The origin of post-injury neointimal cells in the rat balloon injury model

The origin of post-injury neointimal cells in the ratballoon injury model

Luis Rodriguez-Menocal1, Melissa St-Pierre1, Yuntao Wei1, Sheik Khan1, Dania Mateu1, Marian Calfa1,Amir A. Rahnemai-Azar1, Gary Striker1,2, Si M. Pham1, and Roberto I. Vazquez-Padron1*

1Division of Cardiothoracic Surgery, Department of Surgery, Vascular Biology Institute, University of Miami, Miller School ofMedicine, 1600 NW 10th Avenue, RMSB 1063, Miami, FL 33136, USA; and 2Mt. Sinai School of Medicine, 1 Gustave Levy Place,New York, NY 10029, USA

Received 17 June 2008; revised 10 September 2008; accepted 23 September 2008; online publish-ahead-of-print 25 September 2008

Time for primary review: 44 days

Aims The origin of post-injury neointimal cells is still a matter of debate. This study aims to determinethe anatomic source of neointimal cells in one of the most important animal models for the study ofvascular stenosis in response to injury, the rat balloon injury model.Methods and results Chimeric rats were generated by rescuing lethally irradiated animals with greenfluorescent protein (GFP)þ bone marrow (BM) cells from transgenic rats. Neointimal formation wasinduced in the right iliac artery of these animals using a balloon angioplasty catheter. Injured andnon-injured contra-lateral arteries were harvested at 7, 14, and 30 days post-surgery. BM-derived mono-cytes/macrophages (CD68þ GFPþ) were abundant in the media and adventitia of injured vessels har-vested at 7 days as determined by immunofluorescence and confocal microscopy. The number ofGFPþ cells declined in the vascular wall with time. Post-injury neointimal cells were mostly GFP2/smooth muscle actin (SMA)þ, which indicated that those cells originated in the recipient. Only a fewneointimal cells seemed to come from circulating progenitors (GFPþ SMAþ, 2.34%+1.61). The vascularorigin of cells in the neointima was further confirmed by transplanting injured GFP arteries into wild-type recipients. In these grafts, 94.23+0.44% of medial and 92.95+19.34% of neointimal cells wereGFPþ SMAþ. Finally, we tested the capacity of vascular smooth muscle cells (VSMC) to migratethrough the vascular wall using a novel in vivo assay. As expected, VSMC migrated and populated theneointima only in response to injury.Conclusion Our results suggest that neointimal cells in the rat balloon injury model mostly derive frompre-existing vascular cells and that only a small population of those cells come from BM-derivedprogenitors.

KEYWORDSRestenosis;

Neointima;

Balloon injury;

Rat;

Angioplasty;

Cell origin

1. Introduction

Atherosclerosis and post-angioplasty restenosis are histologi-cally characterized by the development of an enlarged neoin-tima. The main cellular component of this neointima is thesynthetic smooth muscle cell. The initial hypothesis explain-ing the pathophysiology of neointima formation proposedthat neointimal cells originated from quiescent medial vascu-lar smooth muscle cells (VSMC) that switch from a contractileto a synthetic phenotype to migrate to the intima and pro-liferate, in response to injury.1–3 This theory was supportedby the apparent increase in proliferation among the tunicamedia cells after injury4–7 and by the ex vivo developmentof neointima in arterial culture systems.8–10 It was alsosupported by the fact that neointima cells expressedVSMC markers, smooth muscle alpha actin (SMA), and

vimentin.11–13 This theory has recently been challenged bySata et al.14,15 who proposed that VSMC of injured and ather-osclerotic arteries are of bone marrow (BM) origin. On theother hand, the idea that cells within the vascular microen-vironment are more plastic is becoming more prevalent.Current data have revealed the plasticity of periadventitialfibroblasts,16,17 adventitial progenitors,18 and/or blood-borne cells19 to differentiate into neointimal cells.

Using one of the most important animal models in studyingvascular stenosis, the rat balloon injury model, we demon-strate that the participation of BM derived-cells in thehealing of injured arteries is limited to inflammatory cellsand that only a small population of neointimal cells comefrom circulating progenitors. Our results clearly show thatin this animal model neointimal cells have mostly a vesselwall origin. We also demonstrate that synthetic VSMC havethe ability to migrate across the vascular wall and populatethe neointima in response to injury.* Corresponding author. Tel: þ1 305 243 1154; fax: þ1 305 243 5636.

E-mail address: [email protected]

Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2008.For permissions please email: [email protected].

Cardiovascular Research (2009) 81, 46–53doi:10.1093/cvr/cvn265

by guest on May 29, 2013

http://cardiovascres.oxfordjournals.org/D

ownloaded from

http://cardiovascres.oxfordjournals.org/

2. Methods

2.1 Transgenic animals

Transgenic inbred Lewis rats that express green fluorescent protein(GFP) under the control of ubiquitin-C promoter20 were obtainedfrom the Rat Resource and Research Center (Columbia, MO) andbred in our laboratories. Wild-type (WT) inbred Lewis rats were pur-chased from Harlan Sprague Dawley, Inc. (Indianapolis, IN). Animalprocedures were approved by the Institutional Committee for Useand Care of Laboratory Animals at the University of Miami andconform to the Guide for the Care and Use of Laboratory Animalspublished by US National Institutes of Health (NIH publication No.85-23, revised 1996).

2.2 Generation of chimeric rats

GFP BM mononuclear cells were isolated from femurs and tibias oftransgenic and WT rats. Recipient animals were lethally irradiatedwith a single dose of 1025 cGy from a Cs-137 source (Nordion,Ontario) and immediately received one dose of 8 � 107 GFP BMcells via the jugular vein. Control chimeric rats were reconstitutedwith WT BM cells. The chimerism was assessed by flow cytometry(BD LSR System I) in peripheral blood 1 month after transplantand in the BM after sacrificing. The 7ADD staining was used togate out dead cells and debris. Flow cytometry data was analysedusing Flowjo 7.2 (Ashland, OR).

2.3 Balloon injury model

All surgeries were under isoflurane anaesthesia. Vascular injury wasperformed according to Gabeler et al.21 Briefly, an aortotomy in theabdominal aorta was made to insert a 2F Fogarty embolectomycatheter to the level of the right iliac artery. The balloon wasinflated to 1.5 atmospheres and retracted to the arteriotomy sitethree times to assure a good vascular injury. The aortic excisionwas repaired with 8.0 sutures. The abdominal cavity was closedby planes using interrupted suture pattern.

2.4 Aortic transplantation

Descending thoracic aortas were balloon-injured before harvesting.Vascular grafts included adventitia and most of the surroundingconnective tissue. GFP and WT thoracic aortas were transplantedto the abdominal aorta of WT and GFP recipients, respectively, byside to-side with interrupted anastomoses. In the control group,non-injured GFP and WT arteries were transplanted into WT recipi-ents. WT injured aortas were also transplanted into chimeric rats.No immunosuppressive or anticoagulant treatment was used inthis study.

2.5 Tissue processing

Arterial specimens were collected 7–30 days after injury or trans-plant and fixed in 4% formalin–PBS. Arteries were cut in threepieces and paraffin-embedded in the same block. Sections of 5 mmthick were mounted on Superfrost/Plus glass slides.

2.6 Morphometric analysis

The area of each vascular layer and lumen was measured on ElasticVan Gieson stained slides. These measurements were used to calcu-late the neointima to media ratio [N/M ¼ N/(M þ N)]. All morpho-metric measurements were performed on digital images using theImage Pro Plus (Media Cybernetics, Inc., Bethesda, MD) computersoftware.

2.7 Immunofluorescence and scanningconfocal microscopy

Sections were re-hydrated by serially immersing them in xylene,alcohol, and water. Antigens were retrieved by boiling slides in

10 mM citrate buffer, pH 6.0 for 20 min. Unspecific binding wasavoided using TBS-FBS 15% (Tris–Borate Saline buffer supplementedwith 15% Fetal Bovine Serum) for 20 min. Sections were incubatedovernight with goat anti-GFP polyclonal antibodies (1:100, Abcam,Cambridge, MA) and mouse anti-human SMA clone 1A4 (1:50,Dako) in TBS-FBS 10%. Bound antibodies were detected with AlexaFluor 488 donkey anti-goat and Alexa Fluor 546 goat anti-mouse(Invitrogene, Carlsbad, CA). Alternatively, mouse anti-rat CD68(1:100, AbD Serotech, Raleigh, NC) monoclonal antibody followedby goat anti-mouse Alexa Fluor 546 (Invitrogene) was used todetect infiltrated monocytes/macrophages. Sections were DAPIcounterstained (Sigma, St Louis, MO) and mounted in Cytoseal(Richard-Allan Scientific, Kalamazoo, MI). The immunofluorescentstaining protocol was validated for potential cross-reactivity ofprimary and secondary antibodies in sections of injured arteriesfrom WT rats. The specificity of anti-GFP antibodies were furthervalidated by western blot.

The stained sections were examined with a confocal scanninglaser microscope Zeiss LSM 510 META (Carl Zeiss MicroImaging,Inc., Thornwood, NY) in an inverted configuration. The system isequipped with four lasers and three confocal detectors, and datawere captured and analysed with Zeiss LSM 510 Meta and ImageBrowser software (Carl Zeiss). Dual antibody-stained images wereacquired with the use of sequential capture mode to avoid potentialfluorescence bleed-through between channels. All images were cap-tured with a plain-neofluor 40�/1.3 Oil DIC objective lens. Up to 12optical slices of 0.5 mm in depth each were recorded for everysample. The three grayscale images were electronically merged toproduce a pseudocolored image in which blue depicts cell nuclei,red depicts either SMA or CD68 immunoreactivity, and greendepicts GFP.

2.8 Vascular smooth muscle cells isolationand characterization

VSMC were isolated from thoracic aortas of GFP rats and were keptunder conventional culture conditions. Primary cell lines werecharacterized by immunostaining using anti-SMA monoclonal anti-body (1:10, Dako) and by RT–PCR. The primers and amplificationconditions for rat GAPDH (450 bp), SMA (292 bp), SM22a (179 bp),SM1/2 (299 bp), SM-MHC (85 bp), CD34 (362 bp), CD31 (225 bp),and eNos (226 bp) have been describe already.22,23

2.9 In vivo migration assays

Thirty million GFP VSMCs between passages 4–10 were embedded in1 mL of BD Matrigel Matrix (BD Biosciences, San Jose, CA). The cellsuspension (250 mL) was applied on the perivascular area of bothinjured and non-injured iliac arteries. Matrigel was allowed to soli-dify for 10 min before closing the surgical excision. VSMC migrationand engraftment in the neointima was detected by immunofluores-cence and confocal microscopy with anti-GFP and SMA antibody asdescribed earlier.

2.10 Statistical analysis

Results were expressed as mean+SEM. Initially, all morphometricand histopathological parameters were tested for normality(Shapiro–Wilk test) and homogeneity of variances (Hartley’s F maxtest). Most of variables in this study were neither normally distrib-uted nor homoscedastic, even after data transformations. There-fore, overall variability among groups was assessed with anon-parametric Kruskal–Wallis one-way ANOVA model. If the nullhypothesis was rejected, the Dunn’s multiple comparison test wasused to delineate differences among groups. All statistical pro-cedures were calculated and/or plotted with Statistic 7.0 (StatSoft,Tulsa, OK) and GraphPad Prism 5 (GraphPad Software, La Jolla, CA)computer software.

The origin of neointimal cells 47



ownloaded from


3. Results

3.1 The origin of neointimal cells in chimeric rats

First, we validated that GFP was constitutively expressed inthe vasculature of transgenic rats during arterial remodel-ling. GFP expression was tracked using immunofluorescenceand confocal microscopy. The neointimal development wassimilar between transgenic and WT rats in response toballoon injury (N/M ratio of 0.19+0.08 vs. 0.21+0.05,n ¼ 5, n.s). GFP was homogenously expressed in endothelial,smooth muscle, and neointimal cells before and after arter-ial remodelling (Figure 1A–F). No cross-reactivity wasobserved in WT sections from injured and non-injuredarteries, which were used as controls (Figure 1G–I).Anti-GFP antibodies were specific for the transgene and nocross-reactivity to rat arterial endogenous proteins wasdetected by western blot (see Supplementary materialonline, Figure S1). Therefore, GFP expression and its

detection using antibodies was a reliable, sensitive methodfor detecting BM-derived cells.

To assess the role of BM-derived cells in the remodelling ofthe arterial wall after injury, chimeras were created by res-cuing lethally irradiated WT rats with GFPþ BM cells fromtransgenic animals. Flow-cytometric analysis 1 month fol-lowing cell transplantation revealed that in chimeras95.40+8.21% of blood peripheral cells were from donorBM (see Supplementary material online, Figure S2). Thisagreed with the 92.43+0.36% of GFPþ cells found in theBM of those animals at sacrifice. Once chimeras were estab-lished, neointimal development was induced by ballooninjuring the right iliac artery. Chimeras were sacrificed at7, 14, and 30 days post-surgery and injured and uninjuredarteries were harvested for histopathological analysis.

Seven days after injury, GFPþ cells were found in the mediaand in the adventitia of the remodelled arteries (Figure 2A).These GFPþ cells were mainly monocytes/macrophages as

Figure 1 Assessment of GFP transgene expression in rat iliac arteries by immunofluorescence and confocal microscopy. Green fluorescent protein (green) andsmooth muscle actin (SMA, red) were detected in injured (A–C) and non-injured (D–F) arteries of green fluorescent protein transgenic rats. Green fluorescentprotein was co-localized with smooth muscle actin in medial and neointimal cells of transgenic rats (yellow). Injured arteries of wild-type rats were used asnegative controls for staining with anti-green fluorescent protein polyclonal antibody to demonstrate that there was no cross-reactivity with local antigens(H–I). Nuclei were counterstained with DAPI (blue). The internal elastic lamina (IEL) that limits the boundaries between the intima and the media is indicated.Magnification �760.

L. Rodriguez-Menocal et al.48



ownloaded from

http://cardiovascres.oxfordjournals.org/cgi/content/full/cvn265/DC1




determined by double immunostaining for CD68 and GFP(Figure 3). Furthermore, those GFPþ cells did not stainpositively for SMA, showing that they were not of thesmooth muscle cell lineage (Figure 2A). The numbers ofinfiltrating BM-derived GFPþ cells in the media and adventi-tia decreased as the arteries remodelled. GFPþ cellsdecreased three-fold from 7–30 days after injury(Figure 2D and E). At this time, only BM-derived cellsremained in the perivascular region of the injured arteries.These cells were negative for macrophage and VSMCmarkers (data not shown). Only a few GFPþ SMAþ neointimalcells (34.12+25.10 cells per mm2 or 2.3%, n ¼ 6) werefound. No GFPþ cells were found in the vascular wall of non-injured arteries, although they were present in the adventi-tial connective tissue. Interestingly, the arterial neointimaof the chimeras was less thickened than that in the WTrats (N/M ratio 0.1784+0.03 vs. 0.29+0.03, P ¼ 0.04,n ¼ 5) (see Supplementary material online, Figure S3).Neointimas of control chimeras (lethal irradiated WT ratsreconstituted with WT BM) were also thinner than thosedeveloped in the injured arteries of non-irradiated rats(N/M ratio 0.15+0.01 vs. 0.29+0.03, P ¼ 0.02, n ¼ 5).

The neointimal development was similar between the GFPand control chimeras (N/M ratio 0.17+0.03 vs. 0.15+0.01, P ¼ 0.43). It excludes the possibility that the trans-gene in the BM cells was playing a role in the differentoutcome of the N/M ratio between GFP chimeras and theWT non-irradiated rats.

3.2 The origin of neointimal cells in arterial grafts

We used an arterial transplant model to assess the contri-bution of vascular wall cells and recipient cells to neointimaldevelopment in rats. Arterial grafts were balloon-injuredbefore transplantation to accentuate development of theneointima after placing the graft. Donor injured aortaswere transplanted including the adventitia and most of thesurrounding connective tissues. Grafted and native arterieswere harvested 1 month after transplant. The N/M ratiosof the GFP and WT aortic grafts transplanted into WT andGFP recipients, respectively, were similar (0.38+0.03 vs.0.45+0.03, P ¼ 0.14). The graft neointima was thickerthan that found in the balloon injury model noted above(N/M ratio of 0.45+0.02 vs. 0.17+0.03, P ¼ 0.002). Most

Figure 2 Post injury neointimal cells are of local origin in the rat balloon-injury model. Injured and non- injured iliac arteries of chimeric rats were harvested at7, 14, and 30 days post-injury and stained with anti-green fluorescent protein (green) and anti-smooth muscle actin (red) antibodies (A–C). Positive cells werevisualized by immunofluorescence and confocal microscopy. Typical GFPþ SMA2 cells at 7 days are labelled with white arrowheads (A). One of the few GFPþ SMAþ

cells is labelled with a yellow arrowhead. Green fluorescent protein cells significantly diminished in the vascular wall at 14 days after surgery (B). Neointimal cellswere GFP2 SMAþ (C). The internal elastic lamina (IEL) limits the boundaries between intima and the media is indicated. Nuclei were DAPI counterstained (blue).Magnification �760. Bar graphs at the bottom represent the number of GFPþ cells in the media (D) and adventitia (E) of injured arteries of chimeric rats atdifferent time points. Bars represent the mean+SEM of n ¼ 5–6.




ownloaded from



of the medial (94.23+0.44% or 395+27 cells per mm2) andneointimal (92.95+19.67% or 1495+127 cells per mm2)cells of aortas from GFP rats transplanted into WT recipientswere GFPþ/SMAþ as determined by immunofluorescence andconfocal microscopy (Figure 4A–C). In contrast, when WTinjured aortas were transplanted into GFP rats, only2.68+1.2% (41.32+11.23 cells per mm2) of the medialcells and 4.88+2.13% (83.33+31.22 cells per mm2) ofthe neointimal cells were doubly positive (Figure 4D–F).Macrophages (CD68þ cells) were rarely found in the vascularwall of the grafts at sacrifice (data not shown). The controlgroups where non-injury GFP and WT aortic grafts weretransplanted into WT recipients developed no neointima,

ruling out the existence of GFP-induced alloreactivity thatcould interfere with our results (see Supplementary materialonline, Figure S4).

The contribution of BM-derived cells to the origin ofneointimal cells was further assessed by transplantinginjured WT aortas into GFP chimeric recipients. The neoin-tima of these grafts was significantly smaller than thoseof non-irradiated recipients (N/M ratios 0.24+0.06 vs.0.41+0.02, P , 0.01, All n ¼ 5). Only 3.20+1.13%(58.71+24.87 cells per mm2) of the neointimal and5.81+3.38 (18.47+14.35 cells per mm2) of the medialcells of those grafts were GFP and SMA positive (see Sup-plementary material online, Figure S5).

Figure 3 Inflammatory cells involved in the arterial response to balloon-injury are of BM origin. The lineage of BM-derived cell in the arterial wall was deter-mined by immunofluorescence and confocal microscopy with specific antibodies. GFPþ cells (green) in the arterial wall of injured arteries at 7 days post-injury(A). Most of those cells were positive for the macrophage marker CD68 (red, B). Typical GFPþ CD68þ cells are labelled with arrowheads (C). Nuclei were counter-stained with DAPI (blue). Magnification �760.

Figure 4 Neointimal cells of aortic isografts are of donor origin. Green fluorescent protein and wild-type injured aortas were transplanted in the abdominalaorta of wild-type (A–C) and green fluorescent protein (D–E) recipients, respectively. GFPþ smooth muscle cells were detected by immunofluorescence and con-focal microscopy using anti-green fluorescent protein (green) and anti-smooth muscle actin antibodies (red). Nuclei were DAPI counterstained (blue). Theinternal elastic lamina (IEL) limits the boundaries between the intima and the media is indicated. Magnification �760.




ownloaded from






3.3 Migration and engraftment of synthetic vascularsmooth muscle cells to the neointima

The above results strongly suggest that neointimal cells orig-inate in the local vascular wall. Therefore, the migration ofsmooth muscle cells into the intima might be a necessaryprerequisite for the formation of a neointima.24 We devel-oped a model to determine if VSMC migrate across the vas-cular wall in response to injury. This consisted of seedingMatrigel-embedded GFPþ VSMC or BM cells outside ofinjured and control arteries. Control rats received Matrigelwhich contained no cells. Prior to seeding, we had foundthat 99.23% of the VSMC used for seeding were GFP positiveby FACS and fluorescence microscopy and expressed theVSMC markers, SMA, SM22a, SM1/2, and SM-MHC by IHC orRT–PCR (Figure 5A–C). These cells were free of endothelialcontamination as shown by RT–PCR (Figure 5C).

The N/M ratio was larger in injured arteries that receivedembedded VSMC than in those treated with BM cells (0.28+0.04 vs. 0.17+0.02, P ¼ 0.08) or matrigel alone (0.15+0.02, P ¼ 0.02). The matrigel-embedded cells did notinduce visible vascular lesions in the control, non-injuredarteries. The number of VSMC (GFPþ and SMAþ) whichmigrated and engrafted into the neointima in response toinjury was 10 times greater in arteries that receivedMatrigel-embedded VSMC cells than in those in which theMatrigel contained BM cells (21.79+5.15 vs. 1.58+0.45%,

P , 0.01, Figure 5D–G). No GFP cells were present in themedia of rats which received Matrigel containing BM cells(data not shown).

4. Discussion

The initial studies of vascular restenosis suggested that thetunica media was the main source of neointimal cellsbased on the observation that VSMC were rapidly labelledwith thymidine after injury.4,7 The current study supportsthis initial hypothesis and utilizes new experimentalapproaches. Our data indicate that most of post-injuryneointimal VSMC originate from local vascular cells andnot from circulating progenitors in rats. The participationof the BM in arterial remodelling after injury appears tobe limited to the inflammatory stage of the vascular remo-delling process. Only a minor population of VSMC in theneointima had a BM origin. Our results also showed thatVSMCs migrate across the vascular wall in response toinjury and participate in the formation of the neointima.

Herein, we demonstrated by transplanting injuredarteries between GFP and transgenic rats and by injuringthe arteries of GFP BM reconstituted chimeras that themain anatomic source of cells in the rat post-injury neoin-tima is the vascular wall. The outcomes from these exper-iments showed that the neointima of balloon-injured GFP

Figure 5 Vascular smooth muscle cells migrated through the media to the neointima in response to vascular injury. Vascular smooth muscle cells constitutivelyexpressed the green fluorescent protein transgene (A) and smooth muscle cell actin (B) as detected by immunohistochemistry. These cells expressed the smoothmuscle cell markers a-SMA, SM22a, SM1/2, and SM-MHC by RT–PCR, but not the endothelial cell markers, CD34, CD31, and eNos (C). Matrigel-embedded GFPþ

vascular smooth muscle cells were seeded around injured arteries of wild-type rats. The GFPþ SMAþ cells (arrowheads) were detected in the neointima 3 weeksafter balloon injury by immunofluorescence and confocal microscopy (D–F). Injured artery treated with BM-embedded cells (H–I). No GFPþ cells were found inthe neointima of those animals. The internal elastic lamina (IEL) limits the boundaries between intima and the media is indicated. Magnification �380.




ownloaded from


arterial grafts transplanted into WT recipients were mainlypopulated with GFPþ/SMAþ cells. Furthermore, only fewGFPþ/SMAþ cells were found in the neointimas of chimericrats that received GFP BM after lethal irradiation. Theseexperimental evidences suggest that neointimal cells in thisanimal model were either from local tunica media cells orfrom adventitial pericytes/smooth muscle progenitors.

These findings contribute substantially to the very confusingliterature existing in the area of the neointimal cell origin. Ourresults are in agreement with a series of recent studies whichshow little or no contribution of BM-derived cells to SMClineages in a variety of models including arteriogenesis,25

atherosclerosis,26 tumour angiogenesis,27,28 and graft vascu-lopathy.18,29 For instance, the fact that post-injury neointimalcells are of local origin concurs with Bentzon et al.26 studieswhich showed that VSMC in atherosclerotic plaques camefrom the local vessel wall and not from circulating progenitorsas claimed by Sata et al.14,15 In addition, neointimal cells inthe mouse vein-to-artery graft models were found to originatefrom adventitial progenitors and not from the BM.18,30

However, our results are in disagreement with other studiessuggesting that a majority of cells in the neointima comesfrom BM-derived circulating progenitors that engraft the vas-cular wall in response to atherogenic stimuli.28,31 Althoughthese results have been partially reproduced, they remaincontroversial.32,33

There are substantial biological and methodological differ-ences between our experiments and those that support a BMorigin of neointimal cells that may account for the contrastingoutcomes. Our studies were performed in rats and vascularinjury was induced using an embolectomy balloon. Reportsfinding great contribution of marrow cells to the neointimalcell population were performed with the wire-injury modelin the mouse. It is well-accepted that species have differentmechanisms of vascular remodelling in response to injury.34

On the other hand, it is also known that the method of inducingarterial damage can affect the recruitment of BM-derivedcells to the site of injury. For example, Tanaka et al.35 foundthat wire injury, but not blood cessation or cuff placement,mobilized BM cells to the neointima. Furthermore, a profounddifference exists between our study and those that claimedthat most of neointimal cells come from BM with respect tothe histological methods utilized to track cells into the vascu-lar wall. Data that support a BM origin of neointimal cells werederived from the direct observation of GFPþ cells in unfixedtissue. This histopathological procedure does not provideenough resolution to discriminate between GFPþ positiveand the negative cells in the arterial background.36–38 Theuse of unfixed tissues may also lead to diffusion of the tracermarker from sectioned cells.39 We detected GFPþ cells infixed tissues by immunofluorescence microscopy using anti-bodies which do not cross-react with self antigens of injuredand non-injured arteries from WT animals. This allowed thedifferentiation of GFPþ cells from the neighbouring negativecells. The GFPþ cells had a defined morphology and thetarget protein was always intracellularly located.

The outcomes of this study also differ from human data.Using sex mismatched specimens, it has been shown thatthere are smooth muscle cells of donor origin in the plaquesof human coronary atherosclerosis.40 The discrepanciesbetween these findings and our results using theballoon-injury models could be explained by the differentpathophysiological mechanisms underlying atherosclerosis

and post-injury stenosis. Even though these vascular occlusivediseases share certain mechanistic elements including VSMCproliferation, they have different dynamics and pathologicalcharacteristics. In addition, the specimens were collectedfrom patients exposed to variables like immunosuppression,chemotherapy, and graft-vs.-host disease that were notincluded in the current study. The differences between thehuman data and our studies could also be explained throughthe known limitation of animal models used for vascular steno-sis/restenosis. Those major limitations in our case are theabsence of a primary lesion (atheroma) in the target vesseland dyslipidaemia that modifies VSMC biology.

The source of the neointimal cells in the rat balloon injurymodel appeared to be from a local site. However, it is notyet proved that neointimal cells are derived from pre-existing contractile VSMC. The existence of stem cellshave been recently documented in the vascular wall,41 peri-vascular area,42 and adventitia.18 The fact that neointimalcells would have an origin in local progenitor cells wouldexplain the arterial repopulation after the massive apoptosisthat occurs after injury.43 It is possible that neointimal cellshave a clonal origin from contractile VSMC precursors thatdevelop a synthetic phenotype after injury.44 Furtherresearch is necessary to find out which of the cell types inthe arterial wall serve as the source of the neointima.

The fact that neointimal vascular smooth muscle cells donot appear to originate from the BM does not imply thatBM-derived cells have no role in the development of theneointima. BM-derived inflammatory cells, mainly macro-phages, are abundant in the arterial wall early afterinjury. The number of those cells decreases with time.Thus, cells derived from the BM may control the remodellingand initiation of healing in the injured blood vessel andthese cells may also provide signals for the mobilizationand recruitment of the VSMC that repopulate the injuredarteries. This hypothesis is supported by the observationthat inhibition of macrophage infiltration after injury signifi-cantly diminishes neointimal formation.45

Finally, we assessed whether VSMC or BM cells can migrateacross the arterial wall in response to injury. We found thatVSMC from primary cultures, but not BM mononuclear cells,migrate from the adventitial region into the intima inresponse to injury. The fact that BM cells do not appear tomigrate through the arterial wall has previously beenshown in the vein-to-graft transplant model.18 It is alsonotable that injured arteries treated with embedded VSMCdeveloped a thicker neointima than those treated with BMor matrigel alone. This suggests that actively proliferatingVSMC derived from the outer layers of the artery could con-tribute to the population of neointimal cells.

In conclusion, our results demonstrate that the majorityof neointimal cells are of local origin in the balloon-injurymodel of the rat. We also found that the function ofBM-derived cells in the pathological vascular remodelling isto serve as a source of inflammatory cells. Therefore,further experiments that dissect molecular mechanisms bywhich local progenitors are recruited and differentiate atthe site of injury are warranted.

Supplementary material

Supplementary material is available at CardiovascularResearch online.




ownloaded from




Conflict of interest: none declared.

Funding

This work was supported by awards from the American Heart Associ-ation (Scientist Development Award 0535167B) and Stanley GlaserFoundation.

References1. Schwartz SM, Campbell GR, Campbell JH. Replication of smooth muscle

cells in vascular disease. Circ Res 1986;58:427–444.2. Clowes AW, Reidy MA, Clowes MM. Mechanisms of stenosis after arterial

injury. Lab Invest 1983;49:208–215.3. Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s.

Nature 1993;362:801–809.4. Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after

arterial injury. I. Smooth muscle growth in the absence of endothelium.Lab Invest 1983;49:327–333.

5. Clowes AW, Schwartz SM. Significance of quiescent smooth musclemigration in the injured rat carotid artery. Circ Res 1985;56:139–145.

6. Velican D, Velican C. Comparative study on age-related changes andatherosclerotic involvement of the coronary arteries of male andfemale subjects up to 40 years of age. Atherosclerosis 1981;38:39–50.

7. Stemerman MB, Weinstein R, Rowe JW, Maciag T, Fuhro R, Gardner R. Vas-cular smooth muscle cell growth kinetics in vivo in aged rats. PNAS 1982;79:3863–3866.

8. Soyombo AA, Angelini GD, Bryan AJ, Jasani B, Newby AC. Intimal prolifer-ation in an organ culture of human saphenous vein. Am J Pathol 1990;137:1401–1410.

9. Jackman RW. Persistence of axial orientation cues in regenerating intimaof cultured aortic explants. Nature 1982;296:80–83.

10. Jackman RW, Anderson SK, Sheridan JD. The aortic intima in organculture. Response to culture conditions and partial endothelial denuda-tion. Am J Pathol 1988;133:241–251.

11. Owens GK. Regulation of differentiation of vascular smooth muscle cells.Physiol Rev 1995;75:487–517.

12. Gabbiani G, Rungger-Brandle E, de Chastonay C, Franke W. Vimentin-containing smooth muscle cells in aortic intimal thickening after endo-thelial injury. Lab Invest 1982;47:265–269.

13. Barja F, Coughlin C, Belin D, Gabbiani G. Actin isoform synthesis andmRNA levels in quiescent and proliferating rat aortic smooth musclecells in vivo and in vitro. Lab Invest 1986;55:226–233.

14. Sata M, Saiura A, Kunisato A, Tojo A, Okada S, Tokuhisa T et al. Hemato-poietic stem cells differentiate into vascular cells that participate in thepathogenesis of atherosclerosis. Nat Med 2002;8:403–409.

15. Sata M, Fukuda D, Tanaka K, Kaneda Y, Yashiro H, Shirakawa I. The role ofcirculating precursors in vascular repair and lesion formation. J Cell MolMed 2005;9:557–568.

16. Sartore S, Chiavegato A, Faggin E, Franch R, Puato M, Ausoni S et al. Con-tribution of adventitial fibroblasts to neointima formation and vascularremodeling: from innocent bystander to active participant. Circ Res2001;89:1111–1121.

17. Li G, Chen S-J, Oparil S, Chen Y-F, Thompson JA. Direct in vivo evidencedemonstrating neointimal migration of adventitial fibroblasts afterballoon injury of rat carotid arteries. Circulation 2000;101:1362–1365.

18. Hu Y, Davison F, Ludewig B, Erdel M, Mayr M, Url M et al. Smooth musclecells in transplant atherosclerotic lesions are originated from recipients,but not bone marrow progenitor cells. Circulation 2002;106:1834–1839.

19. Simper D, Stalboerger PG, Panetta CJ, Wang S, Caplice NM. Smooth muscleprogenitor cells in human blood. Circulation 2002;106:1199–1204.

20. Lois C, Hong EJ, Pease S, Brown EJ, Baltimore D. Germline transmissionand tissue-specific expression of transgenes delivered by lentiviralvectors. Science 2002;295:868–872.

21. Gabeler EE, van Hillegersberg R, Statius van Eps RG, Sluiter W,Gussenhoven EJ, Mulder P et al. A comparison of balloon injury modelsof endovascular lesions in rat arteries. BMC Cardiovasc Disord 2002;2:16.

22. Howson KM, Aplin AC, Gelati M, Alessandri G, Parati EA, Nicosia RF. Thepostnatal rat aorta contains pericyte progenitor cells that form spheroi-dal colonies in suspension culture. Am J Physiol Cell Physiol 2005;289:C1396–C1407.

23. Yoshida T, Kawai-Kowase K, Owens GK. Forced expression of myocardin isnot sufficient for induction of smooth muscle differentiation in multipo-tential embryonic cells. Arterioscler Thromb Vasc Biol 2004;24:1596–1601.

24. Gerthoffer WT. Mechanisms of vascular smooth muscle cell migration.Circ Res 2007;100:607–621.

25. O’Neill TJt, Wamhoff BR, Owens GK, Skalak TC. Mobilization of bonemarrow-derived cells enhances the angiogenic response to hypoxiawithout transdifferentiation into endothelial cells. Circ Res 2005;97:1027–1035.

26. Bentzon JF, Weile C, Sondergaard CS, Hindkjaer J, Kassem M, Falk E.Smooth muscle cells in atherosclerosis originate from the local vesselwall and not circulating progenitor cells in ApoE knockout mice. Arterios-cler Thromb Vasc Biol 2006;26:2696–2702.

27. De Palma M, Venneri MA, Roca C, Naldini L. Targeting exogenous genes totumor angiogenesis by transplantation of genetically modified hemato-poietic stem cells. Nat Med 2003;9:789–795.

28. Roberts N, Jahangiri M, Xu Q. Progenitor cells in vascular disease. J CellMol Med 2005;9:583–591.

29. Li J, Han X, Jiang J, Zhong R, Williams GM, Pickering JG et al. Vascularsmooth muscle cells of recipient origin mediate intimal expansion afteraortic allotransplantation in mice. Am J Pathol 2001;158:1943–1947.

30. Hu Y, Mayr M, Metzler B, Erdel M, Davison F, Xu Q. Both donor and recipi-ent origins of smooth muscle cells in vein graft atherosclerotic lesions.Circ Res 2002;91:e13–e20.

31. Sata M, Tanaka K, Nagai R, Hu Y, Dietrich H, Davison F et al. Origin ofsmooth muscle progenitor cells: different conclusions from differentmodels * response. Circulation 2003;107:e106–e107.

32. Zernecke A, Schober A, Bot I, von Hundelshausen P, Liehn EA, Mopps Bet al. SDF-1falphag/CXCR4 Axis Is Instrumental in Neointimal Hyperplasiaand Recruitment of Smooth Muscle Progenitor Cells. Circ Res 2005;96:784–791.

33. Matsumura G, Miyagawa-Tomita S, Shin’oka T, Ikada Y, Kurosawa H. Firstevidence that bone marrow cells contribute to the construction of tissue-engineered vascular autografts in vivo. Circulation 2003;108:1729–1734.

34. Cooley BC. Murine model of neointimal formation and stenosis in veingrafts. Arterioscler Thromb Vasc Biol 2004;24:1180–1185.

35. Tanaka K, Sata M, Hirata Y, Nagai R. Diverse contribution of bone marrowcells to neointimal hyperplasia after mechanical vascular injuries. CircRes 2003;93:783–790.

36. Sartori M, Weilbaecher D, Valderrama GL, Kubodera S, Chin RC, Berry MJet al. Laser-induced autofluorescence of human arteries. Circ Res 1988;63:1053–1059.

37. Choserot C, Pery E, Goebel JC, Dumas D, Didelon J, Stoltz JF et al. Experi-mental comparison between autofluorescence spectra of constrainedfresh and cryopreserved arteries. Clin Hemorheol Microcirc 2005;33:235–242.

38. Kingsley K, Carroll K, Huff JL, Plopper GE. Photobleaching of arterialautofluorescence for immunofluorescence applications. Biotechniques2001;30:794–797.

39. Jockusch H, Voigt S, Eberhard D. Localization of GFP in frozen sectionsfrom unfixed mouse tissues: immobilization of a highly solublemarker protein by formaldehyde vapor. J Histochem Cytochem 2003;51:401–404.

40. Caplice NM, Bunch TJ, Stalboerger PG, Wang S, Simper D, Miller DV et al.Smooth muscle cells in human coronary atherosclerosis can originatefrom cells administered at marrow transplantation. PNAS 2003;100:4754–4759.

41. Sainz J, Al Haj Zen A, Caligiuri G, Demerens C, Urbain D, Lemitre M et al.Isolation of ‘side population’ progenitor cells from healthy arteries ofadult mice. Arterioscler Thromb Vasc Biol 2006;26:281–286.

42. Howson KM, Aplin AC, Gelati M, Alessandri G, Parati EA, Nicosia RF. Thepostnatal rat aorta contains pericyte progenitor cells that form spheroi-dal colonies in suspension culture. Am J Physiol Cell Physiol 2005;289:C1396–C1407.

43. Perlman H, Maillard L, Krasinski K, Walsh K. Evidence for the rapid onsetof apoptosis in medial smooth muscle cells after balloon injury. Circula-tion 1997;95:981–987.

44. Schwartz SM, Murry CE. Proliferation and the monoclonal origins of ather-osclerotic lesions. Annu Rev Med 1998;49:437–460.

45. Danenberg HD, Fishbein I, Gao J, Monkkonen J, Reich R, Gati I et al.Macrophage depletion by clodronate-containing liposomes reducesneointimal formation after balloon injury in rats and rabbits. Circulation2002;106:599–605.




ownloaded from


Documents

The origin of post-injury neointimal cells in the rat balloon injury model