A novel specific aptamer targets cerebrovascular endothelial cells after ischemic stroke

Cell specific-targeted therapy (CSTT) for acute ischemic stroke remains underdeveloped. Cerebrovascular endothelial cells (CECs) are key components of the blood–brain barrier and are the first brain cells affected by ischemic stroke. After stroke, CEC injury causes insufficient energy supply to neurons and leads to cytotoxic and vasogenic brain edema. Aptamers are short single-stranded RNA or DNA molecules that can bind to specific ligands for cell specific delivery. The expression of vascular cell adhesion molecule-1 (VCAM-1) is increased on CECs after stroke. Herein, we report that an RNA-based VCAM-1-aptamer can specifically target CECs in stroke brains following transient middle cerebral artery occlusion in mice. Our data demonstrate the potential of an RNA-based aptamer as an effective delivery platform to target CECs after stroke. We believe this method will allow for the development of CSTT for treatment of patients with stroke.

www.nature.com/scientificreports/ fluorescence is enriched in ischemic hemispheres (Fig. 1B) and the mean fluorescence intensity (MFI) of Cy5 was significantly higher in mice treated with VCAM-1-aptamer compared with the PBS group and the control aptamer group in ischemic hemispheres (Fig. 1C). Peripheral organs (heart, liver, spleen, and kidney) did not show enriched Cy5 fluorescence (Fig. 1D). Confocal microscopy further confirmed that Cy5 fluorescence overlapped with CD31 + CECs in VCAM-1-aptamer treated stroke mice (Fig. 1E). Quantified relative expression of Cy5 in CD31 + CECs demonstrated that the MFI of Cy5 was significantly higher in stroke brains of VCAM-1-aptamer treated stroke mice (Fig. 1F). These data suggest that the VCAM-1-aptamer specifically targets CECs in stroke mice.

VCAM-1-aptamer does not protect against stroke damage.
To test if this VCAM-1-aptamer affects stroke outcomes, we performed tMCAO in a separate cohort of mice and injected VCAM-1-aptamer (0.5 nmol) compared to PBS control ( Fig. 2A). By analyzing cerebral blood flow (CBF) at four timepoints (prior to tMCAO, 5 min post-tMCAO, 5 min post-reperfusion, and 1-h post-intravenous administration) we found that VCAM-1-aptamer had no significant effect on CBF in both male and female mice (Fig. 2B). In male mice, one mouse from each group died prior to the 22-h experimental endpoint. Males and females had similar degrees of neurological deficits in both the PBS and VCAM-1-aptamer groups (Fig. 2C). Infarct volume did not differ in PBS and VCAM-1-aptmer treated groups when stratified by sex (Fig. 2D).  www.nature.com/scientificreports/ We further evaluated if a higher dose of VCAM-1-aptamer affects stroke outcomes. We administered 2.5 nmol of VCAM-1-aptamer (with or without Cy5 label) to male stroke mice at 6 h post-stroke compared to PBS control (Fig. 3A). While two stroke mice died in PBS treated group and one stroke mouse from each aptamer treated group died prior to the 22-h endpoint, their neurological deficits were not significantly different among these three groups (Fig. 3B). Quantified infarct volumes did not significantly differ either (Fig. 3C). The data have also demonstrated that Cy5 conjugate does not impact neurological deficits (P > 0.99) and infarct volumes (P = 0.95) in stroke mice (Fig. 3B,C). Taken together, these data suggest that the VCAM-1-aptamer does not significantly affect mortality, functional deficits, or infarct volume in stroke mice. www.nature.com/scientificreports/

Discussion
Aptamers have several favorable features including high specificity and binding affinity with specific proteins, small size, low immunogenicity, relatively low cost for production, and ease of store 10 . Therefore, aptamers are one of the best candidates for targeting the specific ligands or receptors that have been applied to therapeutic, biosensing, and engineering fields 10 . The selected VCAM-1-aptamer has high affinity for N-terminal two-domain fragment of VCAM-1, suggesting a potential inhibitory activity on their targets, either by directly blocking VCAM-1 or indirectly blocking its ligand α4β1 integrin expressed on leukocytes to bind VCAM-1 on CECs. While our results have shown high specificity of this VCAM-1-aptamer with CECs in ischemic hemispheres (Fig. 1), we unexpectedly did not observe a significant impact of this aptamer (with or without Cy5 conjugate) on stroke outcomes (Figs. 2, 3), which represents a barrier to developing this aptamer into therapeutic applications. However, our data demonstrated it has low toxicity in stroke mice, suggesting that this aptamer could be a useful and safe tool to target CECs following ischemic stroke. It was previously reported that VCAM-1 has negligible expression evident in normal brain tissue 11 , but is markedly elevated in CECs after stroke and involved in inflammatory and immune responses 3 , which are implicated in the evolution of infarction. Given the high specificity of the VCAM-1-aptamer to CECs shown in our study, it could be utilized as a tool to gauge the degree of inflammation and tissue damage, possibly reflecting the severity of stroke. In vivo imaging of VCAM-1 expression using molecular imaging techniques such as positron emission tomography (PET) or magnetic resonance imaging (MRI) may, therefore, allow for the non-invasive diagnosis and monitoring of stroke and inflammatory diseases. Previous studies have investigated antibodies against VCAM-1. Aniti-VCAM-1 antibody conjugated with a contrast agent has been used as a molecular imaging tool to monitor vascular inflammation in stroke rats 7 . Despite the promising results of anti-VCAM-1 antibody conjugates as molecular imaging tools, the risk of immunogenicity has posed a challenge in the field. Our study has demonstrated herein that the VCAM-1-aptamer does not influence cerebral blood flow, neurological deficits, and infarct volume in the murine tMCAO model. There was no dose response effect on infarct volume when mice received 0.5 nmol (Fig. 2) compared with 2.5 nmol of aptamer (Fig. 3) (P = 0.68). The VCAM-1-aptamer conjugated to Cy5 fluorescence did not significantly impact stroke outcomes either (Fig. 3). It may therefore offer a better option for conjugation to a biosensor for developing molecular MRI, PET, or other in vivo imaging systems.
The extracellular domain of VCAM-1 contains seven Ig-like domains 12 . Ig-like domain pairs of 1 and 4, 2 and 5, and 3 and 6 are highly homologous with each other 12 . Domains 1 and 4 of VCAM-1 are involved in the direct binding of α4β1 integrin, resulting in leukocyte adhesion 12,13 . The VCAM-1-aptamer used in this study is very small (39 oligonucleotides) and only corresponds to VCAM-1 domains 1 and 2, which might not be able to block binding α4β1 integrin and reduce leukocyte adhesion. This could explain why the aptamer failed to protect against stroke damage (Figs. 2, 3) while showing high affinity for CECs (Fig. 1) in our stroke model. Conjugating the VCAM-1-aptamer with other proteins and peptides could potentially block VCAM-1 from binding to α4β1 integrin, offering new strategy to develop novel therapeutics. It remains to be verified if aptamers corresponding to other domains of VCAM-1 could offer therapeutic effects for stroke.
The expression of VCAM-1 on CECs is elevated not only in neurological disorders including stroke, subarachnoid hemorrhage, brain tumor, multiple sclerosis, and traumatic brain injury 14,15 but also involved in various immunological disorders, such as rheumatoid arthritis, asthma, transplant rejection, etc. 16 . Our study warrants further investigation and development as in vivo targeting platform for CECs, and we believe that the VCAM-1-aptamer may be applied to numerous applications in the future. www.nature.com/scientificreports/

Methods
Guidelines. All animal experiments were approved by the Animal Care and Use Committee at the University of Texas Health Science Center at Houston (accredited by the American Association for Accreditation of Laboratory Care) and performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Both male and female C57/BL6J mice (~ 3 to 6 months old) were used in this study. The ARRIVE guidelines (Animal Research: Reporting of In Vivo Experiments) were followed to report animal experiments.

Transient middle cerebral artery occlusion (tMCAO). Focal cerebral ischemia was induced under
isofluorane anesthesia with standard operation procedures. The mice were subjected to 1 h transient middle cerebral artery occlusion (tMCAO) using silicon coated sutures (Cat. #702334, diameter 0.23 mm, Doccol Corporation, MA) followed by reperfusion for 22 h as published previously 1,4,17 . Body temperature was controlled at 37 ± 0.5 °C during occlusion. Experimenters were blinded for data analysis. The inclusion/exclusion criteria were followed as published previously 1,4,17 . Bupivacaine (2 mg/kg, s.c.) was administered to relieve pain after surgery. Immunohistochemistry staining. Brains were fixed in 10% paraformaldehyde for 2 days then in 30% sucrose for 2 days. Brain sections (20 μm) were obtained on Leica cryostat (− 20 °C) then blocked with 5% goat serum staining buffer and stained with anti-mouse-CD31 (R&D systems, Minnesota, USA) overnight, then washed with PBS and stained with FITC-conjugated goat-anti-rabbit (Jackson Immunoresearch, Inc. Pennsylvania, USA) for 2 h. Brain sections were washed with PBS and mounted on glass slides using prolong gold antifade reagent (Life technologies, California, USA). The slides were photographed with a Nikon A1Rsi Confocal microscope (Nikon, Japan) using the Nis Elements software.

Cerebral blood flow.
Neurological deficits. Neurological deficits were determined at 22 h post-tMCAO according to a 0-to 5-point scale neurological score system as published 1,4,17 . The experimenters were blinded to treatments.
Brain infarct volume. The brain sections were stained with 2% Tetranitroblue tetrazolium chloride (TTC, Cat. #T4000, Sigma, Saint Louis, MO) at 37 °C for 30 min then photographed by an image scanner (CanoScan 9000F, Japan). Infarct volume was analyzed using ImageJ (National Institutes of Health) in a double-blinded manner. The infarct volume was expressed as a percentage of contralateral hemisphere.

Statistical analysis.
We performed statistical analysis using Prism 5 software (Graphpad software, California, USA). Differences between two groups were analyzed using unpaired Student's t-tests or Mann-Whitney U tests, one-way analysis of variance (ANOVA), or two-way ANOVA as indicated in the figure legends. P < 0.05 was statistically significant.

Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.