Local delivery of RhoA siRNA by PgP nanocarrier reduces inflammatory response and improves neuronal cell survival in a rat TBI model

Abstract

Traumatic brain injury (TBI) is a leading cause of death and disability with complex pathophysiology including prolonged neuroinflammation, apoptosis, and glial scar formation. The upregulation of RhoA is a key factor in the pathological development of secondary injury following TBI. Previously, we developed a novel cationic, amphiphilic copolymer, poly (lactide-co-glycolide)-graft-polyethylenimine (PgP), as a nanocarrier for delivery of therapeutic nucleic acids. In a rat compression spinal cord injury model, delivery of siRNA targeting RhoA (siRhoA) by PgP resulted in RhoA knockdown; reduced astrogliosis and inflammation; and promoted axonal regeneration/sparing. Here, we evaluated the effect of RhoA knockdown by PgP/siRhoA nanoplexes in a rat controlled cortical impact TBI model. A single intraparenchymal injection of PgP/siRhoA nanoplexes significantly reduced RhoA expression, lesion volume, neuroinflammation, and apoptosis, and increased neuronal survival in the ipsilateral cortex. These results suggest that PgP/siRhoA nanoplexes can efficiently knockdown RhoA expression in the injured brain and reduce secondary injury.

Graphical Abstract

Introduction

Traumatic brain injury (TBI) remains a leading cause of death and disability worldwide with an estimated 10 million new cases each year.¹ In the United States, the Centers for Disease Control and Prevention estimates that over 2.5 million new TBI events occur annually.² The pathology of TBI can be divided into primary and secondary injury stages. The primary injury is caused by mechanical trauma leading to immediate cell death and blood brain barrier (BBB) dysfunction. Secondary injury can persist for months after the initial trauma and provides a window for therapeutic intervention. Secondary injury includes chronic edema, neuroinflammation, decreased cAMP level, dysregulation of ion transport, excitotoxicity, and glial scar formation that result in progressive neuronal and glial cell death.^3–6 Additionally, damaged neurons have limited capacity for regeneration and repair due to potent growth inhibitory signaling from molecules such as Nogo,⁷ myelin-associated glycoprotein, oligodendrocyte myelin glycoprotein,⁸ and chondroitin sulfate proteoglycans (CSPGs).⁹ These molecules exert their inhibitory action commonly through the RhoA/Rho-associated protein kinase (ROCK) signaling pathway.^6,10,11 Secretion of the pro-inflammatory cytokine tumor necrosis factor-alpha (TNF-α) by reactive astrocytes and microglia can also potently activate RhoA/ROCK signaling.^8,11 Activation of the RhoA/ROCK pathway promotes axonal and synaptic retraction,^12,13 induces cell death,^14–16 inhibits axonal regrowth and repair,^17,18 and can modulate astrocyte and microglial reactivity through effects on stress fiber and focal adhesion assembly.^19–21

Inhibition of RhoA/ROCK signaling can enhance axonal regeneration and repair, inhibit apoptosis, reduce neuroinflammation, and improve functional recovery.^8,16,22 RhoA/ROCK inhibition is generally accomplished using pharmaceutical inhibitors including RhoA antagonistic statins or C3 transferase,^23,24 and the ROCK inhibitors, fasudil,^14,22 Y27632,^25–27 or H-1152P.^14,28,29 However, these inhibitors can exhibit dose-dependent off-target effects that can influence therapeutic outcomes.^24,30–32 The ROCK inhibitor Y27632 also inhibits protein kinase N (PRK2) at high doses.^27,31 Additionally, fasudil can exhibit dose-dependent specificity for protein kinase C (PKC) and AMP-activated protein kinase (AMPK) in addition to inhibition of ROCK.^31,33,34 Studies with fasudil in vivo demonstrate variable effects on glial activity with moderate functional recovery.^35–37 Reproducible cell-specific effects are a vital consideration for orchestrating neuroprotection and repair after CNS injury and exclusive inhibition of RhoA/ROCK signaling activity is more ideal for optimizing combinatorial therapeutic strategies with reproducible therapeutic effects in TBI.

RNA interference (RNAi) by small interfering RNAs (siRNAs) is a post-transcriptional gene silencing technology capable of target-specific knockdown of mRNA expression.³⁸ Delivery of siRNAs in vivo is challenging due to endogenous nuclease degradation, activation of an immune response through the toll-like receptor (TLR) pathway, and poor cell entry due to charge repulsion.³⁸ Nanoparticle carriers (nanocarriers) are powerful and versatile tools for surmounting the challenges to siRNA delivery in vivo. Nanocarriers can stably complex siRNA and protect it from nuclease degradation, reduce immune activation, and facilitate intracellular delivery.³⁸ Delivery of siRNA by nanocarriers has achieved efficient knockdown of various pathological targets^39,40 accompanied by reduced neuroinflammation and improved neuroprotection. To our knowledge, this strategy has not yet been investigated for therapeutic knockdown of RhoA for treatment of TBI despite success in spinal cord injury.^41,42

In our group, we have developed a cationic amphiphilic copolymer, poly (lactide-co-glycolide)-graft-polyethylenimine (PgP) for combinatorial delivery of multiple therapeutic agents to sites of CNS injury.^41,43–45 PgP nanocarriers form micellar nanoparticles in aqueous solution and have three important capabilities: 1) electrostatic complexation of therapeutic nucleic acids with the cationic hydrophilic shell, 2) loading of hydrophobic drugs in the hydrophobic core, and 3) surface conjugation of cell-type specific targeting moieties such as ligands or antibodies. In our previous work, we synthesized and characterized PgP nanocarriers demonstrating efficient transfection of plasmid DNA and siRNA.⁴³ We demonstrated that RhoA knockdown by PgP/siRhoA nanoplexes reduced necrotic cavitation and astrogliosis and increased axonal sparing/regeneration in a rat compression spinal cord injury (SCI) model.⁴¹Here, we investigated the effect of RhoA knockdown by PgP/siRhoA polyplex nanoparticles (nanoplexes) on neuroinflammation and neuronal cell survival in a rat controlled cortical impact (CCI) TBI model in vivo. RhoA knockdown by intraparenchymal injection of PgP/siRhoA nanoplexes in the injury site resulted in decreased lesion volume, inflammatory response, and apoptosis and increased neuronal cell survival compared to untreated TBI animals at 7 days post-injury (DPI).

Methods

Materials

Poly (lactide-co-glycolide) (PLGA) (~4 kDa, 50:50) was purchased from Durect Corporation (Cupertino, CA). Anhydrous dimethylformamide (DMF), N-hydroxysuccinimide (NHS), N,N’-dicyclohexylcarbodiimide (DCC), branched polyethylenimine 25 kDa (bPEI), heparin sodium salt from porcine intestinal mucosa, and deuterium oxide (D₂O) were obtained from Sigma-Aldrich (St. Louis, MO). Dialysis tubing (molecular weight cutoff: 50,000) was obtained from Spectrum labs (Rancho Dominguez, CA). The Silencer® Pre-designed siRNA targeting RhoA (ras homolog family member A, NCBI Reference Sequence: NM_057132.3, siRhoA),Silencer Negative Control siRNA (siNT), and RNase A were purchased from ThermoFisher Scientific (Hampton, NH). RNeasy plus mini kit and QuantiTect® SYBR Green PCR Kit were purchased from Qiagen (Valencia, CA). Cresyl violet 0.1% acetic acid pH 1.5 solution was obtained from Polyscientific R&D Corp. (Bay Shore, NY). Mouse monoclonal anti-neuronal nuclei (NeuN), mouse monoclonal anti-macrophage/monocyte clone-ED1 antibodies, and the ApopTag® Fluorescein In Situ Apoptosis Detection Kit were obtained from EMD Millipore (Darmstadt, Germany). Rabbit monoclonal anti-GFAP antibodies were purchased from Abcam (Cambridge, MA). The goat anti-rabbit Dylight-488 secondary antibodies were obtained from Fisher Scientific (Hampton, NH). Donkey anti-mouse Cy3 conjugated IgG antibodies were purchased from Jackson ImmunoResearch (West Grove, PA).

Stability of PgP/siRhoA nanoplexes

The cationic amphiphilic copolymer, PgP (poly (lactide-co-glycolide)-graft-polyethylenimine) was synthesized using PLGA (4 kDa, 50:50) containing carboxylic ends groups and branched PEI (bPEI, 25 kDa) as previously described.⁴³ In our previous study, we observed that PgP/siRhoA nanoplexes at N/P (number of nitrogen atoms in polymer/number of phosphorus atoms in nucleic acid) ratio of 30/1 showed the highest knockdown efficiency in rat neuroblastoma (B35) cells in vitro and in a rat spinal cord injury model in vivo,⁴¹ therefore for this study we prepared PgP/siRhoA nanoplexes (1 μg siRhoA) at N/P ratio of 30/1. Brieflly, PgP/siRhoA nanoplexes at N/P ratio of 30/1 were prepared by adding 50 μl siRhoA (1 μg siRhoA) solution into 50 μl PgP (6 μg PgP) solution with gentle mixing and then incubated at 37 °C for 30 minutes. To assess stability, PgP/siRhoA nanoplexes were incubated with heparin at varying heparin/siRhoA ratios (0 to 8 w/w) at 37 °C for 30 minutes. Samples were immediately electrophoresed on a 2 % (w/v) agarose gel containing ethidium bromide (0.5 μg/mL) for 60 min at 80 V and the gel was imaged on a UV illuminator (Chemidoc-It, UVP).

We also examined the ability of PgP to protect siRhoA from nucleases after incubation in the presence of serum and RNase A by gel electrophoresis analysis. PgP/siRhoA N/P 30/1 nanoplexes (1 μg siRhoA) were incubated at 37 °C for 1 hour in solutions containing 10 % FBS or 0.45 μg RNase A /μg siRNA. For comparison, naked siRNA was incubated under the same conditions. After incubation, PgP/siRhoA nanoplexes were treated with 0.4 % sodium dodecyl sulfate (SDS) to dissociate the siRhoA from PgP, and samples were analyzed by electrophoresis as described above. As controls, naked siRNA and non-dissociated PgP/siRhoA nanoplexes were incubated in water.

Controlled cortical impact TBI model and PgP/siRhoA nanoplex treatment

All surgical procedures and postoperative care were conducted according to NIH guidelines for care and use of laboratory animals under the supervision of the Clemson University Animal Research Committee (Approved Clemson IACUC animal use protocol (AUP) #2015–082). Male Sprague Dawley rats (~350 g, Charles Rivers) were anesthetized with 2–4 % isoflurane gas. The head was shaved, swabbed with betadine, and secured in a stereotaxic frame (Kopf Instruments, Tujunga, CA). A midline incision was made in the scalp and soft tissue was retracted using a micro-retractor to expose the skull surface. A 5 mm circular craniotomy was made on the right hemisphere centered at 3.5 mm lateral and 3.5 mm posterior of the bregma without disturbing the underlying dura. The injury was performed with a TBI impactor (Precision Systems and Instruments, Fairfax Station, VA) using a 3 mm diameter flat tip (impact speed = 3.5 m/s, dwell time = 250 msec, depth = 2 mm). Following injury, hemostasis was achieved by applying GelFoam® Sterile Sponge (Patterson Vet Generics, Devens, MA).

Rats were randomly divided into 4 groups: 1) sham group (Sham), 2) untreated TBI group (TBI), 3) PgP/RhoA siRNA nanoplex injected TBI group (PgP/siRhoA), 4) PgP/non-targeting siRNA nanoplex injected TBI group (PgP/siNT). PgP/siRhoA and PgP/siNT nanoplexes were prepared at 30/1 N/P ratio (20 μg siRhoA/rat) as described above. Immediately after injury a total of 20 μl nanoplex solution or saline (untreated TBI group) was administered by intraparenchymal injection using a Hamilton syringe (30G, beveled tip) at four pre-determined positions surrounding the lesion epicenter at 2 mm depth (5 μl per injection: 3 mm lateral × 3 mm posterior, 3 mm lateral × 4 mm posterior, 4 mm lateral × 3 mm posterior, and 4 mm lateral × 4 mm posterior). Injections were performed at an injection rate of 1 μl/min using a Legato 130 microinjector (KD Scientific, Holliston, MA). After injection, the bone flap was replaced but was not secured with bone cement to relieve intracranial pressure. The scalp was closed with 4–0 vicryl suture. Cefazolin (40 mg/kg, Hikma Farmaceutic) and buprenorphine (0.01 mg/kg, Hospira Inc) were administered after surgery and animals were warmed by a heating pad for recovery.

Analysis of RhoA Knockdown by RT-PCR

To evaluate RhoA knockdown efficiency by PgP/siRhoA nanoplexes in the TBI lesion site, rats (total n=12, n=3 rats/group) were euthanized by CO₂ overdose at 7DPI, the brain retrieved, and the ipsilateral cortical tissue surrounding the injury site was collected and snap-frozen with liquid nitrogen. Total RNA was isolated by RNeasy mini kit (Qiagen, Hilden, Germany) and the quality and quantity of total RNA were evaluated by Take3 (BioTek, Winooski, VT) using a BioTek Synergy microplate reader (Synergy HT, BioTek). Complementary DNA (cDNA) was synthesized by reverse transcription using 0.5 μg total RNA (Retroscript kit, Ambion). Real-time PCR (RT-PCR) was performed using a SYBR Green PCR kit (Qiagen) with target-specific primers in a Rotor-Gene Q Thermocycler (Qiagen). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an endogenous control. The primer sequences for RhoA: forward primer: 5’-TTCGGAGTCGTCGTCTTGAG-3’, reverse primer: 5’-CCACAAGCTCCATCACCAAC-3’. The primer sequences for GAPDH: forward primer: 5’-ATGGCCTTCCGTGTTCCTAC-3’; reverse primer: 5’-TAGCCCAGGATGCCCTTTAG-3’. Relative mRNA expression of RhoA was calculated using the 2^−ΔΔCt method.⁴⁶ Minus reverse transcriptase reactions were performed on a subset of samples to demonstrate a lack of genomic DNA contamination (data not shown).

Tissue Preparation for Immunohistochemistry (IHC)

To evaluate the effect of RhoA knockdown by PgP/siRhoA on the inflammatory response, neuronal cell survival, and apoptosis, immunohistochemistry was performed. Rats (total n=12, n=3 rats/group) were euthanized by cardiac perfusion with 0.9 % saline followed by 4 % paraformaldehyde (PFA) under deep isoflurane anesthesia. The brain was retrieved and post-fixed in the 4 % PFA at 4 °C followed by cryoprotection in 30 % sucrose solution. Brains were rapidly frozen using Ice-IT spray (ThermoFisher Scientific), sectioned coronally (30 μm thick) using a cryostat, and stored at −20 °C in cryoprotectant solution (30 % sucrose, 1 % polyvinylpyrrolidone, and 30 % ethylene glycol in 1X PBS).

Nissl staining

Coronal brain sections were stained for Nissl-bodies with 0.1 % cresyl violet for 10–20 min, rinsed with water and then dehydrated by ethanol gradient. The tissue sections were cleared with xylene and coverslipped with resinous mounting medium (Azer Scientific, Morgantown, PA). Images were taken using an inverted bright-field microscope (Leica Microsystems, Buffalo Grove, IL).

Analysis of RhoA knockdown by IHC

Coronal brain sections were evaluated for RhoA protein expression by immunofluorescent staining (total n=9 sections: n=3 sections/rat, n=3 rats/group). Briefly, sections were mounted to slides, post-fixed with 4 % PFA for 10 minutes, washed with PBS, and incubated with blocking solution (5 % bovine growth serum and 0.05 % Triton X-100 in PBS) for 1 hour at room temperature. The sections were washed, incubated with rabbit anti-RhoA polyclonal antibodies (1:200, ThermoFisher Scientific) overnight at 4°C, and washed before incubation with AlexaFluor® 594-conjugated goat anti-rabbit secondary antibodies (1:2000, Jackson Immunoresearch Laboratories Inc.) for 1 hour at room temperature. Tissue sections were coverslipped using VectaShield mounting media with DAPI (Vector Laboratories, Burlingame, CA). Images were taken using an Axiovert 40 CFL fluorescence microscope (Carl Zeiss, Oberkochen, Germany). The RhoA knockdown at the protein level was quantified by measuring integrated fluorescence intensity of stained tissue sections using ImageJ with normalization to the sham group.

Effect of RhoA knockdown on lesion volume

To measure the effect of RhoA knockdown on lesion cavity volume, brain sections were selected from 2 to 4.5 mm posterior of the bregma with 0.25 mm spacing (n=10 sections/rat, n=3 rats/group) and stained for Nissl-bodies as described above. Cavity areas were measured using ImageJ software and the lesion volume was calculated by Cavalieri’s approximation using the equation $V=d\left({\sum }_{i=1}^{n}yi\right)-\left(t\right)y_{\mathit{\text{max}}}$ where d is the distance between sections, y_i is cross-sectional area in the i-th section, n is the total number of sections, t is the section thickness, and y_max is the maximum possible value of y (defined as the cross-sectional area of the impacting tip at a depth of 2 mm).⁴⁷

Effect of RhoA knockdown on secondary injury

The effect of RhoA knockdown by PgP/siRhoA nanoplexes on secondary injury such as apoptosis, inflammatory response, astrogliosis, and neuronal cell death was evaluated. Briefly, tissue sections (total n=9 sections/group: n=3 sections/rat, n=3 rats/group) were randomly selected 3–4 mm posterior from the bregma for IHC staining. For detection of apoptotic cells, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining was performed using an ApopTag® Plus Fluorescein in situ Apoptosis Detection Kit (EMD Millipore). For detection of activated microglia/macrophages, activated astrocytes, and neuronal nuclei, sections were incubated with mouse monoclonal anti-macrophage/monocyte antibody, CD68 clone-ED1 (1:200, MAB1435, EMD Millipore), polyclonal rabbit anti-GFAP antibody (1:200, ab7260, Abcam), or mouse monoclonal anti-NeuN antibody (1:200, MAB377, EMD Millipore), respectively. Sections were washed and incubated for 1 hour at room temperature with either goat anti-mouse Cy3-conjugated secondary antibody (1:200, 115-165-003, Jackson Immunoresearch) or AlexaFluor 488-conjugated goat anti-rabbit secondary antibody (1:500, A11008, Invitrogen). Finally, sections were washed and coverslipped using VectaShield mounting media with DAPI. Images of the medial border of the lesion were captured using an Axiovert 40 CFL fluorescence microscope. The number of total DAPI positive (+), TUNEL+, CD68 (ED1)+, or NeuN+ cells were counted using ImageJ. The number of DAPI+ cell nuclei was used to calculate the percentage of TUNEL+ or CD68 (ED1)+ cells for each evaluated image. For the activated astrocytes, integrated fluorescence intensity of GFAP-stained tissue sections was measured using ImageJ with normalization to the sham group.

Statistics

All data are presented as mean ± standard deviation (SD). Statistical comparison was performed in SPSS 14.0 software using one-way ANOVA with pairwise comparisons determined by Tukey’s HSD post hoc analysis. The adjusted p-values were calculated by the software. A p-value of less than 0.05 was considered significant.

Results

Stability of PgP/siRhoA nanoplexes

Previously, we demonstrated that PgP nanocarriers form stable complexes with RhoA siRNA at N/P ratios 10/1 or above and PgP/siRhoA nanoplexes showed the highest knockdown efficiency without significant cytotoxicity at an N/P ratio of 30/1.⁴³ Based on this data, we prepared PgP/siRhoA nanoplexes at N/P ratio 30/1 for all the experiments in this study. We first evaluated the stability of PgP/siRhoA nanoplexes at N/P ratio of 30/1 by incubation with varying heparin/siRhoA (w/w) ratios. PgP/siRhoA nanoplexes were completely stable up to 2/1 heparin/siRhoA (w/w) ratio, started to dissociate at ratio of 3/1, and were completely dissociated at ratios of 5/1 or higher (Figure 1A). Next, the ability of PgP to protect siRhoA from nucleases was examined by incubation of naked siRhoA and PgP/siRhoA nanoplexes in the presence of serum or RNase A (Figure 1B). To visualize the intact siRhoA, we dissociated the siRhoA from PgP using SDS prior to electrophoresis. Intact siRhoA bands were observed from PgP/siRhoA nanoplexes incubated with serum and RNase A (lanes 4 and 6), while naked siRhoA incubated with serum and RNasA was completely degraded (lane 3 and 5). Untreated, naked siRhoA migrated through the gel (lane 1) and PgP/siRhoA nanoplexes (lane 2) were retarded in the well and used for comparison.

Figure 1.

Characterization of PgP/siRhoA nanoplex stability by 2 % agarose gel electrophoresis (A) Stability of PgP/siRhoA nanoplex by heparin competition assay using varying heparin/siRNA (0–8 w/w) ratios (B) siRhoA integrity after incubation of naked siRhoA and PgP/siRhoA in the presence of 10 % serum (lanes 3 and 4) or RNase A (lanes 5 and 6). (M: Molecular weight marker; lanes 1, 3, 5: Naked siRhoA; lanes 2, 4, 6: PgP/siRhoA nanoplexes). PgP/siRhoA nanoplexes in lanes 4 and 6 were dissociated with 0.4% SDS after incubation.

RhoA knockdown by PgP/siRhoA nanoplexes

The relative RhoA expression was evaluated by RT-PCR at the mRNA level and IHC at the protein level at 7 DPI of PgP/siRhoA nanoplexes. RhoA mRNA expression was not significantly different in PgP/siRhoA nanoplex treated group compared to that in sham group, while RhoA mRNA expression was significantly increased in the untreated TBI group and the PgP/siNT treated group relative to the sham animal group (Figure 2A). Treatment with PgP/siRhoA nanoplexes resulted in significant knockdown of RhoA mRNA expression compared to the untreated group expression, but the PgP/siNT nanoplex treatment group was not significantly different from the untreated TBI group (Figure 2A). Figure 2B shows representative images of immunohistochemical staining of RhoA protein expressed in brain tissue from the four groups. The normalized fluorescence intensity in the PgP/siRhoA nanoplex treatment group was not significantly different from that in the sham group, while the normalized fluorescence intensity in the untreated TBI animal group and PgP/siNT treated group was significantly higher than that in the sham group. (Figure 2C).

Figure 2.

RhoA knockdown by PgP/siRhoA nanoplex at 7 DPI in TBI lesion site. A) Relative RhoA mRNA expression level determined by RT-PCR. Sham, untreated TBI, and PgP/siNT treated animal groups were used for comparison (n=3/group; One-way ANOVA, Post-hoc: Tukey’s HSD, F_3,8 = 26.369, p<0.0001; ###p<0.001, ##p<0.01, compared to sham and **p<0.01 compared to untreated TBI). (B) Representative images of Nissl-stained coronal brain slices in the ipsilateral cortex. Images of sections stained for RhoA (red) and cell nuclei (DAPI, blue) in the perilesional area denoted by red box in Nissl-stained images. Black scale bar indicates 1 mm and white scale bar indicates 100 μm. (C) Average fluorescence intensity of sections stained for RhoA, normalized relative to the sham group (n=3/group, n=3 sections/rat; One-way ANOVA, Post-hoc: Tukey’s, F_3,8=8.962, p=0.006; #p<0.05 compared to sham).

RhoA knockdown by PgP/siRhoA nanoplexes on lesion cavity volume

Figure 3A shows representative images of Nissl-stained brain sections from 2 to 4.5 mm posterior from the bregma from the four groups. We observed extensive cavity formation in the untreated TBI group. The average lesion cavity volume in the PgP/siRhoA nanoplex treated group was significantly reduced compared to the untreated TBI group. In contrast, the average lesion cavity volume in the PgP/siNT group was not significantly different compared to untreated TBI animals (Figure 3B).

Figure 3.

Lesion cavity volume following RhoA knockdown by PgP/siRhoA nanoplex at 7 DPI. (A) Representative images of Nissl stained brain sections (30 μm) in the ipsilateral cortex. Scale bars indicate 1 mm. (B) Average lesion volumes calculated by Cavalieri approximation using lesion area measurements from ten 0.25 μm spaced sections. (n=3 rats/group, n=3 sections/rat); One-way ANOVA, Post-hoc: Tukey’s, F_{2, 6}=5.426, p=0.045, *p<0.05 compared to TBI)

RhoA knockdown by PgP/siRhoA nanoplexes on secondary injury

The effects of RhoA knockdown by PgP/siRhoA nanoplex on secondary injury were evaluated by IHC at 7 DPI. We first evaluated the effect of RhoA knockdown on the inflammatory response by staining coronal brain sections for CD68 (ED1). Figure 4A shows representative images of CD68 (ED1)+ cells in the ipsilateral cortex from all animal groups. In the untreated TBI group, we observed extensive activation of inflammatory microglia/macrophages in the ipsilateral cortex compared to the sham animal group. Following RhoA knockdown by PgP/siRhoA, the percentage of CD68 (ED1)+ cells was significantly reduced in the ipsilateral cortex compared to untreated TBI animals (Figure 4B).

Figure 4.

Reduced neuroinflammatory response following RhoA knockdown by PgP/siRhoA at 7 DPI. (A) Representative images of Nissl-stained brain sections. Images of immunofluorescent staining of CD68 (ED1, red) and cell nuclei (DAPI, blue) in the perilesional area denoted by red boxes in Nissl-stained images. Black scale bar=1 mm and white scale bars =100 μm. (B) The average percentage of CD68 (ED1)+ cells in each image area. (n=3 rats/group, n=3 sections/rat; One-way ANOVA, Post-hoc: Tukey’s, F_{2, 6}=17.736, p=0.003, **p<0.01 compared to TBI, #p<0.05 compared to PgP/siNT)

To examine the effect of RhoA knockdown on astrocyte activation at 7 DPI, we stained for GFAP positive (GFAP+) astrocytes. In the untreated TBI and PgP/siNT nanoplex treated groups, we observed numerous GFAP+ astrocytes that exhibited cellular hypertrophy and process thickening in the lesion border of the ipsilateral cortex and the normalized fluoresecence intensity was significantly higher than the sham group. The normalized fluorescence intensity in PgP/siRhoA nanoplex treated group was not significantly different than the sham group and was significantly lower than the untreated TBI group (Figure 5).

Figure 5.

Reduced astrogliosis following RhoA knockdown by PgP/siRhoA at 7 DPI. (A) Representative images of Nissl-stained brain sections. Images of GFAP+ (green) reactive astrocytes in the perilesional area denoted by red boxes in Nissl-stained images. White boxes in GFAP images indicate the location of the proceeding magnification. Black scale bar=1 mm. White scale bars=100 μm. (B) Average fluorescence intensity of GFAP-stained sections normalized to that of sham group (n=3 rats/group, n=3 sections/rat; One-way ANOVA, Post-hoc: Tukey’s, F_3,8=7.290, p=0.011; ### p<0.001, ##p<0.01 compared to sham; *p<0.05 compared to TBI; Ŧp<0.05 compared to siNT).

Figure 6A shows representative images of TUNEL positive (TUNEL+) cells in the medial border of the lesion for all animal groups. There was significant apoptosis detected in the ipsilateral cortex in the untreated TBI group. The percentage of TUNEL+ cells in the ipsilateral cortex was significantly decreased in PgP/siRhoA nanoplex treated animals relative to untreated TBI animals (Figure 6B).

Figure 6.

Reduced cellular apoptosis following RhoA knockdown by PgP/siRhoA at 7 DPI. (A) Representative images of Nissl-stained brain sections. Images of TUNEL+ (green) and cell nuclei (DAPI, blue) in the perilesional areas denoted by red boxes in Nissl-stained images. Black scale bar =1 mm. White scale bars =100 μm. (B) The average percentage of TUNEL+ cells was calculated in each image area (n=3 rats/group, n=3 sections/rat; One-way ANOVA, Post-hoc: Tukey’s, F_2,6=5.361, p=0.046, *p < 0.05 compared to TBI)

We also determined the effect of RhoA knockdown by PgP/siRhoA nanoplex treatment on neuronal cell survival at 7 DPI. Figure 7A shows representative images of neuronal nuclei (NeuN) stained cells in the ipsilateral cortex. The number of NeuN-positive cells was significantly decreased in all groups including PgP/siRhoA treatment group compared to that in sham group. However, the number of NeuN-positive cells was significantly increased in the PgP/siRhoA and PgP/siNT treatment groups compared to untreated TBI animal group (Figure 7B).

Figure 7.

Increased neuronal cell survival following RhoA knockdown by PgP/siRhoA at 7 DPI. (A) Representative images of Nissl-stained brain sections. Images of sections stained for NeuN (red) and cell nuclei (DAPI, blue) in the perilesional areas denoted by red boxes in Nissl-stained images. Black scale bar=1 mm. White scale bars=100 μm. (B) The average number of NeuN+ cells per image area. (n=3 rats/group (n=3 sections/rat)), One-way ANOVA, Post-hoc: Tukey’s, F_3,8=21.642, p<0.0001; ***p<0.0001, *p<0.01 compared to sham, #p<0.05 compared to TBI)

Discussion

The pathological mechanisms underlying secondary injury progression following TBI are complex and heterogeneous. The numerous cascades initiated during secondary injury are intended to stabilize the injury and protect undamaged tissue. Aberrant activation of microglia and astrocytes contributes to progressive inflammatory signaling, lesion growth, apoptotic cell death, and neurite growth inhibition.^3–6,48,49 Many of the pathological mechanisms of secondary injury converge on the RhoA/ROCK signaling pathway.^8,14,31,36 Consequently, ROCK and its primary effectors/affecters (RhoA, CSPGs, PTEN, and SOCS3) are important targets for pharmacological intervention.

Inhibition of RhoA/ROCK is generally accomplished by pharmacological inhibitors which can reduce neuroinflammation, increase neuronal survival and improve functional recovery in experimental models of CNS injury.^22,36,50,51 One alternative approach for reducing RhoA/ROCK signaling post-injury is sequence-specific RNA interference using siRNA that can facilitate post-transcriptional knockdown of RhoA mRNA expression. Nanocarriers are often used to deliver therapeutic siRNA in vivo as a means of circumventing endogenous nuclease degradation, inflammatory response, and charge repulsion with the cell membrane.^38,52 In animal models of optic nerve and spinal cord injury, RhoA knockdown enhanced axonal regeneration, improved neuronal cell survival, and reduced inflammation and astrogliosis.^41,42,53 To our knowledge, the present study represents the first report of a polymeric micelle nanocarrier for RhoA siRNA delivery in experimental TBI.

Previously, we developed a cationic, amphiphilic polymer, PgP as a therapeutic nucleic acid and drug delivery carrier. We have demonstrated that PgP can deliver siRNA targeting RhoA to the injured rat spinal cord in vivo and RhoA knockdown by PgP/siRhoA nanoplexes reduced astrogliosis and necrotic cavitation, and also increased axonal sparing/regeneration in a rat compression spinal cord injury model.^41,43 In our previous studies, PgP/siRhoA nanoplexes at N/P ratio 30/1 exhibited sub-200 nm diameter and maintained positive surface charge that is important for intracellular uptake by endocytosis. Formation of stable complexes that can protect siRNAs against nuclease degradation is an important feature for efficient siRNA nanocarriers. Here we demonstrated that PgP/siRhoA nanoplexes are stable against heparin competition (Figure 1A) and can protect siRNA cargo from nuclease degradation in serum and RNase (Figure 1B). We next evaluated the knockdown of RhoA by PgP/siRhoA nanoplexes in the lesion site at 7 DPI. This time point was chosen based on literature^11,54 and our previous study in an SCI model.⁴¹ We observed a significant increase in RhoA mRNA expression in the untreated TBI group and PgP/siNT group at 7 DPI compared to the sham group. The RhoA mRNA expression was significantly reduced in the PgP/siRhoA nanoplex treatment group compared to the untreated TBI group and not significant from the sham group (Figure 2A). In contrast, the PgP/siNT treatment was not significantly different from the untreated TBI group. RhoA protein expression in the ipsilateral cortex was determined by IHC staining against RhoA. The normalized fluorescence intensity in the PgP/siRhoA treated group was reduced compared to that in untreated TBI and PgP/siNT treated group and was not significantly different with that in the sham group (Figure 2B&C). These data indicate that RhoA knockdown by PgP/siRhoA is attributable to sequence specific knockdown.

Upregulation of RhoA/ROCK signaling influences the pathological progression of TBI including neuroinflammation, glial scar formation, and apoptosis. The ultimate consequence of unrestrained secondary injury following focal TBI is a progressive loss of cortical tissue.^48,55 Measurement of lesion volume is often used to quantify tissue sparring as a metric for determining therapeutic success.^55–57 In our study, the lesion volume was significantly reduced following RhoA knockdown by PgP/siRhoA nanoplexes compared to the untreated TBI group at 7 DPI (Figure 3) indicating that PgP/siRhoA treatment promoted sparing of cortical tissue.

Mounting evidence suggests that improvements in neuron survival can be mediated by suppression of microglia and astrocyte activation, thereby inhibiting the development of a neurotoxic environment.^5,49 RhoA signaling mediates the cytoskeletal reorganization required for pathological changes in inflammatory cells and astrocytes, and when properly regulated can enhance tissue sparing and encourage a neuroprotective environment.^31,42,49,58 We observed a significant reduction in CD68 (ED1)+ activated microglia/macrophages within the ipsilateral cortex following PgP/siRhoA treatment (Figure 4). These results are supported by observations in studies of CNS injury where RhoA knockdown⁴² and ROCK pharmacological inhibition^20,58 reduced microglia activation.

The formation of an astrocytic scar is a pathological response that is closely associated with the extent of neuroinflammation. In the untreated TBI animals, we observed significantly increased fluorescence intensity of GFAP+ astrocytes along with thickening of astrocytic processes compared to sham controls indicating reactive astrocytes (Figure 5). Animals treated with PgP/siRhoA nanoplexes exhibited fewer reactive astrocytes at the lesion periphery and fluorescence intensity was not significantly different from the sham group (Figure 5B). This is consistent with other studies that demonstrate reduced reactive astrocyte proliferation and scar formation following Rho/ROCK inhibition.^24,36,41

Upregulation of RhoA signaling can increase apoptosis in neuronal cells through associated activation of PTEN and subsequent inhibition of Akt activity which supports cell survival.^11,54 Consequently, inhibition of Rho/ROCK activation can reduce apoptotic response after CNS injury.^11,54 In the present study, evaluation of cellular apoptosis and neuron cell survival by IHC revealed that RhoA knockdown by PgP/siRhoA nanoplexes significantly decreased cellular apoptosis (Figure 6) and improved neuron cell survival (Figure 7). The neuroprotection and suppression of apoptosis we observed are consistent with studies utilizing ROCK inhibitors in models of CNS injury in vivo.^28,59,60 We observed that neuronal cell survival was also increased in the PgP/siNT treatment group. In our previous study, we observed that the cationic polymer, PgP only treatment was not toxic and improved survival of primary cerebellar granular neurons cultured in hypoxia condition in vitro.⁴⁵ Therefore, we think that PgP may have an effect on neuron survival in PgP/siNT treated group.

In conclusion, we demonstrate that PgP nanocarriers can efficiently bind and protect siRNA cargo and thereby mediate efficient knockdown of RhoA in the injured brain. We show that a single intraparenchymal injection of PgP/siRhoA nanoplexes in the ipsilateral cortex results in significant knockdown of RhoA expression at 7 DPI and subsequently decreased lesion volume, reduced inflammatory cell infiltration, reduced astrogliosis, and enhanced neuroprotection. The ultimate goal of our research is to evaluate PgP as a targeted combinatorial therapeutic delivery carrier for siRNA and drugs. In the future, we will evaluate the synergistic effect of the siRhoA and the phosphodiesterase inhibitor, rolipram on secondary injury and motor and cognitive funciton recovery in a rat TBI model.