Toward development of a dynamic supramolecular peptide therapy for acute ischemic stroke

Abstract

Acute ischemic stroke (AIS) treatment relies on early restoration of blood flow; however, ischemia/reperfusion (I/R) may lead to secondary brain injury. Supramolecular peptide assemblies in which many molecules move collectively by design can activate key cellular pathways by displaying bioactive molecules on their surfaces. In this study, we hypothesized that a highly dynamic assembly formed by a peptide amphiphile (PA) that displays the laminin-mimetic sequence IKVAV (IKVAV-PA), known to promote neuron survival, could be delivered systemically, reach the ischemic brain, and exert therapeutic effects following AIS. C57BL/6 heterozygous CX3CR1^GFP mice underwent 60-min of transient middle cerebral artery occlusion and were administered IKVAV-PA or saline (control) immediately after reperfusion. IKVAV-PA presence and distribution was evaluated by intracranial intravital and wide-field imaging. Cresyl violet staining was performed to quantitate final brain infarct volume at 7 days post stroke. IKVAV-PA formed scaffolds that contain both nanoscale filaments in equilibrium with small micellar aggregates, which is a signature of enhanced epitope dynamicity. Systemically administered IKVAV-PA crossed the blood-brain barrier and was primarily detected within the ischemic hemisphere. Cresyl violet staining demonstrated IKVAV-PA treatment significantly reduced infarct size when compared to saline treated animals. Histological screening of systemic organs suggested good biocompatibility of IKVAV-PAs at 7 days post stroke. We demonstrated the therapeutic potential of systemically delivering IKVAV-PA in a pre-clinical model of ischemic stroke. This work lays the foundation for further studies utilizing supramolecular PA assemblies as an adjunct therapy to reperfusion therapies in order to enhance long-term tissue-level neural regeneration post stroke.

Graphical abstract

Introduction

Acute ischemic stroke (AIS) is a leading cause of death and disability worldwide, with over 500,000 patients affected each year in the United States alone [1]. Current therapies for AIS rely mainly on blood flow restoration, including use of thrombolytics such as Tenecteplase, and mechanical thrombectomy with clot retrieval. Though these approaches have improved survival rates and clinical outcomes [[2], [3], [4]], many AIS patients still suffer from significant disability and long-term cognitive impairment [5]. A growing body of evidence attributes such decline to the occurrence of secondary neurologic injury following AIS [6]. This results from a cascade of biochemical and inflammatory events following ischemia/reperfusion (I/R) injury in the brain [7]. To date, there is no effective therapy that targets key contributors of secondary injury following reperfusion in AIS. Significant advancements in the fields of nanotechnology and regenerative medicine are now primed to address a significant therapeutic unmet need for long-term disability caused by AIS.

There have been previous reports on nanomaterials with therapeutic potential for AIS [8,9], which include the use of self-assembling peptides [10]. Peptide amphiphiles (PAs), explored here for the first time as an AIS therapy, are self-assembling molecules composed of “modified” peptides that can arrange into a variety of nanostructures – fibers, ribbons, sheets, and spheres – depending on their amino acid sequences and ionic environment [[11], [12], [13], [14], [15]]. Given their versatility, the supramolecular assemblies with a large variety of signals have been used to treat many disorders and stimulate tissue repair [[16], [17], [18], [19], [20], [21]]. These include enhancement of vascularization [22], central [17,18] and peripheral [23] nervous system repair, and GLP-1 agonists to enhance insulin production [24]. It was recently shown that these bioactive systems, when designed to display a laminin-mimetic pentapeptide signal IKVAV with high degrees of supramolecular motion, can effectively promote neuron survival as well as axonal extension after a severe injury in vivo in the spinal cord [19]. The PA assembly used therein, the IKVAV-PA, was also recently shown to activate integrin β-1 in human neurons [19,20]. Collectively, the neuroprotective effect of IKVAV-PAs was enhanced by controlling dynamicity of molecules in their assemblies. Furthermore, a recent paper reported enhanced fluorescence in PA nanostructures that are highly dynamic and this may be useful in pre-clinical work to track them within the central nervous system [25]. To date, the therapeutic effects of IKVAV-PA supramolecular systems have not been investigated in the context of AIS.

In the present work, we explored the potential of utilizing bioactive IKVAV-PA supramolecular nanostructures with highly dynamic epitopes as a systemically delivered treatment to augment reperfusion therapy in AIS. Experiments involved the use of a well-validated murine transient middle cerebral artery occlusion (tMCAO) model of ischemia/reperfusion. This AIS model is most representative of large-vessel occlusion occurring in patients subsequently treated with reperfusion therapies such as thrombolytics and mechanical thrombectomy. Augmenting reperfusion therapy with systemically delivered, highly dynamic IKVAV-PA thus has significant translational potential for AIS treatment for patients.

Methods

IKVAV-PA synthesis

IKVAV-PA and IKVAV-AZDye™ 647 PAs were synthesized using standard Fmoc solid-phase peptide synthesis on Rink amide MBHA resin. A CEM Liberty Blue Microwave-assisted Peptide Synthesizer was used to carry out amino acid couplings. Each amino acid and palmitic acid were coupled using 4 molar equivalents of protected amino acid, 8 molar equivalents of ethyl cyanohydroxyiminoacetate (Oxyma), and 4 molar equivalents of N,N′-di-isopropylcarbodiimide (DIC) in N,N-dimethylformamide (DMF) for 2–4 min at 90 °C. Fmoc groups were cleaved using 20 % 4-methylpiperidine and 0.1 M hydroxybenzotriazole (HOBt) in N,N-dimethylformamide (DMF) at 90 °C for 30 s. A solution of 95:2.5:2.5 trifluoroacetic acid (TFA)/triisopropylsilane (TIPS)/water was added to the reaction vessel for 2–3 h to cleave PA molecules off the resin and to deprotect amino acid sidechains. Volatile solvents were concentrated with rotary evaporation, after which PA molecules were precipitated with cold diethyl ether. Crude PA material was then dried on a fritted filter. Preparative scale reverse phase high performance liquid chromatography (Shimadzu Prominence or Waters Prep 150) using a Phenomenex Gemini column (C-18 stationary phase, 5 μm, 100 Å pore size, either 30 × 150 mm or 50 × 250 mm) was used to purify PAs. A mobile phase of acetonitrile and water was used, containing 0.1 % NH₄OH if the sample was acidic or 0.1 % TFA if the sample was basic. Pure fractions were identified using electrospray ionization mass spectrometry (ESI-MS) in positive mode on an Agilent model 6520 Quadrupole Time-of-Flight (Q-ToF) using direct injection. MassHunter Workstation Data Acquisition software was used for instrument operation and MassHunter Qualitative Analysis software was used for data analysis and processing. Excess acetonitrile was removed by rotary evaporation following purification. The samples were then freeze-dried, and the resulting PA powders were stored at −20 to −30 °C.

To conjugate PA with AZDye™ 647, an additional azidolysine residue was appended to IKVAV-PA at the C-terminal, followed by coupling with AZDye™ 647-DBCO using click chemistry. The final product, IKVAV-AZDye™ 647, was purified as described above.

IKVAV-PA preparation for in vivo studies

An overview of PA preparation for in vivo experiments is summarized in Fig. 1d. Briefly, lyophilized PA powders were first sterilized by UV irradiation inside of a cell culture hood and then dissolved in 0.9 % NaCl solution (Ricca Chemical Company, Arlington TX, USA) at a concentration of 5 mg/mL. PA solutions were pH adjusted to approximately 7.4–7.6 by adding 1 M NaOH (Millipore Sigma, Cat. No. 109137, Milwaukee, WI, USA) in 1–2 μL increments. The resulting solution was bath sonicated for 30 min. To make the fluorescently labeled PA assembly, 1 mol% IKVAV-AZDye™ 647 PA was mixed with IKVAV PA at desired volumes and bath sonicated for another 30 min. Following sonication, pH was adjusted again to approximately 7.4–7.6. All PA solutions were aged overnight at room temperature prior to use. All pH adjustments were monitored using Fisherbrand pH Test Paper (Cat. No. 13-640-502, Waltham, MA, USA).

Fig. 1.

Chemical and supramolecular structure of IKVAV-PA supramolecular assemblies. a) Chemical structure of IKVAV-PA. b) Molecular graphics illustration of IKVAV-PA self-assembly into filaments and small aggregates. c) Negative-stained transmission electron micrograph (TEM) of IKVAV-PA supramolecular assemblies. The scale bar corresponds to 500 nm. d) Schematic representation of PA preparation, transient middle cerebral artery occlusion (tMCAO), and retro-orbital administration [70].

Negative stain transmission electron microscopy

Immediately prior to grid preparation, PA solutions were diluted to 1 mM concentration. Samples were drop-cast onto the shiny side of grids (300 mesh copper with 5–6 nm carbon film, Electron Microscopy Sciences, Hatfield, PA, USA) and then negatively stained with a 2 wt% solution of uranyl acetate. Samples were imaged either with a FEI Eagle 4k CCD camera on a FEI Tecnai Spirit G2 microscope or with a JEOL 1400 Flash microscope, both containing LaB6 filaments at 120 kV accelerating voltage.

Animal experimental setup

All animal housing and procedures were performed in accordance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals (Assurance Number A328301). The procedures were approved by Northwestern University Institutional Animal Care and Use Committee (IACUC). For experimental stroke studies we selected 18 female heterozygous B6 CX3CR-1^GFP mice, aged 12–16 weeks, with average body weights between 20 and 28 g. CX3CR1^GFP mice were a gift from Harris Perlmann (Northwestern University, Chicago, IL, USA) and have been described previously [26]. Catchup^IVM mice were a generous gift from Mathias Gunzer (Institute for Analytical Science, Dortmund, Germany) [27] and used with adoptive bone marrow transfer for separate intravital microscopy (IVM) experiments.

Adoptive bone marrow transfer

Bone marrow chimeras were generated to allow real-time visualization of recruited tdTomato + neutrophils and resident microglial cells (CX3CR1^GFP+) in response to systemic injection of fluorescently labeled IKVAV-PA and assess for potential co-localization. Heterozygous C57BL/6 CX3CR1^{GFP +} mice were lethally irradiated with 1000-cGy dose using a gammacell 40 exactor137Cs irradiator (Best Theratronics, Ottawa, Canada). Mice were reconstituted with bone marrow from Catchup^IVM mice. All mice strains were confirmed to be heterozygous for all alleles. Recipients were allowed to recover for one month for complete bone marrow adoption, as per standard protocols used [28,29], and intracranial IVM experiments were performed following tMCAO. Reconstitution was confirmed with fluorescence imaging of blood smear, and flow cytometry as previously indicated [30].

Transient middle cerebral artery occlusion (tMCAO) surgical procedure and validation

All mice underwent tMCAO under aseptic surgical technique and continuous monitoring as previously described [[30], [31], [32]], concordant with ARRIVE 2.0 experimental guidelines [33]. Briefly, mice were anesthetized with isoflurane (1–2 %), body temperature was maintained using a heating system, and lubricant was applied on both eyes. Buprenorphine (0.05 mg/kg, s. c.) was administered to minimize surgical pain. A midline ventral incision was made, and thyroid and connective tissue were separated to expose the left common carotid artery. Following ligation of the external carotid artery and proximal common carotid artery, weight-matched silicon-coated monofilaments (Doccol 602223/602323, MA, USA) were inserted into an incision made within the common carotid artery and advanced to the level of the middle cerebral artery bifurcation and secured in place with suture. The neck incision was closed, and mice were monitored and recovered post anesthesia. Intra-ischemic behavioral assessments were recorded, and a behavioral score consisting of 4 points (contralateral circling behavior, contralateral curling when lifted by tail, contralateral decrease to sensory stimulation of whisker, and ipsilateral head tilt) was used to validate successful ischemia allowing for study inclusion [31]. Mice were re-anesthetized 15 min prior to the 60 min time mark, the silicone monofilament was retracted, and the common carotid was secured. The incision was closed with interrupted nylon sutures, and mice were administered a subcutaneous injection of 500 μL PBS to restore fluid balance. Reperfusion behavior was assessed for resolution of head tilt, improvement in circuling behavior, and resumption of spontaneous movement. Mice displaying “barrel-rolling” behavior, those with signs of excessive pain or lack of any movement were excluded because these behaviors correlate with higher likelihood of technical failures (i.e. subarachnoid hemorrhage and/or incomplete reperfusion). Mice were subsequently recovered in a separate cage on a surgical heating pad set to 37 °C for up to 4 h post-procedure prior to being returned to their home cage. Mice were monitored daily in a temperature-controlled environment with 12 h light/dark cycles in their home cages, with those displaying signs of severely impaired behavior, inadequate pain control, weight loss exceeding 25 % of the original body weight, or other severe health issues being sacrificed and excluded from the study, as previously described [31].

IKVAV-PA administration and detection by intracranial intravital imaging

Adoptive transfer mice (heterozygous C57BL/6 CX3CR1^{GFP +} mice reconstituted with bone marrow from Catchup^IVM mice) underwent tMCAO and were retro-orbitally injected with 100 μL of fluorescently labeled IKVAV-PA by incorporating 1 mol% of AZDye™ 647-conjugated PA molecules (Supplement S2). Fifteen minutes before intravital imaging, mice were retro-orbitally injected with DyLight 550-conjugated rat anti-mouse anti-CD31 (clone 390) mAb for cerebrovascular blood vessel labeling. Intracranial intravital imaging was performed using an Olympus Bx-51WI Fixed Stage Illuminator with a Yokogawa CSU-X1-A1 spinning disk with a 20× water-immersion objective (1.0 numerical aperture), a Hamamatsu EMCCD C9100-50 camera, and a modular laser system with solid-state diode lasers with DPPS modules for 488, 561, and 640 nm and appropriate filters as previously described [34]. Synchronization was managed by a Prosync 2 Controller. Z axis movement and objective positioning were controlled by a Piezoelectric MIPOS100 System. Before the surgery, the mice were anesthetized by intraperitoneal injection of ketamine and xylazine (100 mg/kg and 10 mg/kg body wt, respectively). Mice were placed in a customized stereotactic frame under a heating pad. A midline incision was made across the scalp, and the skin and periosteum were removed to expose the skull surface. Artificial cerebral spinal fluid at 37 °C was applied to the exposed skull surface, and a high-speed drill was used to expose 2 circular areas of the cortical surface (∼1–1.5 mm in diameter) lateral to the superior sagittal suture between the bregma and lambda sutures anteriorly and posteriorly, respectively [35].

Behavioral studies

All mice underwent neurologic scoring with the Bederson Score at day 3 and day 7 post tMCAO given the early reported validity and reliability among scorers [36,37]. Mice were habituated in the testing environment three days prior to baseline behavioral testing and follow-up testing. The modified open field test was performed at baseline, and then again post tMCAO on day 3 and day 7 prior to animal sacrifice [38]. Open field testing was selected to evaluate for any potential differences in mice activity states during the observed recovery period [39]. Mice underwent 1 min of observed novel cage exploration, with recording at baseline, on day 3, and day 7 post tMCAO. Time to exploration of all four quadrants of each cage was tracked and compared across individual mice. Open field testing was capped at 1 min for mice that were unable to complete exploration across all four quadrants within the allotted time.

Tissue processing for histological analysis

At day 7 post tMCAO, mice were sacrificed and perfused with 1x PBS. Seven days was selected as the earliest time point to assess biologic response to IKVAV-PA based on the earliest anticipated effects, and prior work addressing the impact of IKVAV-PA in a spinal cord injury model [19]. Brains and other organs were extracted and then immediately placed in 4 % paraformaldehyde (PFA, Thermo Fisher J19943-K2, Waltham, MA) prepared in 1x PBS. Brains were sectioned in 1–2 mm coronal sections prior to further fixation, as previously described [40]. After 24 h, all organs were sequentially transferred into Falcon tubes containing 15 % and then 30 % sucrose solution prepared in 1x PBS. Once organs sank to the bottom of the Falcon tube (∼2 days), samples were flash frozen in Tissue Tek O.C.T at −80 °C. Brains were cryosectioned at a 50 μm or 10 μm thickness in the coronal plane. All other organs were cryosectioned at a 30 μm thickness.

IKVAV-PA ex vivo detection by fluorescence microscopy

Following brain tissue fixation and preparation, a single 1 mm coronal brain section approximately 3 mm from the olfactory bulb was placed in a Mattek #1.5 dish and covered with a coverslip. Wide-field images were collected using a Nikon Eclipse T12 wide-field microscope with a Nikon objective (NA) equipped with a Nikon DS-Qi2 camera. Images were quantified using FIJI software (version 1.54f). All contrast and brightness adjustments were done linearly.

Cresyl violet staining and infarct quantification

0.1 % cresyl violet solution (Abcam, AB246816-1001, Cambridge, MA, USA) was warmed to 37 °C prior to staining. Brain slices were brought to room temperature and baked at 55 °C for 1 h after which excess O.C.T. was carefully removed with tweezers. Slices were first immersed in 95 % ethanol (Decon Labs, #2701, Swedeland, PA, USA) for 1 min, then dipped twice in 70 % ethanol for 3 s, and hydrated in water for 2 min prior to staining with 0.1 % cresyl violet for 10 min. Immediately following staining, samples were rinsed in water and differentiated for 2–3 min in 95 % ethanol. Samples were then treated with 100 % ethanol twice for 10 s each, followed by two 1 min washes with HistoPrep xylene (Fisherbrand, HC700, Waltham, MA, USA) to clear the tissue. Stained samples were mounted with Cytoseal™ XYL (Thermo Scientific, 8312-4, Waltham, MA, USA) and imaged on a Nikon Ti2 Widefield Microscope, at 4× magnification. Quantification of the infarct region was done over a 2 mm thick coronal section of the brain collected 3 mm caudal to the olfactory bulb. Consecutive 50 μm sections were collected and stained with cresyl violet as described above. Every 8–10th slice was selected for quantification, to allow for equal distribution across the infarct core. Infarct quantification was performed using FIJI software (version 1.54f), with tracing over the entire coronal section and manual segmentation of the infarct area, with subsequent calculation of infarct percent area for the slide. Quantification was performed by a team member blinded to the treatment condition.

Hematoxylin and Eosin (H&E) staining

30 μm cryosections of brain, spleen, kidney, and liver were washed with PBS for 5 min to remove O.C.T. and air-dried. Mayer’s Hematoxylin (ScyTek Laboratories, HMM500, Logan, UT, USA) was applied to organ sections for 2 min, followed by washing under running water for 5 min. The stained sections were quickly dipped in 1 % HCl in 70 % ethanol for 5 s and washed again with running water. Eosin Y solutions (Sigma Aldrich, HT110232, St. Louis, MO, USA) was then applied to the sections for 45 s and washed with water similarly. Stained tissues were dehydrated in a serial manner in 70 % ethanol, 90 % ethanol, 95 % ethanol 2X, 100 % ethanol 2X, and xylene 2X for 3–5 min each. After air drying for 10 min, sections were mounted with Permount™ Mounting Medium (Fisher Chemical™, SP15-100, 20 μL per section) under #1.5 coverslips and stored at room temperature. Bright field images of H&E samples were obtained using a 10× objective on a Leica DM6B Fluorescent Widefield Microscope.

Immunofluorescence (IF)

10 μm frozen brain tissues were washed in PBS followed by deionized water for 10 min to remove O.C.T. After air-drying, tissues were marked with a hydrophobic ink pen and ∼50 μL of blocking buffer (5 % BSA, Thermo Fisher 37525, 0.3 % Triton X-100, MilliPore Sigma X100, in PBS) was added to each section. After 1 h, the blocking buffer was removed and replaced with primary antibodies in 1 % BSA and 0.3 % Triton X-100 (anti-GFP, 1:500, Abcam ab13970; anti-GFAP, 1:500, Cell Signaling Technology CST3670S, Danvers MA, USA) and incubated at 4 °C overnight. Tissues were washed three times with TBST (Cell Signaling Technology 9997S, Danvers MA, USA) before secondary antibodies (Alexa Fluor™ 488 Goat anti Chicken, 1:1000, Invitrogen, A-11039; Alexa Fluor™ 568 Donkey anti Mouse, 1:1000, Invitrogen, A-10037, Waltham, MA, USA) and DAPI (1:500, Thermo Fisher EN62248, Waltham, MA, USA) were added in 1 % BSA and incubated at room temperature for 2 h. Tissues were then washed three times in PBS and mounted with Prolong Antifade Gold Mount (Invitrogen P36930, Waltham, MA, USA). Wide-field fluorescent images were obtained using a 10× objective on a Leica DM6B Fluorescent Widefield Microscope.

Statistical analysis

Statistical analysis was performed using GraphPad Prism software (version 10.4.1). The normality of the data was confirmed using Shapiro-Wilk test. The two-tailed, unpaired Student’s t-test was used. Statistical significance was denoted as follows: ∗∗∗P < 0.001, ∗∗P < 0.01, ∗P < 0.05, ns P > 0.05.

Results

PA design and characterization

We selected palmitoyl-AAGGEEEEG-IKVAV, or IKVAV-PA (Fig. 1a, Supplement Fig. S1), as the candidate PA based on its well-demonstrated capability of promoting neuronal survival and enhancing neurite outgrowth post traumatic injuries [17,19]. To prepare the IKVAV-PA supramolecular assembly, we solubilized IKVAV-PA and IKVAV-AZDye™ 647 PA (Supplement Figs. S2–S3) powder in an isotonic saline and adjusted to neutral pH with NaOH. The resulting PA solution was aged at ambient temperature overnight (Fig. 1d). Negative-staining TEM revealed that after aging, 99 mol% IKVAV-PA and 1 mol% IKVAV-AZDye™ 647 PA self-assembled into a mixture of high-aspect-ratio nanofibers and small micellar nanostructures. Micelles were found to associate closely with the scaffolds provided by nanofibers, causing the fibers to adopt a clustered appearance (Fig. 1b–c). Recent work has proposed that an equilibrium is established between filaments and small aggregates in supramolecular assemblies with enhanced motion of molecules [41,42].

tMCAO animals, PA injection, and PA entry into the ischemic brain

All mice underwent tMCAO within the left hemisphere for technical consistency and reproducibility. Given the extensive reporting on sex differences in the murine tMCAO model, we selected only age and litter-matched female mice for experiments [[43], [44], [45]]. Fourteen female heterozygous C57BL/6 CX3CR1^GFP mice aged 12–16 weeks were included in final study analyses, after exclusion of 4 mice due to pre-specified exclusion criteria. Intra-ischemic behavioral criteria was not met for 2 mice prior to treatment group assignment and they were excluded on the basis of inadequate MCAO. Two mice died prior to the pre-specified 7 day endpoint, with one from each group (IKVAV PA treated and saline control). Autopsy confirmed presence of subarachnoid hemorrhage, suggesting technical failure in these two mice. After tMCAO, immediately following reperfusion, mice underwent retro-orbital injection of either IKVAV-PA or saline (control). Fig. 1d provides a schematic representation of the final experimental workflow. IKVAV-PA was fluorescently labeled prior to injection by incorporating 1 mol% of AZDye™ 647-conjugated PA molecules (Supplement Figs. S2–S3). Systemically administrated IKVAV-PA was able to cross the blood-brain barrier over the ischemic region in contrast to the non-ischemic regions (Fig. 2a). This is due to the well-known principle of transient blood-brain barrier opening following transient ischemic and subsequent reperfusion [46,47], in addition to prior experimental work demonstrating acute blood-brain barrier disruption following tMCAO stroke [46,48,49].

Fig. 2.

Entry of systemically administered IKVAV-PA supramolecular assemblies into the ischemic brain. a) Intravital imaging of AZDye™ 647-labeled IKVAV-PA in the ischemic brains 24 h post retroorbital injection. Neutrophils are identified via red fluorescence protein (tdTomato⁺); microglia are identified with expressed green fluorescent protein (GFP⁺); blood vessel wall is labeled with DyLight 550-conjugated rat anti-mouse anti-CD31 (clone 390) mAb. Scale bars correspond to 50 μm. b) Widefield fluorescent images of AZDye™ 647-labeled IKVAV-PA in mice used in a sham surgery, and in the tMCAO animal model (c). Scale bars correspond to 2 mm. d) Bar graph of PA fluorescence intensity quantified for the infarct hemisphere and contralateral hemisphere for sham vs tMCAO. Fluorescence intensity is averaged across 5 slices for each animal.

Intravital intracranial imaging was performed 24 h post reperfusion using a group of fluorescently labeled IKVAV-PA treated mice (C57BL/6 CX3CR1^GFP mice reconstituted with bone marrow harvested from heterozygous Catchup^IVM mice). Retro-orbitally injected DyLight 550-conjugated rat anti-mouse anti-CD31 (clone 390) mAb was used for cerebrovascular blood vessel labeling. IKVAV-PA fluorescent signals were observed in both the luminal side of the blood vessels and the brain parenchyma outside of them. Interestingly, we also observed occasional co-localization of the vessel walls and IKVAV-PA, suggesting that PAs in the blood circulation most likely traveled across the vessel wall, crossed the blood-brain barrier, and entered the brain tissue (Fig. 2a). To further investigate the distribution of PAs in the ischemic brain, we analyzed brain sections of the IKVAV-PA treated animals with fluorescent microscopy 24 h post injection. The sham animals did not exhibit any difference in PA fluorescence between the two hemispheres. On the other hand, the ischemic hemisphere of the tMCAO animals demonstrated a significant increase in accumulation of PA fluorescence within the infarcted region compared to the contralateral hemisphere, indicating that IKVAV-PA localized to the stroke infarct (Fig. 2b–c).

tMCAO animal response to IKVAV-PA vs. saline treatment

tMCAO animals were monitored for 7 days post injection before histological analysis was performed. Cresyl violet staining of the brains was used to quantify infarct volume in place of 2,3,4-triphenyltetrazolium chloride (TTC), given the inconsistency of TTC staining reported after 72 h post ischemia [[50], [51], [52]]. Saline-injected mice exhibited lighter staining in the sub-cortical infarct regions (26.9 ± 6.3 %) of the left hemisphere compared to the intact contralateral hemisphere, indicative of excessive neuronal and glial cell death. IKVAV-PA treated mice revealed cell death within the same subcortical locations of the ischemic hemisphere, but the overall area of cell death trended to be smaller (10.6 ± 2.4 %, Fig. 3a and . c).

Fig. 3.

Evaluation of the effect of IKVAV-PA treatment on stroke infarct size in tMCAO animals using histological staining. a) Representative cresyl violet staining images of ischemic brains 7 days post IKVAV-PA or saline injection. Section 1-4 are 50 μm coronal sections in serial from anterior to posterior in the mid brain region. b) Zoomed in cresyl violet staining images of the ischemic stroke infarct side (top) and the contralateral side (bottom). c) Quantification of tMCAO mice infarct size relative to the infarct hemisphere across section 1-4 shown in a). N = 7 for the saline group and the IKVAV group. A two-tailed unpaired t-test was used, P = 0.03.

To further probe the cellular response to stroke and recovery post IKVAV-PA or saline treatment, we performed immunofluorescence staining of representative tMCAO brains and examined the stroke infarct with fluorescent microscopy. GFP expressed in microglia and monocytes was re-labeled with an anti-GFP antibody coupled with a green fluorophore to better delineate the mobilization of these immune cells. After stroke injury, astrocytes near the infarct often become reactive and up-regulate glial fibrillary acidic protein (GFAP) expression [53,54]. Therefore, we stained GFAP with a red fluorophore to visualize astrocyte activation. DAPI was used as a counterstain to determine the number and location of all cells. Fluorescent images revealed a lighter DAPI staining in the stroke infarct, consistent with the loss of cells observed in the cresyl violet staining. Nonetheless, we observed an excessive accumulation of microglia and GFAP activation near the stroke site, indicating an elevated immune response and astrocyte reactivity local to the infarct in the quantified sections (Supplement Fig. S5). The IF data indicate that the tMCAO procedure induced substantial neuron death in the infarct accompanied by astrogliosis and the activation of immune response. Consistent with the cresyl violet infarct volume analysis, IKVAV-PA treatment showed a a reduced area of immune response and astrogliosis activation in IF analysis in the sections analyzed. Future studies will quantify the extent of astrogliosis in longer-term time points (>4 weeks post I/R), which will be of significant translational value to assess long-term recovery.

To evaluate functional recovery of tMCAO mice post IKVAV-PA or saline treatment, we employed a modified open field test and examined the latency it took for each animal to explore all four quadrants of the open field. However, we did not observe a significant difference in latency between the IKVAV-PA and the saline group at day 3 or day 7 (Supplement Table). Additionally, the cumulative stroke behaviors of tMCAO animals (based on Bederson Score) were recorded and used for qualitative comparison. Similarly, we did not observe a significant difference in the number of cumulative stroke behaviors between the two groups at day 3 or day 7 (Supplement Table).

Systemic toxicity and biocompatibility screening

Given the promising trend in neurological studies of IKVAV-PA treated animals, we evaluated the biocompatibility of this PA by screening for systemic toxicity by staining the organs and brains for gross pathology. Organ weights serve as an appropriate screen for initial biotoxicity for intravenously administered peptides, in line with routine toxicology practices in murine modeling [55]. Previous work evaluating the role of intravenously delivered self-assembled PAs for treatment of atherosclerosis demonstrated increased biodistribution within the kidneys and liver [56]. For this reason, the capillary-rich kidneys, liver and spleen in addition to the brain were evaluated by H&E staining and observed using a bright field microscope. No evidence of additional strokes was observed in the contralateral hemisphere (Supplement Fig. S4a–b). Kidneys, livers, and spleens from animals in both the saline and the IKVAV-PA group exhibited normal morphology with their native cellular structures preserved (Supplement Fig. S4c) and no statistically significant difference in weights (Fig. 4). The H&E results indicate that IKVAV-PA did not induce significant systemic inflammation or additional structural pathology within the brain.

Fig. 4.

Systemic Biocompatibility of IKVAV-PA compared to saline treated mice in tMCAO by body/organ weights 7 days post IKVAV-PA or saline treatment. a) Total body weight of animals prior to, 3 and 7 days post tMCAO and PA or saline treatment. b-g) Organ weights of tMCAO mice 7 days post IKVAV-PA or saline treatment for brain b), lung c), liver d), spleen e), kidneys f), and heart g). Hematoxylin and Eosin (H&E) staining of representative regions of the brain, kidneys, spleen, and liver did not show any gross histologic differences or markers of injury (Supplement Fig. S4).

Discussion

We explored the potential role of bioactive supramolecular PA assemblies as an adjunct therapy for ischemic stroke using the so-called IKVAV-PA (Fig. 1a, Supplement Fig. S1), which is known to have a significant spinal-cord regenerative effect contributed by enhanced epitope dynamicity. In recent work it has been demonstrated that these assemblies activate the integrin β-1 receptor, leading to neuron survival and axonal extension [19,20]. Following supramolecular self-assembly (Fig. 1d), IKVAV-PA was found, as expected, to form nanostructures with filamentous shape as well as small micellar aggregates (Fig. 1b–c), which is one of the signatures of epitope motion by design to enhance bioactivity. To highlight the involvement of immune cells in stroke progression and recovery, we used CX3CR1^GFP mice and monitored the infiltration of immune cells and their interactions with IKVAV-PA. We found that systemically administered IKVAV-PA was able to cross the blood-brain barrier within the ischemic hemisphere and enter the brain parenchyma. IKVAV-PA deposited within the infarcted region, thus enabling local delivery of its bioactive signals in proximity to the injury site through systemic administration (Fig. 2a,c-d). The overall deposition of IKVAV-PA in the non-ischemic hemisphere was not significant, and similar patterns of distribution were seen in sham-treated mice (Fig. 2b–d). Although the detailed mechanism is not accessible to us at this time, we hypothesize the significant deposition within the ischemic hemisphere can partially be attributed to increased permeability of the blood-brain barrier [[46], [47], [48], [49]] along with changes in the regional microenvironment including elevated levels of inflammatory factors and cytokines, recruitment of immune cells, and accumulation of cell degradation products [57,58]. Fluorescence microscopy revealed an accumulation of immune cells and reactive microglia at the infarct, indicative of a local inflammatory environment that is hostile for neural survival and regrowth (Supplement Fig. S5). Interestingly, by leveraging our unique bone marrow chimera mice (tdTomato + neutrophils transplanted into CX3CR1^GFP+), we observed occasional co-localization between the IKVAV-PA and resident microglial cells, specifically, and not neutrophils. This suggests potential interactions between the resident microglial cells and systemically administered IKVAV-PA within the ischemic infarct. Since IKVAV is known to modulate macrophage phenotype and promote anti-inflammatory responses [59,60], the interactions we observed hold promise for designing PAs with immuno-modulatory functions as potential stroke therapies and warrants further investigation to delineate mechanisms of interaction and effect on neurological function.

Histological and behavioral analysis were performed 7 days after treatment to evaluate the efficacy of IKVAV-PA in promoting stroke-recovery. Cresyl violet staining revealed that IKVAV-PA treated animals have a reduced infarct size compared to the saline treated animals (Fig. 3). Similarly, immunofluorescence imaging showed a trend that IKVAV-PA treatment reduced immune activation and astrogliosis in the stroke brain (Supplement Fig. S5). However, behavioral studies using the neurologic score (Bederson score) and modified open-field test did not show significant functional improvement of the IKVAV-PA treated animals (Supplment Table S6).

The lack of significant behavioral improvement could have been caused by the relatively robust recovery witnessed within murine tMCAO models despite selection of a 60 min occlusion time, the high variability in tMCAO procedures and subsequent animal responses to the procedures [[61], [62], [63], [64]]. Additionally, limitations in the sensitivity of the behavioral testing performed may have minimized detection of any potential behavioral differences within the two groups tested, whereas more sophisticated quantitative assessments might have identified subtle differences not captured by our behavioral testing [65,66]. Sex-specific differences may have been incompletely captured using a cohort of all-female mice, and use of both sexes in larger cohort studies may identify potential sex-specific behavioral differences, if present. Recent studies have highlighted the variability in behavioral testing following tMCAO, leading to lack of reproducibility and conflicting findings confounded further by varying anesthetic regimens [67]. These issues with modeling functional outcomes in murine tMCAO models with disconnect in tissue outcomes (i.e. infarct volumes) were further highlighted in the recent Stroke Preclinical Assessment Network (SPAN) trial [68].

Importantly, we examined the biocompatibility of PAs as a systemically injectable material for stroke therapy. Animal body weight and organ weights were not significantly reduced by IKVAV-PA treatment after 7 days, indicative of no gross systemic toxicity (Fig. 4). Organ and brain H&E staining revealed no systemic inflammation or induced secondary stroke in the brain (Supplement Fig. S4). Based on our preliminary findings and initial toxicity screening, we conclude that IKVAV-PA can be safely injected systemically in tMCAO animals without clear detrimental effects observed at 7 days post stroke.

Combining fluorescence, histology, and systemic toxicity screening, herein we present the highly dynamic IKVAV-PA as a feasible and safe nanomaterial holding great potential as an adjunctive stroke therapeutic to be used with standard-of-care reperfusion therapies. The IKVAV-PA self-assembled into a mixture of small micellar structures in equilibrium with fibrous nanostructures as a result of its enhanced supramolecular dynamics encoded by the peptide sequence [19,20,41,42]. This characteristic is likely to contribute to the ability of the nanostructures to cross the blood-brain barrier within the infarcted region after systemic administration into ischemic stroke animals. The PA nanostructures were shown to circulate to the site of injury and significantly reduce the infarct size. Although a significant functional improvement was not observed for the IKVAV-PA within our short window of observation, future work may take advantage of IKVAV-PA’s ability to promote the slow phenomena of vascularization and axonal regeneration in order to yield long-term functional outcomes. In addition, bioactive PA supramolecular assemblies offer an extremely broad platform compatible with the use of diverse biological signals that can be tailored to extracellular and intracellular targets [16,69]. Future investigations will focus on designing functionalized PAs with additional regenerative signals to promote stroke recovery, balancing the need to suppress potentially harmful early inflammatory signals.

Conclusions

Our study is the first to explore the potential of supramolecular peptide materials as an adjunct therapy for ischemia/reperfusion injury following AIS. Utilizing a pre-clinical murine tMCAO model, we observed permeation of the highly dynamic IKVAV-PA across the blood-brain barrier and deposition in the stroke infarct, which correlated with a significantly reduced infarct size. Furthermore, histological analyses of systemic organs demonstrated good biocompatibility of the dynamic IKVAV-PA with the stroke animals. This work lays the foundation for developing supramolecular peptide amphiphile assemblies to promote tissue-level neural repair post stroke as a complementary therapy to the current standard of care. Further studies are necessary to test longer-term outcomes and efficacy of additional PAs augmenting mechanisms of recovery following ischemic stroke.

Author contributions

ZG: Data Collection, Study Design, Manuscript Preparation and Review.

LHAS: Data Collection, Study Design, Manuscript Preparation and Review.

ZL: Data Collection, Manuscript Review.

FC: Data Collection, Manuscript Review.

CS: Data Collection, Manuscript Preparation and Review.

ZL: Data Collection, Manuscript Review.

RM: Data Collection, Manuscript Review.

EA: Study Design, Manuscript Review.

WAM: Study Design, Manuscript Review.

DPS: Data Collection, Study Design, Manuscript Review.

SIS: Study Design, Manuscript Preparation and Review.

AB: Data Collection, Study Design, Manuscript Preparation and Review.

Declaration of competing interests

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Erika Arias reports financial support was provided by National Institutes of Health. David P. Sullivan reports financial support was provided by National Institutes of Health. William A. Muller reports financial support was provided by National Institutes of Health. Ayush Batra reports financial support was provided by National Institutes of Health. Samuel I. Stupp reports financial support was provided by National Science Foundation. Samuel I. Stupp has patent #US8063014B2 issued to Northwestern University. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Multimedia component 1

Details of peptide amphiphile chemical structure, PA purification and characterization, systemic organ toxicity screening by histology, immunofluorescence analyses of ischemic brain tissues, and behavioral testing scores are included in the supplementary material.