Ratiometric Sensing of Hydrogen Peroxide Utilizing Conformational Change in Fluorescent Boronic Acid Polymers

We demonstrate that the copolymers containing boronic acid and pyrene units can be utilized for the fluorometric sensing of hydrogen peroxide (H2O2) in aqueous solutions. The copolymer exists in a relatively extended conformation in the absence of H2O2, whereas the polymer chain is contracted by the reaction of boronic acid moieties with H2O2 to form phenol groups. This conformational change induces aggregation of the originally isolated pyrene groups. As a result, relative intensity of excimer emission with respect to monomer emission increases with H2O2 concentration. Accordingly, the present methodology enables us to measure H2O2 by means of ratiometric fluorescence change in the range of 0–30 μM.


Introduction
Boronic acids are known to form boronate esters with saccharides through reversible boronic acid-diol interactions (Scheme 1(a)) [1,2]. By utilizing this nature, a variety of saccharide-sensing systems have been developed [3][4][5]. Boronic acids are also known to react with reactive oxygen species (ROS) such as hydrogen peroxide (H 2 O 2 ) to generate the corresponding phenol derivatives (Scheme 1(b)) [6][7][8][9][10][11][12]. This reaction has been utilized for creating colorimetric [13][14][15] and fluorescent [16][17][18][19][20] probes for sensing H 2 O 2 by directly connecting boronic acid moiety with chromophoric or fluorophoric compounds. When these probes react with H 2 O 2 , spectrophotometric changes are induced due to the alteration of the electronic state within the probes caused by the conversion of boronic acid moiety into phenol group. The boronic acid-based H 2 O 2 probes are advantageous over the conventional enzyme-based sensors [21] in terms of durability and reproducibility because they do not use unstable bioorganic substances. One of the drawbacks of the boronic acid-based method may be the necessity of complicated organic synthesis for obtaining the probes.
Our group has been developing saccharide-sensing systems utilizing polymeric boronic acids. Previously, we had reported a novel saccharide detection strategy utilizing saccharide-induced conformational changes in fluorescent boronic acid polymers, which was synthesized from a radical copolymerization of a boronic acid monomer and a pyrene monomer [22]. This polymer takes a relatively contracted conformation in an aqueous solution due to hydrophobic aggregation of pyrene groups. On binding with saccharides, the boronic acid moieties become negatively charged, and the polymer chain changes its conformation to an extended state due to intramolecular electrostatic repulsion between the boronate groups. Thus, the pyrene groups are enforced to be dissociated. These conformational changes can be conveniently detected by monitoring the excimer to monomer intensity ratio in the fluorescence spectra. It occurred to us that a similar mechanism may be applied for developing sensing systems for H 2 O 2 . If the charge state of the boronic acid groups in the fluorescent polymer is altered by the reaction with H 2 O 2 , the conformation of the polymer chain will also be varied due to change in electrostatic interaction within the polymer. Thus, we expect that the concentration of Here, we report our result on the endeavor for novel fluorometric sensing method for H 2 O 2 . Accordingly, we prepared copolymers containing boronic acid, acrylamide, and pyrene units with different compositions. After the reaction of the polymers with H 2 O 2 in aqueous solutions, fluorescence spectral changes of the solutions were measured. As a result, we found that the relative intensity of excimer emission with respect to monomer emission increases with H 2 O 2 concentration, and the present methodology enables us to measure H 2 O 2 by means of ratiometric fluorescence change in the range of 0-30 M.

Synthesis of Polymers.
The outline for the preparation of the fluorescent boronic acid polymers is illustrated in Scheme 2. The radical copolymerization of 1, 2, and 3 was carried out in homogeneous solutions in DMSO/H 2 O mixed solvent in the presence of a polymerization initiator (AIBN) at 70 ∘ C for 12 h under nitrogen atmosphere. For the monomer compositions, see Table 1. The obtained copolymers were purified by reprecipitation using acetone as a nonsolvent and then dried under vacuum. NMR spectra of the copolymers are shown in Figures S1-S5 (see Supplementary Material available online at https://doi.org/10.1155/2017/7829438). From these spectra, successful introduction of boronic acid   Table S1. By using these data, molar ratios of monomer units were estimated (Table S2). It was confirmed that the molar ratios were roughly agreed with compositions in the feed solutions. Molecular weights (Mw) of these copolymers are supposed to be in the range of 13,000-22,000 [22].

Apparatus.
Steady-state fluorescence spectra were recorded on Hitachi F-2500 fluorescence spectrophotometer. UV absorption spectra were obtained by using JASCO V-650 spectrophotometer. pH values were measured using Metrohm 827 pH lab. Aqueous solutions were prepared with distilled water purified by Yamato WG202 system.  (Figure 1). The excimer emission is derived from the dimerization between excited-state and ground state pyrene groups. The excitation spectra were also measured ( Figure S16) and it was found that the spectrum monitored at the excimer emission wavelength (486 nm) was remarkably different from that monitored at monomer emission wavelength (397 nm). The observed excitation spectra indicate that the pyrene groups form aggregations in the ground state probably due to hydrophobic and -stacking interactions [25]. When 1 M of H 2 O 2 was added to the solution, the excimer emission was slightly intensified relative to the monomer emission. With increasing H 2 O 2 concentration, the relative excimer intensity gradually increased and almost saturated above 30 M as shown in Figure 1. This spectral change accompanied a clearly visible variation in the emission color from the solution as shown in Figure 2. By measuring the corresponding UV absorption spectra ( Figure  S17), it was found that a new absorption assignable to phenol moiety emerged at around 290 nm with increasing H 2 O 2 concentration, and the absorbance leveled off above 30 M. To assess the reaction kinetics, time-course of the excimer to monomer intensity ratio was measured with 30 M H 2 O 2 ( Figure S18). The intensity ratio gradually increased with increasing reaction time and then reached constant after 60 min. These observations indicate that the boronic acid groups were quantitatively converted into phenol moieties when H 2 O 2 concentration is larger than 30 M, and the reaction of the boronic acid groups with H 2 O 2 induced a conformational change in the copolymer from expanded to contracted state. It is calculated that the concentration of boronic acid unit in the measurement solutions of B-11 is ca. 3.5 M. (Polymer B-11 was synthesized from 76 mg (0.4 mmol) of 1, 28.4 mg of 2, and 11.4 mg of 3. Supposing that the monomer composition in the copolymer is the same as in the feed solution, the number of moles of boronic acid per unit weight of the copolymer is calculated to be 3.45 mmol g −1 . Therefore, concentration of boronic acid unit in the measurement solutions containing 1 mg L −1 of B-11 is estimated to be 3.45 mol L −1 .) However, the detection curve exhibited saturation at much higher H 2 O 2 concentration. This means that H 2 O 2 does not quantitatively react with the boronic acid unit. Probably the decomposition reaction of H 2 O 2 into H 2 O and O 2 is competing, and/or the reaction is relatively slow so that the reaction with boronic acid is uncompleted within 60 min at lower H 2 O 2 concentrations.

Results and Discussion
The reaction of boronic acid with H 2 O 2 to be converted into phenol was also confirmed by 1 H NMR measurements. Firstly, phenylboronic acid was used as a model compound ( Figure S20). In an alkaline D 2 O solution, phenylboronic acid showed signals in the range of 7.6-7.2 ppm. After the reaction with H 2 O 2 , it showed upfield shifts to 7.2-6.6 ppm that is assignable to the quantitative conversion to phenol. Then, copolymer B-11 was dissolved in alkaline D 2 O and NMR spectra were measured in the absence and presence of H 2 O 2 ( Figure S21). In the absence of H 2 O 2 , it showed broad peaks assignable to the boronic acid moiety at around 7.2 ppm. In the presence of excess amount of H 2 O 2 , the peaks disappeared and new peaks attributable to phenol moiety emerged in the range of 7.0-6.2 ppm.
We also examined detection selectivity against reactive oxygen species [26]. In these cases, measurements were conducted at pH 10.2 with sodium carbonate buffer instead of Good's buffer to avoid causing side reactions of reactive oxygen species with buffer. As the results shown in Figure 3, hypochlorite induced change in the ratio of excimer to monomer emission intensities that is much smaller than the case of H 2 O 2 . Also, t-butyl hydroperoxide (TBHP) showed nearly negligible response.

Effect of pH.
We further investigated the effect of pH on the responsiveness towards H 2 O 2 . Figure 4 shows the pH dependency of the excimer to monomer intensity ratio in the absence and presence (30 M) of H 2 O 2 . At pH 7.4, there was no significant difference of intensity ratios in the absence and presence of H 2 O 2 . Unfortunately, further decrease in pH resulted in precipitation of the copolymer due to lack of boronate negative charges at acidic conditions. In the absence of H 2 O 2 , the relative intensity of excimer emission was almost constant with the increase in pH and then decreased when pH exceeded 10. The decrease in the excimer intensity may be due the extension of the polymer chain that was induced by the ionization of the boronic acid groups into negatively charged boronate groups [27] by which electrostatic repulsion was created within the polymer chain. This idea is supported by the study made by Chandar et al. on the conformations of a pyrene-labeled poly(acrylic acid) using fluorescence of the excimer/monomer emissions. They found that the polymer changed from coiled to extended form with increasing pH [28]. In the presence of 30 M of H 2 O 2 , the relative excimer intensity markedly increased with the increase in pH until 10.9, and it later decreased at pH 11.3. The increase in the excimer intensity suggests that the polymer chain bearing phenol moieties is contracted with the increase in pH. The phenol groups are expected to be partially dissociated at around pH 10 since p a values of monomeric phenols are typically around 10. We therefore speculated that the contraction was caused by the hydrogen bonding between the dissociated phenolate group and the undissociated phenol group [29]. This bonding should be stronger than the hydrogen bonding between phenols only since stronger electrostatic interaction occurs.
To confirm whether the above-mentioned interpretation is appropriate, effect of ionic strength on the response against H 2 O 2 was evaluated ( Figure S19). The response was largely suppressed in the presence of NaCl, which is attributable to disruption of the electrostatic interaction at the higher ionic strength. This result supports the view that the electrostatic attraction between phenol and phenolate is important for the response. The further increase in pH converts the undissociated phenol groups into dissociated phenolate groups, resulting in the creation of electrostatic repulsion between the negatively charged phenolate groups. Thus, the extension of the polymer chain is induced and the excimer emission  intensity decreases. According to these results, it was found that the H 2 O 2 -responsive change in the fluorescence intensity ratio is maximal at pH 10.9.

Effect of Monomer Composition.
In order to evaluate the effect of boronic acid density on the polymer chain, copolymers having different monomer compositions (Table 1) were used. To compare responses of these polymers towards H 2 O 2 , the magnitude of increase in the excimer to monomer intensity ratio was plotted against H 2 O 2 concentration, as shown in Figure 5. When compared with the response of polymer B-11, it is apparent that the decrease in the boronic acid content resulted in depression of the change in the excimer to monomer emission intensity ratios. Polymer B-12, which has a 1 : 2 molar ratio between the boronic acid and acrylamide units, exhibited a slightly diminished response. Moreover, polymer B-13 showed a further diminished response. In the case of polymer B-01, which does not have boronic acid unit, no virtual fluorescence response was induced. These observations clearly support the view that the existence of boronic acid group is indispensable for the occurrence of the responsiveness, and the reaction with H 2 O 2 causes conformational alterations in the copolymer, resulting in the fluorescence spectral changes. Unexpectedly, polymer B-10, which has the highest boronic acid content, showed a much lower response as compared to polymer B-11. Polymer B-10 mainly consists of boronic acid units with a small content of pyrene units and no acrylamide unit. Since the boronic acid unit is sterically bulkier than acrylamide unit, the conformational flexibility of the polymer chain should be relatively restricted due to the steric hindrance between neighboring boronic acid moieties. This could be the reason for the suppression of response against H 2 O 2 .

Conclusions
In conclusion, we established that the copolymers containing boronic acid and pyrene units can be utilized for the fluorometric sensing of H 2 O 2 . The copolymer exists in a relatively extended conformation in the absence of H 2 O 2 , whereas the polymer chain is contracted by the reaction of the boronic acid moieties with H 2 O 2 . This conformational change induces aggregation of the originally isolated pyrene groups as illustrated in Scheme 3. As a result, relative intensity of excimer emission with respect to monomer emission increases with H 2 O 2 concentration. Accordingly, the present methodology enables us to measure H 2 O 2 by means of