Chemical amplification accelerates reactive oxygen species triggered polymeric degradation

Abstract

Chemical amplification is a known strategy for improving the sensitivity of stimuli-responsive polymers. However, the chemical amplification effect has never been fully examined. Many questions remain about its mechanism and efficacy, obstructing its further implementation. Here, we design and demonstrate a reactive oxygen species (ROS) responsive polymer (ROS-ARP) with a chemical amplification strategy to dismiss these concerns. The ROS-ARP is designed to change the hydrophilicity by ROS, revealing a carboxylic acid, which also catalyzes ketal hydrolysis along the polymer backbone. The chemical amplification strategy of ROS-ARP accelerated the polymer degradation up to 17 fold compared to a previously reported ROS-responsive polymer. To investigate the mechanism behind this increased acceleration, we compared the degradation kinetics in various environments. Additionally, other effects such as hydrophilicity changes were excluded. The accelerated degradation of ROS-ARP is evaluated as a potential drug delivery system, demonstrating on-demand cargo release from the formulated polymeric particles.

---

Introduction

On-demand control of polymeric nanoparticle properties has been attractive in a broad range of fields such as cosmetics, environmental engineering, energy and electronics, and biomedical applications.^1–3 Despite its significance, few polymers can degrade or change their properties by relevant stimuli concentrations.^4–6 In particular, it is a challenge to develop polymers degrading by biological stimuli, since they need to be sensitive to subtle change of the triggers.^7–11 For this reason, very few ROS (reactive oxygen species) responsive materials react to pathological ROS concentrations,^12–15 although ROS are important biomarkers in inflammatory diseases.^16–19

One avenue to address this sensitivity challenge is by applying a chemical amplification strategy.^20–23 However, the effect of the strategy and the mechanism are not fully understood. Recently, Song et al.²⁴ reported a hydrogen peroxide sensitive, chemical amplifying polymer. A hydrophilicity switch occurred at the hydrophobic portion of the amphiphilic polymer when exposed to H₂O₂, which caused oxidation to a carboxylic acid, followed by ketal hydrolysis. Although the polymeric particle disassembly was accelerated successfully, it is difficult to evaluate a chemical amplification effect without considering the contribution from a hydrophilicity change by the carboxylic acid. Therefore, questions remain such as whether the revealed acid catalyzes ketal hydrolysis, and whether the chemical amplification strategy works directly on the polymer.

Here, we designed an H₂O₂-sensitive polymer incorporating a chemical amplification strategy for increased polymer degradation efficiency to dispel issues with the strategy (Scheme 1). The designed polymer is composed of three monomers each having different functionalities. The degradation of the polymer arises from the hydrolysis of ketal groups on the backbone. Ketal groups are great candidates because of their rapid catalytic degradation.^25–29 However, they suffer from poor stability and a short shelf life. One way to overcome the stability issue is to add a logic gate system. The logic gate system is based on different ketal hydrolysis rates depending on the hydrophilicity of the polymers the ketal groups are incorporated in.^30–34 In this study, we incorporated two ROS-sensitive functional groups into the polymer to control the hydrolysis rate of the ketals.

Scheme 1 Polymerization scheme and H₂O₂-responsive ROS-ARP degradation mechanism.

First, a sulfide group is incorporated, not only as a Michael addition donor for polymerization but also as an ROS sensitive hydrophilicity switch. Sulfide groups are oxidized to sulfoxides or sulfonyl groups by reacting with various ROS, such as H₂O₂ and hypochlorous acid.^35–37 This oxidation of sulfides changes the polymer's polarity and eventually induces hydrophilicity change of sulfide containing polymers.^38–42

Another ROS-sensitive functional group is hydroxymethyl-phenyl boronic acid pinacol ester (BE). It is well known that BE can be quantitatively cleaved by H₂O₂ with high selectivity and sensitivity.^43–50 This functional group is conjugated with a carboxylic acid of an acryloyl modified lysine to yield an H₂O₂-responsive group masking the acid. Acid release by H₂O₂ amplifies the ROS signal and catalyzes the hydrolysis of the ketals in the polymer backbone.

By polymerizing the monomers with three different functionalities, a chemically amplified highly efficient polymer degradation was expected. To prove that the chemical amplification accelerated the polymer degradation, we compared the degradation kinetics between a newly designed polymer and a control polymer without an amplification strategy. Additionally, a hydrophilicity switch effect was excluded by suppressing ketal hydrolysis following increased water accessibility. Based on the investigated degradation mechanism, on-demand polymeric particle degradation was demonstrated after nanoparticle (NP) formulation. The polymeric NPs were also evaluated for potential use as a drug delivery material by studying H₂O₂-responsive particle degradation, controlled cargo release and cell viability.

Results and discussion

Synthesis of monomers and polymerization

Monomers were synthesized by following Scheme 2. For H₂O₂ sensitive BE monomer synthesis, 4-(hydroxymethyl)phenyl boronic acid pinacol ester was conjugated onto N,N′-di-boc-L-lysine by DCC coupling. By following tert-butyloxycarbonyl (boc) de-protection, compound 2 was synthesized. The ROS sensitive, BE modified monomer, compound 3, was prepared by reacting acryloyl chlorides with compound 2 (Fig. S1†). A pH sensitive ketal monomer was synthesized by following previously reported methods.³²

Scheme 2 Synthesis scheme of H₂O₂-sensitive boronic ester monomer (compound 3) (A) and acid-sensitive ketal monomer (compound 5) (B). Reagents and conditions: (a) DCC, DMAP, DCM, r.t., 15 h, 49%; (b) TFA/DCM, r.t., 1 h, 96%; (c) acryloyl chloride, TEA, DCM, 0 °C, 1 h, 33%; (d) (i) 2-methoxypropene, PTSA, benzene, r.t., 1 h, (ii) TEA, acetic anhydride, benzene, r.t., 15 h, 66% over 2 steps; (e) (i) NaOH/H₂O, reflux, 15 h, (ii) acryloyl chloride, TEA, H₂O/dioxane, −5 °C, 1 h, 90% over 2 steps.

Compounds 3 and 5 were polymerized with 1,3-propandithiol in d₆-DMSO under argon. DBU is used as a catalyst. After 2 hours of reaction time, the crude solution became viscous. Proton peaks of the acryloyl group were completely flattened in the NMR spectrum after overnight incubation. This implies that all the acryloyl groups are consumed and the polymerization completed. This reaction time is remarkably shorter than TEA catalyzed Michael addition polymerization.^20,31 DBU induced faster polymerization since DBU has higher basicity and also supports the reaction by acting as a nucleophile.^51–53 After purification, a pale yellow oily product was obtained and characterized by gel permeation chromatography (GPC) and ¹H-NMR (Fig. S2†). The molecular weight of the polymer was calculated based on a PMMA standard curve. The calculated average molecular weight of the polymer was 10 kDa and the polydispersity index was 1.25. The monomer incorporation ratio was calculated based on H-NMR peaks, and the ratio between BE monomers and ketal monomers was 1 : 1.3.

Stimuli triggered polymer degradation

After obtaining the polymer, H₂O₂ triggered polymer degradation was evaluated by GPC. The degradation of the polymer backbone is triggered by H₂O₂ but ultimately leads to the hydrolysis of ketals on the polymers. Thus, after each time point, polymer solutions in two different H₂O₂ concentrations (0 mM and 10 mM) were compared at different pH values (5.0, 7.4, and 10.0) (Fig. S3 and S4†).

First, the polymer degradation at neutral pH, pH 7.4, with and without H₂O₂ was compared. Fig. 1A is the RI signal of the ROS-ARP obtained by GPC. Using the signal, we calculated and analyzed the molecular weight changes of the ROS-ARP in two different H₂O₂ concentrations at each time point (Fig. 1B). Under 10 mM H₂O₂, 59% of the calculated molecular weight of the polymer decreased in 24 hours, while the polymer in 0 mM H₂O₂ at pH 7.4 degraded only 8% during the same period. Also, UV absorption in the low molecular weight range remarkably increased with incubation time (Fig. S5†). These are considered as peaks of products from the H₂O₂ and BE reaction. To support this, a representative product, 4-hydroxybenzyl alcohol, was injected into the GPC under the same conditions as the polymers. The retention time of 4-hydroxybenzyl alcohol had a similar UV absorption to the aforementioned peaks (Fig. S5†).

Fig. 1 (A) GPC chromatogram of ROS-ARP in 0 mM H₂O₂ (left) and 10 mM H₂O₂ (right) at pH 7.4 (detector; RI, 20% buffer and 80% DMF, 24 hours incubation at 37 °C). (B) Molecular weight change of the polymer with and without H₂O₂ at pH 5.0, pH 7.4 and pH 10.0. Polymer M_W was calculated by GPC with the PMMA standard (n = 3). (C) Half-lives of polymer degradations by 10 mM H₂O₂ in a different pH buffer. (D) Polymer structure and degradation mechanism of the control polymer.

The half-life (t_1/2) of the polymer degradation by 10 mM H₂O₂ was calculated and compared with a control polymer which has been previously published by our group³⁰ (Fig. 1C). This control polymer has sulfide groups and ketal groups in the polymer backbone as ROS-ARP (Fig. 1D). However, the BE group, the ROS sensitive carboxylic acid protecting group, is not included. Therefore, the ketal hydrolysis rate is assumed to rely only on hydrophilicity changes by sulfide oxidation. The t_1/2 of the ROS-ARP at pH 7.4 with 10 mM H₂O₂ was 18.69 hours and it was 17.35 times shorter than the t_1/2 of the control polymer which does not contain BE groups for chemical amplification.

Investigation of the degradation mechanism

The ROS-ARP degraded more efficiently and faster than the previously reported control polymer. Several factors contribute to the accelerated ROS-ARP degradation.

Firstly, the H₂O₂ sensitivity of ROS-ARP is higher than that of the control polymer. Compared with the control polymer, which only has sulfides as an ROS sensitive group, the ROS-ARP additionally has BE groups. The BE group is more sensitive to H₂O₂ than the sulfide.^44–46 Therefore, the hydrophilicity change of the ROS-ARP is faster, and contributes to faster polymer degradation.

Secondly, upon reaction with ROS, the ROS-ARP transforms into a more hydrophilic polymer than its control. The ROS oxidizes the sulfide groups in the control polymer, and the produced sulfoxides induce a hydrophilicity change of the entire polymer. Since the ROS-ARP has two of these sulfide groups in a repeating unit while the control has one, the hydrophilicity change of ROS-ARP by H₂O₂ is greater than the control. Moreover, in the ROS-ARP, H₂O₂ unmasks carboxylic acids, further enhancing its hydrophilicity in addition to the change done by the sulfoxides. Thus, the ROS-ARP has two switches that cause increased hydrophilicity. Furthermore, since a carboxylic acid is more polar than a sulfoxide, the hydrophilicity increase of the ROS-ARP is more drastic than the control. The increased hydrophilicity promotes water access to the polymer, and it accelerates the polymer degradation by the ketal hydrolysis.

Lastly, a local pH change by unmasked carboxylic acids accelerates the ROS-ARP degradation. In the ROS-ARP, the carboxylic acid group is revealed by ROS after the BE group detachment. We hypothesize a possibility that the revealed carboxylic acid catalyzes the ketal hydrolysis in the ROS-ARP backbone in a chemical amplification process. This hypothesis is based on several previous studies applying the strategy.^20,23 However, it has not been clarified whether a hydrophilicity change or a chemical amplification is the primary contributor to polymer degradation. Consequently, the efficiency of chemical amplification is still questionable.

To investigate which effect dominates between hydrophilicity change and local pH change by acid release for accelerating ketal hydrolysis, polymer degradation was evaluated at basic pH (pH 10.0). Under basic conditions, the ketal hydrolysis is extremely suppressed.^25,54 Therefore, the ketal hydrolysis caused by a hydrophilicity change of the polymer is inhibited as no protons are available for catalyzing hydrolysis. In other words, although the hydrophilicity of the polymer is largely increased by ROS, the ROS-ARP degradation is not attributed to increased water accessibility under basic conditions. Note that the hydrolysis rate of amide bonds, the other functional group in the polymer backbone, has negligible difference between pH 5.0 and 10.0.⁵⁵

The molecular weight changes of the ROS-ARP in basic solution are shown in Fig. 1B. Compared to the initial molecular weight, it decreased by 20% right after the H₂O₂ addition and decreased by 62% in 24 hours with 10 mM H₂O₂ at pH 10.0. This confirms that the ketal groups were hydrolyzed even under basic conditions. This means that the polymer was degraded by the unmasked acid, and not by a hydrophilicity change. Moreover, at higher pH, the t_1/2 of ROS-ARP was shorter (Fig. 1C). This is a reverse tendency to the pH dependent ketal hydrolysis rate, but corresponds well to the BE reactivity to H₂O₂. Since the BE and H₂O₂ reaction is faster at a higher pH,^56–59 carboxylate unmasking is also faster under basic conditions. On the other hand, the control polymer's half-life was shorter at lower pH, because the polymer degradation by ketal hydrolysis fully relies on the ROS-triggered hydrophilicity change. These two results support that the BE groups and unmasked carboxylic acids are the main factors in the degradation kinetics of ROS-ARP.

Additionally, ketal hydrolysis is more influenced by carboxylic acids in the same molecule than the surrounding acids, because of the conformational proximity.^60–63 We assume that a similar effect occurs in a polymer, so that carboxylic acids in a polymer catalyze ketal groups in the same polymer backbone more efficiently than the surrounding protons in a solution. Therefore, we conclude that the accelerated ROS-ARP degradation depends more on chemical amplification than hydrophilicity change, due to the H₂O₂-sensitive acid releasing followed by the acid catalyzed ketal hydrolysis.

Nanoparticle formulation and H₂O₂ triggered degradation

After verification of H₂O₂-responsive polymer degradation, nanoparticle formulation and degradation under varying H₂O₂ concentrations were demonstrated. Nanoparticles were formed via nanoprecipitation, and DLS estimated the hydrodynamic nanoparticle size as 208 nm (polydispersity index (PDI) = 0.188) (Fig. S6†). Based on TEM observations, the morphology of the nanoparticles was spherical and the size corresponded to the DLS measurements (Fig. S6†).

The nanoparticles were treated with various H₂O₂ concentrations at pH 7.4 and degradation was observed based on the light scattering intensity measured by DLS (Fig. 2). In 10 mM H₂O₂, the scattering intensity of the nanoparticle solution decreased by 30% immediately after H₂O₂ addition, and by 93% in 24 hours. At 100 μM H₂O₂, the intensity decreased by 25% after 24 hours. At the physiological H₂O₂ concentration in inflammation, 50 μM, 9% of the scattering intensity decreased in the same period.

Fig. 2 Light scattering intensity change of NPs in different H₂O₂ concentrations (0, 50, 100 μM and 1 and 10 mM; pH 7.4, 10 mM phosphate buffer).

The morphology and size population change of NPs in a 10 mM H₂O₂ environment were also evaluated by TEM (Fig. 3A and B). Three representative images of NPs, treated with 0 mM or 10 mM H₂O₂ for 24 hours at 37 °C, were compared for size (Fig. S8†). The size of the nanoparticles was analyzed from the TEM images. In 0 mM H₂O₂, the ROS-ARP NPs maintained their initial mean size of 150 nm (Fig. 3B, left). In 10 mM H₂O₂ solution, the mean size of the NPs was 62 nm based on the images (Fig. 3B, right). The size change of the NPs with H₂O₂ suggests that NPs are degraded by polymer degradation. This size change corresponds to the DLS scattering light intensity change by H₂O₂ (Fig. 2). Furthermore, in the presence of H₂O₂, larger particles (400–700 nm) were observed with a different morphology compared to the NPs in 0 mM H₂O₂ (Fig. 3A). These NPs had lower electron density believed to be caused by swelling of the NPs and polymer degradation.

Fig. 3 (A) Morphology of NPs, with 0 mM (left) and 10 mM H₂O₂ (right). (B) Size population histogram based on TEM images (left; 0 mM H₂O₂, right; 10 mM H₂O₂) (after 24 h incubation at 37 °C).

ROS-responsive payload release

To demonstrate the ROS-dependent payload release from the NPs, IR-780 iodide was loaded into the NPs (IR780-NPs). The IR780-NP solutions were treated with varying concentrations of H₂O₂ and the fluorescence was measured at different time points. Free IR-780 has emission at λ_max = 798 nm in pH 7.4 10 mM phosphate buffer, but the λ_max shifts (λ_max = 823 nm) due to encapsulation.^64,65 Nonetheless, the area under the curve (AUC) comparison of the fluorescence intensity of the loaded and released dye is still difficult as the peaks are in close proximity. Thus, Gaussian fitting of the raw fluorescence intensity signal was used to separate released vs. encapsulated IR-780. The details are described in the ESI (Fig. S9†). The fitting method itself was verified by mixing various amounts (0, 10, 20, 30, 40 and 50 μl) of 0.3 μg ml⁻¹ IR-780 solution with 100 μl of 0.1 mg ml⁻¹ IR780-NP solution. Consequently, the calculated AUC of the free dye increased linearly (R² = 0.9971) while the AUC of the loaded dye (IR780-NPs) signal remained constant (Fig. S10†).

Using the signal separation method, the fluorescence signal from the loaded IR-780 was compared at different incubation concentrations of H₂O₂ and time in 10 mM phosphate buffer (Fig. S11†). Note that the signals of the released IR-780 are not comparable due to the hydrophobic, unstable nature of IR-780 in the buffer. Fig. 4A shows the separated fluorescence signal of the encapsulated IR-780 with 0 mM (Fig. 4A, left) and 10 mM H₂O₂ (Fig. 4A, right). In the presence of H₂O₂, the loaded dye signal clearly decreased. Furthermore, the signal decrease was faster when increasing the H₂O₂ concentration (Fig. 4B and Fig. S11†).

Fig. 4 (A) Fluorescence intensity of IR780-NPs in 0 mM (left) and 10 mM H₂O₂ (right) at pH 7.4, r.t. (B) percent release of dye, based on the AUC of IR780-NP fluorescence intensity (Standard deviation was obtained from Gaussian fitting).

Cell viability

To test the applicability of ROS-ARP nanoparticles as drug delivery vehicles, we assessed the cellular compatibility in Raw 264.7 mouse macrophages (Fig. S12†). The cells tolerated the NPs well up to the maximum tested concentration of 300 μg mL⁻¹, suggesting their potential for drug delivery.

Conclusions

Herein we designed an ROS sensitive degradable polymer applying a chemical amplification strategy for fast and efficient polymer degradation. We successfully synthesized the polymer and verified the chemical amplification effect on its degradation. The ROS-ARP degraded over 17 times faster than its control polymer without an amplification strategy. The accelerated degradation arises from the unmasked carboxylic acid which catalyzes a ketal hydrolysis. As suggested here, this strategy and degradation mechanism may be applied in future biomedical applications such as for drug delivery.

Experimental

General methods and instrumentation

All chemicals and solvents were purchased from Sigma-Aldrich unless specified. Silica gel column chromatography for purification was conducted with an automated CombiFlash® Rf 200 system. The synthesized compounds were confirmed by nuclear magnetic resonance (NMR) and high-resolution mass spectroscopy. Acquired ¹H NMR and ¹³C NMR spectra were based on a Varian spectrometer working at 600 MHz and 150 MHz, respectively. An Agilent 6230 ESI-TOFMS was used for high-resolution mass spectra in positive ion mode. The synthesized polymers and degradation were characterized by size exclusion gel chromatography, an Agilent 1100 Series HPLC system equipped with RI, Agilent 1260 Light Scattering and PDA detectors and a Waters Styragel HR 2 size-exclusion 22 column with 0.1% LiBr/DMF as the eluent and a flow rate of 1 mL min⁻¹ at 37 °C. The molecular weight and polydispersity of polymers were calculated based on monodisperse poly(methylmethacrylate) (PMMA) standards. For particle washing, tangential flow filtration was conducted (Pellicon XL, 500 kDa, Millipore) with 300 mL cell-culture grade water (HyClone) at 50 RPM. Dynamic light scattering (DLS), Malvern Instruments Zetasizer ZS, was utilized for particle characterization and degradation study, and size of the particles was acquired based on culmulants analysis using the scattering of a 4 mW 633 nm laser. The morphology of the nanoparticles was observed by transmission electron microscopy (TEM, Tecnai FEI Spirit). An Agilent Technologies 1260/1290 Infinity HPLC-MS equipped with a ZORBAX 1.8 μm (2.1 × 50 mm) SB-C18 column and a 6120 Quadrapole detector was used for HPLC-MS chromatograms, and it was run in a gradient of water and acetonitrile with 0.1% formic acid.

Synthesis of compound 1

4-(Hydroxymethyl)phenyl boronic acid pinacol ester (1.0 g, 4.27 mmol), N,N′-di-boc-L-lysine (1.48 g, 4.27 mmol) and 4-(dimethylamino)pyridine (DMAP) (260 mg, 2.14 mmol) were dissolved in 20 ml of anhydrous dichloromethane (DCM) under an argon flow. The mixture was cooled down using an ice bath and followed by dropwise addition of 1.32 g (6.4 mmol) of N,N′-dicyclohexylcarbodiimide (DCC) in 3 ml of DCM. After the addition, the resulting solution was stirred overnight at r.t. The crude solution was filtered to remove side products (1,8-diazabicyclo[5.4.0]undec-7-ene, DBU), and then purified with a column. A final product was obtained after purification using gel chromatography with 20% ethyl acetate in hexane (1.04 g, 43.3%).

¹H NMR (600 MHz, CDCl₃) δ 7.80 (d, J = 7.7 Hz), 7.34 (d, J = 7.7 Hz), 5.17 (dd, J = 45.6, 12.7 Hz), 4.32 (dd, J = 11.9, 7.6 Hz), 3.06 (dd, J = 11.2, 5.7 Hz), 1.60 (bs), 1.44 (s), 1.34 (s).

¹³C NMR (151 MHz, d₆-CDCl₃) δ 172.6, 156.1, 155.49, 138.4, 135.0, 127.4, 83.9, 79.9, 79.1, 66.9, 53.3, 40.1, 32.3, 29.5, 22.4, 28.5, 28.4, 24.9.

MS (ESI) m/z: [M + Na]⁺ calcd for C₂₉H₄₇BN₂O₈Na, 585.33; found 585.2.

Synthesis of compound 2

Compound 1 (885 mg, 1.57 mmol) was dissolved in 4 ml of DCM and 4 ml of trifluoroacetic acid (TFA) and stirred for 1 h at r.t. The reaction was monitored by thin-layer chromatography (TLC) and LCMS. When the reaction was completed, a crude solution was dried using a rota-vap to remove TFA. (1.19 g, 93%).

¹H NMR (600 MHz, d₆-DMSO) δ 8.48 (bs, 3H), 7.78 (bs, 3H), 7.70 (d, J = 8.0 Hz, 2H), 7.42 (d, J = 8.0 Hz, 2H), 5.27 (dd, J = 15.2, 12.8 Hz, 2H), 5.15 (bs, 2 equiv. of TFA), 4.10 (dd, J = 11.3, 6.1 Hz, 1H), 2.74 (m, 2H), 1.80 (m, 2H), 1.53 (m, 2H), 1.43–1.31 (m, 2H), 1.29 (s, 12H).

¹³C NMR (151 MHz, d₆-DMSO) δ 169.4, 138.5, 127.5, 83.8, 67.0, 51.8, 38.4, 29.5, 26.5, 21.3.

MS (ESI) m/z: [M + H]⁺ calcd for C₁₉H₃₁BN₂O₄, 363.24; found 363.1.

Synthesis of compound 3

Compound 2 (521 mg, 0.93 mmol) was dissolved in 8 ml of anhydrous DCM and 1 ml (7.44 mmol) of TEA. The solution was cooled down using an ice-bath and followed by acryloyl chloride (165.7 μl, 2.05 mmol) addition. After 30 min of stirring at r.t., the product was washed with 800 ml of 50 mM HCl (aq) and 200 ml of additional DCM. The organic layer was collected and dried using a rota-vap, after treating with MgSO₄ to remove the remaining H₂O. The product was purified again using column chromatography with 95% DCM and 5% MeOH (136 mg, 33%) (Fig. S1†).

¹H NMR (600 MHz, CDCl₃) δ 7.79 (d, J = 7.8 Hz, 2H), 7.32 (d, J = 7.8 Hz, 2H), 6.38 (d, J = 8.1 Hz, 1H), 6.33–6.06 (m, 4H), 5.82 (t, J = 5.5 Hz, 1H), 5.67 (d, J = 10.4 Hz, 1H), 5.60 (d, J = 10.4 Hz, 1H), 5.17 (dd, J = 42.8, 12.5 Hz, 2H), 4.67 (m, 1H), 3.34–3.20 (m, 2H), 1.87 (m, 1H), 1.74 (m, 1H), 1.61–1.45 (m, 2H), 1.39–1.33 (m, 1H), 1.32 (s, 12H).

¹³C NMR (151 MHz, CDCl₃) δ 172.2, 165.9, 138.2, 135.1, 130.6, 127.5, 127.4, 126.4, 84.0, 67.1, 51.8, 38.7, 31.9, 28.5, 24.9, 22.1.

MS (ESI) m/z: [M + Na]⁺ calcd for C₂₅H₃₅BN₂O₆Na, 493.25; found 493.1.

Synthesis of compound 5

Compound 4 was obtained by following the same procedure as in ref. 32. Compound 4 (3.0 g, 7.1 mmol) was dissolved in 6 N NaOH (aq) (9 ml) and refluxed overnight. After the overnight reaction, 9 ml of 1,4-dioxane was added into the resulting solution and cooled down using an ice-bath. TEA (11.9 ml, 85.2 mmol) and acryloyl chloride (1.15 ml, 14.2 mmol) were added dropwise into the mixture, while monitoring the temperature and pH of the solution. The temperature of the solution was kept under 10 °C, and the pH was maintained above pH 13. After the addition, the solution was stirred for 1 h, and extracted with ethyl acetate (EA) and column chromatography followed. After the gel column purification, a light yellow compound was obtained (500 mg, 20%).

¹H NMR (600 MHz, d₆-DMSO) δ 8.15 (t, J = 5.5 Hz, 2H), 6.24 (dd, J = 17.10, 10.10 Hz, 2H), 6.07 (dd, J = 16.9, 2.0 Hz, 2H), 5.56 (dd, J = 10.2, 2.0 Hz, 2H), 5.27 (dd, J = 15.2, 12.8 Hz, 2H), 3.39 (t, J = 5.9 Hz, 4H), 3.23 (dd, J = 11.7, 5.80 Hz, 4H), 1.26 (s, 6H).

¹³C NMR (151 MHz, d₆-DMSO) δ 164.7, 131.7, 125.1, 99.6, 66.4, 59.0, 24.7.

Synthesis of ROS-ARP

102 mg of compound 3 (1 equiv.) and 58.5 mg of compound 5 (1 equiv.) were put together under Ar and dissolved in 500 μl of d6-DMSO. 1,3-Propanedithiol (43.5 μl, 2 equiv.) and DBU (195 μl, 6 equiv.) were added into the reaction mixture. The mixture was stirred at r.t. overnight. The completion of polymerization was confirmed by NMR. The resulting viscous polymer solution was diluted with DCM and precipitated in diethyl ether, and this was repeated 3 times. The purified polymer was evaluated by NMR and GPC (Fig. S2†).

The molecular weight was 10 280 Da and the PDI was 1.25 with the PMMA standard.

¹H NMR (600 MHz, d₆-DMSO, and 5% CD₃OD) δ 8.15 (t, J = 5.5 Hz, 2H), 6.24 (dd, J = 17.10, 10.10 Hz, 2H), 6.07 (dd, J = 16.9, 2.0 Hz, 2H), 5.56 (dd, J = 10.2, 2.0 Hz, 2H), 5.27 (dd, J = 15.2, 12.8 Hz, 2H), 3.39 (t, J = 5.9 Hz, 4H), 3.23 (dd, J = 11.7, 5.80 Hz, 4H), 1.26 (s, 6H).

H₂O₂ sensitive polymer degradation

The ROS-ARP was dissolved in 0.1% LiBr DMF to 5 mg ml⁻¹. 400 μl of the polymer solution was mixed with 100 μl of aqueous solution (10 mM acetate buffer for pH 5.0, 10 mM phosphate buffer for pH 7.4, and 10 mM tris buffer for pH 10.0) and the final H₂O₂ concentration was set to 0 mM and 10 mM. Both samples were incubated for 24 h at 37 °C. Each sample was filtered using a 0.22 μm PVDF syringe filter before the GPC injection (Fig. S3†). The molecular weight of the polymers was calculated by analyzing the peaks between 10 and 16 min of retention time (Fig. S4†). 4-Hydroxybenzyl alcohol, a representative side product of the BE reaction with H₂O₂, was injected into the GPC under the same conditions to compare the retention time (Fig. S5†).

Particle formulation

The ROS-ARP (10 mg) was dissolved in 1 ml of DMF. The resulting polymer solution was added dropwise into 9.0 ml of 1 wt% polyvinyl alcohol (PVA) aqueous solution and stirred for 2 h at r.t. The crude solution was washed using tangential flow filtration to remove excess PVA and DMF. For IR-780 or rhodamine 6G encapsulated particles, 2 mg of encapsulates were dissolved in 1 ml DMF with ROS-ARP. Loading was calculated by dividing the amount of cargo by the weight of particles. The loading of IR-780 was 2 wt%.

The formulated particles were characterized by DLS and TEM (Fig. S6†).

Stimuli triggered nanoparticle degradation

The size distribution of nanoparticles (NPs) with different H₂O₂ concentrations was compared by DLS. 50 μg ml⁻¹ of nanoparticle solution with 2 mM phosphate buffer (pH 7.4) and 0.2% PVA aqueous solution was prepared. Each sample was mixed with a H₂O₂ solution and the final H₂O₂ concentrations were adjusted to 0 μM, 50 μM, 100 μM, 1 mM and 10 mM. All the samples were monitored by DLS for 24 h at r.t. Size distribution was measured at given time points (0, 1, 2, 3, 6, 9, 12 and 24 h). The nanoparticles were also incubated under acidic (pH 5.0) and basic conditions (pH 10.0) by mixing 10 mM acetate buffer or 10 mM of tris buffer.

To observe the morphological change of nanoparticles, 5 μl of 0 mM and 10 mM of H₂O₂ treated nanoparticle solutions were put on a TEM copper grid. The TEM grids were dried under vacuum and imaged by TEM. Size distribution was calculated from three representative images of each sample (Fig. S8†).

Release study

IR-780 was loaded into the ROS-ARP NPs, and 1 mg ml⁻¹ of NP solution was prepared. The NP solution was diluted 10 times with 0, 50, 100 μM and 1, 10 mM H₂O₂, 10 mM phosphate solution. Fluorescence was measured by using a spectrofluorometer (Horiba, FL-1000). The fluorescence intensity of the loaded IR-780 was separated from the released dye Gaussian fitting of the raw signal. All the input parameters are shown in Fig. S9.† The fitting method was verified using various amounts (0, 10, 20, 30, 40 and 50 μl) of 0.3 μg ml⁻¹ IR-780 solution mixed with 100 μl of 0.1 mg ml⁻¹ IR-780 loaded ROS-ARP NP solution. The loaded dye signal was separated out using the above method. Consequently, the calculated AUC of the free dye increased linearly (R² = 0.9971) while the AUC of the loaded dye (IR-780 loaded ROS-ARP NPs) was kept constant (Fig. S10†). These values were converted into AUC changes by free dye concentrations, and a linear calibration curve was obtained (R² = 0.9966, Fig. S10†). The separated IR-780 NP signals by H₂O₂ concentration are shown in Fig. S11.†

Cell viability

Empty ROS-ARP nanoparticles were diluted in sterile DMEM/FBS media and further diluted to appropriate concentrations. Raw 264.7 cells (American Type Culture Collection, ATCC® TIB-71™) were seeded in a 96-well plate (Corning) at a density of 20 000 cells per well before NP addition. The cells were incubated with particle suspensions in quadruplicates for 24 h at 37 °C, in 5% CO₂. Cells were then washed twice in warm PBS and AlamarBlue reagent was added to each well (according to manufacturer's instructions) at a 1 : 10 DMEM concentration and incubated for 4 hours. To quantify the cell viability, the fluorescence excitation/emission was set to 530/590 nm and read using a plate reader (Molecular Devices SpectraMax M5).

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We thank the Air Force office of Scientific Research (AFOSR), FA9550-15-1-0273, for funding. NMR data were acquired at the UCSD Skaggs School of Pharmacy and Pharmaceutical Sciences NMR Facility.

References

1. M. Motornov, Y. Roiter, I. Tokarev and S. Minko, Prog. Polym. Sci., 2010, 35, 174–211.
2. M. A. C. Stuart, W. T. S. Huck, J. Genzer, M. Muller, C. Ober, M. Stamm, G. B. Sukhorukov, I. Szleifer, V. V. Tsukruk, M. Urban, F. Winnik, S. Zauscher, I. Luzinov and S. Minko, Nat. Mater., 2010, 9, 101–113.
3. C. D. H. Alarcon, S. Pennadam and C. Alexander, Chem. Soc. Rev., 2005, 34, 276–285.
4. F. Liu and M. W. Urban, Prog. Polym. Sci., 2010, 35, 3–23.
5. O. Onaca, R. Enea, D. W. Hughes and W. Meier, Macromol. Biosci., 2009, 9, 129–139.
6. A. P. Esser-Kahn, S. A. Odom, N. R. Sottos, S. R. White and J. S. Moore, Macromolecules, 2011, 44, 5539–5553.
7. A. S. Hoffman, Adv. Drug Delivery Rev., 2013, 65, 10–16.
8. G. R. Martin and R. K. Jain, Cancer Res., 1994, 54, 5670–5674.
9. T. P. Szatrowski and C. F. Nathan, Cancer Res., 1991, 51, 794–798.
10. P. Niethammer, C. Grabher, A. T. Look and T. J. Mitchison, Nature, 2009, 459, 996–U123.
11. S. Yang, Z. Tang, D. Zhang, M. Deng and X. Chen, Biomater. Sci., 2017, 5, 2169–2178.
12. C. D. Lux, S. Joshi-Barr, T. Nguyen, E. Mahmoud, E. Schopf, N. Fomina and A. Almutairi, J. Am. Chem. Soc., 2012, 134, 15758–15764.
13. D. L. Zhang, Y. L. Wei, K. Chen, X. J. Zhang, X. Q. Xu, Q. Shi, S. L. Han, X. Chen, H. Gong, X. H. Li and J. X. Zhang, Adv. Healthcare Mater., 2015, 4(1), 69–76.
14. G. Saravanakumar, J. Kim and W. J. Kim, Adv. Sci., 2017, 4, 1600124.
15. Q. R. Deng, X. D. Li, L. P. Zhu, H. He, D. L. Chen, Y. B. Chen and L. C. Yin, Biomater. Sci., 2017, 5, 1174–1182.
16. K. Apel and H. Hirt, Annu. Rev. Plant Biol., 2004, 55, 373–399.
17. I. L. C. Chapple, J. Clin. Periodontol, 1997, 24, 287–296.
18. H. Wiseman and B. Halliwell, Biochem. J., 1996, 313, 17–29.
19. S. Joshi-Barr, C. D. Lux, E. Mahmoud and A. Almutairi, Antioxid. Redox Signaling, 2014, 21, 730–754.
20. J. Olejniczak, V. A. N. Huu, J. Lux, M. Grossman, S. He and A. Almutairi, Chem. Commun., 2015, 51, 16980–16983.
21. E. Han, B. Kwon, D. Yoo, C. Kang, G. Khang and D. Lee, Bioconjugate Chem., 2017, 28, 968–978.
22. M. E. Roth, O. Green, S. Gnaim and D. Shabat, Chem. Rev., 2016, 116, 1309–1352.
23. E. Sella and D. Shabat, Org. Biomol. Chem., 2013, 11, 5074–5078.
24. C. C. Song, R. Ji, F. S. Du, D. H. Liang and Z. C. Li, ACS Macro Lett., 2013, 2, 273–277.
25. E. H. Cordes and H. G. Bull, Chem. Rev., 1974, 74, 581–603.
26. T. H. Fife, Acc. Chem. Res., 1972, 5, 264–272.
27. T. H. Fife and L. K. Jao, J. Org. Chem., 1965, 30, 1492–1495.
28. A. J. Kirby, Acc. Chem. Res., 1984, 17, 305–311.
29. W. Chen, F. H. Meng, R. Cheng and Z. Y. Zhong, J. Controlled Release, 2010, 142, 40–46.
30. E. A. Mahmoud, J. Sankaranarayanan, J. M. Morachis, G. Kim and A. Almutairi, Bioconjugate Chem., 2011, 22, 1416–1421.
31. J. Sankaranarayanan, E. A. Mahmoud, G. Kim, J. M. Morachis and A. Almutairi, ACS Nano, 2010, 4, 5930–5936.
32. S. E. Paramonov, E. M. Bachelder, T. T. Beaudette, S. M. Standley, C. C. Lee, J. Dashe and J. M. J. Frechet, Bioconjugate Chem., 2008, 19, 911–919.
33. M. J. Heffernan and N. Murthy, Bioconjugate Chem., 2005, 16, 1340–1342.
34. B. Liu and S. Thayumanavan, J. Am. Chem. Soc., 2017, 139, 2306–2317.
35. C. Schoneich, Biochim. Biophys. Acta, Proteins Proteomics, 2005, 1703, 111–119.
36. X. H. Fu, Y. A. Ma, Y. Shen, W. X. Fu and Z. B. Li, Biomacromolecules, 2014, 15, 1055–1061.
37. B. L. Allen, J. D. Johnson and J. P. Walker, ACS Nano, 2011, 5, 5263–5272.
38. A. Napoli, M. Valentini, N. Tirelli, M. Muller and J. A. Hubbell, Nat. Mater., 2004, 3, 183–189.
39. Y. T. Chiang, Y. W. Yen and C. L. Lo, Biomaterials, 2015, 61, 150–161.
40. J. R. Kramer and T. J. Deming, J. Am. Chem. Soc., 2012, 134, 4112–4115.
41. B. Liu, D. L. Wang, Y. K. Liu, Q. Zhang, L. L. Meng, H. R. Chi, J. N. Shi, G. L. Li, J. C. Li and X. Y. Zhu, Polym. Chem., 2015, 6, 3460–3471.
42. M. S. Shim and Y. N. Xia, Angew. Chem., Int. Ed., 2013, 52, 6926–6929.
43. C. C. Song, R. Ji, F. S. Du and Z. C. Li, Macromolecules, 2013, 46, 8416–8425.
44. E. Lallana and N. Tirelli, Macromol. Chem. Phys., 2013, 214, 143–158.
45. C. C. Song, F. S. Du and Z. C. Li, J. Mater. Chem. B, 2014, 2, 3413–3426.
46. W. Q. Yang, X. M. Gao and B. H. Wang, Med. Res. Rev., 2003, 23, 346–368.
47. C. Tapeinos and A. Pandit, Adv. Mater., 2016, 28, 5553–5585.
48. M. C. Y. Chang, A. Pralle, E. Y. Isacoff and C. J. Chang, J. Am. Chem. Soc., 2004, 126, 15392–15393.
49. E. W. Miller, A. E. Albers, A. Pralle, E. Y. Isacoff and C. J. Chang, J. Am. Chem. Soc., 2005, 127, 16652–16659.
50. D. Srikun, A. E. Albers, C. I. Nam, A. T. Iavaron and C. J. Chang, J. Am. Chem. Soc., 2010, 132, 4455–4465.
51. C. E. Yeom, M. J. Kim and B. M. Kim, Tetrahedron, 2007, 63, 904–909.
52. J. W. Chan, C. E. Hoyle, A. B. Lowe and M. Bowman, Macromolecules, 2010, 43, 6381–6388.
53. D. P. Nair, M. Podgorski, S. Chatani, T. Gong, W. X. Xi, C. R. Fenoli and C. N. Bowman, Chem. Mater., 2014, 26, 724–744.
54. M. D. Pluth, R. G. Bergman and K. N. Raymond, J. Org. Chem., 2009, 74, 58–63.
55. R. M. Smith and D. E. Hansen, J. Am. Chem. Soc., 1998, 120, 8910–8913.
56. H. C. Brown and G. Zweifel, J. Am. Chem. Soc., 1961, 83, 2544–2511.
57. H. G. Kuivila, J. Am. Chem. Soc., 1954, 76, 870–874.
58. H. G. Kuivila and R. A. Wiles, J. Am. Chem. Soc., 1955, 77, 4830–4834.
59. H. G. Kuivila and A. G. Armour, J. Am. Chem. Soc., 1957, 79, 5659–5662.
60. B. Capon and M. C. Smith, Chem. Commun., 1965, 523–524.
61. T. H. Fife and E. Anderson, J. Am. Chem. Soc., 1971, 93, 6610–6614.
62. G. A. Craze and A. J. Kirby, J. Chem. Soc., Perkin Trans. 2, 1974, 61–66,  10.1039/p29740000061.
63. C. J. Brown and A. J. Kirby, J. Chem. Soc., Perkin Trans. 2, 1997, 1081–1093,  10.1039/a700155j.
64. C. X. Yue, P. Liu, M. B. Zheng, P. F. Zhao, Y. Q. Wang, Y. F. Ma and L. T. Cai, Biomaterials, 2013, 34, 6853–6861.
65. M. L. Viger, G. Collet, J. Lux, V. A. N. Huu, M. Guma, A. Foucault-Collet, J. Olejniczak, S. Joshi-Barr, G. S. Firestein and A. Almutairi, Biomaterials, 2017, 133, 119–131.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7bm00758b

---