Targeted, Stimuli-Responsive Delivery of Plasmid DNA and miRNAs

Dec 29, 2017 - Granulocyte Colony-Stimulating Factor Nanocarriers for Stimulation of the Immune System (Part I): Synthesis and Biodistribution Studies...
0 downloads 0 Views 2MB Size
Subscriber access provided by the University of Exeter

Article

Targeted, stimuli-responsive delivery of plasmid DNA and miRNAs using a facile self-assembled supramolecular nanoparticle system Li-Yen Wong, Bingzhao Xia, Ernst Wolvetang, and Justin J. Cooper-White Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.7b01462 • Publication Date (Web): 29 Dec 2017 Downloaded from http://pubs.acs.org on December 30, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Targeted, stimuli-responsive delivery of plasmid DNA and miRNAs using a facile self-assembled supramolecular nanoparticle system

Li-Yen Wong†, ‡, Bingzhao Xia†, Ernst Wolvetang‡ and Justin Cooper-White†, §, #,*, †

Tissue Engineering and Microfluidics Group, Australian Institute for Bioengineering and

Nanotechnology, The University of Queensland, 4072 Brisbane, Australia ‡

Stem Cell Engineering Group, Australian Institute for Bioengineering and Nanotechnology, The

University of Queensland, 4072 Brisbane, Australia §

Biomedical Manufacturing, CSIRO, Monash University, Clayton, VIC, 3069, Australia

#

School of Chemical Engineering, The University of Queensland, St Lucia, QLD 4072, Australia

*Email correspondence: [email protected]

ACS Paragon Plus Environment

1

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 38

Abstract

Gene therapy is rapidly regaining traction in terms of research activity and investment across the globe, with clear potential to revolutionise medicine and tissue regeneration. Viral vectors remain the most commonly utilised gene delivery vehicles, due to their high efficiency, however, they are acknowledged to have numerous drawbacks, including limited payload capacity, lack of cell-type specificity, and risk of possible mutations in vivo, hence patient safety. Synthetic nanoparticle gene delivery systems can offer substantial advantages over viral vectors. They can be utilised as off-the-shelf components to package genetic material, display targeting ligands, release payloads upon environmental triggers, and enable the possibility of programmed cellspecific uptake and transfection. In this study, we have synthesised three functional polymeric building blocks that, in a rapid, facile, tailorable and stage-wise manner, associate through both electrostatic and non-covalent hydrophobic ‘host-guest’ interactions to form monodisperse selfassembled nanoparticles (SaNP). We show that these SaNPs successfully package significant amounts of microRNA through to plasmid DNA, present desired ligands on their outer surface for targeted receptor-mediated cell-specific uptake and affect efficient translation of packaged plasmids. We confirm that these SaNPs outperform commercially available, gold standard transfection agents in terms of in vitro transfection efficiencies and have very low cytotoxicity. With facile self-assembly and tailorable composition, our SaNP gene delivery system has significant potential in targeted gene therapy applications.

KEYWORDS: cyclodextrin, polyrotaxane, supramolecular, targeted, specific, self-assembled polymer, gene delivery

ACS Paragon Plus Environment

2

Page 3 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1. Introduction One of the major challenges in translating gene therapy into the clinic is the shortage of available efficient and safe vectors that can stably deliver plasmid DNA or RNA in a target-specific manner with minimum toxicity, whilst also protecting such genetic material from degradation1, 2. Viral vectors have been extensively utilised and are currently the gold standard delivery vectors due to their high transfection efficiency but are limited in clinical applications for human use, owing to their safety concerns, such as immunogenicity, stability and insertional mutagenesis1, 2. Furthermore, the lack of cell-type specificity, random integration into the host genome, limited loading capacity, and high cost of production present major drawbacks for generic therapeutic use in future tissue engineering and regenerative medicine applications.

Self-assembling supramolecular nanoparticles have emerged as promising gene delivery vectors, due to their ability to bind, condense and package plasmids, as well as release cargo and increase stability against nucleases3, 4.

These systems can be self-assembled through a number of

mechanisms, but most recently have relied on host-guest complexation of individual components to form the nanoparticles3, 5-8. Host-guest interactions can be tuned by the selection of the two components and can be very robust, even when exposed to substantial changes in solution conditions (pH, temperature). Pendant hydrophobic moieties that can dynamically interact and form highly specific inclusion complexes with one another offer the possibility of multicomponent, multilayered nanoparticle self-assembly, and even the presentation of targeting molecules at the periphery of formed nanoparticle. Of particular relevance to this current work, ‘host’ β-cyclodextrins (βCDs), cyclic toroidal oligosaccharides with a hydrophobic central cavity, can form inclusion complexes with a ‘guest’ hydrophobic molecule (e.g. adamantane

ACS Paragon Plus Environment

3

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 38

(Ad)) by non-covalent interactions9-11,5, 9, 12. Numerous researchers have confirmed the utility of βCD-Ad coupling in nanoparticle assembly and associated delivery of genetic material to a range of cell types, inclusive of some examples of cell-specific uptake9-11,13-15. However, improvements in design and function are still required to ultimately improve their utility, flexibility of application and performance.

For example, the lack of control over component stoichiometry during self-assembly of these constructs currently impacts the ability to tailor the particle size, payload of genetic material and peripheral (surface) composition (e.g. the amount of targeting ligand versus degree of pegylation). This difficulty in explicitly tailoring the stoichiometry of the assembled multiple components is due principally to all currently published systems relying on the peripheral components (i.e. besides the central DNA/RNA binding βCD-modified polycation) being Admodified polymers

15

. Furthermore, these systems have relied on the ‘proton sponge’ effect of

the polycation to engender endosomal escape, whereas in other systems, the inclusion of membrane-disruptive components (such as cyclodextrins) into degradable polyrotaxane constructs has been shown to substantially enhance endosomal escape16. This release of membrane disruptive components ideally would only happen at low (endosomal) pH, allowing for programmable release of packaged cargo, but given their hydrophilic nature, prior to release they should ‘shroud’ the packaged material to protect it from degradation prior to uptake. Lastly, it would be ideal if the outer surface of these nanoparticles could be simply functionalized with multiple receptor targeting ligands, offering the opportunity to have co-receptor, targeted uptake. To achieve tailored self-assembly, targeted uptake, high efficiency release of payload material

ACS Paragon Plus Environment

4

Page 5 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

and, ultimately, efficient transfection, we sought to design a new self-assembling nanoparticle system that would address these particular deficiencies and thus offer novel functionality.

In this paper, we report the synthesis and validation of a novel supramolecular self-assembling nanoparticle (SaNP) system for gene delivery. The three layered system is composed of: layer 1 a cationic β-cyclodextrin-modified polyethyleneimine (CDP) to bind the genetic material; layer 2 - a stimuli (pH)-responsive Tetronic polyrotaxane end-capped with Adamantane (Tet-PRX-Ad) to both protect the payload from enzymatic degradation and enhance endosomal escape in a programmed manner through programmed release of αCDs; layer 3 - a peptide-modified βcyclodextrin (CD-peptide) that functions as a ‘clickable’ ligand for targeted receptor-mediated uptake. To validate the particle system in terms of its ability to target cell-surface receptors, we synthesised a βCD functionalized with a single RGD peptide, chosen to specifically target αvβ3positive U87 cells that have been genetically modified to overexpress this particular integrin pair.

2. Materials and Methods 2.1 Materials Tetronic 1307 (Tet, Mw = 18kDa) was purchased from BASF Corporation Pty. Ltd. (USA). Branched polyethyleneimine (PEI, MW=25kDa), β-cyclodextrin (MW=1135Da, imidazole (Sigma Aldrich), p-toluenesulfonyl chloride (Sigma Aldrich), α-cyclodextrin (MW=984Da), sodium azide (Sigma Aldrich), N,N'-dicyclohexylcarbodiimide (DCC) (Sigma Aldrich), Nhydroxysuccinimide (NHS) (Sigma Aldrich), ammonium chloride (Sigma Aldrich), 1,1,2,2tetrachloroethane (Sigma Aldrich), 4-dimethylaminopyridine (Sigma Aldrich), 1-Adamantyl amine, anhydrous dimethyl sulfoxide (DMSO) (99%), acetone, sodium hydroxide, triethylamine

ACS Paragon Plus Environment

5

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 38

(99%), succinic anyhydride, anhydrous N,N-dimethyl formamide (DMF, 99.8%), copper (II) sulfate pentahydrate (CuSO4. 5H2O), sodium ascorbate (NaOAs), tert-butyl alcohol (t-BuOH), heparin sodium salt, and agarose were purchased from Sigma Aldrich. 6-mono-tosyl-βcyclodextrin (CD-Ts) was synthesized as previously reported14,

17, 18

. Alkyne-terminated RGD

peptide was purchased from China Peptides. RPMI 1640, Opti-MEM Reduced Serum Medium, penicillin/streptomycin, Lipofectamine 2000 and Picogreen reagent were obtained from Invitrogen, Australia. Fetal Bovine Serum (FBS) was purchased from Scientifix. EGFP-encoded plasmid DNA (pMAX-GFP, 3.5kbp) referred to as pGFP was obtained from Lonza Walkersville Inc. Cy3-miR (AM17120) was purchased from Thermofisher Scientific. SYBR Safe DNA Gel Stain was purchased from Invitrogen.

2.2 Plasmid DNA (pDNA) isolation The plasmid DNA (pGFP) encoding GFP was transformed in Escherichia coli XL-1 Blue competent cells and propagated in LB broth (Sigma Aldrich) supplemented with 50µg/ml kanamycin (Sigma Aldrich). The plasmid DNA was then purified using PureLink Quick Plasmid Miniprep Kit (Invitrogen, Australia) and eluted in 10mM TE buffer, pH 8.5. The purity of the plasmids consisting of supercoiled and open circular forms was determined using electrophoresis on 1.0% agarose gel in 1x TAE buffer, and the concentration of DNA was determined by measuring UV absorbance at 260 nm and 280 nm using Nanodrop spectrophotometer (Thermofisher Scientific). The ratio of absorbances at 260nm to 280nm is 1.8-1.9.

2.3 Synthesis and characterisation of nanoparticle components

ACS Paragon Plus Environment

6

Page 7 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

The nanoparticle components were synthesised (as described in Supplementary Information and Scheme S1 (CDP) and S2 (Tet-PRX-Ad)) and all components and their intermediates characterised using FTIR, NMR and Mass Spectrometry to confirm conjugation reactions (please refer to Figures S1-S16 in Supplementary Information). These were assembled into complexes with pDNA and thereafter characterised using DLS to determine nanoparticle size and zeta potential (please refer to Tables S1-3 in Supplementary Information). The FT-IR was performed on a Nicolet 5700 spectrometer (Thermo Fisher Scientific, MA, USA). The samples were scanned between 500 and 4000 cm-1 at a resolution of 1.928 cm-1 in the attenuated total reflection (ATR) mode and an average of 64 scans was used for the analysis. All 1H-NMR spectra were collected at room temperature on a Bruker Avance 700 NMR spectrometer operating at 700 Hz using DMSO-d6 or D2O as the solvent. The samples were run in diffusion experiments, in which high gradient strength of 90% was used to remove all small, unreacted molecules. Integration of peaks were performed on quantitative measurements to calculate the stoichiometric ratio of conjugated compounds. Mass spectrometry experiments were performed in positive ion mode on a BRUKER MicrOTof-Q mass spectrometer with a set capillary of 4000V and flow rate of 4µl/min.

2.4 Preparation of CDP/DNA or Cy3-miR complexes Polyplexes are formed due to interaction between the positively charged surface of PEI and negatively charged backbone from pDNA or Cy3-miR phosphate groups. They undergo selfassembly in aqueous solutions. Different amounts of CDP were complexed with 1µg of pDNA or Cy3-miR according to the calculation of the N/P ratio in 1.5 ml eppendorf tubes. All the tubes were vortexed for 6 seconds and incubated at room temperature for 20 minutes.

ACS Paragon Plus Environment

7

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 38

N/P ratios were calculated based on the following equation:

    (μ) =

N: P x Amount of nucleic acid x Number of moles Concentration of nitrogen content in polymer

2.5 Preparation of supramolecular complexes Nanoparticle components were assembled by sequential addition of the pre-formed CDP/DNA or CDP/Cy3-miR complex, Tet-PRX-Ad and CD-RGD in RNase, DNase-free water as previously reported13, 14. The core of the self-assembled polymers, CDP/DNA (7.22µM for 1 ug of pDNA) or CDP/Cy3-miR complex was first formed at desired N/P ratios prior to assembly with other components. Tet-PRX-Ad (47.5 µM) was added to form inclusion complexes with CDP/DNA or CDP/Cy3-miR, decorating the external CDP chains with Ad-terminated polyrotaxanes (TetPRX-Ad). CD-RGD (2.6mM) was then incubated with the CDP/DNA (or Cy3-miR)/Tet-PRXAd to form the corona of the supramolecular self-assembled system (termed SaNP-RGD). We hypothesize that given the complexation of Ad to βCD is 1:1 and the equilibrium constant lies between 104–105 M−1, we engineered our system to allow sufficient Ad molecules to interact with both CDP and CD-peptide considering the availability of Ad molecules (calculated from NMR) to form inclusion complexes7,

19-23

. These particles were characterised using DLS to

determine nanoparticle size and zeta potential.

2.6 Nanoparticle size and zeta potential measurements using dynamic light scattering The average hydrodynamic diameter and zeta potential were determined using a Zetasizer Nano ZS (Malvern Instruments, Worcestershire, UK). DLS experiments were performed with a

ACS Paragon Plus Environment

8

Page 9 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Zetasizer Nano ZS instrument equipped with a 5-mW helium-neon laser (λ= 633 nm) and thermoelectric temperature controller. Measurements were taken at a 173° scattering angle. SaNP samples for DLS were made up in water to make 173°-scattering intensities around 10,000 counts per second. The raw data were subsequently correlated to the intensity, volume and number mean using a cumulative analysis by the Zetasizer software. The sizes and the standard deviations of assembled SaNPs were calculated by taking the average values of three measurements (Tables S1-3).

2.7 DNA retardation assay The pDNA loading capacity of the PEI, CDP and SaNP-RGD were assessed by gel electrophoresis. The polymer/DNA complexes were prepared as previously described. Polyplexes at different N/P ratios (0.5, 1, 1.5, 2, 2.5, 5, 10, and 20) containing 1 µg of plasmid DNA were mixed with RNase and DNase-free water to a total volume of 10µl. 10µl of each sample plus 2µl of 6x loading dye was loaded on a 1% agarose gel in 1x TAE buffer (40mM Tris-acetate, 1mM EDTA) which was pre-stained with GelRed DNA Gel Stain (Biotium) in a 1:10000 dilution according to manufacturer's protocol. The complexes were incubated for 30 minutes prior to running the gel at 100 V for 40 min and visualized under UV illumination using ChemiDoc MP (BioRad).

2.8 Heparin binding assay using Picogreen The stability of the pDNA-loaded SaNPs against a polyanion competitive exchange was evaluated by Picogreen assay based on the release of pDNA from the polymer complexes. In this

ACS Paragon Plus Environment

9

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

experiment, heparin (heparin ammonium salt from porcine intestinal mucosa, Sigma) was used as the polyanion. The pDNA-loaded SaNP complexes were prepared as described previously. Different heparin concentrations were prepared in water from 0µg/ml up to 500µg/ml. The SaNPs were incubated with 10µl of heparin solution of different concentrations for 30min in 96 well plates before adding 50µl of Picogreen reagent at 1:200 ratio. The fluorescence of PicoGreen (λ ex 485nm, λ em 535 nm) for each sample was measured with a Biotek Synergy HT microplate reader. Relative fluorescence was calculated as:

  )23 4 567 .* 8 − )23 (:*0;) )*+,-  ./0/ (%) = )23 ( 567 /0+ ) − )23 (:*0;)

2.9 Structural morphology and SaNP size using AFM Microscopy Polymer-pDNA complexes were prepared at ratios 5:1, at a final DNA concentration of 0.1ng/ml. For each sample, 50µl of sample was adsorbed onto mica substrate, dried under nitrogen gas and stored under vacuum overnight. The samples were imaged under Cypher AFM (Asylum Research) under tapping mode in air. The resonance frequency of the tapping mode cantilever was set at 291kHz and scanning rate was 0.5Hz. The cantilever used was gold-coated Multi75GD.

2.10 Effects of endosomal pH on SaNP disassembly

The SaNPs were prepared as described previously and incubated in different buffers at pH 7.2 and 5.4. Measurements were taken at 0 min, 5 mins, 15 mins and 24 hrs to evaluate the

ACS Paragon Plus Environment

10

Page 11 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

disassembly of the polymers. Nanoparticle size distributions were recorded from three technical replicates using a Malvern Zetasizer NanoZS instrument.

2.11 Cytotoxicity tests of various polymer concentrations To evaluate the cytotoxicity of the polymer components, U87 cells were seeded at 10,000 cells/cm2 on a BD Falcon 96-well plate and cultured in RPMI-1640 containing 10% fetal bovine serum (FBS) and 1% Penicillin-Streptomycin in 5% CO2 at 37°C for 24 hours prior to treatment with the polymers. The cells were then incubated with different polymer concentrations of 10, 25, 50, 100, 250, 500 µg/ml (n=3) for 4 hours. The culture media was then replaced with fresh media containing 10 times dilution of Cell Counting Kit-8 (CCK8) (Sigma Aldrich). The cells were further incubated for 3-4 hours until a noticeable colour change was observed in the wells. The absorbance was measured at 450nm on a Biotek Synergy HT microplate reader. The cell viability was calculated by the following formula:

100% =5 (/0+ ) − =5 (:*0;)

2.12 Targeted transfection in U87 cells using SaNP-RGD To evaluate targeted transfection of SaNP-RGD on human glioblastoma cell line, U87 cells, a receptor blocking experiment was performed to saturate the αvβ3 receptors with a αvβ3 antibody and to inhibit cellular uptake of SaNP-RGD nanoparticles carrying pGFP or Cy3-miR. Briefly, U87 cells were seeded at 10,000 cells/cm2 and allowed to attach overnight. The cells were then treated in the presence and absence of αvβ3 antibody at a dilution of 1:20 to assess the specificity of the SaNP-RGD complex. The polymers were assembled under desired ratios and allowed to

ACS Paragon Plus Environment

11

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

incubate for 30 mins after the addition of each layer. The cells were then transfected with the polymers for 4 hours prior to media change and a further incubation of 48 hours. The controls used were pGFP or Cy3-miR alone, Lipofectamine 2000 (L2K), PEI25k, and CDP. The cells were then fixed, stained with Hoechst and imaged using fluorescence microscopy with an Olympus BX61 inverted microscope.

Additionally, to assess the specific and targeted delivery of pGFP or Cy3-miR into U87 cells using SaNP-RGD nanoparticles, we transfected U87 cells with SaNP-RGD in the presence and absence of 100µM of RGD peptide. The cells were incubated with the particles for 4 hours prior to a media change and further incubation for another 48 hours. The cells were then fixed, stained with Hoechst and imaged using fluorescence microscopy with an Olympus BX61 inverted microscope.

3. Results The concept of our self-assembled supramolecular nanoparticle system is shown pictorially in Scheme 1. We firstly detail the synthesis of each of these components and their respective intermediates prior to confirming their self-assembly into monodisperse nanoparticles and their ability to deliver genetic material in a programmed manner.

ACS Paragon Plus Environment

12

Page 13 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Scheme 1. The supramolecular self-assembling nanoparticle (SaNP) system was assembled based on varying ratios of CD-PEI (CDP), modified Tet polyrotaxane (Tet-PRX-Ad) and CDpeptide with CDP forming a complex with pDNA or miRNA, followed by the addition of Adterminated Tet polyrotaxane and CD-peptide on the outer shell of the polymer.

ACS Paragon Plus Environment

13

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 38

3.1 Confirmation of synthesis of individual nanoparticle components Tosyl imidazole (Ts-Imi) was synthesised and characterised by 1H-NMR. The tosyl peaks were shown at 7.96 and 7.5 ppm while the imidazole peaks were observed at 8.35, 7.71 and 7.11 (Figure S1a). The ratio of Ts to Imi was 1:1. This was then further conjugated to β-CD to yield tosyl-activated βCD (CD-Ts). CD-Ts was then characterised by 1H-NMR which demonstrated tosyl peaks on β-CD at 7.72 and 7.41, with a 62.3% conjugation efficiency of tosyl groups to βCD (Figure S1b,c).

CDP was produced by the reaction between CD-Ts and PEI. This reaction was performed such that CD-Ts was used in excess to functionalise PEI chains. The grafting of CD-Ts onto PEI was confirmed by 1H-NMR analysis (Figure S2). Two out of four primary amine groups on PEI were functionalised with βCD, which is ideal for yielding protonable primary amine groups for electrostatic binding to pDNA, as well as providing available CDs for forming inclusion complexes with Ad. For the purposes of this study, we chose to use a high molecular weight PEI (HMW PEI) so that there are ample amounts of available amine groups both for condensing and packaging pDNA and for effective buffering capacity to induce endosomal rupture24-26.

CD-azide was then synthesised from intermediate tosyl-activated βCD to be further used for click chemistry. CD-azide was characterised by FTIR and

13

C-NMR. By FTIR, the azide peak

was evident at 2107 cm-1 (Figure S3) and by 13C-NMR, the carbon adjacent to the azide showed a peak at 62.93 and 53.75 ppm (Figure S4).

ACS Paragon Plus Environment

14

Page 15 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Click chemistry was performed using alkyne-terminated RGD peptide and CD-azide. The reaction yielded CD-RGD which was characterised by 1H-NMR and FTIR. The peak from the triazole ring was designated at 8.35ppm (Figure S5). By integration of the peaks, it was observed that one out of seven azide groups were functionalised with the RGD peptide. By FTIR, the near complete disappearance of the azide peak was evident at 2107 cm-1 (Figure S3). We have further confirmed the conjugation of RGD peptide to CD by mass spectrometry, as shown by a peak at m/z = 1762.5 in Figure S16.

Tetronic 1307 was modified by reaction with DMAP, TEA and succinic anhydride to form TetCOOH. This was then further mixed with αCD to form pseudo-polyrotaxane, Tet-PRX-COOH. Due to size restriction, αCD only threads onto the PEO-blocks of the 4-arm PEO-PPO tetronic. The Tet-PRX-COOH polymer was then end-capped with Adamantane to yield Tet-PRX-Ad. The polymers were characterised by 1H-NMR. The CH2-CH2 of the succinyl group was observed between 2.5-2.7ppm (Figure S6). After threading of αCD to Tet-COOH, integration of peaks showed about 50% of αCDs were threaded onto PEO chains of the tetrapolymer, Tet-COOH. Tet-PRX-COOH was further reacted with amine-terminated Adamantane to yield Tet-PRX-Ad with 3 out of 4 arms functionalised with Ad.

3.2 SaNP components display efficient packaging of plasmid DNA The ability of our polymers to condense DNA into polyplexes is a primary requirement for gene delivery applications. The binding and condensation capacity of the branched PEI (25 kDa), CDP and SaNP-RGD at different N/P ratios was examined by agarose gel retardation analysis. Firstly, the cationic polymers, PEI (positive control) and CDP were incubated with pDNA at

ACS Paragon Plus Environment

15

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

various N/P ratios from 0, 0.5, 1, 1.5, 2, 2.5, 5, and 7.5 to determine the binding capacity of the polymer to the anionic pDNA. PEI was completely bound to pDNA at N/P ratio of 2.0 (Figure 1A). In contrast, Figure 1B exemplifies that at an N/P ratio of 2.5, CDP completely retarded pDNA migration down the gel, indicating that the polymer fully complexed with pDNA at this higher N/P ratio. These data demonstrate that a lower amount of PEI is required to fully bind to negatively charged pDNA, a phenomenon that is likely a function of the larger amount of protonable amine groups available for DNA binding on PEI as compared to CDP. The reduction of positive amine groups on CDP is due to the grafting of CD molecules on the PEI chain. This may also be due to the shielding effect of CD which prevents binding of the pDNA to the CDP polymer.

ACS Paragon Plus Environment

16

Page 17 of 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 1. pDNA retardation assay was performed using gel electrophoresis. (A) PEI25k, (B) CDP, (C) SaNP-RGD were complexed with 1µg of DNA from N/P ratio 0.5 to 10. Samples were run on a 1% TAE agarose gel with GelRed. pDNA migration was reduced with increasing N/P ratios; D) Binding capacity of polymers to plasmids by Picogreen assay; E) Heparin binding assay of polymers to plasmids. *p