Detection of HSO3-: A Rapid Colorimetric and Fluorimetric Selective Sensor for Detecting Biological SO2 in Food and Living Cells.

A rapid fluorescent probe based on the conjugate of chromone and benzothiazole moiety was presented, which could selectively respond to HSO3- over other common anions and thiols. The function of the probe relies on nucleophilic addition to break down the π-conjugation. The probe can be used as a signal tool to determine HSO3- levels in sugar-based food and living cells.

Introduction

Recently, anion chemosensors have attracted significant interest due to their crucial role in a wide range of chemical, biological, and environmental processes.[1−7] Sulfur dioxide (SO2) is one of the most largely distributed air pollutants, and it has been investigated extensively in toxicology. Sulfur dioxide is associated with asthma, chronic bronchitis, morbidity, and mortality increase in old people and infants. It also causes breathing difficulty,[8] lung cancer, cardiovascular diseases, and neurological disorders.[9] Inhaled SO2 is easily hydrated to release sulfite (SO32–) and bisulfite (HSO3–) in 3:1 M/M (neutral fluid),[10] and the toxicity of SO2 is due to these two anions. Bisulfite is considered as an essential preservative for several foods, beverages, and pharmaceutical products for preventing oxidation and bacterial growth, and enzymatic reactions during production and storage.[11−13] Despite its valuable properties, bisulfites have harmful effects on tissue, cells, and bio-macromolecules. It can cause visible damage such as necrosis, inhibit cell division, and induce micronucleus, which often lead to the death of a cell.[14,15] Thus, because of its toxicity, the development of a rapid, sensitive, selective, low cost, and smart detection method for the lethal bisulfite pollutant is of significant importance.

A number of conventional analytical techniques for developing colorimetric and fluorimetric bisulfite-based sensors have been reported, such as spectrophotometry,[16,17] spectrofluorimetry,[18] chemiluminescence measurements,[19,20] phosphorimetry,[21] chromatography,[22] and electrochemistry.[23,24] But, these methods require troublesome sample pretreatment, specific reagent preparation, a lot of time, and complicated instrumentation. In addition, some of them are not sensitive enough to determine very low concentrations of HSO3–. On the other hand, fluorescence spectroscopy has been widely used in sensors due to its high sensitivity with less consumption, operational simplicity, real time monitoring, better selectivity, as well as good reproducibility. In recent years, chemodosimetric reactions such as the Michael addition,[25−29] selective reactions with aldehyde or levulinate,[30−34] coordinative interactions,[35,36] and those involving a noncovalent indicator displacement assay[37,38] have been applied for specific HSO3– sensing. Recently, some ratiometric fluorescent chemodosimeters have been reported for bisulfite sensing, which can afford a built-in correction for environmental effects.[39−45] A few groups have developed fluorescent sensors for the rapid detection of HSO3–. Tian et al.[46] and Feng et al.[47] have reported bisulfite probes for rapid detection. But, their probes are applicable only in half water conditions, and they displayed a long response time (>5 min). Thus, it is very important to develop a good, water soluble, rapid, highly selective, and sensitive probe for bisulfite anion.

In this work, we designed a good water soluble chromone-benzothiazole dye (CBD, Scheme ), which is fast-responding, highly sensitive, and selective for the detection of SO2 derivatives. This receptor (CBD) undergoes a nonreversible chemical change with HSO3–, resulting in colorimetric and fluorescent dual responses. Because of the nucleophilic nature of HSO3–, we have developed a reaction-based receptor for HSO3– to avoid the complication due to induced hydrogen bonding. The nucleophilic attack of HSO3– toward this type of dye would interrupt the π-conjugation and block the intramolecular charge transfer (ICT) process. This new HSO3– probe is promising because of its very fast response toward the pollutant (HSO3–), attractive sensing property, and it is easy to synthesize. To prove the feasibility of our probe, we used it for quantitative detection of bisulfate in food samples and living cells.

Scheme 1

Synthetic Routes for CBD Receptor

The synthesis of the receptor CBD is depicted in Scheme . Precursor 1,3-benzothiazol-2-yl-acetonitrile (1)[48] was prepared through a cyclization reaction between o-amino-thiophenol and malononitrile in the presence of acetic acid (eq 1, Scheme ). C–O coupled cyclization with formylation of 2-acetyl-p-cresol was performed using dimethylformamide (DMF)–POCl3 to obtain 3-formyl-6-methylchromone (2, eq 2).[49] Piperidine-mediated condensation of 1,3-benzothiazol-2-yl-acetonitrile (1) with 3-formyl-6-methylchromone (2) furnished the receptor CBD (eq 3). The structure of the new compound (CBD) was confirmed by 1H NMR, 13C NMR, Fourier transform infrared (FT-IR), and high-resolution mass spectrometry (HR-MS) analyses.

Results and Discussion

To understand the selective and sensitive behavior of the probe, absorption was carried out in a CH3CN–H2O solution (1:9, v/v, 10 mM N-(2-hydroxyethyl)piperazine-N′-ethanesulfonic acid (HEPES), pH 7.4) with various anions such as HSO3–, CN–, F–, Cl–, Br–, I–, AcO–, SCN–, SH–, N3–, S2–, HSO4–, PPi, and Pi (Figure S1a, SI). CBD exhibited an absorption peak at 471 nm, which slightly blue shifted (471–467 nm) upon addition of HSO3–, and the blue shifting was gradually increased upon addition of HSO3–. This result clearly indicates that the π-conjugation between aromatic heterocycle 1 and 2 of CBD was destroyed. As a consequence, the yellow-colored solution become colorless, and this could be observed easily by the naked eye (Figure S1a, inset, SI). Thus, nucleophilic addition of HSO3– to the vinyl linkage of the probe interrupts the ICT process and furnishes a new nonconjugated CBD–SO3H adduct. The absorbance at 471 nm changed by almost 21-fold (from 0.0016 to 0.033) in the presence of HSO3– (Figure S1b, SI). In contrast, the addition of other anions resulted in negligible responses (Figure ).

Figure 1

Variation of absorbance for probe CBD in a CH3CN–H2O solution (1:9, v/v, 10 mM HEPES, pH 7.4) in the presence of various anions.

After obtaining good results in absorption spectra, we were interested in studying the fluorescence spectra with this probe. As displayed in Figure a, the CBD chemodosimeter exhibited very weak fluorescence bands centered at 530 nm (Φ = 0.001) upon excitation at 460 nm in the absence of HSO3– or in the presence of other anions, namely, CN–, F–, Cl–, Br–, I–, AcO–, SCN–, SH–, N3–, S2–, HSO4–, PPi, and Pi. Only a drastic change was observed when bisulfite was added to the solution of the chemodosimeter.

Figure 2

(a) Fluorescence emission spectra of CBD (c = 2.0 × 10–5 M) with HSO3– (c = 2.0 × 10–6 M) at pH 7.4 in CH3CN–H2O (1:9, v/v) [gradual addition of 1 equiv of HSO3–, respectively] (λex = 460 nm); the inset shows the naked eye fluorescence change of CBD with addition of HSO3–. (b) Fluorescence change at 530 nm of CBD upon gradual addition of HSO3–.

In the fluorescent spectra, the peak intensity at 530 nm gradually increased and a new peak appeared at 570 nm (Φ = 0.16). When HSO3– was gradually added to the solution of the CBD, the nonfluorescent solution exhibited green fluorescence (Figure a, inset). The emission intensity of the probe at 530 nm showed a drastic change from 9.87 to 610.86 (62-fold) in the presence of bisulfite (Figure b). To test the selective nature of CBD toward HSO3–, we investigated the effect of important physiological anions such as CN–, F–, Cl–, Br–, I–, AcO–, SCN–, SH–, N3–, S2–, HSO4–, PPi, and Pi. Only the nucleophile CN– induced a little change in the fluorescence intensity (Figure a). The competition experiments clearly suggested that the probe CBD is not affected by interference from the other anions (Figure b). CBD displayed remarkable selectivity toward HSO3– over other anions, which might make it suitable for precise HSO3– detection in complex samples. Because SO2 can be endogenously generated in cells during oxidation of H2S or sulfur containing amino acids, a promising HSO3– probe should have high selectivity for HSO3– over other reactive sulfur species such as H2S, Phen, Gly, Cys, Hcy, and GSH. From Figure S2, it is clear that our probe is highly selective toward HSO3– over these sulfur compounds.

Figure 3

(a) Fluorescence response of CBD (2.0 × 10–5 M) toward various anions (1 equiv) at pH 7.4 in CH3CN–H2O (1:9, v/v). (b) The fluorescence intensity of CBD (2.0 × 10–5 M) toward HSO3– (1 equiv) containing 10 equiv of various anions (λex = 460 nm).

From the calibration curve (Figure S3, SI), the limit of detection (LOD) was calculated by using the formula K × Sb1/S,[50] where Sb1 is the standard deviation of blank measurements and S is the slope of the calibration curve. The limit of detection of CBD for HSO3– is 0.45 μM, which is significantly low for the detection of HSO3– found in many chemical systems. Wang et al.[51] reported a probe for bisulfite in which fluorescence recovery is possible in the presence of various reactive oxygen species (ROS). In the case of our probe, nonreversible fast sensing occurred in the presence of various ROS species (Figure S6).

From the time-dependent fluorescence spectra, it was found that CBD behaves as a very fast probe toward the bisulfite ion. The reaction was completed in less than 1 min with a rate constant of 22.1 × 10–2 s–1. This result strongly suggests that CBD is a highly reactive probe toward bisulfite ions (Figures a and S4, SI). The fluorescence lifetime of CBD with and without bisulfate ion was detected by time resolved fluorescent spectra (Figure b). The fluorescence decay curves of the compounds were fitted by utilizing monoexponential functions (SI). The radiative rate constant kr and the total nonradiative rate constant knr of CBD and the CBD–SO3H adduct were calculated according to the equation ′I–1 = kr + knr, where kr = ϕf/′I.[52] For CBD, ′I = 2.20 ns, and for the CBD–SO3H adduct, ′I = 10.91 ns.

Figure 4

(a) Time vs fluorescence spectra of (a) CBD (c = 2.0 × 10–5 M) in the presence of 1 equiv of HSO3– (c = 2.0 × 10–4 M) at pH 7.4 in CH3CN/H2O (1:9, v/v) at different times [(2) 5, (3) 10, (4) 20, (5) 30, (6) 40, (7) 50, and (8) 60 s] (λex = 460 nm). (b) Fluorescence decay curves for CBD and CBD–SO3H adduct in CH3CN.

1H NMR and HR-MS studies were carried out for CBD and the CBD–SO3H adduct to analyze the reaction mechanism. From 1H NMR spectra, we can see that the vinyl protons at 7.859 ppm of CBD disappeared and new upfield signals appeared at 5.257 ppm after reaction with HSO3–. The signals of the aromatic protons were slightly upfield shifted. The appearance of a new symbolic peak at 5.257 ppm for this receptor confirms the nucleophilic addition of HSO3– to the vinyl π-bond (Scheme ). From the mass spectra, CBD showed a main peak at 345.0664 before addition of NaHSO3, which corresponds to the [CBD + H]+ species. After addition of NaHSO3, a peak appeared at 449.0746, which exactly matched with the adduct species [(CBD + HSO3– + H+) + Na].

Scheme 2

Proposed Mechanism for Sensing Bisulfite by CBD Receptor

Next, we were interested in determining bisulfite by using our probe in the solid state (using thin-layer chromatography (TLC) plates). This is an important experiment because we see the sensing of bisulfite with our probe without using any instrumental analyses. To perform this experiment, we prepared TLC plates by immersing the plates into the solution of CBD (2 × 10–4 M) in CH3CN and drying the plates in air (Figure .1). Next, this TLC plate was immersed in a solution of HSO3– and photographs were taken under ambient and UV light (Figure .2). This experiment proves that we can easily and instantly detect HSO3– qualitatively by the naked eye.

Figure 5

Color changes on test paper with (1) CBD and (2) CBD in the presence of NaHSO3.

Our CBD probe could also be used for the measurement of bisulfite in granulated sugar. For this purpose, we prepared a sample solution by dissolving 10 g of sugar in water. Then, this solution was diluted with water to 50 mL. From the data in Table , we see that our probe is able to determine bisulfite content in the sugar sample. So, this probe can easily be used to measure bisulfite level in food samples.

Table 1

Determination of Bisulfite in Food Samples Using CBD Probe

granulated sugar	bisulfite content (μmol/L)	added (μmol/L)	found (μmol/L)	recovery (%)
sample 1		4	3.98	99.5
	5.46	6	5.86	97.6
sample 2		2.5	2.46	98.4
	3.81	3	2.90	96.6

CBD is an excellent intracellular probe for detecting HSO3– due to its permeability as well as stability. Figure a depicts the bioimaging of human peripheral blood mononuclear cells (PBMCs) by CBD, when there is no added HSO3– from the outside. Here, cells show no significant fluorescence. Figure b–f shows increasing green fluorescence with enhanced concentration of HSO3– from 5 to 25 μM. Cell viability is represented in Figure S5 (SI), where up to 60 μM/L concentrations of CBD show around 61.957% viable cells, which predicts that this is a safe probe to use in a biological system. We used 30 μM/L CBD solutions for imaging, which shows a fairly high number of viable cells (78.066%). This confirms the nontoxic nature of the new receptor (CBD).

Figure 6

Human PBMCs (40×) treated with 30 μmol/L CBD under 530 nm fluorescence emission; (a) no added HSO3–, (b–f) with 5, 10, 15, 20, and 25 μM HSO3–, respectively.

Conclusions

We have introduced a smart fluorescent probe for detection of HSO3– in real and biological samples. The method employs the nucleophilic addition of HSO3– to the conjugated double bond of the receptor to block the π-conjugation between the heterocyclic aromatic skeletons. The proposed probe shows rapid (<1 min) and excellent selectivity toward bisulfite over other common anions and biothiols. Thus, we believe that this design concept will find important application and leads the way for further development for detecting SO2 derivatives in biological systems.

Experimental Section

General Methods

All materials were purchased from Sigma-Aldrich Chemicals Private Limited and were used without further purification. 1H NMR and 13C NMR spectra were recorded on Brucker 300 MHz instruments. CDCl3 was used as solvent with tetramethylsilane as an internal standard. Chemical shifts are expressed in δ units and coupling constants in Hz. Melting points were determined on a hot-plate melting point apparatus in an open-mouth capillary and were uncorrected. UV–vis titration experiments were performed on a Perkin Elmer Lambda 750 spectrophotometer and fluorescence experiments were done using a Perkin Elmer LS 55 with a fluorescence cell of 10 mm path. FT-IR spectra were recorded on a JASCO FT/IR-460 plus spectrometer, using KBr disks. Column chromatography was carried out by using silica gel 60 (60–120 mesh).

General Method of UV–Vis and Fluorescence Titrations

For UV–vis and fluorescence titrations, a stock solution of the probe was prepared (c = 2 × 10–5 ML–1) in CH3CN/H2O (1:9, v/v). The solutions of the guest anions and biothiols were prepared (2 × 10–4 ML–1) in CH3CN/H2O (1:9, v/v) at pH 7.4 by using 10 mM HEPES buffer. The solution of the sensor was prepared by an appropriate dilution technique. The spectra of these solutions were recorded by UV–vis and fluorescence methods. All solvents were purchased from domestic suppliers and were used after distillation.

Synthesis of 2-(1,3-Benzothiazol-2-yl)acetonitrile (1)

A mixture of o-aminothiophenol (500 mg, 4.0 mmol) and malononitrile (270 mg, 4.1 mmol) in absolute ethanol (5 mL) was treated with glacial AcOH (0.5 mL) and allowed to stir at room temperature overnight. The formed precipitate was filtered and crystallized from ethanol to afford a yellow product (yield 80%).

Mp: 100–101 °C. 1H NMR (CDCl3, 300 MHz) δ (ppm): 7.991 (d, J = 8.1 Hz, 1H), 7.860 (d, J = 7.8 Hz, 1H), 7.450 (m, 2H), 3.621 (s, 2H). 13C NMR (CDCl3, 75 MHz): 158.12, 152.74, 135.36, 126.62, 125.88, 123.30, 121.62, 114.76, 23.09. FT-IR (KBr): 3056, 2968, 2924, 2264, 1508, 1420, 1226, 1112, 1016, 875, 761 cm–1.

Synthesis of 3-Formyl-6-methylchromone (2)

POCl3 (0.3 mL, 13.33 mmol) was added drop wise with stirring to a solution of 2′-hydroxy-5′-methylacetophenone (500 mg, 3.33 mmol) and DMF (10 mL) at 0 °C. The resulting mixture was stirred at 50 °C under nitrogen for 2 h, when TLC analysis showed that no starting material existed, poured into ice-water (150 mL), and neutralized with NaOH solution (4 M). The solid was collected, washed with water, and dried. Purification by column chromatography (silica gel, EtOAc–hexanes, 1:5) afforded 3-formyl-6-methylchromone (450 mg, 72%). Mp: 172–174 °C. 1H NMR (CDCl3, 300 MHz) δ (ppm): 10.326 (s, 1H), 8.510 (s, 1H), 8.243 (d, J = 7.8 Hz, 1H), 7.726 (m, 1H), 7.478 (m, 2H), 2.410 (s, 3H). 13C NMR (CDCl3, 75 MHz): 188.59, 175.94, 160.68, 156.13, 134.84, 126.63, 126.09, 125.22, 120.23, 118.61, 22.23. FT-IR (KBr): 3053, 2868, 1698, 1646, 1470, 1321, 1109, 951, 775, 555 cm–1.

Synthesis of Receptor (CBD)

2-(1,3-Benzothiazol-2-yl)acetonitrile (1, 250 mg, 1.43 mmol) and 3-formyl-6-methylchromone (270 mg, 1.43 mmol) were dissolved in ethanol. One drop of piperidine was added to the solution. The reaction mixture allowed to stir at room temperature for 3 h. The crude products were filtered, washed with cold ethanol, and dried under vacuum. This crude residue was purified by column chromatography using silica gel (100–200 mesh) and 10% ethyl acetate in petroleum ether as eluent to obtain a colorless gummy liquid, which solidified on cooling (420 mg, 85%). 1H NMR (CDCl3, 300 MHz) δ (ppm): 7.859 (s, 1H), 7.727 (d, J = 8.1 Hz, 2H), 7.582 (m, 2H), 7.464 (t, J = 7.05 Hz, 3H), 6.971 (d, J = 7.05 Hz, 1H), 2.406 (s, 3H). 13C NMR (CDCl3, 75 MHz) δ (ppm): 22.56, 111.55, 119.72, 120.60, 121.64, 122.29, 122.54, 125.48, 125.88, 126.29, 126.86, 127.10, 130.38, 132.93, 134.03, 136.30, 151.20, 159.29, 165.63, 206.80.

HR-MS (ESI TOF) (m/z, %): 345.0664 [(CBD + H+), 100%], calculated: 345.0625. FT-IR (KBr): 3434, 2218, 1675, 1604, 1504, 1353, 1212, 1012, 890, 740, 589 cm–1. Mp: 120–123 °C.

Synthesis of CBD–SO3H Adduct

CBD was mixed with 1.2 equiv of NaSO3H in acetonitrile at room temperature to obtain a yellow solution. On removing the solvent, a solid product was obtained, which was used for 1H NMR and ESI-MS spectroscopy. 1H NMR (CDCl3, 300 MHz) δ (ppm): 7.658 (d, J = 7.5 Hz, 2H), 7.487 (2H, m), 7.29 (d, J = 9 Hz, 3H), 6.901 (d, J = 8.4 Hz, 1H), 5.257 (s, 1H), 2.336 (s, 3H). HR-MS (ESI TOF) (m/z, %): 449.0746 [(CBD + HSO3– + H+ + Na), 100%], calculated: 449.0744. FT-IR (KBr): 3443, 2207, 1675, 1494, 1383, 1203, 1012, 841, 750, 579 cm–1.