Novel Boronate Probe Based on 3-Benzothiazol-2-yl-7-hydroxy-chromen-2-one for the Detection of Peroxynitrite and Hypochlorite

Derivatives of coumarin, containing oxidant-sensitive boronate group, were recently developed for fluorescent detection of inflammatory oxidants. Here, we report the synthesis and the characterization of 3-(2-benzothiazolyl)-7-coumarin boronic acid pinacol ester (BC-BE) as a fluorescent probe for the detection of peroxynitrite (ONOO–), with high stability and a fast response time. The BC-BE probe hydrolyzes in phosphate buffer to 3-(2-benzothiazolyl)-7-coumarin boronic acid (BC-BA) which is stable in the solution even after a prolonged incubation time (24 h). BC-BA is slowly oxidized by H2O2 to form the phenolic product, 3-benzothiazol-2-yl-7-hydroxy-chromen-2-one (BC-OH). On the other hand, the BC-BA probe reacts rapidly with ONOO−. The ability of the BC-BA probe to detect ONOO– was measured using both authentic ONOO– and the system co-generating steady-state fluxes of O2•– and •NO. BC-BA is oxidized by ONOO– to BC-OH. However, in this reaction 3-benzothiazol-2-yl-chromen-2-one (BC-H) is formed in the minor pathway, as a peroxynitrite-specific product. BC-OH is also formed in the reaction of BC-BA with HOCl, and subsequent reaction of BC-OH with HOCl leads to the formation of a chlorinated phenolic product, which could be used as a specific product for HOCl. We conclude that BC-BA shows potential as an improved fluorescent probe for the detection of peroxynitrite and hypochlorite in biological settings. Complementation of the fluorescence measurements by HPLC-based identification of oxidant-specific products will help to identify the oxidants detected.


Introduction
Coumarin skeleton is frequently used to construct a range of fluorescent dyes due to high fluorescent quantum yields and tunable emission wavelengths. The emission of coumarin-based fluorophores can be finely tuned by appropriate substitution in 2Hchromen-2-one skeleton. Fluorescence can be red shifted by the placement of electrondonating groups in the six-or seven-position or electron-accepting groups in the three-or four-position of the skeleton [1].
A widely used example of coumarin dyes is C.I. Disperse Yellow 82 [2] which, in the 7position of the coumarin ring, contains the N,N-diethylamino group and, in the 3-position, a benzimidazole residue. Derivatives of coumarin generally show good photostability which is rather unusual among fluorescent dyes [3]. position, a benzimidazole residue. Derivatives of coumarin generally show good photostability which is rather unusual among fluorescent dyes [3].
The detection of reactive oxygen and nitrogen species is becoming more important these days. An excess of reactive oxygen species (ROS) and reactive nitrogen species (RNS) are implicated in pathologies such as cancer, cardiovascular, and neurodegenerative diseases [23]. To understand how these oxidative stressors participate in cellular function it is important to detect when, where, and what kind of specific products are produced. Significant information about reactive oxidants can be obtained using high performance liquid chromatography, mass spectrometry or other analytical procedure, but only connecting these methods with fluorescence and chemiluminescence approaches provide real-time monitoring.
There are two main fluorescence and chemiluminescence strategies for ROS/RNS detection. The first one is based on aromatic compounds that undergo oxidation to a fluorescent product (redox probes) and, the second, in which compound contains masked fluorophore. These are often called "non-redox" probes, as the fluorescence of the probe is uncovered through nucleophilic attack of the reactive species on the blocking group.
Continuing our research on the synthesis of low molecular weight boronate probes for the detection of peroxynitrite [27][28][29], we focused our efforts on the synthesis of a coumarin probe with the excitation band located above 400 nm. With the exception of coumarin 7-boronic acid, CBA, so far boronate probes based on the coumarin skeleton have been obtained by the boronobenzylation process [30]. However, the oxidative conversion of boronobenzyloxycarbonyl-and boronobenzylcoumarin derivatives not only produces a fluorophore but also releases quinone methide (QM) moiety. The self-immolation of such a moiety results in the delayed formation of a fluorescent product, while fluorophores with direct derivatization by the boronate group produce the fluorescent product instantly upon oxidation [28]. Another potential disadvantage of boronobenzylated probes arises from the fact that QM as an electrophile may influence the redox state of the cell and thus influence the redox environment studied. [31] Here, we report the synthesis and the characterization of a novel 3-(2-benzothiazolyl)-7-coumarin boronic acid pinacol ester, BC-BE, an analogue of the CBA probe with a benzothiazole residue in the 3-position of the coumarin ring (Scheme 1). In aqueous solutions containing a phosphate buffer (pH 7.4), the BC-BE probe undergoes fast hydrolysis to its boronic acid (BC-BA). Oxidation of BC-BE results in instantaneous formation of 3-benzothiazol-2-yl-7-hydroxy-chromen-2-one BC-OH, a fluorescent dye with improved photophysical properties as compared to COH. Scheme 1. Pro-coumarin boronate based probes.

Synthesis
The BC-BE boronate probe was obtained in a two-step protocol starting from the fluorophore BC-OH, prepared from 3-cyano-7-hydroxycoumarin (3) and ortho-aminothiophenol (4) according to the published procedure [32] (Scheme 2). In the first step, the phenolic hydroxyl group was converted into the appropriate triflate BC-OTf with a 99% yield. In the second step, BC-OTf was transformed to BC-BE in a Pd(dppf)Cl 2 -assisted reaction with bis(pinacolato)diboron. Since the probe was designed to detect peroxynitrite, the anticipated minor product, BC-H coumarin derivative was also prepared via a classical one-step condensation between appropriate ortho-hydroxybenzaldehyde (1b) and benzothiazole-2-carbonitrile (2).

Spectroscopic Response of BC-BE
Introduction of the benzothiazolyl group into the coumarin skeleton has only a minor effect on the acid-base properties of the phenolic hydroxyl group. In fact, the reported pKa values of COH and BC-OH are 6.89 [33] and 7.0 [34], respectively. Photophysical properties of the novel boronate probe and the parent hydroxycoumarin BC-OH, as well as COH are compared in Table 1. Figure 1 shows absorption and emission spectra of BC-BA, BC-H and BC-OH. Presented data demonstrate that boronate group significantly reduces the emission of the BC-OH coumarin. BC-BA exhibited a maximal UV-absorption at 371 nm (ε = 25,500 M −1 cm −1 ) and a maximal emission at 473 nm (Φ = 0.17). BC-OH has absorption and emission bands red shifted in comparison with BC-BA. More importantly, this dye has a higher quantum yield of emission than BC-BA and a higher extinction coefficient (resulting in higher brightness) than COH. Bolus addition of peroxynitrite to the solution of the BC-BA probe red shifts the absorption band ( Figure 1c) and turns on fluorescence ( Figure 1d).

Spectroscopic Response of BC-BE
Introduction of the benzothiazolyl group into the coumarin skeleton has only a minor effect on the acid-base properties of the phenolic hydroxyl group. In fact, the reported pKa values of COH and BC-OH are 6.89 [33] and 7.0 [34], respectively. Photophysical properties of the novel boronate probe and the parent hydroxycoumarin BC-OH, as well as COH are compared in Table 1. Figure 1 shows absorption and emission spectra of BC-BA, BC-H and BC-OH. Presented data demonstrate that boronate group significantly reduces the emission of the BC-OH coumarin. BC-BA exhibited a maximal UV-absorption at 371 nm (ε = 25,500 M −1 cm −1 ) and a maximal emission at 473 nm (Φ = 0.17). BC-OH has absorption and emission bands red shifted in comparison with BC-BA. More importantly, this dye has a higher quantum yield of emission than BC-BA and a higher extinction coefficient (resulting in higher brightness) than COH. Bolus addition of peroxynitrite to the solution of the BC-BA probe red shifts the absorption band ( Figure 1c) and turns on fluorescence (Figure 1d).  3 According to quinine sulfate (ref 0.54 in water) [21], 4 NaHCO3-NaOH buffer (10 mM, pH 10.0, containing 1% EtOH) [35], 5 Fluorescence lifetime was determined by TCSPC single photon counting using FL900 spectrofluorometer(Edinburgh Instruments, Livingston, UK).

Reactivity towards Biological Oxidants
The effectiveness of the BC-BA probe for the detection of peroxynitrite was also measured using the system co-generating steady fluxes of O2 •-(from hypoxanthine (HX) and xanthine oxidase XO) and • NO (from spermine-NONOate). The profile of the BC-BA -•-

Reactivity towards Biological Oxidants
The effectiveness of the BC-BA probe for the detection of peroxynitrite was also measured using the system co-generating steady fluxes of O 2 •-(from hypoxanthine (HX) and xanthine oxidase XO) and • NO (from spermine-NONOate). The profile of the BC-BA probe oxidation as a measure of ONOOformation in a matrix of various fluxes of O 2

•-
and • NO is shown in Figure 2a. Next, we determined the stoichiometry of the reaction between BC-BA and peroxynitrite. Using HPLC, we analyzed products formed after the bolus addition of ONOOto the phosphate buffer containing the BC-BA probe. Figure 3a shows that, after 5 min of incubation, we observed the formation of the BC-OH coumarin as the main product. The Next, we determined the stoichiometry of the reaction between BC-BA and peroxynitrite. Using HPLC, we analyzed products formed after the bolus addition of ONOOto the phosphate buffer containing the BC-BA probe. Figure 3a shows that, after 5 min of incubation, we observed the formation of the BC-OH coumarin as the main product. The formation of BC-OH can be observed with the naked eye, as shown in Figure 3c. The hue of fluorescence also changes ( Figure 3d). Figure 3b showed that BC-BA reacts with ONOOforming the BC-OH dye with c.a. 90% yield. These results are consistent with a previously reported reactivity of arylboronate-derived probes toward peroxynitrite [30]. Peroxynitriteinduced oxidation of boronic acid Ar-B(OH) 2 or ester Ar-B(pin) proceeds via two pathways, and typically leads to the formation of minor but ONOO --specific products (ArNO 2 , ArH) in addition to the major phenolic product (ArOH). Therefore, we anticipated that the BC-BA reaction with ONOOwould produce the corresponding minor products (BC-NO 2 and BC-H), as shown in Scheme 3. However, under the experimental conditions used we only detected BC-H (c.a. 9% yield), probably due to the usage of ethanol as an organic co-solvent, which efficiently reduces the phenyl-type radical BC • to BC-H (Scheme 2). Previously, it has shown that in the presence of phenyl radical scavengers (2-propanol), the phenyl radical formed in the radical pathway is almost quantitatively converted into the product in which the boronate moiety is replaced by a hydrogen atom [36,37]. previously reported reactivity of arylboronate-derived probes toward peroxynitrite [30].
Peroxynitrite-induced oxidation of boronic acid Ar-B(OH)2 or ester Ar-B(pin) proceeds via two pathways, and typically leads to the formation of minor but ONOO --specific products (ArNO2, ArH) in addition to the major phenolic product (ArOH). Therefore, we anticipated that the BC-BA reaction with ONOOwould produce the corresponding minor products (BC-NO2 and BC-H), as shown in Scheme 3. However, under the experimental conditions used we only detected BC-H (c.a. 9% yield), probably due to the usage of ethanol as an organic co-solvent, which efficiently reduces the phenyl-type radical BC • to BC-H (Scheme 2). Previously, it has shown that in the presence of phenyl radical scavengers (2-propanol), the phenyl radical formed in the radical pathway is almost quantitatively converted into the product in which the boronate moiety is replaced by a hydrogen atom [36,37].  Boronate probe is also oxidized by other inflammatory oxidants such as hydrogen peroxide and hypochlorous acid [27][28][29][30]. Therefore, we also tested the reactivity of BC-BA toward H2O2 and HOCl. We also compared the kinetic profile of BC-OH formation during the oxidation of BC-BA to the profile of the COH formation from a simple boronate probe, coumarin-7-boronic acid (CBA). Figure 4a shows the buildup of emission at 442 nm during the reaction between those boronic probes and H2O2 (5 mM). It is evident that both probes release products (BC-OH or COH) with comparable reaction rates. However, the signal intensity of BC-OH is 3-fold higher than the signal emission of COH, demonstrating significantly higher brightness of the product, consistent with the fluorescence parameters listed in Table 1. It is also worth emphasizing that in the reaction of the BC-BA probe with H2O2, the BC-OH coumarin is formed as the sole product (Figure 4b). [27][28][29][30]. Boronate probe is also oxidized by other inflammatory oxidants such as hydrogen peroxide and hypochlorous acid [27][28][29][30]. Therefore, we also tested the reactivity of BC-BA toward H 2 O 2 and HOCl. We also compared the kinetic profile of BC-OH formation during the oxidation of BC-BA to the profile of the COH formation from a simple boronate probe, coumarin-7-boronic acid (CBA). Figure 4a shows the buildup of emission at 442 nm during the reaction between those boronic probes and H 2 O 2 (5 mM). It is evident that both probes release products (BC-OH or COH) with comparable reaction rates. However, the signal intensity of BC-OH is 3-fold higher than the signal emission of COH, demonstrating significantly higher brightness of the product, consistent with the fluorescence parameters listed in Table 1. It is also worth emphasizing that in the reaction of the BC-BA probe with H 2 O 2 , the BC-OH coumarin is formed as the sole product (Figure 4b). [27][28][29][30]. Boronate probe is also oxidized by other inflammatory oxidants such as hydrogen peroxide and hypochlorous acid [27][28][29][30]. Therefore, we also tested the reactivity of BC-BA toward H2O2 and HOCl. We also compared the kinetic profile of BC-OH formation during the oxidation of BC-BA to the profile of the COH formation from a simple boronate probe, coumarin-7-boronic acid (CBA). Figure 4a shows the buildup of emission at 442 nm during the reaction between those boronic probes and H2O2 (5 mM). It is evident that both probes release products (BC-OH or COH) with comparable reaction rates. However, the signal intensity of BC-OH is 3-fold higher than the signal emission of COH, demonstrating significantly higher brightness of the product, consistent with the fluorescence parameters listed in Table 1. It is also worth emphasizing that in the reaction of the BC-BA probe with H2O2, the BC-OH coumarin is formed as the sole product (Figure 4b). [27][28][29][30]. Emission spectra recorded after the addition of a micromolar concentration of H 2 O 2 into the phosphate buffer solution containing BC-BA are shown in Figure 5a. The intensity of recorded spectra are lower in comparison with the intensity of fluorescence measured after the reaction of BC-BA with ONOO - (Figure 1d). For comparison, we also studied the reaction of BC-BA with hypochlorous acid. Emission spectra (Figure 5b) reveal that the BC-BA probe is converted to the BC-OH dye. In contrast to the reaction of the probe with hydrogen peroxide, however, the signal intensity is significantly higher. HPLC analysis of the reaction mixture (Figure 6a) shows that besides BC-OH, another product is also formed. We attributed this product to the chlorinated derivative of the coumarin BC-OH(Cl) since the product with the same retention time (6.1 min) was also detected after the addition of HOCl to the BC-OH solution. By analyzing the disappearance of the BC-BA probe and the concentration of products formed with an increase in the amount of added HOCl, it can be seen that the maximum yield of BC-OH reaches only ca. 25%. Moreover, an excess of HOCl causes the disappearance of BC-OH. This is consistent with our previous investigations in which we demonstrated that the phenolic product (luciferin or 6-(2-benzothiazolyl)-2-naphthalenol) released from the boronate probes (LBA, PCL-1 or NAB-BE) undergoes a further reaction with HOCl, leading to the formation of 7 -chloroluciferin [27,28] or 6-(1,3-benzothiazol-2-yl)-1-chloronaphthalen-2-ol [29], a product specific for HOCl. Emission spectra recorded after the addition of a micromolar concentration of H2O2 into the phosphate buffer solution containing BC-BA are shown in Figure 5a. The intensity of recorded spectra are lower in comparison with the intensity of fluorescence measured after the reaction of BC-BA with ONOO - (Figure 1d). For comparison, we also studied the reaction of BC-BA with hypochlorous acid. Emission spectra (Figure 5b) reveal that the BC-BA probe is converted to the BC-OH dye. In contrast to the reaction of the probe with hydrogen peroxide, however, the signal intensity is significantly higher. HPLC analysis of the reaction mixture (Figure 6a) shows that besides BC-OH, another product is also formed. We attributed this product to the chlorinated derivative of the coumarin BC-OH(Cl) since the product with the same retention time (6.1 min) was also detected after the addition of HOCl to the BC-OH solution. By analyzing the disappearance of the BC-BA probe and the concentration of products formed with an increase in the amount of added HOCl, it can be seen that the maximum yield of BC-OH reaches only ca. 25%. Moreover, an excess of HOCl causes the disappearance of BC-OH. This is consistent with our previous investigations in which we demonstrated that the phenolic product (luciferin or 6-(2-benzothiazolyl)-2-naphthalenol) released from the boronate probes (LBA, PCL-1 or NAB-BE) undergoes a further reaction with HOCl, leading to the formation of 7′-chloroluciferin [27,28] or 6-(1,3-benzothiazol-2-yl)-1-chloronaphthalen-2-ol [29], a product specific for HOCl.

The Effect of Compounds on Cell Metabolic Activity
To determine the potential usage of synthesized compounds in cell-based assays, next the impact of probes on the metabolic activity of human colon cancer HT29 cells was studied. Compounds were dissolved in ethanol/dimethylsulfoxide mixture (9:1) and added to the cells in a volume not exceeding 2% of the medium, which had no cytotoxic effect on cells. To determine the cytotoxic potential of studied compounds, cells were incubated for 8 h in their presence at the range of 0-200 µM concentration. As it is presented in Figure 7a-c, compounds BC-OH and BC-BA at 20 µM concentration had no impact on HT29 cells' metabolic activity; therefore, both of them can be potentially used in studies requiring cell incubation for at least 20 min. The direct comparison of cytotoxic potential within the studied range of concentration revealed that compound BC-OH was more cytotoxic than BC-BA. The highest cytotoxicity of BC-OH was detected for 200 µM with decreased metabolic activity by almost 20% compared to the control cells. Microscopic observations performed with calcein AM ester confirmed the lack of cytotoxic effects of both compounds at 20 µM on HT29 cells, as shown in Figure 7d. In healthy cells with active esterases, there is a visible strong cytosolic green fluorescence of calcein. Cells incubated with 200 µM of BC-OH had lower cytoplasmic esterase activity, thus a decreased green fluorescence of calcein was observed, as well as the presence of some less attached and more rounded cells.

The Effect of Compounds on Cell Metabolic Activity
To determine the potential usage of synthesized compounds in cell-based assays, next the impact of probes on the metabolic activity of human colon cancer HT29 cells was studied. Compounds were dissolved in ethanol/dimethylsulfoxide mixture (9:1) and added to the cells in a volume not exceeding 2% of the medium, which had no cytotoxic effect on cells. To determine the cytotoxic potential of studied compounds, cells were incubated for 8 h in their presence at the range of 0-200 µM concentration. As it is presented in Figure 7a-c, compounds BC-OH and BC-BA at 20 µM concentration had no impact on HT29 cells' metabolic activity; therefore, both of them can be potentially used in studies requiring cell incubation for at least 20 min. The direct comparison of cytotoxic potential within the studied range of concentration revealed that compound BC-OH was more cytotoxic than BC-BA. The highest cytotoxicity of BC-OH was detected for 200 µM with decreased metabolic activity by almost 20% compared to the control cells. Microscopic observations performed with calcein AM ester confirmed the lack of cytotoxic effects of both compounds at 20 µM on HT29 cells, as shown in Figure 7d. In healthy cells with active esterases, there is a visible strong cytosolic green fluorescence of calcein. Cells incubated with 200 µM of BC-OH had lower cytoplasmic esterase activity, thus a decreased green fluorescence of calcein was observed, as well as the presence of some less attached and more rounded cells.
The purity of the final compounds was tested by HPLC (Shimadzu) equipped with a photodiode array detector and analytical column-Phenomenex Kinetex Core-Shell C18 (100 mm × 4.6 mm; 2.6 µm). The mobile phase was a gradient prepared from acetonitrile with 0.1% of TFA (component A) and water with 0.1% of TFA (component B). The analytes were eluted by an increase of A concentration from 10-100% over 10 min at the flow rate of 1.5 mL/min. The column temperature was set at 30 °C. 1 H NMR spectra were recorded with a Bruker Avance DPX 250 spectrometer at 250 MHz, respectively (see supplementary materials). Compounds were dissolved in DMSO-d6 and TMS was added as internal reference. Mass spectra [TOF MS (ESI+)] were recorded on a Synapt G2-Si spectrometer (Waters, Milford, MA, USA).
The purity of the final compounds was tested by HPLC (Shimadzu) equipped with a photodiode array detector and analytical column-Phenomenex Kinetex Core-Shell C18 (100 mm × 4.6 mm; 2.6 µm). The mobile phase was a gradient prepared from acetonitrile with 0.1% of TFA (component A) and water with 0.1% of TFA (component B). The analytes were eluted by an increase of A concentration from 10-100% over 10 min at the flow rate of 1.5 mL/min. The column temperature was set at 30 • C. 1 H NMR spectra were recorded with a Bruker Avance DPX 250 spectrometer at 250 MHz, respectively (see Supplementary materials). Compounds were dissolved in DMSOd 6 and TMS was added as internal reference. Mass spectra [TOF MS (ESI+)] were recorded on a Synapt G2-Si spectrometer (Waters, Milford, MA, USA).
Absorption spectra were recorded on UV-Vis-NIR spectrophotometer Jasco-V670. Steady-state and time-resolved fluorescence spectra were recorded on Edinburgh Analytical Instruments FL900.

Cell Culture and Exposure Conditions
Human colon carcinoma cell line HT29 was obtained from ATCC, (Manassas, VA, USA). Cells were grown in DMEM with a 10% fetal bovine serum (FBS) medium supplemented with 100 U/mL penicillin, 100 µg/mL streptomycin, and 25 µg/mL amphotericin B. All cell culture experiments were performed in a humidified 5% CO 2 and 95% atmosphere at 37 • C. All cell culture reagents were obtained from Life Technologies (Carlsbad, CA, USA).

Cell Metabolic Activity
Metabolic activity was evaluated with MTT assay. Briefly, cells were seeded into 96well plate at 1 × 10 4 cells/well density in 100 µL complete medium and grown overnight, and then incubated in the presence of studied compounds for another 8 h. After this, 20 µL of MTT reagent (5 mg/mL) was added for 120 min. After that time, MTT was removed, and formazan precipitates were solubilized by adding 100 µL of DMSO. Absorbance was measured at 570 nm using the Synergy 2 BioTek Microplate Reader (BioTek, Winooski, VT, USA).

Conclusions
We used 3-benzothiazol-2-yl-7-hydroxy-chromen-2-one fluorophore to develop a novel boronate probe for the selected biological oxidants. The BC-BA probe showed similar reactivity toward selected inflammatory oxidants to the previously reported CBA probe [17], producing fluorescent BC-OH phenolic product. In comparison to CBA, BC-BA is significantly more lipophilic, which may improve its cellular uptake. Furthermore, the fluorescence of the oxidation product is red shifted and its brightness is ca. 3-fold higher. This may help in the successful application of the probe for the imaging of biological oxidants in cultured cells. Detection of the minor products characteristic for a specific oxidant (ONOOor HOCl) will allow the unambiguous identification of the oxidants involved in probe oxidation.
Further structure modifications may be introduced to the developed scaffold to modulate the water solubility of the probe, fluorescence properties of the probe and the product and to target the probe to specific subcellular or extracellular compartments.

Supplementary Materials:
The following are available online. Figure S1: HPLC chromatograms of the BC-BA in aqueous solution containing phosphate buffer (0.1 M, pH 7.4), dtpa (10 µM) and EtOH (10%): freshly made solution (above), after 10 min (below). The traces were collected using the absorption detector set at 330 nm, Figure S2