance

What Is Known?

● Arteriovenous (AV) grafts fail because of neointima formation.● Smooth muscle cell (SMC) proliferation is a major reason for intimal

hyperplasia.● In addition to SMCs, the neointima contains several other types of

cells, including fibroblast specific protein 1 (FSP-1)–positive cells.

What New Information Does This Article Contribute?

● FSP-1–positive cells are recruited by a mechanism involving aplatelet-secreted stromal cell–derived factor (SDF-1) and itsreceptor.

● FSP-1 promotes inflammatory cell infiltration into the subendothelialspace in AV grafts. It also functions as paracrine growth factorstimulating SMC growth.

● FSP-1 knockdown in bone marrow cells inhibited neointimaformation.

The hallmark of vein graft failure is neointima formationcharacterized by SMC accumulation, but the mechanism of cellaccumulation in the neointima is poorly understood. In thisstudy, we found that bone marrow–derived FSP-1–positive cellswere recruited into the neointima by a mechanism involving aplatelet-secreted stromal cell–derived factor and its receptor.Our results show that FSP-1 promotes inflammatory cell infil-tration into neointima and that it functions as paracrine growthfactor, which stimulates SMC growth. Ex vivo FSP-1 knockdownin bone marrow attenuated FSP-1–induced inflammatory cellinfiltration, SMC proliferation, and neointima formation. Thesefindings suggest that FSP-1 may have an obligatory role in veingraft failure.



Dow

nloaded from


Jizhong Cheng, Yun Wang, Anlin Liang, Lixin Jia and Jie DuFSP-1 Silencing in Bone Marrow Cells Suppresses Neointima Formation in Vein Graft

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 2011 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/CIRCRESAHA.111.2460252012;110:230-240; originally published online November 23, 2011;Circ Res.

http://circres.ahajournals.org/content/110/2/230World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org/content/suppl/2011/11/23/CIRCRESAHA.111.246025.DC1Data Supplement (unedited) at:

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


Dow

nloaded from

http://circres.ahajournals.org/content/110/2/230

http://circres.ahajournals.org/content/suppl/2011/11/23/CIRCRESAHA.111.246025.DC1

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


Supplemental Material

Methods

Vein Graft procedure. In brief, the right common carotid artery of a male

mouse was mobilized and divided. A cuff was placed on both ends of the

artery and the ends were everted over the cuff and ligated with an 8.0 silk

ligature. Vena cava from donor mice were grafted between the 2 ends of the

carotid artery by “sleeving” the ends of the vein over the artery cuff and

secured with 8.0 silk sutures1. After 4 weeks, the vein grafts were obtained.

Vessel wall thickness was measured as area of the vessel minus that of the

lumen using NIS-Elements BR 3.0 program.

Immunohistochemistry. Sections were blocked with 10% goat serum

(Vector Laboratories, Burlingame, CA) for 30 min and then incubated with

primary antibodies (FSP-1, 1:500; SMA-, 1:3000). Sections were washed in

0.5% Tween 20 in PBS (PBST) and incubated with a biotinylated secondary

antibody(Vector lab) at room temperature1. After PBST washes, tissue

sections were incubated with an Elite® ABCreagent(Vector Laboratories)

followed by a peroxidase substrate kit (Vector Laboratories) according to the

manufacturer’s protocol. The sections were counterstained by hematoxylin.

For double immunofluorescent staining of samples, fluorescent secondary

antibodies were applied to sections; DAPI was used as counter staining.

Pictures were recorded using a Nikon Eclipse 80i fluorescence microscope

(Melville, NY) with a negative control as an isotype control IgG or PBST.

Supplemental Figure Legends:

Supplemental Figure I. Characterization of FSP-1+ cells. A. Flow cytometic

analysis of FSP-1 expression in mouse tissues; cells from arteries, veins,

hearts, and bone marrow from WT or FSP-1-GFP+ mice were collected to

identify GFP+ cells. B. BM cells were isolated from FSP-1-GFP+ mice and

incubated with antibodies to cell markers CD3, B220, leukocyte CD45, and

mononuclear cells CD11b before flow cytometry analysis. C. FSP-1-GFP+ BM

cells were cultured on cover slips and stained with different antibodies: FSP-

1+ cells in red, GFP+ cells in green and nuclei shown in blue. D.

Characterization of FSP-1+ cells in vein grafts. Co-staining of vein grafts for

FSP-1 and macrophage marker F4/80.

Supplemental Figure II. Platelets accumulation in AV grafts. CD41+

platelets accumulated on the lumen surface at 5 days after creation of AV

grafts.

Supplemental Figure III. Recombinant FSP-1 stimulates ERK5

phosphorylation in VSMCs. VSMCs were treated with different doses of

FSP-1 for 30 min after starvation for 24 h, cell lysates were collected and the

phosphorylated ERK5, ERK1/2, p38, JNK and AKT were detected by Western

blots. GAPDH was used as a loading control.

Reference List

1. Cheng J, Du J. Mechanical stretch simulates proliferation of venous

smooth muscle cells through activation of the insulin-like growth factor-1

receptor. Arterioscler Thromb Vasc Biol. Aug 2007;27(8):1744-1751.

cellular biology - circulation...

Documents