Focused Ultrasound-Mediated Blood-Brain Barrier Opening and Long-Circulating Nanoparticles for Systemic Nucleic Acid Delivery to and Genome Editing in the Brain

Abstract

We introduce a two-pronged strategy comprising focused ultrasound (FUS)-mediated blood-brain barrier (BBB) opening and long-circulating biodegradable nanoparticles (NPs) for systemic nucleic acid delivery to the brain. Biodegradable poly(β-amino ester) polymer-based NPs were engineered to stably package various types of nucleic acid payloads and to enable prolonged systemic circulation while retaining excellent serum stability. FUS was applied to a predetermined coordinate within the brain to transiently open the BBB, thereby allowing the systemically administered long-circulating NPs to traverse the BBB and accumulate in the FUS-treated brain region where plasmid DNA or messenger RNA (mRNA) payloads produced reporter proteins in astrocytes and neurons. In contrast, poorly circulating and/or serum-instable NPs, including the lipid NP analogous to a platform used in clinic, were unable to provide efficient nucleic acid delivery to the brain regardless of the BBB-opening FUS. The marriage of FUS-mediated BBB opening and the long-circulating NPs engineered to co-package mRNA encoding CRISPR-associated protein 9 and single-guide RNA resulted in genome editing in astrocytes and neurons precisely in the FUS-treated brain region. The combined delivery strategy provides a versatile means to achieve efficient and site-specific therapeutic nucleic acid delivery to and genome editing in the brain via the systemic route.

Graphical Abstract

INTRODUCTION

Nucleic acid therapy, such as gene therapy and editing, has emerged as a powerful therapeutic modality to potentially treat otherwise incurable brain disorders spanning malignant brain tumors to neurodegenerative diseases.^1-3 Successful nucleic acid therapy, regardless of administration route, requires widespread therapeutic coverage throughout the lesioned areas within the brain.⁴ We have previously demonstrated that nanoparticles (NPs) possessing dense surface coatings with poly(ethylene glycol) (PEG) polymers minimize the adhesive interactions of NPs with brain extracellular matrix (ECM) to enhance NP dispersion in the brain.^5-9 In particular, we showed that NPs comprising PEG-coated surface and biodegradable poly(β-amino ester) (PBAE) polymer/plasmid DNA (pDNA)-complexed core (PEG-PBAE/pDNA NPs) provided widespread transgene expression in healthy or tumor-bearing rodent brains following intracranial administration.^{6, 10} This localized administration method, however, involves an invasive surgical procedure which often causes deleterious effects, including infection, edema, and neuronal damage.¹¹ The limitation can be circumvented by systemic administration, but NPs must address a series of biological barriers, including avoidance of liver accumulation, efficient circulation to cerebral capillaries, and penetration through the blood-brain barrier (BBB), to reach target disease areas and cells within the brain.^{12, 13}

The BBB, a tightly sealed multicellular layer lining the cerebral capillaries.^{12, 13} is the arguably the most challenging barrier to the delivery of therapeutics and NPs to the brain via the systemic route.¹⁴ Molecular targeting and biochemical/osmotic modulation have been explored over the past few decades to enhance the delivery of NPs across the BBB, but the approaches entail several limitations, including suboptimal delivery efficacy, lack of targetability, and safety issues.^{15, 16} More recently, we and others have investigated the use of transcranial focused ultrasound (FUS) as a non-invasive means to open the BBB to promote the penetration of systemically administered NPs into the brain.^17-19 The technique involves perturbation of the BBB by activating intravascular gas-filled microbubbles (MBs; FDA-approved contrasting agents for ultrasound imaging) with FUS applied at a predetermined coordinate within the brain.²⁰ The MBs passing through the cerebral capillaries then oscillate stably in the FUS focal region and generate acoustic radiation force that acts on the vessel wall to open the BBB in a transient, reversible, and targeted manner.^{18, 21, 22} Encouragingly, FUS-mediated BBB opening has been well tolerated among patients with brain disorders in multiple clinical studies^23-26 and has been recently shown to significantly enhance the systemic delivery of albumin-bound paclitaxel (i.e., Abraxane^®) across the BBB in patients with recurrent glioblastoma.²⁷

To fully exploit the FUS-mediated BBB opening and NP penetration, it is conceivable that systemically administered NPs should be able to stably circulate during the time window of the BBB opening. NPs will also need to retain the particle diameters small enough to fit in the vascular openings created by FUS to traverse the BBB. We hypothesized that PEG-PBAE NPs, due to the dense surface PEG coating, would resist non-specific adhesive interactions with hepatic cells and serum proteins to stably circulate and retain colloidal stability following systemic administration. We thus engineered PEG-PBAE NPs to carry either pDNA or messenger RNA (mRNA), extensively characterized the physicochemical properties and physiological stability in vitro, and investigated the ability to stably circulate after system administration in vivo. We then tested whether FUS-mediated BBB opening promoted accumulation of systemically administered PEG-PBAE NPs and production of reporter proteins encoded by respective nucleic acid payloads precisely in the FUS-treated brain region. Finally, we engineered PEG-PBAE NPs to co-package mRNA encoding clustered regularly interspaced palindromic repeats (CRISPR) associated protein 9 (Cas9) and single-guide RNA (sgRNA) and investigated the ability of the NPs, in conjunction with FUS, to mediate somatic genome editing in the brain in vivo.

RESULTS

PEG-PBAE NPs retain the physicochemical properties and colloidal stability in physiological conditions relevant to systemic NP delivery to the brain.

PBAE and PEG-PBAE polymers were synthesized and characterized as we previously reported.⁶ To engineer PEG-PBAE NPs carrying various nucleic acid payloads, we compacted either pDNA or mRNA with a blend of PBAE and PEG-PBAE polymers at a range of polymer-to-nucleic acid weight ratios. We confirmed with gel electrophoresis migration assay that pDNA or mRNA was fully compacted by the polymer mixture at the polymer-to-nucleic acid weight ratio of 60:1 to form pDNA- or mRNA-loaded PEG-PBAE NPs (PEG-PBAE/pDNA or PEG-PBAE/mRNA NPs, respectively) (Figure 1A). In parallel, un-PEGylated PBAE NPs carrying pDNA or mRNA (PBAE/pDNA or PBAE/mRNA NPs, respectively) were prepared with PBAE polymers only at the same weight ratio. Transmission electron microscopy (TEM) revealed that PEG-PBAE NPs, regardless of the type of nucleic acid payloads, exhibited spherical morphology and < 100 nm geometric diameters (Figure 1B). PEG-PBAE/pDNA and PEG-PBAE/mRNA NPs showed comparable hydrodynamic diameters of 61.4 ± 2.5 and 64.8 ± 3.1 nm and ζ-potentials of 2.1 ± 0.3 and 2.5 ± 0.4 mV, respectively (Figure 1C and Table 1). In comparison, PBAE/pDNA and PBAE/mRNA NPs were slightly larger than their respective PEGylated NP counterparts with average hydrodynamic diameters of ~70 nm and possessed positively charged surfaces with ζ-potentials of ~20 mV (Table 1). We next assessed the particle colloidal stability in artificial cerebrospinal fluid (aCSF) by monitoring the particle hydrodynamic diameters over time at 37°C where we found that PEG-PBAE NPs, regardless of the payload type, retained the hydrodynamic diameter <100 nm at least up to 8 hours (Figure 1D). In contrast, PBAE/pDNA and PBAE/mRNA NPs instantaneously aggregated in aCSF, and the particle hydrodynamic diameters reached several microns as early as 30 minutes after the incubation (Figure 1D). To investigate the serum colloidal stability of these NPs, we conducted multiple particle tracking (MPT) analysis. MPT measures the mean square displacement (MSD) values of NPs in biological medium, such as phosphate buffer saline (PBS) and whole mouse serum. MSD is a square of distance traveled by an individual NP and is directly proportional to particle diffusion rate, which is inversely proportional to particle diameter based on the Stokes-Einstein relationship.²⁸ We found that the MSD values of PEG-PBAE NPs were virtually identical in PBS and mouse serum after a 30-minute incubation regardless of the payload type, indicating that particle hydrodynamic diameters were unchanged in mouse serum (Figure 1E). In contrast, the MSD values of both PBAE NP formulations were over an order of magnitude lower in mouse serum compared to in PBS, which reflects particle aggregation in mouse serum (Figure 1E).

Figure 1. PEG-PBAE NPs, but not un-PEGylated PBAE NPs, carrying pDNA or mRNA retain colloidal stability in physiological conditions relevant to the systemic delivery of NPs to the brain.

(A) Representative agarose gel electrophoresis migration assay showing robust packaging of pDNA (left) or mRNA (right) in PEG-PBAE NPs. Numbers indicate the polymer-to-nucleic acid weight ratios. MK: molecular weight marker. (B) Representative transmission electron micrographs of PEG-PBAE NPs carrying pDNA (top) or mRNA (bottom). Scale bar = 200 nm. (C) Hydrodynamic diameters and ζ-potentials of PEG-PBAE NPs carrying pDNA or mRNA. (D) Colloidal stability of PBAE and PEG-PBAE NPs carrying pDNA or mRNA incubated in aCSF at 37 °C for up to 8 hours. (E) Median MSD values of PBAE and PEG-PBAE NPs carrying pDNA or mRNA in PBS or whole mouse serum. MSD is a square of distance traveled by an individual particulate matter within a predetermined time interval (i.e., time scale; τ = 1 s) and thus is directly proportional to the particle diffusion rate. n.s.: no significance, ****p < 0.0001 (one-way or two-way ANOVA).

Table 1.

Physicochemical properties of NPs.

NP	Hydrodynamic diametera ± SD (nm)	PDIa ± SD	ζ-potentialb ± SD (mV)
PBAE/pDNA	73.8 ± 3.1	0.23 ± 0.02	18.4 ± 3.2
PBAE/mRNA	70.4 ± 2.6	0.25 ± 0.03	19.8 ± 2.2
PEG-PBAE/pDNA	61.4 ± 2.5	0.21 ± 0.03	2.1 ± 0.3
PEG-PBAE/mRNA	64.8 ± 3.1	0.19 ± 0.04	2.5 ± 0.4
PEG-PBAE/mRNA+sgRNA	62.2 ± 3.8	0.23 ± 0.03	2.2 ± 0.3
LNP/mRNA	71.1 ± 3.5	0.39 ± 0.03	−3.1 ± 1.7

^aHydrodynamic diameter (number mean) and polydispersity index (PDI) were measured by dynamic light scattering (DLS) in 10 mM NaCl at pH 7.4. Mean ± SD (N =4).

^bζ-potential was measured by laser Doppler anemometry in 10 mM NaCl at pH 7.0. Mean ± SD (N = 4)

PEG-PBAE NPs exhibit prolonged circulation following systemic administration.

We next investigated the ability of PEG-PBAE NPs to stably circulate in bloodstream following systemic administration. We treated male C57BL/6 mice with PBAE NPs, PEG-PBAE NPs, or lipid NPs (LNPs) carrying Cy5-labeled pDNA or mRNA at a nucleic acid dose of 0.5 mg/kg via tail vein injection and conducted live-animal whole-body fluorescence imaging for up to 4 hours. Of note, the LNP formulation tested in this study is analogous to an intramuscular mRNA-based COVID-19 vaccine used in clinic (i.e., Comirnaty^®). PEG-PBAE/pDNA and PEG-PBAE/mRNA NPs exhibited relatively uniform whole-body fluorescence distribution without noticeable attenuation of the intensity over time (Figure 2A). In contrast, fluorescence signals were rapidly accumulated in the livers of the animals treated with PBAE/pDNA NPs or LNPs as early as 5 minutes after the administration, which persisted and were intensified over time (Figure 2A). Major organs and blood samples were harvested from these animals at 4-hour post-administration of NPs to determine the residual amounts of NPs in each compartment. In consistent with the outcomes of the live-animal imaging study (Figure 2A), strong fluorescence was observed in the livers of animals treated with PBAE NPs or LNPs, but the intensity was markedly lower in those of animals treated with PEG-PBAE NPs (Figure 2B). On the other hand, while the fluorescence intensity measured in the serum samples harvested from PBAE NP-, LNP-, and saline-treated animals were virtually identical, the levels were markedly and equally greater for animals treated with PEG-PBAE NPs regardless of the payload type (Figure 2C). Quantitatively, 0.8 ± 0.3%, 31.0 ± 1.2%, 30.4 ± 1.0%, and 0.9 ± 0.5%, respectively, of the systemically administered PBAE/pDNA NPs, PEG-PBAE/pDNA NPs, PEG-PBAE/mRNA NPs, and LNPs remained circulating in the bloodstream at 4-hour post-administration of NPs (Figure 2C).

Figure 2. PEG-PBAE NPs, but not clinically used LNPs, carrying pDNA or mRNA, exhibit long circulation in bloodstream following systemic administration.

(A) Representative whole-body fluorescence images of live animals intravenously treated with PBAE NPs, PEG-PBAE NPs, or LNPs carrying Cy5-labeled pDNA or mRNA over time (N = 3 animals per group). (B) Representative fluorescence images of major organs harvested at 4-hour post-administration of NPs from the animals in Figure 2A. (C) Percentage of circulating nucleic acids determined by quantifying the Cy5 fluorescence intensity of the serum collected at 4-hour post-administration of NPs from the animals intravenously treated with PBAE NPs, PEG-PBAE NPs, or LNPs carrying Cy5-labeled pDNA or mRNA. n.s.: no significance, ****p < 0.0001 (one-way ANOVA).

FUS-mediated BBB opening promotes accumulation of PEG-PBAE NPs in the brain following systemic administration.

We hypothesized that the safe BBB opening by FUS would result in vascular openings with a certain size cut-off. To test this, we engineered densely PEGylated NPs with defined particle diameters. Commercially available red-fluorescent, carboxylated polystyrene (PS) NPs (i.e., Fluosphere^™) possessing different particle diameters, 40, 100, or 200 nm, were densely coated with PEG polymers via chemical conjugation, and effective surface shielding was confirmed by near-neutral surface charges (i.e., ζ-potential) of NPs (Table S1). We then treated male C57BL/6 mice with a blend of single-sized PEGylated PS (PEG-PS) NPs and clinically used MBs (i.e., SonoVue^®) at doses of 5 mg/kg and 0.3 mL/kg, respectively, and immediately applied FUS to the right striatum. Of note, a single-point sonication was given using a 515-kHz transducer at a 0.6-MPa peak negative pressure with a 10-ms burst length for 2 minutes and a pulse repetition frequency of 1 Hz. Four hours after the administration, brain tissues were harvested from the treated animals and striatal NP accumulation was determined by whole-brain fluorescence imaging. We found that 40 and 100 nm PEG-PS NPs were accumulated precisely in the FUS-treated regions of the right striata with comparable fluorescence intensity, but negligible brain accumulation was observed with 200 nm PEG-PS NPs (Figure S1).

We next repeated this study with PBAE and PEG-PBAE NPs carrying Cy5-labeled pDNA at a pDNA dose of 0.5 mg/kg. Four hours after the administration, we found that PEG-PBAE NPs were accumulated in the FUS-treated region of right striatum, similar to our observation with PEG-PS NPs as large as 100 nm in diameters (Figure S1), but PBAE NPs failed to do so (Figure 3A). We next measured the fluorescence intensity of tissue homogenates prepared with left and right brain hemispheres and other major organs harvested from the treated animals. PEG-PBAE NP-treated, but not PBAE NP-treated, animals exhibited strong fluorescence intensity in the FUS-treated (i.e., right) hemisphere (Figure 3B). Fluorescence intensity was markedly and significantly lower in major organs, including heart, liver, and kidney, harvested from PEG-PBAE NP-treated animals compared to PBAE NP-treated animals (Figure 3B), in agreement with our observation with whole-tissue imaging (Figure 3A). Notably, PEG-PBAE NP-treated animals exhibited approximately 5.5-times greater right brain hemisphere-to-liver ratio of fluorescence intensity compared to PBAE NP-treated animals.

Figure 3. PEG-PBAE NPs carrying pDNA accumulate in FUS-treated brain region far greater than other major organs following systemic administration and retain the ability to mediate widespread transgene expression in the brain following serum incubation and subsequent intracranial administration.

(A) Representative fluorescence images and (B) Cy5 fluorescence intensity of brains and other major organs harvested at 4-hour post-administration of NPs from the animals intravenously treated with saline or PBAE or PEG-PBAE NPs carrying Cy5-labeled pDNA and subsequently received a FUS treatment on the right striatum (N = 3 animals per group). (C) Representative fluorescence images showing reporter ZsGreen1 expression in the brain sections at the infusion site within healthy mouse brain striatum 48 hours after the intracranial administration of freshly prepared or serum-incubated PEG-PBAE NPs carrying ZsGreen1-expressing pDNA at a pDNA dose of 0.1 mg/kg (N = 5 animals per group). Blue: nucleus; Green: ZsGreen1. Image-based quantification for (D) area and (E) intensity of ZsGreen1 expression. (F) Luciferase activity of homogenized brain tissues treated with freshly prepared or serum-incubated PEG-PBAE NPs carrying luciferase-expressing pDNA at a pDNA dose of 0.1 mg/kg via intracranial administration. n.s.: no significance, * p < 0.05, ***p < 0.001, ****p < 0.0001 (one-way ANOVA).

Serum incubation does not compromise the ability of PEG-PBAE NPs to mediate widespread transgene expression in mouse brains in vivo.

We tested whether PEG-PBAE NPs retained their ability to efficiently distribute and provide widespread transgene expression in rodent brains⁶ after incubation in mouse serum. PEG-PBAE NPs carrying ZsGreen1- or luciferase-expressing pDNA were freshly prepared or incubated in mouse serum for 30 minutes and intracranially infused into the right striata of mouse brains at a pDNA dose of 0.1 mg/kg via convection enhanced delivery. Forty-eight hours after the administration, brain tissues were harvested and sectioned, followed by confocal microscopy of the infusion planes. We found that freshly prepared and serum-incubated PEG-PBAE NPs provided comparably widespread and high-level reporter transgene expression in mouse brains, as revealed by quantitative analysis of confocal micrographs (Figures 3C-E and S2) and tissue homogenate-based luciferase assay (Figure 3F).

FUS-mediated BBB opening enables dose-dependent reporter transgene expression in the brain by systemically administered PEG-PBAE NPs.

We next investigated whether the combination of FUS-mediated BBB opening and long-circulating PEG-PBAE/pDNA NPs enabled efficient gene transfer to and reporter transgene expression in mouse brains. PEG-PBAE NPs carrying luciferase-expressing pDNA and MBs were co-administered into male C57BL/6 mice via tail vein injection at doses of 1.25 - 2.5 mg/kg (pDNA) and 0.3 mL/kg, respectively, and FUS was immediately applied to the right striatum using the FUS parameters described above. Forty-eight hours after the administration, brain tissues were harvested and subjected to whole-brain bioluminescence imaging and tissue homogenate-based luciferase assay. The luminescence signals were observed in the FUS-treated regions within the right brain hemispheres (Figure 4A), and the level of luciferase transgene expression was significantly greater with the higher dose (2.5 mg/kg) compared to the lower dose (1.25 mg/kg) (Figure 4B). In the subsequent experiment with PEG-PBAE NPs carrying ZsGreen1-expressing pDNA, ZsGreen1 transgene expression was evident in the FUS-treated region (Figure 4C) where the expression was co-localized with GFAP-positive astrocytes and NeuN-positive neurons (Figures 4D-E). As a safe assessment, brain tissues from the treated animals were harvested, sectioned, and stained with hematoxylin and eosin (H&E) for histopathological analysis. A blinded analysis by a board-certified neuropathologist revealed that FUS-treated and contralateral hemispheres were virtually identical with no sign of adverse event at 2- and 14-day post-administration of NPs (Figures 4F and S3).

Figure 4. PEG-PBAE NPs carrying pDNA mediate reporter protein production in astrocytes and neurons precisely in the FUS-treated brain region following systemic administration.

(A) Bioluminescence images of brains harvested at 48-hour post-administration of NPs from animals intravenously treated with PEG-PBAE NPs carrying luciferase-expressing pDNA at a pDNA dose of 1.25 or 2.5 mg/kg and subsequently received a FUS treatment on right striata (N = 5 animals per group). (B) Quantification of luciferase activity in the homogenized hemispheres from the brains in Figure 4A. (C) Representative confocal micrograph of brain harvested at 48-hour post-administration of NPs from animals intravenously treated with PEG-PBAE NPs carrying ZsGreen1-expressing pDNA at a pDNA dose of 2.5 mg/kg pDNA and subsequently received a FUS treatment on right striata (N = 5 animals per group). Blue: nucleus; Green: ZsGreen1. Representative confocal micrographs showing reporter ZsGreen1 production in untreated or FUS-treated areas of (D) GFAP- and (E) NeuN-stained brains from the animals identically treated as in Figure 4C. Blue: nucleus; Green: ZsGreen1; Red: astrocyte (GFAP) or neuron (NeuN). (F) Representative H&E-stained histological images of untreated or FUS-treated areas of the brains harvested from the animals identically treated as in Figure 4C. n.s.: no significance, **p < 0.01, ***p < 0.001, ****p < 0.0001 (one-way ANOVA).

FUS-mediated BBB opening enables reporter mRNA expression by systemically administered PEG-PBAE NPs.

Excellent colloidal stability (Figure 1E) and stable systemic circulation (Figure 2) observed with PEG-PBAE/mRNA NPs and the successful execution of the systemic pDNA delivery study (Figure 4) collectively provided an optimism that our combined delivery strategy would also work for mRNA. We thus tested FUS-mediated PEG-PBAE/mRNA delivery to the brain using the identical experimental design employed to the earlier studies with PEG-PBAE/pDNA NPs. PEG-PBAE NPs or LNPs carrying luciferase-expressing mRNA and MBs were co-administrated into male C57BL/6 mice via tail vein injection at doses of 0.5 mg/kg (mRNA) and 0.3 mL/kg, respectively, and FUS was immediately applied to the right striatum. Whole-brain bioluminescence imaging and homogenate-based luciferase assay revealed robust luciferase mRNA expression by PEG-PBAE NPs in the FUS-treated regions at 24-hour post-administration of NPs, but the mRNA expression by LNPs was negligible (Figures 5A-B). We also confirmed that PEG-PBAE NPs mediated comparably high mRNA expression in the FUS-treated brain regions at 6-hour post-administration, whereas PBAE NPs and LNPs were unable to produce the reporter protein (Figure S4). In contrast, PEG-PBAE NPs exhibited significantly lowered hepatic mRNA expression compared to PBAE NPs and LNPs (Figure S4). The mRNA delivery study was then repeated with PEG-PBAE NPs carrying mCherry-expressing mRNA. We found that mCherry mRNA expression took place in the FUS-treated region (Figure 5C), and the reporter proteins were produced by both GFAP-positive astrocytes and NeuN-positive neurons (Figure 5D). Similar to the observation in the pDNA delivery study above, the combination of FUS-mediated BBB opening and systemically administered PEG-PBAE/mRNA NPs did not manifest any neuropathological features in the treated brains one day after the NP administration (Figure 5F).

Figure 5. PEG-PBAE NPs, but not clinically used LNPs, carrying mRNA mediate reporter protein production in astrocytes and neurons precisely in the FUS-treated brain region following systemic administration.

(A) Bioluminescence images of brains harvested at 24-hour post-administration of NPs from animals intravenously treated with PEG-PBAE NPs carrying luciferase-expressing mRNA at an mRNA dose of 0.5 mg/kg and subsequently received a FUS treatment on right striata (N = 5 animals per group). (B) Quantification of luciferase activity in the homogenized hemispheres from the brains in Figure 5A. (C) Representative confocal micrograph of brain harvested at 24-hour post-administration of NPs from animals intravenously treated with PEG-PBAE NPs carrying mCherry-expressing mRNA at an mRNA dose of 0.5 mg/kg and subsequently received a FUS treatment on right striata (N = 5 animals per group). Blue: nucleus; Green: mCherry. Representative confocal micrographs showing reporter mCherry production in untreated or FUS-treated areas of (D) GFAP- and (E) NeuN-stained brains from the animals identically treated as in Figure 5C. Blue: nucleus; Green: mCherry; Red: astrocyte (GFAP) or neuron (NeuN). (F) Representative H&E-stained histological images of untreated or FUS-treated areas of the brains harvested from the animals identically treated as in Figure 5C. n.s.: no significance, **p < 0.01, ***p < 0.001, ****p < 0.0001 (one-way ANOVA).

Combined delivery strategy of FUS-mediated BBB opening and long-circulating PEG-PBAE NPs provides a means to achieve somatic genome editing in mouse brains in vivo.

We next investigated whether our combined delivery strategy could bring about nucleic acid-based genome editing in the brain. PEG-PBAE NPs were engineered to simultaneously carry Cas9-expressing mRNA and sgRNA 298, and effective packaging of both components were confirmed by gel electrophoresis migration assay (Figures 6A-B). Of note, sgRNA 298 targets the loxP-flanked STOP cassette that prevents the transcription of the tdTomato reporter gene in the Ai9 mouse chromosome.²⁹ Physicochemical properties (i.e., hydrodynamic diameters and ζ-potentials) of PEG-PBAE NPs carrying both mRNA and sgRNA were confirmed to be comparable to those of PEG-PBAE/pDNA and PEG-PBAE/mRNA NPs (Table 1). PEG-PBAE NP and MBs were then co-administrated into female and male Ai9 mice via tail vein injection at doses of 0.5 mg/kg (mRNA and sgRNA each) and 0.3 mL/kg, respectively, and FUS was immediately applied to the right striatum using the FUS parameters described above. Both female and male Ai9 mice exhibited comparably clear tdTomato expression patterns in the FUS-treated regions within the right striata at 5-day post-administration of NPs (Figure 6C), indicating that the combined delivery strategy successfully disrupted the STOP cassette to mediate somatic genome editing. Similar to the earlier reporter pDNA and mRNA delivery studies, tdTomato expression was evident in both GFAP-positive astrocytes and NeuN-positive neurons (Figures 6D-E).

Figure 6. PEG-PBAE NPs carrying Cas9-expressing mRNA and sgRNA targeting the STOP cassette (sgRNA 298) mediate genome editing in astrocytes and neurons precisely in the FUS-treated regions of Ai9 mouse brains following systemic administration.

(A) Agarose gel and (B) TBE gel electrophoresis migration assay showing robust co-packaging of Cas9-expressing mRNA and sgRNA 298 in PEG-PBAE NPs. Numbers indicate the polymer-to-nucleic acid weight ratios. MK: molecular weight marker. (C) Representative confocal micrographs of the brains harvested at 5-day post-administration of NPs from female (upper) and male (bottom) Ai9 mice intravenously treated with PEG-PBAE NPs carrying Cas9-expressing mRNA and sgRNA 298 at an mRNA and an sgRNA dose each of 0.5 mg/kg and subsequently received a FUS treatment on right striata (N = 6 animals per group). Blue: nucleus; Green: tdTomato. Representative confocal micrographs showing reporter tdTomato production in untreated or FUS-treated areas of (D) GFAP- and (E) NeuN-stained brains from the animals identically treated as in Figure 6C. Blue: nucleus; Green: tdTomato; Red: astrocyte (GFAP) or neuron (NeuN).

DISCUSSION

We demonstrate here that long-circulating PEG-PBAE NPs, in conjunction with FUS-mediated BBB opening, promote efficient and safe delivery of various types of nucleic acid payloads precisely to the FUS-treated region within the brain, following systemic administration. Accordingly, the combined delivery strategy provides robust protein production and somatic genome editing in primary brain parenchymal cells in the FUS-treated region in vivo. Importantly, PEG-PBAE NPs exhibit markedly greater accumulation in the FUS-treated brain region than in the liver while the liver is the primary destination of systemically administered NPs,³⁰ as shown with un-PEGylated PBAE NPs and clinically used LNPs. Likewise, FUS-mediated BBB opening has been shown to enhance the accumulation of and genome editing by systemically administered adeno-associated virus (AAV) serotype 9 (AAV9) in the brain.³¹ AAV9 is a serotype that, unlike other serotypes, possesses ability to traverse the BBB and has been approved by the FDA for the treatment of spinal muscular dystrophy (i.e., Zolgensma^®).³² Nevertheless, the amount of AAV9 accumulated in the liver remained two orders of magnitude greater than the amount found in the FUS-treated brain region despite the addition of FUS to enhance the brain accumulation.³¹ Of note, hepatoxicity is a major and significant concern that limits the widespread use of AAV vectors for implementation of gene therapy via the systemic route.^33-35 To this end, the combined delivery strategy of FUS-mediated BBB opening and long-circulating non-viral NPs constitutes an attractive means to enable safe, efficient, and site-specific therapeutic nucleic acid delivery to and therapy in the brain.

Our experimental data demonstrate that the ability of systemically administered nucleic acid delivery NPs to stably circulate and retain colloidal stability in the bloodstream is essential to exploit the FUS-mediated BBB opening to partition into the brain parenchyma. We found that PEG-PBAE NPs stably circulated in the bloodstream following systemic administration at least up to 4 hours, a previously reported time window of FUS-mediated BBB opening.³⁶ Further, comparably modest levels of systemically administered PEG-PBAE NPs were found in liver and other major organs, unlike PBAE NPs and LNPs which exhibited conspicuous and preferential liver accumulation. The findings collectively suggest that a significant amount of systemically administered PEG-PBAE NPs were continuously passing through the cerebral capillaries and thus were available for FUS-mediated BBB penetration during the BBB-opening time window. We note that fluorescence signal beyond the background level was observed around the brains of animals treated with poorly circulating PBAE NPs or LNPs carrying fluorescently labeled nucleic acids. However, FUS failed to promote brain penetration of and subsequent reporter protein production by these formulations. The finding was likely attributed to the inability of these NPs to retain the particle diameters below the size cut-off of the vascular opening created by FUS to traverse the disrupted BBB. In support of this hypothesis, diffusion rates of ~70-nm PBAE NPs were over an order of magnitude lower in whole mouse serum than in PBS, which corresponds to a particle diameter increment near to or possibly beyond a micron in serum, as estimated by the Stokes-Einstein relationship.²⁸ On the other hand, 40- and 100-nm PEG-PS NPs were equally and efficiently accumulated in the FUS-treated brain region following systemic administration, but 200-nm PEG-PS NPs were unable to do so. The data suggest that the size cut-off of the vascular opening established by the FUS parameters employed in this study lies somewhere between 100 and 200 nm, which is far smaller than the size of the serum-incubated PBAE NPs.

Various sonication conditions, including FUS parameters and MB type/concentration, have been explored to achieve safe BBB opening in clinical and preclinical settings.^{18, 19, 37, 38} Mechanical index (MI) indicates the power of a single pulse of FUS application, which is defined as the peak negative pressure of the ultrasound wave divided by the square root of the center frequency of a FUS transducer.³⁹ We here set the MI value to 0.836, which falls in the range of 0.25 – 1.2 previously employed for FUS-mediated BBB opening studies.^{40, 41} Of note, greater MI values generally give rise greater BBB opening efficacy but are associated with greater chances of vascular or tissue damage.^{42, 43} Our histological analysis revealed that a single treatment with SonoVue^® and 0.836-MI FUS did not cause hemorrhage or neuropathological features in the brain 1, 2, and 14 day(s) after the FUS application. Likewise, repeated treatments with SonoVue^® and 0.8-MI FUS were previously shown not to induce intracerebral hemorrhage and behavioral change, as determined by histological analysis and modified Irwin’s test, respectively.⁴⁴ In another study, however, mild yet noticeable intracranial hemorrhage was observed when animals received a single treatment of SonoVue^® and 0.85-MI FUS.⁴⁵ We note that the comparison should be made with caution as FUS-mediated vascular and/or tissue perturbation can also be affected by other FUS parameters not factored in MI, such as burst duration, pulse repetition frequency, and overall exposure time.⁴² Nevertheless, it appears that the MI value utilized in this study may reside around a safely operable borderline, and thus a more comprehensive safety assessment is warranted in future studies. For example, microglial activation is often observed in the brain upon the implementation of FUS-mediated BBB opening.⁴⁶ Although it is generally self-resolved in healthy brains over time,^{47, 48} such a perturbation brings about a cascade of immune responses^{48, 49} and has numerous and case-sensitive implications for neurological discorders.⁵⁰

We and others have previously shown that FUS-mediated BBB opening enables and/or enhances delivery of lipid- or polymer-based NPs carrying pDNA to the brain following systemic administration.^51-56 We demonstrate here that systemically administered PEG-PBAE NPs carrying reporter or genome editing (i.e., Cas9) mRNA are efficiently delivered to the brain with an aid of FUS to produce the respective encoded proteins both in astrocytes and neurons, which has been rarely reported if any. Ogawa et al. recently showed that FUS-mediated BBB opening enhanced the delivery of a specific preclinical LNP formulation carrying reporter mRNA to the brain following systemic administration.⁵⁷ However, production of reporter fluorescence proteins was primarily confined to CD31-positive endothelial cells, lacking fluorescence signal overlapping with astrocytes or neurons.⁵⁷ The discrepancy observed in these two studies may be at least partially attributed to the ability of PEG-PBAE NPs to efficiently penetrate the brain ECM^{6, 10} after traversing the BBB transiently opened by FUS to reach brain parenchymal cells. Importantly, we confirmed in this study that colloidal stability and brain-penetrating property of PEG-PBAE NPs were not compromised after the incubation in mouse serum, underscoring the suitability of the formulation for delivery of mRNA and other therapeutic nucleic acids to the brain via the systemic route.

As mentioned above, PEG-PBAE NPs exhibited the capability to stably package various types of nucleic acid payloads and deliver them precisely to the FUS-treated brain region. Our microscopic analysis revealed that PEG-PBAE/pDNA and PEG-PBAE/mRNA NPs provided comparable striatal coverage area of reporter fluorescence protein production by the respective nucleic acid payloads. The outcome was somewhat expected a priori as PEG-PBAE NPs, regardless of the type of payloads, displayed virtually identical physicochemical properties and physiological colloidal stability, the properties that play a pivotal role on particle distribution in the brain.^{5, 8} However, despite the 5-time lower dose of mRNA compared to pDNA given to animals, the former appeared to provide greater overall level of reporter protein production, as determined by the tissue homogenate-based luciferase assay. The finding likely reflects that while mRNA is readily translated in cytosol,⁵⁸ pDNA must breach an additional challenging intracellular barrier, nuclear envelope, to produce final protein products.⁵⁹ We also found that the striatal coverage area of genome editing in Ai9 mice was smaller than that of reporter protein production in C57BL/6 mice regardless of the use of identical Cas- and mCherry-expressing mRNA doses (i.e., 0.5 mg/kg), respectively. The difference is presumably attributed to the necessity of an orchestrated processing of Cas9 product and sgRNA, beyond the mRNA expression, to disrupt the STOP cassette in the Ai9 mouse chromosome for successful genome editing.⁶⁰ Nevertheless, the notable striatal genome editing was achieved by our combined delivery strategy where the mouse brains were sonicated with a single-element transducer. We expect that a broader coverage of genome editing can be realized, if desired, by using multi-element FUS array which enables BBB opening in a widespread region.^{61, 62} Indeed, the technique has been utilized for systemic AAV9 delivery to the brain, resulting in more widespread vector distribution and subsequent transgene expression in mouse brains compared to a single-element FUS.³¹ We found here that the reporter protein expression patterns in the coronal mouse brain sessions were spherical, unlike the ellipsoidal acoustic pressure distribution created by BBB-opening FUS.⁶³ The discrepancy may reflect that the BBB penetration of PEG-PBAE NPs preferentially took place at the central spherical region of the pressure distribution where the acoustic intensity is greatest and thus the BBB opening efficacy.^{64, 65}

CONCLUSION

We introduce here a combined delivery strategy that comprehensively addresses a series of biological barriers to achieve safe and efficient nucleic acid delivery to the brain following systemic administration. Our engineered biodegradable NP platform is capable of packaging various types of nucleic acid payloads and stably circulating upon entry into the bloodstream while resisting liver accumulation and particle aggregation. On the other hand, FUS allows transient BBB opening at a predetermined location within the brain to promote the accumulation of long-circulating NPs as large as 100 nm in diameters in brain parenchyma. Accordingly, we have experimentally validated that this two-pronged delivery strategy mediates site-specific delivery of various nucleic acid payloads to and genome editing in brain parenchymal cells following systemic NP administration. We envision that our approach, upon clinical development, will provide a versatile means to realize systemic nucleic acid therapy of a wide spectrum of brain disorders.

EXPERIMENTAL SECTION

Polymer synthesis

PBAE polymers were synthesized via Michael addition reactions, as previously described.⁵ Briefly, 1,4-butanediol diacrylate (Millipore Sigma, Burlington, MA, USA) and 4-amino-1-butanol (Millipore Sigma) were reacted at a molar ratio of 1.05:1 or 1.1:1 at 90 °C for 20 hours. PBAE polymers were then collected by centrifuge/precipitation and washed three times in cold diethyl ether (DE) (Thermo Fisher Scientific, Waltham, MA, USA). After drying under vacuum for 5 days, PBAE polymers were dissolved in dichloromethane (DCM) (Thermo Fisher Scientific) at 100 mg/mL concentration and reacted with 1,4-butanediol diacrylate at a quarter molar equivalent of the amount used for the initial PBAE polymerization reaction to ensure that the two terminal ends of the PBAE polymer chains are capped with acrylate groups. The acrylated PBAE polymers were then purified with cold DE and dried under vacuum for 5 days. The polymers were dissolved in deuterated methanol (Thermo Fisher Scientific) at 25 mg/mL concentration and analyzed by nuclear magnetic resonance (NMR). Based on the NMR spectrum, the linear structure of acrylated PBAE polymers was confirmed, and the molecular weight (MW) of the polymer synthesized at a molar ratio of 1.05:1 or 1.1:1 was estimated to be 6.0 ± 0.3 or 3.9 ± 0.2 kDa, respectively (Figures S5A and S5B). In parallel, the number average MW (polydispersity index) of PBAE polymers synthesized at a molar ratio of 1.05:1 or 1.1:1 was determined by gel permeation chromatography to be 6.7 ± 0.5 kDa (1.37) or 4.3 ± 0.3 kDa (1.31), respectively. To cap the terminal acrylate ends with functional groups, the high-MW (~6 kDa) and the low-MW (~4 kDa) acrylated PBAE polymers were dissolved individually in DCM at 100 mg/mL concentration and reacted with 30 molar equivalents of 1,11-diamino-3,6,9-trioxaundecane (Millipore Sigma) and 1,2-diaminoethane (Millipore Sigma), respectively, at room temperature for 16 hours. The end-capped PBAE polymers were purified with cold DE and dried under vacuum for 5 days. Successful end-capping was confirmed with the complete disappearance of the proton peaks of the acrylate functional group in NMR spectrum (Figures S5C and S5D).

To prepare PEG-PBAE polymers, 4 kDa PBAE polymers end-capped with 1,2-diaminoethane were dissolved in DCM at 100 mg/mL concentration and reacted with 3 molar equivalents of 5 kDa methoxy PEG-epoxide (Creative PEGWorks, Durham, NC, USA) for 16 hours at room temperature. The polymer products were then transferred into a tube of regenerated cellulose membrane having a 10 kDa MW cut-off (MWCO) and extensively dialyzed against methanol (Thermo Fisher Scientific) at 4 °C for 3 days. The polymers were then collected by centrifuge/precipitation, washed three times in cold DE, and dried under vacuum for 5 days. PEG-PBAE polymer was also analyzed by NMR to confirm the retention of the PBAE backbone structure and PEG conjugation (Figure S5E). PBAE and PEG-PBAE polymers were stored at −80 °C and dissolved in anhydrous dimethyl sulfoxide at 100 mg/mL to be used to formulate NPs carrying with various nucleic acid payloads.

Acquisition and preparation of nucleic acids

Luciferase-expressing pDNA driven by cytomegalovirus (CMV) promoter, pd1GL3-RL, was a kind gift from Professor Alexander M. Klibanov (M.I.T). ZsGreen-expressing pDNA driven by CMV promoter was purchased from Clonetech Laboratories (Mountainview, CA, USA). Competent E. coli DH5α bacterial cells (New England Biolabs, Ipswich, MA, USA) were transformed with each pDNA. The transformed bacterial cells were placed on an agarose plate containing 50 μg/mL kanamycin (Millipore Sigma) and incubated at 37 °C in a humid chamber. After an overnight incubation, a single colony was picked and expanded in 2 mL of sterilized Lennox Broth (LB) (Millipore Sigma) media containing 50 μg/mL kanamycin. After an overnight incubation, 1 mL of the expanded bacterial cells was transferred into 2 L of sterilized LB media containing 50 μg/mL kanamycin and further incubated overnight. When the optical density at 600 nm wavelength (OD600) of the bacteria culture fell in a range of 3 - 3.5 as measured by a microplate reader (BioTek, Winooski, VT, USA), bacteria cells were lysed and pDNA was purified using the EndoFree Plasmid Giga Kit (Qiagen, Hilden, Germany) as per manufacturer’s protocol. The final pDNA pellet was resuspended in DNase/RNase-free distilled water (Thermo Fisher Scientific).

The mRNA encoding luciferase, mCherry, or streptococcus pyogenes Cas9, modified with 5-methoxyuridine, was purchased from TriLink BioTechnologies (San Diego, CA, USA). The sgRNA 298 modified with 2' O-methyl analog on the first and last 3 bases and 3' phosphorothioate between the first 3 and last 2 bases was purchased from Synthego (Redwood City, CA, USA).

The concentrations of pDNA, mRNA, and sgRNA were measured using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA).

Preparation of PEG-PS NPs

Red-fluorescent, carboxylated PS NPs (Thermo Fisher Scientific) sized 40, 100, or 200 nm in diameters were densely coated with 5kDa methoxy PEG-amine (Creative PEGWorks) via a carboxyl-amine coupling reactions, as described in a previous work.⁶⁶ Briefly, each PS NP suspension was sonicated for 1 hour, washed with DNase/RNase-free distilled water using 100 kDa MWCO Amicon Ultra Centrifugal Filters, and resuspended in 1 mL of DNase/RNase-free distilled water at a PS NP concentration of 10 mg/mL. Methoxy PEG-amine at 5 molar equivalents to the particle surface carboxyl groups was added to 40, 100, or 200 nm PS NP suspension, and N-hydroxysulfosuccinimide (Millipore Sigma) was added to a final concentration of 7 mM. Subsequently, 4 volumes of 200 mM borate buffer (pH 8.2) were added to the NP suspension, followed by the addition of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC, Millipore Sigma) to a final concentration of 2 mM. PS NP suspensions were reacted at 25°C for 4 hours, washed with DNase/RNase-free distilled water, and centrifuged in 100 kDa MWCO Amicon Ultra Centrifugal Filters to collect PEG-PS NPs. PEG-PS NPs were resuspended at 50 mg/mL PS NP concentration and stored at 4°C until use.

Preparation of PBAE, PEG-PBAE NPs or LNPs carrying pDNA, mRNA, or a mixture of mRNA and sgRNA

PBAE or PEG-PBAE NPs were prepared by a formulation method that we have previously reported with minor modifications.^{5, 67} PBAE NPs were prepared with 6 kDa PBAE polymers while PEG-PBAE NP were fabricated with a blend of 6 kDa PBAE and PEG-PBAE polymers at a PBAE weight ratio of 3:2. For the preparation of PEG-PBAE NPs carrying both mRNA and sgRNA, a blend of mRNA and sgRNA at a weight ratio of 1:1 was packaged into the NPs. All NPs were formulated by vigorous mixing of polymer and nucleic acid solutions at a volumetric ratio of 1:5 and various polymer-to-nucleic acid weight ratios ranging from 5 to 90. The mixed solution was incubated at room temperature for 30 minutes for NP assembly. The solution was then transferred into a 100 kDa MWCO Amicon Ultra Centrifugal Filters (Millipore Sigma) and centrifuged at 1,000 × g for 15 minutes at 4 °C. The concentrated NP solution was diluted 10 times with DNase/RNase-free distilled water and re-centrifuged at 1,000 × g for 15 minutes at 4 °C to remove residual polymers. The process was repeated three times, and the final NP solution was concentrated to a nucleic acid concentration of 1 mg/ml for subsequent experiments.

For the preparation of LNPs, a lipid mixture was prepared with LC-0315 ionizable lipid, 1,2-Distearoyl-sn-glycero-3-phosphorylcholine, ALC-0159 PEG lipid, and cholesterol with a molar ratio of 46.3/9.4/1.6/42.7 in absolute ethanol (Thermo Fisher Scientific) at a lipid concentration of 6.32 mg/mL. The mRNA solution was prepared with mRNA in 50 mM sodium acetate (pH 5.0) at an mRNA concentration of 0.21 mg/mL. The lipid mixture and mRNA solutions were mixed at a volume ratio of 1:3 using a NanoAssemblr^® Ignite (Precision NanoSystems, Canada) at a flow rate of 12 mL/min to formulate LNPs. LNPs were washed with PBS (pH 7.4) using 100 kDa MWCO Amicon Ultra Centrifugal Filters, concentrated to a final mRNA concentration of 1 mg/mL, and stored at 4°C until use.

Gel electrophoresis migration assay

The amount of PEG-PBAE NPs equivalent to 100 ng of nucleic acids were loaded into the wells of 1% agarose or 4% - 20% gradient NOVEX tris-borate-EDTA (TBE) gel (Thermo Fisher Scientific). Electrophoresis was conducted at 80 V for 30 minutes in tris-acetate-EDTA or TBE buffer for agarose or TBE gel, respectively. Subsequently, the agarose gel premade with 1X SYBR Safe DNA Gel Stain (Thermo Fisher Scientific) was imaged using a Chemi-Doc imaging system (Bio-RAD, Hercules, CA, USA). TBE gels were further incubated in TBE buffer containing 1X SYBR Safe DNA Gel Stain at room temperature for 10 minutes, washed with fresh TBE buffer for another 10 minutes, and imaged using a Chemi-Doc imaging system.

TEM

PEG-PBAE NPs equivalent to 1 μg of nucleic acids were loaded on a carbon type-B copper grid (Ted Pella, Redding, CA, USA) and air-dried for 6 hours. After rinsing with deionized water for 1 minute, the sample was stained with UranylLess EM Stain (Electron Microscopy Sciences, Hatfield, PA, USA) for 1 minute. The grid was then washed once with deionized water, dried overnight, and imaged with Hitachi H7600 TEM (Hitachi High-Technologies, Tokyo, Japan).

Physicochemical characterization and physiological stability assessment

PEG-PBAE NPs were diluted to a nucleic acid concentration of 5 μg/mL in 10 mM NaCl at pH 7.0. The diluted NP solution was transferred into a UV cuvette or a Capillary Zeta cell (Malvern Instruments, Malvern, UK), and hydrodynamic diameters or ζ-potentials of NPs, respectively, were measured after a 2-minute incubation at room temperature using a Zetasizer Nano ZS (Malvern Instruments, Malvern, UK).

To assess the colloidal stability of NPs in a physiological condition in the brain, PBAE or PEG-PBAE NPs equivalent to a nucleic acid concentration of 5 μg/mL were incubated in aCSF (Harvard Apparatus, Holliston, MA) at 37 °C for up to 8 hours. At each designated time point, a small volume of the incubated NP solution was transferred into a UV cuvette for the measurement of hydrodynamic diameters using a Zetasizer Nano ZS.

MPT analysis

For microscopic observation of NPs, pDNA or mRNA were labeled with Cy5 using a Label IT Tracker Intracellular Nucleic Acid Localization Kit (MirusBio, Madison, WI, USA) as per manufacturer’s protocol and used to formulate PBAE or PEG-PBAE NPs, as described above.

One microliter of PBAE or PEG-PBAE NPs equivalent to 1 μg of nucleic acids was added to and gently mixed with 30 μL PBS or whole serum obtained from mouse blood. After a 30-minute incubation, the motions of NPs in PBS or mouse serum were recorded with fluorescence video microscopy (Axiovert, Carl Zeiss, Stuttgart, Germany) at a frame rate of 15 frames per second (i.e., 67 milliseconds per frame or 15 frames per second). The movies were then analyzed using custom-written software in MATLAB (MathWorks, Natick, MA, USA) to determine MSD values.

Assessment of systemic circulation and biodistribution

C57BL/6 mice (6 – 8-week-old, male) were purchased from Jackson Laboratory (Bar Harbor, ME, USA). Animals were handled in accordance with the guidelines and policies of Johns Hopkins University Institutional Animal Care and Use Committee. The entire chest and abdomen of animals were shaved with a hair trimmer and hair remover cream. Animals were treated with 0.1 mL of PBAE NPs, PEG-PBAE NPs, or LNPs carrying Cy5-labeled pDNA or mRNA at a nucleic acid dose of 0.5 mg/kg via tail vein injection. The fluorescence intensity of three different NP formulations was virtually identical, regardless of the type of nucleic acid payloads, as confirmed by the fluorometrically measured nucleic acid loading efficiency (Figure S6). Animals were then subjected to whole-body live animal imaging at 5-minute and 0.5-, 1-, 2-, and 4-hour post-administration of NPs using an In Vivo Imaging System (IVIS, PerkinElmer, Waltham, MA, USA) at the excitation/emission wavelength of 640/680 nm. Immediately after the final imaging (i.e., 4-hour post-administration of NPs), blood samples were harvested from individual animals which were then subjected to intracardiac perfusion to remove the residual blood. Subsequently, major organs, including brain, heart, liver, kidney, lung, and spleen were harvested and imaged using an IVIS at the excitation/emission wavelength of 640/680 nm. In parallel, the harvested blood samples were centrifuged at 500 × g for 5 minutes at 4°C, and the upper serum layer was collected for the measurement of fluorescence intensity at the excitation/emission wavelength of 640/680 nm using a microplate reader. The percentage of nucleic acids that remained circulating in the bloodstream was calculated based on the estimated average blood volume of 6 – 8-week old mice weighed 21.9 – 25 g (1.876 mL; 0.08 mL/g body weight⁶⁸).

Treatment of animals with FUS and NPs

The heads of 6-8-week-old male C57BL/6 and female/male B6.Cg-Gt(ROSA)26Sortm9(CAG-tdTomato)Hze/J (Ai9, Strain #: 007909, Jackson Laboratory) mice were shaved with a hair trimmer and hair remover cream. Animals were then treated with 0.15 mL of a mixture of PBAE NPs, PEG-PBAE NPs, or LNPs carrying Cy5-labeled or unlabeled nucleic acids and SonoVue^® MBs (Bracco Diagnostics, Milano, Italy) at doses of 1.25 - 0.5 mg/kg (nucleic acid) and 0.3 mL/kg (approximately 1.2 × 106 MBs), respectively, via tail vein injection. Of note, labeled and unlabeled nucleic acids were used for the assessment of particle distribution and reporter protein production, respectively. Animals were immediately anesthetized by isoflurane (Baxter International, Deerfield, IL, USA) and placed on an RK50 stereotactic-guided FUS system (FUS Instruments, Toronto, ON, Canada). FUS was then applied to the right striatum with a single-point sonication using a 515 kHz FUS transducer at a 0.6 MPa peak negative pressure with 10 milliseconds burst length for 2 minutes at a pulse repetition frequency of 1 Hz.

Assessment of brain accumulation and biodistribution

C57BL/6 mice treated with BBB-opening FUS and systemic NPs carrying Cy5-labeled pDNA were euthanized at 4-hours post-administration of NPs. The major organs were then harvested and imaged using an IVIS, as described above. Brains were then separated into two hemispheres: FUS-untreated left and FUS-treated right hemispheres. Subsequently, both brain hemispheres and other major organs were weighed, immersed in reporter lysis buffer (Promega, Madison, WI, USA), and homogenized with a bead-based TissueLyser LT (Qiagen) at a 50/s oscillation rate for 30 minutes at 4oC. The lysates were then subjected to three freeze-and-thaw cycles and centrifuged at 12,000 × g for 20 minutes at 4°C. The supernatants were collected, and fluorescence intensity was measured at the excitation/emission wavelength of 640/680 nm using a microplate reader.

Intracranial NP administration and assessment of transgene expression

The heads of 6-8-week-old male C57BL/6 mice were shaved with a hair trimmer and hair remover cream. Surgical procedures were performed using standard sterile surgical techniques. Animals were anesthetized by intraperitoneal injection of a 100-μL mixture of ketamine (100 mg/kg) and xylazine (10 mg/kg). A midline scalp incision was made to expose the coronal and sagittal sutures (the midline) and a burr hole was drilled 2 mm lateral to the sagittal suture and 0.5 mm posterior to the bregma. A Neuros Syringe (Hamilton, Reno, NV, USA) connected to a 33-gauge needle was filled with 2 μL of freshly prepared or serum-incubated PEG-PBAE NPs carrying ZsGreen1- or luciferase-expressing pDNA at a pDNA concentration of 1 mg/mL. The needle was then lowered into a depth of 3 mm from the cranium to target the striatum, and the NP solution was infused at a rate of 0.2 μL/min for 10 minutes as controlled by a Chemyx Nanojet Injector Module (Chemyx, Stafford, TX). When the infusion was completed, the needle was removed slowly at 1 mm/min to prevent backflow. Following the removal, the skin incision was closed using an Autoclip Wound Closing System (Thermo Fisher Scientific), and a thin layer of bacitracin (Millipore Sigma) was placed over the closed incision.

To determine the distribution of reporter transgene expression, brains of animals intracranially treated with NPs carrying ZsGreen-expressing pDNA were harvested at 48-hour post-administration of NPs, frozen in optimal cutting temperature compound (Thermo Fisher Scientific), and sectioned at a 10 μm thickness using a Leica Cryostat (Leica, Wetzlar, Germany). The brain slices were washed with PBS three times for 1 minute each and counterstained with DAPI (Thermo Fisher Scientific) for 5 minutes. After additional three PBS washes, the slices were mounted with Fluoromount-G Mounting Medium (Thermo Fisher Scientific) and observed with a confocal laser microscope (Zeiss LSM 710, Carl Zeiss, Stuttgart, Germany).

To determine the overall level of reporter transgene expression, brains of separate animals intracranially treated with NPs carrying luciferase-expressing pDNA were harvested at 48-hour post-administration of NPs. The brains were immersed in reporter lysis buffer (Promega, Madison, WI, USA), and homogenized with a bead-based TissueLyser LT at a 50/s oscillation rate for 30 minutes at 4oC. The brain lysates were then subjected to three freeze-and-thaw cycles and centrifuged at 12,000 × g for 20 minutes at 4°C. The supernatants were subjected to homogenate-based luciferase assay to measure the luciferase activity in the relative light unit (RLU) using Luciferase Assay System (Promega) and a 20/20n luminometer (Turner Biosystems, Sunnyvale, CA, USA). In parallel, the protein concentrations of the supernatants were quantified by BCA Protein Assay Kit (Thermo Fisher Scientific). The RLU values were normalized by the protein weight.

Bioluminescence/fluorescence imaging and homogenate-based luciferase assay of brain tissues: assessment of reporter protein production

C57BL/6 mice treated with BBB-opening FUS and systemic NPs carrying luciferase-expressing pDNA or mRNA were euthanized at 48- and 24-hour post-administration of NPs, respectively. Brains were then harvested and immersed in 15 mg/mL luciferin substrate solution (Gold Biotechnology, St. Louis, MO, USA) for 1 minute, followed by bioluminescence imaging using an IVIS for 1 minute. To determine the overall level of luciferase expression, brains were harvested at different time points after NP administration (i.e., 6, 24, or 48 hours), separated into two hemispheres, lysed, and subjected to homogenate-based luciferase assay, as described above. For comparison at 6-hour post-administration of NPs, other major organs were harvested for luciferase assay.

For microscopic observation of reporter protein production, C57BL/6 mice treated with BBB-opening FUS and systemic NPs carrying luciferase-expressing pDNA or mRNA were euthanized at 48- and 24-hour post-administration of NPs, respectively. Likewise, Ai9 mice treated with BBB-opening FUS and systemic NPs carrying Cas9-expressing mRNA and sgRNA 298 were euthanized at 5-day post-administration of NPs. Brains of euthanized animals were then sectioned, and fluorescence emitted from the reporter proteins, either of ZsGreen, mCherry, or tdTomato, were visualized with confocal microscopy, as described above. For immunofluorescence staining, the brain slices were washed three times with PBS for 1 minute each and fixed with 4% paraformaldehyde (Thermo Fisher Scientific) for 5 minutes at room temperature. After three additional PBS washes, the slices were immersed in PBS containing 0.1% Triton X-100 (Thermo Fisher Scientific) for 5 minutes at room temperature. After three additional PBS washes, the slices were incubated in PBS containing 1% bovine serum albumin (BSA) (Thermo Fisher Scientific) for blocking at room temperature for 1 hour. After three additional PBS washes, the slices were incubated with anti-GFAP antibody (Cat. #: NB300-141, Novus Biologicals, Centennial, CO, USA) or anti-NeuN antibody (Cat. #: ab177487, Abcam, Cambridge, UK) at a 1:200 ratio in PBS containing 1% BSA at 4°C for overnight. After three additional PBS washes, the slices were incubated with donkey anti-rabbit IgG H&L conjugated with Alexa Fluor^® 647 at a 1:100 ratio in PBS containing 1% BSA at room temperature for 1 hour. After three additional PBS washes, the slices were counterstained with DAPI, mounted with Fluoromount-G Mounting Medium, and observed with a confocal laser microscope.

In vivo safety assessment

The brains harvested from animal treated with BBB-opening FUS and systemic NPs were paraffin-sectioned, and the brain slices were stained with H&E. Histopathological scoring was conducted by a board-certified neuropathologist (C. G. E., M.D., Ph. D.) in a blinded manner for the assessment of local safety profiles of our combined nucleic acid delivery strategy.

Statistical analysis

Statistical analysis between two groups was conducted using a two-tailed Student’s t-test assuming unequal variances with Welch’s correction. If multiple comparisons were involved, one-way ANOVA was conducted followed by a Tukey’s multiple comparison test. GraphPad Software (GraphPad Software, La Jolla, CA, USA) was used for these statistical analyses.