Optimization and Stability of Cell–Polymer Hybrids Obtained by “Clicking” Synthetic Polymers to Metabolically Labeled Cell Surface Glycans

Re-engineering of mammalian cell surfaces with polymers enables the introduction of functionality including imaging agents, drug cargoes or antibodies for cell-based therapies, without resorting to genetic techniques. Glycan metabolic labeling has been reported as a tool for engineering cell surface glycans with synthetic polymers through the installation of biorthogonal handles, such as azides. Quantitative assessment of this approach and the robustness of the engineered coatings has yet to be explored. Here, we graft poly(hydroxyethyl acrylamide) onto azido-labeled cell surface glycans using strain-promoted azide–alkyne “click” cycloaddition and, using a combination of flow cytometry and confocal microscopy, evaluate the various parameters controlling the outcome of this “grafting to” process. In all cases, homogeneous cell coatings were formed with >95% of the treated cells being covalently modified, superior to nonspecific “grafting to” approaches. Controllable grafting densities could be achieved through modulation of polymer chain length and/or concentration, with longer polymers having lower densities. Cell surface bound polymers were retained for at least 72 h, persisting through several mitotic divisions during this period. Furthermore, we postulate that glycan/membrane recycling is slowed by the steric bulk of the polymers, demonstrating robustness and stability even during normal biological processes. This cytocompatible, versatile and simple approach shows potential for re-engineering of cell surfaces with new functionality for future use in cell tracking or cell-based therapies.


S4
of incidence. Analysis of dried crushed samples were completed following purging the setup with nitrogen for 30 minutes. Scans (100) were obtained between 4000 -400 cm -1 with a resolution of 4 cm -1 . Gain, aperture, scan speed and filter were all set to auto. Standard source and chamber were used along with a triglycine sulfate (TGS) detector.
Fluorimetry. Fluorescence measurements were obtained using a Jasco FP-6500 fluorimeter equipped with a DC-powered 150 W Xenon lamp and holographic grating with 1800 grooves mm -1 modified Rowland mount. Excitation and emission bandwidths were set to 3 nm with a response of 1 sec and sensitivity set to medium. The scanning range was set from 450 to 550 nm, with an excitation wavelength selected at 494 nm, all with an accuracy of ±1.5 nm and reproducibility of ±0.3 nm. A scanning speed of 100 nm.min -1 was chosen and data pitch of 1 nm.

UV-Vis
Spectroscopy. An Agilent Technologies Cary 60 Variable Temperature UV-Vis spectrometer was used to record absorbance measurements between 300 and 800 nm at a scan rate of 600 nm.min -1 and 1 nm data interval. Dual beam mode was used and the system was operated using Agilent software.
Confocal Microscopy. Confocal imaging was completed using a Zeiss LSM 880 inverted microscope with 100x, 63x, 40x and 20x oil immersion objective lenses, equipped with three photomultiplier detectors (GaAsP, multialkali and BiG.2) and multichannel spectral imaging with an ultra-sensitive GASP detector. The UV and VIS Laser Modules allowed selection of six lasers with wavelengths of 633, 594, 561, 543, 514, 488, 458, 405 and 355 nm. Zeiss ZEN (blue edition) 2.3 lite was utilized for image collection and processing. All other imaging was completed using an Olympus CX41 microscope equipped with a UIS-2 20x/0.45/∞/0−2/FN22 lens (Olympus Ltd., Southend on sea, U.K.) and a Canon EOS 500D SLR digital camera. Statistical Analysis. Data were analyzed with a one-way analysis of variance (ANOVA) on ranks followed by a comparison of experimental groups with the appropriate control group (Tukey's post hoc test). R (R Foundation for Statistical Computing, Vienna, Austria) was used for the analyses.

1.3
General procedure for the synthesis of telechelic polymers [1][2][3] Synthesis of 2-(dodecylthiocarbonothioylthio)-2-methylpropionic acid (DMP).    Table S1 to obtain 5 degrees of polymerization (DP). Mesitylene (150 µL) was used as an internal reference and an aliquot was taken in CDCl3 for NMR analysis. The reaction mixture was stirred under N2 for 30 minutes at RT and a further 90 minutes at 70 o C.
An aliquot of the post-reaction mixture was taken for NMR analysis in MeOD, allowing percentage conversion calculations. The polymer was reprecipitated into diethyl ether from methanol three times, yielding a yellow polymer product. The resulting product was dried under vacuum and an aliquot was taken for NMR analysis in MeOD. NMR percentage conversion and SEC results are presented in Table S1.  PFP-p(HEA)n (0.20 g, 1 Eq), and dibenzocyclooctyne-amine (DBCO-NH2; 2 Eq) were stirred in methanol (3 mL) for 16 h. Subsequent addition of propyl amine (1.5 eq) for 2 h was used to ensure complete reduction of the thiocarbonate moiety to a thiol group. The polymer was reprecipitated into diethyl ether from methanol three times, yielding a white polymer product.
The resulting product was dried under vacuum and DMF SEC analysis was completed, Table   S2. An aliquot was also taken for NMR analysis in MeOD. Table S2. SEC results of DBCO-pHEAn.  Fluorescein conjugation was confirmed using fluorimetry following exhaustive dialysis. DMF SEC analysis was completed with the UV-Vis detector set at 494 nm to demonstrate size separation and absorbance overlap, Table S3 and Fig. S12. Percentage dye functionalisation was determined using UV-Vis spectroscopy, Table S3 and Fig. S14 and S15. Deviations from expected Polymer: dye ratio (i.e. 1:1) is expected to be because of error in determining exact molecular weight using SEC. Accutase® solution and cells were strained using nylon cell strainer (40 µm) to ensure single cell analysis. BD Influx™ cell sorter (BD Biosciences) operating parameters were as described above. Control cells, untreated with Ac4ManNAz, were also incubated with DBCO-pHEAn-Fl S17 polymers (5 and 10 mg.mL -1 , 2.5 h) and analyzed using flow cytometry and microscopy to determine the extent of non-specific binding.

DBCO-pHEAn-Fl cell surface degrafting
Flow Cytometry. A549 cells were plated in a 12 well plate at a density of 2 x 10 5 cells.mL -1 with media supplemented with Ac4ManNAz (40 µM) and incubated in a humidified atmosphere of 95% air and 5% CO2 at 37 °C for 96 h. Following this, cells were incubated with DBCO-pHEAn-Fl polymers (5 or 10 mg.mL -1 , 2.5 h) in fresh cell media, washed three times with DPBS and imaged using an Olympus CX41 microscope 0 -72 h post conjugation. Flow sample preparation and operating parameters were completed as previously described above.
Control cells, untreated with Ac4ManNAz, were also incubated with DBCO-pHEAn-Fl polymers (5 and 10 mg.mL -1 , 2.5 h) and analyzed using flow cytometry and microscopy at 0 h and 8 h post conjugation to determine the extent of degrafting of non-specific binding.  Figure S7. Furthermore, the disappearance of a broad carboxylic acid O-H stretch (3253-2375 cm -1 (w, broad)) and a shift in C=O stretch confirmed the conversion of a carboxylic acid to an ester (1778 cm -1 , m-s), Figure S6. Finally, a C-F stretch (1518 cm -1 , s) was noted in the IR of PFP-DMP.

DBCO-p(HEA)n-Fl
Confirmation of fluorescein conjugation was made evident by: (1) overlap between DMF SEC results using both an RI and photodiode array detector indicating that both size and absorbance readings correlate (i.e. polymer and dye conjugation was successful) and (2) the increase in fluorescence intensity compared to the negligible fluorescence intensity found in p(HEA)n-DBCO polymers, Figure S14. Polymer: dye ratio was calculated using UV-Vis spectroscopy, Fig. S15 and S16. Deviation from expected 1:1 ratio is likely due to deviations in determining the exact molecular weight of polymers using SEC.     Figure S16.      Figure S28. Imaging of (live) A549 cells untreated with Ac4ManNAz but treated with DBCOpHEAnFl polymers varying in chain lengths (2.5 h, 10 mg.mL -1 ) to determine the extent of non-specific binding. Images were taken immediately before flow cytometry. Scale bar = S46 Figure S29. Imaging of (live) A549 cells immediately following treatment with Ac4ManNAz