Molecular Recognition‐Mediated Transformation of Single‐Chain Polymer Nanoparticles into Crosslinked Polymer Films

Abstract We describe single‐chain polymer nanoparticles (SCNPs) possessing intramolecular dynamic covalent crosslinks that can transform into polymer films through a molecular recognition‐mediated crosslinking process. The SCNPs utilise molecular recognition with surface‐immobilised proteins to concentrate upon a substrate, bringing the SCNPs into close spatial proximity with one another and allowing their dynamic covalent crosslinkers to undergo intra‐ to interpolymer chain crosslinking leading to the formation of polymeric film. SCNPs must possess both the capacity for specific molecular recognition and a dynamic nature to their intramolecular crosslinkers to form polymer films, and an investigation of the initial phase of film formation indicates it proceeds from features which form upon the surface then grow predominantly in the xy directions. This approach to polymer film formation presents a potential method to “wrap” surfaces displaying molecular recognition motifs—which could potentially include viral, cellular and bacterial surfaces or artificial surfaces displaying multivalent recognition motifs—within a layer of polymer film.

Procedure for preparation of SCPNs P1 (15.4 mg, 1.37 µmol, 1 eq.) and GAL or MAN (7.8 mg, 25 µmol, 18 eq.) were combined in MeOH-d4 (0.25 mL) and dmso-d6 (0.25 mL) before addition of 100 mM NH4OAc, pH 4.5, D2O (1.0 mL). The solution was left to stir at room temperature until 1 H NMR spectroscopic analysis confirmed complete functionalisation of the polymer, determined by total disappearance of the signal corresponding to the aldehyde proton. A 100 µL aliquot was removed and dried under high vacuum for GPC analysis, before addition of succinic dihydrazide (100 µL from 2.0 mg mL -1 stock solution in 100 mM NH4OAc, pH 4.5, D2O, 1.37 µmol, 1 eq.). The reaction mixture was allowed to stir at room temperature overnight, before a further 100 µL aliquot was removed and dried under high vacuum for GPC analysis, which confirmed intrachain crosslinking with increase in retention time indicative of a reduction in hydrodynamic volume, consistent with SCPN formation.

Increasing crosslinking density during SCPN formation
In order to investigate the relationship between the amount of succinic dihydrazide used to crosslink polymer chains and the resultant contraction in hydrodynamic radii during SCPN formation, P1 was functionalised with MAN as described above, yielding glycopolymer P1-MAN, and divided into aliquots to which various molar equivalents of succinic dihydrazide were added. These solutions were dried under high vacuum and subjected to GPC analysis (Fig. S2). Increasing the amount of succinic dihydrazide added to P1-MAN was shown to increase the retention time of the resultant SCPN for crosslinking densities between 1-3 eq., suggesting that the hydrodynamic volume of the SCPNs has been reduced. Crosslinking with greater molar equivalents of succinic dihydrazide, however, led to the formation of inter-chain crosslinked species, as evidenced by the appearance of a shoulder at lower retention time in their GPC chromatograms.

SCPN stability in solution
A solution of SCPN1-MAN, prepared as described earlier, was left to stir at room temperature. 100 µL aliquots were removed at timed intervals over a 24 h period, dried under vacuum and subjected to GPC analysis (Fig. S3). GPC analysis demonstrated that there was no significant aggregation to form larger macromolecular species over this time period.

Expression of E. coli heat labile toxin (LTB)
Cells from a glycerol stock of Vibrio sp60 harbouring plasmid pMMB68 (kindly provided by Prof. Tim Hirst) [2] were used to inoculate growth medium (100 mL, 25 g/L LB mix, 15 g/L NaCl, ampicillin 100 µg/mL). The culture was grown overnight at 30 °C with shaking at 200 rpm, then used to inoculate fresh growth medium (6 x 1 L, 25 g/L LB mix, 15 g/L NaCl, ampicillin 100 µg/mL). These cultures were incubated at 30 °C with shaking at 200 rpm until A600 reached 0.6 before the protein expression was induced by addition of isopropyl β-D-1-thiogalactopyranoside to a concentration of 0.5 mM.
Cultures were incubated (30 °C, 200 rpm) for a further 24 h, then cells were removed by centrifugation (6500 rpm, 15 min). The combined supernatant was treated with ammonium sulphate (550 g/L) and left to stir at 5 °C overnight. Crude protein was isolated by centrifugation (10,000 rpm, 25 min) and redissolved in in 100 mM NaH2PO4, pH 7.0, 500 mM NaCl (60 mL). Insoluble material was removed by centrifugation (10,000 rpm, 10 min) before the solution was passed through a 0.22 µm filter then loaded onto a lactose-sepharose 6B column and eluted with 300 mM lactose, 100 mM NaH2PO4, pH 7.0, 500 mM NaCl. LTB was dialysed against PBS, pH 7.4, freeze-dried and stored at -80 °C

Preparation of biotinyl-LTB
LTB (900 L, 647 M protomer concentration) in PBS was mixed with Pierce EZ-link™ Sulfo-NHS-SS-Biotin (3.5 mg dissolved in 500 L water). After 2 hours at room temperature, the solution was diluted (PBS, 15 mL) and concentrated by centrifugal ultrafiltration (10kDa MW/CO) twice (final volume 2 mL) before freeze-drying in 5 aliquots and storing at -80 °C until required.

Procedure for immobilisation of lectins onto polystyrene surfaces
The bases of Pierce TM streptavidin coated high capacity 96 well plates were extracted with a cork borer and washed with H2O before immersion in a solution of the appropriate biotinylated lectin (1.3 mg mL -1 , H2O) for 1 h at 5 °C. The discs were then washed (H2O) and dried carefully under laminar flow before imaging, or immersion in solutions of SCPNs to prepare polymer films.

Procedure for immobilisation of lectins onto N-hydroxysuccinimide functionalised Si wafer [3]
NHS-functionalised Si wafers were immersed in a solution of Con A or LTB (0.02 mg mL -1 ) in phosphate buffered saline (PBS) at pH 7.4 for 5 h. Substrates were then removed and sonicated in fresh PBS (2 x 15 min), then rinsed with H2O. Surfaces were then immersed in 0.1 M aqueous ethanolamine solution for 1 h, rinsed with H2O and dried under a gentle stream of N2.

Procedure for plasma lithography of films
Electron microscopy grids (1000 mesh x 25 µm pitch, copper, Sigma Aldrich) were affixed to surfaces upon which polymer films had been formed on Si wafer as described above using copper adhesive tape (RS components). The surfaces were treated with oxygen plasma (100 W, 12 cm 3 min -1 , 5 min) and imaged using Tapping Mode™ AFM.

Film reversal experiments
Polymer films of SCPN1-MAN on Con A functionalised polystyrene were prepared as earlier described and subjected to AFM analysis. Substrates were then incubated at 5 °C in (a) a 50 % v/v solution of NH2OH(aq), or (b) a saturated solution of methyl α-mannoside in 100 mM NH4OAc, pH 4.5. Samples were removed after 18 h and the surfaces were examined by optical microscopy and AFM. In the case of samples incubated in methyl α-mannoside solution, polymer film was found to remain on the surface, so the samples were returned to the solution and re-examined after 3 d, when surfaces free from polymer film were observed (Fig. S5).

Fig. S5
Optical microscope images (40 X magnification) of (a) polymer film produced by exposure of SCPN-MAN to a Con A functionalised surface; (b) the same surface after exposure to NH2OH (c) polymer film produced by exposure of SCPN-MAN to a Con A functionalised surface; (d) the same surface after exposure to methyl αmannoside.

Film damage repair experiments
A polymer film of SCPN1-MAN on Con A functionalised Si wafer was generated as previously described, and scratched under PBS (pH 7.4) in a fluid cell using contact mode AFM by applying a voltage of 6.00 V (1.8 µN) in a 16 x 1 µm area, using an Bruker MPP-21100-10 tip where k ≈ 3 N/m and r ≈ 8 nm and the tip has no coating. The area were then imaged (also under PBS in contact mode) using a DNP -s10 tip where k ≈ 0.24 N/m and r ≈ 20 nm revealing a scratch across the surface (Fig.  S6a,b)). The buffer was removed from the fluid cell and replaced with 100 mM NH4OAc, pH 4.5. The sample was left in place for 24 h, then the area was re-examined by AFM (Fig. S6c,d), revealing a more uniform depth profile, suggesting that the polymer film had rearranged to 'heal' itself after damage.

Fig. S6
Contact mode AFM images of a polymer film formed using SCPN1-MAN and Con A functionalised silicon wafer after being subjected to a scratch (a) and the corresponding height profile (b). (c) The same region after 18 h incubation in 100 mM NH4OAc, pH 4.5 and the corresponding height profile, (d).