Current siRNA Targets in the Prevention and Treatment of Intimal Hyperplasia

Abstract

Intimal hyperplasia (IH) is the leading cause of late vein and prosthetic bypass graft failure. Injury at the time of graft implantation leading to the activation of endothelial cells and dedifferentiation of vascular smooth muscle cells to a synthetic phenotype are known causes of IH. Prior attempts to develop therapy to mitigate these cellular changes to prevent IH and graft failure have failed. Small interfering RNA (siRNA) mediated targeted gene silencing is a promising tool to prevent IH. Several studies have been performed in this direction to target genes that are involved in IH. In this review we discuss siRNA targets that are being investigated for prevention and treatment of IH.

Introduction

Intimal hyperplasia (IH)

Intimal hyperplasia remains the most common cause of cardiovascular bypass graft failure and can also occur after angioplasty. IH is the result of an excessive wound healing response of the layers of the arterial or venous vessel wall. The end product is an accumulation of synthetic smooth muscle cells and extracellular matrix in the lumen of the vessel. These processes lead to a successive narrowing of the vessel lumen thus impairing blood flow. Several origins of these activated smooth muscle cells have been discussed, which include media, adventitia, and bone marrow precursor cells (Davies and Hagen, 1994; Goel et al., 2012; Varcoe et al., 2006; Yokote et al., 2003). Many studies have analyzed the molecular and cellular mechanisms of IH and numerous attempts have been made to define pharmacologic strategies to mitigate IH formation in response to angioplasty or bypass grafting (Bhasin et al., 2012; Conte et al., 2006; Wallitt et al., 2007).

Small interfering RNA (siRNA)

siRNAs are non-coding RNAs. Pre-siRNA is cleaved by the endoribonuclease dicer into short double stranded molecules that heterodimerize with complimentary sequences within target mRNA strands (Elbashir et al., 2001). This complex subsequently binds Ago2 and other proteins forming the RNA interference induced silencing complex (RISC), which catalytically degrades that mRNA strand (Lingel et al., 2003). Thus siRNA can be used to achieve temporary gene silencing without the need for viral vectors (Dorsett and Tuschl, 2004; Elbashir et al., 2001; Zamore et al., 2000). Early human trials proved that siRNA could be delivered safely and efficiently (Coelho et al., 2013; Tabernero et al., 2013).

Antisense therapy poses multiple challenges. The Prevent III trial illustrated how convincing pre-clinical studies may not translate into clinical success. Oligodeoxyribonucleotide (DNA based antisense therapy) mediated silencing of the cell cycle regulator E2F in vein grafts did not protect from graft failure (Conte et al., 2006).

Given cellular signaling redundancy it might be necessary to silence multiple genes simultaneously to achieve synergistic, clinically relevant effects.

Further, cell and disease specific siRNA delivery and offside target effects are additional concerns that need to be addressed for improved results.

This review discusses recent studies that used small interfering RNA to silence molecular targets in order to modulate processes that contribute to IH formation (Table 1). Potential targets for the prevention and therapy of atherosclerosis and aortic aneurysm have been recently discussed elsewhere (Pradhan-Nabzdyk et al., 2014).

Table 1.

List of siRNA Targets in Intimal Hyperplasia.

Target	Mechanism	Study Type	References
Growth Factors
TGF-beta1 – SMAD3 and CTGF	VSMC proliferation and migration, MMP modulation, ECM remodeling	Rat model, in vitro	Sun et al., 2012; Suwanabol et al., 2012a; Kundi et al., 2009
ATF / EGFR / Nox1	VSMC migration, ROS production	In vitro	Bakken et al., 2009; Duru et al., 2012
IGF-1 / CIC-2 / AngII / Cx43	VSMC proliferation and migration	In vitro	Cheng et al., 2007; Jia et al., 2008; Jia et al., 2011
Midkine	Cell migration, inflammation	Rabbit model	Banno et al., 2006
Transcription Factors and Cell Cycle Regulators
E2F	VSMC growth	Mouse model, in vitro	Giangrande et al., 2007
C-myc	VSMC proliferation	Rat model, in vitro	Wang et al., 2011
Survivin	VSMC proliferation, migration and apoptosis resistance	In vitro	Nabzdyk et al., 2011
AP-1	VSMC proliferation, differentiation and migration, MMP expression	In vitro	Zhang et al., 2012
Ptc-1 / Shh / GLI2	VSMC proliferation, vascular remodeling	In vitro	Redmond et al., 2013; Li et al., 2010
Immune Mediators
TLR4	Inflammation	Mouse model	Zhang et al., 2011; Karper et al., 2011
AIF-1	VSMC proliferation	Mouse model	Sommerville et al., 2009
IP-10	Chemotaxis, antiangiogenesis	Rabbit model	Zuojun et al., 2012
NF-kappaB	Inflammation	Rat model	Meng et al., 2013
Matrix Proteinases
MMP-2/-9	ECM remodeling, VSMC migration	In vitro	Turner et al., 2007; Hlawaty et al., 2009
MT1-MMP	Serum induced VSMC proliferation	In vitro	Mountain et al., 2012
ADAMTS-7	ECM remodeling	Rat model	Wang et al., 2009
Protein Kinases
CK1alphaLS	VSMC proliferation, ECM composition	Ex vivo	Panchenko et al., 2010
PKCdelta	Loss of function impairs EC migration	In vitro	Bai et al., 2010
MARKCS	VSMC migration and proliferation	In vitro	Monahan et al., 2009
Cell Surface Molecules
Cav1	eNOS signaling, VSMC proliferation	In vitro	Luo et al., 2010
Cdh11	VSMC migration	In vitro	Monahan et al., 2007
Miscellaneous
SCN9A	VSMC migration	In vitro	Meguro et al., 2009
TSP-2	VSMC attachment	In vitro	Yoshida et al., 2011; Nabzdyk et al., 2014
Girdin	VSMC migration, actin filament reorganization	Rabbit model, in vitro	Miyachi et al., 2013
Galphaq	VSMC proliferation, MMP-9 expression	Mouse model, in vitro	Zou et al., 2013
Rab5	VSMC proliferation and migration	In vitro	Ma et al., 2010
HuR	VSMC proliferation	In vitro	Pullmann et al., 2005

Growth Factors

Transforming growth factor-beta (TGF-β1) / Connective tissue growth factor (CTGF)

TGF-β1 has been linked to IH formation in response to vascular injury. For years TGF-β has been conceived as an inhibitor of proliferation. However, in a rat model of internal jugular vein-to-carotid artery grafting nanoparticle-mediated TGF-β1 silencing decreased MMP-1 and increased TIMP-1 expression leading to an overall decrease in vein graft IH (Sun et al., 2012).

Furthermore, TGF-β via its signaling protein Smad3 can promote VSMC proliferation and IH partly via nuclear p27 export (Tsai et al., 2009). TGF-β activates ERK-MAPK in vascular smooth muscle cell (VSMC) via Smad3 in vitro and in vivo. In VSMC Smad3 silencing blocked TGF-β induced ERK-MAPK phosphorylation (Suwanabol et al., 2012a). Smad3 overexpression in rat carotid arteries increased p38 phosphorylation, while p38 silencing blocked TGF-beta-induced Akt phosphorylation and TGF-β/Smad3 mediated SMC proliferation (Suwanabol et al., 2012b).

Increased levels of Smad3 were also detected in rat carotid artery medias after angioplasty injury. Adenoviral mediated Smad3 overexpression revealed hallmarks of adaptive remodeling including neointimal thickening and luminal expansion. Supernatant taken from Smad3-overexpressing SMC led to adventitial fibroblast transformation, proliferation, and collagen production. The profibrotic factor CTGF is involved in fibroblast proliferation, angiogenesis, and extracellular matrix (ECM) synthesis and is considered a downstream mediator of certain TGF-β signaling. CTGF silencing in these Smad3-overexpressing SMC mitigated the effects of their supernatant on adventitial fibroblasts (Kundi et al., 2009). CTGF mRNA was also found to be upregulated in VSMC in response to high glucose environment. CTGF silencing in VSMC attenuated high glucose-induced ECM deposition, proliferation, and migration (Liu et al., 2007).

Amino-terminal fragment (ATF) / Epidermal growth factor receptor (EGFR) / NADPH-oxidase 1 (Nox1)

The ATF of urokinase has been shown to activate EGFR via A Disintegrin and Metalloproteinase Domains (ADAM). ATF increased VSMC migration via EGFR phosphorylation, which was blocked by EGFR silencing (Bakken et al., 2009). Further, silencing of radical oxygen species producing NADPH-oxidase 1 (Nox1) prevented ATF-mediated EGFR activation and cell migration (Duru et al., 2012).

Angiotensin II (AngII) / Connexin-43 (Cx43)

AngII is a potent vasoconstrictor and plays a crucial role in vascular remodeling in response to injury. More recently AngII has been recognized as a growth factor with pleiotropic effects. AngII caused hypertrophy and hyperploidy in VSMC (Geisterfer et al., 1988). Further, via angiotensin II type 1 receptor (AT-1R) AngII led to an upregulation of gap junction protein Cx43 in venous SMC and subsequently increased proliferation. These AngII mediated effects were blocked by extracellular signal-regulated kinases (ERK 1/2), p38 MAPK, and JNK silencing, respectively. Further, Cx43 silencing inhibited AngII mediated SMC migration and proliferation; the latter possibly via decreased expression of cyclin E (Jia et al., 2008).

AngII induced AT-1R and insulin-like growth factor-1 (IGF-1) protein expression in VSMC. IGF-1 also increased AT-1R expression, and likewise AngII increased IGF-1R expression. IGF-1R and AT-1R silencing blocked these effects. Further, ERK1/2 silencing blocked AT-II and IGF-1 signaling crosstalk. Both AngII and IGF-1 led to Cx43 upregulation via ERK1/2 phosphorylation and activation of transcription factor activator protein 1 (AP-1) (Jia et al., 2011).

Insulin-like growth factor (IGF)-1 / Chloride Channel 2 (ClC-2)

IGF-1 is believed to contribute to IH formation possibly through dysregulation of VSMC migration and proliferation. IGF-1 has shown to upregulate ClC-2, a member of the volume-regulated Cl- channel family. ClC-2 silencing inhibited IGF-1-induced VSMC proliferation (Cheng et al., 2007).

Midkine (MK)

Midkine is a heparin-binding growth factor and MK deficiency in mice is associated with decreased IH formation in a restenosis model. Topical application of MK siRNA mixed with atelocollagen to rabbit jugular vein-to-carotid artery interposition vein grafts mitigated local inflammatory cell recruitment and cell proliferation, and led to a greater than 90% IH reduction compared with controls (Banno et al., 2006).

Transcription Factors and Cell Cycle Regulators

E2F

The family of E2F transcription factors is involved in a variety of cellular functions including proliferation and differentiation. The PREVENT III trial showed that nonselective E2F silencing did not confer protection from IH formation (Conte et al., 2006). It is known, that several E2F isoforms modulate IH pathology. While E2F3 has shown to increase VSMC proliferation, E2F4 decreased VSMC growth. In a mouse model of bypass grafting selective E2F3 silencing mitigated VSMC growth and IH formation (Giangrande et al., 2007).

C-myc

C-myc is a potent promoter of mitosis. Its activation also influences VSMC proliferation and c-myc silencing decreased VSMC proliferation in vitro and mitigated IH in a rat model of jugular vein-to-carotid artery interposition grafting (Wang et al., 2011).

Survivin (SVV)

Recent research has implicated SVV in the development of IH. It is known that nuclear SVV is essential for mitosis, whereas mitochondrial SVV has a cytoprotective function. In VSMC SVV siRNA depleted predominantly the nuclear SVV pool, whereas the larger mitochondrial pool appeared less affected. Cell cycle regulators p53 and p27 were significantly upregulated at the protein level. SVV silencing blocked VSMC proliferation, arresting VSMC growth. SVV knockdown reduced migration to PDGF-AB, and actin-phalloidin staining revealed disorganized actin filaments and polygonal cell shape. However, apoptosis (DNA content and annexin V flow cytometry) was not directly induced by SVV silencing, and sensitivity to apoptotic stimuli was unchanged (Nabzdyk et al., 2011).

AP-1

The activator protein-1 (AP-1) transcription factor upon binding to its specific DNA sequence promotes VSMC proliferation, dedifferentiation, and migration. Conversely, AP-1 silencing leads to increased VSMC SM alpha-actin protein expression, decreased proliferation, and suppressed serum-induced MMP-2 expression (Zhang et al., 2012).

Patched receptor-1 (Ptc-1) / Sonic Hedgehog (Shh) / GLI family zinc finger 2 (Gli2)

Ptc-1 is a transmembrane receptor for sonic hedgehog a key regulator of embryogenesis and tumorigenesis. Low blood flow in vivo after ligation injury increased Ptc-1/Notch expression, VSMC growth, and vascular remodeling. Ptc-1 silencing by perivascular siRNA delivery blocked these effects (Redmond et al., 2013).

VSMC of murine vein graft IH lesions were found to have increased Shh signaling. Gli2 is a mediator of Shh signaling. Shh/Gli2 signaling induced VSMC proliferation, while Gli2 silencing led to VSMC cell cycle arrest and decreased levels of cyclin D1, cyclin E, and phosphorylated retinoblastoma (pRB). Cyclin D1 silencing also abrogated Shh induced VSMC proliferation (Li et al., 2010).

Immune Mediators

Toll like receptor 4 (TLR4)

TLR4 is a transmembrane receptor and pleiotropic activator of the innate immune system. TLR4 and other pro-inflammatory cytokines are upregulated in murine wire-injury-induced carotid arteries and in platelet-derived growth factor (PDGF)-incubated VSMC. TLR4 silencing ameliorated IH formation in injured mouse carotid arteries and decreased proliferation and migration of LPS and PDGF activated VSMC (Zhang et al., 2011). Focal TLR4 upregulation was also observed in fresh human saphenous veins (huSV) and huSV that served as coronary artery vein bypass graft. TLR4 silencing via perivascular lentiviral delivery decreased IH formation in murine vein grafts (Karper et al., 2011).

Allograft inflammatory factor-1 (AIF-1)

AIF is induced by interferon and other cytokines and found to be upregulated in response to vascular injury. AIF-1 activation has shown to increase VSMC proliferation contributing to neointimal lesions. Adenovirus mediated AIF-1 silencing in balloon injured rat carotid arteries mitigated IH. Further, AIF silencing in VSMC decreased mitogen-activated protein kinase p38 activation compared with control. AIF-1-mediated VSMC proliferation is p38 kinase dependent, while AIF-1-mediated VSMC migration is p38 independent (Sommerville et al., 2009).

Interferon-gamma-inducible protein (IP)-10 or C-X-C motif chemokine 10 (CXCL10)

IP-10 is produced and released by a variety of cells including macrophages, endothelial cells, and fibroblasts in response to interferon gamma. IP-10 has been implicated in the formation of IH in response to vascular injury possibly because of its chemotactic and antiangiogenic effects. IP-10 silencing reduced intima-to-media ratio in a rabbit model of carotid artery injury. The exact mechanism however remains elusive (Zuojun et al., 2012).

Nuclear factor kappaB (NF-kappaB)

NF-kappaB is a key regulator of inflammatory injury response and has been linked to stenosis progression. In a rat model of external jugular vein to abdominal aorta grafting, NF-kappaB silencing blunted inflammatory factor activation and IH formation (Meng et al., 2013).

Proteinases

Matrix Metalloproteinases (MMP), membrane type 1 metalloproteinase (MT1-MMP), A disintegrin and metalloproteinase with thrombospondin motifs (ADAMTs)

Matrix metalloproteinases (MMP) are considered a major contributor of IH formation. MMP activation enhances SMC invasion, a crucial step of IH development in saphenous vein (SV) bypass grafts. Both MMP-2 and MMP-9 silencing led to reduced MMP-2 and MMP-9 secretion and mitigated SV-SMC invasion through a Matrigel barrier (Turner et al., 2007).

MT1-MMP, a transmembrane activator of MMP-2, is upregulated by estrogen. MT-1-MMP knockdown decreased basal MMP-2, but did not change VSMC invasion or proliferation. Interestingly, estrogen stimulation of MT1-MMP-silenced VSMC still increased MMP-2 activity and cell invasion. However, MT1-MMP silencing did inhibit estrogen and serum-induced VSMC proliferation (Mountain et al., 2012).

Members of the ADAMTS proteinase family are also involved in ECM degradation. In injured rat carotid arteries ADAMTS-7 degraded ECM cartilage oligomeric matrix protein (COMP). ADAMTS-7 silencing by perivascular siRNA delivery decreased IH compared to controls (Wang et al., 2009).

Leukotrienes (LT) are derivatives of arachidonic acid that elicit pleiotropic inflammatory effects. LT are also involved in vascular injury response and presence of Leukotriene B4 (LBT4) has been associated with increased levels of MMP-2 and MMP-9. LTB4-induced VSMC migration can be mitigated by MMP-2 silencing (Hlawaty et al., 2009).

Protein Kinases

Protein kinase CK1alphaLS (CK1alphaLS)

Amongst other functions, protein kinase CK1alpha regulates cell proliferation and differentiation. Alternative exon splicing leads to up to four isoforms. While the L-insert aids in targeting the kinase to the nucleus, the role of nuclear CK1alphaLS has been unclear. VSMC proliferation was decreased after L-insert-targeted CK1alphaLS silencing. Moreover, CK1alphaLS in silencing in cultured human arteries reduced VSMC proliferation, IH, and proteoglycan deposition. Lastly CK1alphaLS silencing led to a decrease of its target substrate heterogeneous nuclear ribonucleoprotein C and in differential expression or splicing of chemokine receptor CXCR4, MMP3, and others (Panchenko et al., 2010).

Protein kinase C delta type (PKCdelta)

PKCdelta is known to regulate cell proliferation and apoptosis and has been linked to IH formation. Further, PKCdelta knockout mice revealed delayed reendothelialization after arterial wire injury. Short hairpin RNA mediated PKCdelta silencing in endothelial cells also revealed migratory impairment (Bai et al., 2010).

Myristoylated alanine-rich C kinase substrate (MAR-CKS)

MARCKS is a protein kinase C (PKC) substrate, which was found to be upregulated in canine IH lesions. MARCKS silencing in VSMC decreased migration and proliferation partially via cyclin-dependent kinase inhibitor p27(kip1) upregulation. Interestingly, MAR-CKS silencing did not alter EC migration or proliferation (Monahan et al., 2009).

Cell Surface Molecules

Caveolin-1 (Cav1)

Cav1 is the integral component of plasma membrane caveolae and is involved in clathrin-independent endocytosis, eNOS signaling, mechanotransduction, and other forms of cell signaling. Significant ERK1/2 upregulation as well as a decreased cav1 expression were detected in rat VSMC exposed to different static pressures (0–180 mmHg). ERK1/2 phosphorylation and VSMC proliferation were increased by cav1 silencing (Luo et al., 2010).

Cadherin 11

Gene chip analysis of canine vein grafts showed increased mRNA expression for the cell surface adhesion molecule cadherin 11/osteoblast-cadherin (CDH11) in response to injury. Inhibition of CDH11 decreased SMC migration and proliferation in vitro (Monahan et al., 2007).

Miscellaneous

Na Channel Na(V)1.7 - SCN9A

Voltage-gated Na(+) channel currents (I(Na)) are expressed in several types of SMC. Na(V)1.7 is also expressed in aortas after balloon injury. Different from native aorta, cultured aortic SMCs expressed SCN9A, which encodes for Na(V)1.7. SCN9A silencing inhibited cell migration, while not affecting cell proliferation (Meguro et al., 2009).

TSP-2

Thrombospondin-2 (TSP-2) is an antiangiogenic matricellular protein that is upregulated in neointimal smooth muscle cells after prosthetic arterial bypass graft placement. TSP-2 silencing led to increased HAoSMC attachment to fibronectin in vitro (Yoshida et al., 2011). TSP-2 silencing could be achieved in AoSMC that infiltrated polyethyleneimine-TSP-2-siRNA-coated electrospun Dacron graft materials (Nabzdyk et al., 2014).

Girdin

Actin-binding protein girdin is upregulated in injured vein grafts and has previously linked to arterial remodeling. In a rabbit vein graft model, perivascular girdin siRNA delivery mitigated IH. Further, girdin silencing in SMC reduced cell migration and proliferation illustrating girdin’s relevant role in actin filament rearrangement (Miyachi et al., 2013).

G protein-coupled receptor alpha-q (Galphaq)

G protein-coupled receptors are involved in IH formation in parts by enhancing SMC proliferation. Galphaq increases vascular smooth muscle cell proliferation in vitro. In a murine femoral artery wire injury model Galphaq expression increased in a time-dependent manner. Galphaq siRNA applied externally to injured mouse femoral arteries reduced SMC proliferation but not cell migration. Galphaq silencing also reduced IH development and temporarily decreased MMP-9 but not MMP-2 expression (Zou et al., 2013).

Rab5a

Rab5a, a member of the Rab family of GTPases and key regulator of clathrin endosome formation, is upregulated in rat anastomotic IH lesions. Rab5a silencing reduced proliferation and migration of rat thoracic aorta VSMC (Ma et al., 2010).

Human antigen R (HuR)

HuR is an RNA-binding protein of the HU/ELAV family that stabilizes various mRNAs, including those of growth factors, cell cycle regulators, and cytokines.

Elevated levels of HuR have been detected in neointimal lesions. HuR silencing decreased basal and platelet-derived growth factor (PDGF) stimulation of VSMC proliferation (Pullmann et al., 2005).

Conclusion

Similar to atherosclerosis, IH has been a tremendous challenge for the field of vascular biology inflicting an enormous disease and financial burden on patients and the health care system, respectively. To date no effective pharmacologic strategy has been developed to sufficiently ameliorate IH formation after bypass grafting. Drug eluting stents (DES) for coronary angioplasty have been used for over a decade (Bangalore et al., 2012). DES release immunosuppressive compounds such as sirolimus and everolimus (inhibit mTOR pathway) or mitosis inhibitors such a paclitaxel (Bangalore et al., 2012). While these devices have shown to reduce IH formation after angioplasty, this harsh ablative approach renders patients at high risk for instent thrombosis. These patients are thus placed on a strict anti-platelet therapy (clopidogrel) for up to twelve months (El-Hayek et al., 2014; Loh et al., 2014). This approach has so far not been adopted for prosthetic bypass grafting and is not easily applicable for vein bypass grafting. Gentler ways to modulate gene expression and thus vascular remodeling in response to vascular injury are needed.

While the failure of the Prevent III trial may have been discouraging, it was critical in proving that non-viral gene silencing in vein bypass grafts is feasible and safe (Conte et al., 2006). It further illustrated, that more in depth analyses are needed to better understand the signaling networks in the cells of the injured vessel wall. A temporospatial gene expression analysis of canine vein graft EC and SMC revealed that thousands of genes are altered within 24 hours of implantation (Bhasin et al., 2012). A near return to baseline gene expression was observed after fours weeks for both EC and SMC (Bhasin et al., 2012). Such bioinformatic approaches analyze entire gene networks and thus may identify more suitable siRNA targets for the prevention of IH. Aside from target identification, additional challenges remain in developing siRNA delivery modalities that allow for reliable gene silencing across the entire vessel wall.

Collaboration between bioinformaticians, cardiovascular scientists/physicians, and chemical engineers may lead to the development of novel siRNA therapy approaches for the prevention of IH in the future.