Vascular Repair by Circumferential Cell Therapy Using Magnetic

Jan 6, 2016 - gene and cell therapy with custom-made magnetic fields enables circumferential re-endothelialization of vessels and improvement of vascu...
0 downloads 0 Views 3MB Size
Download PDF
Vascular Repair by Circumferential Cell Therapy Using Magnetic Nanoparticles and Tailored Magnets Sarah Vosen,† Sarah Rieck,† Alexandra Heidsieck,‡ Olga Mykhaylyk,§ Katrin Zimmermann,∥ Wilhelm Bloch,⊥ Dietmar Eberbeck,⊗ Christian Plank,§ Bernhard Gleich,‡ Alexander Pfeifer,∥ Bernd K. Fleischmann,*,†,△ and Daniela Wenzel*,†,△ †

Institute of Physiology I, Life&Brain Center, and ∥Institute of Pharmacology and Toxicology, University Clinic Bonn, Bonn 53127, Germany ‡ Zentralinstitut für Medizintechnik (IMETUM), TU München, München 85748, Germany § Institute of Experimental Oncology and Therapy Research, TU München, München 81675, Germany ⊥ Institute of Cardiovascular Research and Sport Medicine, German Sport University Cologne, Cologne 50735, Germany ⊗ Physikalisch-Technische Bundesanstalt Berlin, Berlin 10587, Germany S Supporting Information *

ABSTRACT: Cardiovascular disease is often caused by endothelial cell (EC) dysfunction and atherosclerotic plaque formation at predilection sites. Also surgical procedures of plaque removal cause irreversible damage to the EC layer, inducing impairment of vascular function and restenosis. In the current study we have examined a potentially curative approach by radially symmetric re-endothelialization of vessels after their mechanical denudation. For this purpose a combination of nanotechnology with gene and cell therapy was applied to site-specifically re-endothelialize and restore vascular function. We have used complexes of lentiviral vectors and magnetic nanoparticles (MNPs) to overexpress the vasoprotective gene endothelial nitric oxide synthase (eNOS) in ECs. The MNP-loaded and eNOS-overexpressing cells were magnetic, and by magnetic fields they could be positioned at the vascular wall in a radially symmetric fashion even under flow conditions. We demonstrate that the treated vessels displayed enhanced eNOS expression and activity. Moreover, isometric force measurements revealed that EC replacement with eNOS-overexpressing cells restored endothelial function after vascular injury in eNOS−/− mice ex and in vivo. Thus, the combination of MNP-based gene and cell therapy with custom-made magnetic fields enables circumferential re-endothelialization of vessels and improvement of vascular function. KEYWORDS: magnetic nanoparticles, magnetic fields, cell replacement, gene therapy, endothelium, endothelial nitric oxide synthase (eNOS), vascular function of nitric oxide (NO) production.5 In order to restore endothelial function, cell and gene therapy have come into focus, in particular because these strategies could enable longterm repair. For the prevention of restenosis the seeding of ECs at the site of angioplasty6,7 and the overexpression of vascular growth factors and eNOS have been explored.8,9 Even though cells and gene vectors have been locally applied within vessels using catheters, restoration of blood flow resulted in detach-

C

ardiovascular diseases are the leading cause of death worldwide, and they are mainly based on vascular dysfunction and atherosclerosis.1 Novel pharmacological and in particular interventional treatment strategies have improved the prognosis of patients with vascular disease; however, such therapeutic regimes still have multiple limitations and side effects: Pharmacological compounds with endothelium-protective action display limited half-life and need to be applied at an early stage of disease. Surgical and interventional revascularization approaches such as endarterectomy, angioplasty, and bypass grafting result in myointimal thickening and an increased risk for restenosis.2,3 Restenosis has been shown to be due to endothelial cell (EC) loss4 and the ensuing decrease © XXXX American Chemical Society

Received: August 11, 2015 Accepted: December 29, 2015

A

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX

Article

www.acsnano.org

Article

ACS Nano

Figure 1. Strategy of radially symmetric endothelial cell (EC) replacement in mouse aortas ex vivo using lentivirus (LV)/magnetic nanoparticle (MNP) complexes. (a) Schematic diagram of the experimental strategy for EC replacement in mouse aorta (left = transversal view, right = longitudinal view). (b) Transmitted (top) and fluorescence (bottom) light pictures of ECs 72 h after transduction with LV(eGFP)/SO complexes; bar = 100 μm. (c) Quantification of eGFP+ ECs 72 h after transduction with LV(eGFP)/MNP complexes (LV 30′: lentivirus for 30 min, LV oN: LV overnight, LV/PEI(300): 300 fg iron/IP PEI-Mag2, LV/SO(300): 300 fg iron/IP SO, LV/SO(4000): 4000 fg iron/IP SO). (d) Magnetic responsiveness experiments. E(t)/E(t0) indicates the relative extinction of the cell suspension in the photometer, E(t0) is the initial extinction, *p < 0.05, ***p < 0.001. (c) One way ANOVA, Tukey’s multiple comparison test.

MNP(300), LV/MNP(4000)).16,17 At 72 h after transduction, MNPs were visible in the cytoplasm of ECs and prominent eGFP expression was detected (Figure 1b). Importantly, LV/ SO and LV/PEI could strongly enhance the transgene expression compared to standard overnight transduction without MNPs, as demonstrated by the number of eGFP+ ECs (Figure 1c). Efficient cell retention of ECs at the vascular wall by a magnetic field requires high numbers of strongly magnetic cells. We therefore quantified the number of magnetically retained ECs in a magnetic rack (Figure S1a−d) and their iron content (Figure S1e,f), measured their magnetic responsiveness (Figure 1d), and also performed Prussian blue iron stainings (Figure S1g−k). We found that 72 h after transduction by application of complexes consisting of LV/SO(4000) the cells were loaded with about 30 pg iron/cell and were strongly magnetic in the applied magnetic fields. By contrast, complexes of LV/PEI could not be used, because of their limited uptake by ECs and toxic effects even at low concentrations (LV/PEI(300), Figure S2a) that resulted in cell detachment at higher concentrations (LV/PEI(4000), data not shown). Although LV/SO(4000) complexes did not show any obvious toxicity in the LDH assay, we also wanted to exclude more subtle effects on EC biology. We therefore assessed the expression of the key endothelial marker vascular endothelial growth factor receptor 2 (VEGFR2) (Figure S2b−g), the integrity of the actin cytoskeleton (Figure S2h−m), and vascular network formation of LV/SO(4000)-loaded ECs in matrigel (Figure S2n−t). All these assays did not reveal any difference between MNP-loaded and native or LV-treated ECs 72 h after transduction; strong magnetic cell labeling and preserved EC characteristics were also found 120 h after treatment (Figure S3a−h, Figure S4a−t). We also explored the effect of LV/SO(4000) transduction on EC growth, because this cell biological parameter appeared

ment of ECs or vectors and in adverse side effects at remote sites.4,10 In order to improve the therapeutic impact and to reduce offtarget effects, in earlier studies we and others have taken advantage of the combined use of magnetic nanoparticles (MNPs) and magnetic fields. For the local retention of pharmacological substances, viral vectors, and cells vascular stents were employed.11−14 The use of stents, however, also causes adverse side effects, as it is known to damage the vessel, promoting in-stent stenosis. We have therefore in our earlier work started to explore the utility of magnetic fields directly targeting MNPs to enhance transduction rates and to enable overexpression of reporter genes under blood flow in vivo.15 Since these initial studies suggested the feasibility of such an approach, in the current study we have developed this much further by establishing a magnetic targeting strategy for circumferential positioning of lentivirus (LV)/MNP-treated ECs that overexpress a functional gene under flow conditions (Figure 1a). For this purpose, we have generated a custom-made radially symmetric magnetic field based on the optimal arrangement of 12 bar magnets. This strategy resulted in radially symmetric retention of MNP-loaded eNOS-overexpressing ECs at the wall of mechanically injured vessels ex and in vivo, yielding even an improvement of vascular function.

RESULTS AND DISCUSSION In order to identify an MNP type that enables both an efficient lentiviral transduction and magnetization, we first compared two MNPs with different physicochemical properties. ECs were transduced with complexes of LV containing an enhanced green fluorescence protein (eGFP) expression cassette and SOMag5 (LV/SO) or PEI-Mag2 nanoparticles (LV/PEI) at 300 or 4000 fg iron/infectious particle (IP) at MOI 50 (LV/ B

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

vascular wall; quantification in aortic sections revealed that about 27% of the injected cells could be retained at the vessel wall (n = 3). For clinical applications an EC replacement strategy with functional cells appears particularly attractive, because this could prevent restenosis and other complications after invasive procedures such as plaque removal or stent implantation. We therefore examined this approach ex vivo and in vivo using genetically modified ECs overexpressing eNOS. We have chosen this strategy to improve the viability of the ECs and also to enhance vascular function. ECs were transduced with LV(eNOS-IRES-eGFP)/SO(4000) complexes. First, we determined the transduction kinetics and found strongly elevated numbers of transduced cells at all time points investigated compared to standard overnight transduction with the same LV (Figure 3a−c). Accordingly, eNOS expression in these cells was increased at mRNA and protein levels by 95- and 6-fold compared to native ECs, respectively (Figure S9a−c). The highest transduction rates of ECs were found after 120 h (Figure 3a), a time point when cells displayed normal growth rates after transduction with LV/SO(4000) (Figure S5). Next, nitric oxide production after eNOS overexpression was

particularly relevant for EC viability and the re-endothelialization of denuded vessels in vivo. Treatment with LV/SO(4000) led to reduced growth and proliferation rate of ECs during the first 72 h (Figure S5a−c), but this effect disappeared at later stages (72 to 120 h after transduction) (Figure S5a,b,d); there was no increase in EC apoptosis at any time point (Figure S6). Finally, we also studied the fate of LV/MNP complexes after application to ECs using fluorescent-labeled SO particles. After 72 and 120 h the particles appeared to be located within the cytoplasm surrounding the nucleus of the ECs (Figure S7a−f). This was corroborated by electron microscopy showing cytoplasmatic localization at 72 h (Figure S7g,h), 168 h (Figure S7i), and even 240 h (Figure S7j) after LV/SO(4000) loading of cells. Thus, we could identify a suitable MNP type and the required dose in order to ensure both strong transduction and magnetization of ECs. For efficient cell replacement therapy under flow conditions a magnet configuration allowing cell retention at a large vascular surface area is needed. The optimal field geometry would enable a radially symmetric retention of ECs at the vascular wall. For this purpose, as done in an earlier project,18 numerical simulations were performed yielding as ideal configuration four groups of three bar magnets (mean residual magnetization 1.195 T each) that are arranged in an alternating orientation perpendicular to the vessel. This magnet configuration was predicted to provide a homogeneous and strong radially symmetric magnetic gradient field19 (Figures 1a, 2a,b,c).

Figure 2. Characterization of magnet configuration required for radially symmetric cell deposition in vessels. (a) Schematic diagram of magnet configuration. (b) Gradient of magnetic flux density in a vessel generated by this magnet configuration. (c) Plexiglas holder containing magnets; the central cavity incorporates the vessel. (d) Transmitted (top) and fluorescence (bottom) light pictures of mouse aorta after perfusion with LV(eGFP)/SO(4000)-treated ECs (eGFP ECs) and magnet configuration shown in (a); bar = 2 mm. (e) Transversal section of the same aorta, bar = 100 μm. (f) Magnification of the boxed area of the vascular wall in (e), bar = 10 μm; blue = Hoechst, green = eGFP, magenta = alpha smooth muscle actin.

Figure 3. ECs treated with LV(eNOS-IRES-eGFP)/SO complexes. (a) Quantification of eGFP+ ECs at different time points after 30 min of transduction with LV(eNOS-IRES-eGFP)/SO complexes at 4000 fg iron/IP, MOI 50 or overnight (oN) transduction. (b, c) Fluorescence light pictures of ECs 120 h after transduction with LV(eNOS-IRES-eGFP)/SO (arrows indicate eGFP+eNOS+ cells (green = native eGFP (b), red = eNOS staining (c)), bar = 100 μm. (d) Radioimmunoassay of eNOS activity in ECs 120 h after treatment with LV(eNOS-IRES-eGFP)/SO (eNOS ECs) compared controls. (e) Diaminofluorescein (DAF) assay of NO production in ECs 120 h after 30 min of transduction with LV(eNOS)/SO; ECs: native ECs, LN: 100 μM eNOS inhibitor L-NAME, SO: 4000 fg iron/IP SO-Mag5, LV 30′: LV(eNOS) for 30 min, LV oN: LV(eNOS) overnight, LV(eNOS)/SO: complexes of LV(eNOS) and 4000 fg iron/IP for 30 min, LV(rrl)/SO: transduction with complexes containing control virus, SNP: sodium nitroprusside, positive control. (f) Nitric oxide production in response to the Ca2+ ionophore A23187, *p < 0.05, **p < 0.01, ***p < 0.001. (d, e) One-way ANOVA, Tukey’s multiple comparison test; (a, f) Student’s t test.

Next, we experimentally tested this using an ex vivo flow-loop model in endothelium-denuded murine aortas (Figure S8a−c, Figure 2c). ECs treated with LV(eGFP)/SO(4000) overexpressing the reporter gene eGFP were perfused through the vessel in a recirculating manner, and the cell retention at the vascular wall by the magnetic field was assessed. Our analysis revealed, as predicted, a radially symmetric distribution of eGFP+ ECs along the vessel wall (Figure 2d−f), proving that LV/MNP complexes and magnetic fields can be successfully used for the radially symmetric re-endothelialization of the C

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

Figure 4. EC replacement in eNOS−/− aortas using genetically engineered ECs. (a) Quantitative RT-PCR for eGFP (left) and eNOS (right) expression in denuded eNOS−/− aortas (w/o endothelium) reseeded with eNOS ECs, values normalized to 18S-rRNA. (b) Western blot of denuded eNOS−/− aortas reseeded with either eGFP or eNOS ECs. (c) Quantification of Western blot, values normalized to β-actin. (d) Radioimmunoassay of eNOS activity in denuded eNOS−/− aortas reseeded with eNOS ECs compared to controls. (e) cGMP ELISA of denuded eNOS−/− aortas reseeded with eNOS ECs compared to control, values normalized to total protein content. (f) Original traces of isometric force measurements in eNOS−/− aortas reseeded with eGFP ECs. (g) Original traces of isometric force measurements in eNOS−/− aortas reseeded with eNOS ECs. (h) Ratios of constriction amplitudes with and without L-NAME in eNOS−/− aortas reseeded with eNOS ECs compared to controls; the aortic rings are taken from at least 4 animals per group; *p < 0.05, **p < 0.01, ***p < 0.001. (a, d, h) One-way ANOVA, Tukey’s multiple comparison test; (c, e) Student’s t test.

mounted in a wire-myograph (Figure S10e), and the phenylephrine (Phe)-induced contractile response of each vascular ring was measured without and in the presence of the NOS inhibitor N-nitro-L-arginine methyl ester hydrochloride (LNAME, 100 μM): In the case of enhanced generation of the vasorelaxant NO by eNOS-overexpressing ECs one would expect increased contraction upon pharmacological eNOS inhibition with L-NAME, whereas in control aortas L-NAME application should not alter force generation. Our functional data revealed that vascular rings reseeded with eNOS ECs showed indeed increased contractions in the presence of LNAME; these measurements even yielded similar values to those in native aortas with intact endothelium (Figure S10f−h). This suggests that EC replacement with eNOS-overexpressing ECs can restore vasoactive NO production and hence vascular function. To unequivocally exclude that this effect could be caused by a few residual ECs in the denuded aortas, the same type of experiment was also performed in aortas from eNOS−/− mice. After radially symmetric EC replacement using eNOS ECs we found again strongly elevated levels of eNOS mRNA (Figure 4a) and protein (Figure 4b,c) in these aortas.

measured, and we found that ECs transduced with LV/SO complexes containing an eNOS expression cassette (eNOS ECs) displayed strongly elevated levels of eNOS activity and NO production compared to the controls of native ECs and ECs transduced with LV(eGFP)/SO(4000) (eGFP ECs) (Figure 3d,e). Importantly, NO production could be enhanced 2.2-fold in eNOS ECs by application of the Ca2+ ionophore A23187 (Figure 3f), indicating intact cellular regulation. Next, we tested the potential of these cells to be retained at the vascular wall by a magnetic field in a radially symmetric manner and determined the consequences of this treatment on vascular function. First, this was examined in wild-type aortas after mechanical removal of the EC layer (Figure S10a,b). After reseeding with eNOS ECs and homogenization of the wild-type aortas strongly elevated levels of eNOS mRNA were found (Figure S10c). Immunostainings of aortic cryosections revealed that eNOS+ (red) and eGFP+ (green) ECs were attached to the vascular wall in a circumferential manner (Figure S10d). The functional impact of this increased NO production on vascular tone was assessed in isometric force measurements. For this purpose aortic rings with MNP-containing eNOS ECs were D

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano

120 h after EC transduction, the neck region and several organs were examined for eGFP expression. Only if magnets had been applied was a strong eGFP+ signal found along the treated ACCs (Figure 5d), whereas no eGFP+ signals could be detected in the brain (Figure 5e), one of the main areas of blood supply of the carotid artery. In clear contrast, in animals without magnet application, no eGFP fluorescence could be detected in the ACC (Figure 5f), whereas prominent eGFP+ spots were found in the brains of these mice (Figure 5g). This distribution pattern was also confirmed by qRT-PCR analysis of the eGFP expression in the ACC and the brain (Figure 5h). EGFP+ ECs were still found in the ACC 7 days after the procedure. Because these control experiments demonstrated that the magnets were strong enough to retain magnetically labeled ECs in the ACC even under physiological flow conditions, we next treated injured vessels of eNOS−/− mice with eNOS ECs (100 000 cells per mouse). Two days later, the mice were sacrificed and the carotid arteries isolated and analyzed. Fluorescence microscopy of cryosections demonstrated that eGFP+ cells were distributed in a homogeneous fashion along the inner surface of denuded vessels and covered at least half of their circumference (Figure 6a). The grafted ECs were directly

Moreover, elevated NO production was confirmed by measuring eNOS activity (Figure 4d) and cGMP concentration (Figure 4e). Importantly, enhanced contraction of aortic rings in the presence of L-NAME underscored that the NO released by the reseeded eNOS-overexpressing ECs in vessels is vasoactive and regulates vascular tone ex vivo (Figure 4f−h). Next, we wondered whether this experimental strategy could also be effective for the treatment of vascular injury in vivo. To test this, we used the wire-injury model20 and removed the resident EC layer in carotid arteries of eNOS−/− mice. The carotid artery was chosen, because it is a predilection site for atherosclerotic lesions and invasive therapeutic procedures,21 resulting in an irreversible loss of the EC layer. The experimental in vivo situation in rodent models is characterized by restricted spatial access to the carotid artery. We therefore used a slightly altered magnet configuration generating a semiradial magnetic gradient field (Figure 5a−c) to ensure re-

Figure 6. Functional effects of EC replacement using eNOSoverexpressing cells in eNOS−/− mice. (a) Overlay of transmitted and fluorescence light picture of a section of eNOS−/− ACC after cell replacement with eNOS ECs and magnet on the ACC, bar = 50 μm. (b) Immunofluorescence staining of the consecutive section; blue = Hoechst, green = eGFP, red = eNOS, bar = 20 μm. (c) Original traces of isometric force measurements of eNOS−/− ACCs reseeded with eGFP ECs. (d) Original traces of isometric force measurements of eNOS−/− ACCs reseeded with eNOS ECs. (e) Ratios of constriction amplitudes with and without L-NAME in eNOS−/− ACCs reseeded with eNOS ECs compared to controls; the aortic rings are taken from at least 4 animals per group, **p < 0.01, ***p < 0.001. (e) One-way ANOVA, Tukey’s multiple comparison test.

Figure 5. EC replacement in the common carotid artery (ACC) of mice in vivo. (a) Metallic holder containing two groups of three bar magnets each resembling half of the magnet configuration used for ex vivo experiments. (b) Schematic diagram of magnet configuration used for in vivo experiments. (c) Gradient of magnetic flux density in a vessel generated by the magnet configuration shown in (b). (d, e) Transmitted (left) and fluorescence (right) light pictures of the neck region (d) and the brain (e) of a mouse after cell replacement with eGFP ECs and magnet on the ACC; bar = 1 mm. (f, g) Transmitted (left) and fluorescence (right) light pictures of the neck region (f) and the brain (g) of a mouse after cell replacement with eGFP ECs without magnet on the ACC; bar = 1 mm. (h) Quantitative RT-PCR of eGFP expression in ACCs (left) and in brains (right), ***p < 0.001. (h) Student’s t test.

attached to the denuded vascular wall and displayed preserved eGFP fluorescence up to a length of 2.6 mm within the ACC (Figure 6a), and immunostainings revealed strong eNOS signals in these cells (Figure 6b). To investigate the functional impact of our cell therapy on vascular tone, isometric force measurements of the ACCs with the protocol described above were performed. In eNOS−/− vessels treated with eNOS ECs significantly higher contractile responses (∼1.3-fold) were detected in the presence of L-NAME. This effect was absent in controls, where either no ECs or only eGFP ECs were applied (Figure 6c−e). These findings recapitulate our ex vivo results (Figure 2) and prove that even our semiradially

endothelialization of a large part of the circumference of the carotid artery. We tested the replacement strategy first with eGFP ECs in the wire-injury model and placed the magnets directly on top of the common carotid artery (ACC). ECs were injected into the ACC via the external carotid artery while blood flow was maintained in the ACC and the internal carotid artery. After 30 min the magnets were removed, the skin incision was closed, and the mice could recover. Two days later, E

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano symmetric cell replacement strategy with magnetic fields improves vascular function in vivo.

METHODS Animals. CD1 wild-type (WT) mice, C57BL/6 WT mice (both Charles River, Sulzfeld, Germany), or eNOS-deficient mice (eNOS−/−, kindly provided by Dr. A. Goedecke, University of Duesseldorf) were used. Animal housing and experiments were approved by the local ethics committee and carried out according to the guidelines of the German law of protection of animal life with approval by the local government authorities (Landesamt fü r Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen (LANUV), NRW, Germany). Numerical Simulations of Magnetic Fields. The calculations of the magnetic fields used for aortic perfusion experiments were based on a cylindrical model of the mouse aorta with an inner diameter of 0.75 mm and a length of 30 mm. These values correspond to the dimensions of the aorta in the ex vivo flow-loop system. To obtain a radially symmetric magnetic field gradient within the vessel, 12 cuboidal rare-earth permanent magnets (10 mm × 2 mm, neodymium iron boron (NdFeB), N35, 1.195 T; HKCM Engineering, Eckernförde, Germany) were arranged in four groups of three magnets. The four triplets are perpendicular to each other, and this configuration covers the whole length of the vessel. The magnets within the triplet have the same orientation, while the different triplets have an opposite alignment. Magnetic flux densities and the gradients of magnetic flux densities were calculated by means of finite element calculations with the AC/DC module of the package Comsol Multiphysics 4.3 (Comsol Multiphysics GmbH, Goettingen, Germany), as described before in detail.19 Lentiviral Vectors. Experiments were conducted with selfinactivating, VSV.G pseudotyped lentiviral vectors that were purified by ultracentrifugation of cell culture supernatant from transfected HEK293T producer cells as described before.32 Enhanced green fluorescent protein and human eNOS were cloned under control of the cytomegalovirus promoter into the vector constructs resulting in LV(eGFP), LV(eNOS), and LV(eNOS-IRES-eGFP). For detection of NO production by the fluorescence indicator diaminofluorescein LV(eNOS) was applied. In all other experiments LV(eNOS-IRESeGFP) was used. A lentivirus without transgene expression cassette LV(rrl) served as control virus. The infectious virus titer (infectious particles (IPs)/mL) was determined by transduction of HEK293T cells and flow cytometry analysis.32 The physical titer (virus particles (VP)/mL) was calculated based on the concentration of active viral reverse transcriptase as measured using an enzyme-linked immunosorbent assay (ELISA).23 Virus concentrations ranging from 2.7 × 108 to 1.6 × 1010 IP/mL and 3.2 × 1010 to 3.6 × 1011 VP/mL were used. Magnetic Nanoparticles. In this study PEI-Mag2 and SO-Mag5 MNPs of the core−shell type were used. The magnetic core is composed of magnetite. The core of PEI-Mag2 has a diameter of 9 nm and is coated with a self-assembling layer of fluorsurfactant ZONYL FSA and branched polyethylenimine with a molecular weight of 25 kDa.33 The core of SO-Mag5 has a diameter of 6.8 nm and is surrounded by a silica coating with surface phosphonate groups.34 The saturation magnetization of the core material at room temperature (RT) was 62 and 94 Am2/kg(Fe) for PEI-Mag2 and SO-Mag5 particles, respectively. The electrokinetic potential was ζ = +55.4 ± 1.6 mV for PEI-Mag2 and ζ = −38.4 ± 2 mV for SO-Mag5 as determined in H2O. Transduction and Magnetic Cell Labeling. For all experiments bovine artery endothelial cells were cultivated in endothelial cell growth medium and used until passage 10. Unless otherwise stated cells were seeded on a 24-well plate at a density of 20 000 cells/well for immunofluorescence stainings or on a six-well plate at a density of 200 000 cells/well for all other experiments. The transduction and magnetic cell labeling were performed similarly to already described protocols.16 LV/MNP complexes were generated by incubation of LVs and MNPs in 800 μL of Hank’s balanced salt solution containing magnesium and calcium (HBSS++) for 20 min at RT. Twelve hours after seeding the cells were transduced with LV/MNP complexes at 300 fg iron/IP (2.04 × 105 PEI-Mag2 particles/IP; 1.045 × 105 SOMag5 particles/IP) or 4000 fg iron/IP (1.3915 × 106 SO-Mag5 particles/IP) using a multiplicity of infection (MOI) of 50 and

CONCLUSION Taken together, our study demonstrates that the use of LV/ MNP complexes in combination with specifically designed magnetic fields enables radially symmetric EC replacement in vessels under physiologic blood flow. This approach has two distinct advantages, namely, site-specific positioning and prevention of thrombosis or ischemia because of preserved blood flow during the procedure. The strategy appears particularly promising, as we can genetically engineer the ECs to overexpress vasoactive genes,8,22 improving their survival and also their function. Our earlier work revealed that different MNP types are suited either for efficient virus binding or for cell magnetization.16,23 Since it is not possible to predict the behavior of an MNP type based on its physical properties (charge, size, coating), in the current study we have performed in vitro assays and identified SO-Mag5 particles that can be used for the transduction and the magnetization of ECs in a one-step procedure. Compared to a protocol with subsequent transduction and MNP loading, this saves time and is less stressful for the cells; in addition, SO-Mag5 particles did not induce toxic effects on EC biology and function. While stents have been shown to retain magnetic cells mainly at the stent struts,14 our approach provides homogeneous circumferential distribution of MNP-loaded cells at the vascular wall without implantable devices and their potential detrimental effects on the vascular wall. In addition, our myograph measurements of re-endothelialized vessels reveal improved vascular function ex vivo and in vivo, underscoring functional engraftment of the eNOS-overexpressing ECs. We have chosen lentiviral vectors for the ex vivo transduction of ECs, because of their efficient transduction of nondividing ECs and their stable long-term gene expression.24 For the application in humans, in particular for short-term therapies, also nonintegrating viral vectors or nonviral gene transfer shuttles such as chemically modified mRNAs could be tested.25,26 We have employed bPAECs for EC replacement, because this EC type could be easily transduced by lentiviral vectors, they efficiently take up MNPs, and they produce large amounts of NO after LV/ MNP-dependent eNOS overexpression.16 In addition, these cells are highly proliferative, enabling the detection of even minimal adverse effects of LV/MNP treatment. For the translation into humans either endothelial progenitor cells, human ECs derived from induced pluripotent stem cells (iPS), or ECs obtained by direct reprogramming could be used; all these cell types display a high regenerative potential27−29 and enable an autologous cell replacement strategy. As an alternative approach for the modulation of vascular function in future studies, it would be very interesting to explore the feasibility of MNP-based site-specific transduction of dysfunctional endothelium in vessels or the retention of NO donors30 at the inner surface of the vessel. Irrespective of the therapeutic agent used, for targeting larger human vessels also in deep tissues future efforts in the development of magnetic devices are required. For example a magnet configuration using a rotating pattern of permanent magnets according to the Halbach array could increase field gradients for large vessels in humans.31 This would help to extend our approach of sitespecific cell therapy using MNPs to almost any therapeutically relevant vascular bed within the body. F

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano Author Contributions

incubated on a magnetic plate (Chemicell, Berlin, Germany) for 30 min at 37 °C. Then, LV/MNP complexes were removed and the cells were incubated in medium for 3 or 5 days before experiments were performed. Ex Vivo Perfusion Experiments. For ex vivo perfusion experiments mouse aortas were used. Experiments were conducted as described previously.16 The vessels were isolated, and connective tissue was removed. Then, the endothelium of the aorta was removed with a stainless steel wire (diameter 0.8 mm for CD1 aortas and 0.6 mm for C57BL/6 aortas, HMC H.Meyer &Co. Spezialstahl GmbH, Duesseldorf, Germany). Intercostal arteries were cauterized using a soldering gun (WHS 40D, Weller Tools GmbH, Besigheim, Germany), and vessels were mounted on cannulae in an organ chamber filled with Dulbecco’s modified Eagle medium (DMEM) without supplements. Tubes or vessels were enclosed by a custommade magnet holder incorporating the magnets (see above). The cannulae were connected to a roller pump. Perfusion was performed with 200 000 transduced ECs (see above) in 10 mL of 37 °C warm DMEM for 30 min at a flow rate of 4.5 mL/min. Aortas were analyzed directly after perfusion. Pictures of the aortas were taken by a Leica MZ 16 F stereomicroscope (Microsystems GmbH, Wetzlar, Germany). In Vivo Cell Replacement. For in vivo cell replacement the mouse model of arterial wire-injury was used.20 Mice were anaesthetized using isoflurane. They received immunosuppression (cyclosporine A, 20 mg/kg, ip), analgesia, and antibiosis (carprofen, 5 mg/kg, sc, cefuroxime, 20 mg/kg, ip) prior to surgery. Denudation of the left common carotid artery was performed using a stainless steel wire with a diameter of 0.2 mm and a slightly angled tip (HMC H.Meyer & Co. Spezialstahl GmbH). For injection of cells an injection device composed of a 1 mL syringe (B. Braun Melsungen AG, Melsungen, Germany), a butterfly with a 27 G needle (B. Braun Melsungen AG), and a gel loading pipet (Eppendorf, Hamburg, Germany) was used. eGFP-overexpressing cells were injected into CD1 WT mice, and eNOS-overexpressing cells were injected into eNOS−/− animals. A total of 100 000 transduced ECs (4000 fg iron/IP, MOI 50) were injected in 200 μL of endothelial cell growth medium (Provitro, Berlin, Germany) via the external carotid artery into the common carotid artery while blood flow was maintained via the internal carotid artery. In some experiments a custom-made magnet holder incorporating six cuboidal rare-earth permanent bar magnets corresponding to magnet configuration A (height = 10 mm, width = 2 mm, NdFeB, N35, HKCM Engineering) was placed above the blood vessel for 30 min. Then the external carotid artery was tied off, the skin incision was closed, and the mice recovered. To prevent cell rejection mice were daily treated with cyclosporine A. At day 2 or 7 after surgery mice were further analyzed. Statistical Analysis. Data are expressed as mean ± SEM. Student’s t test or one-way analysis of variance (ANOVA) with Tukey’s post hoc test was performed. Differences were considered statistically significant when p < 0.05.

S.V. and S.R. performed in vitro, ex vivo, and in vivo cell replacement experiments and analyzed the data, A.H. and B.G. performed and analyzed numerical simulations, K.Z. and A.P. designed and produced lentiviral vectors, O.M. and C.P. generated MNPs and performed and analyzed magnetic responsiveness measurements, W.B. performed and analyzed electron microscopy experiments, D.E. quantified MNPs, and B.K.F. and D.W. designed the study, analyzed the data, and wrote the paper. S. Vosen and S. Rieck contributed equally. Notes

The authors declare the following competing financial interest(s): C.P. is a co-founder and shareholder of OZ Biosciences S.A., a company selling reagents for magnetofection.

ACKNOWLEDGMENTS We thank A. Gödecke, University of Düsseldorf, Germany, for providing eNOS−/− mice and C. Rudolph for helpful discussions. The work was supported by funding to the junior research group “Magnetic nanoparticles (MNPs)endothelial cell replacement in injured vessels” by the Ministry of Innovation, Science, Research and Technology of the State of North Rhine-Westphalia (D.W.) and the DFG Research Unit FOR 917 Nanoguide (K.Z., C.P., B.G., A.P., B.K.F.). REFERENCES (1) Bonetti, P. O.; Lerman, L. O.; Lerman, A. Endothelial Dysfunction: a Marker of Atherosclerotic Risk. Arterioscler., Thromb., Vasc. Biol. 2003, 23, 168−175. (2) Pauletto, P.; Sartore, S.; Pessina, A. C. Smooth-Muscle-Cell Proliferation and Differentiation in Neointima Formation and Vascular Restenosis. Clin. Sci. 1994, 87, 467−479. (3) Patel, S. D.; Waltham, M.; Wadoodi, A.; Burnand, K. G.; Smith, A. The Role of Endothelial Cells and Their Progenitors in Intimal Hyperplasia. Ther. Adv. Cardiovasc. Dis. 2010, 4, 129−141. (4) Kipshidze, N.; Dangas, G.; Tsapenko, M.; Moses, J.; Leon, M. B.; Kutryk, M.; Serruys, P. Role of the Endothelium in Modulating Neointimal Formation: Vasculoprotective Approaches to Attenuate Restenosis After Percutaneous Coronary Interventions. J. Am. Coll. Cardiol. 2004, 44, 733−739. (5) Myers, P. R.; Webel, R.; Thondapu, V.; Xu, X. P.; Amann, J.; Tanner, M. A.; Jenkins, J. S.; Pollock, J. S.; Laughlin, M. H. Restenosis Is Associated With Decreased Coronary Artery Nitric Oxide Synthase. Int. J. Cardiol. 1996, 55, 183−191. (6) Thompson, M. M.; Budd, J. S.; Eady, S. L.; James, R. F.; Bell, P. R. A Method to Transluminally Seed Angioplasty Sites With Endothelial Cells Using a Double Balloon Catheter. Eur. J. Vasc. Surg. 1993, 7, 113−121. (7) Conte, M. S.; Birinyi, L. K.; Miyata, T.; Fallon, J. T.; Gold, H. K.; Whittemore, A. D.; Mulligan, R. C. Efficient Repopulation of Denuded Rabbit Arteries With Autologous Genetically Modified Endothelial Cells. Circulation 1994, 89, 2161−2169. (8) Yla-Herttuala, S.; Alitalo, K. Gene Transfer As a Tool to Induce Therapeutic Vascular Growth. Nat. Med. 2003, 9, 694−701. (9) von der Leyen, H. E.; Dzau, V. J. Therapeutic Potential of Nitric Oxide Synthase Gene Manipulation. Circulation 2001, 103, 2760− 2765. (10) Yla-Herttuala, S. Cardiovascular Gene Therapy With Vascular Endothelial Growth Factors. Gene 2013, 525, 217−219. (11) Chorny, M.; Fishbein, I.; Yellen, B. B.; Alferiev, I. S.; Bakay, M.; Ganta, S.; Adamo, R.; Amiji, M.; Friedman, G.; Levy, R. J. Targeting Stents With Local Delivery of Paclitaxel-Loaded Magnetic Nanoparticles Using Uniform Fields. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 8346−8351.

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.5b04996. Material and Methods and supplementary figures (PDF)

AUTHOR INFORMATION Corresponding Authors

*(B. K. Fleischmann) Tel: xx49/228/6885/200. Fax: xx49/ 228/6885/201. E-mail: bernd.fl[email protected]. *(D. Wenzel) Tel: xx49/228/6885/216. Fax: xx49/228/6885/ 201. E-mail: [email protected]. Present Address △

(B.K.F. and D.W.) Institute of Physiology I, University of Bonn, Sigmund-Freud-Straße 25, 53127 Bonn, Germany. G

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX

Article

ACS Nano (12) Chorny, M.; Fishbein, I.; Tengood, J. E.; Adamo, R. F.; Alferiev, I. S.; Levy, R. J. Site-Specific Gene Delivery to Stented Arteries Using Magnetically Guided Zinc Oleate-Based Nanoparticles Loaded With Adenoviral Vectors. FASEB J. 2013, 27, 2198−2206. (13) Pislaru, S. V.; Harbuzariu, A.; Gulati, R.; Witt, T.; Sandhu, N. P.; Simari, R. D.; Sandhu, G. S. Magnetically Targeted Endothelial Cell Localization in Stented Vessels. J. Am. Coll. Cardiol. 2006, 48, 1839− 1845. (14) Polyak, B.; Fishbein, I.; Chorny, M.; Alferiev, I.; Williams, D.; Yellen, B.; Friedman, G.; Levy, R. J. High Field Gradient Targeting of Magnetic Nanoparticle-Loaded Endothelial Cells to the Surfaces of Steel Stents. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 698−703. (15) Hofmann, A.; Wenzel, D.; Becher, U. M.; Freitag, D. F.; Klein, A. M.; Eberbeck, D.; Schulte, M.; Zimmermann, K.; Bergemann, C.; Gleich, B.; Roell, W.; Weyh, T.; Trahms, L.; Nickenig, G.; Fleischmann, B. K.; Pfeifer, A. Combined Targeting of Lentiviral Vectors and Positioning of Transduced Cells by Magnetic Nanoparticles. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 44−49. (16) Wenzel, D.; Rieck, S.; Vosen, S.; Mykhaylyk, O.; Trueck, C.; Eberbeck, D.; Trahms, L.; Zimmermann, K.; Pfeifer, A.; Fleischmann, B. K. Identification of Magnetic Nanoparticles for Combined Positioning and Lentiviral Transduction of Endothelial Cells. Pharm. Res. 2012, 29, 1242−1254. (17) Kilgus, C.; Heidsieck, A.; Ottersbach, A.; Roell, W.; Trueck, C.; Fleischmann, B. K.; Gleich, B.; Sasse, P. Local Gene Targeting and Cell Positioning Using Magnetic Nanoparticles and Magnetic Tips: Comparison of Mathematical Simulations With Experiments. Pharm. Res. 2012, 29, 1380−1391. (18) Dames, P.; Gleich, B.; Flemmer, A.; Hajek, K.; Seidl, N.; Wiekhorst, F.; Eberbeck, D.; Bittmann, I.; Bergemann, C.; Weyh, T.; Trahms, L.; Rosenecker, J.; Rudolph, C. Targeted Delivery of Magnetic Aerosol Droplets to the Lung. Nat. Nanotechnol. 2007, 2, 495−499. (19) Heidsieck, A.; Vosen, S.; Zimmermann, K.; Wenzel, D.; Gleich, B. Analysis of Trajectories for Targeting of Magnetic Nanoparticles in Blood Vessels. Mol. Pharmaceutics 2012, 9, 2029−2038. (20) Lindner, V.; Fingerle, J.; Reidy, M. A. Mouse Model of Arterial Injury. Circ. Res. 1993, 73, 792−796. (21) Erickson, K. M.; Cole, D. J. Carotid Artery Disease: Stenting Vs Endarterectomy. Br. J. Anaesth. 2010, 105 (Suppl 1), i34−i49. (22) Sato, J.; Mohacsi, T.; Noel, A.; Jost, C.; Gloviczki, P.; Mozes, G.; Katusic, Z. S.; O’Brien, T.; Mayhan, W. G. In Vivo Gene Transfer of Endothelial Nitric Oxide Synthase to Carotid Arteries From Hypercholesterolemic Rabbits Enhances Endothelium-Dependent Relaxations. Stroke 2000, 31, 968−975. (23) Trueck, C.; Zimmermann, K.; Mykhaylyk, O.; Anton, M.; Vosen, S.; Wenzel, D.; Fleischmann, B. K.; Pfeifer, A. Optimization of Magnetic Nanoparticle-Assisted Lentiviral Gene Transfer. Pharm. Res. 2012, 29, 1255−1269. (24) Nayerossadat, N.; Maedeh, T.; Ali, P. A. Viral and Nonviral Delivery Systems for Gene Delivery. Adv. Biomed. Res. 2012, 1, 27. (25) Sen, S.; Strappe, P. M.; O’Brien, T. Gene Transfer in Endothelial Dysfunction and Hypertension. Methods Mol. Med. 2005, 108, 299−314. (26) Kormann, M. S.; Hasenpusch, G.; Aneja, M. K.; Nica, G.; Flemmer, A. W.; Herber-Jonat, S.; Huppmann, M.; Mays, L. E.; Illenyi, M.; Schams, A.; Griese, M.; Bittmann, I.; Handgretinger, R.; Hartl, D.; Rosenecker, J.; Rudolph, C. Expression of Therapeutic Proteins After Delivery of Chemically Modified MRNA in Mice. Nat. Biotechnol. 2011, 29, 154−157. (27) Wassmann, S.; Werner, N.; Czech, T.; Nickenig, G. Improvement of Endothelial Function by Systemic Transfusion of Vascular Progenitor Cells. Circ. Res. 2006, 99, e74−e83. (28) Zampetaki, A.; Kirton, J. P.; Xu, Q. Vascular Repair by Endothelial Progenitor Cells. Cardiovasc. Res. 2008, 78, 413−421. (29) Tan, K. S.; Tamura, K.; Lai, M. I.; Veerakumarasivam, A.; Nakanishi, Y.; Ogawa, M.; Sugiyama, D. Molecular Pathways Governing Development of Vascular Endothelial Cells From ES/IPS Cells. Stem Cell Rev. 2013, 9, 586−598.

(30) Gao, M.; Liu, S.; Fan, A.; Wang, Z.; Zhao, Y. Nitric oxidereleasing graft polymer micelles with distinct pendant amphiphiles. RSC Adv. 2015, 5, 67041−67048. (31) Halbach, K. Design of Permanent Multipole Magnets with Oriented Rare Earth Cobalt Materials. Nucl. Instrum. Methods 1980, 169, 1−10. (32) Pfeifer, A.; Hofmann, A. Lentiviral Transgenesis. Methods Mol. Biol. 2009, 530, 391−405. (33) Mykhaylyk, O.; Antequera, Y. S.; Vlaskou, D.; Plank, C. Generation of Magnetic Nonviral Gene Transfer Agents and Magnetofection in Vitro. Nat. Protoc. 2007, 2, 2391−2411. (34) Mykhaylyk, O.; Sobisch, T.; Almstatter, I.; Sanchez-Antequera, Y.; Brandt, S.; Anton, M.; Doblinger, M.; Eberbeck, D.; Settles, M.; Braren, R.; Lerche, D.; Plank, C. Silica-Iron Oxide Magnetic Nanoparticles Modified for Gene Delivery: a Search for Optimum and Quantitative Criteria. Pharm. Res. 2012, 29, 1344−1365.

H

DOI: 10.1021/acsnano.5b04996 ACS Nano XXXX, XXX, XXX−XXX