Intraarterial route increases the risk of cerebral lesions after mesenchymal cell administration in animal model of ischemia

1 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 www.nature.com/scientificreports Intraarterial route increases the risk of cerebral lesions after mesenchymal cell administration in animal model of ischemia Bárbara Argibay1, Jesse Trekker2,3, Uwe Himmelreich3, Andrés Beiras4, Antonio Topete5,6, Pablo Taboada5, María Pérez-Mato1, Alba Vieites-Prado1, Ramón Iglesias-Rey1, José Rivas7, Anna M. Planas8,9, Tomás Sobrino1, José Castillo1 & Francisco Campos1 Mesenchymal stem cells (MSCs) are a promising clinical therapy for ischemic stroke. However, critical parameters, such as the most effective administration route, remain unclear. Intravenous (i.v.) and intraarterial (i.a.) delivery routes have yielded varied outcomes across studies, potentially due to the unknown MSCs distribution. We investigated whether MSCs reached the brain following i.a. or i.v. administration after transient cerebral ischemia in rats, and evaluated the therapeutic effects of both routes. MSCs were labeled with dextran-coated superparamagnetic nanoparticles for magnetic resonance imaging (MRI) cell tracking, transmission electron microscopy and immunohistological analysis. MSCs were found in the brain following i.a. but not i.v. administration. However, the i.a. route increased the risk of cerebral lesions and did not improve functional recovery. The i.v. delivery is safe but MCS do not reach the brain tissue, implying that treatment benefits observed for this route are not attributable to brain MCS engrafting after stroke. Ischemic stroke, caused by interruption of the blood supply to the brain, is one of the most important causes of morbidity and mortality worldwide. Currently, the control of systemic parameters, such as body temperature, blood pressure, and glycemia, has considerably improved the outcome of stroke patients. However, in the absence of protective therapy, an early artery reperfusion, i.e. mechanical or enzymatic thrombolysis, remains the primary goal of treatment for acute ischemic stroke1,2. Cell based therapies have emerged as a promising tool for the treatment of both acute and delayed phases of stroke. In this regard, mesenchymal stem cells (MSCs) are one of the best candidates for stem cell therapy of ischemic stroke owing to their multipotentiality, ability to release growth factors, and immunomodulatory capacities3. Thus, this transdifferentiation can produce cells with a neural lineage4–7, induce neurogenesis8–10, angiogenesis8–10 and synaptogenesis11, and activate endogenous restorative processes through production of cytokines and trophic factors8,12–14. Moreover, the regulation of cerebral blood flow (CBF), the blood brain barrier (BBB)12, and other neuroprotective mechanisms, such as the reduction of apoptosis, inflammation, demyelination, and increased astrocyte survival8,9,15,16, have been involved as beneficial mechanisms of MSCs after of stroke3. 1Clinical Neurosciences Research Laboratory, Clinical University Hospital, Health Research Institute of Santiago de Compostela (IDIS), Universidade de Santiago de Compostela, Santiago de Compostela, Spain. 2IMEC, Department of Life Science Technology, Leuven 3001, Belgium. 3Biomedical MRI, Department of Imaging and Pathology, KU Leuven, Leuven 3000, Belgium. 4Department of Morphological Sciences, Universidade de Santiago de Compostela, Santiago de Compostela, Spain. 5Grupo de Física de Coloides y Polímeros, Departamento de Física de la Materia Condensada, Universidade de Santiago de Compostela, Santiago de Compostela, Spain. 6Departamento de Fisiología, Centro Universitario de Ciencias de la Salud, Universidad de Guadalajara, Guadalajara 44340, México. 7Applied Physics Department, Campus Vida, Universidade de Santiago de Compostela, Santiago de Compostela, Spain. 8Department of Brain Ischemia and Neurodegeneration, Institut d’ Investigacions Biomèdiques de Barcelona (IIBB), Consejo Superior de Investigaciones Científicas (CSIC), Barcelona, Spain. 9August Pi i Sunyer Biomedical Research Institute (IDIBAPS), Barcelona, Spain. Correspondence and requests for materials should be addressed to J.C. (email: [email protected]) or F.C. (email: [email protected]) Received: 31 August 2016 accepted: 09 December 2016 Published: 16 January 2017 OPEN www.nature.com/scientificreports/ 2 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 Functional recovery in animal models of focal cerebral ischemia has been observed when MSCs were injected intravenously (i.v.) or intraarterially (i.a.)17–20, however, there is not agreement yet about the optimal administration route. Intravenous injections are minimally invasive, and cell tracking studies following that route have shown that most administered cells remain trapped in the lungs, liver, and spleen21, indicating that a reduced number of cells reach the brain22. Intraarterial administration is a promising strategy to direct the majority of injected cells to the brain23, but the fate of injected cells following this route remains unknown due to high variation in the reported results. Indeed, recent studies have shown that approximately 21% of the cells delivered via i.a. carotid injection were observed in the ipsilateral hemisphere24. Conversely, other studies have reported that 24 h after injection, 95% of the delivered cells were found in the spleen25. Additional studies have indicated that i.a. carotid cell administration is a safe delivery strategy that can overcome limitations of i.v. administration, since it represents a more direct route; however, new findings have associated a higher mortality to i.a. administration compared to the i.v. route21. Despite of the discrepancies about the best route for cell administration, it has also not been well established if the therapeutic effect described for MCSs after i.a. and i.v. administration requires the diffusion of cells through the BBB and the engraftment in the cerebral parenchyma tissue. Therefore, an in vivo analysis of the cellular fate and biodistribution of both administration routes is an important and necessary step towards the further development of minimally invasive stem cell therapy for central nervous system diseases, including stroke. To this end, the objective of this study was to perform an analysis of cell tagging by magnetic resonance imaging (MRI) contrast agents (CAs) and subsequent MRI analysis to address this challenge26. In this study, first, we synthesized dextran-coated superparamagnetic nanoparticles (D-MNPs), validated their use as CAs for cell tracking in MRI, and evaluated the cellular viability of MSCs after labeling, including their detection by MRI. Secondly, the optimal route and cell dosage were evaluated for i.a. administration. Third, cellular biodistribution patterns following i.a. and i.v. administration were investigated. Finally, the therapeutic effects of MSCs administered through either route were compared in an animal model of ischemic stroke. Results Synthesis and characterization of D-MNPs. D-MNPs were synthesized in the presence of dextran following the chemical co-precipitation method described in the Methods section. Transmission electron microscopy (TEM) micrographs (Fig.1A) showed a mean core size of 3.7 ± 0.8 nm. The core crystal structure determined by X-ray diffraction (XRD) (Fig.1B) showed peaks at 2θ positions of ca. 30.2°, 35.6°, 43.2°, 57.1°, and 62.7°, corresponding to the (220), (311), (400), (511), and (440) planes of magnetite, respectively, with a lattice parameter of 8.33 ± 0.02 Å and a crystallite size of 4.8 ± 0.5 nm derived from the Scherrer equation. Vibrating sample magnetometer (VSM) measurements showed that cores exhibited superparamagnetic behavior at room temperature (Fig.1C). The hydrodynamic size of D-MNPs measured by dynamic light scattering (DLS) was 94 ± 3 nm with Z-potential values of ca. − 11 ± 3 mV. The presence of the polymeric coating was confirmed by thermal gravimetric analysis (TGA) and fourier-transform infrared (FTIR). The TGA spectrum (Fig.1D) showed two peaks, corresponding to mass losses at approximately 280 °C and 322 °C, in the range of dextran decomposition temperature (150–380 °C)27, demonstrating that about 66 wt% of the D-MNPs can be attributed to dextran. The FTIR spectrum (Fig.1E) exhibited several polysaccharide-characteristic absorption bands at 3351 cm−1 due to O-H stretching; vibrational modes ν (C-H) and δ (C-H) corresponding to 2906 cm−1, 1420 cm−1, and 1380 cm−1; water molecular bending assigned to 1620 cm−1, 1145 cm−1; and 1010 cm−1 peaks due to C-O vibrations; and small peaks at 908 cm−1, 877 cm−1, and 746 cm−1, corresponding to α -glucopyranose ring deformation modes. There were also two absorption bands at 578 cm−1 and 430 cm−1, corresponding to Fe-O vibration modes28,29. MRI contrast in T2-weighted images at different Fe concentrations (0.09 mmol/L; 0.05 mmol/L, and 0.02 mmol/L) of D-MNPs were measured (Fig.1F). A linear relationship was found between R2(R2 = 1/T2) and MNP concentration, resulting in a transverse relaxivity (r2) of 701 ± 16 L mmol−1 s−1. Characterization of mesenchymal stem cells labeling with D-MNPs. The analysis of the cell properties showed that D-MNP labeling did not reduce the proliferation rate of MSCs with respect to the control when evaluated as cell counts at P0 (12 h), P1 (2 days), and P2 (5 days) after labeling (Fig.2A). No significant differences were observed between D-MNPs and the control in terms of cell viability determined by lactate dehydrogenase (LDH) assay (Fig.2A). Vascular endothelial growth factor (VEGF) release was not affected by D-MNP internalization with respect to control cells (Fig.2A) at the different time-points analyzed (P0, P1, and P2). The amount of internalized iron oxide nanoparticles was determined in labeled cells by inductive coupled plasma optical emission spectroscopy (ICP-OES) (Fig.2A). Internalized iron at P0 was 7.7 ± 0.7 pg Fe/cell, decreasing to 2.4 ± 0.2 pg Fe/cell at P1, and 0.90 ± 0.09 pg Fe/cell at P2. MRI analysis of labeled cells and non-labeled cells was also performed. T2*-maps were calculated from T2*-weighted images. R2* relaxivities of 500 cells/μ L in agar were obtained for each condition (Fig.2A). At P0, the R2* values were 36.0 ± 1.2 s−1 and 68.0 ± 0.9 s−1 for the control and D-MNPs, respectively; at P1, these values were 21 ± 4 s−1 and 31.8 ± 1.1 s−1 for the control and D-MNPs, respectively; and at P2, these values were 19 ± 2 s−1 and 19.6 ± 0.8 s−1 for the control and D-MNPs, respectively, according to the decrease in internalized iron observed at the different timepoints. Transmission electron microscopy (TEM) images of labeled MSCs showed D-MNPs engulfed in cellular compartments and distributed along the cell (Fig.2B). No D-MNPs were observed attached at the surface of the cells. In addition, the matrigel assay did not show differences in angiogenic capacity between control and D-MNP-labeled cells at P0 and P2 in terms of ring formation and distribution (Fig.2C). Finally, flow cytometry analysis revealed no differences in the membrane phenotype markers of MSCs (CD90+ , CD73+ and CD45− ) between D-MNPs and non-labeled MSCs at P0 and P2 (Fig.2D). No differences between cell passages 9 and 17 were observed regarding to VEGF release, ring formation or membrane phenotype markers. Moreover, analysis of labeled cells with increasing amounts of D-MNPs did not affect cellular size (size average in suspension 15 ± 2 μ m). www.nature.com/scientificreports/ 3 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 Intra-arterial cell tracking of D-MNPs MSCs. To optimize i.a. cell delivery, previously we tested various combinations of surgical procedures in an animal model of cerebral ischemia as it is detailed in Fig.3. In all the Figure 1. Characterization of D-MNPs. (A) TEM micrograph showing the morphology and size distribution of D-MNP cores; (B) The XRD spectrum of D-MNP cores synthesized through chemical co-precipitation shows the characteristic peaks of magnitude at 2θ positions of ca. 30.2°, 35.6°, 43.2°, 57.1°; and 62.7°; (C) Magnetization curve of D-MNPs demonstrates superparamagnetic behavior at room temperature; (D) TGA spectrum of D-MNPs and (E) FTIR spectrum of D-MNPs provide evidence for the presence of dextran in the D-MNP formulation; (F) T2weighted MRI of 0.09 mM, 0.05 mM, and 0.02 mM D-MNPs. Iron concentrations demonstrate the suitability of D-MNPs as contrast agents (CA) for MRI analysis. www.nature.com/scientificreports/ 4 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 Figure 2. Time elapsed study of MSCs labeled with D-MNPs. (A) Cell count after labeling at P0, P1, and P2 compared to the control. Viability after labeling at P0, P1, and P2 compared to the control as determined by LDH assay. VEGF secretion of D-MNPs and the control at P0, P1, and P2. Iron content per cell at P0, P1, and P2, as determined by ICP. Relaxation times (R2*) of labeled and non-labeled cells at P0, P1, and P2. T2*-maps of control and D-MNP labeled cells at P0, P1, and P2. (B) TEM of a MSC labeled with D-MNPs appearing as dispersed black regions within the cellular cytoplasm. Magnification in one region indicates evidence of D-MNP encapsulation in endosomes. (C) Matrigel-angiogenic assay in Control and D-MNP labeled MSCs at P0 and P2. (D) Flow cytometry analysis of Control and D-MNP-labeled MSCs at P0 and P2 for CD90/CD73/ CD45 analysis. P0 (12 h), P1 (3 days) and P2 (5 days) after D-MNP labeling. www.nature.com/scientificreports/ 5 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 experimental conditions MSC administration (1 × 106 MNP labeled MSCs suspended in 300 μ L) was carried out 4 hours after transient middle cerebral artery occlusion (tMCAo). In administration protocol A (Fig.3A), a high variation was observed in cerebral MSC detection. MSC administration following procedure A resulted in brain engraftment in only 3 out of 8 animals, and no associations with mortality were found. In this procedure, the ipsilateral common carotid (Ip-CC) remained closed from the beginning of tMCAo that could limit the entrance of labeled cells after i.a. administration. To overcome this effect, in protocol B (Fig.3B), cerebral blood flow in the Ip-CC was reestablished at the time of administration. In this experimental condition, the Ip-CC was opened during the injection, ensuring high engraftment efficiency, but most of the animals died (6 out of 8 rats died within 24 h of cell delivery). This high mortality could be due to secondary intracerebral hemorrhage since the Ip-CC was closed for 4 h prior to injection, which could induce vessel damage. Figure 3. Several combinations of i.a. MSCs delivery were considered in this study to elucidate the most efficient and safe administration, here the schemes of different experimental procedures for tMCAo and i.a. cannulation for brain stem cell delivery. Black arrows denote differences between the indicated case and the previous case. CASE A: The Ip-CC and Con-CC were closed during tMCAo; immediately after reperfusion of the MCA, the con-CC was also reperfused. MSCs were administered after 4 h through the Ip-EC with the Ip-CC closed. After administration, the Ip-EC and Ip-CC remained closed. CASE B: The Ip-CC and Con-CC were closed during tMCAo; immediately after reperfusion of the MCA, the con-CC was perfused as well. MSCs were administered after 4 h through the Ip-EC with the Ip-CC opened. After administration, the Ip-EC remained closed, while the Ip-CC remained open. CASE C: The Ip-CC and Con-CC were closed during tMCAo; immediately following reperfusion of the MCA, both CCs were perfused. MSCs were administered after 4 h later through the Ip-EC with the Ip-CC opened. After administration, the Ip-EC remained closed and the Ip-CC remained open. CASE D: The Con-CC remained intact during all procedures. Immediately after reperfusion of the MCA, the con-CC was reperfused as well. MSCs were administered after 4 h through the Ip-EC with the Ip-CC opened. After administration, the Ip-EC remained closed, while the Ip-CC remained open. See SupplementalFigureS1 for arterial nomenclature. N = 8 per experimental case. The figure was produced, in part, by using Servier Medical Art, (www.servier.com/Powerpoint-image-bank). www.nature.com/scientificreports/ 6 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 In protocol C (Fig.3C), the Ip-CC was opened after tMCAo to prevent the possible vessel damage caused by prolonged occlusion. Under these conditions, all the animals showed hypointense signals in MRI, but exhibited difficulties in breathing when they awoke from anesthesia, and 4 out of 8 rats died within the next 5 days. Coronal MR T2*-weighted images showed dispersed cells along the forebrain and also in the cerebellum suggesting cell trafficking through the posterior communicating artery, which connects the vertebro-basilar system with the carotid system (diagram of cerebrovascular anatomy is represented in the SupplementaryFigureS1). After occlusion of both common carotids (Fig.3A–C), blood flow to the forebrain is provided by the vertebro-basilar system through the posterior communicating arteries, which suffer vasodilatation. After cell injection and under these conditions, the regional localization of cells along the brain and cerebellum depends on the CBF distribution in the circle of Willis. To test whether blood supply through the common carotid (Con-CC) could prevent cell migration towards the cerebellum, in protocol D (Fig.3D) this artery was not closed during tMCAo. The results showed uniform cerebral cell distribution and no mortality. Typical brain engraftment of D-MNP labeled MSCs could be observed in T2*-weighted images (Fig.4A) after this delivery route. All these results are summarized in the SupplementalTableS1. Figure 4. (A) First echo of MR T2* weighted image of one brain slice of a Wistar rat 8 h after the onset of cerebral ischemia, and 4 h after i.a. administration of 1 × 106 MSCs labeled with D-MNPs in case D. Labeled cells can be observed as black punctate patterns in the right hemisphere of the animal. (B) Cerebral blood flow in three regions of the middle cerebral artery determined in healthy animals by 3 Doppler probes before, during, and after administration of PBS, 0.25 × 106 non-labeled MSCs, and 1 × 106 non-labeled MSCs; MR T2-weighted images of several brain slices 24 h after cell delivery. Multifocal ischemia is observed all along the brain after administration of 1 × 106 MSCs (indicated with yellow arrows). After the administration of 0.25 × 106 MSCs, an isolated micro-ischemia was observed in one animal (indicated with a yellow arrow). No edema (middle line shift) was observed for all conditions studied. N = 6 per experimental group. www.nature.com/scientificreports/ 7 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 To determine the safety of different cell doses in i.a. administration procedure D, two cell doses, 0.25 × 106 and 1 × 106 non-labeled MSCs in PBS (used as the cell vehicle) were injected in healthy animals and were compared to vehicle administration. CBF was monitored during delivery, and brain was evaluated 24 h after administration by MR T2-weighted images. MR images showed absence of lesions in the cerebral tissue of rats receiving PBS (Fig.4B). In contrast, administration of 1 × 106 MSCs induced multifocal lesions in all brain slices (Fig.4B), and injection of 0.25 × 106 MSCs produced only one isolated lesion in one of the animals without evidence of brain edema. Multifocal lesions caused by the 1 × 106 dose of MSCs were also associated with a reduction of CBF determined by laser Doppler flowmetry. Accordingly, we limited the cell delivery in this study to 0.25 × 106 MSCs as a safe dose for i.a. administration. To further study the distribution of MSCs and to determine the possible cause of multifocal lesions observed previously, healthy animals were treated (procedure D) with 1 × 106 MSCs labeled with D-MNPs and CFSE (Cell Proliferation Kit is used for in vivo labeling of cells) for MRI and histology detection, respectively. T2*-weighted images showed D-MNPs labeled MSCs in the brain of all animals. Immunohistological analysis revealed positive staining for nuclei (Hoechst), MSCs (CFSE), and vessels (CD31) mainly in the ipsilateral hemisphere along all cerebral sections. CFSE positive MSCs were mainly observed in small vessels (Fig.5A,B) but not in wide vessels, and we found no evidence of cells extravasated to the brain parenchyma. Immunohistological analysis of lungs and heart did not reveal CFSE positive MSCs. In line with these results, TEM analysis showed MSCs in small vessels (Fig.6), however, labeled MSCs were also detected in the brain parenchyma (Fig.7). MSCs were identified readily due to their contrasting cytoplasmic color compared to the surroundings and scattered dark endosomes within the cellular membrane. Magnification in one region showed a dark, punctate pattern inside the endosome membrane cristae, with similar distribution, size, and morphology to those previously observed in vitro (Fig.2B). Adjacent to the endosomes, mitochondria were also identified by their cristae, demonstrating the viability of MSCs after injection. Intravenous cell tracking of MSCs. Cell tracking after i.v. administration of 1 × 106 MSCs labeled with D-MNPs and CFSE showed no cerebral engraftment, noted as hypointensities in MR T2*-weighted images, in any of the rats. This in vivo result was confirmed by histological analysis. No positive CFSE staining for MSCs was found in the brain and heart by using fluorescence optical microscopy, whereas positive CFSE-labeled cells were found to be distributed along the lungs (SupplementalFigureS2). Neurorecovery study. Once the optimal injection protocol and MSC dosage for safe i.a. administration and the distribution of MSCs after i.v. and i.a. delivery in rats were established, we aimed to evaluate the neurorecovery effect produced by i.a. and i.v. administration of MSCs after cerebral ischemia. D-MNP labeled MSCs were detected readily in brain by MR T2*-weighted images at 4 h, 24 h, and 3 days after i.a. injection only (SupplementalFigureS3). Infarct volume analysis revealed that both administration route did not reduce the infarct volume within 14 days after tMCAo, compared to the groups receiving the vehicle. In fact, lesion size significantly increased after i.a. delivery of PBS or MSCs with respect to i.v. treatments (Fig.8A,B). The administration of MSCs, either i.a. or i.v., did not reduce edema volume in the acute phase of cerebral ischemia. Moreover, a significant increase in edema volume was observed 24 h after i.a. administration (*P < 0.05) (Fig.8C). Functional recovery after MSC delivery was also evaluated by the cylinder test, and no significant differences were observed between groups (Fig.8D). Nonetheless, the group of rats receiving i.v. MSCs showed a reduction of the use of the impaired forelimb, but the difference did not reach statistical significance compared to the i.v. control group. The plasma concentrations of VEGF at different time points were below the limit of detection of the ELISA assay used in this study. The presence of neural progenitors in the subventricular region was examined with doublecortin (DCX) and Ki-67 immunostaining (Fig.9A). Co-localization of DCX and Ki-67 was higher for MSC administered groups compared to controls, particularly for DCX staining, a marker of neuronal precursor cells and immature neurons (Fig.9A and C). Infarct and peri-infarct regions were examined to evaluate angiogenesis through co-localization of CD31 and Ki-67, but no differences in marker staining were observed between groups (Fig.9B and C). Discussion MSCs have received special attention in recent years as a promising therapeutic candidate for stroke due to their potential multi-mechanistic effects on the damaged brain30. However, parameters such as the therapeutic window and cell dosage or administration route are still under discussion31,32. For a better understanding of how stem cell therapies can succeed, non-invasive monitoring of cells is required, and MRI is one of the most appropriate tools for this use because it provides excellent soft tissue contrast and high resolution33–35. MRI can provide anatomical and pathological information and, by tagging cells with MRI CAs such as superparamagnetic nanoparticles, MRI can establish the relationship between cellular biodistribution and outcome after stroke. Indeed, the fate and biodistribution of injected MSCs after stroke is a critical aspect for therapeutic efficacy; therefore, the administration route is a decisive parameter. The most common routes of MSC transplantation are i.v. or i.a. infusion, or direct intra-tissue injection36. Although intraparenchymal (i.p.) administration provides direct cell delivery to the lesion, it results in poor lesion distribution and is highly invasive due to the necessary craniotomy23. Many studies have explored i.a. and i.v. administration in animal models of ischemic stroke with different degrees of success in terms of outcome37–39 and brain engraftment23,40–42, but only a few have established a relationship between them43–45. In this study, we administered MSCs labeled with D-MNPs to address the relationship between cell allocation and animal recovery after stroke. Superparamagnetic iron oxide nanoparticles are the most frequently used CAs for cell tracking owing to the high hypointense MR signal in T2 and T2*-weighted images at low nanoparticle concentrations46. Several methods have been described for synthesizing these particles47. However, chemical co-precipitation is the easiest, most www.nature.com/scientificreports/ 8 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 economical, and efficient in terms of producing a large amount of nanoparticles48,49. Our synthesis resulted in well-coated, size-uniformed, slightly negatively charged and homogeneously dispersed nanoparticle suspensions in aqueous solutions, similar to other nanoparticles previously used28,29,50–52. D-MNPs exhibited superparamagnetic behavior at room temperature, reflecting their suitability as a general MRI CA. Although these nanoparticles can be used directly as a CA for MRI53–55, in vivo cell tracking requires nanoparticle internalization into the cells, which is a challenging procedure, as both are negatively charged56. To overcome this pitfall, Poly-L-lysine (PLL) (see methods for cell labeling with D-MNPs) was used because it has been widely accepted for nanoparticle cell tagging based on its low toxicity profile and high iron content per cell at low incubation concentrations57–61. Indeed, labeling was highly efficient, as particles were found inside the cells when observed by TEM, reducing the risk of unspecific contrast and possible MRI misinterpretation59,62. Quantitatively, the internalization of D-MNPs at P0 is in accordance with previous studies which have demonstrated the requirement of PLL for efficient cell labeling with commercial and in-house-synthesized dextran coated nanoparticles, obtaining comparable nanoparticle internalizations, up to 20 pg/cell59,63,64. However, since the administered cells Figure 5. Histological cell localization after intraarterial (i.a.) administration of CFSE labeled MSCS. Positive staining of injected MSCs can be seen in the small diameter vessels (A and B) but not in big vessels (A). CFSE is co-localized with CD31 and Hoechst. (Scale bar 100 μ m). www.nature.com/scientificreports/ 9 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 must survive and interact with other cells in the host, an in vitro study of long-term detection and stem cell biology was necessary. Thus, labeled cells were seeded sequentially at P1 (2days) and P2 (5days), and their internalized iron and consequently T2*-MRI signals were found to be reduced. However, D-MNP-labeled MSCs were still detected by MRI immediately after labeling (P0) and 2 days later (P1). The MRI signal from labeled cells at P2 could not be distinguished from the control; the lower detectability limit for labeled cells has been reported previously to be 2–5 pg Fe/cell61,65. This dilution factor over time is related to the in vitro proliferation rate of MSCs (approximately two cell divisions, each lasting 24 h), which suggests that other cell types with therapeutic potential but different proliferation ratios could be labeled for longer or shorter periods of time. Despite the indispensable MRI detection capacities of D-MNPs, cellular viability after labeling must be guaranteed for further effective therapy. One of the most relevant mechanisms of action of MSCs includes the release of VEGF, which promotes the development of fibroblasts, endothelial cells, and tissue progenitor cells, and performs tissue regeneration and repair66. In this study, labeled and non-labeled cells secreted equal levels of VEGF immediately after labeling and also 5 days after tagging, and MSCs maintained angiogenic capacity for tubular formation. Unchanged surface markers CD45−, CD90+, and CD73+ measured by flow cytometry67–70 after 5 days support the biocompatible nature of D-MNPs for MSC tagging. Figure 6. Electron transmission micrograph of rat brain cortex 4 h after intra-arterial (i.a.) delivery of D-MNP-labeled MSCs. (A) Dilated brain vessel surrounded by neuropil (Scale bar 5 μ m). (B) Magnification of vessel dilation in (A). Red blood (RB) corpuscles are on the right, and platelets(PTs) and MSCs can be observed inside the vessel (Scale bar 2 μ m). (C) Magnification of the upper vessel expansion in (B). One well-defined MSC is identified by the encapsulated D-MNPs (black punctate encapsulated pattern). (Scale bar 1 μ m). (D) Longitudinal section of a brain vessel where two D-MNPs labeled MSCs can be observed. Encapsulated D-MNPs are noted by dark punctuate patterns (Scale bar 2 μ m). Figure 7. Transmission electron micrograph of rat brain cortex 4 h after intra-arterial (i.a.) delivery of D-MNP-labeled MSCs. (A) One single MSC labeled with D-MNPs was found in the brain parenchyma surrounded by neuropil (NP) (Scale bar 2 μ m). (B) Magnification of the region selected in A, where D-MNPs are observed as black punctate regions compartmentalized within a membrane. Three mitochondria (MT) can be noted as well (Scale bar 0.5 μ m). (C) One single MSC labeled with D-MNPs was found in the brain parenchyma surrounded by neuropil (NP) and by a neuron (NEU) (Scale bar 2 μ m). (D) Magnification of the region selected in C, where D-MNPs can be distinguished as black punctate regions compartmentalized within the membrane (Scale bar 0.5 μ m). www.nature.com/scientificreports/ 16 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 55. Tassa, C., Shaw, S. Y. & Weissleder, R. Dextran-coated iron oxide nanoparticles: a versatile platform for targeted molecular imaging, molecular diagnostics, and therapy. Acc Chem Res 44, 842–852 (2011). 56. Kim, D. Y. et al. Effects of the Surface Charge of Stem Cell Membranes and DNA/Polyethyleneimine Nanocomplexes on Gene Transfection Efficiency. Journal of Biomedical Nanotechnology 11, 522–530 (2015). 57. Arbab, A. S. et al. Comparison of transfection agents in forming complexes with ferumoxides, cell labeling efficiency, and cellular viability. Mol Imaging 3, 24–32 (2004). 58. Yu, M. X., Chen, W. L., Zhou, Q. & Gao, P. Study on ASTC-a-1 cells labeled with superparamagnetic iron oxide and its magnetic resonance imaging. Exp Biol Med (Maywood) 235, 1053–1061 (2010). 59. Trekker, J. et al. Sensitive in vivo cell detection using size-optimized superparamagnetic nanoparticles. Biomaterials 35, 1627–1635 (2014). 60. Roeder, E. et al. Dose-response of superparamagnetic iron oxide labeling on mesenchymal stem cells chondrogenic differentiation: a multi-scale in vitro study. PLoS One 9, e98451 (2014). 61. Struys, T. et al. Magnetic resonance imaging of human dental pulp stem cells in vitro and in vivo. Cell Transplant 22, 1813–1829 (2013). 62. Ruehm, S. G., Corot, C., Vogt, P., Kolb, S. & Debatin, J. F. Magnetic resonance imaging of atherosclerotic plaque with ultrasmall superparamagnetic particles of iron oxide in hyperlipidemic rabbits. Circulation 103, 415–422 (2001). 63. Zhang, H. In Molecular Imaging and Contrast Agent Database (MICAD) (2004). 64. Ketkar-Atre, A. et al. Variability in contrast agent uptake by different but similar stem cell types. Int J Nanomedicine 8, 4577–4591 (2013). 65. Crabbe, A. et al. Effects of MRI contrast agents on the stem cell phenotype. Cell Transplant 19, 919–936 (2010). 66. Ma, S. et al. Immunobiology of mesenchymal stem cells. Cell Death Differ 21, 216–225 (2014). 67. Dominici, M. et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 8, 315–317 (2006). 68. Boxall, S. A. & Jones, E. Markers for characterization of bone marrow multipotential stromal cells. Stem Cells Int 2012, 975871 (2012). 69. Chamberlain, G., Fox, J., Ashton, B. & Middleton, J. Concise review: mesenchymal stem cells: their phenotype, differentiation capacity, immunological features, and potential for homing. Stem Cells 25, 2739–2749 (2007). 70. Jones, E. & McGonagle, D. Human bone marrow mesenchymal stem cells in vivo. Rheumatology 47, 126–131 (2008). 71. Lu, D. et al. Intraarterial administration of marrow stromal cells in a rat model of traumatic brain injury. J Neurotrauma 18, 813–819 (2001). 72. Guo, L. et al. A novel method for efficient delivery of stem cells to the ischemic brain. Stem Cell Res Ther 4, 116 (2013). 73. Riou, A. et al. MRI assessment of the intra-carotid route for macrophage delivery after transient cerebral ischemia. NMR Biomed 26, 115–123 (2013). 74. Du, S. et al. Intra-arterial delivery of human bone marrow mesenchymal stem cells is a safe and effective way to treat cerebral ischemia in rats. Cell transplantation 23 Suppl 1, S73–82 (2014). 75. Mitkari, B. et al. Intra-arterial infusion of human bone marrow-derived mesenchymal stem cells results in transient localization in the brain after cerebral ischemia in rats. Experimental neurology 239, 158–162 (2013). 76. Ishizaka, S. et al. Intra-arterial cell transplantation provides timing-dependent cell distribution and functional recovery after stroke. Stroke; a journal of cerebral circulation 44, 720–726 (2013). 77. Santillan, A. et al. Cannulation of the internal carotid artery in mice: a novel technique for intra-arterial delivery of therapeutics. Journal of neuroscience methods 222, 106–110 (2014). 78. Fagan, S. C., Lapchak, P. A., Liebeskind, D. S., Ishrat, T. & Ergul, A. Recommendations for preclinical research in hemorrhagic transformation. Transl Stroke Res 4, 322–327, doi: 10.1007/s12975-012-0222-5 (2013). 79. Janowski, M. et al. Cell size and velocity of injection are major determinants of the safety of intracarotid stem cell transplantation. J Cereb Blood Flow Metab 33, 921–927 (2013). 80. Villalobos, V. et al. Ultrastructural changes of the olfactory bulb in manganese-treated mice. Biocell 33, 187–197 (2009). 81. Liu, L. et al. From blood to the brain: can systemically transplanted mesenchymal stem cells cross the blood-brain barrier? Stem Cells Int 2013, 435093 (2013). 82. Schrepfer, S. et al. Stem cell transplantation: the lung barrier. Transplant Proc 39, 573–576 (2007). 83. Iihoshi, S., Honmou, O., Houkin, K., Hashi, K. & Kocsis, J. D. A therapeutic window for intravenous administration of autologous bone marrow after cerebral ischemia in adult rats. Brain Res 1007, 1–9 (2004). 84. Zhang, L. et al. Delayed administration of human umbilical tissue-derived cells improved neurological functional recovery in a rodent model of focal ischemia. Stroke 42, 1437–1444 (2011). 85. Zhang, L. et al. Intravenous administration of human umbilical tissue-derived cells improves neurological function in aged rats after embolic stroke. Cell Transplant (2012). 86. Ramos-Cabrer, P., Justicia, C., Wiedermann, D. & Hoehn, M. Stem cell mediation of functional recovery after stroke in the rat. PLoS One 5, e12779 (2010). 87. Suzuki, J. et al. Bilateral cortical hyperactivity detected by fMRI associates with improved motor function following intravenous infusion of mesenchymal stem cells in a rat stroke model. Brain Res 1497, 15–22 (2013). 88. Chopp, M. & Li, Y. Treatment of stroke and intracerebral hemorrhage with cellular and pharmacological restorative therapies. Acta Neurochir Suppl 105, 79–83 (2008). 89. Zhang, Z. G. & Chopp, M. Neurorestorative therapies for stroke: underlying mechanisms and translation to the clinic. Lancet Neurol 8, 491–500 (2009). 90. Zhang, Z. G. et al. VEGF enhances angiogenesis and promotes blood-brain barrier leakage in the ischemic brain. J Clin Invest 106, 829–838 (2000). 91. Choi, J. H. et al. Cell proliferation and neuroblast differentiation in the rat dentate gyrus after intrathecal treatment with adiposederived mesenchymal stem cells. Cellular and molecular neurobiology 31, 1271–1280 (2011). 92. Yoo, S. W. et al. Mesenchymal stem cells promote proliferation of endogenous neural stem cells and survival of newborn cells in a rat stroke model. Exp Mol Med 40, 387–397 (2008). 93. Chen, J. et al. Intravenous administration of human bone marrow stromal cells induces angiogenesis in the ischemic boundary zone after stroke in rats. Circ Res 92, 692–699 (2003). 94. Chen, J. et al. Neuroprotective effect of human placenta-derived cell treatment of stroke in rats. Cell Transplant 22, 871–879 (2013). 95. Ankrum, J. & Karp, J. M. Mesenchymal stem cell therapy: Two steps forward, one step back. Trends in molecular medicine 16, 203–209 (2010). 96. Li, Y. & Chopp, M. Marrow stromal cell transplantation in stroke and traumatic brain injury. Neuroscience letters 456, 120–123 (2009). 97. Eckert, M. A. et al. Evidence for high translational potential of mesenchymal stromal cell therapy to improve recovery from ischemic stroke. J Cereb Blood Flow Metab 33, 1322–1334 (2013). 98. Longa, E. Z., Weinstein, P. R., Carlson, S. & Cummins, R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20, 84–91 (1989). www.nature.com/scientificreports/ 17 Scientific RepoRts | 7:40758 | DOI: 10.1038/srep40758 Acknowledgements This study has been partially supported by grants from Axencia Galega de Innovación (Xunta de Galicia), the Instituto de Salud Carlos III (PI13/00292; PI14/01879), the Spanish Research Network on Cerebrovascular Diseases RETICS-INVICTUS (RD12/0014), Xunta de Galicia (Consellería Educación GRC2014/027), the European Commission program FEDER and Promoting Active Ageing program: Functional Nanostructures For Alzheimer’s Disease At Ultra-Early Stages” (Pana_686009), a Research and Innovation Project, funded within the EU Horizon 2020 Programme”. Furthermore, this study was also co-funded within the POCTEP (Operational Programme for Cross-border Cooperation Spain-Portugal) program (0681_INVENNTA_1_E), co-financed by the ERDF (European Regional Development Fund). T. Sobrino (CP12/03121) and F. Campos (CP14/00154) are recipients of a research contract from Miguel Servet Program of Instituto de Salud Carlos III. Finally, P. Taboada thanks Mineco and Xunta de Galicia for funding through projects MAT2013-40971-R and EM2013-046, respectively. J Trekker is the recipient of an innovation grant from the IWT-Vlaanderen. The funders had no role in the study design, data collection, and analysis, decision to publish, or preparation of the manuscript. Author Contributions B.A.: Conception and design; collection and/or assembly of data; manuscript writing, interpretation. J.T.: Collection and/or assembly of data and manuscript writing. U.H.: Conception and design; data analysis, interpretation and manuscript writing. A.B.: Collection and/or assembly of data and manuscript writing. A.T.: Collection and/or assembly of data. P.T.: Collection and/or assembly of data, data analysis, interpretation and manuscript writing. M.P.-M.: Collection and/or assembly of data. A.V.-P.: Collection and/or assembly of data. R.I.-R.: Collection and/or assembly of data and manuscript writing. J.R.: Manuscript writing and manuscript writing. A.P.: Manuscript writing; data analysis and interpretation and manuscript writing. T.S.: Manuscript writing; data analysis and interpretation and manuscript writing. J.C.: Conception and design; data analysis and interpretation; final approval of manuscript. F.C.: Conception and design, data analysis and interpretation; manuscript writing; final approval of manuscript. Additional Information Supplementary information accompanies this paper at http://www.nature.com/srep Competing financial interests: The authors declare no competing financial interests. How to cite this article: Argibay, B. et al. Intraarterial route increases the risk of cerebral lesions after mesenchymal cell administration in animal model of ischemia. Sci. Rep. 7, 40758; doi: 10.1038/srep40758 (2017). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ © The Author(s) 2017