In vivo Neuroprotective Effect of a Self-assembled Peptide Hydrogel

Abstract

Traumatic brain injury (TBI) is associated with poor intrinsic healing responses and long-term cognitive decline. A major pathological outcome of TBI is acute glutamate-mediated excitotoxicity (GME) experienced by neurons. Short peptides based on the neuroprotective extracellular glycoprotein ependymin have shown the ability to slow down the effect of GME — however, such short peptides tend to diffuse away from target sites after in vivo delivery. We have designed a self-assembling peptide containing an ependymin mimic that can form nanofibrous matrices. The peptide was evaluated in situ to assess neuroprotective utility after an acute fluidpercussion injury. This biomimetic matrix can conform to the intracranial damaged site after delivery, due its shear-responsive rheological properties. We demonstrated the potential efficacy of the peptide for supporting neuronal survival in vitro and in vivo. Our study demonstrates the potential of these implantable acellular hydrogels for managing the acute (up to 7 days) pathophysiological sequelae after traumatic brain injury. Further work is needed to evaluate less invasive administrative routes and long-term functional and behavioral improvements after injury.

1. INTRODUCTION

Traumatic brain injuries (TBI) contribute to millions of emergency department visits, hospitalizations, and deaths annually in the US — leading causes include car crashes, falls, and assaults [1]. Such injuries cause neuronal death, inflammation, and glial scarring — which may lead to cognitive deficits years or even decades later. The surge of extracellular glutamate level at the acute stage after TBI has been linked to increased neuronal death [2, 3]. This phenomenon is referred to as glutamate-mediated excitotoxicity (GME). The neurons in the brain are post-mitotic and have muted intrinsic regenerative potential. Unlike those in the peripheral nervous system, neurons in the central nervous system (CNS) cannot regenerate axons following axonal tear [4], which may result from the mechanical impact of TBI [5].

In the absence of a strong physiological regenerative response, neuroprotective and regenerative therapies may be useful in managing the impact of TBI in the acute phase (up to 7 days). Neurotrophic factors, such as brain-derived neurotrophic factor (BDNF), can aid in the neuronal survival in vitro [6] as well as in vivo [7]. Peptide-based hydrogels have been used to deliver such neurotrophic growth factors for a sustained period [8] and aid axonal survival in the CNS [9]. One pronounced disadvantage of such drug delivery strategies is that the neurotrophic factors can diffuse away from the delivery site, leading to time-dependent attenuation of activity. We can obviate this problem by encoding bioactive neuroprotective epitopes directly onto the sequence of the therapeutic peptides that self-assemble into injectable nanofibrous hydrogels, resembling the extracellular matrix (ECM) [10, 11]. This strategy enables sustained epitope presentation on ECM-mimetic scaffolds, while maintaining native-like material properties.

Self-assembling peptides based on a β-sheet motif can form high-aspect-ratio nanofibers, resulting in thixotropic hydrogels [12]. These scaffolds have been evaluated for drug delivery, wound healing, and modulation of angiogenesis, dentinogenesis, and inflammation [12–14]. We designed a neuroprotective peptide (referred to as SLen) containing a central β-sheet-forming peptide backbone [(SL)₆] [10], attached to a neuroprotective peptide domain KKDGDGDFAIDAPE (referred to as CMX-9236) [15]. CMX-9236 is a 14 amino acid domain of ependymin, a neuroprotective extracellular glycoprotein [15]. CMX-9236 promotes neuronal survival in ischemic rat models [15], by activating the transcription factor AP-1 (a heterodimer of leucine zipper proteins Fos and Jun), which regulates CREB (cAMP response element-binding protein) and promotes synaptic plasticity and neurite growth [16, 17]. Neurotrophic factors exert their neuroprotective effect and modulate synaptic functions, in part, by activating AP-1 [18]. In vitro, CMX-9236 blocks excitotoxic effects of glutamate in primary rat cortical neuron cultures [15]. However the small peptides such as CMX-9236 are not suitable for clinical application in vivo as they are prone to rapid diffusion and clearance [19]. Attaching the peptide moiety to a self-assembling domain can ensure that the peptide is physically retained (as a hydrogel) in the injury site.

As the neuroprotective moiety is negatively charged, we designed the self-assembling domain to be negatively charged as well (glutamate residues are negatively charged at neutral pH), resulting in a resulting net negative charge (−5) for the functionalized self-assembling peptide [(SL)₆-E-G-KKDGDGDFAIDAPE] in physiological conditions. The net charge on SLen confers flexibility in the storage and formulation of the peptide. It is soluble in low ionic strength conditions, as the net charge of the building blocks prevent self-assembly into nanofibers, due to electrostatic repulsion among the building blocks. In presence of multivalent positively charged ions (such as Ca⁺²), due to shielding of like-charge repulsion among the building blocks [20], SLen self-assembled into β-sheet nanofibers, which are ionically crosslinked into a hydrogel (Fig. 1A).

Fig. 1. Hierarchical self-assembly and biophysical characterization of the SLen nanofibrils.

(A) The core self-assembling domain, comprising alternating hydrophilic and hydrophobic residues, has a high propensity to form β-sheet nanofibers. (B–C) Scanning electron microscopy (SEM) images of the nanofibrous mesh underlying the SLen hydrogel (critical point dried and sputter-coated with gold) at two different magnifications (hydrogel bolus on a stage imaged in the inset of panel B). (D) Atomic force microscopy (AFM) on diluted SLen nanofibers allows the determination of the height (~2.2 nm) and width (~10 nm) of individual SLen nanofibers (AFM is more accurate in determining heights compared to widths of nanostructures). Please note that the SLen nanofibers are relatively brittle and we cannot determine the length of the nanofibers accurately through AFM, as the spin-coating process may lead to calving of the nanofibers. (E) The circular dichroism (CD) spectrum indicates a β-sheet secondary structure (characteristic minimum at 216 nm) for the self-assembled nanostructures, (F) Fourier transform infrared (FTIR) spectrum of SLen also has a typical β-sheet signature absorbance peak at 1620 cm⁻¹. (G) Rheometric analysis of the hydrogel (concentration: 20 mg/mL) has a storage modulus of ~1000 Pa at low (1%) strain (G’>G”), but the gel undergoes shear thinning (G”>G’) at high strain (>40%).

Another potential advantage of using negatively charged nanofibrous matrices is that similarly charged peptide scaffolds show remarkably lower immune response from the host tissue in vivo [21]. In this article, we see that direct intracortical implantation of the neuroprotective peptide hydrogel SLen may potentially enhance neuronal survival in the acute phase after traumatic brain injury in vivo.

2. MATERIALS AND METHODS

2.1. Synthesis and Characterization of Peptides:

SLen was synthesized by standard Fmoc solid-phase peptide synthesis protocol [20] and purified by high-performance liquid chromatography and dialysis. The dialyzed peptide was lyophilized and stored in the powder form at −80°C until formulation. The peptide hydrogel was formulated as: 20 mg/mL SLen (6.6 mM); 16.5 mM CaCl₂ aq. solution; 298 mM sucrose solution; pH 7.

2.2. Scanning Electron Microscopy & Atomic Force Microscopy:

The underlying nanostructure of the SLen hydrogel was probed with scanning electron microscopy (SEM) and atomic force microscopy (AFM). The 2% (w/v) hydrogel was critical point dried, sputter-coated with gold and imaged in a JEOL JSM-7900F High-Performance FE-SEM machine. For AFM, the 2% (w/v) hydrogel was diluted 10X in deionized water and deposited on a flat mica surface, followed by spin-coating. The coated mica plate was imaged with a Bruker Dimension ICON AFM machine in ScanAsyst PeakForce mode.

2.3. Biophysical Characterizations:

The 2% hydrogel was diluted 100X in deionized water and analyzed with a Jasco J810 circular dichroism (CD) spectropolarimeter. The hydrogel was deposited and dried on an attenuated total reflectance (ATR) set-up and characterized via a PerkinElmer Spectrum 100 Fourier transform infrared (FTIR) spectrometer. For rheometric analysis, the hydrogel was deposited on the stage of a Malvern Kinexus Ultra+ rheometer and subjected to a strain sweep (1% to 100% strain), as well as a shear strain cycling at 1 rad/s (repeated transitions between 1% and 100% strain).

2.4. Cytocompatibility with Fibroblasts in vitro:

3T3 fibroblasts were cultured following a previous protocol [13, 14] and seeded after the first passage. They were seeded at a density of 1000 cells/well in a 96-well plate and cultured in fibroblast media (DMEM, 10% FBS, 1% Pen-Strep) for 24 h prior to the introduction of 200 μL of condition media (0.02% and 0.002% SLen) and the control (fibroblast media) (n=4). The condition media was changed at day 1, day 4, and day 6. At 6 days, a live/dead assay was performed following the manufacturer’s protocol (Invitrogen).

After 30 minutes incubation, the live/dead solution was aspirated and 200 μL of phosphate buffered saline (PBS) was added to image the cells with fluorescence microscopy. Each well was imaged n=3 for both Calcein AM (green) and EthD-1 (red), manually counted, and averaged for analysis. Live and dead cells were quantified using the NIH ImageJ processing program [22]. The images were converted to 8-bit versions, threshold-adjusted, and then the particles were analyzed using set parameters of size (pixel²) 120–infinity, and 0.00–1.00 circularity.

2.5. Effects of SLen on Glutamate-mediated Excitotoxicity in vitro:

Cortices were extracted from the embryos of a Sprague-Dawley timed pregnant rats (Charles River) at day 16 of the gestational period. The cortices were digested in trypsin-EDTA (0.25%) at 37°C for 30 min and agitated with a pipette to release cells, then strained with 70 μm and 40 μm pore filters to separate out primary cortical neurons. Neurons were counted and seeded at a density of 50,000 cells/cm² on coverslips (diameter: 12 mm) coated with 10 μg/mL poly-D-lysine and supplemented with neurobasal medium with 2% Gibco B27 AO+ neuronal culture system (50X), 1% penicillin-streptomycin, and 0.2% glutamate solution. Neurons were cultured in the neurobasal media for 12 days with media changes every 2 days. At Day 12, neurons exhibited neurite branching (Fig. 2C). Neurons were then incubated with 100 μM L-glutamic acid for 24 h to induce excitotoxicity (see Fig. S2 in the Supporting Information). Previous work demonstrated a range of concentrations and incubation times for L-glutamate to induce excitotoxicity [23, 24]. After performing a dose response assessment, we found that 100 μM L-glutamate with 24 h incubation caused severance of synapses and fragmentation of neurites.

Fig. 2. In vitro assays to test cytocompatibility and neuroprotective properties of the peptide hydrogel SLen.

(A) Live/dead staining for cytocompatibility of SLen in fibroblasts, and, (B) quantification of live versus dead cells at different concentration of SLen. The cytocompatibility data suggests a low risk of off-site toxicity to somatic cells. (C) Bright-field image of primary cortical neurons cultured in vitro, (D) live/dead staining for cortical neurons cultured with SLen. (E–G) Live/dead staining indicate SLen is neuroprotective against L-glutamate-mediated excitotoxicity in the in vitro model of primary cortical neurons. Representative immunostaining of (H–J) the neuronal marker NeuN (green), and (K–M) the neurofilament protein (red); (N–P) merged images of NeuN and neurofilament staining. (Q) Quantification of live/dead staining in control neurons, control neurons + SLen, L-glutamate exposed neurons, and L-glutamate acid exposed neurons + SLen (representative images in panel E, F, and G). (R) Cell counting of positive NeuN staining, and (S) quantification of Neurofilament staining fluorescence intensity show a significant difference between L-glutamate exposed neurons with and without SLen. Data were analyzed using one-way ANOVA post-hoc independent-samples t-tests with Bonferroni correction (N=4). Scale bar: 100 μm. (*p < 0.05).

Our preliminary data indicated that SLen exerts dose-dependent effects specifically on neurite integrity and growth. After performing a dose response assessment, we found that 0.005 mM concentration of SLen did not have any protective effects on the injured neurons and 0.5 mM concentration caused neurons to aggregate into large neuro-spheres, but 0.05 mM SLen exhibited beneficial effects (see Fig. S2 in the Supporting Information). After removal of L-glutamate, neurobasal media supplemented with 0.05 mM SLen was added and media was changed every day for 3 days. At Day 16, Live/Dead cell viability kit (Invitrogen) was used to evaluate the neuronal survival. Four images were taken per hydrogel (n= 4 wells per condition), and the overall in vitro assay was repeated 4 times with separately isolated neurons.

2.6. Animals:

Twelve-weeks-old male Sprague-Dawley rats were purchased from Charles River Laboratory (Wilmington, MA). The animals were kept in a 12 h dark-light cycle at room temperature with food and water. All procedures were approved by the Rutgers-Newark Institutional Animal Care and Use Committee and followed the guidelines for the Care and Use of Laboratory Animals (Public Health Services (NIH) Assurance no. A3158–01; USDOA registration no. 22-R-0153; AAALAC accreditation no. 000534).

2.7. Fluid Percussion Injury (FPI):

Sprague Dawley male rats (250–300 g; 8–11 weeks old) were randomly selected and subjected to lateral FPI for a moderate TBI [25]. Sham rats with surgery were used as control. 12 rats were used for each group (Sham, FPI, FPI+SLen) (total 36 rats). All rats were anesthetized with a mixture of ketamine (100mg/kg), and xylazine (10mg/kg), administered via intraperitoneal injection. The rats were placed in a stereotaxic frame. Craniotomy (3.0 mm) was performed over the left parietal skull, 2.5 mm lateral from the midline and 3.0 mm caudal from bregma, with the dura intact. A Luer-lock hub was implanted on the skull and sealed by a cyanoacrylate gel. Methyl methacrylate (Henry Schein, Melville, NY, USA) was applied to secure the hub. The secured hub was filled with sterile saline. Twenty-four hours after surgery, animals were randomly picked for either sham or FPI. FPI was induced by releasing the pendulum hammer onto the piston of the fluid filled cylinder to induce the injury (see Fig. S3 in the Supporting Information). For a moderate injury (1.8–2.0 atm.), the recorded apnea and righting reflex time were 15–20 s and 8–10 min, respectively.

2.8. Therapeutic Intervention:

After FPI and anesthesia, the rats were placed in a stereotaxic frame to be treated with 5 μL SLen hydrogel (2% w/v). The locations of injection according to the stereotaxic coordinates were: AP=−3, L=−2.5 from Bregma, dorsoventral (DV) = 0.5 mm. Then the injury hub was removed, and the head sutured. The rats were returned to a heating pad until their wake-up and then they were returned to their cages.

2.9. Tissue Processing:

Animals were euthanized at day 7 post-FPI by intraperitoneal injection of 0.1 ml of ketamine (80–100 mg/kg) + xylazine (5–10 mg/kg) mixture, using a 26-gauge needle, as approved by the panel on euthanasia of the American Veterinary Medical Association (AVMA). This was followed by slow transcardial perfusion with ice-cold phosphate buffered saline (PBS) and 4% paraformaldehyde in PBS. Brain regions with the impact sites were recut and immersed in 4% paraformaldehyde overnight, then transferred to 30% sucrose containing PBS. 20 μm thick coronal sections were sectioned by cryostat from each sample for immunofluorescence staining.

2.10. Western Blot:

Fresh brain tissues were collected from the impact/surgery sites and homogenized with CelLytic-M (Sigma) lysis buffer (500 μL buffer for every 10 mg of tissue) on ice using sonication. Then the homogenized tissue was centrifuged at 13,000 rpm for 20 min at 4°C and the supernatant was collected and assessed for protein concentration by bicinchoninic acid (BCA) assay (Thermo Fisher). Tissue lysate was formulated as 1 μg/μL, and then loaded as 20 μg/lane in 4–15% SDS-PAGE gradient gels (Thermo Fisher), transferred on PVDF membranes and blocked with 5% milk. The gel was then incubated overnight with primary antibodies at 4°C, washed, and incubated with horse-radish peroxidase conjugated to secondary antibodies (1:10000 dilution) for 1 h at room temperature. West Pico chemiluminescence substrate (Thermo Fisher) was used to detect immunoreactive bands. Data were analyzed by densitometry with ImageJ.

2.11. Immunofluorescence Microscopy:

Both in vitro and in vivo constructs were analyzed through immunofluorescence. Coverslips with cells were fixed in 4% paraformaldehyde for 15 min at room temperature, followed by washing with 0.1% Triton X-100 in PBS for 5 min at room temperature. Brain coronal sections (20 μm thick) were rinsed with PBS and fixed in acetone-methanol (1:1 volumetric ratio) for 10 minutes at −20°C. Fixed slides were then blocked by 3% bovine serum albumin (BSA) with 0.1% Triton X- 100 at room temperature for 1 h. Slides were then incubated with the following primary antibodies overnight at 4°C: anti-vWF (vWF: von Willebrand factor), anti-NeuN, anti-myelin basic protein, anti-alpha smooth muscle actin, anti-neurofilament (Table S1 in the Supporting Information). Sample slides were washed with PBS and then incubated with Alexa Fluor 488/594 conjugated anti-mouse/goat/rabbit/sheep immunoglobulin-G (IgG) for 1 h at room temperature. After washing, sample slides were mounted with golden mounting containing DAPI (Thermo-Fisher). Images were captured by fluorescent microscopy (Eclipse TE2000-U, Nikon microscope, Melville, NY) and analyzed with ImageJ.

2.12. Neuronal cell counting:

Immunostained brain sections were imaged at 10X magnification by a fluorescent microscope. NeuN positive cells were quantified using ImageJ. Cells in the injured cortex of 3 groups including sham, FPI+PBS and FPI+SLen (n = 8 per group) were counted. 4 cortical sections from each rat were imaged for cell counting. For each brain section, NeuN positive cells were counted in the region of interest. Results in the same treatment group were averaged and represented graphically.

2.13. Evaluation Methodology and Statistical Analyses:

Statistical analysis of the data was performed using SPSS 24 (IBM). Percentage of live cells differed significantly among conditions (one-way ANOVA, F(3,8) = 69.47, η2 = 0.96, p < 0.001). Post hoc independent samples were tested with Student’s t test, with Bonferroni correction (α adjusted to 0.013). Positive NeuN count differed significantly among conditions (Kruskal-Wallis chi-squared = 27.42, df = 3, p < 0.001). Post hoc two-sample Wilcoxon tests were carried out with Bonferroni correction (α adjusted to 0.013).

3. RESULTS & DISCUSSION

3.1. Self-assembly and Biophysical Characterization:

SLen contains a central domain with alternating hydrophilic and hydrophobic residues [20, 26], which has a high propensity to form β-sheet nanofibers with a hydrophobic core (Fig. 1A). In presence of cationic counter-ions, SLen forms a hydrogel at a concentration of 20 mg/mL. When the hydrogel is critical-point dried (to preserve the hydrated scaffold architecture) and imaged by scanning electron microscopy (SEM), the interlinked nanofibrous mesh constituting the hydrogel becomes apparent (Fig. 1B–C). The dimensions of the individual nanofibers can be more easily measured by diluting the hydrogel to yield a homogeneous solution (2 mg/mL), spin-coating it on a mica plate, and characterizing by atomic force AFM. The individual nanofibers have uniform heights of 2.2 ± 0.2 nm and widths of 10 ± 1 nm (Fig. 1D). Circular dichroism spectra of SLen (0.2 mg/mL) show a characteristic β-sheet signature minimum at 216 nm (Fig. 1E). FTIR spectrum obtained in an attenuated total reflectance (ATR) set-up supports the β-sheet secondary structure of the peptide (amide-I absorbance at 1620 cm⁻¹) (Fig. 1F).

3.2. Material Properties:

As the self-assembly of the hydrogel is mediated purely by non-covalent interactions among the peptide building blocks, the assembled matrix is responsive towards mechanical shear [20]. The hydrogel has a storage modulus (G’) of ~1000 Pa at a low shear strain, which is an order of magnitude higher than the loss modulus (G”) (Fig. 1G). The hydrogel liquefies at high shear strain (i.e., G”>G’) (Fig. 1G) and recovers its material properties when the strain is lowered (Fig. S1 in the Supporting Information). Such material properties are important for facile implantation of the hydrogels in vivo (via syringe aspiration and injection), as the injected peptide formulation can efficiently conform to the injured site and re-assemble into a solid hydrogel, seconds after injection into the target tissue. The post-shear recovery in storage modulus of the material, which corresponds to the re-constitution of the hydrogel, is robust with respect to several cycles of high and low strains (Fig. S1 in the Supporting Information). Thus, these material features enable the hydrogel to retain solid-like in situ placement, with advantages of fluid-like flow (e.g., flowing through a syringe needle and conforming to the physical void of the injection site). In the developing brain, the mechanical features of the tissue guide axonal projections and growth [27]. SLen hydrogel has stiffness (1000 Pa) comparable to the native brain ECM [28, 29], one of the softest tissues in mammals [30], as well as to other neuroprotective hydrogels developed recently [31]. Additionally, hydrogels in this range of stiffness have also been shown to be optimal for preferential differentiation of neuronal stem cells into neurons, which could be potentially favorable for long-term efficacy of the implant [32, 33].

3.3. Cytocompatibility and Neuroprotection in vitro:

SLen’s potential off-site cytotoxicity was evaluated in a dose-dependent fashion against NIH 3T3 fibroblasts (Fig. 2A–B), by Live/dead staining. Over a range of concentrations, we noted excellent cytocompatibility of SLen with fibroblasts. For ease of imaging and characterization, cytocompatibility was evaluated at subgelation concentrations of the peptide. This initial screen warranted evaluation of these constructs with cells specific to the neuronal niche. To evaluate this, we tested the effect of the peptide on primary cortical neuron culture by an excitotoxicity assay (Fig. 2C). First, by live/dead staining, SLen was shown to be neuro-compatible in vitro (Fig. 2D, 2Q). Predictably, excitotoxic insult by L-glutamate leads to death of most of the neurons (compare Fig. 2E and 2F, also see Fig. 2Q). However, the presence of 0.05 mM SLen in culture exerts a neuroprotective effect even in the presence of a toxic amount of extracellular glutamate (Fig. 2G). We examined neuronal cell bodies (NeuN) and the axonal projections (neurofilament) by immunostaining after the excitotoxicity assay (Fig. 2E–P). Excitotoxic glutamate damage led to a significant loss of neuronal cell bodies (Fig. 2R) and disrupted neuronal branches (Fig. 2S), both of which showed signs of abrogation in presence of SLen (Fig. 2R–S). These in vitro studies suggested SLen’s potential effect on neuronal bodies and branches against glutamate-mediated excitotoxicity.

3.4. Effect of the Peptide on Neuronal Growth and Axonal Sprouting Post-injury:

To validate the proof-of-concept that SLen could promote the sprouting of injured neuronal dendrite branches and axonal extensions, we scraped off half of the neuronal culture seeded on glass cover slips causing injury to neurons (Fig. S4 in the Supporting Information). Neurons were then grown in the presence or absence of SLen. We then evaluated the differential sprouting of neuronal dendrite branches and axonal elongation at day 1 and 3 after injury. We observed a significant enhancement of neuronal dendritic branches and axonal extension in neuronal culture that were exposed to SLen at day 1 and day 3 compared with injury alone (Fig. S4). The promoting effects of SLen on dendritic branches or axonal grown were highly distinguishable from those of injured neurons without SLen.

3.5. Effect of the Peptide Hydrogel on the Injured Brain Microenvironment:

Fluid percussion injury (FPI) is an established TBI model that mimics concussion-like scenarios [25]. It’s a mixed injury model which produces focal damage including perisomatic axotomy and vascular disruption, as well as diffuse axonal damage, leading to impaired axonal transport and neuronal atrophy [25]. In our study, a moderate level of FPI led to a drastic loss in the number of neurons (as observed in NeuN and myelin basic protein staining) at day 7 after the injury (compare Fig. 3A–C vs. 3D–F). We detected a significant decrease of neurons in the cortex by counting cell bodies stained by NeuN after FPI — this effect was diminished with treatment of the peptide hydrogel SLen (Fig. 3G–I, for quantification see Fig. 3J). As an indication of white matter/axonal damage, staining of the myelin basic protein (MBP) following TBI revealed that TBI results in a high degree of disorganization of myelin in the rat cortex compared to sham (Fig. 3D). SLen injection following TBI attenuated the disruption of myelin organization (Fig. 3G). The signal embedded in the sequence of the peptides cannot diffuse away rapidly after implantation, since the epitopes are immobilized covalently in the building blocks of the hydrogel, which is retained physically in situ.

Fig. 3. In vivo response to the peptide hydrogels.

Representative images of brain sections immunostained for Myelin basic protein (A, D, G), NeuN (B, E, H) and merged images (C, F, I) in Sham, FPI, and FPI treated with SLen, (J) Neuronal cell counting in cortex for NeuN staining in sham, FPI, FPI+ SLen. N=12 (twelve animals for each condition). Scale bar: 200 μm. * denotes p < 0.05.

3.6. Prevention of Cortical Atrophy:

In order to evaluate whether peptide hydrogel was able to protect injured neurons at the impact site, we performed histological analysis and Nissl staining of neurons in cortical tissue sections from Sham, FPI+PBS and FPI+SLen. Our results showed that brain injury indeed caused a significant loss of cells in the cortical area (Fig. 4, FPI+PBS), and exposure to SLen post-injury appeared to prevent this FPI-induced cell loss (Fig. 4, FPI+SLen), compared with sham controls. The loss of neurons in the impacted cortical area was clearly visible in Nissl staining of neuron density in FPI+PBS compared with that of FPI+SLen or sham control (Fig. 4, bottom panel). Our results from these staining analyses indicate that treatment of SLen exhibited a protective effect on cortical neurons in post-TBI, suggesting a therapeutic potential in brain injury management.

Fig. 4. Cortical atrophy post-FPI.

Histological staining of brain cortical tissue cross-sections (10 μm thick) in Sham, FPI+PBS and FPI+SLen, 2 weeks after injury. Representative image of H&E staining (upper panel, 5X), Nissl staining (middle panel, 5X), and Nissl staining (bottom panel, 40X). Upper/middle panels scale bars are 200 μm long, bottom scale bars are 25 μm long.

3.7. Reduction of Hyperphosphorylation of Tau-protein Post-FPI:

Tauopathy was also investigated as an indication of neuronal dysfunction since hyperphosphorylated tau can induce microtubule instability that results in neuronal cell death. Detection of hyperphosphorylated tau protein (P-tau) in the impacted cortex has shown to be a hallmark for neuronal injury pathology at post TBI [34], in which hyperphosphorylation of tau protein in animal model of moderate TBI happened to occur in hours to months after the injury [35]. In the present studies, we found that the expression of P-tau was significantly increased in the impacted cortical at 2 weeks post-injury. The levels of FPI-induced P-tau were significantly attenuated by SLen treatment compared with sham controls (Fig. S5A–C in the Supporting Information). These qualitative data were further validated by quantitative Western blot analysis, which indicated that the FPI-induced increase in P-tau protein levels were significantly lower after SLen exposure (Fig. S5D). These findings showed that SLen treatment impeded the hyperphosphorylation of P-tau protein after traumatic brain injury, indicating neuroprotective potential of SLen.

3.8. Limitations & Future Work:

The focus of this study is on SLen-mediated recovery and neuroprotection post-TBI in the short-term (up to 7 days). We acknowledge that our characterization is focused on cell-level viability and tissue-level staining studies, and do not directly show neuronal activity or functional improvement. We aim to investigate behavioral and functional outcomes including anxiety and stress measures of mental health in rodent and higher animal models incorporating both diffuse and blunt injury models with less invasive intravenous dosing over a longer timeframe. However, these initial results may hold promise especially for our soldiers and athletes who experience repeated head trauma.

4. CONCLUSIONS

We present a biomaterial scaffold that can prevent some of the neurotoxic outcomes after fluid percussion injury in rats. We have developed a self-assembling peptide based injectable hydrogel containing a neuroprotective domain that can potentially help stem some of the acute negative effects of brain injury. We demonstrate that the hydrogel can be injected into the cortex after an induced fluid percussion injury to generate a consistent healing microenvironment for neurons.

Highlights:

• Mechanical impact driven injuries of the brain do not heal naturally.

• Self-assembled hydrogels recapitulate features of the extracellular matrix.

• Incorporation of a neuroprotective domain in a self-assembling peptide.

• Injectable hydrogel shown to improve response to traumatic brain injury in vivo.