Aqueous photo-RAFT polymerization under ambient conditions: synthesis of protein–polymer hybrids in open air

A photoinduced reversible addition-fragmentation chain-transfer (photo-RAFT) polymerization technique in the presence of sodium pyruvate (SP) and pyruvic acid derivatives was developed. Depending on the wavelength of light used, SP acted as a biocompatible photoinitiator or promoter for polymerization, allowing rapid open-to-air polymerization in aqueous media. Under UV irradiation (370 nm), SP decomposes to generate CO2 and radicals, initiating polymerization. Under blue (450 nm) or green (525 nm) irradiation, SP enhances the polymerization rate via interaction with the excited state RAFT agent. This method enabled the polymerization of a range of hydrophilic monomers in reaction volumes up to 250 mL, eliminating the need to remove radical inhibitors from the monomers. In addition, photo-RAFT polymerization using SP allowed for the facile synthesis of protein–polymer hybrids in short reaction times (<1 h), low organic content (≤16%), and without rigorous deoxygenation and the use of transition metal photocatalysts. Enzymatic studies of a model protein (chymotrypsin) showed that despite a significant loss of protein activity after conjugation with RAFT chain transfer agents, the grafting polymers from proteins resulted in a 3–4-fold recovery of protein activity.


Table of contents
.
UV-Vis spectra were recorded using an Agilent 8453 spectrophotometer.
Femtosecond transient absorption measurements were conducted using the Solstice Ti:sapphire regenerative amplifier from Spectra Physics and an optical detection system provided by Ultrafast Systems (Helios).The source for the pump and probe pulses was the fundamental emission at 800 nm.The fundamental output was split into two beams: a pump (95%) and a probe (5%).The pump beam was directed through the TOPAS-Prime automated optical parametric amplifier from the Spectra Physics to obtain the desired excitation wavelength in the range 290-2600 nm.The probe beam was directed to the Helios: a CCD-based pump-probe TA spectrometer from Ultrafast Systems LLC with an optical delay line allowing delay between the pump and probe up to 3.2 ns.
For the detection of the transients, a white light continuum was used, which was generated from 5% of the fundamental beam by passing it through a sapphire or calcium fluoride crystal.For transient UV-vis measurements a quartz cell with 2 mm optical path of solution was used with the absorbance of about 0.2 at the excitation wavelength, the sample solution was stirred by a Tefloncoated bar.All experiments were performed at room temperature.Global analysis of the transient absorption data was made using the Surface Explorer software (Ultrafast Systems).
After adjusting the DMSO volume to (500 µL, 10% v/v), the volumetric flask was topped up with water.The mixture was vortexed for 3 min, transferred to an open-to-air glass vial (1 mL), and irradiated with different lights to initiate polymerization.Polymerizations were stopped by turning the light off and residues were mixed in characterization solvents (DMF for GPC and Similar procedures were performed for carrying out polymerization of other pyruvic acid derivatives (PADs) except for the use of EP (0.15 g, 1.3 mmol), PA (0.11 g, 1.3 mmol), OA (0.11 g, 1.3 mmol) and SO (0.085 g, 0.63 mmol).
After adjusting the DMSO volume (500 µL, 10% v/v), the volumetric flask (5 mL) was topped up with water.The mixture was vortexed for 3 minutes, transferred to a 1 mL open-cap glass vial, and was irradiated with different lights (UV, green and blue) to initiate the polymerization.
Polymerization was stopped by turning the light off and dilution of samples in characterization solvent (DMF for GPC or D 2 O for NMR).

General procedure for photo-RAFT polymerization with SP and without DMSO
OEOMA 500 (750 mg, 1.5 mmol) and SP were mixed in a volumetric flask (5.0 mL), followed by the addition of CPADB (4.2 mg, 0.015 mol).The mixture was heated up to 40 °C for 10 min to increase the solubility of CPADB in the mixture.Next, the volumetric flask (5 mL) was topped up with water.The mixture was vortexed for 10 min, transferred to an open-cap glass vial (1 mL), and irradiated with light to initiate polymerization.

Kinetic analysis of photo-RAFT polymerization
The kinetics of the RAFT polymerizations were monitored under irradiation with three different light sources (UV, blue and green).In the typical reaction under UV light, the mixture was prepared following the aforementioned protocol, and the mixture without SP served as a reference.For the reactions under blue and green lights, the reactant ratios were adjusted as follows: OEOMA 500 (2.0 g, 800 mM), SP (0.14 g, 255 mM) and CPADB (8.5 mg, 4.0 mM) dissolved in water.The total reaction volume, and the ratio of water and DMSO (9/1 v/v) were kept identical.When working under UV light, the aliquots (100 µL) were withdrawn in 10minute intervals, while a withdrawal period of 60 minutes was selected when carrying out the RAFT polymerizations under blue and green lights.The purified products were analyzed by 1 H NMR and GPC techniques.

Discussion on the role of SP in visible light induced PI-RAFT polymerization
The observed increase in rate of polymerization, suppression of induction period and enhanced oxygen tolerance in the presence of SP upon visible light irradiation suggest the interaction of SP with ground or excited states of CPADB.This was investigated by several other experiments.
First, the polymerization of OEOMA 500 with blue light irradiation (450 nm) was carried out using different SP concentrations.As shown in Figure S4, the conversions of OEOMA 500 are proportional to the concentration of SP used for the polymerization and increased with higher SP concentration.CPADB exhibited interesting wavelength-dependent photochemistry (Figure S5) and therefore it was of interest to study properties of CPADB excited to S 2 or S 1 .Qualitative analysis of the different photochemistry depending on the excitation wavelength was performed by steady-state irradiation experiments using 370 nm and 505 nm LED diode.Steady-state irradiation of CPADB with 370 nm LED diode led to almost complete disappearance of the CPADB absorbance within 3 minutes of irradiation (Figure S5A).Surprisingly, irradiation of the CPADB in the visible range with 505 nm LED diode did not cause any meaningful changes in the UV-VIS spectra even after 4 h of irradiation (Figure S5B).This conclusion is in an agreement with previous work by Thum et al.where it was shown that the excitation of dithioesters into their S 1 band (i.e., visible light) results in negligible production of radicals due to the slightly higher bond dissociation energies of C-S bond compared to the lowest singlet and triplet excited state energies of dithioesters. 30 Ultrafast transient absorption experiments were also conducted to monitor the possible influence of the SP on the deactivation pathways of 1 S * of the CPADB.The transient absorption spectra recorded for CPADB in the presence of SP just after the excitation is similar to the analogous spectra recorded for CPADB (Figure S7).However as presented in Figure S7B, formation of an additional band in the region of 500-550 nm was detected at longer time delays.Interestingly the new species with the absorption in the 500-550 nm does not decay over the 3 ns period (Figure S8).By comparing the monitored changes in the UV-vis spectra and ultrafast transient absorption experiments of CPADB with and without SP upon irradiation, it can be concluded that there is an interaction between SP and the photoexcited CPADB into S 1 state.The most possible interaction is the electron transfer from SP to the excited CPADB, followed by decarboxylation of SP and formation of other decomposition products.

Temporal control of photo-RAFT polymerization
A vial (4.8 mL) was charged with OEOMA 500 (543 µL, 300 mM), SP (0.11 g, 255 mM) dissolved in water, and CPADB (3.2 mg, 1.5 mM) in DMSO.The total reaction volume (including monomer) was set to 3.8 mL while adjusting the DMSO content to 10 % v/v.The homogeneity was achieved by a high-speed vortex, and the opened vial was placed into an air-conditioned photoreactor fitted with the UV-light source.The reactor was exposed to periodic irradiation/dark phases that each lasted for around 20 min.Before withdrawing the aliquots (100 µL), the reactor was shortly stirred using a vortex device to ensure homogeneity prior to analyzing by 1 H NMR and GPC techniques.

In situ chain extension experiment
The in-situ chain extension experiment was carried out by targeting DP = 50 for the first block and DP = 150 for the second block.Briefly, OEOMA 500 (0.25 g, 0.5 mmol) and SP (28.2 mg, 0.25 mmol) were mixed in a volumetric flask (1.0 mL), followed by the addition of CPADB (2.8 mg, 0.010 mmol) stock solution in DMSO.After adjusting the DMSO volume (100 µL, 10% v/v), the volumetric flask (1.0 mL) was topped up with water.The mixture was vortexed for 3 minutes, transferred to a 1 mL open-cap glass vial, and was irradiated with light (3 h) to carry out the polymerization.Polymerization was stopped by turning the light off, and aliquot of polymerization (250 µl) was mixed with a fresh OEOMA 500 (62.5 mg, 0.12 mmol) and water (250 µL) in a glass vial (1 mL).The mixture was vortexed and irradiated with UV light for 0.5 h.

Procedure for photo-RAFT polymerization at different target DP
The target degrees of polymerization (DP) were varied by adjusting the CPADB concentration, while the concentrations of all the other components, such as OEOMA 500 (300-800 mM), SP (128-256 mM), and DMSO (10% v/v) were kept constant.After preparing the mixture, it was vortexed for 3 min, transferred into five open-to-air glass vials (1 mL), and was irradiated with UV light to initiate polymerization.

Scale up of photo-RAFT polymerization
Polymerization at five different scales was carried out in completely open to air round bottom flasks (RBF) of different sizes (10 mL, 25 mL, 50 mL, 100 mL, and 200 mL) and stirred at 250 RPM for 1 h.The mixture was irradiated with UV light.In all cases, monomers without standard inhibitor purification procedures were used for polymerization.

Synthesis of 2-(methylsulfinyl)ethyl methacrylate (MSEMA)
A round bottom flask (250 mL) was charged with a stir bar, MAA (10.3 g, 0.12 mol), EDC (24 g, 0.12 mol), 2-(methylthio)ethanol (10 g, 0.11 mol) and DCM (150 mL) and placed in an ice bath.When the temperature was around 0 °C, DMAP (2.0 g, 16 mmol) was added dropwise into the reaction solution.The reaction was carried out for 24 h at 0 °C.The product was isolated by three extractions with 1 M HCl solution; the organic phase was dried with MgSO 4 and filtered through neutral alumina.DCM was removed by blowing air over the surface overnight.
The purified product was directly used for the oxidation with H 2 O 2 (1.4 equiv.)which was added dropwise.The reaction mixture was left under stirring overnight.The product was isolated by performing three extractions with DCM.The organic phase was treated with a small amount of manganese dioxide (MnO 2 ) to completely remove the peroxide and MgSO 4 to remove any trace of water.After filtration, DCM was removed by air blowing overnight dried a vacuum desiccator for 24 h.

Figure S2 .Figure S3. 1 H
Figure S2.Effect of SP concentration in photo-RAFT polymerization of OEOMA 500 with SP upon irradiation with UV for 1 h: final conversions (A) and kinetics of polymerization at different SP concentrations (B).
Effect of SP concentration in photo-RAFT polymerization of OEOMA 500 with SP upon irradiation with blue light for 4 h: final conversions (A) and kinetics of polymerization at different SP concentrations (B).

Figure S5 .
Figure S5.Comparison of decomposition of CPADB in DMSO-water (1:2 v/v) under A) 370 nm and B) 505 nm irradiation as measured by UV-Vis spectroscopy.Unlike SP-free conditions which exhibited nearly identical UV-vis spectra, when the steady state irradiation of CPADB has been performed in the presence of SP (255 mM), the intensity of the CPADB band at 475-575 nm decreases upon light irradiation (FigureS6, red square).This was also accompanied by an increase in overall absorption, which was attributed to the light scattering as some photoproducts of SP and CPADB reaction is possibly not well soluble in this solvent mixture.

Figure S6 .
Figure S6.Decomposition of CPADB in DMSO-water (1:2 v/v) in the presence of SP (255 mM) under 505 nm irradiation as measured by UV-VIS spectroscopy.

Figure S7 .
Figure S7.Transient absorption spectra obtained at different time delays for A) CPADB (14 mM) and B) CPADB in the presence of SP (255 mM) in DMSO-water (1:2 v/v) following the 470 nm laser excitation.

Figure S8 .
Figure S8.Absorption time profiles at 530 nm measured for the CPADB in the presence of SP (255 mM) following the 470 nm laser excitation.

Figure S10 .
Figure S10.The relationship between k p app and volume of the polymerization mixture for SP-RAFT polymerization.
g, 6.8 μmol, 102 μmol amine group) in 100 mM sodium phosphate buffer (pH 8.0) in ice bath, a solution of CPADB-NHS (77 mg, 204 μmol) in DMSO (1.0 mL) was added dropwise.The mixture was stirred at 4 °C for 16 h.CT-macro-CPADB was purified by dialysis using 12-14 kDa molecular cutoff dialysis membrane in 25 mM sodium phosphate buffer (pH 7.0) and deionized water at 4 °C for 24 h.CTmacro-CPADB was obtained as light pink powder after lyophilization.The number of CPADB on CT was estimated by fluorescamine assay.dialysis tube in deionized water and then characterized by aqueous SEC equipped with MALS detector.PPH kinetic assay.N-Succinyl-Ala-Ala-Pro-Phe-pNA (0 to 80 μL of 6.0 mM in DMSO) was added to sodium phosphate buffer (988 to 910 μL of 100 mM, pH 8.0).Native CT or conjugate solution (10 μL of 4.0 μM of CT) was added to the substrate solution.The initial rate of hydrolysis of the peptide substrate was monitored by recording the increase in absorption at 412 nm using an UV−vis spectrometer.The Michaelis−Menten parameters (k cat , K M , and k cat /K M ) were determined by nonlinear curve fitting (equation for Michaelis−Menten parameters) of plots of initial rate versus substrate concentration using the Enzfitter software.

Table of literature examples of oxygen-tolerant RAFT polymerization in waterTable S2 .
Literature examples of oxygen tolerant RAFT polymerization in water.

Table S3 .
Properties of POEOMA 500 synthesized with different target DP and concentrations of OEOMA 500 by photo-RAFT polymerization in the presence of CPADB and SP.a)

Table S4 .
1H NMR spectrum of MSEMA in DMSO-d 6 .were mixed.The rate of the hydrolysis was determined by recording the increase in absorbance at 412 nm for the first 20 s after mixing.k cat , K M and k cat /K M values were calculated using GraphPad Prism software when plotting substrate concentration versus initial hydrolysis velocity.

Table S5 .
Michaelis-Menten parameters of native CT after treatment with PBS (control), light, SP, and SP with light.