Ceramide-Coated Balloon Catheters Limit NeointimalHyperplasia After Stretch Injury in Carotid ArteriesRoger Charles, Lakshman Sandirasegarane, Jong Yun, Nicole Bourbon, Ronald Wilson,
Raymond P. Rothstein, Steven W. Levison, Mark Kester
Abstract—Neointimal hyperplasia at the site of surgical intervention is a common and deleterious complication of surgeryfor cardiovascular diseases. We hypothesized that direct delivery of a cell-permeable growth-arresting lipid via theballoon tip of an embolectomy catheter would limit neointimal hyperplasia after stretch injury. We have previouslydemonstrated that sphingolipid-derived ceramide arrested the growth of smooth muscle cell pericytes in vitro. Here, weshow that ceramide-coated balloon catheters significantly reduced neointimal hyperplasia induced by balloonangioplasty in rabbit carotid arteries in vivo. This ceramide treatment decreased the number of vascular smooth musclecells entering the cell cycle without inducing apoptosis. In situ autoradiographic studies demonstrated that inflating theballoon catheter forced cell-permeable ceramide into the intimal and medial layers of the artery. Intercalation ofceramide into vascular smooth muscle cells correlated with rapid inhibition of trauma-associated phosphorylation ofextracellular signal–regulated kinase and protein kinase B. These studies demonstrate the utility of cell-permeableceramide as a novel therapy for reducing neointimal hyperplasia after balloon angioplasty.(Circ Res. 2000;87:282-288.)
Restenosis still persists as a major complication in themaintenance of vessel patency after percutaneous
transluminal coronary angioplasty (PTCA). Restenosis is aconsequence of multiple factors, including vessel recoil,negative vascular remodeling, residual plaque burden, andneointimal hyperplasia.1,2 Neointimal hyperplasia reflectsthe migration and proliferation of vascular smooth muscle(VSM) cells with subsequent deposition of extracellularmatrix components at the site of injury.1,3 Considerableevidence indicates that, in restenosis, growth factors stim-ulate the VSM cells to proliferate, resulting in a thickeningof the tunica intima.4 Nearly 40% of all patients developsignificant luminal narrowing within 6 months after angio-plasty procedures.1 Consequently, despite the initial ther-apeutic benefits of angioplasty, within a few months aftersurgery, blood flow through the affected vessels can againbecome compromised. Conventional therapies, which in-clude angiotensin-converting enzyme inhibitors, antico-agulants, and statins, are ineffective in preventing orreducing neointimal hyperplasia after stretch injury.1,5
Endovascular radiation therapy has shown some success inboth animal and human trials, yet the long-term deleteriouseffects of this therapy on the artery have not beenadequately evaluated.1,6 We propose that direct delivery, tothe site of vascular injury, of a cell-permeable lipid thatblocks growth factor–mediated signaling cascades has the
potential to reduce neointimal hyperplasia without sys-temic complications.
Sphingolipids are ubiquitous membrane lipids that serve assubstrates for the formation of second messengers.7 Cer-amide, a second messenger derived from cytokine receptor–activated sphingomyelin catabolism, stimulates differentia-tion, inhibits proliferation, and has been associated withapoptosis.7 We previously demonstrated that increasing en-dogenous ceramide concentration by inhibition of ceramidecatabolism induces growth arrest in smooth muscle peri-cytes.8 Moreover, cell-permeable ceramide (C6-ceramide)mimics the effect of interleukin-1 to inhibit both tyrosinekinase receptor–linked and G protein receptor–linked mito-genesis in A7r5 aortic smooth muscle cells and rat glomerularmesangial cells.8–10 In vitro, ceramide inhibits VSM cellproliferation by differentially regulating members of themitogen-activated protein kinase (MAPK) cascade. Ceramidestimulates c-jun N-terminal kinases (JNKs), whereas it sup-presses extracellular signal–regulated kinases (ERKs).10,11 Inaddition, ceramide could regulate mitogenesis by inhibitingcell survival kinases, such as protein kinase B (PKB).12 Theexperiments described here were designed to determinewhether a cell-permeable ceramide could diminish VSM cellproliferation in vivo and, if so, to characterize the mecha-nisms responsible for this effect.
Received February 29, 2000; revision received June 23, 2000; accepted June 23, 2000.From the Departments of Pharmacology (R.C., L.S., J.Y., N.B., M.K.), Comparative Medicine (R.W.), and Neuroscience and Anatomy (R.P.R.,
S.W.S.), Pennsylvania State University, Milton S. Hershey Medical Center, Hershey, Pa.Correspondence to Dr Mark Kester, Department of Pharmacology, Pennsylvania State University, College of Medicine, 500 University Dr, Hershey,
Materials and MethodsAnimal ModelWe chose the carotid artery of the New Zealand white rabbit as amodel system for neointimal hyperplasia after stretch injury. Therabbit carotid artery responds to stretch injury with marked, rapid,and reproducible neointimal hyperplasia.13–15In addition, this modelserves as an excellent source for explanted and cultured smoothmuscle cells so that in vivo studies can be confirmed with in vitroexperiments.13
The details of the balloon angioplasty procedure are described inthe online Materials and Methods (available at http://www.circresa-ha.org). Briefly, the left carotid artery was exposed and a smallincision was made in the vessel 20 mm above the bifurcation of theinternal and external carotid. A 3F Intimax arterial embolectomycatheter from Applied Medical Vascular Division was insertedretrograde into the common carotid artery 70 mm below the incision.The balloon was inflated to 4 mm, which distended the vessel 3-fold.The inflated balloon was withdrawn 50 mm, deflated, rotated 120°,and inserted back to the original position in the common carotid.This procedure was repeated 3 times.
Lipid TherapeuticsThe lipid gels were applied to the catheters by dipping the balloons 10times into a DMSO/ethanol, 1:1 vol/vol solution containing 5 mmol/LC6-ceramide (D-erythro-N-hexanoylsphingosine) or dihydro-C6-ceramide (D-erythro-N-hexanoylsphinganine) (Biomol), interspersedwith drying under nitrogen. The coated latex balloon catheter, insertedinto 50 mmol/L ceramide solution, remained intact after 50 inflations asevidenced by enveloping the balloon with a loading dye. In situautoradiography with radiolabeled C6-ceramide was used to documentthe pharmacokinetics of ceramide transfer to carotid arteries, and detailsof these methods can be found in the online Materials and Methods16–19
(available at http://www.circresaha.org).
ImmunohistochemistryAn adapted “ABC” (avidin-biotin-peroxidase complex) procedurewas used to stain fora smooth muscle cell actin and proliferating cellnuclear antigen (PCNA) 2 weeks after angioplasty.18,20
Apoptosis MeasurementWe initially assessed apoptosis of primary VSM cells isolated fromrabbit carotid arteries by fluorescence-activated cell sorting afterpropidium iodide staining.15 To confirm these measurements, wealso assessed apoptosis in situ by quantifying the percentage ofpyknotic propidium iodide– or hematoxylin-stained nuclei per arte-rial section as well as by in situ end labeling of nicked DNA.21
An expanded Materials and Methods section is available online athttp://www.circresaha.org.
ResultsExperiments were designed to determine whether ceramide-coated embolectomy catheters would diminish neointimalhyperplasia after balloon-induced stretch injury. Initial stud-ies assessed the extent of neointimal hyperplasia in rabbitcarotid arteries after balloon angioplasty as a function of time.Animals were euthanized at 15 and 60 minutes as well as at1, 2, 4, and 6 weeks after balloon injury. Marked neointimalhyperplasia was observed as early as 1 week and peaked at 4weeks (Figure 1A). Neointimal hyperplasia was not observedin damaged arteries 15 and 60 minutes after angioplasty.Sham-treated carotid arteries showed no signs of neointimalhyperplasia at any time point. On the basis of these results, wechose to investigate the effects of ceramide on dynamic VSMgrowth 2 weeks after balloon injury.
Figures 1B through 1E show hematoxylin and eosin(H&E)–stained cryostat sections of rabbit carotid arteries 2
weeks after balloon injury. In addition to the sham-treatedcontrol artery (B), the 3 treatment groups included a vehicle-treated balloon (C), a C6-ceramide–coated balloon (D), and adihydro-C6-ceramide–coated balloon (E). Quite strikingly,C6-ceramide treatment significantly reduced the neointimalhyperplasia induced by balloon angioplasty. A quantitativeanalysis revealed that balloon catheters coated with C6-ceramide diminished the number of neointimal concentriccell layers by'50% (Figure 2A). This corresponds to areduction of neointimal thickness from 0.2160.06 to0.1260.09 mm. As a control for the lipid vehicle, noncoatedballoon embolectomy catheters always induced the samedegree of neointimal hyperplasia as vehicle-coated balloons.In accordance with Komukai et al15 and Negoro et al,22 wealso quantified neointimal stenosis as a ratio of neointimal/medial cross-sectional areas and showed a 92% reduction ofstenosis with ceramide treatment (Figure 2B). Stretch injuryinduced a slight but significant increase in medial hypertro-phy that was not reduced by ceramide treatment (Figure 2C).Dihydro-C6-ceramide, an inactive analogue of C6-ceramide,did not significantly reduce neointimal hyperplasia, nor did itreduce medial hypertrophy after balloon injury (Figures 2Athrough 2C). Thus, the selective reduction in neointimalhyperplasia after stretch injury requires bioactive ceramide,and this effect cannot be mimicked using structurally similar
Figure 1. C6-ceramide but not dihydro-C6-ceramide limited neo-intimal hyperplasia after balloon angioplasty in rabbit carotidarteries. A, Time course of neointimal hyperplasia after angio-plasty. Each time point depicts 2 to 6 arteries. B through E,Representative H&E-stained sections from excised arteries 2weeks after angioplasty. These photomicrographs are represen-tative of 5 to 8 arteries. B, Sham-treated control artery. C,Artery treated with a DMSO/ethanol (1:1, vol/vol)–coated bal-loon. D, Artery treated with a C6-ceramide–coated balloon. E,Artery treated with dihydro-C6-ceramide, a biologically inactiveform of ceramide. Bar5200 mm.
Charles et al Ceramide Limits Neointimal Hyperplasia 283
by guest on December 3, 2015http://circres.ahajournals.org/Downloaded from
but inactive lipids. It can be inferred that the effects ofceramide are due to biochemical actions and not to lipophilicproperties.
We next assessed the pharmacokinetics of ceramide trans-fer and delivery from the balloon catheter to the damagedartery. Using[3H]C6-ceramide as a tracer, we calculated that'70610 nmol of C6-ceramide was applied to the balloon as
a gel from a solution of 5mmol of C6-ceramide. Figure 3Ashows that, after insertion and inflation,'1262 nmol re-mained on the balloon. This translates to roughly 58 nmol ofC6-ceramide being transferred from the balloon catheterduring the angioplasty procedure. To test whether inflation ofthe balloon within the carotid artery was essential for optimaltransfer of the ceramide, we repeated the surgical procedureusing noninflated balloons. The recovered ceramide mass onthe inserted but noninflated balloon was 1463 nmol. We nextasked whether the difference in ceramide mass between theinflated and noninflated balloons ('2 nmol) corresponded tothe calculated mass of [3H]C6-ceramide isolated from dam-aged carotid arteries. Rabbit carotid arteries treated withradiolabeled lipid were homogenized, and lipid products wereseparated by thin-layer chromatography (TLC) (Figure 3A,inset). The mass of intact ceramide isolated 15 minutes afterangioplasty was 2.760.4 nmol for inflated balloon treatmentsand 0.760.2 nmol for noninflated balloon treatments. Theamount of ceramide recovered from excised tissues did notdiffer significantly from the amount of ceramide transferredto the tissue as a consequence of balloon inflation. As thetransferred ceramide was initially delivered to 0.0365 cm3 ofcarotid artery luminal volume, the effective concentration ofceramide at the site of balloon injury was estimated to be1.5 mmol/L. Thus, we suggest that an effective and repro-ducible dose of ceramide can be delivered to the damagedartery as a consequence of the balloon inflation.
We next used in situ autoradiography to document arterialpenetrance for [3H]C6-ceramide transferred from the ballooncatheter after angioplasty (Figures 3B through 3D). Com-pared with unlabeled arteries (panel B), [3H]C6-ceramide wasobserved throughout the medial layers of the artery 15minutes after angioplasty (panel C). This increase in pixelintensity reflects an increase in intact ceramide, as at this timepoint '8964% of the radiolabel comigrates with authenticC6-ceramide standards. Pixel intensity was more intense ininflated (panel C) versus noninflated (panel D) arteries.Expressed as pixel density per square millimeter for 10randomly selected blocks with background values subtracted,medial staining was increased 4.760.2-fold for ceramide-coated inflated versus noninflated balloons. Again, this sup-ports the finding that balloon inflation leads to maximaldelivery and penetrance. Thus, a lipid-coated balloon deliversa therapeutic dose of ceramide to tissues underlying the site ofvascular stretch injury. These studies also suggest that ashort-term application of cell-permeable ceramide is suffi-cient to completely penetrate injured arteries and to reduceintimal proliferation despite an inflammatory milieu.
We next assessed degradation of the rapidly intercalatedradiolabeled ceramide by TLC. For the 15-minute postangio-plasty time point, 8964% of the TLC-separated lipid comi-grated with authentic C6-ceramide standards. This corre-sponded to a recovered mass of 2.760.4 nmol of ceramide.At 60 minutes after angioplasty, 1.360.6 nmol of ceramidewas recovered. Thus,'50% radiolabel can still be recoveredas intact ceramide in 1 hour. This decrease in ceramide masscorresponded to an increase in TLC-separated gangliosidesand cerebrosides but not sphingosines (data not shown).
Figure 2. C6-ceramide but not dihydro-C6-ceramide reducedthe thickness and area of the stenotic lesion. A, Thickness ofthe neointimal lesion in 5 to 8 arteries 2 weeks after angioplasty.Ceramide-treated arteries had fewer neointimal concentric celllayers as assessed by counting stained nuclei in 8 randomlyselected locations for each artery. B, Extent of stenosis as aratio of neointimal/medial surface areas. C6-ceramide but notdihydro-C6-ceramide reduces the stenotic lesion. C, Thicknessof the medial layer in the same group of experimental animals.Even though balloon angioplasty significantly induced medialhypertrophy, C6-ceramide- or dihydro-C6-ceramide–coated bal-loon treatments were not significantly different from vehicle-coated balloon treatments. Data are mean6SEM. Stars indicateP,0.01, Kruskal-Wallis 1-way ANOVA, and Dunn method formultiple comparisons.
284 Circulation Research August 18, 2000
by guest on December 3, 2015http://circres.ahajournals.org/Downloaded from
To prevent thrombus formation, patients routinely receiveanticoagulants before PTCA. Thus, the consequences ofanticoagulation therapy on the effectiveness of ceramidetherapy were investigated. Neither ceramide- nor vehicle-treated balloon angioplasty induced thrombus formation.Lovenox, a low molecular weight heparin, administeredsubcutaneously (2.5 mg/kg) for 7 days after surgery, did notby itself diminish neointimal hyperplasia.23 Nor did it aug-ment ceramide-induced inhibition of neointimal hyperplasia(data not shown). These findings suggest that ceramidetreatment is equally effective in both anticoagulated anduntreated rabbits.
We next examined the effects of ceramide treatment onVSM cell growth in vivo 2 weeks after angioplasty. Immu-nohistochemical techniques were used to identify VSM cellsusing smooth muscle cell–specifica actin antibody (Figures4A and 4B) and cell growth using PCNA antibody (Figures4C through 4F). The positive staining with the actin antibodyindicates that VSM cells or myofibroblasts were a majorcomponent of balloon injury–induced neointimal formation(panel B). Also, this photomicrograph shows dramatic bal-loon angioplasty–induced ruffling and dispersion of VSMcells in the medial layer. PCNA is synthesized in early G1 andS phases of the cell cycle and thus can be used as a marker for
Figure 3. Pharmacokinetics of delivery of C6-ceramide to thesite of VSM injury. A, Decrease in the amount of radioactiveC6-ceramide associated with the balloon catheter after insertionwith or without inflation. Inset, Mass of radiolabeled ceramiderecovered by TLC from carotid arteries subjected to ballooninsertion with or without inflation; 4 balloons or arteries for eachcondition. Data are mean6SEM; stars indicate P,0.01. Bthrough D, Representative autoradiograms for vehicle-coated (B)and ceramide-coated inflated (C) and noninflated (D) catheter–treated carotid arteries, 15 minutes after angioplasty.
Figure 4. Ceramide-treated catheters reduced PCNA expres-sion in VSM cells 2 weeks after angioplasty. A and B, a smoothmuscle actin staining for control and balloon-injured arteries,respectively. C through F, PCNA staining for control, balloon-injured, ceramide-coated balloon-injured, and dihydro-C6-ceramide-coated balloon-injured carotid arteries, respectively.Bars5200 mm. Arrows point to PCNA-positive cells that can beseparated from nonspecific stained cells by the intensity ofstaining as well as by the increased size of proliferating nuclei.These immunohistochemical micrographs are representative of4 to 8 separate arteries. LUM indicates lumen; MED, media;NEO, neointima; and ADV, adventitia.
Charles et al Ceramide Limits Neointimal Hyperplasia 285
by guest on December 3, 2015http://circres.ahajournals.org/Downloaded from
cell proliferation. In Figures 4C through 4F, representativephotomicrographs depicting PCNA-positive staining areshown for control, balloon-injured, ceramide-treated, anddihydro-ceramide–treated carotid arteries, respectively. Thepercentage of PCNA-positive cells in balloon-injured arteries(2.8% 60.1%) was dramatically increased compared withcontrol vessels (0.260.1%). C6-ceramide (0.660.2%) but notdihydro-C6-ceramide (1.960.3%) diminished the number ofPCNA-positive cells in the neointimal layer but not in themedial layer of the carotid artery (n54 to 8 experimentalarteries, P,0.05, 1-way repeated-measures ANOVA fol-lowed by the Bonferronit test). These data suggest thatceramide reduces neointimal hyperplasia by diminishing thepercentage of VSM cells or myofibroblasts that enter the cellcycle after trauma to the vessel wall.
Cell-permeable ceramide as well as exogenous sphingo-myelinase can mimic tumor necrosis factor, Fas ligand, orionizing radiation to induce apoptosis in hemopoietic celllines.7 However, it remains controversial as to whetherceramide induces VSM cell apoptosis. It is possible thatexcessive or unregulated apoptosis can result in formation ofaneurysms or plaque rupture during vascular remodeling.Thus, fluorescence-activated cell sorting after propidiumiodide staining was used to determine whether C6-ceramide
induced significant apoptosis in primary VSM cells isolatedfrom rabbit carotid arteries. Primary cultured rabbit VSMcells treated with 5mmol/L C6-ceramide or dihydro-C6-ceramide for either 24 or 40 hours showed,1% apoptoticcell death. As a control, okadaic acid treatment (100 nmol/L)significantly induced apoptosis after 24 hours (5263%) and40 hours (6962%) (Figure 5A). To confirm these studies,apoptosis was assessed in situ at time points when apoptoticmedial cells were identified after balloon angioplasty injury.14
Minimal pyknotic nuclei were evident in either vehicle-treated or ceramide-treated arteries at 15 to 60 minutes afterangioplasty (Figure 5B). In data not shown, pyknotic nucleiwere not observed in sections from ceramide-treated arteries2 weeks after angioplasty. In addition, we were unable toobserve any evidence of apoptotic cells in stretch-injuredarteries at any time point by in situ end labeling of nickedDNA (data not shown). Taken together, it is suggested thatcell-permeable ceramide limits stenosis by arresting VSMcell growth without inducing significant apoptosis.
We next investigated both early morphological and bio-chemical determinants for an inflammatory or proliferativephenotype in stretch-injured VSM cells. Figures 6A through6D show H&E staining of control (Figure 6A), vehicle-coated(Figure 6B), or ceramide-coated (Figures 6C and 6D) arteriesat 15 or 60 minutes after angioplasty. Surprisingly, littleevidence was noted for severe clinical damage at these earlytime points. Morphologically, even though there were earlyand reproducible changes in the integrity of the endotheliallining after balloon injury, there was minimal VSM cellularnecrosis or apoptosis. There was also minimal evidence ofmacrophage or neutrophil invasion (panel D), which wasconfirmed by immunohistochemistry with antibodies to either
Figure 5. C6-ceramide did not induce appreciable cellular apo-ptosis in carotid artery explants or in VSM cells in situ. A, Apo-ptosis by fluorescence-activated cell sorting after propidiumiodide staining in explanted and cultured cells from untreatedcarotid arteries. n54 cultures, P,0.05, Student t test. B, Apo-ptosis in situ by quantifying the number of pyknotic nuclei inH&E-stained arterial sections at 15 to 60 minutes after angio-plasty. These experiments were analyzed by a double-blindmethod; 1-way repeated-measures ANOVA, 6 arteries. Data aremean6SEM; P.0.05.
Figure 6. Minimal morphological injury was observed afteracute stretch injury. Representative H&E-stained sections fromcontrol and stretch-injured carotid arteries 15 and 60 minutesafter angioplasty. A through C, Control, vehicle-treated balloon,and C6-ceramide-coated balloon arteries 15 minutes after angio-plasty, respectively. The arrow reflects endothelial cells presentin panel A but not in panels B or C. D, Carotid artery 60 minutesafter angioplasty. The small arrow marks neutrophils and thelarge arrow macrophages; the phenotypes of which were con-firmed by immunohistochemistry. These photomicrographs arerepresentative of 3 arteries for each condition and time point.Bar550 mm.
286 Circulation Research August 18, 2000
by guest on December 3, 2015http://circres.ahajournals.org/Downloaded from
macrophages (RAMII, DAKO) or neutrophils (LY6G,Pharmingen) (data not shown). Positive controls for theseantibodies included thrombolytic arteries.
Even though there was little evidence of severe clinicaldamage, there were significant elevations in the phosphory-lation states of critical kinases involved in the proliferativeresponse to stretch injury. Evidence from in vitro studiessuggests that ceramide arrests cell growth by inhibiting thegrowth factor–induced ERK cascade and possibly by inhib-iting the PKB cascade.10–12Thus, to elucidate mechanisms bywhich ceramide limits neointimal hyperplasia, the phosphor-ylation states of ERK2 and PKBa were investigated usingfreshly excised carotid arteries after angioplasty (Figure 7).Phosphorylation of ERK2 and PKBa was increased at 15minutes and 24 hours after balloon injury. With ceramidetreatment, the phosphorylation states of these kinases weresignificantly diminished and remained at basal levels for upto 24 hours. These data suggest that ceramide-mediatedinhibition of ERK and PKB phosphorylation are very earlyevents in minimizing the proliferative and inflammatoryresponses of VSM cells to stretch injury.
DiscussionRapid advances in the field of sphingolipid-based signaltransduction have identified several metabolic products aspotential targets for pharmacological manipulation. Receptor-generated ceramide has been implicated in growth regulation,apoptosis, and cellular differentiation in vitro.7,8 The presentstudy extends these observations to an in vivo model ofarterial stretch injury as a consequence of balloon angio-
plasty. We demonstrate that a cell-permeable ceramide selec-tively limits neointimal hyperplasia without inducing signif-icant apoptosis. In addition, we further demonstrate the utilityof delivering cell-permeable ceramide directly to the site ofvascular injury by applying the bioactive lipid as a gel on theballoon catheter itself. Our studies indicate that the efficacyof therapy might be a consequence of physical force, whichtransfers ceramide from the inflated balloon to the site ofvascular lesions. Pharmacokinetic studies indicate that thetransferred ceramide rapidly penetrates the medial layer ofthe VSM at a therapeutic dose. Thus, intra-arterial delivery ofcell-permeable ceramide has a high likelihood of clinicalsuccess, as the failure of experimentally effective therapies tosucceed in clinical trials is often the consequence of subop-timal doses being delivered to the site of injury for theappropriate duration.24
It is noteworthy that stretch injury resulted in rapid changesin ERK and PKB activities that preceded marked signs ofinflammation. The sustained phosphorylation of these kinasesmost likely reflects continuous remodeling of damaged arter-ies. The downregulation of both ERK and PKB activitieswithin 15 minutes of ceramide treatment argues very stronglyfor the seminal roles played by these mitogenic and cellsurvival pathways in the pathology of neointimal hyperplasia.The rapid inhibition of kinase activity precedes any substan-tive morphological changes as assessed by H&E staining. Weare intrigued by the observation that ceramide treatmentinhibits PKB activity leading to growth arrest without apo-ptosis. This might reflect the fact that the cell cycle transcrip-tion factor E2F is downstream of PKB.25 Regardless ofmechanism, direct administration of ceramide to the site ofvascular injury results in a chronic inhibition of kinasesignaling cascades linked to mitogenesis.
Even though altered ceramide metabolism has been impli-cated in atherosclerosis, diabetes mellitus, and cancer, cer-amide analogues have not yet been considered as therapeuticsfor proliferative vascular diseases.26–28 Increased concentra-tions of lactosyl- and glycoceramide conjugates at the ex-pense of endogenous ceramide were noted in models ofatherosclerosis and diabetes mellitus,27–29and this diminishedlevel of ceramide correlated with VSM cell proliferation andvasoconstriction.7,27 Thus, it is logical to consider the use ofexogenous ceramide analogues as antimitogenic agents.
We have used an animal model that responds to stretchinjury with significant and reproducible neointimal hyperpla-sia. However, restenosis in humans reflects other mecha-nisms, such as vessel recoil and negative vascular remodel-ing, in addition to neointimal hyperplasia.1,2 The interactionsbetween these complications are only now being identified.Growth factors that induce neointimal hyperplasia also con-tribute to vessel narrowing caused by recoil through inflam-matory and myofibroproliferative mechanisms.5 In addition,adventitial proliferation and fibrosis may also contribute tonegative vascular remodeling.6 Therefore, it is possible thatdelivery of antiproliferative, cell-permeable lipid therapeuticsthat block growth factor signaling cascades can contribute toa decrease in restenosis after PTCA through multiplemechanisms.
Figure 7. ERK2 and PKBa phosphorylation was diminishedafter ceramide-coated balloon angioplasty in rabbit carotidarteries. A, Representative Western blot for ERK2 and PKBaprobed using phosphorylation-specific antibodies. Lysates fromNIH3T3 cells treated with or without platelet-derived growth fac-tor were used as positive and negative controls, respectively. Band C, Immunoblot data. n54 carotid preparations for eachtime point and condition; mean6SEM. Stars indicate P,0.05,Student t test.
Charles et al Ceramide Limits Neointimal Hyperplasia 287
by guest on December 3, 2015http://circres.ahajournals.org/Downloaded from
Documentation that cell-permeable ceramide can be usedas an efficacious treatment for neointimal hyperplasia afterstretch injury has important ramifications for control ofdysregulated smooth muscle proliferation not only afterangioplasty but also after stent placement, hemodialysisaccess failure, and diabetic retinopathy. In fact, neointimalformation is more significant after stenting than after balloonangioplasty.30 Our studies demonstrating that ceramide deliv-ery is an effective treatment in a model of neointimalhyperplasia after stretch injury argue for the applicability andefficacy of ceramide-coated stents. The ability to deliver thebioactive lipid directly at the site of injury has strong clinicalpotential. In addition to delivering this drug on the tip ofballoon catheters or through infusion ports, antimitogenicceramide analogues can be delivered as components ofconventional or cationic liposomal vectors, potentially aug-menting the efficacy of gene transfer and targeting strategies.
In this report, we have demonstrated that intra-arterialdelivery of ceramide analogues via the balloon tip of embo-lectomy catheters is technically feasible and targets the drugprecisely where it is needed. Use of endogenous lipid-derivedmetabolites as well as lipomimetic drugs promises highefficacy with low toxicity. This study establishes ceramideanalogue–coated balloon catheters as an efficacious therapyto reduce neointimal hyperplasia after stretch injury. More-over, this study documents a signal transduction mechanismresponsible, in part, for ceramide-induced VSM growth arrestin vivo.
AcknowledgmentsThis work was supported by grants from the W.W. Smith CharitableTrust and from the NIH (RO1 DK53715). We thank Xuwen Peng,DVM; Kang Li, PhD; and Joy Ellwanger, CAHT, for their assistancewith the animal surgeries and tissue sectioning. We thank ElliotVesell, MD (Chairman, Department of Pharmacology, Penn StateUniversity), and Ralph Damiano Jr, MD (Chief, CardiovascularSurgery, Penn State University), for insightful discussion.
References1. Casterella PJ, Teirstein PS. Prevention of coronary restenosis.Cardiol
Rev. 1999;7:219–231.2. Pasterkamp G, de Kleijn DP, Borst C. Arterial remodeling in atheroscle-
rosis, restenosis and after alteration of blood flow: potential mechanismsand clinical implications.Cardiovasc Res. 2000;45:843–852.
3. Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s.Nature. 1993;362:801–809.
4. Schwartz SM, Majesky MW, Murry CE. The intima: development andmonoclonal responses to injury.Atherosclerosis. 1995;118(suppl):S125–S140.
5. Rosanio S, Tocchi M, Patterson C, Runge MS. Prevention of restenosisafter percutaneous coronary interventions: the medical approach.ThrombHaemost. 1999;82(suppl 1):164–170.
6. Wilcox JN, Waksman R, King SB, Scott NA. The role of the adventitiain the arterial response to angioplasty: the effect of intravascularradiation.Int J Radiat Oncol Biol Phys. 1996;36:789–796.
7. Kester M. Spingolipid metabolites and the cellular phenotype.TrendsGlycosci Glycotechnol. 1997;9:447–460.
8. Coroneos E, Martinez M, McKenna S, Kester M. Differential regulationof sphingomyelinase and ceramidase activities by growth factors andcytokines: implications for cellular proliferation and differentiation.J Biol Chem. 1995;270:23305–23309.
9. Mandal A, Wang Y, Ernsberger P, Kester M. Interleukin-1-inducedether-linked diglycerides inhibit calcium-insensitive protein kinase Cisotypes: implications for growth senescence.J Biol Chem. 1997;272:20306–20311.
10. Coroneos E, Wang Y, Panuska JR, Templeton DJ, Kester M. Sphin-golipid metabolites differentially regulate extracellular signal-regulatedkinase and stress-activated protein kinase cascades.Biochem J. 1996;316:13–17.
11. Westwick JK, Bielawska AE, Dbaibo G, Hannun YA, Brenner DA.Ceramide activates the stress-activated protein kinases.J Biol Chem.1995;270:22689–22692.
12. Summers SA, Garza LA, Zhou H, Birnbaum MJ. Regulation of insulin-stimulated glucose transporter GLUT4 translocation and Akt kinaseactivity by ceramide.Mol Cell Biol. 1998;18:5457–5464.
13. Shinomiya M, Shirai K, Saito Y, Yoshida S. Inhibition of intimalthickening of the carotid artery of rabbits and of outgrowth of explants ofaorta by probucol.Atherosclerosis. 1992;97:143–148.
14. Pollman MJ, Hall JL, Gibbons GH. Determinants of vascular smoothmuscle cell apoptosis after balloon angioplasty injury: influence of redoxstate and cell phenotype.Circ Res. 1999;84:113–121.
15. Komukai M, Wajima YS, Tashiro J, Shinomiya M, Saito Y, Morisaki N.Carvastatin suppresses intimal thickening of rabbit carotid artery afterballoon catheter injury probably through the inhibition of vascular smoothmuscle cell proliferation and migration.Scand J Clin Lab Invest. 1999;59:159–166.
16. Jacobs LS, Kester M. Sphingolipids as mediators of effects of platelet-derived growth factor in vascular smooth muscle cells.Am J Physiol.1993;265:C740–C747.
17. Levison SW, Ducceschi MH, Young GM, Wood TL. Acute exposure toCNTF in vivo induces multiple components of reactive gliosis.ExpNeurol. 1996;141:256–268.
18. Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex(ABC) in immunoperoxidase techniques: a comparison between ABC andunlabeled antibody (PAP) procedures.J Histochem Cytochem. 1981;29:577–580.
19. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purifi-cation.Can J Biochem Physiol. 1959;37:911–917.
20. Oberhoff M, Herdeg C, Baumbach A, Shamet K, Kranzhofer A, WeingartnerO, Rubsamen K, Kluge M, Karsch KR. Time course of smooth muscle cellproliferation after local drug delivery of low-molecular-weight heparin usinga porous balloon catheter.Cathet Cardiovasc Diagn. 1997;41:268–274.
21. Levison SW, Rothstein RP, Brazel CY, Young GM, Albrecht PJ.Selective apoptosis within the rat subependymal zone: a plausiblemechanism for determining which lineages develop from neural stemcells.Dev Neurosci. 2000;22:106–115.
22. Negoro N, Hoshiga M, Seto M, Kohbayashi E, Ii M, Fukui R, Shibata N,Nakakoji T, Nishiguchi F, Sasaki Y, Ishihara T, Ohsawa N. The kinaseinhibitor fasudil (HA-1077) reduces intimal hyperplasia throughinhibiting migration and enhancing cell loss of vascular smooth musclecells.Biochem Biophys Res Commun. 1999;262:211–215.
23. Oberhoff M, Novak S, Herdeg C, Baumbach A, Kranzhofer A, Bohnet A,Horch B, Hanke H, Haase KK, Karsch KR. Local and systemic deliveryof low molecular weight heparin stimulates the reendothelialization afterballoon angioplasty.Cardiovasc Res. 1998;38:751–762.
24. Schwartz RS. Pathophysiology of restenosis: interaction of thrombosis,hyperplasia, and/or remodeling.Am J Cardiol. 1998;81:14E–17E.
26. Ichikawa S, Sakiyama H, Suzuki G, Hidari KI, Hirabayashi Y. Expressioncloning of a cDNA for human ceramide glucosyltransferase that catalyzesthe first glycosylation step of glycosphingolipid synthesis [publishedcorrection appears inProc Natl Acad Sci U S A. 1996;93:12654].ProcNatl Acad Sci U S A. 1996;93:4638–4643.
27. Johns DG, Charpie JR, Webb RC. Is ceramide signaling a target forvascular therapeutic intervention?Curr Pharm Des. 1998;4:481–488.
28. Zador IZ, Deshmukh GD, Kunkel R, Johnson K, Radin NS, Shayman JA.A role for glycosphingolipid accumulation in the renal hypertrophy ofstreptozotocin-induced diabetes mellitus.J Clin Invest. 1993;91:797–803.
29. Chatterjee S. Sphingolipids in atherosclerosis and vascular biology.Arte-rioscler Thromb Vasc Biol. 1998;18:1523–1533.
30. von Birgelen C, Kutryk MJ, Serruys PW. Three-dimensional intra-vascular ultrasound analysis of coronary stent deployment, andin-stent neointimal volume: current clinical practice and the conceptsof TRAPIST, ERASER, and ITALICS.J Invasive Cardiol. 1998;10:17–26.
288 Circulation Research August 18, 2000
by guest on December 3, 2015http://circres.ahajournals.org/Downloaded from
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:
by guest on December 3, 2015http://circres.ahajournals.org/Downloaded from
