Application of deacetylated-poly-N-acetyl glucosamine nanoparticles for the delivery of miR-126 for the treatment of cecal-ligation puncture induced sepsis

Abstract

Sepsis is an acute inflammatory syndrome in response to infection. In some cases, excessive inflammation from sepsis results in endothelial dysfunction and subsequent increased vascular permeability leading to organ failure. We previously showed that treatment with endothelial progenitor cells, which highly express microRNA-126 (miR-126), improved survival in mice subjected to cecal ligation and puncture (CLP) sepsis. MiRNAs are important regulators of gene expression and cell function, play a major role in endothelial homeostasis and may represent an emerging therapeutic modality. However, delivery of miRNAs to cells in vitro and in vivo is challenging due to rapid degradation by ubiquitous RNases. Herein, we developed a nanoparticle delivery system separately combining deacetylated poly-N-acetyl glucosamine (DEAC-pGlcNAc) polymers with miRNA-126–3p and miRNA-126–5p and testing these combinations in vitro and in vivo. Our results demonstrate that DEAC-pGlcNAc polymers have an appropriate size and zeta potential for cellular uptake and when complexed, DEAC-pGlcNAc protects miRNA from RNase A degradation. Further, DEAC-pGlcNAc efficiently encapsulates miRNAs as evidenced by preventing their migration in an agarose gel. The DEAC-pGlcNAc-miRNA complexes were taken up by multiple cell types and the delivered miRNAs had biological effects on their targets in vitro including pERK and Dlk-1. In addition, we found that delivery of DEAC-pGlcNAc alone or DEAC-pGlcNAc:miRNA-126–5p nanoparticles to septic animals significantly improved survival, preserved vascular integrity and modulated cytokine production. These composite studies support the concept that DEAC-pGlcNAc nanoparticles are an effective platform for delivering miRNAs and that they may provide therapeutic benefit in sepsis.

INTRODUCTION

Sepsis impacts over 1,000,000 people a year and is one of the leading causes of death in intensive care units across the United States [1–4]. It is the most expensive condition to treat in US hospitals, with an economic burden of $20 billion in 2011 and increasing cost of 11.9% annually [1]. Sepsis is characterized by a massive release of inflammatory cytokines that facilitates the activation and trafficking of immune cells. Further, sepsis is frequently complicated by endothelial dysfunction that disrupts the microcirculation and leads to organ failure [5]. Therefore, therapies that blunt cytokine and chemokine release and mitigate endothelial cell damage in sepsis could have enormous therapeutic benefit.

The development of therapies that exploit RNA interference strategies have shown potential for treating cardiovascular, cancer, and inflammatory pathologies. MicroRNAs (miRNAs) are noncoding RNAs that are 21–25 nucleotides in length. MiRNA impacts post-transcriptional regulation of genes by causing gene destabilization and preventing messenger RNA translation [6]. MiRNAs may be effective in preventing endothelial cell damage and improving repair mechanisms in response to cellular stress and inflammation observed in sepsis [7]. MiR-126 is the most abundantly expressed miRNA in endothelial cells and is important for maintaining vascular integrity and angiogenesis [8]. Genetic deletion of pre-miR-126 impedes vascular development, halts angiogenesis and promotes vascular permeability [9–11], while suppression of miR-126 in the inflammatory microenvironment promotes enhanced expression of VEGF which similarly leads to increased vascular permeability [12].

We recently demonstrated that treatment of murine sepsis with endothelial progenitor cells (EPCs) increases circulating levels of miR-126 while reducing vascular leak and improving survival [13]. Moreover, we observed that EPCs release exosomes with abundant levels of miR-126 [14,15]. MiR-126 is known to inhibit a variety of genes, which impact endothelial cell homeostasis. Specifically, miR-126–3p targets sprouty-related EVH1 domain-containing protein 1 (Spred1), an inhibitor of angiogenesis, leading to phospho-ERK signaling and cofilin-mediated VE-cadherin stabilization [9]. MiR-126–5p targets the NOTCH1 inhibitor Dlk1 enhancing endothelial cell proliferation [16]. Given the potential beneficial effects of miR-126 on endothelial cell function, we postulated that delivery of miR-126 may improve outcomes in sepsis.

Several properties of nucleic acids impair their cellular uptake and potential therapeutic efficacy including their anionic charge, hydrophilicity, and susceptibility to ribonuclease degradation [17]. Thus, the development of a safe and effective miRNA carrier would be of importance for understanding the therapeutic value of miRNAs in vivo. We developed a nanoparticle delivery system that is both safe and effective. Deacetylated poly-N-acetyl glucosamine (DEAC-pGlcNAc) has excellent biocompatibility profiles and is biodegradable [18]. DEAC-pGlcNAc is cationic which allows it to form electrostatic interactions with anionic nucleotides like miRNA. Moreover, nanofibers consisting of pGlcNAc have been shown to be antimicrobial [19] while simultaneously supporting wound healing with a favorable safety profile [20]. We tested the ability of DEAC-pGlcNAc to form nanoparticles with miRNAs and thereby serve as a miRNA carrier both in vitro and in vivo and determine its effects in a model of acute sepsis. We characterized the properties of these nanoparticle complexes, analyzed their ability to protect miRNAs, examined their ability to be taken up by cells and modulate cellular function and measured their impact on inflammatory cytokine release and survival in a murine model of sepsis. We hypothesized that cationic DEAC-pGlcNAc would form nanoparticles with miRNAs and thereby would successfully deliver active miRNA-126 with minimal toxicity and improve sepsis survival.

MATERIALS AND METHODS

miRNA

Commercially available hsa-miR-126–5p (5’CAUUAUUACUUUUGGUACGCG), hsa-miR-126–3p, (5’UCGUACCGUGAGUAAUAAUGCG) and Allstars negative siRNA scrambled control were purchased from Qiagen (Valencia, CA). Additionally, Cy3 labeled hsa-miR-126–5p was purchased from GE Healthcare Dharmacon (Lafayette, CO).

Cell Culture

Pooled-human umbilical vein endothelial cells (HUVECs) were purchased from Lonza (Walkersville, MD) and cultured in Endothelial Basal Media-2 supplemented with EBM-2 SingleQuots (Walkersville, MD). Cells were cultured at 37°C in a humidified atmosphere with 5% CO₂. Cell culture media were renewed every 48 hours. Cells were passaged at 70% confluence using Trypsin/EDTA (Lonza) and were used at passages 3–5. NIH 3T3 embryonic fibroblast cells were a gift from Dr. Xian Zhang at the Medical University of South Carolina. Cells were cultured in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal bovine calf serum.

Nanoparticle Preparation

Sterile filtered 70% deacetylated-poly-N-acetyl-glucosamine (DEAC-pGlcNAc) 40mg/ml (molecular weight ~40,000) and sterile sodium sulfate (50mM) were obtained from Marine Polymer Technologies (Burlington, MA). Ratio dependent amounts of miR-126–3p and 5p (0.5µg-25µg) were added to 100ul of a 50mM sodium sulfate solution at room temperature in order to investigate encapsulation efficiency. Addition of DEAC-pGlcNAc (1.1 mg) was followed by high speed vortexing for 20 seconds. Complex self-assembly occurred at room temperature for 15 minutes. The nanopolymer mixture was neutralized using 0.5M NaOH with additional vortexing for 10 seconds. Following neutralization, the final pH of the complexes was 7. For RNase A and encapsulation studies, the formulations were pelleted by centrifugation (15,000 rpm, 4°C for 1 hour, 5424R Centrifuge, Beckman) for further analysis. The N/P ratio was calculated using the following equation:

$$\frac{N}{P}=\frac{numberofmolesofcationicDEAC-pGlcNACx150aminegroups}{numberofmolesofmiRNAx44phosphategroups}$$

Four different ratios of DEAC-pGlcNAc were generated 50:1, 700:1, 1400:1 and 2700:1 which contained 24.48µg, 1.8, 0.9µg and 0.45µg of miRNA, respectively.

Nanoparticle characterization

Following formulation, DEAC-pGlcNAc:miRNA nanoparticles were centrifuged and supernatant was removed.Nanoparticles were then suspended in 20 µl of 50mM sodium sulfate solution and placed on a stainless steel TEM grid and covered with 2% uranyl acetate for 30 seconds. Samples were rinsed in ultra-purified water and allowed to dry in a desiccator and imaged using a JEOL 1010 transmission electron microscope (Peabody, MA) with a Hamamatsu C4742–95 digital camera (Hamamatsu City, Japan). We determined size and zeta (Ζ) potential of the DEAC-pGlcNAc nanocomplexes using dynamic light scattering methodology (DLS), (Zetasizer, Nano ZS, Malvern, UK and ZetaPALS, Brookhaven Instruments Corp).

Electrophoretic Mobility Gel Shift Assay

The binding of DEAC-pGlcNAc and miR-126 was evaluated using a 4% (w/v) agarose E-gel and iBase system containing ethidium bromide. Briefly, nanocomplexes were formed and 20 µl of sample were added to the gel. The iBase electrophoresis was run based on the manufacturer’s recommendations. The miR-126 bands were analyzed using a UVP transilluminator (Alpha Innotech Corporation, Santa Clara, CA).

Evaluation of encapsulation efficiency

To evaluate the encapsulation efficiency of miRNA by DEAC-pGlcNAc, a standard curve using naked miRNA was generated. Briefly, various concentrations of miRNA were electrophoresed in a 4% (w/v) agarose E-gel using the iBase system for 30 minutes. Simultaneously, 4% SDS treated nanocomplexes were resolved in the same gels. The gel was analyzed using a gel doc system and bands were analyzed using image J. The standard curve was generated based on band intensity and the concentrations of encapsulated RNA were calculated based on the standard curve.

RNase A Protection Assay

The ability of DEAC-pGlcNAc to protect miRNA from RNase A enzyme was evaluated. Nanocomplexes and control naked miR-126–5p were treated with 0.18 µg RNase A for 1 hour at 37°C. Complexes were pelleted using centrifugation (15,000RPM, 25C) and treated with 4 µl EDTA (0.25M) for 10 minutes. Pellets were then re-suspended in 2% SDS and allowed to sit for 30 minutes prior to gel electrophoresis. Treated nanocomplexes and naked miRNA underwent gel electrophoresis using an 4% (w/v) agarose E-gel using the iBase system for 30 minutes and were then analyzed.

MTT Assay

The viability of HUVECs following treatment with various ratios of DEAC-pGlcNAc:miRNA nanocomplexes was assessed using the MTT (3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyl tetrazolium bromide) assay. Approximately 2.0×10⁴ cells/well were allowed to adhere overnight in a 96 well plate. Media were replenished and cells were treated with varying N/P molar ratio nanocomplexes. After a 24 hour incubation, cells were washed and the MTT solution was added. Cells were incubated at 37°C for 4 hours and the formazan was dissolved in acidic isopropanol. The viability of cells was assessed using a Synergy 4 microplate reader (BioTek) at absorbance 570nm with background subtraction at 630–690nm.

Fluorescence Microscopy

Complexes were formed using fluorescently labeled DEAC-pGlcNAc and Cy3 labeled-miRNA. HUVECs were treated for 24 hours with 25µl of the total complex, yielding molar concentrations of 55nM, 27.5 nM, and 14nM, respectively. Complexes were removed and cells were washed with PBS and fixed using 4% paraformaldehyde (PFA) for 20 minutes. Upon PFA removal, slides were mounted with a glass coverslip using ProLong® Gold antifade reagent with DAPI (ThermoFisher Scientific, Eugene, OR). Fluorescent images were acquired using an Olympus IX73 Research Inverted microscope (Olympus) equipped with a LUCPlan FLN 40X/0.60 Ph2 objective.

In vitro cellular uptake and quantitative Real Time RT-PCR of miRNAs

HUVECs and NIH 3T3 fibroblasts were seeded at a density of 1.0×10⁵ cells per well. Cells were allowed to adhere overnight and then treated with 25 µl of nanocomplexes for 24 hours. MiRNA was isolated using MiRNeasy Kit (Qiagen). RNA integrity was assessed using a BioTek Gen5™ (Winooski, Vermont) plate reader Take 3 software. For each reaction, 1 µg of miRNA was used for cDNA synthesis using a miScript II RT Kit (Qiagen). cDNA product was amplified with miScript SYBR Green PCR kit. A CFX96 Real Time PCR Detection System (Bio-Rad, Hercules, CA) was used to assess changes in hsa-miR-126–3p and 5p, and RNU6B as internal control.

Fibroblast growth factor mediated ERK phosphorylation

HUVECs seeded in 6-well plates were incubated with nanocomplexes at 50:1 containing 1.0µg per well of negative siRNA (Allstars, Qiagen) or miR-126–3p for 24 hours. HUVECs were completely serum starved with EBM-2 basal medium for 16 hours and then treated with fibroblast growth factor (FGF-2, 10 ng/ml) for 15 or 45 minutes. HUVECs were then placed on ice, washed in ice cold PBS and lysed with RIPA Buffer (50mM Tris-HCl pH 8.0, 150 mM NaCl, 1% Nonidet P-40 (NP-40), 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulphate (SDS), 1 mM sodium orthovanadate, 1 mM NaF) supplemented with a protease/phosphatase inhibitor and okadaic acid (20nM; Cell Signaling). Protein lysates were resolved in a 4–12% Bis-Tris Protein Gel at 100–200V for 1.5 hours. Proteins were transferred to a PVDF membrane at 100V for 1 hour and blocked using Odyssey Blocking Buffer (Li-Cor) for 1 hour. Blots were probed with phosphoERK 42/44 (1:1000), total ERK 42/44 (1:1000) overnight at 4C. Blots were analyzed using an Odyssey Licor Imaging System (Li-COR).

DLK-1 Western Blots

For DLK-1 experiments, HUVECs seeded in 6-well plates were incubated with nanocomplexes at a 50:1 ratio containing 1.0µg of negative siRNA (Allstars, Qiagen) or miR-126–5p for 48 hours. The cellular proteins were extracted and lysates were processed as described above. Blots were probed for α-tubulin (1:1000; Cell Signaling), or DLK-1 (1:1000; Abcam) overnight at 4°C and IRDye secondary antibodies (1:10000) for 1 hour. Blots were analyzed using an Odyssey Licor Imaging System (Li-COR).

Cecal-Ligation and Puncture (CLP)

CLP was performed in CD-1 male mice, aged 7–8 week old as previously described [21]. Briefly, after ligation with a 5–0 suture, a 22-gauge needle was used to produce two punctures in the cecum, after which the surgical site was closed with clips. All mice received saline subcutaneously after closure of the abdominal opening. Investigations were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health and were approved by the Institutional Animal Care and Use Committee at the Medical University of South Carolina. Mice were subjected to CLP and administered PBS, DEAC-pGlcNAc (5mg/kg), or DEAC-pGlcNAc:miR-126–3p or DEAC-pGlcNAc:miR-126–5p nanocomplexes (700:1 complex; 0.4µg per mouse/ per injection) via tail vein injection 2 and 6 hours after CLP. The antibiotic imipenem (25 mg/kg, subcutaneously) was administered at 6, 24, and 48 hours after CLP.

Blood Collection and Serum Isolation and Analysis

CD-1 mice previously subjected to CLP surgery were euthanized using isoflurane and blood was collected from the inferior vena cava 24 hours post-CLP surgery. Blood samples were kept at 25°C for 30 minutes. Samples were centrifuged at 10,000g for 10 minutes at 25°C. The serum fraction was collected and stored at −80°C prior to blood chemistry analysis. Blood chemistries were analyzed in the mouse pathology facility at MUSC. Mouse serum samples were shipped on dry ice to Eve Technologies (Calgary, Alberta, Canada) where a mouse cytokine-32 plex discovery assay was conducted.

Vascular Permeability Assay

CD-1 mice were subjected to CLP and administered PBS, DEAC-pGlcNAc, or DEAC-pGlcNAc:miR-126–3p or DEAC-pGlcNAc:miR-126–5p nanocomplexes via tail vein injection 2 and 6 hours after CLP as described above. Vascular permeability was quantified 24 hours post CLP using an Evans blue dye as previously described [13]. Briefly, 1% Evans blue (Sigma) dye was dissolved in saline solution. Two-hundred microliters of Evans blue dye was slowly injected into the tail vein of the mice using a syringe void of air bubbles. After 40 min, the mice were sacrificed, perfused via the heart with PBS, and the lung and kidney tissue were collected. The lung and kidney weights were measured and placed in 1 ml of formamide (Avantor) at 60°C for 24 h to extract Evans blue. Following centrifugation at 2,000 rpm for 10 min, the supernatant was collected. Evans blue in the lung and kidney tissue were quantified using spectrophotometric analysis at 620nm. Evans blue concentrations were extrapolated from a standard curve. The permeability index was calculated based on the extrapolated concentrations and normalized to the tissue weight.

Statistical Analysis

Data are expressed as mean ± the standard error of the mean (SEM). RT-PCR data were analyzed using ANOVA with a Tukey’s post-test or a Kruskal-Wallis test if data was not normally distributed. P-ERK western blot data was analyzed using a two-way ANOVA, matching both factors. A Student’s t-test was used to analyze Dlk-1 data. Survival analysis was performed using a Gehan-Breslow-Wilcoxon test. The Evans blue data was analyzed using a one-way ANOVA. Cytokine expression data were analyzed using the Kruskal-Wallis or ANOVA tests as appropriate. All statistical analysis was performed using GraphPad Prism 7 software. A p-value of <0.05 was used to reject the null hypothesis.

RESULTS

DEAC-pGlcNAc polymer effectively binds and protects miRNA from RNase A degradation

We chose to use DEAC-pGlcNAc nanoparticles as a delivery system for the miRNAs because they have non-specific antibacterial properties [19] and we hypothesized that miR-126 might provide additive or synergistic effects for the treatment of sepsis. DEAC-pGlcNAc nanoparticles were prepared as described in the methods. We performed an electrophoretic mobility shift assay using agarose gel to determine how effectively DEAC-pGlcNAc bound miRNA. The assay demonstrated that across all concentrations, miRNA was thoroughly bound to DEAC-pGlcNAc and unable to migrate through the gel matrix (Figure 1A). The nanocomplexes and control naked miRNA were also incubated with RNase A for 60 minutes. Minimal miRNA degradation was observed in the 50:1, 700:1 and 1400:1 complexes incubated with RNase A while substantial degradation did occur in the 2700:1 complex (Figure 1B, C). Densitometry analysis suggested that DEAC-pGlcNAc at 700 and 1400 molar excess protected 90% of the miRNA from degradation (Figure 1C, densitometry analysis not shown). Additionally, we sought to determine the optimal encapsulation efficiency of 700:1, 1400:1, and 2700:1 nanocomplexes. At a ratio of 700:1 approximately 80% of the miRNA was encapsulated while at ratios of 1400:1 and 2700:1 only about 50% of the miRNA was encapsulated (Figures 1D& E).

Figure 1. Synthesis and characterization of DEAC-pGlcNAc nanoparticles.

(A) Gel retardation of miRNA-126 by DEAC-pGlcNAc in a 5% agarose gel. Representative of 3 independent assays. (B) (C) Naked and DEAC-pGlcNAc -complexed miRNAs at a 50:1, 700:1, 1400:1, and 2700:1 ratio were incubated in the presence and absence of RNase A (0.1ug) for 1 hour at 37⁰C. (D) Naked miRNA at different microgram concentrations were used to establish a standard curve (lanes 2, 4–8) using an encapsulation efficiency assay. Nanocomplexes treated with 4% SDS are displayed in lanes 9–12. (E) Quantification of encapsulation efficiency based on the standard curve established using Image J and GraphPad Prism.

Optimization of the DEAC-pGlcNAc:miR-126 complex

In order to confirm that the nanocomplexes facilitate the cellular uptake of miRNAs, we incubated fluorescently labeled DEAC-pGlcNAc:cy3-miRNA-126–5p complexes with HUVECs for 24 hours and then analyzed the subcellular localization of the complex by fluorescent microscopy (Figure 2A). Cy3-positive cells were observed in the 700:1, 1400:1, and 2700:1 N/P ratio groups while Cy3 fluorescence was absent in cells incubated with Cy3-miRNA-126–5p alone (data not shown).

Figure 2. Optimization of in vitro cellular delivery of DEAC-pGlcNAc:miRNAs.

(A) Fluorescent microscopy analysis of Cy3-labeled miRNA-126 5p (Cy3-miRNA-126;red) uptake in HUVECs post-24 hour incubation at 37°C. Nuclei were counterstained with DAPI (blue). The arrows point to Cy3 co-localization with DAPI stained nuclei. (B) HUVECs incubated for 24 hours with DEAC-pGlcNAc alone or in complex with varying amounts of miRNA.

Cationic polymers can impact the metabolic activity and viability of the cell, thus we wanted to determine the impact of DEAC-pGlcNAc:miR-126 on cellular metabolic activity using an MTT colorimetric assay. HUVECs were exposed to DEAC-pGlcNAc:miR-126–5p nanoparticles for 24 hours at N/P ratios of 700:1, 1400:1, and 2700:1. Cells incubated in media without particles were used as a control. Differences in cell metabolic activity were not observed between control and experimental groups (Figure 2B). Based on these findings, the 700:1 N/P ratio was used for further characterization.

Characterization of the DEAC-pGlcNAc:miRNA nanocomplex

Dynamic light scattering (DLS) is a measurement of fluctuations in scattered light intensity due to Brownian motion of spherically shaped particles. Based on using DLS analysis, we observed that DEAC-pGlcNAc polymers alone had an average size of 287 nm (Figure 3A). The 700:1 was 204 nm (Figure 3A). The 700:1 nanoparticles had a zeta potential of +16.4mV and a polydispersity index (PDI) of 0.479 suggesting that the nanoparticles were fairly monodisperse (Figure 3B &C). TEM imaging further confirmed that the 700:1 nanoparticles were spherical in shape and monodispersed (Figure 3D).

Figure 3. Analysis of nanoparticles using dynamic light scattering (DLS) and TEM.

Size distribution of (A) DEAC-pGlcNAc alone and (B) DEAC-pGlcNAc:miRNA nanoparticles size at a 700:1 molar ratio as measured by DLS after resuspension in an acidic (HCl) water solution, pH 3.0. (C) Table showing measurements of average nanoparticle size, polydispersity, and zeta potential. (D) DEAC-pGlcNAc /miRNA nanoparticles visualized using transmission electron microscopy (TEM); scale bar: 50nm at a magnification of 200000X and 80kV power.

The DEAC-pGlcNAc:miRNA nanocomplex delivers miR-126 expression in vitro

To further quantify miRNA delivery following treatment of HUVECs with nanocomplexes, we performed qPCR. Incubation with DEAC-pGlcNAc:miR-126–3p complexes demonstrated significant increases in miR-126–3p expression (9-fold) compared to control and DEAC-pGlcNAc alone (Figure 4A); while incubation with DEAC-pGlcNAc:miR-126–5p complexes resulted in a non-significant trend toward increased cellular miR-126–5p (Figure 4B). We further examined miRNA delivery in NIH 3T3 cells which have low basal miR-126–3p and −5p levels. Incubation of NIH 3T3 cells with DEAC-pGlcNAc:miR-126–3p nanocomplexes resulted in a 1525±310 fold cellular increase (Figure 4C). Similarly, incubation with −5p complexes augmented expression 3772±3248 fold which failed to meet statistical significance due to inter-experimental variability (Figure 4D). Collectively, these data suggest that DEAC-pGlcNAc complexes can efficiently deliver miRNAs to enhance their cellular concentrations in vitro.

Figure 4. DEAC-pGlcNAc mediates miRNA-126 delivery.

Real-time PCR analysis of (A) miR-126–3p and (B) −5p levels in HUVECs after 24 hours. miR-126 levels were normalized to U6. Real-time PCR analysis of NIH 3T3 fibroblast cell line (C) miR-126–3p and (D) −5p levels after 24 hours DEAC-pGlcNAc complex treatment. miR-126 levels were normalized to U6. Data are mean ± SEM of 3 independent samples performed in duplicate, where *P<0.05, ***P<0.001, compared to control; #P<0.05 compared to DEAC-pGlcNAc; ϕϕP<0.01 compared to miR-126 alone using one-way ANOVA.

Previous studies have shown that miR-126–3p suppresses Spred-1 mRNA translation [9]. Spred-1 directly targets Rac-1 with subsequent suppression of ERK phosphorylation. Thus, we examined whether utilizing a 50:1 N/P molar ratio nanoparticles with reduced DEAC-pGlcNAc concentrations would induce in vitro biological effects as we previously demonstrated that the 50:1 complex sufficiently binds miRNA (Figure 1B) and protects against RNase A degradation (Figure 1C).

In efforts to further characterize the 50:1 N/P ratio particle size and effective delivery of miRNA, TEM, DLS, and qRT-PCR were performed. Using DLS technology, we confirmed that the mean diameter of the complexes was 190nM with a polydispersity index (PDI) of 0.34 (Figure 5A). To assess the functionality of the 50:1 complex, first we performed qPCR following HUVEC treatment with the 50:1 complex after 24 hours. Quantitation by RT-PCR was used to measure the changes in miR-126 expression compared to control (Figures 5B & C). In cells pre-incubated with 50:1 DEAC-pGlcNAc:miR-126–3p nanoparticles, FGF significantly increased the phosphorylation of ERK at 15 minutes in cells compared to DEAC-pGlcNAc:NsiRNA controls (p<0.001) (Figure 5D). Similarly, treatment of HUVECs with 50:1 DEAC-pGlcNAc:miR-126–5p nanocomplexes significantly reduced baseline expression of the miR-126–5p target Dlk1 (Figure 5E).

Figure 5. DEAC-pGlcNAc 50:1 complex mediates enhanced pERK expression and suppression of Dlk-1 in HUVECs.

(A) A table showing average 50:1 nanoparticle size and polydispersity. (B,C) Quantitative RT-PCR analysis of the relative expression of miR-126 3p and −5p levels in HUVECs treated with 50:1 DEAC-pGlcNAc:miR-126–3p or −5p nanocomplexes for 24 hours. Samples were normalized to U6. #P<0.05 compared to DEAC-pGlcNAc as assessed by One-Way ANOVA. (D) Western blot and densitometry bar graph of pERK and ERK expression induced by FGF (10ng/ml) at 0,15, and 45 minutes following DEAC-pGlcNAc:nsiRNA or -miR-126–3p nanopolymer treatment (50:1) for 30 hours. (E) Western blot and densitometry bar graph of DLK1 expression levels in HUVECs following a 48-hour incubation with DEAC-pGlcNAc:NsiRNA or -miR-126–5p nanopolymer treatment (50:1). Analysis for panels D &E was performed using a 2-way ANOVA with matching. Data are mean ±SEM of 6 independent experiments, where *P<0.05, **P<0.01 compared to NsiRNA control and ##p<0.01 compared to the DEAC-pGlcNAc:miR-126–3p control, ϕϕϕ p<0.001 compared to DEAC-pGlcNAc:miR-126–3p + 15 minute FGF (10ng/ml).

DEAC-pGlcNAc:miR-126–5p nanoparticles improve survival in septic mice

Although our in vitro studies showed that the 50:1 N/P ratio had more robust biological effects, based on the RNase A assay, we postulated that higher concentrations of DEAC-pGlcNAc may provide better protection from circulating RNases in vivo. Thus, we utilized the 700:1 complex for in vivo studies. Delivery of miR-126 has previously been shown to have therapeutic effects on vascular pathologies [22]; thus, we wanted to determine if DEAC-pGlcNAc:miR-126 delivery would improve survival outcomes in septic mice. Sepsis was induced by CLP in CD-1 male mice followed by antibiotic treatment. PBS, DEAC-pGlcNAc alone, DEAC-pGlcNAc:miR-126–3p, or DEAC-pGlcNAc:miR-126–5p were administered via intravenous tail vein injection at 2 and 6 hours post-CLP surgery. Treatment with DEAC-pGlcNAc:miR-126–5p significantly improved survival at 168 hours (7 days) post-CLP surgery compared to PBS controls (66.6% vs 25%; p<0.01) while treatment with DEAC-pGlcNAc:miR-126–3p did not change survival (6.25% vs 25%; p<0.01). (Figure 6). While we observed improved survival with delivery of DEAC-pGlcNAc alone, this increase in survival was not significantly different (p=0.055) than the survival of CLP-PBS controls.

Figure 6. Delivery of miR-126–5p improves survival in CLP mice.

(A) CD-1 mice (15–16 per group) were subjected to CLP surgery. Mice received tail vein injections of either PBS (n=16), DEAC-pGlcNAc (n=16), and DEAC-pGlcNAc:miR-126–3p (n=16) or −5p (n=15) nanopolymer treatments at 2 and 6 hours post-CLP surgery. Mice received intermittent treatments of imipenem at 6h, 24h, and 48h post-CLP. (B) Survival of CLP mice exposed to PBS, DEAC-pGlcNAc, DEAC-pGlcNAc:miR-126–3p, DEAC-pGlcNAc:miR-126–5p with survival monitored for 7 days. Differences in survival were assessed using Gehab-Breslow-Wilcoxon test, where **p<0.01 compared to CLP-PBS and #p<0.05 compared to CLP + DEAC-pGlcNAc:3p.

DEAC-pGlcNAc nanoparticles decrease vascular leakage and suppress the inflammatory immune response in CLP-induced sepsis

Mice underwent sham or CLP surgery and were injected with DEAC-pGlcNAc nanoparticles at 2 and 6 hours post-CLP. An Evans blue dye assay revealed that DEAC-pGlcNAc, DEAC-pGlcNAc −3p, and DEAC-pGlcNAc-5pnanoparticles significantly decreased lung vascular leakage (48±28%, 42±29%, 40±32%, respectively, p<0.01) and kidney vascular leakage (61±23%, 43±33%, 49±26%, respectively, p<0.01, Figure 7A). Serum samples were collected 24 hours post-CLP. DEAC-pGlcNAc and DEAC-pGlcNAc:miR-126–5p significantly reduced serum IL-6 (12±0.7 and 6.9±0.4, p<0.01) and KC (9.4±0.25 and 4.7±0.10, p<0.01) levels. In contrast, DEAC-pGlcNAc-miR-126–3p had no significant effect on those important inflammatory mediators. DEAC-pGlcNAc, DEAC-pGlcNAc:miR-126–5p and DEAC-pGlcNAc:miR-126–3p significantly reduced IL-1β, MP-1α, MCP-1, and IL-10 (Figure 7B).

Figure 7. DEAC-pGlcNAc nanoparticle effects on vascular leakage and inflammatory mediators in CLP-induced sepsis.

CD-1 male mice were subjected to CLP surgery. Animals received tail vein injections of either PBS, DEAC-pGlcNAc, DEAC-pGlcNAc:miR-126–3p or −5p nanopolymer treatments at 2 and 6 hours post-CLP surgery. (A) Vascular leakage in the lung and kidney were determined using an Evans blue dye assay 24 hours post-CLP surgery. (B) Serum was collected from whole blood and analyzed for cytokines. Data were analyzed using a One-Way ANOVA or Kruskal-Wallis test, n=4–7 mice per group, *p<0.05, **p<0.01, ***p<0.001 compared to sham. #p<0.05, ##p<0.01 compared to CLP-PBS.

DISCUSSION

The development of miRNA therapies requires effective platforms for miRNA delivery into the cytoplasm of cells. Current delivery systems including viral vectors, liposomes, and chitosan have proven to be suboptimal due to their complex assembly and inefficient delivery of nucleic acids in vitro and in vivo [23]. To reduce these limitations, we used a DEAC-pGlcNAc nanoparticle/miRNA complex that, when formed, yielded efficient delivery of miRNAs to HUVECs and NIH 3T3 fibroblasts. Our results demonstrate that the DEAC-pGlcNAc:miRNA-126–5p nanocomplex enhances intracellular levels of miR-126 in vitro and improves sepsis outcomes in vivo.

There are several potential advantages of the DEAC-pGlcNAc polymer as a miRNA delivery platform in sepsis. Its known antimicrobial effects make it an ideal delivery platform for sepsis [19]. Linder et al. previously demonstrated that poly-N-acetyl-glucosamine (sNAG) nanofibers, an analogue of DEAC-pGlcNAc have antibacterial effects in wound healing. Specifically, they showed that sNAG enhanced both α- and β-defensin expression in endothelial cells and keratinocytes while reducing bacterial infections in cutaneous wounds infected with Staphylococcus aureus [19]. Thus, its anti-bacterial effects, are a potential advantage to its use as a miRNA delivery vehicle in sepsis. Its ability to efficiently encapsulate miRNA and protect it from RNase degradation is critically important to ensuring stability of miRNA in the in vivo environment. Further, the nanocomplex diameter (~200nm) allows it to avoid uptake by the liver and reticuloendothelial system during circulation while its cationic zeta potential suggests moderate stability [24].

Our data demonstrate that DEAC-pGlcNAc:miRNA complexes effectively deliver miRNA to cells leading to alterations in cellular function. It is worth noting that the magnitude of delivered intracellular miRNA appeared to differ by cell type with HUVECs exhibiting significantly smaller increases in miR-126 levels than 3T3 fibroblasts. Although this could be a property of cell-specific delivery efficiency, we suspect that this observation is more likely dictated by the significantly higher baseline levels of miR-126 [13] in endothelial cells as compared to fibroblasts where miR-126 is not normally expressed. This is supported by the robust amount of intracellular fluorescently tagged miR-126 seen within HUVECs after treatment with the nanocomplexes. Interestingly, although the 700:1 concentration of nanocomplexes appeared to provide optimal miRNA protection, our in vitro data suggested that this concentration did not allow for optimal miRNA activity upon intracellular delivery (data not shown). We hypothesized that this concentration may impede miRNA release from the nanocomplex leading to impaired miRNA function. Notably, the 50:1 concentration of DEAC-pGlcNAc:miRNA still delivered intact miRNA intracellularly while allowing the miRNA to dissociate from the complex and inhibit gene translation. Future investigation into the optimal concentration for both miRNA delivery and function will be important.

Our findings revealed that DEAC-pGlcNAc nanoparticles loaded with miR-126–5p significantly improved survival in animals subjected to CLP compared to PBS controls. These findings complement our previous studies showing that EPCs that are rich in miR-126–5p significantly improve CLP survival [13]. Survival rates between −5p and −3p nanoparticles were statistically different with −3p also failing to improve survival compared to PBS treatment in sepsis. This suggest that the delivery of miR-126–3p using DEAC-pGlcNAc nanoparticles was ineffective and potentially deleterious. Interestingly, there appeared to be a trend toward improved survival when treating with DEAC-pGlcNAc nanoparticles alone (not coupled to miR-126–3p or −5p) compared to the PBS control. However, there was no significant difference compared to miR-126–5p loaded nanoparticles.

Circulating cytokines and oxidative stress leads to endothelial dysfunction and increased vascular permeability in CLP sepsis. We observed that all formulations of nanoparticles with or without miRNA preserved vascular integrity compared to CLP-PBS animals 24 hours post-CLP surgery. However, vascular integrity measured at 24 hours may not be predictive of long-term outcome in CLP sepsis. Interestingly, DEAC-pGlcNAc and DEAC-pGlcNAc:miR-126–5p but not DEAC-pGlcNAc:miR-126–3p treatment reduced the hallmark sepsis cytokine levels of IL-6, KC, and IFN-gamma. The aforementioned cytokines are most predictive of sepsis outcomes. The failure to decrease these inflammatory mediators may provide a partial explanation for why DEAC-pGlcNAc:miR-126–3p failed to improve the mortality.

One possible explanation for the ineffectiveness of miR-126–3p might be that it functions as a miR-126–5p antagomir thus reducing the effects of innate miR-126–5p. In human sepsis [14] there is an increase in plasma miR-126 levels and if similar trends persist in experimental animal models of sepsis, excess miR-126–3p may neutralize the beneficial effects of miR-126–5p. Further, the miR-126/VEGF axis modulates angiogenesis and junctional protein stability in vivo which has been linked to increased vascular permeability in sepsis. MiR-126–3p promotes activation of VEGF mediated pathways including the phosphorylation of ERK, which has been proven to be detrimental to vascular integrity in pathophysiologic conditions [25,26].

One limitation of this study was that we dosed mice based on the polymer concentration (5mg/kg) rather than the miRNA concentration. Thus, we only administered 720ng (18ng/g) of miRNA to CLP mice. The large amounts of DEAC-pGlcNAc delivered may have blunted the potential impacts of −5p on the disease process. In addition, we only treated animals at 2 and 6 hours post-surgery as opposed to continuous treatments over the course of sepsis. Future studies will examine the impacts of 50:1 nanocomplexes that contain increased amounts of miRNA in order to determine the presence of a dose-response effect. Additionally, targeting of nanocomplexes to the endothelium using the αvβ3 integrin ligand LXW7 modification [27] or RGD peptide might allow us to target the endothelium in vivo and parse out the effects of miR-126–5p on endothelial function in the sepsis microenvironment.

CONCLUSION

The DEAC-pGlcNAc polymer serves as a viable option for delivery of miRNAs, DEAC-pGlcNAc:miR-126–5p nanoparticle complexes improved survival in CLP animals and reduced the inflammatory cytokine response. Collectively, these studies raise the possibility that DEAC-pGlcNAc and/or complexed with miR-126–5p may represent a novel therapeutic approach for human sepsis.