Stealth nanorods via the aqueous living crystallisation-driven self-assembly of poly(2-oxazoline)s

The morphology of nanomaterials critically influences their biological interactions. However, there is currently a lack of robust methods for preparing non-spherical particles from biocompatible materials. Here, we combine ‘living’ crystallisation-driven self-assembly (CDSA), a seeded growth method that enables the preparation of rod-like polymer nanoparticles, with poly(2-oxazoline)s (POx), a polymer class that exhibits ‘stealth’ behaviour and excellent biocompatibility. For the first time, the ‘living’ CDSA process was carried out in pure water, resulting in POx nanorods with lengths ranging from ∼60 to 635 nm. In vitro and in vivo study revealed low immune cell association and encouraging blood circulation times, but little difference in the behaviour of POx nanorods of different length. The stealth behaviour observed highlights the promising potential of POx nanorods as a next generation stealth drug delivery platform.


Materials and Methods
All moisture-sensitive reactions were carried out in oven (100 °C) dried glassware under a positive nitrogen pressure using standard Schlenk techniques. All compounds were purchased from Sigma-Aldrich and used without further purification unless stated otherwise. Methyl ptoluenesulfonate was purified by fractional distillation under reduced pressure. Anhydrous acetonitrile was stored over activated 3 Å molecular sieves prior to use. 2-Methyl-2-oxazoline was purified by distillation over BaO. 2-Isopropyl-2-oxazoline (iPrOx) was synthesised following the procedure outlined in ref.
Nuclear magnetic resonance (NMR) spectra were recorded on a Bruker Avance III 400 MHz spectrometer using an external lock and referenced internally to a resonance from residual protonated solvent. Chemical shifts (δ) are reported in parts per million (ppm). NMR solvents (CDCl3 and D2O) were purchased from Cambridge Isotope Laboratories, Inc and used as received.
Size-exclusion chromatography (SEC) analyses of polymer samples were performed using a Shimadzu modular system comprising a DGU-20A3R degasser unit, an SIL-20A HT autosampler, a 10.0 µm bead-size guard column (50 × 7.8 mm) followed by three KF-805L columns (300 × 8 mm, bead size: 10 µm, pore size maximum: 5000 Å), a SPD-20A UV/Vis detector, and an RID-10A differential refractive-index detector. The temperature of columns was maintained at 40 °C using a CTO-20A oven. The eluent was 0.03 wt% LiBr in N,Ndimethylacetamide (DMAc, CHROMASOLV Plus for HPLC) and the flow rate kept at 1.0 mL/min using a LC-20AD pump. A molecular weight calibration curve was produced using commercial narrow molecular weight distribution polystyrene standards with molecular weights ranging from 500 to 2 × 10 6 Da. Polymer solutions were analysed at 2 mg/mL after filtration through 0.45 µm PTFE membrane.
Dynamic light scattering (DLS) and zetapotential measurements were performed using a Malvern Zetasizer Nano ZS. Cloud point temperature was determined by DLS analysis of filtered 10 mg/mL polymer solutions in Milli-Q water in a disposable sizing cuvette (Malvern,Zen0040). Polymer solutions were heated in 2 °C increments from 30 to 70 °C, equilibrating for 120 s and recording 3 measure of at least 10 runs at each temperature. Zetapotential measurements were carried out at 25 °C on unfiltered 1 mg/mL nanorod suspensions in Milli-Q water in a disposable folded capillary cuvette (Malvern, DTS1070). Values shown represent the mean and standard deviation of three measurements comprised of a minimum of ten runs each.
Transmission electron microscopy (TEM) was performed using either a Tecnai F20 or a Thermo Scientific Talos L120C microscope, both instruments were operated at 200 kV. Samples for TEM analysis were prepared by cooling polymer suspensions to room temperature followed by the addition of 5 µL of colloidal solution to a negative glow discharge treated carbon coated copper grid (ProSciTech). After approximately 60 s excess sample solution was removed by contact with filter paper. Samples to be stained with UranyLess (Electron Microscopy Sciences) were then placed onto a single drop of UranyLess solution for 50 s. Excess stain solution was removed by contact with filter paper and the grid allowed to air dry prior to imaging.
Analysis of TEM images was carried out using ImageJ software (National Institute of Health). Number average (Ln) and weight average particle length (Lw) were calculated using the equations below.
Atomic force microscopy (AFM) was performed using a Bruker dimension icon SPM (USA). A triangular silicon nitride cantilever (ScanAsyst Air probe, spring constant 0.4 N/m, Bruker) with a tip radius of 2-10 nm was used for the topographic analysis. Samples were scanned with the PeakForce Tapping mode with a scan rate of 0.70 Hz. While scanning, the peak force and feedback gain were dynamically optimized by the ScanAsyst auto control. NanoScope Analysis 2.0 software was used to process the raw data and height analyses. Samples were prepared by adding 50 µL of nanoparticle solution to freshly cleaved mica. Excess solvent was removed by blotting with filter paper and the substrate dried under N2 flow.
Differential scanning calorimetry (DSC) was performed using a PerkinElmer DSC8500 coupled to a PerkinElmer Intracooler 2 cooling system at a scanning rate of 10 °C·min -1 . Samples were sealed in 50 µL aluminium pans featuring a pinhole lid.
X-ray diffraction was carried out using a Bruker D8 Advance with Cu K-alpha radiation (λ = 1.5406 Å) and Lynxeye XE position sensitive detector running in 1D mode. Scatter was collected using Bragg-Brentano reflection geometry, a viable divergence slit was used with a constant sample illumination width at 17 mm. Diffraction peak positions and d-spacings were determined using Bruker DIFFRAC.EVA software.
Sonication. A Sonics VC 505 ultrasonic processor was used fitted with a 1/8 " tapered probe.
PMeOx48-b-PiPrOx47. Methyl p-toluenesulfonate (35.6 µL, 0.24 mmol, 1 eq), anhydrous acetonitrile (4.9 mL) and 2-methyl-2-oxazoline (1.0 mL, 12 mmol, 50 eq) were transferred under nitrogen to an oven-dried greaseless Schlenk flask equipped with magnetic stirrer. The flask was then sealed and the reaction mixture heated with stirring to 80 °C for 7.5 h. A 0.3 mL aliquot was removed from the reaction mixture for molecular weight analysis of the PMeOx block, then 2-isopropyl-2-oxazoline (1.3 mL, 11 mmol, 50 eq) added, the flask sealed and the reaction mixture stirred at 80 °C for a further 16 h. To quench the polymerisation acetic acid (64 µL, 1.1 mmol, 5 eq) and triethylamine (156 µL, 1.1 mmol, 5 eq) were added and the reaction mixture stirred for 5 h at 60 °C. The resulting block copolymer was then precipitated directly into diethyl ether, followed by two subsequent precipitations from CH2Cl2 into diethyl ether. The resulting colourless solid was dialysed against water for 3 d using 12-14 kDa MWCO dialysis tubing (Cellu-Sep T3) and isolated by freeze-drying. Yield = 1.3 g (59%) PiPrOx40-NH2. In a 50 mL round-bottom flask fitted with magnetic stirrer PiPrOx40-phthalimide (1.0 g, 0.21 mmol, 1 eq) was dissolved in ethanol (25 mL) followed by the addition of hydrazine monohydrate (162 µL, 3.3 mmol, 16 eq). The reaction mixture was then heated to reflux for 22 h, cooled to room temperature and filtered. Subsequently, 5 M HCl (aq) was added to adjust the pH value of the solution to 2 and the resulting precipitate removed by filtration. Volatiles were then removed under vacuum and the residue dissolved in water. Saturated NaOH (aq) was added to increase the pH value of the solution to 10. This solution was then extracted with CH2Cl2 and washed three times with water. The organic layer was dried over MgSO4 and concentrated under vacuum. PiPrOx40-Cy5. In an oven-dried greaseless Schlenk flask, PiPrOx40-NH2 (25 mg, 0.006 mmol, 1 eq), Cy5-NHS (8 mg, 0.013 mmol, 2.2 eq) and triethylamine (100 µL, 0.717 mmol, 120 eq) were dissolved in anhydrous N,N-dimethylformamide (1.5 mL). The reaction mixture was stirred under nitrogen in absence of light at room temperature for 14 h. Volatiles were removed under vacuum and the residue was redissolved in 5.0 mL of CHCl3. Excess Cy5 was removed by preparative size exclusion column chromatography (Bio-BeadsTM S-X1, 600 -14000 MW exclusion range) and PiPrOx40-Cy5 isolated as a blue solid by freeze-drying. Yield = 24 mg (96% PMeOx48-b-PiPrOx39-NH2. In a 50 mL round-bottom flask fitted with magnetic stirrer PMeOx48-b-PiPrOx39-phthalimide (0.7 g, 0.08 mmol, 1 eq) was dissolved in ethanol (25 mL) followed by the addition of hydrazine monohydrate (47 µL, 0.86 mmol, 11 eq). The reaction mixture was then heated to reflux for 22 h, cooled to room temperature and filtered. Subsequently, 5 M HCl (aq) was added to adjust the pH value of the solution to 2 and the resulting precipitate removed by filtration. Volatiles were then removed under vacuum and the residue dissolved in water. Saturated NaOH (aq) was added to increase the pH value of the solution to 10. This solution was transferred to 12-14 kDa MWCO dialysis tubing (Cellu-Sep T3), dialysed against water for 3 d and isolated by freeze-drying.
Nanoparticle sonication. Seed micelles were prepared by ultrasonication of micelles prepared by spontaneous nucleation (annealed at 70 °C for 1 week and used immediately). The ultrasonic processor was operated at 20% amplitude using an on:off pulse sequence of 2 s:2 s and samples cooled with an ice bath. A total sonication time of 40 minutes was used to prepare seeds from each polymer. Summary of nanoparticle length analysis is shown in Table S1 and histograms shown in Figure S10.
Seeded growth PMeOx48-b-PiPrOx47. Samples were prepared from an aqueous colloidal solution of PMeOx48-b-PiPrOx47 seeds (1 mg/mL) and an aqueous solution of PMeOx48-b-PiPrOx47 (1 mg/mL). Solutions were prepared as described in σ corresponds to standard deviation of particle length. Table S2 and heated to 60 °C for 16 h before cooling to 23 °C. Summary of nanoparticle length analysis is shown in σ corresponds to standard deviation of particle length. Table S2 and histograms shown in Figure S13.
Nanorod kinetic stability. To study the kinetic stability of POx nanorods, a 1 mg/mL D2O solution of nanorods (Ln = 275 nm, Lw/Ln = 1.08) was prepared by annealing a mixture of 1.6 mL of PMeOx48-b-PiPrOx47 seeds (1 mg/ml in D2O) and 6.4 mL of PMeOx48-b-PiPrOx47 unimer (1 mg/mL in D2O) at 60 °C for 16 h. This solution was split into two aliquots and which were stored at either 23 or 37 °C. Nanorod stability was monitored by 1 H NMR spectroscopy and TEM.
Cy5-labelled PMeOx48-b-PiPrOx47 nanorods. Samples were prepared from an aqueous colloidal solution of PMeOx48-b-PiPrOx47 seeds (1 mg/mL) and aqueous solutions of PMeOx48-b-PiPrOx47 (1 mg/mL) and PiPrOx40-Cy5 (1 mg/mL). Solutions were prepared as described in Table S3 and S4 and heated to 60 °C for 16 h before cooling to 23 °C. Summary of nanoparticle length analysis is shown in σ corresponds to standard deviation of particle length. Table S2 and histograms shown in Figure S25.
Cy5 labelled nanorods used in the murine ex vivo biodistribution assays were concentrated to 10 mg/mL by reducing the volume of the nanorod solutions ten-fold using a centrifugal concentrator with a MWCO of 30 kDa. TEM analysis of these samples was carried out after concentration by centrifugal filtration, with samples diluted from 10 mg/mL to 1 mg/mL prior to drop casting.

Biology
Human blood immune cell association assay. The method described here is based closely on those described by Caruso, Kent and coworkers in suplimentary reference [2]. Human whole blood was taken via venupuncture from a single donor and stored in sodium heparin. Plasma-stripped blood was obtained by centrifugal washing of whole blood 4x with phosphate buffered saline (PBS). 90-100 µl of whole and plasma-stripped blood was added to FACS tubes and allowed to equilibrate at 37˚C for 10 min prior to addition of Cy5-labelled POx nanorods or free polymer (stock concentration: 1 mg/ml in water) to a final concentration of either 10 µg/ml. Samples were incubated in the dark at 37˚C (95% humidity, 5% CO2) for 1 h. All samples were then transferred to ice for further processing: first, red blood cell lysis was carried out utilising BD PharmLyse, with samples then incubated with appropriate antibodies for 30 min. Immune cell types identified by this assay include: CD66b+ neutrophils, CD14+ monocytes, CD19+ B cells, CD3+ T cells, CD56+ natural killer (NK) cells, and Lin1-HLA-DR+ dendritic cells (DCs); all antibodies were purchased from BD Biosciences. Samples were then washed in FACS wash buffer (2 mM EDTA, 0.5% bovine serum albumin in PBS) before cells were fixed with 10% formaldehyde. Characterisation was carried out on a BD LSRFortessa flow cytometer at medium speed, 5 decades and the FSC threshold set at 5,000. Data were analysed in FlowJo, graphed in GraphPad Prism and represent n = 2 independent experiments and n = 3 technical repeats. Raw cell counts are shown in the Supporting Schemes and Tables section.
Ex Vivo Biodistribution. Mice 8 weeks of age were intravenously injected with 100 µl of 10 mg/ml nanorod solution and 24 h post injection euthanized by CO2. The heart, brain, stomach, liver, spleen, kidney, lung, muscle and blood were collected and wet-weighed. The organs were placed on a petri dish and imaged using an IVIS Lumina II imaging system (Caliper Life Sciences). All imaging parameters were kept constant and were as follows: Bin, (M)4; FOV: 12.5; f2; 2 s. The total radiant efficiency (p/s)/(µW/cm 2 ) of the organs as well as the standards (injected samples) were measured and background corrected. Based on the weight of the individual organs and total radiant efficiency, the data were then converted to %ID tissue g -1 . For statistical analysis with a group size of n=3 per nanorod sample, data were compared by two-way ANOVA with Tukey multiple comparisons post hoc test. Statistical significance was accepted at p<0.05. Ethical approval for animal experiments were granted from the Alfred Research Alliance Animal Ethics Committee under E/1625/2016/M.

Control Short
Medium Long  Table S6. Raw cell counts for flow cytometry analysis of immune cell assocation in plasma-stripped blood. Figure            A Figure S15. TEM images of PMeOx24-b-PiPrOx50 nanorods. A) Fibre fragments prepared by ultrasonication. B) Nanorods prepared by seeded growth at 60 °C (0.5 mg/mL) with a unimer to seed ratio of 2. C) Nanorods prepared by seeded growth at 60 °C (0.5 mg/mL) with a unimer to seed ratio of 4. Scale bars equal to 500 nm.  Figure S18. Zeta potential plots for (A) PMeOx48-b-PiPrOx47 nanorods prepared by seeded growth (munimer/mseed = 4), and (B) Cy5-labelled PMeOx48-b-PiPrOx47 nanorods prepared by seeded growth (munimer/mseed = 4, 1 wt% PiPrOx40-Cy5). Zeta potential measured at 1 mg/mL in Milli-Q water, plots represent the mean of 3 individual measurements. A B Figure S20. Normalised Cy5 mean fluorescence intensity (MFI) associated with different immune cell types in 'washed' (plasma-stripped) blood after incubation with POx-Cy5 nanorods for 1 h. Insets: comparative MFI of POx nanorods interacting with B cells (blue) and monocytes (grey) in the presence (whole) and absence (washed) of plasma. Error is standard error of mean (n = 2); * p < 0.05 by a twoway ANOVA with Sidak's correction for multiple comparisons. Figure S21. Immune cell association of POx-Cy5 nanorods after incubation for 1 h as a percentage of total cell population, in human (A) whole and (B) washed blood. Error is standard error of mean (n = 2); **** p < 0.0001 by a two-way ANOVA with Sidak's correction for multiple comparisons.