Renovating Neural Networks With Viral‐Mediated Gene Transfer From A Tissue Contacting Matrix Mimic

ABSTRACT

Neurodegenerative diseases such as Huntington's Disease (HD) have a significant impact on healthcare accessibility and costs. A fatal genetic condition, characterized by the progressive loss of striatal neurons, HD is hindered by the lack of endogenous repair in the adult brain. Recent efforts toward protecting neural circuits through neurotrophic support using brain‐derived neurotrophic factor (BDNF) have been suboptimal due to the protein's short half‐life and limited diffusion. Addressing this, adeno‐associated viral vectors (AAV) can be employed as a delivery tool to spatially transduce cells, enabling the localised production of BDNF with consequential neuron protection and/or plasticity, yet present their own constraints. To overcome these known challenges of AAV gene delivery, an injectable, physiologically stable hydrogel‐mimic of the brain's extracellular matrix was fabricated to encapsulate the AAVs. This smart system both shielded and constrained the AAV; optimising transfection and therefore elevated and sustained BDNF presentation at the target site. Here, we achieved high neuroprotection using AAVDJ‐BDNF delivered through a hydrogel formed via self‐assembling peptide nanoscaffolds. These findings support the notion that the spatiotemporal release of BDNF to striatal neurons, facilitated by engineered biomaterial delivery systems, demonstrates tremendous promise by enhancing the efficacy of gene therapy targeted at slowing neurodegenerative disease progression.

Fmoc‐DDIKVAV self‐assembling peptidemediated delivery of AAV‐BDNF significantly enhances neuroprotection in the striatum of a mouse model compared with AAV‐BDNF injection alone. This synergistic integration of biomaterial scaffolding and gene therapy holds substantial promise for slowing the progression of neurodegenerative diseases such as Huntington's disease (HD).

1. Introduction

Many neurodegenerative diseases are incurable and compounded by limited treatment options. This problem is particularly acute for neurological conditions, often leading to progressive decline in cognitive and/or motor functions [1, 2, 3, 4]. For a number of neurodegenerative diseases growth factors, such as glial‐cell‐derived neurotrophic factor (GDNF) and brain‐derived neurotrophic factor (BDNF), have been shown to be beneficial for protecting neurons and promoting plasticity [5, 6, 7]. However, clinical delivery of these proteins face significant challenges, including (i) instability in physiological environments (with evidence that BDNF degrades rapidly, with a half‐life of <5 h in vitro and undetectable after 24 h) [8], (ii) difficulty crossing the blood‐brain barrier [9] and, where infusion approaches are adopted, the added challenges of (iii) high costs (iv) off‐target effects, (v) localised trauma and infection, as well as (vi) adverse immune reactions [10].

To address these challenges, preclinical research is increasingly focused on minimally invasive strategies for targeted and prolonged delivery of growth factors at physiologically relevant concentrations [8, 11]. Developing effective drug/growth factor delivery vehicles that can provide the localized, sustained release through a single, well‐defined injection while protecting growth factors from degradation is critical [12].

Growth factor delivery can protect and restore neurons lost in neurodegenerative diseases like Alzheimer's disease, Parkinson's disease (PD), and Huntington's Disease (HD). HD, a monogenic disorder, is caused by excessive cytosine, adenine, and guanine (CAG) trinucleotide repeats on the short arm of chromosome 4p16.3 in the huntingtin (Htt) gene, leading to a dysfunctional mHTT protein which accumulates to form toxic aggregates within neurons [13, 14, 15, 16]. This results in progressive degeneration, particularly of striatal and cortical neurons, manifesting as involuntary movements (chorea), cognitive decline, and emotional disturbances [2]. While FDA‐approved drugs like tetrabenazine and deutetrabenazine alleviate chorea, they have severe side effects, and fail to address the root cause [17]. Preclinical and clinical findings have indicated that BDNF can provide potential benefits for HD by protecting host neurons from degenerative processes, including by way of viral delivery. In this context, adeno‐associated viral (AAV) vectors have emerged as highly effective tools within the central nervous system (CNS) [12, 18, 19]

Gene therapy has benefitted from recent advancements in modifying and mutating the AAV capsid, the feature which plays a pivotal role in determining AAV tropism, transduction and targeting specific cells for gene delivery [20]. In our previous study, we observed that among the most commonly employed AAVs for gene delivery, including AAV‐DJ, AAV‐2, and AAV‐5, the AAV‐DJ exhibited the highest transduction efficiency in both in vitro (such as with primary cortical neurons, astrocytes, and human astrocytes) and in vivo experiments [12]. The AAV‐DJ viral variant is a bioengineered virus that integrates components from AAV2, AAV8, and AAV9, incorporating the heparan sulfate proteoglycan (HSPG) binding residues from AAV2. Moreover, the inclusion of capsid regions from AAV8 and AAV9 serotypes enhances the variant's interactions with cellular receptors and co‐receptors, thereby improving the AAV‐DJ specificity to transduced cells [19].

Despite such achievements and the emergence of multiple candidate serotypes, fundamental constraints persist, including low transduction efficiencies, off‐target effects, virion dissemination, and, most importantly, neutralization by the host immune system [12, 21]. Numerous attempts have been undertaken to devise strategies for enhancing the mechanism, efficiency, and specificity of viral vector delivery by leveraging implantable or tissue‐contacting biomaterials as delivery vectors [22].

An important aspect of this advancement will be to develop a tissue‐integrated, non‐immunogenic biomaterial that can then be engineered toward inclusion and rationed delivery of AAVs as a comprehensive, tissue‐specific and clinically applicable package. Hydrogels, a class of hydrophilic polymer networks, have great potential in this space, primarily because their structure can be readily engineered via well‐established chemical approaches to exhibit similar mechanical and structural properties to match aspects of the extracellular matrix (ECM) of various tissues [23, 24, 25, 26, 27, 28].

Recently, Self‐assembled peptide (SAP) hydrogels have demonstrated their potential for regenerative medicine, and have been of significant interest to researchers, as they are easily fabricated via the spontaneous assembly of biologically relevant peptide sequences under physiologically relevant conditions. This class of hydrogels have the advantages of a fully synthetic approach to yield a nanofibrous, porous structure, yet retain an inherent biocompatibility with tuneable chemical and mechanical characteristics with no solvents, crosslinkers, or catalysts required [27, 29, 30].

Injectable, tissue‐matched and biocompatible hydrogels formed by the self‐assembly of minimalist (less than 7 amino acids) peptide sequences, have previously been engineered by our group [27, 31, 32, 33, 34, 35, 36, 37]. To initiate assembly, these short sequences are N‐terminally conjugated to fluorenylmethyloxycarbonyl (Fmoc) aromatic capping groups. Under physiological conditions, these peptide derivatives utilise physical interactions to self‐assemble into nanoscale fibrillar architectures to yield biofunctional and bioavailable peptide epitopes at high density that promote cellular interactions [38, 39, 40]. In a demonstration of their clinical promise, we have shown these Fmoc‐SAPs can be both neuroprotective [38] and anti‐inflammatory when delivered in vivo [41]. To tailor the hydrogel toward CNS tissue, we included the widely validated peptide epitope from laminin, IKVAV (isoleucine–lysine–valine–alanine–valine), which is an abundant protein in the brain's ECM. This protein and its epitope have been shown to play an important ECM‐related role that facilitates various neural cell behaviours such as adhesion, migration, proliferation, and differentiation [42]. However, it's utility in minimalist SAPs has been limited as it does not assemble into the required fibrillar structures under physiological conditions. Therefore, to enable the assembly of this sequence at physiological pH, aspartic acid (D) residues were included as the N‐terminal residues of the SAP; these were introduced to modify the pKa of the peptide derivative whilst still ensuring the C‐terminal IKVAV sequence was bioavailable [39]. The SAPs undergo a thermodynamically driven self‐organisation to stabilise in a nanofibrillar arrangement with a hydrophobic core (Fmoc) that allows the solvated C‐terminal sequence to be presented on the surface [43]. These then undergo longitudinal alignment of individual fibrils through surface‐mediated supramolecular associations to entrap the solvent (any dissolved components) within a physiologically stable, supramolecular hydrogel underpinned by decorated intertwined bundles, enabling the formation of the interconnected nanofibrous network [44, 45].

Inspired by the critical gaps in the development of novel treatment strategies for neurodegenerative diseases such as PD and HD, this study explores the ability of an injectable Fmoc‐SAP hydrogel system to sustain and control delivery (in vitro and in vivo) of viral vectors encoding BDNF. The result was a system whereby a therapeutic growth factor could be sustained within the degenerated area without side effects or immune response stimulation.

2. Experimental Procedures

2.1. Materials Preparation and Characterization

2.1.1. Solid Phase Peptide Synthesis (SPPS)

Synthesis of Fmoc‐DDIKVAV peptide derivative involved solid‐phase peptide synthesis (SPPS) on Wang resin in a rotating glass reactor vessel, with a scale of 0.4 mmol. Briefly, SPPS involves stepwise deprotection of the N‐terminal Fmoc group of the amino acid attached to the resin, followed by coupling of the residue in the sequence. Deprotection occurred in a solution of 20% piperidine (PIP) in DMF wasthen carried out for 20 min. Next, the coupling step was performed with the reaction of a solution containing the next Fmoc‐protected amino acid (2.00 mmol) in HBTU (1.92 mmol), HOBt (2.00 mmol), along with N, N‐diisopropylethylamine (DIPEA) (4.8 mmol, 0.80 mL) dissolved in DMF (8.00 mL) and then reacted with mixing for 60 min. To validate the effective deprotection of the N‐terminal residue, a Kaiser test was implemented to detect the presence of free amines. The deprotection and coupling procedures were repeated until the DDIKVAV sequence was obtained. Upon completion, the Fmoc‐DDIKVAV peptide resin was washed with ethanol and then left to dry under a vacuum for 48 h. Cleavage of the residue was performed using a solution containing trifluoroacetic acid (TFA) and triethyl silane (TES) (2.5%) together with 2.5% distilled water for a total of 2 h, with agitation every 20 min to ensure mixing. The resin was then removed by filtration with glass wool. Excess TFA was then evaporated through nitrogen gas sparging to a final volume of 5 mL. A hydrochloric acid (HCl) salt exchange was then performed to remove any TFA from the final product. Briefly, the TFA salt of the peptide in solution was transferred dropwise to an excess of HCl (35 mL, 10 mm) heated to 50 ° C with stirring on a hot plate. The volume was then reduced via evaporation by sparging the peptide solution with nitrogen gas on a hot plate for approximately 2 h to remove the TFA. The resultant HCL salt of Fmoc‐DDIKVAV underwent precipitation in cold ether and was subsequently centrifuged five times for 12 min each, at 4000 rpm, to remove organic components. The crystalline Fmoc‐DDIKVAV peptide was then placed in a constant vacuum for a minimum of 48 h. After grinding the collected crystals to a fine powder with a metal micro‐spatula for 10 min, the peptide was further subjected to a regular vacuum for an additional seven days.

2.1.2. High‐Performance Liquid Chromatography (HPLC) and Mass Spectroscopy (MS)

A solution of 1 mg/mL concentration was prepared by dissolving 1 mg of the peptide in a mixture comprising 0.5% ammonia solution (NH₄OH), 25% acetonitrile (ACN), and 74.5% water. The peptide separation process utilized a high‐performance liquid chromatography (HPLC) system (Agilent Analytical 1290 LC System), employing a reversed‐phase C18 column (ZORBAX Eclipse Plus C18, 95 Å, 150 × 4.6 mm, 5 µm) for chromatographic analysis. The mobile phases included water with 0.1% TFA as Mobile A and ACN with 0.1% TFA as Mobile B. The mixture was filtered using a 0.22 µm PTFE filter (Merck Millipore, USA) before use. The gradient program for the mobile phase was as follows: A/B (v/v)/time (min); starting at 98:2 for 0 min, holding at 98:2 until 2 min, changing to 35:65 at 20 min, maintaining 35:65 until 23 min, and returning to 98:2 at 30 min. This was executed at a constant flow rate of 0.9 mL/min. UV detection at 280 nm was used to monitor the eluent. The column was kept at ambient temperature, and each injection volume was set to 30 µL. Peak purity was verified using a diode array detector.

2.1.3. Preparation of Fmoc‐DDIKVAV SAP Hydrogel

The hydrogelation was conducted using the pH switch technique, as detailed in previous studies [39, 46]. In brief, Fmoc‐DDIKVAV powders (10 mg) were mixed in 200 µL deionized water (DI water). Then, sodium hydroxide (NaOH, 0.5 m) was gradually added dropwise with continuous vortexing until the peptide completely dissolved. Subsequently, HCl, 0.1 m was incorporated in a similar manner until the solution's pH, measured using an Oaktron pH 700 micro pH electrode (Thermo Scientific), aligned with physiological pH levels. Finally, phosphate‐buffered saline (PBS, pH 7.4) was added to achieve the target peptide concentration of 15 mg/mL.

2.1.4. Zeta Potential Measurement for Surface Charge

The measurement of the ζ‐potential for Fmoc‐DDIKVAV SAP hydrogel was conducted using the Zetasizer Nano ZS (Redbadge) model ZEN3600 from Malvern Instruments, UK. For this procedure, the hydrogel was re‐dispersed in 1 mL of DI water with a dilution ratio of 1:100 and transferred into a capillary zeta cell using a syringe. The data gathered from these measurements were subsequently examined and interpreted through the Zetasizer Marvel Software developed by Malvern Instruments.

2.1.5. Spectroscopic Analysis of Secondary Structure

Fourier transform infrared (FTIR) spectroscopy analysis was carried out with an Alpha Platinum Attenuated Total Reflectance FTIR instrument (from Bruker Optics, Germany), using approximately 20 µL of Fmoc‐DDIKVAV hydrogel in its undiluted form. For Circular dichroism (CD) assessments, a Chirascan CD Spectrometer (from Applied Photophysics Limited) was employed, examining roughly 50 µL of hydrogel samples diluted to a 1:10 ratio in DI water. The operational settings were a pathlength of 10 mm, a step size of 0.5 nm, and a bandwidth of 1 nm. Post‐acquisition, the gathered data were averaged and refined using the Chirascan software.

2.1.6. Rheology

Rheological properties were examined using a Kinexus Pro+ Rheometer (Malvern Instruments) equipped with a flat plate setup (20 mm diameter with solvent trap, Upper Geometry: PU20 SR1351 SS, Lower Geometry: PLS55 C0177 SS), maintaining a gap size of 0.2 mm. The frequency sweeps ranged from 0.1 to 100 Hz, applying an oscillatory strain of 0.1% at a temperature of 37 ° C. These rheological tests were additionally employed to estimate the mesh size in the Fmoc‐DDIKVAV hydrogel systems.

2.1.7. SAP Hydrogel Injectability Testing (Shear Thinning and Recovery Behaviour)

To assess the shear‐thinning properties and recovery kinetics of the hydrogel, an oscillatory rheological test was conducted. Approximately 250 µL of Fmoc‐DDIKVAV hydrogel was positioned in the centre of the rheometer plate. The experiment involved conducting multiple frequency sweeps ranging from 0.1 to 100 Hz, under a constant shear strain of 0.1%. The hydrogel was subjected to step‐strain cycles, where the strain was alternated from 0.1% to 100% for a duration of 1 min at a frequency of 1 Hz, this cycle was repeated four times.

Additionally, a linear shear rate sweep was implemented, ranging from 0.01 to 100 s⁻¹, to thoroughly analyse the hydrogel's shear‐thinning behaviour.

2.1.8. Mesh Size

To assess the mesh size of the Fmoc‐DDIKVAV hydrogel, the theory of rubber elasticity (RET) analytical methodis utilized to establish a relationship between the storage modulus (G') and the mesh size (r_mesh) through a specific mathematical formula [47], where R is the gas constant, T is the absolute temperature, and N_Av is the Avogadro's number:

$$rmesh={\left(6RT\frac{6RT}{\pi NA\nu }\frac{1}{G}\right)}^{1/3}$$ (1)

This formula enables the estimation of hydrogel mesh size using measurements of G’ (storage modulus). Typically, reported mesh sizes for hydrogels fall within a range of 5 to 100 nanometers [48]. Regulating the mesh size of hydrogels involves not just the concentration of peptides and cross‐linkers but also external factors like temperature and pH.

2.1.9. Transmission Electron Microscopy (TEM)

Negative stain Transmission Electron Microscopy (TEM) was performed using a Hitachi H7100FA TEM system operating with a tungsten filament at 100 kV. Copper grids coated with Formvar were subjected to glow discharge at 15 mA for 30 s. These grids were then covered with a small amount of hydrogel (4 µL) and incubated for around 30 s, followed by a rinsing step in DI water and intermittent blotting. The negative staining of the structures involved immersing the grids in an ultra‐fine uranyl formate (UF) solution (0.75%, 25 µL) for a brief period, then blotting and repeating the immersion for another 30 s. The grids were allowed to air dry for a minimum of 2 h prior to TEM imaging.

2.1.10. Atomic Force Microscopy (AFM)

We further characterize the fabricated Fmoc‐DDIKVAV hydrogel using a Multimode 8 (Bruker Biosciences Corporation, USA) AFM system. The hydrogel was diluted ten‐fold, then dropped onto freshly cleaved mica surfaces. AFM then imaged the hydrogel with ScanAsyst‐Air probes constructed from silicon nitride in the proprietary ScanAsyst mode. The scan size was set to 5 × 5 µm.

2.1.11. Cryo‐Scanning Electron Microscopy (cryoSEM)

Small aliquots (∼20 µL) of the hydrogel samples were securely attached to a cryoSEM stage using the Tissue‐Tek O.C.T. Compound (Sakura, CA, USA) mixed with colloidal graphite. The samples were then plunged into semi‐solid nitrogen slush to rapidly freeze. The frozen samples were quickly transferred into the cryo‐chamber, where they were held at −140 °C and fractured by a cold scalpel. The samples were then sublimated by raising the temperature to −90 °C for 10 mins, before being cooled back to −140 °C. Samples were coated with platinum before being imaged. All images were taken using a Hitachi SU7000 (Hitachi, Japan) with an accelerating voltage of 5 kV.

2.1.12. Preparation of Adeno‐Associated Viral Vectors (AAVs)

AAV vector production was carried out at the Vector and Genome Engineering Facility (VEGF) located at the Children's Medical Research Institute (CMRI) in Sydney, Australia. In brief, the production process involved the transient transfection of Human Embryonic Kidney 293 (HEK293 ATCC) cells, which are adherent, using polyethyleneimine (PEI, PolyPlus, Cat#115‐100). The transfection used a molar ratio of 1:1:2 for pRep2CapX:pTransgene:pAd5. The vector particles were then isolated through a standard iodixanol gradient purification method, which included four gradient layers (15%, 25%, 40%, and 60%) comprising PBS‐MgKNa, Optiprep, and Phenol Red (the latter only in the 25% and 60% gradients). The vector particles were collected from a thin layer located between the 40% and 60% gradient buffers. For the final buffer exchange and concentration, Amicon Ultra‐4 Centrifuge Filter Units with Ultracel‐100 kDa membranes (EMD Millipore, Cat# Z648043) were used. The buffer consisted of PBS(Gibco, Cat# 14190), 50 mm NaCl (Gibco, Cat#24740), and 0.001% Pluronic F68 (v/v) (Gibco, Cat# 24040). The vector samples were quantified using a real‐time quantitative PCR master mix (qPCR, Bio‐Rad, Cat# 172–5125) with eGFP‐specific primers (Forward: 5’‐ GAGGTGAAGTTCGAGGGC‐3’; Reverse: 3’‐CTTGTGCCCCAGGATGTTG‐5’). A serially diluted linearized plasmid was used to create a standard curve for quantification, as described in previous research [49]. We achieved a viral titer of at least 2.49 × 10¹³ viral genomes (vg)/mL for rAAVDJ‐CMV‐BDNF‐P2A‐GFP.

2.2. Release Profiles

2.2.1. Release Profile of AAVDJ‐BDNF Viral Vector

To determine release patterns, rAAVDJ‐CMV‐BDNF‐P2A‐GFP (AAVDJ‐BDNF), was combined in a predetermined ratio with the Fmoc‐DDIKVAV solution before the gelation process. The resulting SAP‐virus mixtures were then set to gel using the previously described method, after which cast into a 96‐well plate. Each well was gently topped with 100 µL of PBS, and the plate was incubated at 37 ° C in a 5% CO₂ environment. At various time intervals (1, 4, 8, 24, 48, 72, and 120 h), the PBS was refreshed, and the supernatants were collected for analysis. The quantity of vectors released into these supernatants was measured using a qPCR Adeno‐Associated Virus Titration (Titer) Kit from Abcam, Australia.

2.2.2. Quantitative PCR

AAV titers were determined using the qPCR Adeno‐Associated Virus Titration (Titer) Kit (Abcam) as per the manufacturer's instructions. In brief, supernatants were mixed with Virus Lysis Buffer in a 1:1 ratio and incubated for 10 min at 70 ° C to extract viral DNA. Subsequently, a combination of 2X qPCR MasterMix (12.5 µL), Reagent Mix (10 µL), and Nuclease‐free H2O (2.5 µL) was prepared with 2.5 µL of the extracted viral DNA and either STD1 or STD2 (AAV Standard 1 or 2). The qPCR was performed on the StepOne Plus Real‐Time PCR System (Applied Biosystems) using the recommended cycling protocol: initial enzyme activation phase for 2 min at 95 ° C, 40 cycles of 15s at 95 ° C) followed by 1 min annealing/extension at 60 ° C.

2.3. Biological Testing in Vitro

2.3.1. Mouse Primary Cortical Neurons Culture

All animal procedures and techniques were carried out in compliance with the Australian National Health and Medical Research Council's (NHMRC) published Code of Practice for the Use of Animals in Research, and was confirmed by the ANU Animal Care and Use Committee (Animal Ethics Protocol Number: A2020/20). Swiss mice were provided by the Australian Phenomics Facility (APF). Mice were time‐mated overnight, and visualization of a vaginal plug the next morning classified as embryonic day 0.5 (E0.5). At E14.5 of gestation, embryos were collected from pregnant dams, decapitated, and cortices dissected and stripped of meninges. The cortices were then washed three times with Hanks’ Balanced Salt Solution (HBSS 1X, Gibco) and dissociated in 2.5% trypsin (10X, Gibco) with HBSS and DNase (10X, Sigma) at 37^° C for 20 min. After three washes in HBSS, cells were seeded at a density of 3 × 10⁵ cells/well in a 24‐well plate (previously coated overnight with Poly‐D‐lysine (PDL, 10 µg/mL) in modified Eagle's medium (MEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco) for 3 h. Subsequently, the medium was removed, and fresh serum‐free medium, including Neurobasal medium (NBM, Gibco), B27 supplement (Gibco), glutamine, and gentamicin (Gibco) added. Neurons were incubated at 37^° C with 5% CO₂ for 5 days before transduction.

2.3.2. Mouse Striatal Neuron Cells Culture (Lateral Ganglionic Eminence (LGE))

Striatal tissue was similarly obtained and dissociated from E14.5 mouse embryos. To isolate the lateral gamblionic eminence (LGE, primitive striatum), the cerebral hemispheres was lifted to expose underlying ganglionic eminences, with the larger and more dorsal eminence identified as the LGE, and accordingly dissected away from the overlying cortex [50]. The LGE tissues were then collected and dissociated in 2.5% trypsin with HBSS and DNase at 37 ° C for 15 min. After 3X HBSS washes, cells were seeded at a density of 3 × 10⁵ cells/well in 24‐well plate in MEM supplemented with 10% FBS for 3 h. After that, the medium was removed, and fresh serum‐free medium, including NBM, B27 supplement, glutamine, and gentamicin, was added. Striatal neurons were incubated at 37 ° C with 5% CO₂ for 5 days before the transfection procedure.

2.3.3. AAVDJ‐BDNF Transduction Studies in Vitro

The efficiency of transduction by the chosen AAVDJ‐BDNF was assessed on both mouse primary cortical and LGE neurons. For this purpose, cells were plated in a 24‐well plate at a density of 3 × 10⁵ cells per well. After 5 days, primary cortical or striatal neuron cells were transduced with the AAVDJ‐BDNF at the multiplicity of transduction (MOT) of 10⁴ and 10³ vg/cell, respectively. Non‐transduced cells (MOT = 0) served as a negative control. Cells were transduced in the viral vectors‐containing medium for 24 h, after which time the media was replaced with fresh serum‐ free medium containing NBM, B27 supplement, glutamine, and gentamicin. After a further 72 h, cultures were fixed using 4% paraformaldehyde (PFA), followed by immunostaining.

2.3.4. Immunofluorescence Procedures on Transduced Cells

The fixed cultures were washed with PBS (3 × 5 min), followed by overnight incubation at RT in primary antibodies (Goat polyclonal anti‐GFP (GFP, 1:1000, Abcam), Mouse monoclonal anti‐β tubulin, class III C‐term (βIII Tubulin 1:2000, Promega; neuronal marker), recombinant Rabbit monoclonal anti‐MAP2 (MAP2, 1:500, Abcam; neuronal marker), and rabbit polyclonal anti‐GABA (GABA, 1:300, Sigma–Aldrich; striatal marker)) diluted in PBST (0.3%) (Triton‐100X (Sigma–Aldrich) in PBS) and 5% normal donkey serum (DS, Merck). The following day, after PBS 5‐min washes for three times, non‐specific binding of antibodies was blocked with a blocking solution (0.3% PBST with 10% DS) at RT for 1 h with gentle shaking. Subsequently, the blocking solution was decanted, and the cells were incubated in the dark for 2 h at RT with fluorescent secondary antibody solution (Alexa Fluor 488 anti‐rabbit/mouse and 555 anti‐goat, 1:500) containing 2% DS and 0.3% PBST, on the shaker. Hoechst nuclear marker counter‐staining (Life Technology) at a dilution of 1:5000 in PBST (0.3%) was applied in the dark for 10 min under shaking. Finally, virally transduced and immunostained GFP‐positive cells were visualised using a Leica Stellaris 8 confocal microscope.

Subsequent to immunostaining and imaging, the adherent cells were manually detached using a cell scrapper and subjected to Flow Cytometry analysis (Fusion, BD Bioscience).

2.3.5. Reverse‐Transcription Polymerase Chain Reaction (RT‐PCR) Analysis

To analyse BDNF mRNA expression in transduced cells, reverse transcription (RT)‐ PCR was conducted. Briefly, the TRIZOL Plus RNA purification kit (Invitrogen, 12183555) was used to isolate total mRNA from transduced mouse primary cortical neurons and striatal cells with AAVDJ‐BDNF viral vector. RNA concentration was measured using a Nanodrop 2000C spectrophotometer (ThermoFisher). According to the manufacturer's instructions, the first‐strand cDNA was synthesized from 220 ng of total mRNA using a SuperScript VILO cDNA synthesis kit (Invitrogen, 11766050). The presence of viral‐derived BDNF was analysed by amplifying from 32 ng of total RNA using MyTaq Red Mix (Bioline, BIO‐25043/S) and primers specific for viral‐derived human BDNF primers (hBDNF‐F: 5'‐CAAACATCCGAGGACAAGGT‐3', and hBDNF‐R:3' TCGCCAGCCAATTCTCTTT‐5') and primers specific for mouse glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) (mGAPDH‐F: 5'‐GGAGAAACCTGCCAAGTATGA‐3' and mGAPDH‐R: 3'‐ GGGTGCAGCGAACTTTATTG‐5' were used. PCR reactions were performed according to MyTaq Red Mix manufacturer instructions with an annealing step at 58 °C and 35 cycles in BIO‐RAD T100 Thermal Cycler. Finally, PCR products were separated on a 1.5% agarose gel by electrophoresis, and product sizes were estimated by comparing them to a 100 bp DNA ladder (G2101, Promega). hBDNF and mGAPDH primers are estimated to amplify 617 and 440 bp products, respectively.

2.3.6. Enzyme‐Linked Immunosorbent Assay (ELISA)

An ELISA was used to quantify the protein level present in release profile samples (transduced mouse primary cortical and LGE/striatal neurons with AAVDJ‐BDNF viral vectors). Following the manufacturer's instructions, BDNF protein release was quantified using a Human/Mouse BDNF DuoSet ELISA kit (R&D Systems). Briefly, 100 µL/well of capture antibody (R&D Systems: LOT# BBL1517011) solution at 2.0 µg/mL in PBS was added to 96‐well plates and incubated at RT overnight. The next day, the wells were washed 3 times with 100 µL of washing buffer solution containing 0.05% (v/v) Tween20 in PBS, followed by 10 min gentle oscillation (this procedure was repeated after each incubation period). Afterward, to block all non‐specific bounds, the wells were incubated for 1 h at RT with 100 µL/well of blocking reagent solution (1% (w/v) BSA blocking solution in PBS). Next, 100 µL/well of all samples and standard solutions (23.4‐1500 pg/mL) were added and incubated for 2 h at RT, followed by 2 h incubation in 100 µL/well of the detection (primary) antibody (R&D Systems: LOT#BAL0517011) solution at 25.0 ng/mL in blocking reagent solution at RT. Then, the streptavidin‐HRP (secondary) antibody (R&D Systems: LOT# P139521) was diluted 200‐fold in blocking reagent solution, added to each well (100 µL/well), and incubated at RT for a further 20 min. The plates were covered with aluminium foil to decrease light exposure after introducing a secondary antibody until the end of the procedure. Finally, a colour reagent solution was prepared by mixing colour reagent A (PART# 895000) and colour reagent B (PART# 895001), and 100 µL of it was added to each well, which was allowed to react for 20 min at RT. This reaction was then stopped by adding (50 µL/well) stop solution (2N Sulfuric acid, PART# 895926), and absorbance at 450 nm was immediately measured using a Tecan plate reader spectrophotometer. A calibration curve of standard solutions was applied to quantify the protein present.

2.3.7. Post‐Release Transduction Efficiency of the AAVDJ‐BDNF Viral Vector on Different Rodent Cells In Vitro

To verify the functionality of the AAVDJ‐BDNF after release from hydrogel, the supernatant was collected from the cast AAV + hydrogel mixture and subsequently tested on in vitro cultures of mouse primary cortical neurons and striatal cells. For this experiment, 1 µL of the AAVDJ‐BDNF, with a concentration of 2.49 × 10¹³ vg/mL, was incorporated in Fmoc‐DDIKVAV solution prior to the gelation process. The resultant SAP‐viral vector solutions were then gelled, transferred to a 96‐well plate, and allowed to stabilize. After an hour, 100 µL of PBS was applied to each hydrogel, and the plates were incubated at 37 ° C in a 5% CO₂ atmosphere. The PBS supernatants were collected at 24‐hour intervals over five days and stored at −80 ° C for future use.

Mouse primary cortical neurons and striatal neuron cells were cultured separately in 24‐well plates at a density of 3 × 10⁵ cells per well for five days. The collected PBS supernatants were then added to these confluent cell cultures, followed by a 24 h incubation period at 37 ° C with 5% CO2. After 24 h of transduction, the viral medium was removed, and the cells were incubated for an additional 72 h. The transduction efficiency was then assessed by fixing the cells with 4% PFA and performing immunostaining for neuronal or striatal markers as detailed earlier. The cells were visualized using a Leica Stellaris 8 confocal microscope. Then, cells were manually detached and further subjected to Flow Cytometry analysis (Fusion, BD Bioscience).

2.4. In Vivo Studies

2.4.1. Animals

In vivo animal experiments were approved by the Florey Institute of Neuroscience and Mental Health animal ethics committee. Animals were grouped housed in individually ventilated cages on a 12:12 h light/dark cycle with ad libitum access to food and water. A total of 25 adult female Swiss mice were used with animals distributed across the following groups (n = 5 mice/group): QA; QA + AAV‐GFP; QA + SAP‐AAV‐GFP; QA + AAV‐BDNF; and QA + SAP‐AAV‐BDNF.

2.4.2. Surgical Procedures

All animal surgeries were performed under general anaesthetic using 2%–5% isoflurane inhalation (Baxter; Deerfield, IL, USA). According to allocated groups, mice received an intrastriatal injection (1 µL) of the AAVs (in the presence or absence of the SAP hydrogel) at the following stereotaxic coordinates: 1.0 mm anterior, 2.0 mm lateral to bregma, and 3.0 mm below the dura surface. After 13 days, a single unilateral injection of quinolinic acid (QA, Sigma–Aldrich; 60 nmol in 0.5 µL, at the same stereotaxic coordinates) was used to create an excitotoxic insult and local inflammatory response within the striatum (mimicking a mouse model of HD).

2.4.3. Tissue Processing and Immunohistochemistry

Immunohistochemistry was performed to assess the ability of the AAVDJ‐BDNF viral tool alone and in combination with the Fmoc‐DDIKVAV hydrogel system to protect local striatal neurons and prevent tissue atrophy. In brief, 6weeks after QA injection, all mice received a lethal overdose of sodium pentobarbitone (100 mg/kg, Virbac) and were transcardially perfused with warmed (37 ° C) Tyrode buffer followed by 30 mL of chilled PFA (4% w/v). The brains were subsequently post‐fixed in the same fixative for 2 h before immersion in 20% sucrose solution overnight. The brains were then serially sectioned at 40 µm (1:12 series) using a freezing microtome (Leica).

For immunohistochemistry, sections were washed in PBS prior to the addition of a blocking solution (5% donkey serum and 0.3% Triton‐X‐100 (Ameresco, USA)) for 20 mins at RT. Subsequently, primary antibodies were added, including: Rabbit polyclonal to NeuN marker (1:1000, Abcam, to label neurons), Chicken polyclonal to GFP (1:1000, Abcam, to identify transduced cells), and Rabbit anti‐Iba1 (1:500, WAKO, 019–19741, to resolve reactive microglia within the host brain).

The following day, the sections were washed (3 × 5 min in PBS), followed by incubation in a blocking solution for 1 h. Tissues were then incubated with the following secondary antibodies: Alexa Fluor 488 anti‐rabbit (1:200) or 647 anti‐chicken (1:200) for 2 h. Finally, the tissue was washed (3 × 5 min in PBS), incubated in Hoechst (Life Technology) to identify nuclei, washed and the tissue mounted on gelatinized slides using a fluorescent mounting medium (Dako, Australia). Sections were visualized under a Zeiss LSM 800 Airyscan Microscope. Tissue atrophy (neuronal cell loss revealed by the absence of NeuN+ cells) calculated according to Cavalieri's principal using section thickness, the sum of areas depleted from NeuN+ cells, and the interval. Microglial density (Iba‐1) was measured as optical density at different fields of view (FOV) lateral of the needle stab injury (20x) and analysed using ImageJ.

2.5. Statistical Analysis

Statistical analysis of the data was conducted using one‐way analysis of variance (ANOVA), followed by Tukey's post hoc tests, utilizing OriginLab software. The levels of statistical significance between different groups were determined as follows: ^* p <0.05, ^** p <0.01, ^*** p <0.001, and ^**** p <0.0001. Additionally, the results are presented as the mean ± standard deviation (SD).

3. Results

3.1. In Vitro AAVDJ‐BDNF Transduction on Mouse Primary Cortical Neurons and Striatal Neurons

Transduction efficiency of the rAAVDJ‐CMV‐BDNF‐P2A‐GFP (AAVDJ‐BDNF) on mouse primary cortical neurons (CTX) as well as lateral ganglionic eminence (LGE)/medium spiny striatal neurons (MSN) were assessed in vitro (Figure 1). CTX and LGE were examined for multiplicity of transduction (MOT) = 10⁴ vg/cell (vector per cell dose), this MOT resulted in cell toxicity in LGE cells. Therefore, the subsequent experiments were conducted employing MOT = 10⁴ and 10³ for CTX and LGE, respectively. Transduced cells were visualized using GFP, AAVDJ‐BDNF reporter (Figure 1A), immunolabeling 72 h post‐transduction. As illustrated in Figure 1B–D, AAVDJ‐BDNF could effectively transduce both CTXs and LGE neurons. This visual observation for transduction efficiency is supported by quantification of cells co‐expressing GFP (AAV reporter) together with neuronal markers β3‐tubulin (immature neurons, green), MAP2 (mature neurons, green) or GABA (LGE‐derived striatal neurons, magenta) (Figure 1E,F). Flow cytometry analysis of GFP positive cells revealed 79.1 ± 5.4% and 56.6 ± 1.8% of CTX and LGE cells were transduced with AAVDJ‐BDNF, respectively. As expected, CTX cells exhibited higher transduction efficiency compared to LGE as higher MOT was used.

FIGURE 1.

In vitro transduction of Primary Cortical Neurons (CTX), and striatal neurons (lateral Ganglionic Eminence (LGE)) using rAAVDJ‐CMV‐BDNF‐P2A‐GFP (AAVDJ‐BDNF) viral vector. (A) Schematic of the AAVDJ‐BDNF construct. A representative image of (B,C) transduced CTX and (D) LGE cells stained with transduction efficiency marker GFP (green), Map2/ β3‐Tubulin neuronal markers (red), GABA (magenta) and Hoechst (Blue). CTX were transduced with 10⁴ viral genomes (vg)/cell, and LGE cells underwent transduction with 10³ vg/cell of AAV‐BDNF. (E) Flow Cytometry data indicating the proportion of cells positive for both GFP and neural markers (β3‐tubulin) as well as GABA in the AAVDJ‐BDNF viral vector system. (F) Quantification of the percentage of transduced cells corresponding to the representative images, showing co‐expression with neural and striatal markers. The data is presented as the mean ± SD. Analysis was conducted using ANOVA followed by Tukey's post‐hoc test for statistical differentiation between groups, with ^**** indicating p <0.0001. Scale bar = 100 µm.

3.2. BDNF Expression Analysis in Cells

To determine if the AAVDJ‐BDNF transduced cells expressed hBDNF mRNA, RT‐PCR was performed as shown in Figure 2A. In agreement with our previous findings from in vitro transduction efficiency experiments, the AAVDJ‐BDNF transduced CTX and LGE cells expressed significantly higher intensity of hBDNF mRNA compared to non‐transduced CTX and LGE cells (MOT = 0). The mGAPDH was used as a housekeeping control. Next, to determine if the expressed hBDNF protein can also be released, the ratio of BDNF protein release present in the supernatant samples from AAVDJ‐BDNF transduced CTX and LGE cells was quantified using a Human/Mouse BDNF ELISA kit. As shown in Figure 2B, both transduced CTX and LGE cells significantly (p<0.0001 for transduced CTX, and p<0.01 for transduced LGE cells) released BDNF protein compared to the control samples (MOT = 0), which are in agreement with the results achieved from hBDNF mRNA expression test (RT‐PCR analysis). To confirm the efficient transduction of CTX and LGE with AAVDJ‐BDNF and the ability to promote the hBDNF protein secretion, next we investigated the Fmoc‐DDIKVAV SAP hydrogel's capability to mimic the brain's mechanical properties and the viral vectors entrapment.

FIGURE 2.

BDNF mRNA and protein expression in transduced cells. (A) RT‐PCR analysis of the hBDNF mRNA expression analysed by agarose gel electrophoresis. Control, non‐transduced mouse primary cortical neurons as well as striatal cells; AAVDJ‐BDNF viral vector. According to the results, no hBDNF was detected in non‐transduced primary cortical neurons and striatal neuron cells (control). However, hBDNF was detected in transduced primary cortical neuron cells and striatal neurons by the AAVDJ‐BDNF viral vector tool. (B) Assessment of BDNF protein release via ELISA. ELISA results indicate a notable release of BDNF protein in supernatant samples from primary cortical and striatal neurons infected with AAVDJ‐BDNF viral vector, in comparison to non‐transduced cells (control). The data is shown as mean ± SD. Statistical analysis was performed using ANOVA followed by Tukey's post‐hoc test to identify significant differences between groups, with ^**** indicating p <0.0001 and ^** signifying p <0.01.

3.3. Fmoc‐DDIKVAV Hydrogel Biomaterial System Characterization

To confirm the purity of the peptide and therefore the assembly, HPLC (Figure 3A) and MS (Figure 3B) analyses of the Fmoc‐DDIKVAV biomaterial system showed a single peak, confirming a high purity (>94%) and only a single peptide species. The expected mass (980.49 m/z) closely matched the observed mass (981.1 m/z), further confirming the integrity of the peptide.

FIGURE 3.

Fmoc‐DDIKVAV hydrogel characterization. (A) HPLC analysis of Fmoc‐DDIKVAV. The chromatogram shows a distinct peak, indicating the sample's purity. (B) MS profile of Fmoc‐DDIKVAV hydrogel system. The spectrum highlights a predominant single component along with minor degradation products. (C) Optical image of generated Fmoc‐DDIKVAV. (D) The FTIR spectra of Fmoc‐DDIKVAV hydrogel system show a major peak around 1630 cm^{− 1} and a minor peak at 1690 cm^{− 1}, indicating anti‐parallel β‐sheet arrangements. (E) CD spectra of Fmoc‐DDIKVAV reveal β‐sheet structures, with significant transitions below 220 nm. (F) The TEM image reveals that the nanofibers are intertwined, creating fine fibrils; the scale bar represents 200 nm. (G) cryoSEM image of the nanofibers demonstrating the nano and microstructural network formed by the fibres. (H) The AFM image demonstrates the nanofibrous architecture of Fmoc‐DDIKVAV hydrogel. (I) The mesh size (ξ) (nm) of Fmoc‐DDIKVAV was determined using the equation from the theory of rubber elasticity. (J) Assessment of surface ζ‐potential in Fmoc‐DDIKVAV system using zetasizer. (K) Rheological analysis verifies that the fabricated hydrogel exhibits viscoelastic properties, as indicated by a storage modulus (G′) that exceeds the loss modulus (G″). (L) Conducting an oscillatory rheological test to observe the variations in the hydrogel's modulus over time, further illustrating its shear‐thinning characteristics. (M) Analysis of the SAP hydrogel's recovery process. This involved initially applying a low shear rate of 0.01 s⁻¹ for 30 s, followed by a high shear rate of 100 s⁻¹ for another 30 s, and concluding with a return to the initial low shear rate for 15 min. This procedure helps to understand how the hydrogel regains its original viscosity after the cessation of stress, (N) AAVDJ‐ release characteristic from the Fmoc‐SAP hydrogel biomaterial. The release pattern of the AAVDJ‐BDNF was analyzed through qPCR, quantifying vector‐encoded DNA in the Fmoc‐DDIKVAV hydrogel at specified intervals (1, 4, 8, 24, 48, 72, and 120 h). The presented data reflect the mean + SD, with three replicates at each timepoint. Statistical analysis was performed using ANOVA followed by Tukey's post‐hoc test to identify significant differences between groups, with ^**** indicating p < 0.0001 and ^** signifying p < 0.01.

Then the structure of the Fmoc‐DDIKVAV SAP hydrogel was examined through a series of characterization experiments. The optical images of the fabricated Fmoc‐DDIKVAV hydrogel confirmed a stable and clear gel formation buffered to pH 7.4 (Figure 3C). To confirm the formation of the supramolecular assembly and secondary structure of the fabricated hydrogel, FTIR and CD spectroscopic characterization techniques was used. As illustrated in Figure 3D, the FTIR spectra of the single component Fmoc‐DDIKVAV hydrogel between 1550 and 1750 cm⁻¹, showed characteristic major and minor peaks, which is absent before gel formation. The major peak around 1630 cm⁻¹ and a minor peak around 1690 cm⁻¹ correspond to the formation of predominantly anti‐parallel β‐sheets among the peptide sequences. Additionally, a smaller peak observed near 1660 cm⁻¹ indicates the presence of α‐helical structures within the π‐β configuration. CD spectroscopy analysis of the hydrogel, as shown in Figure 3E, supports the existence of a predominantly β‐sheet structure within this system. This is evidenced by the cotton transitions occurring in the 190–200 nm region, reinforcing the data obtained from FTIR analysis. Next, TEM was used to demonstrate and validate the formation of the interconnected nanofibrous structures that underpin the Fmoc‐DDIKVAV hydrogel system (Figure 3F). A stable, homogenous nanofibrous network was observed for the hydrogels, with a branched intertwined architecture. This shows that the non‐covalent interactions observed spectroscopically drove the development of highly ordered nanofibrous structures that were strong enough to overcome changes in peptide charge, sequence, and length. Further, wet atomic force microscopy (AFM) confirmed the higher order of the nanofibrous bundled structure of the hydrogel in their hydrated, aqueous state as shown in Figure 3H. This was further confirmed through cryoSEM, which demonstrated a nanofibrous structure that bundled into a higher order structure, creating a hierarchical nano and micro‐porous structure (Figure 3G).

The size of the mesh formed by the nanofibers is crucial as it dictates the diffusion of solutes through the hydrogel, particularly when larger and charged molecules can form steric interactions with the peptide network. Herein, as shown in Figure 3I, the mesh size (ξ) (nm) value of the Fmoc‐DDIKVAV calculated was 20 nm based on the equation of the theory of rubber elasticity (Equation 1). It is well‐established that the size of AAV is 24–26 nm [51]. When the mesh size is smaller than the payload size, the payload is physically entrapped inside the network of hydrogel. In this case, the mesh is non‐covalently formed, meaning the release of the immobilized payload can happen through a variety of mechanisms such as diffusion, network degradation, swelling, or deformation of the network [51, 52]. This method is widely used to control drug release, where smaller mesh sizes proportionally slow the release of AAV from hydrogels. A slower release rate of AAVs is ideal, as it is anticipated to increase transduction efficiency. Previous studies conducted by our research team have demonstrated the proficient viral binding ability of our hydrogels, particularly those incorporating the IKVAV sequence, using lentivirus particles [32]. Next, the peptide sequence‐specific surface charge of the hydrogel was evaluated using a Marvel Zetasizer at a dilution factor of 1/100 to measure the ζ‐potential value, as shown in Figure 3J. The data reveal that Fmoc‐DDIKVAV exhibits an intermediate negative ζ‐potential value. It is important to acknowledge the role of surface charges, as they are provided by the amino acids in the sequence, and as such play a crucial part in not only biological signalling, but must also provide the electrostatic bonding and stabilization of the constituent components of the self‐assembled materials.

The utility of hydrogels mimicking the physical features of the brain relies significantly on mechanical factors such as it's stiffness and elasticity [53]. To determine these mechanical properties of the system, a rheological analysis was performed, as shown in Figure 3K. Here, it was observed the hydrogel system was stableover a rangeof frequencies, with an elastic modulus of ∼1kpa with mechanical characteristics within the typical stiffness range for the rodent brain. In order to implant the hydrogel using a microinjection protocol, it is important that the material can flow under the shear generated from a syringe, and then recover its mechanical properties upon the conditions experienced post implantation. To evaluate the shear‐thinning and recovery behaviour of the Fmoc‐DDIKVAV hydrogel biomaterial, an injectability test was performed where the material was allowed to gel, then a step‐strain cycle was applied to fully break the material down and the reformation and final gelation was recorded (Figure 3L). This SAP system exhibits shear‐thinning properties, meaning its gel viscosity diminishes as the shear rate increases, facilitating injectability. Importantly, the mechanical properties were restored over four cycles, confirming the stability of the material once applied to the brain via a single injection.

3.4. Release Profile of AAVDJ‐BDNF Viral Vector

To ensure a consistent and uniform dispersion of the vector particles, the AAVDJ‐BDNF was thoroughly mixed under vortexing into the Fmoc‐DDIKVAV prior to the gelation phase to enable a sustained and controlled delivery of the AAVDJ‐BDNF vector. Then the release profile of the AAVDJ‐BDNF was analyzed using quantitative Real‐Time PCR (qPCR), as depicted in Figure 3N to ensure the delivery during the predetermined timeframes required.

The experimental results showed that the Fmoc‐DDIKVAV hydrogel exhibited an initial burst release at early time points, followed by a stable plateau phase at 72 and 120 h. This release profile indicates that the hydrogel enables an early, consistent release of AAVDJ‐BDNF and then maintains a controlled, sustained dispersal over time. This is optimal as the extended timeframe of sustained and metered delivery potentially enhances the efficacy of the vector delivery in targeted gene therapy applications.

3.5. Transduction Efficacy After Release of AAVDJ‐BDNF From the Fmoc‐SAP System (In Vitro)

We assessed the transduction capabilities of the AAVDJ‐BDNF targeting CTX as well as LGE cells, following its delivery from the SAP‐AAV hydrogel. The SAP‐AAV hydrogel demonstrated a significant and sustained transduction efficiency of the viral vector over a span of 5 days. This was evidenced in both CTX, as highlighted in Figure 4A and LGE cells as detailed in Figure 4B.

FIGURE 4.

Hydrogel encapsulated AAVDJ‐BDNF prolonged payload release while maintaining its biofunctionality. Analysis of AAVDJ‐BDNF viral vector's effectiveness post‐release from Fmoc‐DDIKVAV hydrogel on (A) CTX and (B) LGE over 5 days. The neurons were immunostained with GFP (green) to highlight GFP‐tagged virally transduced cells, and counterstained with Hoechst (blue) for total cell visualization. Additionally, the cortical and striatal neurons were marked with the neural marker MAP2 (red) as well as the GABA striatal marker (magenta), respectively. The intense GFP fluorescence indicates a high transduction efficiency of both cells by AAV, confirming the virus's viability and efficacy after being released from the Fmoc‐SAP system, Scale bar = 50 µm. (C) Representative Flow Cytometry data showing the proportion of GFP and MAP2 positive cells and GFP and GABA co‐positive cells. (D) The analysis of transduction levels, expressed as the percentage of neurons positive for GFP and the designated neuronal and LGE markers. The results are presented as the mean ± SD, with three samples for each timepoint.

The observed post‐release transduction efficiency of the AAVDJ‐BDNF is a critical indicator of the viral vector's viability and functional activity when delivered through the Fmoc‐DDIKVAV hydrogel system (Figure 4C,D). Notably, the peak of the viral release, and consequently its highest transduction potential, was recorded after 5 days. The findings underscore the hydrogel system's ability not only to maintain the viability and activity of the viral vector but also to effectively control its release, thereby enhancing its transduction efficiency in target neural cells.

3.6. SAP‐Delivery of AAV‐BDNF Enhances Striatal Neuroprotection Against QA‐Induced Excitotoxicity

Next, we performed in vivo studies (Figure 5A) to determine whether the direct injection of the AAVDJ‐BDNF viral vector into the mouse striatum, or its delivery in combination with the Fmoc‐DDIKVAV hydrogel, could result in efficient transduction of the host, and subsequently the functional capability of sustained BDNF to protect local neurons from neurodegeneration induced by an excitotoxin (QA) challenge.

FIGURE 5.

The synergistic effect of AAVDJ‐BDNF and SAP hydrogel reduces tissue atrophy. (A) Schematic overview of the designed experiment. (B) Quantification of striatal tissue atrophy corresponding to the representative images (C), shown as the volume (mm³) of NeuN‐negative tissue surrounding the injection site in mice implanted with the indicated materials. The analysis was conducted using ANOVA complemented by Tukey's post‐hoc test to discern statistical variances between groups, with significance levels indicated by ^**** p <0.0001, ^*** p <0.001, ^** p <0.01. Notably, the least amount of tissue degeneration was observed in animals that received QA + AAVDJ‐BDNF+SAP post QA lesioning. (C) In vivo implants of different viral vectors with or without Fmoc‐SAP system in the mouse striatum 8 weeks post‐implantation illustrate the effects of different materials on Huntington's Disease (HD) mouse models. Immunostaining highlights the impact of QA‐induced lesions in the striatum (right hemisphere), evident by the absence of NeuN‐positive cells (green) in the core lesion area. Contrastingly, in mice receiving QA lesions followed by AAVDJ‐BDNF and Fmoc‐DDIKVAV implantation, there is a notable reduction in lesion size and increased neuroprotection, as seen by the denser population of NeuN‐positive cells near the injection site. Scale bars in the images are set at 500 µm.

In mice injected with QA alone, the lesion encompassed most of the striatum, as indicated by the lack of NeuN‐positive cells compared to the contralateral hemisphere (Figure 5B,C). On the contrary, it is obvious that the size of the lesion declined in animals that received identical QA lesions yet received prior injections of AAVDJ‐BDNF viral vector alone or in combination with the Fmoc‐DDIKVAV hydrogel system. The core of the lesion in these animals was minimal and restricted to a small area surrounding the injection site but still, numerous healthy NeuN‐positive perikarya and dendrites adjacent to the needle track were identified immediately outside of this central core, followed by animals treated with alone AAVDJ‐BDNF viral tool reaching the second best position in surviving neurons. As can be seen in the animals that received an injection of empty vector (AAVDJ‐GFP) without Fmoc‐DDIKVAV hydrogel, indicated the lowest rate of NeuN‐positive cells within the needle track among the other groups.

To compare tissue atrophy among different groups, the volume of the area devoid of NeuN‐positive neuron cells around the lesion site (mm³) was measured, which estimates the neural loss generated by QA and the protection mediated by implants of AAVDJ‐BDNF alone and in conjunction with Fmoc‐DDIKVAV delivery system. This analysis confirmed the lowest rates of tissue atrophy (neural loss) in the striatum of AAVDJ‐BDNF plus Fmoc‐DDIKVAV or AAVDJ‐BDNF alone‐treated animals, respectively, compared to the control animals that received either QA injection alone or plus an empty viral vector (AAVDJ‐GFP) with or without Fmoc‐DDIKVAV hydrogel delivery system (Figure 5B). Cell transduction can trigger cellular stress responses, change cell membrane permeability, interfere with normal endocytosis and trafficking pathways, and elicit an immune response, all of which may affect how transduced cells take up external substances like QA [54]. Herein, the lower tissue atrophy seen in the empty vector‐treated group compared with QA treated animals could be related to the level of QA uptake by transduced cells. Additionally, in mice that received the AAV‐GFP through SAP hydrogel, greater cell survival was observed. The SAP hydrogel may provide physical buffering or limit QA diffusion, thereby passively reducing the extent of lesion propagation, supporting neural survival, and secondary neuronal loss. This study is the first demonstration that in vivo targeted delivery of BDNF neurotrophic factor using AAV‐DJ viral vehicle can produce a significant protection of neurons following QA‐lesioning. In addition, the highest protection of NeuN‐positive neuron cells by Fmoc‐DDIKVAV+ AAVDJ‐BDNF highlights the potential for Fmoc‐SAP hydrogel biomaterials to not only deliver a functional virus but also provide considerable support to vulnerable neuron cells against neurotoxic insults.

To investigate whether the implantation of AAVs ± SAP hydrogel exacerbates the immune response in vivo, the Iba‐1 antibody was used to detect local microglia. We have studied the activation of microglia in close proximity of the injection site as well as the whole right hemisphere of the brain as illustrated in Figure 6.

FIGURE 6.

Incorporation of AAVDJ‐BDNF within SAP hydrogel attenuates the host inflammatory response. (A) Representative overview of the core section of the brain in the 5 different groups stained for Iba‐1 (red) and GFP (green), Scale bar = 500 µm. (B) Quantification of microglial activation density within the right hemisphere of the core sections corresponding to the images in panel A. (C,D) Quantified measurements of the density of microglial activation within FOV1 and FOV2 in the striatum of mice corresponding to the images in panel G,H. Data are represented as mean ± SD (n = 4 per group). ^** p <0.01, ^*** p <0.001, ^**** p <0.0001. (E) Schematic diagram of the overview of the images section (shown as magenta) and selected field of view (FOV) for analysis and higher magnification images. (F) Representative images of the core section of the injection site in different groups stained for Iba‐1 (magenta), scale bar = 500 µm. (G,H) High magnification of the fluorescent images of the different FOV1 and FOV2 of different groups stained for Iba‐1, Scale bar = 100 µm.

As shown in Figure 6A, animals receiving QA alone exhibited the strongest microglial activation within the right hemisphere, consistent with the high Iba‐1 intensity observed in the representative images. This qualitative increase is supported by the quantitative measurements in Figure 6B. In contrast, the lowest inflammatory response was detected in animals treated with AAVDJ‐BDNF delivered within the SAP hydrogel prior to QA, demonstrating significantly reduced microglial activation compared with animals that received AAVDJ‐BDNF without the SAP gel (Figure 6B). This confirms that the incorporation of the Fmoc‐SAP hydrogel delivery system can provide support for AAVDJ‐BDNF in accordance with decreasing unwanted immune responses.

We also examined two distinct fields of view (FOV) perpendicular to the injection site to investigate the activation of Iba‐1 in the vicinity of the implant. As illustrated in Figure 6G,H, FOV1 demonstrated that the injection of AAV‐GFP (empty vector) resulted in the highest intensity of microglial activation compared to all other experimental groups, significantly surpassing even the QA group (Figure 6C). This finding underscores the critical role of SAP hydrogel in shielding the AAVs from triggering immune responses. Furthermore, the density of inflammatory cells was notably reduced in the AAVDJ‐BDNF + SAP group across both FOV1 and FOV2 (Figure 6C,D). This observation, in conjunction with the tissue atrophy results presented in Figure 5, confirms the presence of a conducive microenvironment facilitated by the synergistic effects of AAVDJ‐BDNF and SAP hydrogel.

In summary, the results presented provide a robust foundation for the continued investigation of our engineered AAV delivery system within the protective milieu of injectable and tissue‐matched Fmoc‐SAP biomaterials. The observed success in transduction, BDNF expression, sustained release, and in vivo therapeutic outcomes collectively highlight the potential of this system as a transformative approach in the realm of gene therapy for otherwise untreatable neurodegenerative disorders such as Huntington's disease.

4. Conclusions

Here, we provide evidence that Fmoc‐SAP biomaterials can provide an effective tissue‐tuned material for the delivery of AAV. In vitro experiments demonstrated significant transduction of rodent cells, highlighting the efficacy of our engineered system in delivering genetic material to target cells. This achievement is particularly noteworthy, as effective transduction is a critical step in ensuring the therapeutic impact of gene therapy.

In this study, we used intrastriatal quinolinic acid (QA) injection as a pharmacological model of Huntington's disease (HD)‐ like excitotoxicity. The QA model produces rapid and reproducible striatal neuronal loss and, in our hands, was used to mimic imminent neuronal injury. The gel was applied before QA administration, meaning that our experiments primarily address the neuroprotective potential of the gel at the time of the insult, rather than the rescue of already established degeneration. A limitation of this work is that we did not test the gel in transgenic ‘standard’ HD models that exhibit slowly progressive neurodegeneration and behavioural decline, nor did we assess therapeutic delivery after neuronal loss has occurred. Future studies will extend this proof‐of‐concept by applying the gel in genetic HD models and in delayed‐treatment paradigms, with longitudinal behavioural and histopathological readouts and tracking the protein levels of BDNF in the brain to evaluate its ability to slow or rescue ongoing neuronal degeneration.

Importantly, the virally transduced cells exhibited not only successful transduction but also the continued expression and subsequent release of conformationally viable BDNF protein. This dual functionality indicates that our engineered system offers the therapeutic goal of neuroprotection as it not only delivers the protective genetic material but also facilitates the expression and release of a key neurotrophic factor.

The prolonged release of the virus observed after its incorporation within Fmoc‐SAP biomaterials is a crucial finding. This sustained release profile suggests that the biomaterials contribute not only to the protection of the viral vectors from immune clearance but also to the preservation of their functionality post‐release improving the probability of transfection of co‐localised cells. The retained functional morphology of the Fmoc‐SAPs post‐release is a promising feature, as it ensures that the biomaterial continues to exert its biomimetic structure and activity even after the initial delivery of the genetic payload.

Moving beyond the relatively static in vitro setting, the dynamic in vivo experiments provide compelling evidence of the biomaterial system's therapeutic potential in an active tissue. The observed neural protection, delivery of a functional virus, and the reduction in unwanted immune responses in vivo collectively validate the efficacy of our engineered AAV delivery Fmoc‐SAP biomaterial environment.

These results not only advance our understanding of the potential applications of gene therapy in neurodegenerative disorders but also emphasize the critical role that advanced tissue tuned biomaterials such as Fmoc‐SAPs to attenuate the host response, play in unlocking the precision and effectiveness of gene delivery systems. The success observed in these experiments positions our developed system as a promising candidate for further development toward clinical applications, ultimately offering a breakthrough technology in the treatment of conditions like Huntington's disease. (Supporting Information S1)

Funding

Australian Research Council FT230100220, National Health and Medical Research Council, GNT1135657

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: smll72542‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.